From magnus at rubidium.dyndns.org Sun Feb 1 00:10:55 2009 From: magnus at rubidium.dyndns.org (Magnus Danielson) Date: Sun, 01 Feb 2009 01:10:55 +0100 Subject: [time-nuts] Lady Heather Update In-Reply-To: <49846D8D.6080401@rubidium.dyndns.org> References: <49846D8D.6080401@rubidium.dyndns.org> Message-ID: <4984E88F.7030303@rubidium.dyndns.org> Magnus Danielson skrev: > Mark Sims skrev: >> Here is an updated version of the good Lady Heather's GPS Disciplined Oscillator control program for the Thunderbolt. This version adds command line support for setting the video mode: >> >> /VS = 800x600 >> /VM = 1024x768 (default) >> /VL = 1280x960 (also good for 1280x1024) >> >> >> /? for command line help >> >> >> Again, the compiled version is for DOS/WIN98 and mayby WIN ME, WIN/NT DOS mode on systems with a VESA compatible video BIOS. > > Fails to run on my old Win2k box... even with /2 it fails to find COM1 > while the Thunderbolt software does... After doing some attempts to hack the configuration so that the serial port does show up as COM1 rather than the laptops builtin modem it works, but only in text-mode. Ah well, a small grace as it is. I failed to toggle to Kalman filter. Got it working at least. Cheers, Magnus From magnus at rubidium.dyndns.org Sun Feb 1 00:18:02 2009 From: magnus at rubidium.dyndns.org (Magnus Danielson) Date: Sun, 01 Feb 2009 01:18:02 +0100 Subject: [time-nuts] Lady Heather Update In-Reply-To: References: Message-ID: <4984EA3A.2070608@rubidium.dyndns.org> Mark Sims skrev: > Hello again Magnus, Hi Mark, > If the program is started up and it is not seeing the Tbolt on the COM > port you should see the "Log: OFF" indicator, two lines of text that > are the plot area header, and the plot axes/grid. Pressing the > SPACE bar should bring up a help screen in place of the plot grid. > If all that is working, the problem is probably in the COM port. > It wants the tbolt to be at factory com settings (9600,N,8,1) That was not the problem. It failed with a big windows thing saying it would not find the COM1 and then the programmed died. > The program talks to the COM port UART directly. It probably won't > work with a USB converter. The default is COM1, the /2 command line > option should select COM2 (but I don't have a machine here with two > serial ports to test it... the com port code was lifted from another > program that did work with COM2). The trouble here was that COM1 was "felt" by the OS to be used for something else. Very strange. I had to force COM2 to become COM1 to solve it. /2 did not help as it actually was COM1 in hardware. > There are several command line options. use /? to see them. I found them by tossing random commands at it... fairly quickly. My remaining problem is to get it to run in graphical mode. Cheers, Magnus From jmiles at pop.net Sun Feb 1 01:10:05 2009 From: jmiles at pop.net (John Miles) Date: Sat, 31 Jan 2009 17:10:05 -0800 Subject: [time-nuts] Lady Heather Update In-Reply-To: <4984EA3A.2070608@rubidium.dyndns.org> Message-ID: I'm trying to bring the Win32 version up to date with the graphical version, but am dealing with a bug that apparently puts the Thunderbolt into some kind of mode where it returns junk data over the serial line. My Win32 build runs for a minute or so, draws some graphs that don't make sense, and then I have to go into TBoltMon.exe and do a warm reset to make it resume sending normal serial updates. Is this basically what the DOS version normally looks like (attached), obvious errors aside? -- john, KE5FX > -----Original Message----- > From: time-nuts-bounces at febo.com [mailto:time-nuts-bounces at febo.com]On > Behalf Of Magnus Danielson > Sent: Saturday, January 31, 2009 4:18 PM > To: Discussion of precise time and frequency measurement > Subject: Re: [time-nuts] Lady Heather Update > > > Mark Sims skrev: > > Hello again Magnus, > > Hi Mark, > > > If the program is started up and it is not seeing the Tbolt on the COM > > port you should see the "Log: OFF" indicator, two lines of text that > > are the plot area header, and the plot axes/grid. Pressing the > > SPACE bar should bring up a help screen in place of the plot grid. > > If all that is working, the problem is probably in the COM port. > > It wants the tbolt to be at factory com settings (9600,N,8,1) > > That was not the problem. It failed with a big windows thing saying it > would not find the COM1 and then the programmed died. > > > The program talks to the COM port UART directly. It probably won't > > work with a USB converter. The default is COM1, the /2 command line > > option should select COM2 (but I don't have a machine here with two > > serial ports to test it... the com port code was lifted from another > > program that did work with COM2). > > The trouble here was that COM1 was "felt" by the OS to be used for > something else. Very strange. I had to force COM2 to become COM1 to > solve it. /2 did not help as it actually was COM1 in hardware. > > > There are several command line options. use /? to see them. > > I found them by tossing random commands at it... fairly quickly. > > My remaining problem is to get it to run in graphical mode. > > Cheers, > Magnus > > _______________________________________________ > time-nuts mailing list -- time-nuts at febo.com > To unsubscribe, go to > https://www.febo.com/cgi-bin/mailman/listinfo/time-nuts > and follow the instructions there. > -------------- next part -------------- A non-text attachment was scrubbed... Name: window.gif Type: image/gif Size: 24803 bytes Desc: not available Url : http://www.febo.com/pipermail/time-nuts/attachments/20090131/86760c45/attachment-0001.gif From jra at febo.com Sun Feb 1 03:05:37 2009 From: jra at febo.com (John Ackermann N8UR) Date: Sat, 31 Jan 2009 22:05:37 -0500 Subject: [time-nuts] HP 3456A stability upgrade In-Reply-To: <4984CE83.3060807@febo.com> References: <4984CE83.3060807@febo.com> Message-ID: <49851181.3070703@febo.com> I need to clarify something about my earlier posting regarding the 3456A upgrade. The upgrade involves more than dropping in a new part; there are other circuit changes also required. Lion Electronics does the conversion, but doesn't supply the parts separately (after all, he's in business!). John From dforbes at dakotacom.net Sun Feb 1 04:00:27 2009 From: dforbes at dakotacom.net (David Forbes) Date: Sat, 31 Jan 2009 21:00:27 -0700 Subject: [time-nuts] Boatanchor docs: HP AC-4 counter module manual Message-ID: O Nuts of Time, I was recently asked to make a big PDF file of the HP AC-4 counter modules manual that I have had on my website for a few years as a series of jpeg files. So I did... http://www.nixiebunny.com/hpac4/hpac4.pdf These modules are used in the ancient HP 521 to 524 series of counters. Each module handles one digit. They have four vacuum tubes and either a nixie tube or ten neons in a thermometer arrangement. I have a 521C and a 522B counter on hand but they are a bit too slow to handle the 10 MHz of modern times, having a maximum counting frequency of 200 kilocycles per second. -- --David Forbes, Tucson, AZ http://www.cathodecorner.com/ From magnus at rubidium.dyndns.org Sun Feb 1 12:09:14 2009 From: magnus at rubidium.dyndns.org (Magnus Danielson) Date: Sun, 01 Feb 2009 13:09:14 +0100 Subject: [time-nuts] Boatanchor docs: HP AC-4 counter module manual In-Reply-To: References: Message-ID: <498590EA.3070903@rubidium.dyndns.org> David, David Forbes skrev: > O Nuts of Time, > > I was recently asked to make a big PDF file of the HP AC-4 counter > modules manual that I have had on my website for a few years as a > series of jpeg files. So I did... > > http://www.nixiebunny.com/hpac4/hpac4.pdf > > These modules are used in the ancient HP 521 to 524 series of > counters. Each module handles one digit. They have four vacuum tubes > and either a nixie tube or ten neons in a thermometer arrangement. > > I have a 521C and a 522B counter on hand but they are a bit too slow > to handle the 10 MHz of modern times, having a maximum counting > frequency of 200 kilocycles per second. A pair of phantastron setups will divide it into range. Should use up about 3 tubes. Cheers, Magnus From jmiles at pop.net Sun Feb 1 12:50:05 2009 From: jmiles at pop.net (John Miles) Date: Sun, 1 Feb 2009 04:50:05 -0800 Subject: [time-nuts] Lady Heather's Window In-Reply-To: <498498F8.3010608@rubidium.dyndns.org> Message-ID: Here's my quick-and-dirty Windows port (including the modified source code): http://www.ke5fx.com/tbolt/setup.exe (600K) Tested on XP, should be fine on Win9x/2K, no telling with Vista. As with Mark's original program, the default is COM1; you'll need to modify your Windows shortcut to run "heather /x" (without the quotes) for COMx. -- john, KE5FX > -----Original Message----- > From: time-nuts-bounces at febo.com [mailto:time-nuts-bounces at febo.com]On > Behalf Of Magnus Danielson > Sent: Saturday, January 31, 2009 10:31 AM > To: Richard W. Solomon; Discussion of precise time and frequency > measurement > Subject: Re: [time-nuts] Lady Heather Update > > > Richard W. Solomon skrev: > > I tried it on my Compaq 486 running WIN98SE I get a black screen > > with Logoff on it and a couple of other lines that make no sense ? > > > > Who has the Rosetta Stone ?? > > I do not know... British Museum maybe? > > I tried to recompile in my Cygwin environment but with no luck, failed > to compile. > > It seems to be beyond me... I know nothing about Windows... I just need > to run the damn thing for some apps... > > Cheers, > Magnus > > From jmiles at pop.net Sun Feb 1 12:53:27 2009 From: jmiles at pop.net (John Miles) Date: Sun, 1 Feb 2009 04:53:27 -0800 Subject: [time-nuts] Lady Heather's Window In-Reply-To: Message-ID: By the way, I can't run the DOS version here at all, so let me know if anything looks significantly different. -- john, KE5FX > -----Original Message----- > From: John Miles [mailto:jmiles at pop.net] > Sent: Sunday, February 01, 2009 4:50 AM > To: Discussion of precise time and frequency measurement > Subject: Lady Heather's Window > > > Here's my quick-and-dirty Windows port (including the modified > source code): > > http://www.ke5fx.com/tbolt/setup.exe (600K) > > Tested on XP, should be fine on Win9x/2K, no telling with Vista. > > As with Mark's original program, the default is COM1; you'll need > to modify your Windows shortcut to run "heather /x" (without the > quotes) for COMx. > > -- john, KE5FX > > > > -----Original Message----- > > From: time-nuts-bounces at febo.com [mailto:time-nuts-bounces at febo.com]On > > Behalf Of Magnus Danielson > > Sent: Saturday, January 31, 2009 10:31 AM > > To: Richard W. Solomon; Discussion of precise time and frequency > > measurement > > Subject: Re: [time-nuts] Lady Heather Update > > > > > > Richard W. Solomon skrev: > > > I tried it on my Compaq 486 running WIN98SE I get a black screen > > > with Logoff on it and a couple of other lines that make no sense ? > > > > > > Who has the Rosetta Stone ?? > > > > I do not know... British Museum maybe? > > > > I tried to recompile in my Cygwin environment but with no luck, failed > > to compile. > > > > It seems to be beyond me... I know nothing about Windows... I just need > > to run the damn thing for some apps... > > > > Cheers, > > Magnus > > > > From magnus at rubidium.dyndns.org Sun Feb 1 14:00:38 2009 From: magnus at rubidium.dyndns.org (Magnus Danielson) Date: Sun, 01 Feb 2009 15:00:38 +0100 Subject: [time-nuts] Lady Heather's Window In-Reply-To: References: Message-ID: <4985AB06.5090601@rubidium.dyndns.org> John Miles skrev: > Here's my quick-and-dirty Windows port (including the modified source code): > > http://www.ke5fx.com/tbolt/setup.exe (600K) > > Tested on XP, should be fine on Win9x/2K, no telling with Vista. > > As with Mark's original program, the default is COM1; you'll need to modify > your Windows shortcut to run "heather /x" (without the quotes) for COMx. It runs fine functionally, however, the font is minimalistic and I can't read anything written. But the graphs seems to work etc. Also, I can't change the display at run-time, which would be nice. Cheers, Magnus From magnus at rubidium.dyndns.org Sun Feb 1 14:45:39 2009 From: magnus at rubidium.dyndns.org (Magnus Danielson) Date: Sun, 01 Feb 2009 15:45:39 +0100 Subject: [time-nuts] Lady Heather's Window In-Reply-To: <4985AB06.5090601@rubidium.dyndns.org> References: <4985AB06.5090601@rubidium.dyndns.org> Message-ID: <4985B593.3070205@rubidium.dyndns.org> Magnus Danielson skrev: > John Miles skrev: >> Here's my quick-and-dirty Windows port (including the modified source code): >> >> http://www.ke5fx.com/tbolt/setup.exe (600K) >> >> Tested on XP, should be fine on Win9x/2K, no telling with Vista. >> >> As with Mark's original program, the default is COM1; you'll need to modify >> your Windows shortcut to run "heather /x" (without the quotes) for COMx. > > It runs fine functionally, however, the font is minimalistic and I can't > read anything written. But the graphs seems to work etc. Also, I can't > change the display at run-time, which would be nice. /vl does not improve the size of fonts, just the overall windown size. It is notable that it maps to the wrong font as the right and lower part of the text-area is left empty, but the everything seems to work, just below the limit to unreadable. I lack the Windows magic wand capability... Cheers, Magnus From didier at cox.net Sun Feb 1 15:10:39 2009 From: didier at cox.net (Didier) Date: Sun, 1 Feb 2009 09:10:39 -0600 Subject: [time-nuts] Lady Heather's Window, font size In-Reply-To: <4985B593.3070205@rubidium.dyndns.org> References: <4985AB06.5090601@rubidium.dyndns.org> <4985B593.3070205@rubidium.dyndns.org> Message-ID: <0EDF51BAB7D84E268CD0145EFB50BFBE@didierhp> I have found it difficult to find out from within a Windows program how big or small a particular font is rendered. Windows has a function called TextWidth (and TextHeight) that does not work correctly for all fonts. It seems to be video card driver dependant too, as the same fonts look different (and have different size) on different video cards. I wrote a small VB program that you can use to find out if windows is lying for a particular font. It's at http://www.ko4bb.com/VB_Projects/TextWidth/TextWidth.exe This programs sets the size of the colored background to what windows says the font takes. You can see some fonts work well, and some don't. I have found that most of the time, Windows returns a size greater than is actually used. That can be useful if you use these functions by letting you choose fonts that work. Didier > -----Original Message----- > From: time-nuts-bounces at febo.com > [mailto:time-nuts-bounces at febo.com] On Behalf Of Magnus Danielson > Sent: Sunday, February 01, 2009 8:46 AM > To: Discussion of precise time and frequency measurement > Subject: Re: [time-nuts] Lady Heather's Window > > Magnus Danielson skrev: > > John Miles skrev: > >> Here's my quick-and-dirty Windows port (including the > modified source code): > >> > >> http://www.ke5fx.com/tbolt/setup.exe (600K) > >> > >> Tested on XP, should be fine on Win9x/2K, no telling with Vista. > >> > >> As with Mark's original program, the default is COM1; > you'll need to > >> modify your Windows shortcut to run "heather /x" (without > the quotes) for COMx. > > > > It runs fine functionally, however, the font is minimalistic and I > > can't read anything written. But the graphs seems to work > etc. Also, I > > can't change the display at run-time, which would be nice. > > /vl does not improve the size of fonts, just the overall > windown size. > It is notable that it maps to the wrong font as the right and > lower part of the text-area is left empty, but the everything > seems to work, just below the limit to unreadable. > > I lack the Windows magic wand capability... > > Cheers, > Magnus > > _______________________________________________ > time-nuts mailing list -- time-nuts at febo.com To unsubscribe, > go to https://www.febo.com/cgi-bin/mailman/listinfo/time-nuts > and follow the instructions there. From brooke at pacific.net Sun Feb 1 16:16:25 2009 From: brooke at pacific.net (Brooke Clarke) Date: Sun, 01 Feb 2009 08:16:25 -0800 Subject: [time-nuts] Lady Heather's Window, font size In-Reply-To: <0EDF51BAB7D84E268CD0145EFB50BFBE@didierhp> References: <4985AB06.5090601@rubidium.dyndns.org> <4985B593.3070205@rubidium.dyndns.org> <0EDF51BAB7D84E268CD0145EFB50BFBE@didierhp> Message-ID: <4985CAD9.9000803@pacific.net> Hi Didier: I've been working with my PC display adapter and it's related software. In WIN XP when I right click the desktop \ properties \ settings \advanced \ general there is an option for DPI with 96 the default, 120 the option I choose for larger text in general applications, or a custom setting. Also regarding fonts. There are fixed fonts, like courier, and True Type variable fonts that scale, like Times New Roman. Have Fun, Brooke Clarke http://www.prc68.com Didier wrote: > I have found it difficult to find out from within a Windows program how big > or small a particular font is rendered. Windows has a function called > TextWidth (and TextHeight) that does not work correctly for all fonts. It > seems to be video card driver dependant too, as the same fonts look > different (and have different size) on different video cards. > > I wrote a small VB program that you can use to find out if windows is lying > for a particular font. > > It's at http://www.ko4bb.com/VB_Projects/TextWidth/TextWidth.exe > > This programs sets the size of the colored background to what windows says > the font takes. You can see some fonts work well, and some don't. I have > found that most of the time, Windows returns a size greater than is actually > used. > > That can be useful if you use these functions by letting you choose fonts > that work. > > Didier > >> -----Original Message----- >> From: time-nuts-bounces at febo.com >> [mailto:time-nuts-bounces at febo.com] On Behalf Of Magnus Danielson >> Sent: Sunday, February 01, 2009 8:46 AM >> To: Discussion of precise time and frequency measurement >> Subject: Re: [time-nuts] Lady Heather's Window >> >> Magnus Danielson skrev: >>> John Miles skrev: >>>> Here's my quick-and-dirty Windows port (including the >> modified source code): >>>> http://www.ke5fx.com/tbolt/setup.exe (600K) >>>> >>>> Tested on XP, should be fine on Win9x/2K, no telling with Vista. >>>> >>>> As with Mark's original program, the default is COM1; >> you'll need to >>>> modify your Windows shortcut to run "heather /x" (without >> the quotes) for COMx. >>> It runs fine functionally, however, the font is minimalistic and I >>> can't read anything written. But the graphs seems to work >> etc. Also, I >>> can't change the display at run-time, which would be nice. >> /vl does not improve the size of fonts, just the overall >> windown size. >> It is notable that it maps to the wrong font as the right and >> lower part of the text-area is left empty, but the everything >> seems to work, just below the limit to unreadable. >> >> I lack the Windows magic wand capability... >> >> Cheers, >> Magnus >> >> _______________________________________________ >> time-nuts mailing list -- time-nuts at febo.com To unsubscribe, >> go to https://www.febo.com/cgi-bin/mailman/listinfo/time-nuts >> and follow the instructions there. > > > _______________________________________________ > time-nuts mailing list -- time-nuts at febo.com > To unsubscribe, go to https://www.febo.com/cgi-bin/mailman/listinfo/time-nuts > and follow the instructions there. > > From didier at cox.net Sun Feb 1 16:47:34 2009 From: didier at cox.net (Didier) Date: Sun, 1 Feb 2009 10:47:34 -0600 Subject: [time-nuts] Lady Heather's Window, font size In-Reply-To: <4985CAD9.9000803@pacific.net> References: <4985AB06.5090601@rubidium.dyndns.org> <4985B593.3070205@rubidium.dyndns.org><0EDF51BAB7D84E268CD0145EFB50BFBE@didierhp> <4985CAD9.9000803@pacific.net> Message-ID: <6BE0D4B640EE4FDB851C4694EE111A77@didierhp> Hi Brooke, Windows' error in the response to TextWidth is not a function of fixed versus proportional spacing, bitmap versus scalable font. I have not found a hard rule. Simpler fonts tend to be more accurately reported, but that's not universal. Experiment with my little program on different computers, you will be surprised of the differences. This came about because I tried to emulate a TTY screen and make the window exactly 80 characters wide, while offering the capability to change the size of the window. I have not been able to do what I wanted and at the end, I only selected one font which worked consistently over a few sizes, and I simply offer the user a few selections rather than let the user scale the window continuously. Didier > -----Original Message----- > From: time-nuts-bounces at febo.com > [mailto:time-nuts-bounces at febo.com] On Behalf Of Brooke Clarke > Sent: Sunday, February 01, 2009 10:16 AM > To: Discussion of precise time and frequency measurement > Subject: Re: [time-nuts] Lady Heather's Window, font size > > Hi Didier: > > I've been working with my PC display adapter and it's related > software. In WIN XP when I right click the desktop \ > properties \ settings \advanced \ general there is an option > for DPI with 96 the default, 120 the option I choose for > larger text in general applications, or a custom setting. > > Also regarding fonts. There are fixed fonts, like courier, > and True Type variable fonts that scale, like Times New Roman. > > Have Fun, > > Brooke Clarke > http://www.prc68.com > > Didier wrote: > > I have found it difficult to find out from within a Windows program > > how big or small a particular font is rendered. Windows has > a function > > called TextWidth (and TextHeight) that does not work > correctly for all > > fonts. It seems to be video card driver dependant too, as the same > > fonts look different (and have different size) on different > video cards. > > > > I wrote a small VB program that you can use to find out if > windows is > > lying for a particular font. > > > > It's at http://www.ko4bb.com/VB_Projects/TextWidth/TextWidth.exe > > > > This programs sets the size of the colored background to > what windows > > says the font takes. You can see some fonts work well, and > some don't. > > I have found that most of the time, Windows returns a size greater > > than is actually used. > > > > That can be useful if you use these functions by letting you choose > > fonts that work. > > > > Didier > > > >> -----Original Message----- > >> From: time-nuts-bounces at febo.com > >> [mailto:time-nuts-bounces at febo.com] On Behalf Of Magnus Danielson > >> Sent: Sunday, February 01, 2009 8:46 AM > >> To: Discussion of precise time and frequency measurement > >> Subject: Re: [time-nuts] Lady Heather's Window > >> > >> Magnus Danielson skrev: > >>> John Miles skrev: > >>>> Here's my quick-and-dirty Windows port (including the > >> modified source code): > >>>> http://www.ke5fx.com/tbolt/setup.exe (600K) > >>>> > >>>> Tested on XP, should be fine on Win9x/2K, no telling with Vista. > >>>> > >>>> As with Mark's original program, the default is COM1; > >> you'll need to > >>>> modify your Windows shortcut to run "heather /x" (without > >> the quotes) for COMx. > >>> It runs fine functionally, however, the font is > minimalistic and I > >>> can't read anything written. But the graphs seems to work > >> etc. Also, I > >>> can't change the display at run-time, which would be nice. > >> /vl does not improve the size of fonts, just the overall windown > >> size. > >> It is notable that it maps to the wrong font as the right > and lower > >> part of the text-area is left empty, but the everything seems to > >> work, just below the limit to unreadable. > >> > >> I lack the Windows magic wand capability... > >> > >> Cheers, > >> Magnus > >> > >> _______________________________________________ > >> time-nuts mailing list -- time-nuts at febo.com To unsubscribe, go to > >> https://www.febo.com/cgi-bin/mailman/listinfo/time-nuts > >> and follow the instructions there. > > > > > > _______________________________________________ > > time-nuts mailing list -- time-nuts at febo.com To unsubscribe, go to > > https://www.febo.com/cgi-bin/mailman/listinfo/time-nuts > > and follow the instructions there. > > > > > > > _______________________________________________ > time-nuts mailing list -- time-nuts at febo.com To unsubscribe, > go to https://www.febo.com/cgi-bin/mailman/listinfo/time-nuts > and follow the instructions there. From had at to-way.com Sun Feb 1 17:25:12 2009 From: had at to-way.com (Had) Date: Sun, 01 Feb 2009 09:25:12 -0800 Subject: [time-nuts] Lady Heather's Window In-Reply-To: References: Message-ID: <20090201172515.BACB9CBCD3E@mail-in02.adhost.com> Morning John, Just loaded up your build of Lady Heather on a XP Pro (SP3) machine. Runs just fine and the graphing is readable but the upper text portion is so tiny that it is un readable. Tried Brooks trick of changing the default DPI to 125 but it has no effect on the LH program. Any thoughts as to what I can do so I can read the text portions. Thanks Had K7MLR At 04:53 AM 2/1/2009, you wrote: >By the way, I can't run the DOS version here at all, so let me know if >anything looks significantly different. > >-- john, KE5FX > > > -----Original Message----- > > From: John Miles [mailto:jmiles at pop.net] > > Sent: Sunday, February 01, 2009 4:50 AM > > To: Discussion of precise time and frequency measurement > > Subject: Lady Heather's Window > > > > > > Here's my quick-and-dirty Windows port (including the modified > > source code): > > > > http://www.ke5fx.com/tbolt/setup.exe (600K) > > > > Tested on XP, should be fine on Win9x/2K, no telling with Vista. > > > > As with Mark's original program, the default is COM1; you'll need > > to modify your Windows shortcut to run "heather /x" (without the > > quotes) for COMx. > > > > -- john, KE5FX From magnus at rubidium.dyndns.org Sun Feb 1 17:27:18 2009 From: magnus at rubidium.dyndns.org (Magnus Danielson) Date: Sun, 01 Feb 2009 18:27:18 +0100 Subject: [time-nuts] Lady Heather's Window, font size In-Reply-To: <0EDF51BAB7D84E268CD0145EFB50BFBE@didierhp> References: <4985AB06.5090601@rubidium.dyndns.org> <4985B593.3070205@rubidium.dyndns.org> <0EDF51BAB7D84E268CD0145EFB50BFBE@didierhp> Message-ID: <4985DB76.4010006@rubidium.dyndns.org> Didier skrev: > I have found it difficult to find out from within a Windows program how big > or small a particular font is rendered. Windows has a function called > TextWidth (and TextHeight) that does not work correctly for all fonts. It > seems to be video card driver dependant too, as the same fonts look > different (and have different size) on different video cards. The trouble is that the windows port of Lady Heathers choose "system default" which is not the system default one sets in the Display Configuration. It seems like it picks Terminal 4 point. It needs to at least choose size... but preferably also choose font. > I wrote a small VB program that you can use to find out if windows is lying > for a particular font. > > It's at http://www.ko4bb.com/VB_Projects/TextWidth/TextWidth.exe > > This programs sets the size of the colored background to what windows says > the font takes. You can see some fonts work well, and some don't. I have > found that most of the time, Windows returns a size greater than is actually > used. > > That can be useful if you use these functions by letting you choose fonts > that work. Well, it is useful as such, but it seems like the graphic on my display is not obstructed from incorrectly given size, it simply chooses the wrong size to begin with.... Cheers, Magnus From w1ksz at earthlink.net Sun Feb 1 17:51:47 2009 From: w1ksz at earthlink.net (Richard W. Solomon) Date: Sun, 1 Feb 2009 10:51:47 -0700 (GMT-07:00) Subject: [time-nuts] Lady Heather's Window Message-ID: <15224908.1233510707474.JavaMail.root@mswamui-chipeau.atl.sa.earthlink.net> I have the program running on an old Compaq 486 Laptop with a WIN98 OS. I see the data, but I cannot seem to get the graphical display. Also the Space Bar has no effect. How does one get the graph ?? 73, Dick, W1KSZ -----Original Message----- >From: Magnus Danielson >Sent: Feb 1, 2009 7:00 AM >To: Discussion of precise time and frequency measurement >Subject: Re: [time-nuts] Lady Heather's Window > >John Miles skrev: >> Here's my quick-and-dirty Windows port (including the modified source code): >> >> http://www.ke5fx.com/tbolt/setup.exe (600K) >> >> Tested on XP, should be fine on Win9x/2K, no telling with Vista. >> >> As with Mark's original program, the default is COM1; you'll need to modify >> your Windows shortcut to run "heather /x" (without the quotes) for COMx. > >It runs fine functionally, however, the font is minimalistic and I can't >read anything written. But the graphs seems to work etc. Also, I can't >change the display at run-time, which would be nice. > >Cheers, >Magnus > >_______________________________________________ >time-nuts mailing list -- time-nuts at febo.com >To unsubscribe, go to https://www.febo.com/cgi-bin/mailman/listinfo/time-nuts >and follow the instructions there. From pe1rks at xs4all.nl Sun Feb 1 18:31:00 2009 From: pe1rks at xs4all.nl (S. Nestra) Date: Sun, 01 Feb 2009 19:31:00 +0100 Subject: [time-nuts] Lady Heather's Window In-Reply-To: References: Message-ID: <4985EA64.3050103@xs4all.nl> I had the same problem of the ultra small text under XP. One of you already tried to enlarge the window with /vl. I have found that the opposite of this works: /vs. So if you use "heather /vs" you get the big letters! Works for me, Stijn PE1RKS John Miles schreef: > Here's my quick-and-dirty Windows port (including the modified source code): > > http://www.ke5fx.com/tbolt/setup.exe (600K) > > Tested on XP, should be fine on Win9x/2K, no telling with Vista. > > As with Mark's original program, the default is COM1; you'll need to modify > your Windows shortcut to run "heather /x" (without the quotes) for COMx. > > -- john, KE5FX > > >> -----Original Message----- >> From: time-nuts-bounces at febo.com [mailto:time-nuts-bounces at febo.com]On >> Behalf Of Magnus Danielson >> Sent: Saturday, January 31, 2009 10:31 AM >> To: Richard W. Solomon; Discussion of precise time and frequency >> measurement >> Subject: Re: [time-nuts] Lady Heather Update >> >> >> Richard W. Solomon skrev: >>> I tried it on my Compaq 486 running WIN98SE I get a black screen >>> with Logoff on it and a couple of other lines that make no sense ? >>> >>> Who has the Rosetta Stone ?? >> I do not know... British Museum maybe? >> >> I tried to recompile in my Cygwin environment but with no luck, failed >> to compile. >> >> It seems to be beyond me... I know nothing about Windows... I just need >> to run the damn thing for some apps... >> >> Cheers, >> Magnus >> >> > > > _______________________________________________ > time-nuts mailing list -- time-nuts at febo.com > To unsubscribe, go to https://www.febo.com/cgi-bin/mailman/listinfo/time-nuts > and follow the instructions there. > -- Real radios have motors. From had at to-way.com Sun Feb 1 18:47:21 2009 From: had at to-way.com (Had) Date: Sun, 01 Feb 2009 10:47:21 -0800 Subject: [time-nuts] Lady Heather's Window In-Reply-To: <4985EA64.3050103@xs4all.nl> References: <4985EA64.3050103@xs4all.nl> Message-ID: <20090201184725.A5AE2CBCD27@mail-in02.adhost.com> Stijn, Thank you, that works for me also. 73 Had K7MLR At 10:31 AM 2/1/2009, you wrote: >I had the same problem of the ultra small text under XP. One of you >already tried to enlarge the window with /vl. I have found that the >opposite of this works: /vs. So if you use "heather /vs" you get the big >letters! > >Works for me, > >Stijn PE1RKS From magnus at rubidium.dyndns.org Sun Feb 1 19:03:24 2009 From: magnus at rubidium.dyndns.org (Magnus Danielson) Date: Sun, 01 Feb 2009 20:03:24 +0100 Subject: [time-nuts] Lady Heather's Window In-Reply-To: <4985EA64.3050103@xs4all.nl> References: <4985EA64.3050103@xs4all.nl> Message-ID: <4985F1FC.10904@rubidium.dyndns.org> Stijn, S. Nestra skrev: > I had the same problem of the ultra small text under XP. One of you > already tried to enlarge the window with /vl. I have found that the > opposite of this works: /vs. So if you use "heather /vs" you get the big > letters! > > Works for me, YES!!! It works for me too! I get a larger window too! Obviously there is something wrong with that part of the code, maybe Windows just scales the window size down to fit the screen. Great... now it is useful to me! :D Cheers, Magnus > Stijn PE1RKS > > John Miles schreef: >> Here's my quick-and-dirty Windows port (including the modified source code): >> >> http://www.ke5fx.com/tbolt/setup.exe (600K) >> >> Tested on XP, should be fine on Win9x/2K, no telling with Vista. >> >> As with Mark's original program, the default is COM1; you'll need to modify >> your Windows shortcut to run "heather /x" (without the quotes) for COMx. >> >> -- john, KE5FX >> >> >>> -----Original Message----- >>> From: time-nuts-bounces at febo.com [mailto:time-nuts-bounces at febo.com]On >>> Behalf Of Magnus Danielson >>> Sent: Saturday, January 31, 2009 10:31 AM >>> To: Richard W. Solomon; Discussion of precise time and frequency >>> measurement >>> Subject: Re: [time-nuts] Lady Heather Update >>> >>> >>> Richard W. Solomon skrev: >>>> I tried it on my Compaq 486 running WIN98SE I get a black screen >>>> with Logoff on it and a couple of other lines that make no sense ? >>>> >>>> Who has the Rosetta Stone ?? >>> I do not know... British Museum maybe? >>> >>> I tried to recompile in my Cygwin environment but with no luck, failed >>> to compile. >>> >>> It seems to be beyond me... I know nothing about Windows... I just need >>> to run the damn thing for some apps... >>> >>> Cheers, >>> Magnus >>> >>> >> >> _______________________________________________ >> time-nuts mailing list -- time-nuts at febo.com >> To unsubscribe, go to https://www.febo.com/cgi-bin/mailman/listinfo/time-nuts >> and follow the instructions there. >> > From wje at quackers.net Sun Feb 1 19:07:32 2009 From: wje at quackers.net (wje) Date: Sun, 01 Feb 2009 14:07:32 -0500 Subject: [time-nuts] Some pics of the Datron 491x DC reference cell Message-ID: <4985F2F4.8070807@quackers.net> I've uploaded some photos of the Datron 10 volt cell and reference board, as requested by a few members. I've kept them full-resolution so you can see fine detail. But, that means they are about 3 mb each. The schematics for the cell board I previously posted are also still on the FTP site. DatronCell.jpg is an image of the entire cell board. The reference board is piggybacked on it. Note the thermal shield around the LTZ1000, and the completely shielded sections on the board. Those contain the digital circuitry. The cell board itself goes into a shielded enclosure along with the other three cells and the output buffer amp. DatronLTZCloseup.jpg is exactly that, a closeup of the LTZ with thermal shield removed. The interesting detail here is the mounting. Note the cutouts on the board to provide thermal isolation. This board is also very thin; I didn't measure it, but it's on the order of 0.5 - 1 mm thick. DatronRefCloseup.jpg is a closeup image of the rest of the reference board. The ref board is permanently soldered in place on the cell board, so I can't show the back of it. The ftp login is: ftp ftp.quackers.net login: ftp at quackers.net password: (none required) Or, directly from a browser. I know this works for Firefox, I haven't tried it with anything else: ftp://ftp at quackers.net:ftp at ftp.quackers.net/ Some of you are also waiting for the schematic for the LTZ board itself; I'm recreating it now (since I lost it) and will post when ready. -- Bill Ezell ---------- They said 'Windows or better' so I used Linux. From holrum at hotmail.com Sun Feb 1 21:29:55 2009 From: holrum at hotmail.com (Mark Sims) Date: Sun, 1 Feb 2009 21:29:55 +0000 Subject: [time-nuts] Some pics of the Datron 491x DC reference cell In-Reply-To: References: Message-ID: I tried to access the photos... got to an empty pub directory... Cutouts in a circuit board around a reference chip are usually more to relieve mechanical stress from the chip package caused by the PC board expanding/contracting/warping. This stress can affect the refereence voltage more than you might expect. Malone does this on his little Xicor X6008C voltage reference boards. (BTW that chip has been discontinued). That is why I prefer his boards over the other one that is available on Ebay that does not cut out around the ref chip. If you can live with that, you can get a packaged 5V / 0.01% reference for $14.50 shipped... see Ebay item 250366551770... the guy that builds them is closing them out... and he ships worldwide. _________________________________________________________________ Windows Live? Hotmail??more than just e-mail. http://windowslive.com/howitworks?ocid=TXT_TAGLM_WL_t2_hm_justgotbetter_howitworks_012009 From w1ksz at earthlink.net Sun Feb 1 21:39:47 2009 From: w1ksz at earthlink.net (Richard W. Solomon) Date: Sun, 1 Feb 2009 16:39:47 -0500 (EST) Subject: [time-nuts] Lady Heather Question Message-ID: <29501338.1233524387396.JavaMail.root@mswamui-chipeau.atl.sa.earthlink.net> Having failed to get the program working properly on an old 486 Laptop, I installed it on a more modern machine running XP. When I try to open the program I get an error message about Com Port Not Available ?? How does one select the COM Port ? Thanks, Dick, W1KSZ From bruce.griffiths at xtra.co.nz Sun Feb 1 21:50:22 2009 From: bruce.griffiths at xtra.co.nz (Bruce Griffiths) Date: Mon, 02 Feb 2009 10:50:22 +1300 Subject: [time-nuts] Some pics of the Datron 491x DC reference cell In-Reply-To: References: Message-ID: <4986191E.4010708@xtra.co.nz> Mark Sims wrote: > I tried to access the photos... got to an empty pub directory... > > Cutouts in a circuit board around a reference chip are usually more to relieve mechanical stress from the chip package caused by the PC board expanding/contracting/warping. This stress can affect the refereence voltage more than you might expect. Malone does this on his little Xicor X6008C voltage reference boards. (BTW that chip has been discontinued). That is why I prefer his boards over the other one that is available on Ebay that does not cut out around the ref chip. If you can live with that, you can get a packaged 5V / 0.01% reference for $14.50 shipped... see Ebay item 250366551770... the guy that builds them is closing them out... and he ships worldwide. > _________________________________________________________________ > Windows Live? Hotmail??more than just e-mail. > http://windowslive.com/howitworks?ocid=TXT_TAGLM_WL_t2_hm_justgotbetter_howitworks_012009 > _______________________________________________ > time-nuts mailing list -- time-nuts at febo.com > To unsubscribe, go to https://www.febo.com/cgi-bin/mailman/listinfo/time-nuts > and follow the instructions there. > > Second method works fine (Linux machine) with Opera, Firefox, Epithany, Seamonkey however only Epithany and Opera save the jpeg files correctly. Bruce From drfrank.stellmach at freenet.de Sun Feb 1 22:11:30 2009 From: drfrank.stellmach at freenet.de (Dr. Frank Stellmach) Date: Sun, 01 Feb 2009 23:11:30 +0100 Subject: [time-nuts] Some pics of the Datron 491x DC reference cell Message-ID: <49861E12.8030704@freenet.de> Bill, thank you very much for the schematics and the photographs! Using XP / firefox, I was able to retrieve the files with the given special link, and also very easily with the FileZilla Client. The most interesting part, i.e. the precision resistor circuitry, obviously resides on the bottom side of the ref board. Would be great to get a photo also from that side, as Datron made some efforts to get rid of strain and heat conduction effects. Best regards - Frank Stellmach From bruce.griffiths at xtra.co.nz Sun Feb 1 22:20:11 2009 From: bruce.griffiths at xtra.co.nz (Bruce Griffiths) Date: Mon, 02 Feb 2009 11:20:11 +1300 Subject: [time-nuts] Some pics of the Datron 491x DC reference cell In-Reply-To: <49861E12.8030704@freenet.de> References: <49861E12.8030704@freenet.de> Message-ID: <4986201B.50400@xtra.co.nz> Dr. Frank Stellmach wrote: > Bill, > > thank you very much for the schematics and the photographs! > Using XP / firefox, I was able to retrieve the files with the given > special link, and also very easily with the FileZilla Client. > > The most interesting part, i.e. the precision resistor circuitry, > obviously resides on the bottom side of the ref board. > > Would be great to get a photo also from that side, as Datron made some > efforts to get rid of strain and heat conduction effects. > > Best regards - Frank Stellmach > > Frank The precision resistors are more likely to be inside the metal can labeled vishay. Bruce From magnus at rubidium.dyndns.org Sun Feb 1 22:30:42 2009 From: magnus at rubidium.dyndns.org (Magnus Danielson) Date: Sun, 01 Feb 2009 23:30:42 +0100 Subject: [time-nuts] Lady Heather Question In-Reply-To: <29501338.1233524387396.JavaMail.root@mswamui-chipeau.atl.sa.earthlink.net> References: <29501338.1233524387396.JavaMail.root@mswamui-chipeau.atl.sa.earthlink.net> Message-ID: <49862292.4030304@rubidium.dyndns.org> Richard W. Solomon skrev: > Having failed to get the program working properly on an old 486 Laptop, > I installed it on a more modern machine running XP. When I try to open > the program I get an error message about Com Port Not Available ?? > > How does one select the COM Port ? /2 should get it to run on COM2... however... due to my previous experience check your Hardware Driver settings and FORCE your com-port to COM1. That worked for me... but maybe the new Windows-port works better. Cheers, Magnus From drfrank.stellmach at freenet.de Sun Feb 1 22:39:27 2009 From: drfrank.stellmach at freenet.de (Dr. Frank Stellmach) Date: Sun, 01 Feb 2009 23:39:27 +0100 Subject: [time-nuts] Some pics of the Datron 491x DC reference cell Message-ID: <4986249F.7080402@freenet.de> Bruce, good hint, I think you are right.. Then it's a custom specific R array, metal foil technolgy (S105 or similar) from Vishay, perhaps oil filled. Makes it very difficult to read the circuitry and design a replacement with discrete resistors. Frank From bruce.griffiths at xtra.co.nz Sun Feb 1 22:46:48 2009 From: bruce.griffiths at xtra.co.nz (Bruce Griffiths) Date: Mon, 02 Feb 2009 11:46:48 +1300 Subject: [time-nuts] Some pics of the Datron 491x DC reference cell In-Reply-To: <4986249F.7080402@freenet.de> References: <4986249F.7080402@freenet.de> Message-ID: <49862658.8040900@xtra.co.nz> Dr. Frank Stellmach wrote: > Bruce, > > good hint, I think you are right.. > > Then it's a custom specific R array, metal foil technolgy (S105 or > similar) from Vishay, perhaps oil filled. > > Makes it very difficult to read the circuitry and design a replacement > with discrete resistors. > > Frank > > > Frank The core circuit probably doesn't differ too much from the positive reference circuit shown on the LTZ1000 datasheet. It certainly wont be identical to the negative reference shown on that datasheet which has the heater pins reversed (as shown it will cause a large current to flow in the parasitic shunt diode connected across the heater resistor). NB use the latest datasheet which has a corrected schematic showing the location and polarity of the parasitic diodes. Bruce From drfrank.stellmach at freenet.de Sun Feb 1 23:07:00 2009 From: drfrank.stellmach at freenet.de (Dr. Frank Stellmach) Date: Mon, 02 Feb 2009 00:07:00 +0100 Subject: [time-nuts] Some pics of the Datron 491x DC reference cell Message-ID: <49862B14.8090209@freenet.de> > NB use the latest datasheet which has a corrected schematic showing the > location and polarity of the parasitic diodes. > Bruce Thank you, I already got that... When I used the first datasheet, I reversed the heater as there was no hint, just to discover those hidden parasitic diodes. I obviously did not damage the device (hope so), but sent that info to LT, and then they did the correction. Frank From bruce.griffiths at xtra.co.nz Sun Feb 1 23:21:30 2009 From: bruce.griffiths at xtra.co.nz (Bruce Griffiths) Date: Mon, 02 Feb 2009 12:21:30 +1300 Subject: [time-nuts] Some pics of the Datron 491x DC reference cell In-Reply-To: <49862B14.8090209@freenet.de> References: <49862B14.8090209@freenet.de> Message-ID: <49862E7A.706@xtra.co.nz> Dr. Frank Stellmach wrote: > > NB use the latest datasheet which has a corrected schematic showing the > > location and polarity of the parasitic diodes. > > >> Bruce >> > > Thank you, I already got that... > > When I used the first datasheet, I reversed the heater as there was no hint, just to discover those hidden parasitic diodes. > I obviously did not damage the device (hope so), but sent that info to LT, and then they did the correction. > > Frank > > > Frank Yes but they didn't correct the negative reference circuit. I notified them about that - maybe that will be corrected in the next revision. I've located my Vishay brochure on resistor networks from the 70's. The TO-5 can resistor networks used 50 mil x 50 mil (50mW) and or 150 mil x 50 mil (100mW) resistor chips wire bonded to the header pins. Either 8 pin(2 different pin circle diameter options) or 10 pin variants were available. Other than the resistor chip specifications (TCR, pwer dissipation, etc) there's not much else. Bruce From jmiles at pop.net Sun Feb 1 23:36:59 2009 From: jmiles at pop.net (John Miles) Date: Sun, 1 Feb 2009 15:36:59 -0800 Subject: [time-nuts] Lady Heather's Window In-Reply-To: <4985F1FC.10904@rubidium.dyndns.org> Message-ID: Ah, I should have mentioned that. You need to run it on a desktop that's at least one 'standard size' larger than the window, or my graphics library will halve the size of the window, which renders the text unreadable. The text is still pretty cramped when running normally, but it's not a standard Windows font -- it's a hardwired 8x8 raster font that's part of an ancient platform-independent graphics library I wrote back when the Win95 Chicago beta came out. So unfortunately there are no real options for improvement there without rewriting a bunch of code to use actual Windows fonts. One thing that would help is running it in fullscreen mode, which I forgot to enable. Grab the newest build at http://www.ke5fx.com/tbolt/setup.exe and add an -f option to your Windows shortcut if you want it to start in fullscreen mode (or just use alt-Enter or the maximize button to toggle fullscreen mode after running.) Especially if you use it with /vs, the font will appear much larger than it does in a window. -- john, KE5FX > -----Original Message----- > From: time-nuts-bounces at febo.com [mailto:time-nuts-bounces at febo.com]On > Behalf Of Magnus Danielson > Sent: Sunday, February 01, 2009 11:03 AM > To: Discussion of precise time and frequency measurement > Subject: Re: [time-nuts] Lady Heather's Window > > > Stijn, > > S. Nestra skrev: > > I had the same problem of the ultra small text under XP. One of you > > already tried to enlarge the window with /vl. I have found that the > > opposite of this works: /vs. So if you use "heather /vs" you > get the big > > letters! > > > > Works for me, > > YES!!! It works for me too! I get a larger window too! > > Obviously there is something wrong with that part of the code, maybe > Windows just scales the window size down to fit the screen. > > Great... now it is useful to me! :D > > Cheers, > Magnus > > > Stijn PE1RKS > > > > John Miles schreef: > >> Here's my quick-and-dirty Windows port (including the modified > source code): > >> > >> http://www.ke5fx.com/tbolt/setup.exe (600K) > >> > >> Tested on XP, should be fine on Win9x/2K, no telling with Vista. > >> > >> As with Mark's original program, the default is COM1; you'll > need to modify > >> your Windows shortcut to run "heather /x" (without the quotes) > for COMx. > >> > >> -- john, KE5FX > >> > >> > >>> -----Original Message----- > >>> From: time-nuts-bounces at febo.com [mailto:time-nuts-bounces at febo.com]On > >>> Behalf Of Magnus Danielson > >>> Sent: Saturday, January 31, 2009 10:31 AM > >>> To: Richard W. Solomon; Discussion of precise time and frequency > >>> measurement > >>> Subject: Re: [time-nuts] Lady Heather Update > >>> > >>> > >>> Richard W. Solomon skrev: > >>>> I tried it on my Compaq 486 running WIN98SE I get a black screen > >>>> with Logoff on it and a couple of other lines that make no sense ? > >>>> > >>>> Who has the Rosetta Stone ?? > >>> I do not know... British Museum maybe? > >>> > >>> I tried to recompile in my Cygwin environment but with no luck, failed > >>> to compile. > >>> > >>> It seems to be beyond me... I know nothing about Windows... I > just need > >>> to run the damn thing for some apps... > >>> > >>> Cheers, > >>> Magnus > >>> > >>> > >> > >> _______________________________________________ > >> time-nuts mailing list -- time-nuts at febo.com > >> To unsubscribe, go to https://www.febo.com/cgi-bin/mailman/listinfo/time-nuts >> and follow the instructions there. >> > _______________________________________________ time-nuts mailing list -- time-nuts at febo.com To unsubscribe, go to https://www.febo.com/cgi-bin/mailman/listinfo/time-nuts and follow the instructions there. From jmiles at pop.net Sun Feb 1 23:40:47 2009 From: jmiles at pop.net (John Miles) Date: Sun, 1 Feb 2009 15:40:47 -0800 Subject: [time-nuts] Lady Heather Question In-Reply-To: <49862292.4030304@rubidium.dyndns.org> Message-ID: The DOS version had a bug that caused it to ignore the /2 altogether (it hardwired the COM port back to 1 after parsing the command-line arts). That's fixed in the Windows build. Mark: the Windows source code should still be able to compile for DOS at your end, if you change the #define back. If you'll switch to my copy (and switch to .cpp compilation) it would be easy to keep them in sync. -- john, KE5FX > -----Original Message----- > From: time-nuts-bounces at febo.com [mailto:time-nuts-bounces at febo.com]On > Behalf Of Magnus Danielson > Sent: Sunday, February 01, 2009 2:31 PM > To: Richard W. Solomon; Discussion of precise time and frequency > measurement > Subject: Re: [time-nuts] Lady Heather Question > > > Richard W. Solomon skrev: > > Having failed to get the program working properly on an old 486 Laptop, > > I installed it on a more modern machine running XP. When I try to open > > the program I get an error message about Com Port Not Available ?? > > > > How does one select the COM Port ? > > /2 should get it to run on COM2... however... due to my previous > experience check your Hardware Driver settings and FORCE your com-port > to COM1. That worked for me... but maybe the new Windows-port > works better. > > Cheers, > Magnus > > From magnus at rubidium.dyndns.org Mon Feb 2 00:17:41 2009 From: magnus at rubidium.dyndns.org (Magnus Danielson) Date: Mon, 02 Feb 2009 01:17:41 +0100 Subject: [time-nuts] Lady Heather Question In-Reply-To: References: Message-ID: <49863BA5.70808@rubidium.dyndns.org> John Miles skrev: > The DOS version had a bug that caused it to ignore the /2 altogether (it > hardwired the COM port back to 1 after parsing the command-line arts). > That's fixed in the Windows build. Do you use Windows serial line routines or maintain Marks more direct access? For a Windows port the OS serial line routines should be the right thing. We should avoid having to force the configuration like I had to do initially. Cheers, Magnus From jmiles at pop.net Mon Feb 2 00:33:16 2009 From: jmiles at pop.net (John Miles) Date: Sun, 1 Feb 2009 16:33:16 -0800 Subject: [time-nuts] Lady Heather Question In-Reply-To: <49863BA5.70808@rubidium.dyndns.org> Message-ID: I use the normal Win32 device-communications API (via CreateFile/ReadFile). Same serial-interface module that I wrote to talk to Prologix GPIB-USB dongles. It should support any COM port number you give it. The graph ends up looking a lot neater with 10-second updates than with 1-second updates -- I wonder if that should be the default. The PPS and ref traces in particular seem to break up a lot at 1-second timescales. -- john, KE5FX > -----Original Message----- > From: time-nuts-bounces at febo.com [mailto:time-nuts-bounces at febo.com]On > Behalf Of Magnus Danielson > Sent: Sunday, February 01, 2009 4:18 PM > To: Discussion of precise time and frequency measurement > Subject: Re: [time-nuts] Lady Heather Question > > > John Miles skrev: > > The DOS version had a bug that caused it to ignore the /2 altogether (it > > hardwired the COM port back to 1 after parsing the command-line args). > > That's fixed in the Windows build. > > Do you use Windows serial line routines or maintain Marks more direct > access? For a Windows port the OS serial line routines should be the > right thing. We should avoid having to force the configuration like I > had to do initially. > > Cheers, > Magnus > > _______________________________________________ > time-nuts mailing list -- time-nuts at febo.com > To unsubscribe, go to > https://www.febo.com/cgi-bin/mailman/listinfo/time-nuts > and follow the instructions there. > From magnus at rubidium.dyndns.org Mon Feb 2 01:04:11 2009 From: magnus at rubidium.dyndns.org (Magnus Danielson) Date: Mon, 02 Feb 2009 02:04:11 +0100 Subject: [time-nuts] Lady Heather's Window In-Reply-To: References: Message-ID: <4986468B.6000903@rubidium.dyndns.org> John Miles skrev: > Ah, I should have mentioned that. You need to run it on a desktop that's at > least one 'standard size' larger than the window, or my graphics library > will halve the size of the window, which renders the text unreadable. I think a more flexible fashion of setting size would be a good thing. The preferred fasion is always being able to do standard resize by dragging the edge, but saying x*y size at command line kind of works. > The text is still pretty cramped when running normally, but it's not a > standard Windows font -- it's a hardwired 8x8 raster font that's part of an > ancient platform-independent graphics library I wrote back when the Win95 > Chicago beta came out. So unfortunately there are no real options for > improvement there without rewriting a bunch of code to use actual Windows > fonts. Avoiding the shrinking should be the main thing to do then, in this case it hurts more than it helps. Oh... The oscillator ADEV positions at 2000 s and 5000 s is displaced over the 500s and 1000s respectively, 2 digits too much to the left and halfway up on the mentioned values, seems like 16 left and 16 up if the font is 8 x 8 as you say. > One thing that would help is running it in fullscreen mode, which I forgot > to enable. Grab the newest build at http://www.ke5fx.com/tbolt/setup.exe > and add an -f option to your Windows shortcut if you want it to start in > fullscreen mode (or just use alt-Enter or the maximize button to toggle > fullscreen mode after running.) Especially if you use it with /vs, the font > will appear much larger than it does in a window. Fullscreen works. Usefull. Dropping shrinking and making size more flexible would be beneficial. I haven't checked if the fullscreen version has the position bug mentioned above... I will see in the morning. Cheers, Magnus From jmiles at pop.net Mon Feb 2 01:40:03 2009 From: jmiles at pop.net (John Miles) Date: Sun, 1 Feb 2009 17:40:03 -0800 Subject: [time-nuts] Lady Heather's Window In-Reply-To: <4986468B.6000903@rubidium.dyndns.org> Message-ID: > I think a more flexible fashion of setting size would be a good thing. > The preferred fasion is always being able to do standard resize by > dragging the edge, but saying x*y size at command line kind of works. Yep, the code could support arbitrary x*y sizes, but he has several groups of related parameters that are tied to the discrete size options -- and also, I can only go into fullscreen mode in standard VGA/EVGA resolutions. So I probably won't put a lot of work into that. Left as an exercise to the reader, as it were. If I hack Mark's code too extensively, it'll be hard to catch back up when he posts new updates. > Avoiding the shrinking should be the main thing to do then, in this case > it hurts more than it helps. Yes, that's true in all cases, really. It just confuses people. When I first wrote this code, there were some severe performance penalties associated with odd scale factors on many systems, so halving the window size in the event where the desktop area wasn't big enough to accommodate its client area seemed like a good idea. I also didn't want users running with a downsampled window without intending to, since that's like running an LCD at a resolution other than its native one, only worse. The half-size window is a pretty good clue that things are configured suboptimally. I'll probably just "fix" the problem with a readme file, since there aren't any easy hacks that work well in all cases where the desktop resolution is too small for the window. Forcing fullscreen mode in those cases would also be an option. > Oh... The oscillator ADEV positions at 2000 s and 5000 s is displaced > over the 500s and 1000s respectively, 2 digits too much to the left and > halfway up on the mentioned values, seems like 16 left and 16 up if the > font is 8 x 8 as you say. Not sure I follow you there; is this happening in a specific resolution? The table seems to look OK in the default res (attached) - is this similar to what you're seeing? -- john, KE5FX -------------- next part -------------- A non-text attachment was scrubbed... Name: scrn.gif Type: image/gif Size: 39246 bytes Desc: not available Url : http://www.febo.com/pipermail/time-nuts/attachments/20090201/71f3b3a4/attachment-0001.gif From magnus at rubidium.dyndns.org Mon Feb 2 01:59:09 2009 From: magnus at rubidium.dyndns.org (Magnus Danielson) Date: Mon, 02 Feb 2009 02:59:09 +0100 Subject: [time-nuts] Lady Heather's Window In-Reply-To: References: Message-ID: <4986536D.4040700@rubidium.dyndns.org> John Miles skrev: >> I think a more flexible fashion of setting size would be a good thing. >> The preferred fasion is always being able to do standard resize by >> dragging the edge, but saying x*y size at command line kind of works. > > Yep, the code could support arbitrary x*y sizes, but he has several groups > of related parameters that are tied to the discrete size options -- and > also, I can only go into fullscreen mode in standard VGA/EVGA resolutions. > So I probably won't put a lot of work into that. Left as an exercise to the > reader, as it were. If I hack Mark's code too extensively, it'll be hard to > catch back up when he posts new updates. Maybe just detect the cases and selects among them. >> Avoiding the shrinking should be the main thing to do then, in this case >> it hurts more than it helps. > > Yes, that's true in all cases, really. It just confuses people. When I > first wrote this code, there were some severe performance penalties > associated with odd scale factors on many systems, so halving the window > size in the event where the desktop area wasn't big enough to accommodate > its client area seemed like a good idea. I can see that. > I also didn't want users running with a downsampled window without intending > to, since that's like running an LCD at a resolution other than its native > one, only worse. The half-size window is a pretty good clue that things are > configured suboptimally. Well... yes. > I'll probably just "fix" the problem with a readme file, since there aren't > any easy hacks that work well in all cases where the desktop resolution is > too small for the window. Forcing fullscreen mode in those cases would also > be an option. I think it would be more usefull. >> Oh... The oscillator ADEV positions at 2000 s and 5000 s is displaced >> over the 500s and 1000s respectively, 2 digits too much to the left and >> halfway up on the mentioned values, seems like 16 left and 16 up if the >> font is 8 x 8 as you say. > > Not sure I follow you there; is this happening in a specific resolution? > The table seems to look OK in the default res (attached) - is this similar > to what you're seeing? No, I was using the /vs as needed for the smallest resolution in order to avoid shrinking. Cheers, Magnus From richiem at hughes.net Mon Feb 2 06:35:22 2009 From: richiem at hughes.net (Richard Moore) Date: Sun, 1 Feb 2009 22:35:22 -0800 Subject: [time-nuts] Bill Ezell's ftp In-Reply-To: References: Message-ID: <01EAA8A7-C52C-4838-A436-5A256DC433D3@hughes.net> Mark (and Bill), I had to use my FTP client (Transmit) to get to anything. Using the browser did not work -- took me to an empty folder named "pub". Everything is there when I connect using Transmit on my iMac. Thanks Bill, for all the work! Greatly appreciated. Dick Moore On Feb 1, 2009, at 1:39 PM, time-nuts-request at febo.com wrote: > Message: 7 > Date: Sun, 01 Feb 2009 14:07:32 -0500 > From: wje > Subject: [time-nuts] Some pics of the Datron 491x DC reference cell > To: Discussion of precise time and frequency measurement > > Message-ID: <4985F2F4.8070807 at quackers.net> > Content-Type: text/plain; charset=ISO-8859-1; format=flowed > > I've uploaded some photos of the Datron 10 volt cell and reference > board, as requested by a few members. > I've kept them full-resolution so you can see fine detail. But, that > means they are about 3 mb each. > The schematics for the cell board I previously posted are also > still on > the FTP site. > > DatronCell.jpg is an image of the entire cell board. The reference > board > is piggybacked on it. Note the thermal shield around the LTZ1000, and > the completely shielded sections on the board. Those contain the > digital > circuitry. The cell board itself goes into a shielded enclosure along > with the other three cells and the output buffer amp. > > DatronLTZCloseup.jpg is exactly that, a closeup of the LTZ with > thermal > shield removed. The interesting detail here is the mounting. Note the > cutouts on the board to provide thermal isolation. This board is also > very thin; I didn't measure it, but it's on the order of 0.5 - 1 mm > thick. > > DatronRefCloseup.jpg is a closeup image of the rest of the > reference board. > > The ref board is permanently soldered in place on the cell board, so I > can't show the back of it. > > The ftp login is: > > ftp ftp.quackers.net > login: ftp at quackers.net > password: (none required) > > Or, directly from a browser. I know this works for Firefox, I haven't > tried it with anything else: > > ftp://ftp at quackers.net:ftp at ftp.quackers.net/ > > Some of you are also waiting for the schematic for the LTZ board > itself; > I'm recreating it now (since I lost it) and will post when ready. > > -- > Bill Ezell > ---------- > They said 'Windows or better' > so I used Linux. > > > > > ------------------------------ > > Message: 8 > Date: Sun, 1 Feb 2009 21:29:55 +0000 > From: Mark Sims > Subject: [time-nuts] Some pics of the Datron 491x DC reference cell > To: > Message-ID: > Content-Type: text/plain; charset="Windows-1252" > > > I tried to access the photos... got to an empty pub directory... > > Cutouts in a circuit board around a reference chip are usually more > to relieve mechanical stress from the chip package caused by the PC > board expanding/contracting/warping. This stress can affect the > refereence voltage more than you might expect. Malone does this on > his little Xicor X6008C voltage reference boards. (BTW that chip > has been discontinued). That is why I prefer his boards over the > other one that is available on Ebay that does not cut out around > the ref chip. If you can live with that, you can get a packaged > 5V / 0.01% reference for $14.50 shipped... see Ebay item > 250366551770... the guy that builds them is closing them out... > and he ships worldwide. > _________________________________________________________________ > Windows Live? Hotmail??more than just e-mail. > http://windowslive.com/howitworks? > ocid=TXT_TAGLM_WL_t2_hm_justgotbetter_howitworks_012009 > > > ------------------------------ > > Message: 9 > Date: Sun, 1 Feb 2009 16:39:47 -0500 (EST) > From: "Richard W. Solomon" > Subject: [time-nuts] Lady Heather Question > To: time-nuts at febo.com > Message-ID: > <29501338.1233524387396.JavaMail.root at mswamui- > chipeau.atl.sa.earthlink.net> > > Content-Type: text/plain; charset=UTF-8 > > Having failed to get the program working properly on an old 486 > Laptop, > I installed it on a more modern machine running XP. When I try to open > the program I get an error message about Com Port Not Available ?? > > How does one select the COM Port ? > > Thanks, Dick, W1KSZ > > > > ------------------------------ > > _______________________________________________ > time-nuts mailing list > time-nuts at febo.com > https://www.febo.com/cgi-bin/mailman/listinfo/time-nuts > > End of time-nuts Digest, Vol 55, Issue 4 > **************************************** From sar10538 at gmail.com Mon Feb 2 23:17:46 2009 From: sar10538 at gmail.com (Steve Rooke) Date: Tue, 3 Feb 2009 12:17:46 +1300 Subject: [time-nuts] Racal-Dana 1992 Opt 4B or 4E, what's best? Message-ID: <1231b6a80902021517l73912ec4vd4fbd14c7d3d554e@mail.gmail.com> Has anyone experience with these two options as the specifications I have seem to slightly favour the Opt 4B (9423) over 4E (404386) high stability ocxos? 73, Steve -- Steve Rooke - ZL3TUV & G8KVD & JAKDTTNW Omnium finis imminet From EWKehren at aol.com Tue Feb 3 01:01:55 2009 From: EWKehren at aol.com (EWKehren at aol.com) Date: Mon, 2 Feb 2009 20:01:55 EST Subject: [time-nuts] Racal-Dana 1992 Opt 4B or 4E, what's best? Message-ID: I have option E but am only aware of E and A. My manual does not refer to an option B. What would you like to know specifically? Bert in Miami **************Who's never won? Biggest Grammy Award surprises of all time on AOL Music. (http://music.aol.com/grammys/pictures/never-won-a-grammy?ncid=emlcntusmusi00000001) From sar10538 at gmail.com Tue Feb 3 03:17:54 2009 From: sar10538 at gmail.com (Steve Rooke) Date: Tue, 3 Feb 2009 16:17:54 +1300 Subject: [time-nuts] Racal-Dana 1992 Opt 4B or 4E, what's best? In-Reply-To: References: Message-ID: <1231b6a80902021917g6df52644m187d05033dc0082a@mail.gmail.com> Hi Bert, If you go to the Bama site for this counter: http://bama.edebris.com/manuals/racal/1991/ You can see references to the Opt 4B, 9423, ocxo in: http://bama.edebris.com/download/racal/1991/RACAL-DANA-1991-1992-UserManual.pdf http://bama.edebris.com/download/racal/1991/RACAL-DANA-1991-1992-ServiceManual.pdf And the Opt 4E, 404386, ocxo in: http://bama.edebris.com/download/racal/1991/RACALDANA-1991-1992-IM.pdf As for what I wanted to know, well I wondered if anyone had experience if one was better than the other for drift, stability, thermal stability, etc. The 4E unit appears to be the more substantial but the specs in the above pdfs seem to favour the 4B with regard to thermal stability. This is really all very well anyway as I would like to use the counter locked to GPS for most of the time but there may be times when it could be used as stand-alone. I guess the only way to tell is to run a test on them both but before I looked into this I wondered if anyone had already done this, that's all. 73, Steve 2009/2/3 : > I have option E but am only aware of E and A. My manual does not refer to an > option B. What would you like to know specifically? Bert in Miami > **************Who's never won? Biggest Grammy Award surprises of all time on > AOL Music. > (http://music.aol.com/grammys/pictures/never-won-a-grammy?ncid=emlcntusmusi00000001) > _______________________________________________ > time-nuts mailing list -- time-nuts at febo.com > To unsubscribe, go to https://www.febo.com/cgi-bin/mailman/listinfo/time-nuts > and follow the instructions there. > -- Steve Rooke - ZL3TUV & G8KVD & JAKDTTNW Omnium finis imminet From EWKehren at aol.com Tue Feb 3 09:35:29 2009 From: EWKehren at aol.com (EWKehren at aol.com) Date: Tue, 3 Feb 2009 04:35:29 EST Subject: [time-nuts] Racal-Dana 1992 Opt 4B or 4E, what's best? Message-ID: Sorry I cannot help. If you look at the publication dates it seams to me that the E is a replacement for the B with slightly relaxed specification. Bert **************Who's never won? Biggest Grammy Award surprises of all time on AOL Music. (http://music.aol.com/grammys/pictures/never-won-a-grammy?ncid=emlcntusmusi00000001) From n8fgv at usa.net Tue Feb 3 23:32:21 2009 From: n8fgv at usa.net (Daniel Schultz) Date: Tue, 03 Feb 2009 18:32:21 -0500 Subject: [time-nuts] LPRO and switching power supply? Message-ID: <634NBcXfV7116S11.1233703941@cmsweb11.cms.usa.net> What type of power supply are you using to provide the 24 volts to the LPRO? The Users Guide mentions (on page 23) that there is a susceptibility to input noise and suggests that users not connect fans to the same power supply, but it does not give any quantitative data, except for mentioning that "This topic is discussed further in the LPRO Data Manual" which I cannot find. Have you had success running the LPRO from a switch mode power supply? Dan Schultz N8FGV From df6jb at ulrich-bangert.de Wed Feb 4 10:32:25 2009 From: df6jb at ulrich-bangert.de (Ulrich Bangert) Date: Wed, 4 Feb 2009 11:32:25 +0100 Subject: [time-nuts] LPRO and switching power supply? In-Reply-To: <634NBcXfV7116S11.1233703941@cmsweb11.cms.usa.net> Message-ID: Dan, I run the LPRO (as well as an FRK-L) from an 24 V / 25 W closed frame switched power supply from Mean Well like http://www.meanwell.com/search/ned-100/NED-100-catalog.pdf without any problems. Please note that it does not necessarily mean that the output signal may not be even a bit "cleaner" with an analogue supply but I find no general problems with it. 73s and my best regards Ulrich, DF6JB > -----Ursprungliche Nachricht----- > Von: time-nuts-bounces at febo.com > [mailto:time-nuts-bounces at febo.com] Im Auftrag von Daniel Schultz > Gesendet: Mittwoch, 4. Februar 2009 00:32 > An: time-nuts at febo.com > Betreff: [time-nuts] LPRO and switching power supply? > > > What type of power supply are you using to provide the 24 > volts to the LPRO? The Users Guide mentions (on page 23) that > there is a susceptibility to input noise and suggests that > users not connect fans to the same power supply, but it does > not give any quantitative data, except for mentioning that > "This topic is discussed further in the LPRO Data Manual" > which I cannot find. > > Have you had success running the LPRO from a switch mode power supply? > > Dan Schultz N8FGV > > > > _______________________________________________ > time-nuts mailing list -- time-nuts at febo.com > To unsubscribe, go to > https://www.febo.com/cgi-> bin/mailman/listinfo/time-nuts > and > follow the instructions there. > From sr75pro at free.fr Wed Feb 4 11:44:44 2009 From: sr75pro at free.fr (Sylvain RICHARD) Date: Wed, 04 Feb 2009 12:44:44 +0100 Subject: [time-nuts] LPRO and switching power supply? In-Reply-To: <634NBcXfV7116S11.1233703941@cmsweb11.cms.usa.net> References: <634NBcXfV7116S11.1233703941@cmsweb11.cms.usa.net> Message-ID: <49897FAC.8070200@free.fr> Daniel Schultz a ?crit : > What type of power supply are you using to provide the 24 volts to the LPRO? > The Users Guide mentions (on page 23) that there is a susceptibility to input > noise and suggests that users not connect fans to the same power supply, but > it does not give any quantitative data, except for mentioning that "This topic > is discussed further in the LPRO Data Manual" which I cannot find. > > Have you had success running the LPRO from a switch mode power supply? > > Dan Schultz N8FGV > Dan, I have no experience with the LPRO, but you may find the results of connecting various power supplies to a Thunderbolt interesting: http://www.leapsecond.com/pages/tbolt/noise.htm You may already have seen that page, but it's an eye-opener. Especially when you can't measure phase noise yourself. -- Sylvain RICHARD From CHJ at taconic.net Wed Feb 4 12:49:44 2009 From: CHJ at taconic.net (Christopher Hilton-Johnson) Date: Wed, 4 Feb 2009 07:49:44 -0500 Subject: [time-nuts] Tracor 600 Rubidium Frequency Standard Message-ID: <4D62E05B511B16439FE408E5DB93985F0732A4@behemothus.Pchjhome.local> I am a newbie to this group Apologies if I am breaching policy with this, but I have recently acquired a Tracor 600 Rubidium Frequency standard. Although there is some brief info on the net about other Tracor rubidium products, I cannot find any information about this at all. I know Tracor are now part of BAE systems, but have had no luck in getting info from them In particular I need manuals and schematics. Can your group help? Kind regards Chris HJ From hmurray at megapathdsl.net Wed Feb 4 18:06:59 2009 From: hmurray at megapathdsl.net (Hal Murray) Date: Wed, 04 Feb 2009 10:06:59 -0800 Subject: [time-nuts] LPRO and switching power supply? In-Reply-To: Message from "Daniel Schultz" of "Tue, 03 Feb 2009 18:32:21 EST." <634NBcXfV7116S11.1233703941@cmsweb11.cms.usa.net> Message-ID: <20090204180700.32AFEBCDD@ip-64-139-1-69.sjc.megapath.net> > Have you had success running the LPRO from a switch mode power supply? I've been using ETS240170UTC-P5P-SZ, $25 from Digikey. 24V, 1.7A I don't have the gear to see what the spectrum looks like. The 10 MHz output is off a bit, but seems stable. I haven't tried to adjust it. -- These are my opinions, not necessarily my employer's. I hate spam. From holrum at hotmail.com Wed Feb 4 05:45:55 2009 From: holrum at hotmail.com (Mark Sims) Date: Wed, 4 Feb 2009 05:45:55 +0000 Subject: [time-nuts] Tbolt temperature sensor In-Reply-To: References: Message-ID: Well, I replaced the DS1620 temperature sensor in my 2005 vintage Thunderbolt that shows the (mostly) flat temperature curve that is quantized to 0.25/0.75C increments (earlier models produce a smooth temperature curve that tracks minute temperature changes). The new DS1620 chip shows the same behavior as the old one. Conclusion is that either Trimble changed the firmware (but did not change the reported firmware revision number/date) or Dallas Semiconductor changed the DS1620 high resolution read mode in a way that messes up the Tbolt firmware (the DS1620 that I used did have a similar date code to the one I changed out). My bet is Dallas Semi changed their chip in a way that is not compatible with the Tbolt firmware. If Trimble changed their firmware to not do the high resolution read cycle, they would not have done the first steps of the high res read cycle (mask off the LSB, subtract 0.25C) which actually reduces the temperature resolution from nine bits to eight bits. Trimble could have easily missed the problem if they did not plot the temperature curves. The unit reports a proper temperature, it just does not have the fine resolution of the earlier units... which could easily affect holdover performance. Does anybody out there have a late model (like 2005 date code) Tbolt (like from the second batch of TAPR units) that shows the smoothly changing temperature curves? The pictures that Asa posted show the flat temperature curve. All my 2002-2003 vintage units show the smoothly changing curve. Lady Heather shows the date of manufacture on the third line down. _________________________________________________________________ Windows Live? Hotmail??more than just e-mail. http://windowslive.com/howitworks?ocid=TXT_TAGLM_WL_t2_hm_justgotbetter_howitworks_012009 From tvb at LeapSecond.com Wed Feb 4 21:33:48 2009 From: tvb at LeapSecond.com (Tom Van Baak) Date: Wed, 4 Feb 2009 13:33:48 -0800 Subject: [time-nuts] Tbolt temperature sensor References: Message-ID: <20587202C5A943079CD12C23E3BC8232@pc52> Hi Mark, I've kept aside several Rev A through Rev E TBolt's here just in case someone ever found any difference among them. So what chip or P/N or date code should I look at on these units to help you on your interesting quest? /tvb From SAIDJACK at aol.com Thu Feb 5 00:56:54 2009 From: SAIDJACK at aol.com (SAIDJACK at aol.com) Date: Wed, 4 Feb 2009 19:56:54 EST Subject: [time-nuts] Tbolt temperature sensor Message-ID: Hi Mark, from my experience even a resolution of 0.25C is several orders of magnitude too low for use in holdover temperature compensation on an OCXO. This would result in fairly massive EFC changes on the Thunderbolt ovens, especially if they are single ovens with significant thermal sensitivity. The average error for a typical single oven OCXO due to this would be ~1.25E-010, see below. To make this work somewhat smoothly, one would need at least 0.001C or better 0.0001C resolution. I can't get into too much detail, but our units allow for down to ~0.000025C resolution by using a 24 bit ADC for the temperature sensing. Keep in mind that for temperature compensation the absolute temperature is not of interest, but rather the change of temperature over time. This kind of resolution is not possible with I2C based absolute temp sensors, one needs analog circuitry feeding high-resolution Delta-Sigma ADC's. These circuits are not accurate on the absolute temperature, but they do have very high resolution for relative temperature changes. A resolution of 0.25C would give you only 40 individual DAC settings for a 10C change as could be expected on a typical diurnal temperature change on a lab bench. This would make the frequency jump around and really sink the ADEV performance of the unit during holdover. To give you an example: a typical single-oven OCXO has about 1ppb per degree C change. This would mean the unit could only be adjusted by 2.5E-010 steps with the Dallas Temp sensor as a reference. So the average error would be 1.25E-010, which results in a massive 12.5 microseconds average drift error in 1000s intervals due to the temperature quantization error! The math looks better for double oven OCXO's, but not much. I am pretty sure the Dallas chip is only used for environmental monitoring of the absolute temperature, and not for compensation anywhere inside the PLL loop. Any Thunderbolt designers on this list?? bye, Said In a message dated 2/4/2009 12:45:51 Pacific Standard Time, holrum at hotmail.com writes: My bet is Dallas Semi changed their chip in a way that is not compatible with the Tbolt firmware. If Trimble changed their firmware to not do the high resolution read cycle, they would not have done the first steps of the high res read cycle (mask off the LSB, subtract 0.25C) which actually reduces the temperature resolution from nine bits to eight bits. Trimble could have easily missed the problem if they did not plot the temperature curves. The unit reports a proper temperature, it just does not have the fine resolution of the earlier units... which could easily affect holdover performance. From tvb at LeapSecond.com Thu Feb 5 01:49:54 2009 From: tvb at LeapSecond.com (Tom Van Baak) Date: Wed, 4 Feb 2009 17:49:54 -0800 Subject: [time-nuts] Tbolt temperature sensor References: Message-ID: > To make this work somewhat smoothly, one would need at least 0.001C or > better 0.0001C resolution. Hi Said, Can you provide some real data, to help me believe this claim? When you get below 1C or 0.1C resolution it starts to matter from which direction the temperature gradient is headed, or from which angle air is flowing. Or which side is up. Or what the humidity is, etc. This, because temperature gradients break any steady state model you have based on a point source (e.g., temp sensor) or average temperature measurement (e.g., oven current). Then there is the matter of the rate at which temperature changes; slow temp changes and rapid temp changes affect an OCXO quite differently. This, due to different thermal time constants of all the metal and insulation materials in and around the OCXO or GPSDO. > To give you an example: a typical single-oven OCXO has about 1ppb per degree > C change. This would mean the unit could only be adjusted by 2.5E-010 steps > with the Dallas Temp sensor as a reference. So the average error would be > 1.25E-010, which results in a massive 12.5 microseconds average drift error in > 1000s intervals due to the temperature quantization error! Are you assuming a design where one takes a Dallas reading each second and stuffs it into the EFC every second? No one would actually do that. Instead if one averages the temperature sensor over 10, 100 or more seconds you avoid large EFC steps. Everything else in a GPSDO is all about slow averaging; there's no reason temperature adaptation cannot be treated the same. I suppose I should add a random temperature cycle hold-over test to the abuse I inflict on each GPSDO here. Are you saying the Fury would do much better than others in this regard? Again, given this is time-nuts, some real data would be nice. /tvb From magnus at rubidium.dyndns.org Thu Feb 5 02:21:24 2009 From: magnus at rubidium.dyndns.org (Magnus Danielson) Date: Thu, 05 Feb 2009 03:21:24 +0100 Subject: [time-nuts] Tbolt temperature sensor In-Reply-To: References: Message-ID: <498A4D24.1060002@rubidium.dyndns.org> Tom Van Baak skrev: >> To make this work somewhat smoothly, one would need at least 0.001C or >> better 0.0001C resolution. > > Hi Said, > > Can you provide some real data, to help me believe this claim? > > When you get below 1C or 0.1C resolution it starts to matter > from which direction the temperature gradient is headed, or > from which angle air is flowing. Or which side is up. Or what > the humidity is, etc. Recall that a typical thunderbolt is a fairly closed up box. The case is a great heat-conductor so gradients and changes caused from external sources is damped by the relatively high heat conductivity in the shell as compared to the head conductivity into the box and especially the oscillator. Inside the box there are some local heat sources dispersed throughout, but the OCXO heats up pretty good by itself. There is not much self convection in there. One also must recall that an OCXO oven has a damping of temperature swing. This gain of the oven is of interest here. While we surely could have use for more bits, an OCXO ambient temperature of 0-70 C may become say a 0,07 degree shift, a gain of 0,01. A 0,25 C shift will become 0,0025 C shift at the crystal. Is the resolution usefull? Yes, but it is has limits. The important thing being, which limits was it intended to solve? While I agree that absolute temperature is not necessary, high resolution can be spoiled by other sources of temperature, such as enabling another dump source, shift of sats etc. all causing shift in power usage and changing the gradients inside the box. To get any real performance one has to keep the temperature sensor tightly tied to the OCXO while well isolate from the other stuff. For the Thunderbolt target application I think it would be severly overkill. > This, because temperature gradients break any steady state > model you have based on a point source (e.g., temp sensor) > or average temperature measurement (e.g., oven current). Actually, any CHANGE in temperature gradients will do that. If you have a static temperature gradient you will not really notice. Gradients have however a tendency to change... as power burn changes in different places etc. > Then there is the matter of the rate at which temperature changes; > slow temp changes and rapid temp changes affect an OCXO > quite differently. This, due to different thermal time constants of > all the metal and insulation materials in and around the OCXO > or GPSDO. Even a very passive filter making convection onto the OCXO case basically none causes a very dramatic change and the response curve is quite different. The most important thing is that the oven actually can track along fast enought. > >> To give you an example: a typical single-oven OCXO has about 1ppb per degree >> C change. This would mean the unit could only be adjusted by 2.5E-010 steps >> with the Dallas Temp sensor as a reference. So the average error would be >> 1.25E-010, which results in a massive 12.5 microseconds average drift error in >> 1000s intervals due to the temperature quantization error! > > Are you assuming a design where one takes a Dallas reading > each second and stuffs it into the EFC every second? No one > would actually do that. Instead if one averages the temperature > sensor over 10, 100 or more seconds you avoid large EFC steps. > Everything else in a GPSDO is all about slow averaging; there's > no reason temperature adaptation cannot be treated the same. Except that temperature changes may be fairly quick... I have detuned OCXOs just by walking by them... pushing colder air onto them as I come walking. The OCXO needs to do up-front compensation, but the EFC path may do after the fact compensation if propper dynamic models exist and is fairly valid. I don't think 0,25 C resolution is worthless, but it can still do good work. > I suppose I should add a random temperature cycle hold-over > test to the abuse I inflict on each GPSDO here. Are you saying > the Fury would do much better than others in this regard? There are many interesting aspects to such thermal dynamic tests. > Again, given this is time-nuts, some real data would be nice. As always. Cheers, Magnus From sar10538 at gmail.com Thu Feb 5 03:29:57 2009 From: sar10538 at gmail.com (Steve Rooke) Date: Thu, 5 Feb 2009 16:29:57 +1300 Subject: [time-nuts] Tbolt temperature sensor In-Reply-To: <498A4D24.1060002@rubidium.dyndns.org> References: <498A4D24.1060002@rubidium.dyndns.org> Message-ID: <1231b6a80902041929u211490cufc9e3178748248bd@mail.gmail.com> 2009/2/5 Magnus Danielson : > Tom Van Baak skrev: > > I don't think 0,25 C resolution is worthless, but it can still do good work. > Sorry for the heavy editing but I have been following this thread closely. Now if the temperature circuit only reacts to changes of 0.25C then doesn't that represent quite a bit of hysteresis in the system, and is liable to cause overshoot, or am I missing something here. 73, Steve -- Steve Rooke - ZL3TUV & G8KVD & JAKDTTNW Omnium finis imminet From holrum at hotmail.com Thu Feb 5 00:03:29 2009 From: holrum at hotmail.com (Mark Sims) Date: Thu, 5 Feb 2009 00:03:29 +0000 Subject: [time-nuts] Tbolt temperature sensor In-Reply-To: References: Message-ID: Hello Tom, All my earlier Tbolts report 2002-2003 manufacturing date codes. They are rev 2.22 and 3.00 firmware. They all show the nice smooth high res temperature curves that are usually sinusoidal. I know of around a dozen units that respond this way. All the later build Tbolts that I have seen (mine, Asa's, and one other) are from the March to May 2005 time frame. They all show the basically flat temperature curve that is quantized to 0.25/0.75C readings (except when a firmware filter kicks in for a short time when the value steps between readings). The temperature sensor chip is a DS1620 chip in a small SOIC-8 package next to the RS-232 connector. I guess the questions are: 1) Is the difference a firmware change (even though the "problem" units all report the same 3.0 firmware levels as the earlier units) or is it a change in the sensor chip that causes the high-res temperature read technique listed in the DS1620 data sheet to fail on the Tbolts? 2) Which units are affected by the issue? Can it be isolated to a date code or some other marking on the DS1620 package? Is there a Tbolt manufacturing date or range where the issue first pops up or occurs over? Lady Heather shows the Tbolt manufacturing dates and firmware dates. She also plots the temperature curve. _________________________________________________________________ Windows Live?: Keep your life in sync. http://windowslive.com/howitworks?ocid=TXT_TAGLM_WL_t1_allup_howitworks_022009 From bruce.griffiths at xtra.co.nz Thu Feb 5 04:16:07 2009 From: bruce.griffiths at xtra.co.nz (Bruce Griffiths) Date: Thu, 05 Feb 2009 17:16:07 +1300 Subject: [time-nuts] Tbolt temperature sensor In-Reply-To: <1231b6a80902041929u211490cufc9e3178748248bd@mail.gmail.com> References: <498A4D24.1060002@rubidium.dyndns.org> <1231b6a80902041929u211490cufc9e3178748248bd@mail.gmail.com> Message-ID: <498A6807.9020709@xtra.co.nz> Steve Rooke wrote: > 2009/2/5 Magnus Danielson : > >> Tom Van Baak skrev: >> >> I don't think 0,25 C resolution is worthless, but it can still do good work. >> >> > > Sorry for the heavy editing but I have been following this thread > closely. Now if the temperature circuit only reacts to changes of > 0.25C then doesn't that represent quite a bit of hysteresis in the > system, and is liable to cause overshoot, or am I missing something > here. > > 73, Steve > According to the data sheet the readout resolution in normal 9 bit mode is actually 0.5C not 0.25C. The firmware appears to offset the temperature output data by -0.25C from the DS1620 reading. The high resolution mode has 12 bits resolution or about 0.0625C. Conversion time from temperature to digital is about 750ms. Bruce From SAIDJACK at aol.com Thu Feb 5 05:06:43 2009 From: SAIDJACK at aol.com (SAIDJACK at aol.com) Date: Thu, 5 Feb 2009 00:06:43 EST Subject: [time-nuts] Tbolt temperature sensor Message-ID: Hi Tom, trying to send this message, but attachment is too big. John has to approve the attached image: Hi Tom, > To make this work somewhat smoothly, one would need at least 0.001C or > better 0.0001C resolution. >Hi Said, >Can you provide some real data, to help me believe this claim? Please remember I talked about resolution, not accuracy. This type of resolution is easy to achieve with a 24 bit ADC and a thermistor; keeping in mind that 24 bits of resolution are 16 million steps. Again, this doesn't mean we will have accuracy nearing this type of resolution. See attached image, these are three single-oven FireFly-II GPSDO's running in our garage, exposed to at least 20 degrees C of temperature change. You can see a significant amount of EFC change on all units (about 4mV pk to pk on each unit), and this is mostly done through the temperature compensation (red trace is EFC, blue trace is temperature/OCXO-current). You can see this by the fact that the red EFC trace almost perfectly follows the blue temperature trace in inverse, and the units are well locked throughout the ordeal so compensation is working really well. If it wasn't, then the UTC phase would have to have a larger error on it to create this kind of EFC range. When doing the math, this comes out to about 25 micro Degrees per ADC LSB. Again this is resolution, and I would not be surprised if the noise level of the temp sensor is 10x or 100x higher than this, but as you said we can do averaging on these numbers to get rid of noise. This also doesn't mean the thermistor can generate accuracy to this level, it may have higher hysterisis etc, but again the resolution comes for free with a 24 bit ADC. >When you get below 1C or 0.1C resolution it starts to matter >from which direction the temperature gradient is headed, or >from which angle air is flowing. Or which side is up. Or what >the humidity is, etc. Yes, but we don't mind about that if the temperature sensor sits right on the crystal inside the OCXO well shielded and insulated from the outside world as it would be for most OCXO's. >This, because temperature gradients break any steady state >model you have based on a point source (e.g., temp sensor) >or average temperature measurement (e.g., oven current). >Then there is the matter of the rate at which temperature changes; >slow temp changes and rapid temp changes affect an OCXO >quite differently. This, due to different thermal time constants of >all the metal and insulation materials in and around the OCXO >or GPSDO. Absolutely. Fast changing transients are deadly for any OCXO. But if a GPSDO is hermetically sealed in an enclosure such as the Thunderbolt, then the unit will low-pass filter any thermal change. As long as the thermal change is slower than the heater's natural loop time constant then this should not be a problem since the heater can track the change. We are in trouble if the PID cannot follow the thermal gradient. > To give you an example: a typical single-oven OCXO has about 1ppb per degree > C change. This would mean the unit could only be adjusted by 2.5E-010 steps > with the Dallas Temp sensor as a reference. So the average error would be > 1.25E-010, which results in a massive 12.5 microseconds average drift error in > 1000s intervals due to the temperature quantization error! >Are you assuming a design where one takes a Dallas reading >each second and stuffs it into the EFC every second? No one >would actually do that. My point exactly. But I think this is what Mike said about the Thunderbolt in his question. My point was that this is not done, and he should not be worried about the change to a lower resolution temp sensor affecting the signal quality from his Thunderbolt. >Instead if one averages the temperature >sensor over 10, 100 or more seconds you avoid large EFC steps. >Everything else in a GPSDO is all about slow averaging; there's >no reason temperature adaptation cannot be treated the same. Respectfully I do not agree :) This will only work if the temp sensor jumps back and forth between two LSB's due to noise (PWM, dithering). If it only changes 0.25C every 10 minutes, then you would see a 2.5E-010 change in the frequency when this jump happens. It's just too coarse, even if it is low-pass filtered. A typical GPSDO has 5E-013 DAC LSB resolution, so reducing this to 2.5E-010 would be detrimental. >I suppose I should add a random temperature cycle hold-over >test to the abuse I inflict on each GPSDO here. Are you saying >the Fury would do much better than others in this regard? No, I am saying that other vendors also don't use I2C temp sensors for their temperature compensation because their resolution is not enough, and one could only measure temperature at one particular point on the PCB, not the crystal itself. I would think most GPSDO's use the same method as Fury. >Again, given this is time-nuts, some real data would be nice. See the attached image, it would be good to see how the competitors units' correlate OCXO EFC to temperature. I am sure they have very high resolution as well. I would think that the resolution would be at least as good as 1 to 5 EFC LSB steps per temperature sensor step to avoid quantization errors due to temp sensor resolution. bye, Said From tvb at LeapSecond.com Thu Feb 5 05:57:33 2009 From: tvb at LeapSecond.com (Tom Van Baak) Date: Wed, 4 Feb 2009 21:57:33 -0800 Subject: [time-nuts] Tbolt temperature sensor References: Message-ID: <869E2D4801E54585B7C6A798EAC0C1DB@pc52> Cool. Thanks for the plots. I can see the oven current change (temp), as well as the EFC changing to counteract it. As far as I can tell the three GPSDO in the photo are GPS locked. Is that correct? I mean, they stay within a couple of 10 ns over the span of 70+ hours, so it's not likely in holdover mode. In that case, how do you distinguish in the plots between the GPS phase locking reaction to the OCXO starting to go off-frequency and software temperature compensation? How much of the correction in the plots is due to normal phase locking against GPS and how much is due to your temperature sensing and compensation? The plots look like what any GPSDO would do. Temp goes down; OCXO changes rate; GPS TIC detects divergence; and DAC changes EFC to compensate for everything. /tvb ----- Original Message ----- From: SAIDJACK at aol.com To: tvb at leapsecond.com ; time-nuts at febo.com Sent: Wednesday, February 04, 2009 8:58 PM Subject: Re: [time-nuts] Tbolt temperature sensor Hi Tom, > To make this work somewhat smoothly, one would need at least 0.001C or > better 0.0001C resolution. >Hi Said, >Can you provide some real data, to help me believe this claim? Please remember I talked about resolution, not accuracy. This type of resolution is easy to achieve with a 24 bit ADC and a thermistor; keeping in mind that 24 bits of resolution are 16 million steps. Again, this doesn't mean we will have accuracy nearing this type of resolution. See attached image, these are three single-oven FireFly-II GPSDO's running in our garage, exposed to at least 20 degrees C of temperature change. You can see a significant amount of EFC change on all units (about 4mV pk to pk on each unit), and this is mostly done through the temperature compensation (red trace is EFC, blue trace is temperature/OCXO-current). You can see this by the fact that the red EFC trace almost perfectly follows the blue temperature trace in inverse, and the units are well locked throughout the ordeal so compensation is working really well. If it wasn't, then the UTC phase would have to have a larger error on it to create this kind of EFC range. When doing the math, this comes out to about 25 micro Degrees per ADC LSB. Again this is resolution, and I would not be surprised if the noise level of the temp sensor is 10x or 100x higher than this, but as you said we can do averaging on these numbers to get rid of noise. This also doesn't mean the thermistor can generate accuracy to this level, it may have higher hysterisis etc, but again the resolution comes for free with a 24 bit ADC. >When you get below 1C or 0.1C resolution it starts to matter >from which direction the temperature gradient is headed, or >from which angle air is flowing. Or which side is up. Or what >the humidity is, etc. Yes, but we don't mind about that if the temperature sensor sits right on the crystal inside the OCXO well shielded and insulated from the outside world as it would be for most OCXO's. >This, because temperature gradients break any steady state >model you have based on a point source (e.g., temp sensor) >or average temperature measurement (e.g., oven current). >Then there is the matter of the rate at which temperature changes; >slow temp changes and rapid temp changes affect an OCXO >quite differently. This, due to different thermal time constants of >all the metal and insulation materials in and around the OCXO >or GPSDO. Absolutely. Fast changing transients are deadly for any OCXO. But if a GPSDO is hermetically sealed in an enclosure such as the Thunderbolt, then the unit will low-pass filter any thermal change. As long as the thermal change is slower than the heater's natural loop time constant then this should not be a problem since the heater can track the change. We are in trouble if the PID cannot follow the thermal gradient. > To give you an example: a typical single-oven OCXO has about 1ppb per degree > C change. This would mean the unit could only be adjusted by 2.5E-010 steps > with the Dallas Temp sensor as a reference. So the average error would be > 1.25E-010, which results in a massive 12.5 microseconds average drift error in > 1000s intervals due to the temperature quantization error! >Are you assuming a design where one takes a Dallas reading >each second and stuffs it into the EFC every second? No one >would actually do that. My point exactly. But I think this is what Mike said about the Thunderbolt in his question. My point was that this is not done, and he should not be worried about the change to a lower resolution temp sensor affecting the signal quality from his Thunderbolt. >Instead if one averages the temperature >sensor over 10, 100 or more seconds you avoid large EFC steps. >Everything else in a GPSDO is all about slow averaging; there's >no reason temperature adaptation cannot be treated the same. Respectfully I do not agree :) This will only work if the temp sensor jumps back and forth between two LSB's due to noise (PWM, dithering). If it only changes 0.25C every 10 minutes, then you would see a 2.5E-010 change in the frequency when this jump happens. It's just too coarse, even if it is low-pass filtered. A typical GPSDO has 5E-013 DAC LSB resolution, so reducing this to 2.5E-010 would be detrimental. >I suppose I should add a random temperature cycle hold-over >test to the abuse I inflict on each GPSDO here. Are you saying >the Fury would do much better than others in this regard? No, I am saying that other vendors also don't use I2C temp sensors for their temperature compensation because their resolution is not enough, and one could only measure temperature at one particular point on the PCB, not the crystal itself. I would think most GPSDO's use the same method as Fury. >Again, given this is time-nuts, some real data would be nice. See the attached image, it would be good to see how the competitors units' correlate OCXO EFC to temperature. I am sure they have very high resolution as well. I would think that the resolution would be at least as good as 1 to 5 EFC LSB steps per temperature sensor step to avoid quantization errors due to temp sensor resolution. bye, Said From holrum at hotmail.com Thu Feb 5 06:25:08 2009 From: holrum at hotmail.com (Mark Sims) Date: Thu, 5 Feb 2009 06:25:08 +0000 Subject: [time-nuts] Tbolt temperature sensor In-Reply-To: References: Message-ID: Actually, what is apparently going on with the newer Thunderbolts is worse than that. The reported temperature seems to only end in 0.25C or 0.75C, so it has an effective 0.5C resolution. The basic DS1620 resolution is 9 bits, but the Tbolt firmware reduces that to 8 bits since the first step in the high res temp algorithm is to mask off the lower bit. Whatever is going on they are not getting the extra resolution that they think they are, and in fact they are reducing the basic resolution of the chip. The firmware does seem do do some filtering on those values since whenever the reading steps you can see some smoothing going on. A lot of times the temperature value oscillatates around the step point. The filter apparently does not have any hysteresis. The older Thunderbolts produced a nice smooth curve. The high res temperature reading (Bruce says is 12 bit/0.0625C) coupled with the firmware filtering gave temperature curves with microdegree scale resolution. The newer ones clunk around with effectively 0.5C resolution. _________________________________________________________________ Windows Live?: Keep your life in sync. http://windowslive.com/howitworks?ocid=TXT_TAGLM_WL_t1_allup_howitworks_022009 From SAIDJACK at aol.com Thu Feb 5 08:59:08 2009 From: SAIDJACK at aol.com (SAIDJACK at aol.com) Date: Thu, 5 Feb 2009 03:59:08 EST Subject: [time-nuts] Tbolt temperature sensor Message-ID: Hi Tom, yes, they are all GPS locked. here is the same OCXO type without proper temp compensation: _http://www.jackson-labs.com/images/gpsstat.htm_ (http://www.jackson-labs.com/images/gpsstat.htm) You can see that now the phase has to change considerably to compensate for the EFC voltage change (because tempco has not been properly established yet). To have the TIC compensate, it needs to detect a phase offset first. This error will be prevented if the tempco already compensates properly for temperature changes. As the proper tempco is established, the phase offsets diminish in magnitude. bye, Said In a message dated 2/4/2009 21:58:18 Pacific Standard Time, tvb at LeapSecond.com writes: In that case, how do you distinguish in the plots between the GPS phase locking reaction to the OCXO starting to go off-frequency and software temperature compensation? How much of the correction in the plots is due to normal phase locking against GPS and how much is due to your temperature sensing and compensation? From bruce.griffiths at xtra.co.nz Thu Feb 5 09:01:57 2009 From: bruce.griffiths at xtra.co.nz (Bruce Griffiths) Date: Thu, 05 Feb 2009 22:01:57 +1300 Subject: [time-nuts] Tbolt temperature sensor In-Reply-To: References: Message-ID: <498AAB05.1060501@xtra.co.nz> Mark Sims wrote: > Actually, what is apparently going on with the newer Thunderbolts is worse than that. The reported temperature seems to only end in 0.25C or 0.75C, so it has an effective 0.5C resolution. The basic DS1620 resolution is 9 bits, but the Tbolt firmware reduces that to 8 bits since the first step in the high res temp algorithm is to mask off the lower bit. Whatever is going on they are not getting the extra resolution that they think they are, and in fact they are reducing the basic resolution of the chip. > > The firmware does seem do do some filtering on those values since whenever the reading steps you can see some smoothing going on. A lot of times the temperature value oscillatates around the step point. The filter apparently does not have any hysteresis. > > The older Thunderbolts produced a nice smooth curve. The high res temperature reading (Bruce says is 12 bit/0.0625C) coupled with the firmware filtering gave temperature curves with microdegree scale resolution. The newer ones clunk around with effectively 0.5C resolution. > Not quite right. According to the datasheet the resolution with the standard 9 bit output is actually 0.5C. The MSBit is actually a sign bit. Despite what the text may perhaps implies, 1/2 of a LSB is actually subtracted before combining with the residual count and other data (see attachment for formula). The resolution isnt in fact being truncated at all. An offset of 1/2 LSB is being subtracted from the coarse reading, this has no effect on the resolution. Bruce -------------- next part -------------- A non-text attachment was scrubbed... Name: TQ9133.png Type: image/png Size: 50138 bytes Desc: not available Url : http://www.febo.com/pipermail/time-nuts/attachments/20090205/573b7007/attachment-0001.png From GandalfG8 at aol.com Thu Feb 5 10:14:55 2009 From: GandalfG8 at aol.com (GandalfG8 at aol.com) Date: Thu, 5 Feb 2009 05:14:55 EST Subject: [time-nuts] Tbolt temperature sensor Message-ID: In a message dated 05/02/2009 06:25:52 GMT Standard Time, holrum at hotmail.com writes: Actually, what is apparently going on with the newer Thunderbolts is worse than that. The reported temperature seems to only end in 0.25C or 0.75C, so it has an effective 0.5C resolution. The basic DS1620 resolution is 9 bits, but the Tbolt firmware reduces that to 8 bits since the first step in the high res temp algorithm is to mask off the lower bit. Whatever is going on they are not getting the extra resolution that they think they are, and in fact they are reducing the basic resolution of the chip. The firmware does seem do do some filtering on those values since whenever the reading steps you can see some smoothing going on. A lot of times the temperature value oscillatates around the step point. The filter apparently does not have any hysteresis. The older Thunderbolts produced a nice smooth curve. The high res temperature reading (Bruce says is 12 bit/0.0625C) coupled with the firmware filtering gave temperature curves with microdegree scale resolution. The newer ones clunk around with effectively 0.5C resolution. -------------------------------- But have you actually established this has anything whatsoever to do with the oscillator conditioning and, if not, what effects are you suggesting it has on the conditioned output? Isn't it likely that a temperature sensor adjacent to the RS232 connector is just going to monitor unit temperature for environmental purposes, perhaps, for example, to give the option for flagging up an overheating situation, in which case surely 0.5C resolution is more than adequate and the "clunkiness" isn't really an issue? I agree it's always nice to know what's happening, and why, but I suspect the performance of this sensor has no relevance to the oscillator performance itself. regards Nigel GM8PZR regards Nigel GM8PZR From df6jb at ulrich-bangert.de Thu Feb 5 10:23:44 2009 From: df6jb at ulrich-bangert.de (Ulrich Bangert) Date: Thu, 5 Feb 2009 11:23:44 +0100 Subject: [time-nuts] Tbolt temperature sensor In-Reply-To: Message-ID: Gents, has anyone checked which DALLAS part is REALLY inside the box? I put forward this question because DALLAS has parts in their portfolio that sell as "genuine" 12-bit resolution ones and other ones that sell as 9-bit resolution ones where additional 3 bits of resolution can be used by some "tricks" The software for these two types is NOT the same. Best regards Ulrich Bnagert > -----Ursprungliche Nachricht----- > Von: time-nuts-bounces at febo.com > [mailto:time-nuts-bounces at febo.com] Im Auftrag von GandalfG8 at aol.com > Gesendet: Donnerstag, 5. Februar 2009 11:15 > An: time-nuts at febo.com > Betreff: Re: [time-nuts] Tbolt temperature sensor > > > > In a message dated 05/02/2009 06:25:52 GMT Standard Time, > holrum at hotmail.com > writes: > > Actually, what is apparently going on with the newer > Thunderbolts is worse > than that. The reported temperature seems to only end in > 0.25C or 0.75C, so > it has an effective 0.5C resolution. The basic DS1620 > resolution is 9 > bits, but the Tbolt firmware reduces that to 8 bits since > the first step in the > high res temp algorithm is to mask off the lower bit. > Whatever is going on > they are not getting the extra resolution that they think > they are, and in > fact they are reducing the basic resolution of the chip. > > The firmware does seem do do some filtering on those values > since whenever > the reading steps you can see some smoothing going on. A > lot of times the > temperature value oscillatates around the step point. The > filter apparently > does not have any hysteresis. > > The older Thunderbolts produced a nice smooth curve. The high res > temperature reading (Bruce says is 12 bit/0.0625C) coupled > with the firmware filtering > gave temperature curves with microdegree scale resolution. > The newer ones > clunk around with effectively 0.5C resolution. > > > > -------------------------------- > But have you actually established this has anything > whatsoever to do with > the oscillator conditioning and, if not, what effects are you > suggesting it has > on the conditioned output? > > Isn't it likely that a temperature sensor adjacent to the > RS232 connector is > just going to monitor unit temperature for environmental > purposes, perhaps, > for example, to give the option for flagging up an > overheating situation, in > which case surely 0.5C resolution is more than adequate and > the "clunkiness" > isn't really an issue? > > I agree it's always nice to know what's happening, and why, > but I suspect > the performance of this sensor has no relevance to the > oscillator performance > itself. > > regards > > Nigel > GM8PZR > > regards > > Nigel > GM8PZR > _______________________________________________ > time-nuts mailing list -- time-nuts at febo.com > To unsubscribe, go to > https://www.febo.com/cgi-> bin/mailman/listinfo/time-nuts > and > follow the instructions there. > From bruce.griffiths at xtra.co.nz Thu Feb 5 11:14:24 2009 From: bruce.griffiths at xtra.co.nz (Bruce Griffiths) Date: Fri, 06 Feb 2009 00:14:24 +1300 Subject: [time-nuts] Tbolt temperature sensor In-Reply-To: References: Message-ID: <498ACA10.7000100@xtra.co.nz> Ulrich As in the difference between the DS620 (10, 11, 12, or 13 bit directly with no additional calculation) and the DS1620 (9 bits + 3bits extension via the procedure in the datasheet)? The original post did state that the sensor was a DS1620. Mark also stated in a later post that he had replaced the chip a new DS1620 with no effect on the resolution of the reported temperature. Bruce Ulrich Bangert wrote: > Gents, > > has anyone checked which DALLAS part is REALLY inside the box? > > I put forward this question because DALLAS has parts in their portfolio > that sell as "genuine" 12-bit resolution ones and other ones that sell > as 9-bit resolution ones where additional 3 bits of resolution can be > used by some "tricks" The software for these two types is NOT the same. > > Best regards > Ulrich Bnagert > > >> -----Ursprungliche Nachricht----- >> Von: time-nuts-bounces at febo.com >> [mailto:time-nuts-bounces at febo.com] Im Auftrag von GandalfG8 at aol.com >> Gesendet: Donnerstag, 5. Februar 2009 11:15 >> An: time-nuts at febo.com >> Betreff: Re: [time-nuts] Tbolt temperature sensor >> >> >> >> In a message dated 05/02/2009 06:25:52 GMT Standard Time, >> holrum at hotmail.com >> writes: >> >> Actually, what is apparently going on with the newer >> Thunderbolts is worse >> than that. The reported temperature seems to only end in >> 0.25C or 0.75C, so >> it has an effective 0.5C resolution. The basic DS1620 >> resolution is 9 >> bits, but the Tbolt firmware reduces that to 8 bits since >> the first step in the >> high res temp algorithm is to mask off the lower bit. >> Whatever is going on >> they are not getting the extra resolution that they think >> they are, and in >> fact they are reducing the basic resolution of the chip. >> >> The firmware does seem do do some filtering on those values >> since whenever >> the reading steps you can see some smoothing going on. A >> lot of times the >> temperature value oscillatates around the step point. The >> filter apparently >> does not have any hysteresis. >> >> The older Thunderbolts produced a nice smooth curve. The high res >> temperature reading (Bruce says is 12 bit/0.0625C) coupled >> with the firmware filtering >> gave temperature curves with microdegree scale resolution. >> The newer ones >> clunk around with effectively 0.5C resolution. >> >> >> >> -------------------------------- >> But have you actually established this has anything >> whatsoever to do with >> the oscillator conditioning and, if not, what effects are you >> suggesting it has >> on the conditioned output? >> >> Isn't it likely that a temperature sensor adjacent to the >> RS232 connector is >> just going to monitor unit temperature for environmental >> purposes, perhaps, >> for example, to give the option for flagging up an >> overheating situation, in >> which case surely 0.5C resolution is more than adequate and >> the "clunkiness" >> isn't really an issue? >> >> I agree it's always nice to know what's happening, and why, >> but I suspect >> the performance of this sensor has no relevance to the >> oscillator performance >> itself. >> >> regards >> >> Nigel >> GM8PZR >> >> regards >> >> Nigel >> GM8PZR >> _______________________________________________ >> From df6jb at ulrich-bangert.de Thu Feb 5 12:23:53 2009 From: df6jb at ulrich-bangert.de (Ulrich Bangert) Date: Thu, 5 Feb 2009 13:23:53 +0100 Subject: [time-nuts] Tbolt temperature sensor In-Reply-To: <498ACA10.7000100@xtra.co.nz> Message-ID: Bruce, I have been thinking about the DS18B20 with programmable resolution. Best regards Ulrich > -----Ursprungliche Nachricht----- > Von: time-nuts-bounces at febo.com > [mailto:time-nuts-bounces at febo.com] Im Auftrag von Bruce Griffiths > Gesendet: Donnerstag, 5. Februar 2009 12:14 > An: Discussion of precise time and frequency measurement > Betreff: Re: [time-nuts] Tbolt temperature sensor > > > Ulrich > > As in the difference between the DS620 (10, 11, 12, or 13 bit > directly with no additional calculation) and the DS1620 (9 > bits + 3bits extension via the procedure in the datasheet)? > > The original post did state that the sensor was a DS1620. > Mark also stated in a later post that he had replaced the > chip a new DS1620 with no effect on the resolution of the > reported temperature. > > Bruce > > Ulrich Bangert wrote: > > Gents, > > > > has anyone checked which DALLAS part is REALLY inside the box? > > > > I put forward this question because DALLAS has parts in their > > portfolio that sell as "genuine" 12-bit resolution ones and > other ones > > that sell as 9-bit resolution ones where additional 3 bits of > > resolution can be used by some "tricks" The software for these two > > types is NOT the same. > > > > Best regards > > Ulrich Bnagert > > > > > >> -----Ursprungliche Nachricht----- > >> Von: time-nuts-bounces at febo.com > >> [mailto:time-nuts-bounces at febo.com] Im Auftrag von > GandalfG8 at aol.com > >> Gesendet: Donnerstag, 5. Februar 2009 11:15 > >> An: time-nuts at febo.com > >> Betreff: Re: [time-nuts] Tbolt temperature sensor > >> > >> > >> > >> In a message dated 05/02/2009 06:25:52 GMT Standard Time, > >> holrum at hotmail.com > >> writes: > >> > >> Actually, what is apparently going on with the newer > >> Thunderbolts is worse > >> than that. The reported temperature seems to only end in > >> 0.25C or 0.75C, so > >> it has an effective 0.5C resolution. The basic DS1620 > >> resolution is 9 > >> bits, but the Tbolt firmware reduces that to 8 bits since > >> the first step in the > >> high res temp algorithm is to mask off the lower bit. > >> Whatever is going on > >> they are not getting the extra resolution that they think > >> they are, and in > >> fact they are reducing the basic resolution of the chip. > >> > >> The firmware does seem do do some filtering on those values > >> since whenever > >> the reading steps you can see some smoothing going on. A > >> lot of times the > >> temperature value oscillatates around the step point. The > >> filter apparently > >> does not have any hysteresis. > >> > >> The older Thunderbolts produced a nice smooth curve. The high res > >> temperature reading (Bruce says is 12 bit/0.0625C) coupled > >> with the firmware filtering > >> gave temperature curves with microdegree scale resolution. > >> The newer ones > >> clunk around with effectively 0.5C resolution. > >> > >> > >> > >> -------------------------------- > >> But have you actually established this has anything > >> whatsoever to do with > >> the oscillator conditioning and, if not, what effects are you > >> suggesting it has > >> on the conditioned output? > >> > >> Isn't it likely that a temperature sensor adjacent to the > >> RS232 connector is > >> just going to monitor unit temperature for environmental > >> purposes, perhaps, > >> for example, to give the option for flagging up an > >> overheating situation, in > >> which case surely 0.5C resolution is more than adequate and > >> the "clunkiness" > >> isn't really an issue? > >> > >> I agree it's always nice to know what's happening, and why, > >> but I suspect > >> the performance of this sensor has no relevance to the > >> oscillator performance > >> itself. > >> > >> regards > >> > >> Nigel > >> GM8PZR > >> > >> regards > >> > >> Nigel > >> GM8PZR > >> _______________________________________________ > >> > > > _______________________________________________ > time-nuts mailing list -- time-nuts at febo.com > To unsubscribe, go to > https://www.febo.com/cgi-> bin/mailman/listinfo/time-nuts > and > follow the instructions there. > From didier at cox.net Thu Feb 5 13:15:22 2009 From: didier at cox.net (Didier) Date: Thu, 5 Feb 2009 07:15:22 -0600 Subject: [time-nuts] Tbolt temperature sensor In-Reply-To: <498AAB05.1060501@xtra.co.nz> References: <498AAB05.1060501@xtra.co.nz> Message-ID: <095B10EC7CDB44DB834605C2F3D16C3B@didierhp> > -----Original Message----- > From: time-nuts-bounces at febo.com > [mailto:time-nuts-bounces at febo.com] On Behalf Of Bruce Griffiths > Sent: Thursday, February 05, 2009 3:02 AM > To: Discussion of precise time and frequency measurement > Subject: Re: [time-nuts] Tbolt temperature sensor > > Mark Sims wrote: > > Actually, what is apparently going on with the newer > Thunderbolts is worse than that. The reported temperature > seems to only end in 0.25C or 0.75C, so it has an effective > 0.5C resolution. The basic DS1620 resolution is 9 bits, > but the Tbolt firmware reduces that to 8 bits since the first > step in the high res temp algorithm is to mask off the lower > bit. Whatever is going on they are not getting the extra > resolution that they think they are, and in fact they are > reducing the basic resolution of the chip. > > > > The firmware does seem do do some filtering on those values > since whenever the reading steps you can see some smoothing > going on. A lot of times the temperature value oscillatates > around the step point. The filter apparently does not have > any hysteresis. > > > > The older Thunderbolts produced a nice smooth curve. The > high res temperature reading (Bruce says is 12 bit/0.0625C) > coupled with the firmware filtering gave temperature curves > with microdegree scale resolution. The newer ones clunk > around with effectively 0.5C resolution. > > > Not quite right. > According to the datasheet the resolution with the standard 9 > bit output is actually 0.5C. > The MSBit is actually a sign bit. > Despite what the text may perhaps implies, 1/2 of a LSB is > actually subtracted before combining with the residual count > and other data (see attachment for formula). > The resolution isnt in fact being truncated at all. > An offset of 1/2 LSB is being subtracted from the coarse > reading, this has no effect on the resolution. > > Bruce > I have been casually following this thread because I use the DS1629 in several of my projects and it also has a temperature sensor so I logically assumed (that word again) that the 1620 and 1629 had a similar way to measure temperature. The 1629 has 0.5 degree resolution but higher resolution is available if you are willing to put up with additional work, just like the 1620. However, both devices uses completely different methods to measure temperature: the 1620 uses a bandgap reference and an analog temperature sensor while the 1629 uses a temperature sensistive oscillator and another that is stable. Since the 1629 is also a real time clock that can run from a lithium battery for a long time, I suspect they had to use a lower power scheme for it than the bandgap reference. Interesting Didier KO4BB From tvb at LeapSecond.com Thu Feb 5 15:12:01 2009 From: tvb at LeapSecond.com (Tom Van Baak) Date: Thu, 5 Feb 2009 07:12:01 -0800 Subject: [time-nuts] Tbolt temperature sensor References: Message-ID: <4F750A0E93C74B3FA8386DFFA4CE80C5@pc52> Ah, yes, I see that one also shows the same correlation between oven current (temp) and EFC voltage, but the TI wanders a lot more. What are the large rapid swings; almost like ringing? I remember a few years ago there was a long thread about the effect of changes in oven current on EFC potential due to common ground lines. In your plots the EFC correction seems to match so well with oven current (without clear evidence of the usual thermal lag) could it be that what you're seeing, and what you're correcting, is not resonator temperature at all but simply the voltage offset due to ground currents? One way to tell for sure is to take your raw syst:stat data and plot the normalized difference between oven current and -efc. If you see evidence of time lags then its a good sign that it's temperature. If the two track well, at short and at long periods, then I wonder if it's not temperature but current that's being compensated for. /tvb ----- Original Message ----- From: SAIDJACK at aol.com To: tvb at leapsecond.com Cc: time-nuts at febo.com Sent: Thursday, February 05, 2009 12:59 AM Subject: Re: [time-nuts] Tbolt temperature sensor Hi Tom, yes, they are all GPS locked. here is the same OCXO type without proper temp compensation: http://www.jackson-labs.com/images/gpsstat.htm You can see that now the phase has to change considerably to compensate for the EFC voltage change (because tempco has not been properly established yet). To have the TIC compensate, it needs to detect a phase offset first. This error will be prevented if the tempco already compensates properly for temperature changes. As the proper tempco is established, the phase offsets diminish in magnitude. bye, Said In a message dated 2/4/2009 21:58:18 Pacific Standard Time, tvb at LeapSecond.com writes: In that case, how do you distinguish in the plots between the GPS phase locking reaction to the OCXO starting to go off-frequency and software temperature compensation? How much of the correction in the plots is due to normal phase locking against GPS and how much is due to your temperature sensing and compensation? From SAIDJACK at aol.com Thu Feb 5 17:31:12 2009 From: SAIDJACK at aol.com (SAIDJACK at aol.com) Date: Thu, 5 Feb 2009 12:31:12 EST Subject: [time-nuts] Tbolt temperature sensor Message-ID: Hello Tom, >Ah, yes, I see that one also shows the same correlation between >oven current (temp) and EFC voltage, but the TI wanders a lot >more. What are the large rapid swings; almost like ringing? This is a brand new crystal, and so the unit is learning about it, and the aging is rather high in the first couple of days. The "ringing" is caused by the heater going on in the lab. The phase errors happen because the loop needs them to properly change the EFC voltage (proportional and integral gain). After some more time this unit will perform as well as the three I had sent earlier (without the large phase swings). By the way, I disconnected the GPS antenna last night, and you can now see that compensation continues and the EFC curve still follows the thermal curve. It does look to me like the unit is over-compensating a little bit though; it has not finished fully learning about the OCXO yet. >I remember a few years ago there was a long thread about the >effect of changes in oven current on EFC potential due to >common ground lines. We checked for this in some extensive testing, and it is not an issue, even with oven currents going to 0.5A (this oven only needs about 0.05A). There is no measurable effect due to ground currents causing voltage drops. You can see this by turning temperature compensation completely off, and running a double oven in a thermal chamber. It performs the same even with ground loops removed (by soldering the OCXO can directly to the PCB ground through massive straps etc). There is no change in performance. This does require some tricky PCB layout to work properly though. >In your plots the EFC correction seems to match so well with >oven current (without clear evidence of the usual thermal lag) >could it be that what you're seeing, and what you're correcting, >is not resonator temperature at all but simply the voltage offset >due to ground currents? This is hard to see due to the long time scale, there is a small lag, but way to small to see in 50+ hour plots. BTW: the ground current effect is not an issue in our designs, since it is the total current that we measure and correct for. Ground loops could change the temperature compensation gain up or down, but will be fully measured and compensated for. bye, Said From SAIDJACK at aol.com Thu Feb 5 18:22:13 2009 From: SAIDJACK at aol.com (SAIDJACK at aol.com) Date: Thu, 5 Feb 2009 13:22:13 EST Subject: [time-nuts] Tbolt temperature sensor Message-ID: Hi Tom, I re-connected the antenna to that unit just now, so it is now GPS locked again. Forgot to mention: another reason why ground loop currents for these single oven units are not an issue is that the current changes very little with temperature. In the plots you can see that the OCXO current is typ. less than 50mA, and the variation is typically less than 4mA peak to peak. A 4mA change is a very small change to affect ground-loop induced voltage offset errors across the ground plane, OCXO pin, etc. Nevertheless we do Kelvin sensing to avoid any issues for ovens with higher current variations. bye, Said In a message dated 2/5/2009 09:32:31 Pacific Standard Time, SAIDJACK at aol.com writes: >In your plots the EFC correction seems to match so well with >oven current (without clear evidence of the usual thermal lag) >could it be that what you're seeing, and what you're correcting, >is not resonator temperature at all but simply the voltage offset >due to ground currents? From holrum at hotmail.com Thu Feb 5 18:44:08 2009 From: holrum at hotmail.com (Mark Sims) Date: Thu, 5 Feb 2009 18:44:08 +0000 Subject: [time-nuts] Tbolt temperature sensor In-Reply-To: References: Message-ID: I doubt the temperature sensor is just used for environmental monitoring. If it was, they would most likely have just used the basic 0.5C resolution reading and have been done with it. Instead, they do the high res read routine and then feed that into a filter to get very smooth high resolution (microdegree scale) values. The whole purpose of a GPSDO with an expensive double oven OCXO is to provide extremely high quality holdover performance when GPS signals go away. To achieve this level of performance one needs to compensate for the affects of temperature on the parts of the system outside the oven. 0.5C resolution is not up to the task of maintaining parts per trillion accuracy. There does not appear to be any other temperature sensor in the Tbolt. ------------ Isn't it likely that a temperature sensor adjacent to the RS232 connector is just going to monitor unit temperature for environmental purposes, perhaps, for example, to give the option for flagging up an overheating situation, in which case surely 0.5C resolution is more than adequate and the "clunkiness" isn't really an issue? _________________________________________________________________ Windows Live?: E-mail. Chat. Share. Get more ways to connect. http://windowslive.com/howitworks?ocid=TXT_TAGLM_WL_t2_allup_howitworks_022009 From bruce.griffiths at xtra.co.nz Thu Feb 5 19:16:58 2009 From: bruce.griffiths at xtra.co.nz (Bruce Griffiths) Date: Fri, 06 Feb 2009 08:16:58 +1300 Subject: [time-nuts] Tbolt temperature sensor In-Reply-To: References: Message-ID: <498B3B2A.3070003@xtra.co.nz> Mark Sims wrote: > I doubt the temperature sensor is just used for environmental monitoring. If it was, they would most likely have just used the basic 0.5C resolution reading and have been done with it. Instead, they do the high res read routine and then feed that into a filter to get very smooth high resolution (microdegree scale) values. > > The whole purpose of a GPSDO with an expensive double oven OCXO is to provide extremely high quality holdover performance when GPS signals go away. To achieve this level of performance one needs to compensate for the affects of temperature on the parts of the system outside the oven. 0.5C resolution is not up to the task of maintaining parts per trillion accuracy. There does not appear to be any other temperature sensor in the Tbolt. > > > > ------------ > Isn't it likely that a temperature sensor adjacent to the RS232 connector is > just going to monitor unit temperature for environmental purposes, perhaps, > for example, to give the option for flagging up an overheating situation, in > which case surely 0.5C resolution is more than adequate and the "clunkiness" > isn't really an issue? > Mark Unless there is sufficient noise or dither merely filtering/averaging a large number of temperature readings will not increase the resolution to below 1lsb. Either internal noise or temperature fluctuations with an amplitude of at least 1 lsb or so are necessary. Bruce From tvb at LeapSecond.com Thu Feb 5 19:21:49 2009 From: tvb at LeapSecond.com (Tom Van Baak) Date: Thu, 5 Feb 2009 11:21:49 -0800 Subject: [time-nuts] Tbolt temperature sensor References: Message-ID: The reason I'm asking this is because I can't figure out from your descriptions or from the plots a clear way to distinguish between unintended oven current-induced changes in applied EFC voltage vs. direct temperature-induced changes in OCXO frequency output. It seems both would have the same effect. I wonder if there is a clear test that would tell you one way or the other. Maybe change oven voltage quite suddenly. That should cause a change in oven current. If the measured EFC voltage or the measured frequency also changes suddenly then that might indicate it's not really temperature that's being compensated for. Can I do these tests on my double oven Fury? Or is this something only for the single oven version. /tvb ----- Original Message ----- From: SAIDJACK at aol.com To: time-nuts at febo.com ; tvb at leapsecond.com Sent: Thursday, February 05, 2009 10:22 AM Subject: Re: [time-nuts] Tbolt temperature sensor Hi Tom, I re-connected the antenna to that unit just now, so it is now GPS locked again. Forgot to mention: another reason why ground loop currents for these single oven units are not an issue is that the current changes very little with temperature. In the plots you can see that the OCXO current is typ. less than 50mA, and the variation is typically less than 4mA peak to peak. A 4mA change is a very small change to affect ground-loop induced voltage offset errors across the ground plane, OCXO pin, etc. Nevertheless we do Kelvin sensing to avoid any issues for ovens with higher current variations. bye, Said In a message dated 2/5/2009 09:32:31 Pacific Standard Time, SAIDJACK at aol.com writes: >In your plots the EFC correction seems to match so well with >oven current (without clear evidence of the usual thermal lag) >could it be that what you're seeing, and what you're correcting, >is not resonator temperature at all but simply the voltage offset >due to ground currents? From GandalfG8 at aol.com Thu Feb 5 19:25:19 2009 From: GandalfG8 at aol.com (GandalfG8 at aol.com) Date: Thu, 5 Feb 2009 14:25:19 EST Subject: [time-nuts] Tbolt temperature sensor Message-ID: In a message dated 05/02/2009 18:44:44 GMT Standard Time, holrum at hotmail.com writes: The whole purpose of a GPSDO with an expensive double oven OCXO is to provide extremely high quality holdover performance when GPS signals go away. To achieve this level of performance one needs to compensate for the affects of temperature on the parts of the system outside the oven. 0.5C resolution is not up to the task of maintaining parts per trillion accuracy. There does not appear to be any other temperature sensor in the Tbolt --------------- I'm still not convinced. There's going to be another very important temperature sensor inside the Thunderbolt and that's the one inside the ovens, perhaps even two in a double oven. I understand where you're coming from but, given the apparent lack of evidence to support the sudden deterioration in performance your findings would suggest, does it not seem more likely that in fact this sensor is not involved in the conditioning process? regards Nigel GM8PZR From tvb at LeapSecond.com Thu Feb 5 19:28:59 2009 From: tvb at LeapSecond.com (Tom Van Baak) Date: Thu, 5 Feb 2009 11:28:59 -0800 Subject: [time-nuts] Tbolt temperature sensor References: Message-ID: > The whole purpose of a GPSDO with an expensive double > oven OCXO is to provide extremely high quality holdover > performance when GPS signals go away. To achieve this > level of performance one needs to compensate for the affects > of temperature on the parts of the system outside the oven. Mark, Is this really true? What is the typical tempco of parts outside the oven? Compared to the spec'd tempco of the OCXO itself. In holdover mode, the tempco of the GPS receiver and the phase comparator have no effect. Maybe just the DAC tempco is all that's left? Does anyone know if 58503/Z3801-series has a temperature sensor? /tvb From newell at cei.net Thu Feb 5 19:34:56 2009 From: newell at cei.net (Scott Newell) Date: Thu, 05 Feb 2009 13:34:56 -0600 Subject: [time-nuts] Tbolt temperature sensor In-Reply-To: References: Message-ID: <200902051934.n15JYuMe007858@host22.the-web-host.com> At 12:44 PM 2/5/2009 , Mark Sims wrote: > >I doubt the temperature sensor is just used for environmental monitoring. If it >was, they would most likely have just used the basic 0.5C resolution reading >and have been done with it. Instead, they do the high res read routine and >then feed that into a filter to get very smooth high resolution (microdegree >scale) values. Maybe shoot some cold spray or place a heater up against the Dallas temp sensor chip and see if the EFC voltage moves? Obviously, run this test in holdover mode. -- newell N5TNL From SAIDJACK at aol.com Thu Feb 5 19:54:10 2009 From: SAIDJACK at aol.com (SAIDJACK at aol.com) Date: Thu, 5 Feb 2009 14:54:10 EST Subject: [time-nuts] Tbolt temperature sensor Message-ID: Hello Tom, >The reason I'm asking this is because I can't figure out from >your descriptions or from the plots a clear way to distinguish >between unintended oven current-induced changes in applied >EFC voltage vs. direct temperature-induced changes in OCXO >frequency output. It seems both would have the same effect. As long as all effects are fairly linear over a given temperature range, then it doesn't matter where the thermal sensitivity is coming from. You may even have canceling effects (if for example the oven has a negative tempco, and the DAC/Vref have a positive tempco). So in other words, even the unintended temp-induced EFC changes will be compensated, because we measure absolute current versus absolute EFC voltage when the unit is locked, and calculate the parameters from this measurement, which includes all unintended thermal effects. >I wonder if there is a clear test that would tell you one way or >the other. You could test specifically for the unintended effects: turn off temperature compensation and GPS locking, and put the unit into a thermal chamber and see how it performs. On our double ovens we usually achieve better to significantly overall performance than the OCXO thermal spec itself. This shows that we don't "make things worse" by the DAC, DAC-reference, or grounding thermal effects. Since our PCB can perform better than the double ovens we use, this means the PCB is about 50x to 100x or more better than the performance of the single ovens, since that's the relationship of double/single oven performance. One caveat: since the double ovens are so extremely good, it can actually happen that the electronic compensation could under some circumstances make things worse than if it was just turned-off. This can happen if you operate your oven in a small thermal range (say +/-5C) where the current/temp relationship is nice and linear, and then almost sudden go -50C lower than usual, where the relationship may be parabolic etc. For applications that require large thermal ranges (say fighter aircraft that can descend 30000 feet in mere minutes or less) we actually disable the tempco statistics, and rather measure the unit over a 100C range in a thermal chamber, and establish the best tempco parameter that will work from -25C to +75, and will improve the units performance over the entire temperature range without adding additional error anywhere. In all of our Fury and FireFly GPSDO's the user has the option to put a hard-coded tempco value, and disable statistics gathering. To check for ground loops: solder a wide (say 1/2 inch) copper string from the OCXO case to the power supplies' ground pin so the oven ground current will flow through this instead of the OCXO ground pin. We do not see any change in performance doing this, and with a current clamp you can verify that most of the current is actually flowing through this new ground. >Maybe change oven voltage quite suddenly. That should cause >a change in oven current. If the measured EFC voltage or the >measured frequency also changes suddenly then that might >indicate it's not really temperature that's being compensated for. This would not work, since it would change the oven current for sure, and this would end up causing an immediate EFC voltage change since it would be interpreted as a thermal change (which it is not over the long term of course after the oven settles down). That's why we have a highly accurate, low noise 10.45V regulator on our boards to generate the OCXO voltage, and don't use the raw supply voltage. >Can I do these tests on my double oven Fury? Or is this something >only for the single oven version. You can do these tests on the double oven Fury. The tempco numbers are usually around 10 - 50 for a good double oven, and around 1500 to 4000 for a good single oven. This means the compensation is usually 100 to ~300 times more aggressive on a single than on a double oven unit. You may want to upgrade the firmware to the last release (rev 1.21) for these types of tests. On a typical double oven we thus see very little correlation between OCXO current and EFC voltage locked or unlocked :) bye, Said From djl at montana.com Thu Feb 5 19:56:11 2009 From: djl at montana.com (Don Latham) Date: Thu, 5 Feb 2009 12:56:11 -0700 Subject: [time-nuts] Tbolt temperature sensor References: Message-ID: <001a01c987cb$cb62f220$0700a8c0@OFFICE2> Here's even more fun. Do the OXCO ovens use a proportional servo or bang-bang? In the either case, the thermal time constant is very important, perhaps more so for the latter. Don ----- Original Message ----- From: To: Sent: Thursday, February 05, 2009 12:25 PM Subject: Re: [time-nuts] Tbolt temperature sensor > > In a message dated 05/02/2009 18:44:44 GMT Standard Time, > holrum at hotmail.com > writes: > > The whole purpose of a GPSDO with an expensive double oven OCXO is to > provide extremely high quality holdover performance when GPS signals go > away. To > achieve this level of performance one needs to compensate for the affects > of > temperature on the parts of the system outside the oven. 0.5C resolution > is > not up to the task of maintaining parts per trillion accuracy. There > does > not appear to be any other temperature sensor in the Tbolt > > > --------------- > I'm still not convinced. > > There's going to be another very important temperature sensor inside the > Thunderbolt and that's the one inside the ovens, perhaps even two in a > double > oven. > > I understand where you're coming from but, given the apparent lack of > evidence to support the sudden deterioration in performance your findings > would > suggest, does it not seem more likely that in fact this sensor is not > involved > in the conditioning process? > > regards > > Nigel > GM8PZR > _______________________________________________ > time-nuts mailing list -- time-nuts at febo.com > To unsubscribe, go to > https://www.febo.com/cgi-bin/mailman/listinfo/time-nuts > and follow the instructions there. From brooke at pacific.net Thu Feb 5 19:56:06 2009 From: brooke at pacific.net (Brooke Clarke) Date: Thu, 05 Feb 2009 11:56:06 -0800 Subject: [time-nuts] Tbolt temperature sensor In-Reply-To: References: Message-ID: <498B4456.4000600@pacific.net> Hi Tom: AFAICR the HP/Agilent GPS receivers compensate for temperature and crystal oscillator drift when in holdover. You can see the oscillator drift compensation in the Allan plots on your web page. That's why the HP units have better performance. AFAICR Mr. Allan is on contract with Agilent as a consultant and he has a patent on this concept. Have Fun, Brooke Clarke http://www.prc68.com Tom Van Baak wrote: >> The whole purpose of a GPSDO with an expensive double >> oven OCXO is to provide extremely high quality holdover >> performance when GPS signals go away. To achieve this >> level of performance one needs to compensate for the affects >> of temperature on the parts of the system outside the oven. > > Mark, > > Is this really true? What is the typical tempco of parts outside > the oven? Compared to the spec'd tempco of the OCXO itself. > > In holdover mode, the tempco of the GPS receiver and the > phase comparator have no effect. Maybe just the DAC tempco > is all that's left? > > Does anyone know if 58503/Z3801-series has a temperature > sensor? > > /tvb > > > _______________________________________________ > time-nuts mailing list -- time-nuts at febo.com > To unsubscribe, go to https://www.febo.com/cgi-bin/mailman/listinfo/time-nuts > and follow the instructions there. > > From dave.g0dja at tiscali.co.uk Thu Feb 5 20:04:47 2009 From: dave.g0dja at tiscali.co.uk (Dave Ackrill) Date: Thu, 05 Feb 2009 20:04:47 +0000 Subject: [time-nuts] Tbolt temperature sensor In-Reply-To: <200902051934.n15JYuMe007858@host22.the-web-host.com> References: <200902051934.n15JYuMe007858@host22.the-web-host.com> Message-ID: <498B465F.8080807@tiscali.co.uk> I've been reading the discussions and have to admit that alot of it is out of my depth. My question is this, does any of this mean that I should not use the Thunderbolt to lock the 10MHz OXCO input of my newly bought, and still being put in a box, 10GHz transverter? I guess that it will be light years ahead of the 'stability' of my old wideband FM gear, possibly not quite up to moon bounce but not bad overall. I am getting a bit of deja vu here as I think I probably asked a similar question the last time the thunderbolt was discussed regarding accuracy of the 10MHz output... Dave From gsteinba52 at aol.com Thu Feb 5 20:05:50 2009 From: gsteinba52 at aol.com (gsteinba52 at aol.com) Date: Thu, 05 Feb 2009 15:05:50 -0500 Subject: [time-nuts] Tbolt temperature sensor Message-ID: <8CB55E4468CCA90-990-15C9@webmail-mf10.sysops.aol.com> Why the doubt? Couldn't this just be one input to the Kalman filter? Does the external electronics package require the same temperature regulation as the OCXO? Why not put a cold pack on the DS chip and see if the OXCO output swings as wildly as you expect (I think it would be irresponsibly poor design to control the OXCO outside of the crystal component). Perhaps the early units had microdegree resolution for ambient air, but that doesn't mean the application *requires* microdegree resolution. Jerry Message: 6 Date: Thu, 5 Feb 2009 18:44:08 +0000 From: Mark Sims Subject: [time-nuts] Tbolt temperature sensor To: Message-ID: Content-Type: text/plain; charset="Windows-1252" I doubt the temperature sensor is just used for environmental monitoring. If it was, they would most likely have just used the basic 0.5C resolution reading and have been done with it. Instead, they do the high res read routine and then feed that into a filter to get very smooth high resolution (microdegree scale) values. The whole purpose of a GPSDO with an expensive double oven OCXO is to provide extremely high quality holdover performance when GPS signals go away. To achieve this level of performance one needs to compensate for the affects of temperature on the parts of the system outside the oven. 0.5C resolution is not up to the task of maintaining parts per trillion accuracy. There does not appear to be any other temperature sensor in the Tbolt. ------------ Isn't it likely that a temperature sensor adjacent to the RS232 connector is just going to monitor unit temperature for environmental purposes, perhaps, for example, to give the option for flagging up an overheating situation, in which case surely 0.5C resolution is more than adequate and the "clunkiness" isn't really an issue? From djl at montana.com Thu Feb 5 20:10:51 2009 From: djl at montana.com (Don Latham) Date: Thu, 5 Feb 2009 13:10:51 -0700 Subject: [time-nuts] Tbolt temperature sensor References: <200902051934.n15JYuMe007858@host22.the-web-host.com> <498B465F.8080807@tiscali.co.uk> Message-ID: <002301c987cd$d7f98830$0700a8c0@OFFICE2> Probably Quite adequate for moon-bounce Don ----- Original Message ----- From: "Dave Ackrill" To: "Discussion of precise time and frequency measurement" Sent: Thursday, February 05, 2009 1:04 PM Subject: Re: [time-nuts] Tbolt temperature sensor > I've been reading the discussions and have to admit that alot of it is > out of my depth. > > My question is this, does any of this mean that I should not use the > Thunderbolt to lock the 10MHz OXCO input of my newly bought, and still > being put in a box, 10GHz transverter? > > I guess that it will be light years ahead of the 'stability' of my old > wideband FM gear, possibly not quite up to moon bounce but not bad > overall. > > I am getting a bit of deja vu here as I think I probably asked a similar > question the last time the thunderbolt was discussed regarding accuracy > of the 10MHz output... > > Dave > > _______________________________________________ > time-nuts mailing list -- time-nuts at febo.com > To unsubscribe, go to > https://www.febo.com/cgi-bin/mailman/listinfo/time-nuts > and follow the instructions there. From jra at febo.com Thu Feb 5 20:31:42 2009 From: jra at febo.com (John Ackermann N8UR) Date: Thu, 05 Feb 2009 15:31:42 -0500 Subject: [time-nuts] Tbolt temperature sensor In-Reply-To: <498B465F.8080807@tiscali.co.uk> References: <200902051934.n15JYuMe007858@host22.the-web-host.com> <498B465F.8080807@tiscali.co.uk> Message-ID: <498B4CAE.7090502@febo.com> Here's my view as a ham, not as a time-nut (well, as much not-as-a-time-nut as I can manage): Even at 10 GHz, it's easy to get sufficient frequency accuracy; 1x10e-10 is only a 1 Hertz error, and I doubt anyone is going to be unable to find you if you're even 100 times further off than that. However, depending on the mode you want to use, and the way you are getting from 10 MHz to 10 GHz, short term frequency stability may be important, and phase noise may be even more important. Remember that phase noise increases 20dB for every decade of multiplication, so a noisy oscillator can start to sound pretty fuzzy on 10 GHz CW. Particularly if you're considering mobile/portable operation where a GPSDO has the disadvantage of needing to be told where it is, I think a better choice may well be a low noise free-running OCXO that you calibrate every month or so against the GPSDO in the shack. 73, John ---- Dave Ackrill wrote: > I've been reading the discussions and have to admit that alot of it is > out of my depth. > > My question is this, does any of this mean that I should not use the > Thunderbolt to lock the 10MHz OXCO input of my newly bought, and still > being put in a box, 10GHz transverter? > > I guess that it will be light years ahead of the 'stability' of my old > wideband FM gear, possibly not quite up to moon bounce but not bad overall. > > I am getting a bit of deja vu here as I think I probably asked a similar > question the last time the thunderbolt was discussed regarding accuracy > of the 10MHz output... > > Dave > > _______________________________________________ > time-nuts mailing list -- time-nuts at febo.com > To unsubscribe, go to https://www.febo.com/cgi-bin/mailman/listinfo/time-nuts > and follow the instructions there. From bruce.griffiths at xtra.co.nz Thu Feb 5 20:41:41 2009 From: bruce.griffiths at xtra.co.nz (Bruce Griffiths) Date: Fri, 06 Feb 2009 09:41:41 +1300 Subject: [time-nuts] Tbolt temperature sensor In-Reply-To: <8CB55E4468CCA90-990-15C9@webmail-mf10.sysops.aol.com> References: <8CB55E4468CCA90-990-15C9@webmail-mf10.sysops.aol.com> Message-ID: <498B4F05.5020402@xtra.co.nz> Jerry No one has seriously suggested that the temperature sensor is used to control the oven temperature. It is likely to merely an input to the Kalman filter to allow for compensation of the EFC DAC gain and offset drift as well as the effects of finite OCXO oven thermal gain. Thermal hysteresis will limit the stability of the external EFC DAC after such correction. The question is really is what effect (if any) does the apparent 0.5C quantisation error have on the performance compared to the earlier versions that have greater resolution? If the external DAC has a gain and offset tempco of 10ppm/C and the OCXO EFC range is 1ppm, then 0.5C temperature resolution can potentially cause 5E-12 steps in the fractional OCXO frequency after correction for DAC gain drift. If the DAC experiences wide temperature swings then thermal hysteresis may cause variations of several ppm in the DAC gain and/or offset, in which case there may be little point in using a sensor with a resolution of better than 0.5C. The sensor itself may have significant thermal hysteresis. Bruce gsteinba52 at aol.com wrote: > Why the doubt? > Couldn't this just be one input to the Kalman filter? Does the external > electronics package require the same temperature regulation as the OCXO? > Why not put a cold pack on the DS chip and see if the OXCO output swings as > wildly as you expect (I think it would be irresponsibly poor design to control > the OXCO outside of the crystal component). > Perhaps the early units had microdegree resolution for ambient air, but that > doesn't mean the application *requires* microdegree resolution. > > Jerry > > > > > > Message: 6 > Date: Thu, 5 Feb 2009 18:44:08 +0000 > From: Mark Sims > Subject: [time-nuts] Tbolt temperature sensor > To: > Message-ID: > Content-Type: text/plain; charset="Windows-1252" > > > I doubt the temperature sensor is just used for environmental monitoring. If it > was, they would most likely have just used the basic 0.5C resolution reading > and have been done with it. Instead, they do the high res read routine and > then feed that into a filter to get very smooth high resolution (microdegree > scale) values. > > The whole purpose of a GPSDO with an expensive double oven OCXO is to provide > extremely high quality holdover performance when GPS signals go away. To > achieve this level of performance one needs to compensate for the affects of > temperature on the parts of the system outside the oven. 0.5C resolution is not > up to the task of maintaining parts per trillion accuracy. There does not > appear to be any other temperature sensor in the Tbolt. > > > > ------------ > Isn't it likely that a temperature sensor adjacent to the RS232 connector is > just going to monitor unit temperature for environmental purposes, perhaps, > for example, to give the option for flagging up an overheating situation, in > which case surely 0.5C resolution is more than adequate and the "clunkiness" > isn't really an issue? > > From hmurray at megapathdsl.net Fri Feb 6 00:14:15 2009 From: hmurray at megapathdsl.net (Hal Murray) Date: Thu, 05 Feb 2009 16:14:15 -0800 Subject: [time-nuts] Tbolt temperature sensor In-Reply-To: Message from "Tom Van Baak" of "Thu, 05 Feb 2009 11:28:59 PST." Message-ID: <20090206001416.B183FBCDD@ip-64-139-1-69.sjc.megapath.net> > Does anyone know if 58503/Z3801-series has a temperature sensor? I don't think so. I just scanned the Z3801A manual and couldn't find any mention of it. It sounds handy enough that I expect they wouldn't hide it if they had one. -- These are my opinions, not necessarily my employer's. I hate spam. From holrum at hotmail.com Thu Feb 5 21:17:43 2009 From: holrum at hotmail.com (Mark Sims) Date: Thu, 5 Feb 2009 21:17:43 +0000 Subject: [time-nuts] Tbolt temperature sensor In-Reply-To: References: Message-ID: Well, I did the experiment.... I put the unit into holdover, spritzed the DS1620 temp sensor with cold spray, no immediate change in the DAC voltage was seen. Then about 10 seconds later the DAC reading shot up. After a minute or so, the temp reading was rapidly rising its way to normal and the DAC voltage was dropping, lagging by about 10 seconds. Looks like the temperature definitely sensor affects holdover compensation... And yes Bruce, I agree that the software filtering does not increase the accuracy of the readings. I think they do it to smooth out the control voltage changes that the temp sensor readings lead to. _________________________________________________________________ Windows Live?: Keep your life in sync. http://windowslive.com/howitworks?ocid=TXT_TAGLM_WL_t1_allup_howitworks_022009 From bruce.griffiths at xtra.co.nz Fri Feb 6 01:00:59 2009 From: bruce.griffiths at xtra.co.nz (Bruce Griffiths) Date: Fri, 06 Feb 2009 14:00:59 +1300 Subject: [time-nuts] Tbolt temperature sensor In-Reply-To: References: Message-ID: <498B8BCB.9000201@xtra.co.nz> Mark Sims wrote: > Well, I did the experiment.... I put the unit into holdover, spritzed the DS1620 temp sensor with cold spray, no immediate change in the DAC voltage was seen. Then about 10 seconds later the DAC reading shot up. After a minute or so, the temp reading was rapidly rising its way to normal and the DAC voltage was dropping, lagging by about 10 seconds. > > Looks like the temperature definitely sensor affects holdover compensation... > > And yes Bruce, I agree that the software filtering does not increase the accuracy of the readings. I think they do it to smooth out the control voltage changes that the temp sensor readings lead to. > Mark Smoothing will not occur unless there is some dither in the system. Small temperature fluctuations will suffice as will sufficient noise in the temperature measurement. Bruce From GandalfG8 at aol.com Fri Feb 6 01:23:06 2009 From: GandalfG8 at aol.com (GandalfG8 at aol.com) Date: Thu, 5 Feb 2009 20:23:06 EST Subject: [time-nuts] Tbolt temperature sensor Message-ID: In a message dated 06/02/2009 00:51:21 GMT Standard Time, holrum at hotmail.com writes: Well, I did the experiment.... I put the unit into holdover, spritzed the DS1620 temp sensor with cold spray, no immediate change in the DAC voltage was seen. Then about 10 seconds later the DAC reading shot up. After a minute or so, the temp reading was rapidly rising its way to normal and the DAC voltage was dropping, lagging by about 10 seconds. Looks like the temperature definitely sensor affects holdover compensation... ----------------- Probably can't argue with that then:-) Another thought I had earlier, is it possible perhaps that Trimble were running out of space for a firmware upgrade and cut back on the complexity of this routine having decided that precision of ambient temperature measurement was sufficiently of a "secondary" nature not to warrant the extra code they'd used when more space was available? regards Nigel GM8PZR From stanley_reynolds at yahoo.com Fri Feb 6 18:09:34 2009 From: stanley_reynolds at yahoo.com (Stanley Reynolds) Date: Fri, 6 Feb 2009 10:09:34 -0800 (PST) Subject: [time-nuts] Efratom Modular System 8008-400-8 Message-ID: <643157.33301.qm@web30301.mail.mud.yahoo.com> This is and old ~1987 Rubidium Standard which has a calibrate module "MDC". In the past I have attempted to calibrate but has always failed with the error led turned on. I have used various GPSDOs as primary standards with these attempts. I was able to calibrate with a FTS-5030 Cesium standard last night. My guess is the primary standard must need a certain stability that the GPSDOs lack. Does anyone have experience with this ? Stanley From stanley_reynolds at yahoo.com Fri Feb 6 19:32:33 2009 From: stanley_reynolds at yahoo.com (Stanley Reynolds) Date: Fri, 6 Feb 2009 11:32:33 -0800 (PST) Subject: [time-nuts] Fw: Efratom Modular System 8008-400-8 Message-ID: <562083.92806.qm@web30301.mail.mud.yahoo.com> ----- Forwarded Message ---- From: Stanley Reynolds To: time nuts Cc: stanley_reynolds at yahoo.com Sent: Friday, February 6, 2009 12:09:34 PM Subject: Efratom Modular System 8008-400-8 This is and old ~1987 Rubidium Standard which has a calibrate module "MDC". In the past I have attempted to calibrate but has always failed with the error led turned on. I have used various GPSDOs as primary standards with these attempts. I was able to calibrate with a FTS-5030 Cesium standard last night. My guess is the primary standard must need a certain stability that the GPSDOs lack. Does anyone have experience with this ? Stanley From holrum at hotmail.com Fri Feb 6 02:07:16 2009 From: holrum at hotmail.com (Mark Sims) Date: Fri, 6 Feb 2009 02:07:16 +0000 Subject: [time-nuts] Tbolt temperature sensor In-Reply-To: References: Message-ID: The firmware on the old and new units reports the same revision levels and dates. I am fairly certain that whatever is going on is in the DS1620 chip. Also, if they were going to remove the high res temp code they probably would not have left in the "subtract 0.25C" step and the software filter. I have contacted Maxim to see if they know of any changes in the DS1620 chip between 2003 and 2005. _________________________________________________________________ Windows Live?: E-mail. Chat. Share. Get more ways to connect. http://windowslive.com/howitworks?ocid=TXT_TAGLM_WL_t2_allup_howitworks_022009 From holrum at hotmail.com Fri Feb 6 02:52:27 2009 From: holrum at hotmail.com (Mark Sims) Date: Fri, 6 Feb 2009 02:52:27 +0000 Subject: [time-nuts] Tbolt temperature sensor In-Reply-To: References: Message-ID: Some more (possibly irrelevant) information in the DS1620 temperature sensor: The units that I have checked that give the good temperature curve all have rev D1 DS1620 chips. The ones with the flat line temperatures all have Rev E2 DS1620 chips. The chip rev is the last two characters of the middle line on the package... _________________________________________________________________ Windows Live?: Keep your life in sync. http://windowslive.com/explore?ocid=TXT_TAGLM_WL_t1_allup_explore_022009 From holrum at hotmail.com Sat Feb 7 04:01:51 2009 From: holrum at hotmail.com (Mark Sims) Date: Sat, 7 Feb 2009 04:01:51 +0000 Subject: [time-nuts] Tbolt temperature sensor In-Reply-To: References: Message-ID: I got some info back from Maxim on the DS1620. The rev D chips use the dual oscillator design described in the data sheet. The rev E chips are based on a temperature compensated bandgap design. The high resolution read routine is based upon reading counter values and residuals... it's beginning to look like the newer chips don't do it... I've got more questions to ask Maxim... _________________________________________________________________ Windows Live?: Keep your life in sync. http://windowslive.com/explore?ocid=TXT_TAGLM_WL_t1_allup_explore_022009 From holrum at hotmail.com Sat Feb 7 05:24:10 2009 From: holrum at hotmail.com (Mark Sims) Date: Sat, 7 Feb 2009 05:24:10 +0000 Subject: [time-nuts] A real time-nut In-Reply-To: References: Message-ID: The lady just down the road from me just turned 110 years old... and has it more together than a lot of 70 year olds... _________________________________________________________________ Windows Live?: Keep your life in sync. http://windowslive.com/howitworks?ocid=TXT_TAGLM_WL_t1_allup_howitworks_022009 From df6jb at ulrich-bangert.de Sat Feb 7 09:49:50 2009 From: df6jb at ulrich-bangert.de (Ulrich Bangert) Date: Sat, 7 Feb 2009 10:49:50 +0100 Subject: [time-nuts] Tbolt temperature sensor In-Reply-To: Message-ID: <87211EDB401144E1B638B7BD6CEE51B3@athlon> Mark, in this case the DS1620 may perhaps be exchanged against a DS18S20 which definitely supports the high res reading. Bought chips & programmed for them at the end of last year. Best regards Ulrich Bangert > -----Urspr?ngliche Nachricht----- > Von: time-nuts-bounces at febo.com > [mailto:time-nuts-bounces at febo.com] Im Auftrag von Mark Sims > Gesendet: Samstag, 7. Februar 2009 05:02 > An: time-nuts at febo.com > Betreff: [time-nuts] Tbolt temperature sensor > > > > I got some info back from Maxim on the DS1620. The rev D > chips use the dual oscillator design described in the data > sheet. The rev E chips are based on a temperature > compensated bandgap design. > > The high resolution read routine is based upon reading > counter values and residuals... it's beginning to look like > the newer chips don't do it... I've got more questions to > ask Maxim... > _________________________________________________________________ > Windows Live?: Keep your life in sync. > http://windowslive.com/explore?ocid=TXT_TAGLM_WL_t1_allup_expl ore_022009 _______________________________________________ time-nuts mailing list -- time-nuts at febo.com To unsubscribe, go to https://www.febo.com/cgi-bin/mailman/listinfo/time-nuts and follow the instructions there. From wb6bnq at cox.net Sat Feb 7 09:59:36 2009 From: wb6bnq at cox.net (WB6BNQ) Date: Sat, 07 Feb 2009 01:59:36 -0800 Subject: [time-nuts] Tbolt temperature sensor References: <87211EDB401144E1B638B7BD6CEE51B3@athlon> Message-ID: <498D5B88.69ECFC9@cox.net> Ulrich, The problem is the software structure is different for the DS18S20. So the Tbolt would not know what to do with it. Bill....WB6BNQ Ulrich Bangert wrote: > Mark, > > in this case the DS1620 may perhaps be exchanged against a DS18S20 which > definitely supports the high res reading. Bought chips & programmed for > them at the end of last year. > > Best regards > Ulrich Bangert > > > -----Urspr?ngliche Nachricht----- > > Von: time-nuts-bounces at febo.com > > [mailto:time-nuts-bounces at febo.com] Im Auftrag von Mark Sims > > Gesendet: Samstag, 7. Februar 2009 05:02 > > An: time-nuts at febo.com > > Betreff: [time-nuts] Tbolt temperature sensor > > > > > > > > I got some info back from Maxim on the DS1620. The rev D > > chips use the dual oscillator design described in the data > > sheet. The rev E chips are based on a temperature > > compensated bandgap design. > > > > The high resolution read routine is based upon reading > > counter values and residuals... it's beginning to look like > > the newer chips don't do it... I've got more questions to > > ask Maxim... > > _________________________________________________________________ > > Windows Live?: Keep your life in sync. > > http://windowslive.com/explore?ocid=TXT_TAGLM_WL_t1_allup_expl > ore_022009 > _______________________________________________ > time-nuts mailing list -- time-nuts at febo.com > To unsubscribe, go to > https://www.febo.com/cgi-bin/mailman/listinfo/time-nuts > and follow the instructions there. > > _______________________________________________ > time-nuts mailing list -- time-nuts at febo.com > To unsubscribe, go to https://www.febo.com/cgi-bin/mailman/listinfo/time-nuts > and follow the instructions there. From bruce.griffiths at xtra.co.nz Sat Feb 7 10:03:09 2009 From: bruce.griffiths at xtra.co.nz (Bruce Griffiths) Date: Sat, 07 Feb 2009 23:03:09 +1300 Subject: [time-nuts] Tbolt temperature sensor In-Reply-To: <87211EDB401144E1B638B7BD6CEE51B3@athlon> References: <87211EDB401144E1B638B7BD6CEE51B3@athlon> Message-ID: <498D5C5D.3000003@xtra.co.nz> Ulrich If and only if it can be shown that the thermostat function output pins (Thigh, Tlow, Tcom) of the DS1620 aren't used. Bruce Ulrich Bangert wrote: > Mark, > > in this case the DS1620 may perhaps be exchanged against a DS18S20 which > definitely supports the high res reading. Bought chips & programmed for > them at the end of last year. > > Best regards > Ulrich Bangert > > >> -----Urspr?ngliche Nachricht----- >> Von: time-nuts-bounces at febo.com >> [mailto:time-nuts-bounces at febo.com] Im Auftrag von Mark Sims >> Gesendet: Samstag, 7. Februar 2009 05:02 >> An: time-nuts at febo.com >> Betreff: [time-nuts] Tbolt temperature sensor >> >> >> >> I got some info back from Maxim on the DS1620. The rev D >> chips use the dual oscillator design described in the data >> sheet. The rev E chips are based on a temperature >> compensated bandgap design. >> >> The high resolution read routine is based upon reading >> counter values and residuals... it's beginning to look like >> the newer chips don't do it... I've got more questions to >> ask Maxim... >> _________________________________________________________________ >> Windows Live?: Keep your life in sync. >> http://windowslive.com/explore?ocid=TXT_TAGLM_WL_t1_allup_expl >> > ore_022009 > _______________________________________________ > time-nuts mailing list -- time-nuts at febo.com > To unsubscribe, go to > https://www.febo.com/cgi-bin/mailman/listinfo/time-nuts > and follow the instructions there. > > > _______________________________________________ > time-nuts mailing list -- time-nuts at febo.com > To unsubscribe, go to https://www.febo.com/cgi-bin/mailman/listinfo/time-nuts > and follow the instructions there. > > From bruce.griffiths at xtra.co.nz Sat Feb 7 10:14:33 2009 From: bruce.griffiths at xtra.co.nz (Bruce Griffiths) Date: Sat, 07 Feb 2009 23:14:33 +1300 Subject: [time-nuts] Tbolt temperature sensor In-Reply-To: <498D5B88.69ECFC9@cox.net> References: <87211EDB401144E1B638B7BD6CEE51B3@athlon> <498D5B88.69ECFC9@cox.net> Message-ID: <498D5F09.6000400@xtra.co.nz> Perhaps the solution is to use a single chip microprocessor with a high resolution temperature sensor to emulate the DS1620 ? WB6BNQ wrote: > Ulrich, > > The problem is the software structure is different for the DS18S20. So the Tbolt > would not know what to do with it. > > Bill....WB6BNQ > > > Ulrich Bangert wrote: > > >> Mark, >> >> in this case the DS1620 may perhaps be exchanged against a DS18S20 which >> definitely supports the high res reading. Bought chips & programmed for >> them at the end of last year. >> >> Best regards >> Ulrich Bangert >> >> >>> -----Urspr?ngliche Nachricht----- >>> Von: time-nuts-bounces at febo.com >>> [mailto:time-nuts-bounces at febo.com] Im Auftrag von Mark Sims >>> Gesendet: Samstag, 7. Februar 2009 05:02 >>> An: time-nuts at febo.com >>> Betreff: [time-nuts] Tbolt temperature sensor >>> >>> >>> >>> I got some info back from Maxim on the DS1620. The rev D >>> chips use the dual oscillator design described in the data >>> sheet. The rev E chips are based on a temperature >>> compensated bandgap design. >>> >>> The high resolution read routine is based upon reading >>> counter values and residuals... it's beginning to look like >>> the newer chips don't do it... I've got more questions to >>> ask Maxim... >>> _________________________________________________________________ >>> Windows Live?: Keep your life in sync. >>> http://windowslive.com/explore?ocid=TXT_TAGLM_WL_t1_allup_expl >>> >> ore_022009 >> _______________________________________________ >> time-nuts mailing list -- time-nuts at febo.com >> To unsubscribe, go to >> https://www.febo.com/cgi-bin/mailman/listinfo/time-nuts >> and follow the instructions there. >> >> _______________________________________________ >> time-nuts mailing list -- time-nuts at febo.com >> To unsubscribe, go to https://www.febo.com/cgi-bin/mailman/listinfo/time-nuts >> and follow the instructions there. >> > > > _______________________________________________ > time-nuts mailing list -- time-nuts at febo.com > To unsubscribe, go to https://www.febo.com/cgi-bin/mailman/listinfo/time-nuts > and follow the instructions there. > > From magnus at rubidium.dyndns.org Sat Feb 7 12:49:23 2009 From: magnus at rubidium.dyndns.org (Magnus Danielson) Date: Sat, 07 Feb 2009 13:49:23 +0100 Subject: [time-nuts] A real time-nut In-Reply-To: References: Message-ID: <498D8353.3080603@rubidium.dyndns.org> Mark Sims skrev: > The lady just down the road from me just turned 110 years old... and has it more together than a lot of 70 year olds... So you don't think your crystal therapy will work? :) Cheers, Magnus From wje at quackers.net Sun Feb 8 14:28:47 2009 From: wje at quackers.net (wje) Date: Sun, 08 Feb 2009 09:28:47 -0500 Subject: [time-nuts] Datron 4910/11/12 LTZ1000 Reference Board (finally) Message-ID: <498EEC1F.4080607@quackers.net> I've finished retracing the LTZ ref board I promised months ago. While it somewhat resembles the LTZ1000 data sheet example, it differs significantly in a number of details. The heater control is far more complex, using a PWM scheme. Take the schematic with a bit of a grain of salt... tracing it is difficult because there is no access to the back of the board, at least not without more surgery than I'm willing to do on my secondary reference! In particular, the internal resistor connections in the Vishay precision resistor network are just guesses. If anyone has any info about Vishay TO5 canned precision networks from the 80's, please share. I'm confident about the heater side circuitry, but less so about the output side. You will also need to refer to the Datron cell schematics already posted, since significant portions of the heater control as well as the output buffer amp are on that board. If anyone notices any highly questionable topology, let me know and I'll try to confirm it. I haven't spent the time to do much of a sanity check myself. One final note - although not apparent from the schematic, the ref board itself makes extensive use of guarding. This is critical for ultra-precision refs, since even picoamps of leakage in the wrong place can easily lead to multi-PPM errors. I would estimate that fully 50% of the total trace length on the board is guard traces. As a bonus, I've also scanned the 10V reference from my Solartron 7081 8.5 digit voltmeter. This is an HP 3458 class voltmeter. The interesting (and amazing) thing about this reference is that it does NOT use an LTZ1000, or even an ovenized reference. It uses a 1N829 Zener with some clever linear and non-linear temp compensation. My meter routinely achieves a 1-year drift of less than 1ppm compared against my (calibrated, NIST-traceable) Datron ref. They are all on my FTP site. Look for DatronRef.PDF, SolartronReadme.txt, SolartronRef-1.PDF, and SolartronRef-2.PDF. FTP: ftp ftp.quackers.net login: ftp at quackers.net password: or via a browser: ftp://ftp at quackers.net:ftp at ftp.quackers.net/ Enjoy. -- Bill Ezell ---------- They said 'Windows or better' so I used Linux. From holrum at hotmail.com Sun Feb 8 17:05:32 2009 From: holrum at hotmail.com (Mark Sims) Date: Sun, 8 Feb 2009 17:05:32 +0000 Subject: [time-nuts] Thunderbolt disciplining process In-Reply-To: References: Message-ID: Here is nice screen shot of a Lady Heather run that John Miles did. You can infer quite a bit about the Thunderbolt disciplining process. It appears that the tbolt basic disciplining action is to vary the DAC voltage to keep minimize the PPS error. Every change in the PPS error curve gets immediately mirrored in the DAC curve. You can also see that the disciplining processes varies the DAC voltage with temperature. The envelope of the DAC voltage curve mirrors the temperature curve, but delayed around 45 seconds (huge changes in the temperature curve get to the DAC in around 10 seconds). The later model tbolts with the rev E2 DS1620 temperature chip do not produce that nice high resolution temperature curve and the envelope of the DAC voltage curve is much flatter. It does not wander above/below the plot center line. I think one can conclude that the earlier production tbolts have better disciplining and holdover performance than the later ones. If I can find an earlier rev DS1620 chip and interesting test would be to run late production tbolt in holdover mode for 24 hours, record the oscillator drift, change the DS1620 chip, and run the test again. My bet is you would see a significantly smaller holdover error. _________________________________________________________________ Windows Live?: Keep your life in sync. http://windowslive.com/howitworks?ocid=TXT_TAGLM_WL_t1_allup_howitworks_022009 -------------- next part -------------- A non-text attachment was scrubbed... Name: ex2.gif Type: image/gif Size: 56086 bytes Desc: not available Url : http://www.febo.com/pipermail/time-nuts/attachments/20090208/8f615fbc/attachment-0001.gif From newell at cei.net Sun Feb 8 17:49:59 2009 From: newell at cei.net (Scott Newell) Date: Sun, 08 Feb 2009 11:49:59 -0600 Subject: [time-nuts] Thunderbolt disciplining process In-Reply-To: References: Message-ID: <200902081750.n18Ho5w1027365@host22.the-web-host.com> At 11:05 AM 2/8/2009, Mark Sims wrote: >If I can find an earlier rev DS1620 chip and interesting test would >be to run late production tbolt in holdover mode for 24 >hours, record the oscillator drift, change the DS1620 chip, and >run the test again. My bet is you would see a significantly smaller >holdover error. What package? I may have a '1620 sample in my junk box from back when they were still Dallas Semi. -- newell N5TNL From bruce.griffiths at xtra.co.nz Sun Feb 8 19:25:22 2009 From: bruce.griffiths at xtra.co.nz (Bruce Griffiths) Date: Mon, 09 Feb 2009 08:25:22 +1300 Subject: [time-nuts] Datron 4910/11/12 LTZ1000 Reference Board (finally) In-Reply-To: <498EEC1F.4080607@quackers.net> References: <498EEC1F.4080607@quackers.net> Message-ID: <498F31A2.9040309@xtra.co.nz> The internal details shown for the LTZ1000 are incorrect (see the latest datasheet where it has been corrected). The parasitic diode connections are inaccurate. Bruce wje wrote: > I've finished retracing the LTZ ref board I promised months ago. While > it somewhat resembles the LTZ1000 data sheet example, it differs > significantly in a number of details. The heater control is far more > complex, using a PWM scheme. Take the schematic with a bit of a grain of > salt... tracing it is difficult because there is no access to the back > of the board, at least not without more surgery than I'm willing to do > on my secondary reference! In particular, the internal resistor > connections in the Vishay precision resistor network are just guesses. > If anyone has any info about Vishay TO5 canned precision networks from > the 80's, please share. I'm confident about the heater side circuitry, > but less so about the output side. You will also need to refer to the > Datron cell schematics already posted, since significant portions of the > heater control as well as the output buffer amp are on that board. If > anyone notices any highly questionable topology, let me know and I'll > try to confirm it. I haven't spent the time to do much of a sanity check > myself. > > One final note - although not apparent from the schematic, the ref board > itself makes extensive use of guarding. This is critical for > ultra-precision refs, since even picoamps of leakage in the wrong place > can easily lead to multi-PPM errors. I would estimate that fully 50% of > the total trace length on the board is guard traces. > > As a bonus, I've also scanned the 10V reference from my Solartron 7081 > 8.5 digit voltmeter. This is an HP 3458 class voltmeter. The interesting > (and amazing) thing about this reference is that it does NOT use an > LTZ1000, or even an ovenized reference. It uses a 1N829 Zener with some > clever linear and non-linear temp compensation. My meter routinely > achieves a 1-year drift of less than 1ppm compared against my > (calibrated, NIST-traceable) Datron ref. > > They are all on my FTP site. Look for DatronRef.PDF, > SolartronReadme.txt, SolartronRef-1.PDF, and SolartronRef-2.PDF. > > FTP: > > ftp ftp.quackers.net > login: ftp at quackers.net > password: > > or via a browser: > > ftp://ftp at quackers.net:ftp at ftp.quackers.net/ > > Enjoy. > > From holrum at hotmail.com Sun Feb 8 18:00:28 2009 From: holrum at hotmail.com (Mark Sims) Date: Sun, 8 Feb 2009 18:00:28 +0000 Subject: [time-nuts] Thunderbolt disciplining process In-Reply-To: References: Message-ID: The tbolt uses the DS1620S chip in the SOIC-8 package. The DIP package has quite a bit of thermal mass and would probably not work as well. The chip revision is the last two characters of the date code line (second line on the SOIC chips). I am looking for a rev D chip. ------------------ What package? I may have a '1620 sample in my junk box from back when they were still Dallas Semi. _________________________________________________________________ Windows Live?: Keep your life in sync. http://windowslive.com/explore?ocid=TXT_TAGLM_WL_t1_allup_explore_022009 From w1ksz at earthlink.net Sun Feb 8 19:53:16 2009 From: w1ksz at earthlink.net (Richard W. Solomon) Date: Sun, 8 Feb 2009 14:53:16 -0500 (EST) Subject: [time-nuts] Thunderbolt disciplining process Message-ID: <23619963.1234122796972.JavaMail.root@mswamui-backed.atl.sa.earthlink.net> Where does one find this chip ? Those little letters are hard to read !! 73, Dick, W1KSZ -----Original Message----- >From: Mark Sims >Sent: Feb 8, 2009 1:00 PM >To: time-nuts at febo.com >Subject: [time-nuts] Thunderbolt disciplining process > > >The tbolt uses the DS1620S chip in the SOIC-8 package. The DIP package has quite a bit of thermal mass and would probably not work as well. The chip revision is the last two characters of the date code line (second line on the SOIC chips). I am looking for a rev D chip. > >------------------ > >What package? I may have a '1620 sample in my junk box from back >when they were still Dallas Semi. > > > >_________________________________________________________________ >Windows Live?: Keep your life in sync. >http://windowslive.com/explore?ocid=TXT_TAGLM_WL_t1_allup_explore_022009 >_______________________________________________ >time-nuts mailing list -- time-nuts at febo.com >To unsubscribe, go to https://www.febo.com/cgi-bin/mailman/listinfo/time-nuts >and follow the instructions there. From brooke at pacific.net Sun Feb 8 20:17:03 2009 From: brooke at pacific.net (Brooke Clarke) Date: Sun, 8 Feb 2009 12:17:03 -0800 Subject: [time-nuts] Thunderbolt disciplining process Message-ID: <000601c98a2a$3fd6efc0$6501a8c0@brooke3i1qaluo> Hi Dick: It's U9 listed in the IC table at: http://www.prc68.com/I/ThunderBolt.shtml#IC which you can also see in the lower left corner the board drawing at: http://www.prc68.com/I/ThunderBolt.shtml#Dwg This is a pdf drawing so you can use the "+" button a number of times. Have Fun, Brooke Clarke http://www.prc68.com Richard W. Solomon wrote: > Where does one find this chip ? Those little letters are hard to read !! > > 73, Dick, W1KSZ > > -----Original Message----- >> From: Mark Sims >> Sent: Feb 8, 2009 1:00 PM >> To: time-nuts at febo.com >> Subject: [time-nuts] Thunderbolt disciplining process >> >> >> The tbolt uses the DS1620S chip in the SOIC-8 package. The DIP package has quite a bit of thermal mass and would probably not work as well. The chip revision is the last two characters of the date code line (second line on the SOIC chips). I am looking for a rev D chip. >> >> ------------------ >> >> What package? I may have a '1620 sample in my junk box from back >> when they were still Dallas Semi. >> >> >> >> _________________________________________________________________ >> Windows LiveT: Keep your life in sync. >> http://windowslive.com/explore?ocid=TXT_TAGLM_WL_t1_allup_explore_022009 >> _______________________________________________ >> time-nuts mailing list -- time-nuts at febo.com >> To unsubscribe, go to https://www.febo.com/cgi-bin/mailman/listinfo/time-nuts >> and follow the instructions there. > > > _______________________________________________ > time-nuts mailing list -- time-nuts at febo.com > To unsubscribe, go to https://www.febo.com/cgi-bin/mailman/listinfo/time-nuts > and follow the instructions there. From newell at cei.net Sun Feb 8 20:24:10 2009 From: newell at cei.net (Scott Newell) Date: Sun, 08 Feb 2009 14:24:10 -0600 Subject: [time-nuts] Thunderbolt disciplining process In-Reply-To: References: Message-ID: <200902082024.n18KOGDI024779@host22.the-web-host.com> At 12:00 PM 2/8/2009, Mark Sims wrote: >The tbolt uses the DS1620S chip in the SOIC-8 package. The DIP >package has quite a bit of thermal mass and would probably not work >as well. The chip revision is the last two characters of the date >code line (second line on the SOIC chips). I am looking for a rev D chip. I have a DS1620S (looks wider than an SOIC-8, I think it's 208 mil), date coded 9614C2. Any help? (I'm located in the US.) -- newell N5TNL From pvince at theiet.org Sun Feb 8 20:50:42 2009 From: pvince at theiet.org (Peter Vince) Date: Sun, 08 Feb 2009 20:50:42 +0000 Subject: [time-nuts] New Scientist article on optical clocks Message-ID: <39627.1234126242@uk2.net> The 7th of February edition of "New Scientist" has an article talking, in general terms, about how time-keeping has improved, and how optical clocks are set to supersede Caesium with accuracies of parts in 10^18. The author, a freelance writer, was reporting his talk with Patrick Gill, the head "Time Lord" at the British "National Physical Laboratory". The complete article can be read online on the New Scientist website: http://www.newscientist.com/article/mg20126941.900-a-clock-more-accurate-than-time-itself.html?full=true (this tinyurl version might get through the system better: http://tinyurl.com/bt5uds ) Enjoy! Peter From tvb at LeapSecond.com Sun Feb 8 22:52:56 2009 From: tvb at LeapSecond.com (Tom Van Baak) Date: Sun, 8 Feb 2009 14:52:56 -0800 Subject: [time-nuts] Thunderbolt disciplining process References: Message-ID: <25FFDB6F0DDC4D54B5E234641E6ABE81@pc52> > It appears that the tbolt basic disciplining action is to vary > the DAC voltage to keep minimize the PPS error. Every Is there a GPSDO that doesn't work this way? > You can also see that the disciplining processes varies the > DAC voltage with temperature. The envelope of the DAC > voltage curve mirrors the temperature curve, but delayed > around 45 seconds (huge changes in the temperature > curve get to the DAC in around 10 seconds). Mark, is there a way to tell if the voltage envelope you see is simply the result of the OCXO frequency changing in response to ambient temperature change? Or is your thought that somehow the OCXO is immune from ambient temperature and the onboard temperature sensor is doing all the work to change the DAC? Do you have parallel runs with both style temp chips? Also are these free-run or GPS locked runs? Maybe the LH plot you attached got cut off; what is the x-axis scale for this plot? How many seconds (or hours?) wide. /tvb From tvb at LeapSecond.com Sun Feb 8 23:06:14 2009 From: tvb at LeapSecond.com (Tom Van Baak) Date: Sun, 8 Feb 2009 15:06:14 -0800 Subject: [time-nuts] Tbolt temperature sensor References: Message-ID: <476481AFC28A4EE1A04D0FBDB468B633@pc52> >>The reason I'm asking this is because I can't figure out from >>your descriptions or from the plots a clear way to distinguish >>between unintended oven current-induced changes in applied >>EFC voltage vs. direct temperature-induced changes in OCXO >>frequency output. It seems both would have the same effect. > > As long as all effects are fairly linear over a given temperature > range, then it doesn't matter where the thermal sensitivity is > coming from. I was thinking it might matter to an engineer. It's in our nature to try to understand why something works the way it does. > You may even have canceling effects (if for example the oven > has a negative tempco, and the DAC/Vref have a positive tempco). > So in other words, even the unintended temp-induced EFC > changes will be compensated, because we measure absolute > current versus absolute EFC voltage when the unit is locked, > and calculate the parameters from this measurement, which > includes all unintended thermal effects. I still don't like unintended consequences. Did anyone try the experiment to move your OCXO off-board by a foot of cable? That might help isolate the root cause of the effect that you're compensating for. Also, do you know if all single/double OCXO show the same correlation of frequency output change vs. oven current consumption change? This might be relevant for anyone taking a Fury board and using their own external oscillator. Rick - are there similar effects to what Said sees in his board on the 10811? Or is this one reason the 10811 has separate power supplies for the oscillator and oven circuits? /tvb From magnus at rubidium.dyndns.org Sun Feb 8 23:06:58 2009 From: magnus at rubidium.dyndns.org (Magnus Danielson) Date: Mon, 09 Feb 2009 00:06:58 +0100 Subject: [time-nuts] Thunderbolt disciplining process In-Reply-To: <25FFDB6F0DDC4D54B5E234641E6ABE81@pc52> References: <25FFDB6F0DDC4D54B5E234641E6ABE81@pc52> Message-ID: <498F6592.40002@rubidium.dyndns.org> Tom Van Baak skrev: >> It appears that the tbolt basic disciplining action is to vary >> the DAC voltage to keep minimize the PPS error. Every > > Is there a GPSDO that doesn't work this way? Yes. There exists GPSDOs which aims for frequency stability and traceability over PPS accuracy. It's fine and fits their purposes for cal-labs. Cheers, Magnus From SAIDJACK at aol.com Sun Feb 8 23:49:15 2009 From: SAIDJACK at aol.com (SAIDJACK at aol.com) Date: Sun, 8 Feb 2009 18:49:15 EST Subject: [time-nuts] Tbolt temperature sensor Message-ID: Hello Tom, it's pretty straight forward to calculate the effects of ground-loop due to cable loss; we can calculate this based on the impedance of the wire. From holrum at hotmail.com Mon Feb 9 00:18:05 2009 From: holrum at hotmail.com (Mark Sims) Date: Mon, 9 Feb 2009 00:18:05 +0000 Subject: [time-nuts] Thunderbolt disciplining process In-Reply-To: References: Message-ID: Hello Tom, Yes, the unit could be disciplining based upon the oscillator output. The DAC changes are much more correlated with the PPS signal than the OSC signal. The plots are GPS locked. Time scale is one minute per major horizontal division. If the DAC envelope changes were due to outside temperature getting into the OCXO, the same envelope would show up on the units with the "flat line" temperature sensor. This does not appear to happen. The DAC envelope offset clearly tracks the temperature sensor changes. Old units show lots of fine scale temperature changes that are reflected in the DAC curve. New units show a very flat temperature curve without the corresponding changes to the DAC envelope. My feeling is the temperature sensor reading supplements the OCXO oven and helps compensate for system level dependencies on temperature. These would be be more important during periods of holdover. _________________________________________________________________ Windows Live?: Keep your life in sync. http://windowslive.com/explore?ocid=TXT_TAGLM_WL_t1_allup_explore_022009 From GandalfG8 at aol.com Tue Feb 10 21:29:20 2009 From: GandalfG8 at aol.com (GandalfG8 at aol.com) Date: Tue, 10 Feb 2009 16:29:20 EST Subject: [time-nuts] Datum/Symmetricom Starlite GPS Message-ID: Hi All I have some GPS PCB modules that I've identified from an online data sheet as being Datum Starlite modules, later sold as the Symmetricom Starlite and described as being a "cut down" version of the Starloc unit but with similar performance. Unfortunately that online data sheet is about all the information I can find. If anyone has any information or documentation they would be willing to share, or could point me in the right direction, that would be very much appreciated. regards Nigel GM8PZR From jmiles at pop.net Wed Feb 11 05:41:08 2009 From: jmiles at pop.net (John Miles) Date: Tue, 10 Feb 2009 21:41:08 -0800 Subject: [time-nuts] Initial release of Lady Heather for Windows In-Reply-To: Message-ID: Mark and I have unified the Lady Heather code base between the DOS and Windows versions, fixed various bugs and UI issues, and added some useful features. This is my "version 1.0" release: http://www.ke5fx.com/heather/readme.htm The app is a decent graphical replacement for tboltmon.exe at this point, IMHO. If you currently run tboltmon on your Thunderbolt, give this release a try. As before, if your Windows desktop is smaller than 1280x1024, you will need to add /vs to your shortcut, or use /f to run in fullscreen mode. -- john, KE5FX From rk at timing-consultants.com Wed Feb 11 09:17:28 2009 From: rk at timing-consultants.com (Rob Kimberley) Date: Wed, 11 Feb 2009 09:17:28 -0000 Subject: [time-nuts] Datum/Symmetricom Starlite GPS In-Reply-To: References: Message-ID: Hi Nigel, The StarLoc units were designed for similar performance to the T'bolt. I may have some info tucked away somewhere, but moved house again last week, and still in a state of some chaos! If I can find anything relevant I'll send it over. Rob K -----Original Message----- From: time-nuts-bounces at febo.com [mailto:time-nuts-bounces at febo.com] On Behalf Of GandalfG8 at aol.com Sent: 10 February 2009 21:29 To: time-nuts at febo.com Subject: [time-nuts] Datum/Symmetricom Starlite GPS Hi All I have some GPS PCB modules that I've identified from an online data sheet as being Datum Starlite modules, later sold as the Symmetricom Starlite and described as being a "cut down" version of the Starloc unit but with similar performance. Unfortunately that online data sheet is about all the information I can find. If anyone has any information or documentation they would be willing to share, or could point me in the right direction, that would be very much appreciated. regards Nigel GM8PZR _______________________________________________ time-nuts mailing list -- time-nuts at febo.com To unsubscribe, go to https://www.febo.com/cgi-bin/mailman/listinfo/time-nuts and follow the instructions there. From ch at murgatroid.com Wed Feb 11 18:14:16 2009 From: ch at murgatroid.com (christopher hoover) Date: Wed, 11 Feb 2009 10:14:16 -0800 Subject: [time-nuts] Initial release of Lady Heather for Windows (John Miles) In-Reply-To: References: Message-ID: <00b201c98c74$8cec4840$a6c4d8c0$@com> "John Miles" wrote: > This is my "version 1.0" release: > > http://www.ke5fx.com/heather/readme.htm > > The app is a decent graphical replacement for tboltmon.exe at this > point, IMHO. If you currently run tboltmon on your Thunderbolt, > give this release a try. Is it known to run on Vista? How 'bout Vista x64? I'm trying on Vista x64 with a GPSTM (related but not actually thunderbolt), I get: Serial receive error, code 0x2. Serial port is fine, connected, no other processes attached, etc. Can I send you any more debugging info? I'll have to dig out one of my Thunderbolts to see if the GPSTM itself is the problem. -ch From Doug2 at suddenlink.net Wed Feb 11 22:07:31 2009 From: Doug2 at suddenlink.net (Doug Hutton) Date: Wed, 11 Feb 2009 16:07:31 -0600 Subject: [time-nuts] Initial release of Lady Heather for Windows References: <00b201c98c74$8cec4840$a6c4d8c0$@com> Message-ID: I just downloaded and ran it on Windows 7 and it works fine on a group-buy Thunderbolt.. Doug Hutton -------------------------------------------------- From: "christopher hoover" Sent: Wednesday, February 11, 2009 12:14 PM To: Subject: Re: [time-nuts] Initial release of Lady Heather for Windows (JohnMiles) > "John Miles" wrote: >> This is my "version 1.0" release: >> >> http://www.ke5fx.com/heather/readme.htm >> >> The app is a decent graphical replacement for tboltmon.exe at this >> point, IMHO. If you currently run tboltmon on your Thunderbolt, >> give this release a try. > > Is it known to run on Vista? How 'bout Vista x64? > > I'm trying on Vista x64 with a GPSTM (related but not actually > thunderbolt), > I get: > > Serial receive error, code 0x2. > > Serial port is fine, connected, no other processes attached, etc. > > Can I send you any more debugging info? > > I'll have to dig out one of my Thunderbolts to see if the GPSTM itself is > the problem. > > -ch From jmiles at pop.net Wed Feb 11 23:27:16 2009 From: jmiles at pop.net (John Miles) Date: Wed, 11 Feb 2009 15:27:16 -0800 Subject: [time-nuts] Initial release of Lady Heather for Windows In-Reply-To: Message-ID: Thanks for the confirmation -- I don't believe anyone else has tried it on Windows 7 before. Did you have to run it as admin? Christopher, I imagine there's either a WoW64 compatibility problem or (more likely) the unit you're using isn't 100% compatible with Trimble's TSIP implementation for the Thunderbolt. It would be interesting to hear if you can try 64-bit Vista with a real Thunderbolt; otherwise there are too many variables at play to make an educated guess as to what's going on in your case. Error code 0x02 means that it found the serial port and opened it correctly, but timed out trying to perform a read operation. -- john, KE5FX > -----Original Message----- > From: time-nuts-bounces at febo.com [mailto:time-nuts-bounces at febo.com]On > Behalf Of Doug Hutton > Sent: Wednesday, February 11, 2009 2:08 PM > To: Discussion of precise time and frequency measurement > Subject: Re: [time-nuts] Initial release of Lady Heather for Windows > > > I just downloaded and ran it on Windows 7 and it works fine on > a group-buy Thunderbolt.. > > Doug Hutton > > > -------------------------------------------------- > From: "christopher hoover" > Sent: Wednesday, February 11, 2009 12:14 PM > To: > Subject: Re: [time-nuts] Initial release of Lady Heather for Windows > (JohnMiles) > > > Is it known to run on Vista? How 'bout Vista x64? > > > > I'm trying on Vista x64 with a GPSTM (related but not actually > > thunderbolt), > > I get: > > > > Serial receive error, code 0x2. > > From holrum at hotmail.com Thu Feb 12 00:37:17 2009 From: holrum at hotmail.com (Mark Sims) Date: Thu, 12 Feb 2009 00:37:17 +0000 Subject: [time-nuts] Initial release of Lady Heather for Windows In-Reply-To: References: Message-ID: Runs great under WIN98SE... A 200 MHz laptop barely breaks a sweat (spends maybe 2 seconds) calculating the OADEVS over 250,000 points in real time. _________________________________________________________________ Windows Live?: E-mail. Chat. Share. Get more ways to connect. http://windowslive.com/howitworks?ocid=TXT_TAGLM_WL_t2_allup_howitworks_022009 From Doug2 at suddenlink.net Thu Feb 12 04:03:00 2009 From: Doug2 at suddenlink.net (Doug Hutton) Date: Wed, 11 Feb 2009 22:03:00 -0600 Subject: [time-nuts] Initial release of Lady Heather for Windows References: Message-ID: <72753C742FED45CE857567396EB3E0A9@DougPC> John - I did have to run it as Admin. I tried a few things on a Guest logon including a USB serial port but got the 0x02 error. Just clicking on the Lady Heather icon worked on the first try using a real serial port.. Your display is beautiful - you did good. Doug, W5JUV Lubbock -------------------------------------------------- From: "John Miles" Sent: Wednesday, February 11, 2009 5:27 PM To: "Discussion of precise time and frequency measurement" Subject: Re: [time-nuts] Initial release of Lady Heather for Windows > Thanks for the confirmation -- I don't believe anyone else has tried it on > Windows 7 before. Did you have to run it as admin? > > Christopher, I imagine there's either a WoW64 compatibility problem or > (more > likely) the unit you're using isn't 100% compatible with Trimble's TSIP > implementation for the Thunderbolt. It would be interesting to hear if > you > can try 64-bit Vista with a real Thunderbolt; otherwise there are too many > variables at play to make an educated guess as to what's going on in your > case. Error code 0x02 means that it found the serial port and opened it > correctly, but timed out trying to perform a read operation. > > -- john, KE5FX > >> -----Original Message----- >> From: time-nuts-bounces at febo.com [mailto:time-nuts-bounces at febo.com]On >> Behalf Of Doug Hutton >> Sent: Wednesday, February 11, 2009 2:08 PM >> To: Discussion of precise time and frequency measurement >> Subject: Re: [time-nuts] Initial release of Lady Heather for Windows >> >> >> I just downloaded and ran it on Windows 7 and it works fine on >> a group-buy Thunderbolt.. >> >> Doug Hutton >> >> >> -------------------------------------------------- >> From: "christopher hoover" >> Sent: Wednesday, February 11, 2009 12:14 PM >> To: >> Subject: Re: [time-nuts] Initial release of Lady Heather for Windows >> (JohnMiles) >> > >> > Is it known to run on Vista? How 'bout Vista x64? >> > >> > I'm trying on Vista x64 with a GPSTM (related but not actually >> > thunderbolt), >> > I get: >> > >> > Serial receive error, code 0x2. >> > > > > _______________________________________________ > time-nuts mailing list -- time-nuts at febo.com > To unsubscribe, go to > https://www.febo.com/cgi-bin/mailman/listinfo/time-nuts > and follow the instructions there. From jmiles at pop.net Thu Feb 12 04:56:29 2009 From: jmiles at pop.net (John Miles) Date: Wed, 11 Feb 2009 20:56:29 -0800 Subject: [time-nuts] Initial release of Lady Heather for Windows In-Reply-To: <72753C742FED45CE857567396EB3E0A9@DougPC> Message-ID: > John - > > I did have to run it as Admin. I tried a few things on a Guest logon > including a > USB serial port but got the 0x02 error. Just clicking on the > Lady Heather > icon > worked on the first try using a real serial port.. > So if you have a real serial port you don't need to run as admin, but if you have a USB adapter you do, correct? > Your display is beautiful - you did good. Thanks! The display's all Mark's doing. -- john From kyndbo at gmail.com Thu Feb 12 08:57:37 2009 From: kyndbo at gmail.com (Helge Kyndbo) Date: Thu, 12 Feb 2009 09:57:37 +0100 Subject: [time-nuts] Initial release of Lady Heather for windows Message-ID: <33899fa20902120057h1fa1e04bt81b0b416295b8d67@mail.gmail.com> Thanks John for the your release of windows version Lady Heather. I like the GPS time cursor, It has been running stable for 18 hours now on a laptop with XP-home SP1-2. Thanks Helge > > > Message: 2 > Date: Tue, 10 Feb 2009 21:41:08 -0800 > From: "John Miles" > Subject: [time-nuts] Initial release of Lady Heather for Windows > To: "Discussion of precise time and frequency measurement" > > Message-ID: > Content-Type: text/plain; charset="us-ascii" > > Mark and I have unified the Lady Heather code base between the DOS and > Windows versions, fixed various bugs and UI issues, and added some useful > features. This is my "version 1.0" release: > > http://www.ke5fx.com/heather/readme.htm > > The app is a decent graphical replacement for tboltmon.exe at this point, > IMHO. If you currently run tboltmon on your Thunderbolt, give this release > a try. > > As before, if your Windows desktop is smaller than 1280x1024, you will need > to add /vs to your shortcut, or use /f to run in fullscreen mode. > > -- john, KE5FX > > > > > > > From cdelect at juno.com Thu Feb 12 16:31:13 2009 From: cdelect at juno.com (Corby Dawson) Date: Thu, 12 Feb 2009 08:31:13 -0800 Subject: [time-nuts] SRS PRS10 Message-ID: <20090212.083113.3164.2.cdelect@juno.com> Hi, just wanted to let everyone know that I have an SRS PRS10 mounted in an Efratom MRT chassis on eBay right now. 320340140628 Corby Dawson ____________________________________________________________ Click for free info on online doctorate degrees and make up to $250k/ year. http://thirdpartyoffers.juno.com/TGL2141/fc/PnY6rw3YN20xoxfIgUGyfTnL5EJAEschcuGOtlifevpW2ZDh0ajQl/ From richiem at hughes.net Thu Feb 12 18:20:27 2009 From: richiem at hughes.net (Richard Moore) Date: Thu, 12 Feb 2009 10:20:27 -0800 Subject: [time-nuts] Lady Heather Win port In-Reply-To: References: Message-ID: <3E5AADEB-0E49-44D2-BDBC-093726E98CBA@hughes.net> John and Mark -- Thanks. I'm running it fine under XP sp3. I had to change my USB>serial port adapter's port number using the device manager in Windows. LH couldn't find port 1 unused (it wasn't - default allocated to comms), and would fail to boot. I couldn't figure out how to get into the shortcut's command line system to reconfig the port number. Once I got port 1 cleared, LH booted OK. The display is great looking. I booted it last night and went to bed; MS updated the system overnight and re-booted automatically, so I lost data. Am rerunning now. Good work! Dick Moore From Doug2 at suddenlink.net Thu Feb 12 18:42:10 2009 From: Doug2 at suddenlink.net (Doug Hutton) Date: Thu, 12 Feb 2009 12:42:10 -0600 Subject: [time-nuts] Initial release of Lady Heather for Windows References: Message-ID: <5E52C7E16E944F99B6C114137649C28E@DougPC> John - As I poked around, with Heather and Windows 7, things got a little more involved. If TBolt is connected to real serial port 1, TBoltMon and Heather can be run either as Admin or Guest (Standard user), starting from a desktop icon or the Start Key list.. However, the Run box from the Start Key is not available to a Standard user, so I couldn't figure out any way to use the command-line switches to change the Serial Port to anything other than 1 if you are not logged on as Admin. As Admin, I can run TBoltMon just fine on a U232-P9 USB Serial Port adapter, but using this Run command line - C:\Program Files\heather\heather.exe /4 Heather gets Serial receive error, code 0x2 with the exact same connection. Heather formats the window, pauses a couple of seconds, then gets the error. It looks like Dick Moore and I are seeing similar stuff. Doug Hutton, W5JUV Lubbock -------------------------------------------------- From: "John Miles" Sent: Wednesday, February 11, 2009 10:56 PM To: "Discussion of precise time and frequency measurement" Subject: Re: [time-nuts] Initial release of Lady Heather for Windows > > So if you have a real serial port you don't need to run as admin, but if > you > have a USB adapter you do, correct? > >> Your display is beautiful - you did good. > > Thanks! The display's all Mark's doing. > > -- john > > > > > _______________________________________________ > time-nuts mailing list -- time-nuts at febo.com > To unsubscribe, go to > https://www.febo.com/cgi-bin/mailman/listinfo/time-nuts > and follow the instructions there. > From chris at ven-tel.com Thu Feb 12 22:59:00 2009 From: chris at ven-tel.com (christopher bumgarner) Date: Thu, 12 Feb 2009 14:59:00 -0800 Subject: [time-nuts] Navsymm/Symmetricom DCD523 antenna Message-ID: <32fa84d70902121459n3ca4c66fk88809e93c67443c4@mail.gmail.com> Greetings List, I'm looking for some Navsymm GTR antennas. The original manufacturer Navsymm was a company out of the United Kingdom. They were bought some time ago by Telecom Solutions, a subsidiary of Symmetricom. The model number on the base of the unit is DCD-LPR-G and the part number is 090-42110-01. This antenna goes with a DCD523 Digital Clock Distributor Network Synchronization System. I can pay handsomely for each antenna found or trade any equipment you may be in search of. I have pictures available, please email if these will help. Please let me know if there is anyway you can help me find some of these antennas. Thanks, Chris chris at ven-tel.com From flash452 at yahoo.com Fri Feb 13 04:30:45 2009 From: flash452 at yahoo.com (Flash Crews) Date: Thu, 12 Feb 2009 20:30:45 -0800 (PST) Subject: [time-nuts] CMC Allstar 12 GPS engines Message-ID: <934299.10433.qm@web51608.mail.re2.yahoo.com> hi i was trying to get more info on the CMC Allstar 12 GPS engines you wrote about in 2007 as i have over 350 of them to sell but i to cant seem to find any info please help From mccorkle at ptialaska.net Fri Feb 13 05:21:32 2009 From: mccorkle at ptialaska.net (Richard H McCorkle) Date: Thu, 12 Feb 2009 20:21:32 -0900 (AKST) Subject: [time-nuts] CMC Allstar 12 GPS engines In-Reply-To: <934299.10433.qm@web51608.mail.re2.yahoo.com> References: <934299.10433.qm@web51608.mail.re2.yahoo.com> Message-ID: <16024.206.174.20.67.1234502492.squirrel@mymail.acsalaska.net> Try this link for the Allstar Manual ftp://ftp.novatel.ca/outgoing/support/Documents/Manuals/CMC/ Richard > > hi i was trying to get more info on the CMC Allstar 12 GPS engines you wrote > about in 2007 as i have over 350 of them to sell but i to cant seem to find any > info please help > > > > > _______________________________________________ > time-nuts mailing list -- time-nuts at febo.com > To unsubscribe, go to https://www.febo.com/cgi-bin/mailman/listinfo/time-nuts > and follow the instructions there. > From ka2cdk at cox.net Fri Feb 13 14:33:12 2009 From: ka2cdk at cox.net (Thomas A. Frank) Date: Fri, 13 Feb 2009 09:33:12 -0500 Subject: [time-nuts] Even better than a leap second Message-ID: http://www.engadget.com/2009/02/13/this-afternoon-party-like- its-1234567890-unix-time/ This afternoon, party like it's 1234567890 Unix time by Ross Miller, posted Feb 13th 2009 at 5:41AM Number aficionados, you probably already know what happens today. As of 6:31:30PM ET this afternoon it will officially be 1234567890 Unix time, which started at zero and has been counting seconds since the stroke of midnight on January 1, 1970, not counting leap seconds. We suggest you put on your best watch or other geek chic and enjoy that one fateful second of sequential bliss -- as the story goes, 1234567891, party over, oops, out of time. In case you've got other plans and can't celebrate the milestone, we'll see you December 22, 2282 at 3:13:30PM ET for the merriment of 9876543210. and http://www.1234567890day.com/ 1234567890 Day It's time to party like it's 1234567890 ? 'cause it is! 8 Hours 59 Minutes 5 Seconds What is this all about anyway? It's time to party like it's 1234567890 ? 'cause it is! On this Friday, Feb 13 at exactly 3:31:30 PM (PST), Unix time (WTF is Unix time?) will equal '1234567890'. But what can I do? For those based in San Francisco, you can join us for the countdown to 1234567890: Friday, Feb 13 at exactly 2:31 (1 hr early for some drinks) The 21st Amendment 563 Second Street San Francisco http://www.21st-amendment.com On Upcoming.org http://upcoming.yahoo.com/event/1814440 Other 1234567890 parties around the world Vancouver, BC at the Irish Heather Eureka, California at the Lost Coast Brewery Helena, Montana at the Gold Bar Seattle, Washington at here Edmonton, Alberta at the Sherlock Holmes Pub, downtown on Rice Howard Way Los Angeles, California at Harry O's Boston time_t party at the Westin Boston Waterfront lobby bar Buenos Aires, Argentina at Sugar (4619 Costa Rica in Palermo) Yerevan, Armenia at Maret's Kitchen Saloon Slatina, Croatia at St. Patric's Pub Bujumbura, Burundi at Club du Lac Tanganyika Nairobi, Kenya at Tamasha in the Hurlingham area Lappeenranta, Finland at the Old Cock restaurant Omaha, Nebraska at BrightMix (5018 Leavenworth St) Indianapolis, Indiana at the Rock Bottom Brewery (downtown) Minot, North Dakota at Off the Vine (Main Street) Vienna, Austria at here (Free shot at time_t 1234567890 for everyone showing up with a device capable of displaying the current timestamp)! Sioux Falls, South Dakota at Nasser Law Offices, P.C. 204 South Main Avenue (rsvp) Budapest, Hungary at Szilvupl? Copenhagen, Denmark Email for details Denver, Colorado at The Broker (downtown) more Raleigh, North Carolina at Amedeo's Bozeman, Montana at Spectators Bar and Grill Washington, DC at Brickskeller (DC PHP Dev Group) Vernon, New York the only cafe (state route 5) NYC, New York at Jelly Potsdam, New York at The Clarkson Open Source Institute lab, Clarkson University San Diego, California at the Living Room Cafe near SDSU San Diego, California at Porter's Pub on UCSD Asunci?n, Paraguay at Britannia Pub Melbourne, Australia at the Beefyville Berkeley, California at Soda Hall Dubai UAE Irish Pub, Near Tennis Stadium Pearland, Texas BJ's Pizza Honolulu, Hawaii Lunch gathering at ITS, Keller Hall, University of Hawaii, Manoa Philadelphia, Pennsylvania at the Abbaye (637 N 3rd St) San Francisco at Noise Bridge Patna, India at Elevens (Exhibition Road) ##1234567890 on irc.freenode.net Nimbot 1234567890 Jam Shoot us an email to add your event here... On twitter, follow utcwatch: http://twitter.com/utcwatch ? Copyleft 2009 1234567890 Inc. ? Brought to you by Colin, Andy, and the numbers 123456789 and 0 From magnus at rubidium.dyndns.org Fri Feb 13 14:52:50 2009 From: magnus at rubidium.dyndns.org (Magnus Danielson) Date: Fri, 13 Feb 2009 15:52:50 +0100 Subject: [time-nuts] Even better than a leap second In-Reply-To: References: Message-ID: <49958942.3000402@rubidium.dyndns.org> Thomas A. Frank skrev: > http://www.engadget.com/2009/02/13/this-afternoon-party-like- > its-1234567890-unix-time/ > > This afternoon, party like it's 1234567890 Unix time > > by Ross Miller, posted Feb 13th 2009 at 5:41AM > > Number aficionados, you probably already know what happens today. As > of 6:31:30PM ET this afternoon it will officially be 1234567890 Unix > time, which started at zero and has been counting seconds since the > stroke of midnight on January 1, 1970, not counting leap seconds. We > suggest you put on your best watch or other geek chic and enjoy that > one fateful second of sequential bliss -- as the story goes, > 1234567891, party over, oops, out of time. In case you've got other > plans and can't celebrate the milestone, we'll see you December 22, > 2282 at 3:13:30PM ET for the merriment of 9876543210. Of the three time_t interpretations... which is it? I am sure the party will be long enought to enjoy all three interpretations, but never the less... Need to calculate the different between UTC-TAI again at 1 Jan 1970... I don't recall if it was 8 seconds and 82 or 85 us... something like that. Hmm... I think I just came up with a fourth interpretation of when the time_t... need to check details. Is it wise to use such an ambiguous timing marker? Cheers, Magnus From jgd at johngsbbq.com Fri Feb 13 23:39:01 2009 From: jgd at johngsbbq.com (John DeArmond) Date: Fri, 13 Feb 2009 18:39:01 -0500 Subject: [time-nuts] Even better than a leap second In-Reply-To: References: Message-ID: <1234568341.6832.5.camel@neonjohn> On Fri, 2009-02-13 at 09:33 -0500, Thomas A.Frank wrote: > It's time to party like it's 1234567890 ? 'cause it is! On this > Friday, Feb 13 at exactly 3:31:30 PM (PST), Unix time (WTF is Unix > time?) will equal '1234567890'. A screen shot of this momentous occasion is attached :-) -- John DeArmond Tellico Plains, Occupied TN http://www.neon-john.com <-- email from here http://www.johndearmond.com <-- Best damned Blog on the net -------------- next part -------------- A non-text attachment was scrubbed... Name: 1234567890a.jpeg Type: image/jpeg Size: 36022 bytes Desc: not available Url : http://www.febo.com/pipermail/time-nuts/attachments/20090213/8451a1d5/attachment-0001.jpeg From holrum at hotmail.com Sat Feb 14 17:43:22 2009 From: holrum at hotmail.com (Mark Sims) Date: Sat, 14 Feb 2009 17:43:22 +0000 Subject: [time-nuts] Tbolt temperature sensor In-Reply-To: References: Message-ID: Thanks to a fellow list member I obtained a rev C2 DS1620 temperature sensor chip and replaced the rev E2 chip on my "flat line" Thunderbolt board. The board is now a "curvy line" board. So the difference is definitely a DS1620 chip difference and not a tbolt firmware difference. I made several 24 hour holdover runs with the old chip. I am going to let the unit run for a week or so to let the filters learn the new chip and repeat the holdover runs. Hopefully something will pop out of the data, but I am not too hopeful. There was not a lot of consistency in the data from the first holdover runs. Any ideas of what to look for or how to analyze the data? _________________________________________________________________ Windows Live?: Keep your life in sync. http://windowslive.com/howitworks?ocid=TXT_TAGLM_WL_t1_allup_howitworks_022009 From sam at canardpc.com Sat Feb 14 19:30:00 2009 From: sam at canardpc.com (Samuel D. [x86/CPC]) Date: Sat, 14 Feb 2009 20:30:00 +0100 Subject: [time-nuts] DIY Frequency extension for HP Agilent 53181A, 53131A or 53181A Message-ID: <003901c98eda$a1128510$e3378f30$@com> Hi there, First, let me introduce myself : I?m a 28-years-old electronic engineer working for a press magazine as Hardware & Test Chief Editor. I?m from France; so, please forget about my bad English. Since years, I?m obsessed with precision measurements and I spend the last 15 months finding the best way to accurately measure a couple of femtoamperes or nanovolts. For 2009, my new goal is to be able to measure the mHz part of a GHz signal so, I need ultra high accuracy frequency standard and other fancy things like that. But here is the problem: I?m not rich and I don?t have enough money to build Agilent?s and Symmetricon?s stuff at MSRP. In the other hand, I love precision equipment and high quality, well-known hardware. After all, its 90% hobby and 10% professional, so, why spend thousands of dollars on a brand new Rubidium standard when you can build one with a Datum module found on eBay for a small fraction of that price? That?s what I will do in the upcoming weeks. By the way, as you know, the two first equipments to own for precision time measurements are a high stability frequency reference and a good counter. Building the first one is quite easy but a good counter is something expensive. Fortunately, I just bought an Agilent 53131A for ?350 on eBay and I?m really happy with it. As usual, I will now spend some time to see how I can improve it. HP/Agilent 53181A, 53131A & 53132A are probably the most used frequency counters in the world. They come with two kinds of options: three advanced time-bases (an OCXO with 2x10e-7, 1.5x10e-8 or 3x10e-9 monthly aging rate) and four frequency extensions. These counter?s max base frequency is 225 MHz and may go up to 1.5, 3.0, 5.0 or 12.4 GHz with an additional PCB linked to a BNC or type-N connector acting as Channel 2 (for 53131A) or Channel 3 (for 53131A and 53132A). As often with Agilent, the price for those extensions is clearly and indecently overpriced: more than $2000 for the 3 GHz option, the price of the counter itself. Overpriced, Really? Yes! And if you still doubt, here is an anecdote from my experience: the Agilent 34420A Nanovolt Meter comes with a low-thermal input connector. Last year, I wanted to build a custom cable and I asked Agilent the price for a male connector. They quoted that small part more than ?100. After lots or research I found the exact same connector, originally built by Lemo, and bought it new for less than $15. Ok, come back to the 53131A frequency options. The extension sold by Agilent is nothing more than a prescaler (probably /16, /32, /64) mounted on a small PCB and connected to the counter with a standard ribbon cable. The PCB is really small and there is no more than 20 components built-in as you can see here : http://www.x86.fr/temp/003.jpg (3 GHz option) and http://www.x86.fr/temp/124.jpg (12.4 GHz option). From holrum at hotmail.com Sat Feb 14 21:39:33 2009 From: holrum at hotmail.com (Mark Sims) Date: Sat, 14 Feb 2009 21:39:33 +0000 Subject: [time-nuts] DIY Frequency extension for HP Agilent 53181A, 53131A or 53181A In-Reply-To: References: Message-ID: I would HIGHLY recommend that you buy a GPS disciplined oscillator. You can pick up a Trimble Thunderbolt on Ebay for less than $150 from the Chinese sellers (or $300 ready to go with power supply and antenna which you can buy for a lot less). The frequency output of these can rival that of a cesium beam unit. You will need one to set the frequency of the rubidium oscillator. I would avoid the Lucent units. They are poorly documented, very big, and most cannot be made to properly lock the oscillator to the GPS signal. _________________________________________________________________ Windows Live?: Keep your life in sync. http://windowslive.com/howitworks?ocid=TXT_TAGLM_WL_t1_allup_howitworks_022009 From holrum at hotmail.com Sat Feb 14 21:49:03 2009 From: holrum at hotmail.com (Mark Sims) Date: Sat, 14 Feb 2009 21:49:03 +0000 Subject: [time-nuts] Tbolt temperature sensor In-Reply-To: References: Message-ID: I wrote some code to process my holdover logs from the Thunderbolt with the newer "flat line" rev E2 temperature sensor. The data is a lot more consistent than I thought it was. I am seeing around 9.5 uS +- 0.5 uS of drift in the PPS output per degree C across 4 different 24 hour runs. My analysis is very simple... scan the log and calculate (MAX_PPS-MIN_PPS)/(MAX_TEMP-MIN_TEMP) It will be interesting to see how the unit performs with the old revision C2 ("better"?) temperature sensor. _________________________________________________________________ Windows Live?: E-mail. Chat. Share. Get more ways to connect. http://windowslive.com/howitworks?ocid=TXT_TAGLM_WL_t2_allup_howitworks_022009 From mskop at bigpond.net.au Sun Feb 15 01:07:42 2009 From: mskop at bigpond.net.au (Max Skop) Date: Sun, 15 Feb 2009 12:07:42 +1100 Subject: [time-nuts] DIY Frequency extension for HP Agilent 53181A, 53131A or 53181A References: <003901c98eda$a1128510$e3378f30$@com> Message-ID: <1872EF473D184CE5A01C929E443207D0@p4253> Hi Samuel and all, Welcome to the group. I also have a 53131A and would welcome any way to enhance or upgrade the performance of this instrument. Please count me in on clone option parts. As can be seen from the picure of the 3Ghz option there is nothing special in its construction. Just four stages of amplification and a divider. Only need to ID the parts to copy and determine the pinout of the connector. The mounting holes would also have to be accurately positioned. Max ----- Original Message ----- From: "Samuel D. [x86/CPC]" To: Sent: Sunday, 15 February, 2009 6:30 AM Subject: [time-nuts] DIY Frequency extension for HP Agilent 53181A,53131A or 53181A > Hi there, > > First, let me introduce myself : I?m a 28-years-old electronic engineer > working for a press magazine as Hardware & Test Chief Editor. I?m from > France; so, please forget about my bad English. Since years, I?m obsessed > with precision measurements and I spend the last 15 months finding the > best way to accurately measure a couple of femtoamperes or nanovolts. For > 2009, my new goal is to be able to measure the mHz part of a GHz signal > so, I need ultra high accuracy frequency standard and other fancy things > like that. > > But here is the problem: I?m not rich and I don?t have enough money to > build Agilent?s and Symmetricon?s stuff at MSRP. In the other hand, I love > precision equipment and high quality, well-known hardware. After all, its > 90% hobby and 10% professional, so, why spend thousands of dollars on a > brand new Rubidium standard when you can build one with a Datum module > found on eBay for a small fraction of that price? That?s what I will do in > the upcoming weeks. > By the way, as you know, the two first equipments to own for precision > time measurements are a high stability frequency reference and a good > counter. Building the first one is quite easy but a good counter is > something expensive. Fortunately, I just bought an Agilent 53131A for ?350 > on eBay and I?m really happy with it. As usual, I will now spend some time > to see how I can improve it. > > HP/Agilent 53181A, 53131A & 53132A are probably the most used frequency > counters in the world. They come with two kinds of options: three advanced > time-bases (an OCXO with 2x10e-7, 1.5x10e-8 or 3x10e-9 monthly aging rate) > and four frequency extensions. These counter?s max base frequency is 225 > MHz and may go up to 1.5, 3.0, 5.0 or 12.4 GHz with an additional PCB > linked to a BNC or type-N connector acting as Channel 2 (for 53131A) or > Channel 3 (for 53131A and 53132A). > > As often with Agilent, the price for those extensions is clearly and > indecently overpriced: more than $2000 for the 3 GHz option, the price of > the counter itself. Overpriced, Really? Yes! And if you still doubt, here > is an anecdote from my experience: the Agilent 34420A Nanovolt Meter comes > with a low-thermal input connector. Last year, I wanted to build a custom > cable and I asked Agilent the price for a male connector. They quoted that > small part more than ?100. After lots or research I found the exact same > connector, originally built by Lemo, and bought it new for less than $15. > > Ok, come back to the 53131A frequency options. The extension sold by > Agilent is nothing more than a prescaler (probably /16, /32, /64) mounted > on a small PCB and connected to the counter with a standard ribbon cable. > The PCB is really small and there is no more than 20 components built-in > as you can see here : http://www.x86.fr/temp/003.jpg (3 GHz option) and > http://www.x86.fr/temp/124.jpg (12.4 GHz option). From a good source, the > prescaler is not built by Agilent and all components are available from > third-party. > > My target is to create a clone of that extension. I think it will not cost > more than $50. Then, I?m planning to offer a ?DIY frequency extension kit? > for sales for all time enthusiasts and without making profit. Every owner > of a 53181A, 53131A will be able to add the nice freq extender for a > fraction of the MSRP. But I need some help. The first thing to do is to > find someone with an original -003, -005 or -124 extension and willing to > scan (or take pictures of) the card in high resolution in order to > reproduce the schematic and layout used by Agilent. Knowing the pinout of > the J2 connector will be useful too. Then, I will create a PCB and find > the good replacement components...etc. > > Do you think this project is a good idea and may interest somebody other > than me? > > Please share your thoughts ;) > > Regards, > > --------------------- > Samuel DEMEULEMEESTER > Presse Non Stop - Canard PC > www.canardpc.com > MSN : sam at x86.fr > > > > _______________________________________________ > time-nuts mailing list -- time-nuts at febo.com > To unsubscribe, go to > https://www.febo.com/cgi-bin/mailman/listinfo/time-nuts > and follow the instructions there. From richard at karlquist.com Sun Feb 15 02:46:46 2009 From: richard at karlquist.com (Rick Karlquist) Date: Sat, 14 Feb 2009 18:46:46 -0800 (PST) Subject: [time-nuts] DIY Frequency extension for HP Agilent 53181A, 53131A or 53181A In-Reply-To: <1872EF473D184CE5A01C929E443207D0@p4253> References: <003901c98eda$a1128510$e3378f30$@com> <1872EF473D184CE5A01C929E443207D0@p4253> Message-ID: <43fc33a655d8644ca3c013b6be3b892c.squirrel@webmail.sonic.net> Max Skop wrote: > Hi Samuel and all, > Welcome to the group. > I also have a 53131A and would welcome any way to enhance or upgrade the > performance of this instrument. Please count me in on clone option parts. > As can be seen from the picure of the 3Ghz option there is nothing special > in its construction. Just four stages of amplification and a divider. In 1987, I was the project manager for the HP 5334B frequency counter. In those days, we had a 1.3 GHz option called a "C channel". I evaluated various off the shelf dividers at the time and built some C channel boards very similar to the one you are describing. It is actually very difficult to make a prescaler that really works well using off the shelf dividers. The prescaler tends to be insensitive to low frequencies. Attempting to fix this with a lot of gain (4 stages!) has the problem that any wideband noise from the source is also amplified. Off the shelf dividers are made to work in frequency synthesizers where they have a clean signal from an oscillator. They don't do well with general signals from noisy sources. The HP 5386 used an HP made frequency divider that had "static" flip flops instead of the "dynamic" ones in all the off the shelf dividers. It is the only prescaler I am familiar with that actually works well. The custom IC in it cost HP a fair amount just because it was "home made" in the Santa Rosa fab, which was a money sink. If you want to build your own prescaler, you might not want to copy the Agilent one. Unless you need the sensitivity, you would be better off with less gain. Also, you should choose a divider that works well at the frequency you want to measure. AFAIK, all OTS dividers have an optimum frequency range, which varies from divider to divider. When testing prescalers, the thing to watch out for is that the lower order digits will bounce around. This is usually a symptom of prescaler error, assuming that the source is stable. Rick Karlquist N6RK From sam at canardpc.com Sun Feb 15 17:04:28 2009 From: sam at canardpc.com (Samuel D. [x86/CPC]) Date: Sun, 15 Feb 2009 18:04:28 +0100 Subject: [time-nuts] DIY Frequency extension for HP Agilent 53181A, 53131A or 53181A In-Reply-To: <43fc33a655d8644ca3c013b6be3b892c.squirrel@webmail.sonic.net> References: <003901c98eda$a1128510$e3378f30$@com> <1872EF473D184CE5A01C929E443207D0@p4253> <43fc33a655d8644ca3c013b6be3b892c.squirrel@webmail.sonic.net> Message-ID: <001901c98f8f$76f4cda0$64de68e0$@com> Thanks for your answers. About the components, I think my first prototype will use some very well-known prescaler and amplifiers like Hittite HMC363 and Mini-circuits ERA-1SM, only to validate the design. I will use a 100 MHz high-pass filter, but overall, it will be a wide-band, low-sensitivity design, exactly like the original option. But before thinking about the best components to use for the amps and the divider, we need to know how to interface the prescaler with the counter. There is some very important questions to answer : - What's the divider ratio used ? Perhaps /16 for 3 GHz, /32 for 5 GHz, and /64 for 12.4 GHz, but we really need to be sure. - What's the gain of each one of the four amplifiers present ? - How does the counter aware about the option used ? There is no "intelligence" on the option board, so, the counter may detect the type of prescaler with some pull-up or pull-down resistors somewhere. - What is the exact pinout for the J2 connector ? In order to get an answer, we need to find somebody with a genuine option board willing to spend an hour for the project... :-) -----Message d'origine----- De?: time-nuts-bounces at febo.com [mailto:time-nuts-bounces at febo.com] De la part de Rick Karlquist Envoy??: dimanche 15 f?vrier 2009 03:47 ??: Discussion of precise time and frequency measurement Objet?: Re: [time-nuts] DIY Frequency extension for HP Agilent 53181A, 53131A or 53181A Max Skop wrote: > Hi Samuel and all, > Welcome to the group. > I also have a 53131A and would welcome any way to enhance or upgrade the > performance of this instrument. Please count me in on clone option parts. > As can be seen from the picure of the 3Ghz option there is nothing special > in its construction. Just four stages of amplification and a divider. In 1987, I was the project manager for the HP 5334B frequency counter. In those days, we had a 1.3 GHz option called a "C channel". I evaluated various off the shelf dividers at the time and built some C channel boards very similar to the one you are describing. It is actually very difficult to make a prescaler that really works well using off the shelf dividers. The prescaler tends to be insensitive to low frequencies. Attempting to fix this with a lot of gain (4 stages!) has the problem that any wideband noise from the source is also amplified. Off the shelf dividers are made to work in frequency synthesizers where they have a clean signal from an oscillator. They don't do well with general signals from noisy sources. The HP 5386 used an HP made frequency divider that had "static" flip flops instead of the "dynamic" ones in all the off the shelf dividers. It is the only prescaler I am familiar with that actually works well. The custom IC in it cost HP a fair amount just because it was "home made" in the Santa Rosa fab, which was a money sink. If you want to build your own prescaler, you might not want to copy the Agilent one. Unless you need the sensitivity, you would be better off with less gain. Also, you should choose a divider that works well at the frequency you want to measure. AFAIK, all OTS dividers have an optimum frequency range, which varies from divider to divider. When testing prescalers, the thing to watch out for is that the lower order digits will bounce around. This is usually a symptom of prescaler error, assuming that the source is stable. Rick Karlquist N6RK _______________________________________________ time-nuts mailing list -- time-nuts at febo.com To unsubscribe, go to https://www.febo.com/cgi-bin/mailman/listinfo/time-nuts and follow the instructions there. From surplus1 at prodigy.net Sun Feb 15 20:09:41 2009 From: surplus1 at prodigy.net (surplus1) Date: Sun, 15 Feb 2009 12:09:41 -0800 Subject: [time-nuts] Kinemetrics Truetime 60-DC Message-ID: <00b001c98fa9$5825ff60$90eef804@pavilion> I am looking for a manual for the Kinemetrics Truetime 60-DC time receiver. John Bartholomew surplus1 at prodigy.net Thanks From GandalfG8 at aol.com Sun Feb 15 17:42:16 2009 From: GandalfG8 at aol.com (GandalfG8 at aol.com) Date: Sun, 15 Feb 2009 12:42:16 EST Subject: [time-nuts] DIY Frequency extension for HP Agilent 53181A, 53131A or 53181A Message-ID: In a message dated 15/02/2009 17:05:21 GMT Standard Time, sam at canardpc.com writes: - What's the divider ratio used ? Perhaps /16 for 3 GHz, /32 for 5 GHz, and /64 for 12.4 GHz, but we really need to be sure. - What's the gain of each one of the four amplifiers present ? - How does the counter aware about the option used ? There is no "intelligence" on the option board, so, the counter may detect the type of prescaler with some pull-up or pull-down resistors somewhere. - What is the exact pinout for the J2 connector ? In order to get an answer, we need to find somebody with a genuine option board willing to spend an hour for the project... ------------------- If you download the 53132A component level information manual from the Agilent site you'll find a schematic of the channel 3 board with values and J2 pinout. The 53131A version of this manual doesn't seem to include this. regards Nigel GM8PZR From had at to-way.com Sun Feb 15 18:05:52 2009 From: had at to-way.com (Had) Date: Sun, 15 Feb 2009 10:05:52 -0800 Subject: [time-nuts] Kinemetrics Truetime 60-DC In-Reply-To: <00b001c98fa9$5825ff60$90eef804@pavilion> References: <00b001c98fa9$5825ff60$90eef804@pavilion> Message-ID: <20090215180556.25D18E0480D@mail-in03.adhost.com> John, Go to www.to-way.com it is in the time-nut page. Had K7MLR At 12:09 PM 2/15/2009, you wrote: >I am looking for a manual for the Kinemetrics Truetime 60-DC time >receiver. John Bartholomew surplus1 at prodigy.net Thanks >_______________________________________________ >time-nuts mailing list -- time-nuts at febo.com >To unsubscribe, go to https://www.febo.com/cgi-bin/mailman/listinfo/time-nuts >and follow the instructions there. From surplus1 at prodigy.net Sun Feb 15 21:23:18 2009 From: surplus1 at prodigy.net (surplus1) Date: Sun, 15 Feb 2009 13:23:18 -0800 Subject: [time-nuts] Kinemetrics Truetime 60-DC References: <00b001c98fa9$5825ff60$90eef804@pavilion> <20090215180556.25D18E0480D@mail-in03.adhost.com> Message-ID: <00cf01c98fb3$a074d8e0$90eef804@pavilion> Hello, For some reason I cannot open the Kinemetrics Truetime 60-DC page. I can open any of the others listed but not the one that I want. Any advice?? Thanks, John in WV----- Original Message ----- From: "Had" To: "Discussion of precise time and frequency measurement" Sent: Sunday, February 15, 2009 10:05 AM Subject: Re: [time-nuts] Kinemetrics Truetime 60-DC > > John, > > Go to www.to-way.com it is in the time-nut page. > > Had > K7MLR > > > At 12:09 PM 2/15/2009, you wrote: > >I am looking for a manual for the Kinemetrics Truetime 60-DC time > >receiver. John Bartholomew surplus1 at prodigy.net Thanks > >_______________________________________________ > >time-nuts mailing list -- time-nuts at febo.com > >To unsubscribe, go to https://www.febo.com/cgi-bin/mailman/listinfo/time-nuts > >and follow the instructions there. > _______________________________________________ > time-nuts mailing list -- time-nuts at febo.com > To unsubscribe, go to https://www.febo.com/cgi-bin/mailman/listinfo/time-nuts > and follow the instructions there. From sam at canardpc.com Sun Feb 15 18:39:07 2009 From: sam at canardpc.com (Samuel D. [x86/CPC]) Date: Sun, 15 Feb 2009 19:39:07 +0100 Subject: [time-nuts] DIY Frequency extension for HP Agilent 53181A, 53131A or 53181A In-Reply-To: References: Message-ID: <002101c98f9c$af9d94e0$0ed8bea0$@com> > If you download the 53132A component level information manual from the > Agilent site you'll find a schematic of the channel 3 board with values and J2 > pinout. The 53131A version of this manual doesn't seem to include this. Excellent ! I only checked the 53131A datasheet without success. I'm checking the schematics... -----Message d'origine----- De?: time-nuts-bounces at febo.com [mailto:time-nuts-bounces at febo.com] De la part de GandalfG8 at aol.com Envoy??: dimanche 15 f?vrier 2009 18:42 ??: time-nuts at febo.com Objet?: Re: [time-nuts] DIY Frequency extension for HP Agilent 53181A, 53131A or 53181A In a message dated 15/02/2009 17:05:21 GMT Standard Time, sam at canardpc.com writes: - What's the divider ratio used ? Perhaps /16 for 3 GHz, /32 for 5 GHz, and /64 for 12.4 GHz, but we really need to be sure. - What's the gain of each one of the four amplifiers present ? - How does the counter aware about the option used ? There is no "intelligence" on the option board, so, the counter may detect the type of prescaler with some pull-up or pull-down resistors somewhere. - What is the exact pinout for the J2 connector ? In order to get an answer, we need to find somebody with a genuine option board willing to spend an hour for the project... ------------------- If you download the 53132A component level information manual from the Agilent site you'll find a schematic of the channel 3 board with values and J2 pinout. The 53131A version of this manual doesn't seem to include this. regards Nigel GM8PZR _______________________________________________ time-nuts mailing list -- time-nuts at febo.com To unsubscribe, go to https://www.febo.com/cgi-bin/mailman/listinfo/time-nuts and follow the instructions there. From mccorkle at ptialaska.net Sun Feb 15 19:01:59 2009 From: mccorkle at ptialaska.net (Richard H McCorkle) Date: Sun, 15 Feb 2009 10:01:59 -0900 (AKST) Subject: [time-nuts] DIY Frequency extension for HP Agilent 53181A, 53131A or 53181A In-Reply-To: <001901c98f8f$76f4cda0$64de68e0$@com> References: <003901c98eda$a1128510$e3378f30$@com> <1872EF473D184CE5A01C929E443207D0@p4253> <43fc33a655d8644ca3c013b6be3b892c.squirrel@webmail.sonic.net> <001901c98f8f$76f4cda0$64de68e0$@com> Message-ID: <16805.206.174.20.67.1234724519.squirrel@mymail.acsalaska.net> Hi Samuel, I suggest starting by downloading the component level information for the 53131A and 53132A from tha Agilent site at the following links. http://cp.literature.agilent.com/litweb/pdf/5989-6308EN.pdf http://cp.literature.agilent.com/litweb/pdf/5989-6307EN.pdf The 53132A document includes a schematic of the channel C board with parts list and may assist in getting you started. Richard > Thanks for your answers. > > About the components, I think my first prototype will use some very > well-known prescaler and amplifiers like Hittite HMC363 and Mini-circuits > ERA-1SM, only to validate the design. I will use a 100 MHz high-pass filter, > but overall, it will be a wide-band, low-sensitivity design, exactly like > the original option. > > But before thinking about the best components to use for the amps and the > divider, we need to know how to interface the prescaler with the counter. > There is some very important questions to answer : > > - What's the divider ratio used ? Perhaps /16 for 3 GHz, /32 for 5 GHz, and > /64 for 12.4 GHz, but we really need to be sure. > - What's the gain of each one of the four amplifiers present ? > - How does the counter aware about the option used ? There is no > "intelligence" on the option board, so, the counter may detect the type of > prescaler with some pull-up or pull-down resistors somewhere. > - What is the exact pinout for the J2 connector ? > > In order to get an answer, we need to find somebody with a genuine option > board willing to spend an hour for the project... > > :-) > > > > -----Message d'origine----- > De?: time-nuts-bounces at febo.com [mailto:time-nuts-bounces at febo.com] De la > part de Rick Karlquist > Envoy??: dimanche 15 f?vrier 2009 03:47 > ??: Discussion of precise time and frequency measurement > Objet?: Re: [time-nuts] DIY Frequency extension for HP Agilent 53181A, > 53131A or 53181A > > Max Skop wrote: >> Hi Samuel and all, >> Welcome to the group. >> I also have a 53131A and would welcome any way to enhance or upgrade the >> performance of this instrument. Please count me in on clone option parts. >> As can be seen from the picure of the 3Ghz option there is nothing special >> in its construction. Just four stages of amplification and a divider. > > In 1987, I was the project manager for the HP 5334B frequency counter. > In those days, we had a 1.3 GHz option called a "C channel". I > evaluated various off the shelf dividers at the time and built > some C channel boards very similar to the one you are describing. > It is actually very difficult to make a prescaler that really works > well using off the shelf dividers. The prescaler tends to be insensitive > to low frequencies. Attempting to fix this with a lot of gain > (4 stages!) has the problem that any wideband noise from the source > is also amplified. Off the shelf dividers are made to work in > frequency synthesizers where they have a clean signal from an oscillator. > They don't do well with general signals from noisy sources. > The HP 5386 used an HP made frequency divider that had "static" > flip flops instead of the "dynamic" ones in all the off the shelf > dividers. It is the only prescaler I am familiar with that actually > works well. The custom IC in it cost HP a fair amount just because it > was "home made" in the Santa Rosa fab, which was a money sink. > > If you want to build your own prescaler, you might not want to copy > the Agilent one. Unless you need the sensitivity, you would be > better off with less gain. Also, you should choose a divider that > works well at the frequency you want to measure. AFAIK, all OTS > dividers have an optimum frequency range, which varies from divider > to divider. > > When testing prescalers, the thing to watch out for is that the > lower order digits will bounce around. This is usually a symptom > of prescaler error, assuming that the source is stable. > > Rick Karlquist N6RK > > > _______________________________________________ > time-nuts mailing list -- time-nuts at febo.com > To unsubscribe, go to > https://www.febo.com/cgi-bin/mailman/listinfo/time-nuts > and follow the instructions there. > > > _______________________________________________ > time-nuts mailing list -- time-nuts at febo.com > To unsubscribe, go to https://www.febo.com/cgi-bin/mailman/listinfo/time-nuts > and follow the instructions there. > From richard at karlquist.com Sun Feb 15 19:39:03 2009 From: richard at karlquist.com (Rick Karlquist) Date: Sun, 15 Feb 2009 11:39:03 -0800 (PST) Subject: [time-nuts] DIY Frequency extension for HP Agilent 53181A, 53131A or 53181A In-Reply-To: <001901c98f8f$76f4cda0$64de68e0$@com> References: <003901c98eda$a1128510$e3378f30$@com> <1872EF473D184CE5A01C929E443207D0@p4253> <43fc33a655d8644ca3c013b6be3b892c.squirrel@webmail.sonic.net> <001901c98f8f$76f4cda0$64de68e0$@com> Message-ID: <9abfe85577d22164ea7067350004c98b.squirrel@webmail.sonic.net> Samuel D. [x86/CPC] wrote: > - How does the counter aware about the option used ? There is no > "intelligence" on the option board, so, the counter may detect the type of > prescaler with some pull-up or pull-down resistors somewhere. Don't be surprised if the option is activated in software. Therefore, even installing the genuine board may not work. Of course, you can always build a prescaler in an external box and feed it into the front panel connector intended for low frequenices. You can use the math functions on the counter to scale the display. Rick Karlquist N6RK From pieter.ibelings at gmail.com Sun Feb 15 19:41:49 2009 From: pieter.ibelings at gmail.com (RFSPACE) Date: Sun, 15 Feb 2009 14:41:49 -0500 Subject: [time-nuts] DIY Frequency extension for HP Agilent 53181A, 53131A or 53181A References: <003901c98eda$a1128510$e3378f30$@com><1872EF473D184CE5A01C929E443207D0@p4253><43fc33a655d8644ca3c013b6be3b892c.squirrel@webmail.sonic.net><001901c98f8f$76f4cda0$64de68e0$@com> <9abfe85577d22164ea7067350004c98b.squirrel@webmail.sonic.net> Message-ID: There is probably a voltage on one of the pins that detects the presence of the board. Pieter ----- Original Message ----- From: "Rick Karlquist" To: "Discussion of precise time and frequency measurement" Sent: Sunday, February 15, 2009 2:39 PM Subject: Re: [time-nuts] DIY Frequency extension for HP Agilent 53181A, 53131A or 53181A > Samuel D. [x86/CPC] wrote: >> - How does the counter aware about the option used ? There is no >> "intelligence" on the option board, so, the counter may detect the type >> of >> prescaler with some pull-up or pull-down resistors somewhere. > > Don't be surprised if the option is activated in software. > Therefore, even installing the genuine board may not work. > Of course, you can always build a prescaler in an external > box and feed it into the front panel connector intended for > low frequenices. You can use the math functions on the counter > to scale the display. > > Rick Karlquist N6RK > > > _______________________________________________ > time-nuts mailing list -- time-nuts at febo.com > To unsubscribe, go to > https://www.febo.com/cgi-bin/mailman/listinfo/time-nuts > and follow the instructions there. From richard at karlquist.com Sun Feb 15 19:53:49 2009 From: richard at karlquist.com (Rick Karlquist) Date: Sun, 15 Feb 2009 11:53:49 -0800 (PST) Subject: [time-nuts] DIY Frequency extension for HP Agilent 53181A, 53131A or 53181A In-Reply-To: <16805.206.174.20.67.1234724519.squirrel@mymail.acsalaska.net> References: <003901c98eda$a1128510$e3378f30$@com> <1872EF473D184CE5A01C929E443207D0@p4253> <43fc33a655d8644ca3c013b6be3b892c.squirrel@webmail.sonic.net> <001901c98f8f$76f4cda0$64de68e0$@com> <16805.206.174.20.67.1234724519.squirrel@mymail.acsalaska.net> Message-ID: <10521b8d00d5278f3bf56d329e34fd84.squirrel@webmail.sonic.net> Richard H McCorkle wrote: > Hi Samuel, > > I suggest starting by downloading the component level information for > the 53131A and 53132A from tha Agilent site at the following links. > > http://cp.literature.agilent.com/litweb/pdf/5989-6308EN.pdf > http://cp.literature.agilent.com/litweb/pdf/5989-6307EN.pdf > > The 53132A document includes a schematic of the channel C board with > parts list and may assist in getting you started. I see that they have a factory select resistor that unbalances the input slightly to keep the prescaler from oscillating with no input. This a simple, obvious "solution" to this problem. It is also non-optimum, for three reasons. It desensitizes the prescaler, it uses a factory select part that may not work consistently over temperature, and it still results in a range of input power where the counter displays garbage. In the 5334B C channel, I used a diode detector that turned on the counter at a repeatable input power. Thus you never got garbage on the display due to low input power. The detector used the well known circuit where you bias the diode on with a few 100 microamps of DC and have a compensating diode that is not excited by RF. A difference amp subtracts the compensating diode voltage. Simple, but effective. It is also interesting to note that in 1987 I used the MB506 prescaler. Here they have simply upgraded to the MB510. Not likely to be any better on noisy sources. Rick Karlquist N6RK From pieter.ibelings at gmail.com Sun Feb 15 20:08:52 2009 From: pieter.ibelings at gmail.com (RFSPACE) Date: Sun, 15 Feb 2009 15:08:52 -0500 Subject: [time-nuts] DIY Frequency extension for HP Agilent 53181A, 53131A or 53181A References: <003901c98eda$a1128510$e3378f30$@com><1872EF473D184CE5A01C929E443207D0@p4253><43fc33a655d8644ca3c013b6be3b892c.squirrel@webmail.sonic.net><001901c98f8f$76f4cda0$64de68e0$@com><16805.206.174.20.67.1234724519.squirrel@mymail.acsalaska.net> <10521b8d00d5278f3bf56d329e34fd84.squirrel@webmail.sonic.net> Message-ID: <69EF7180556A462F819309A4A8CEF07F@xps630> Samuel, The prescaler option is detected by pin 9 and 10 of the 10 pin header. If you look at the page3 of 4 of the 53132A, you see a pair of pins called CH3CODE1 and CH3CODE0. Those tell the processor that the board is plugged in. If you go to the page for the 12.4 GHz prescaler, you can see that the pins 5 and 10 can be pulled high or low to tell the counter what board it is. The pinout on the main board and the prescaler board is different to confuse the hacker. The pinout on the prescaler board is: 6-1 7-2 8-3 9-4 10-5 the pinout on the main board is: 2-1 4-3 6-5 8-7 10-9 This is why the pins 10 and 5 that are used to ID the board, match the schematic pins 9 and 10. It looks really simple to build this board. I would like to add 3GHz to my HP53131A. Currently I use a programmable divider from Hittite. I have it set to /10 and do the math in my head. Pieter, N4IP ----- Original Message ----- From: "Rick Karlquist" To: ; "Discussion of precise time and frequency measurement" Sent: Sunday, February 15, 2009 2:53 PM Subject: Re: [time-nuts] DIY Frequency extension for HP Agilent 53181A, 53131A or 53181A > Richard H McCorkle wrote: >> Hi Samuel, >> >> I suggest starting by downloading the component level information for >> the 53131A and 53132A from tha Agilent site at the following links. >> >> http://cp.literature.agilent.com/litweb/pdf/5989-6308EN.pdf >> http://cp.literature.agilent.com/litweb/pdf/5989-6307EN.pdf >> >> The 53132A document includes a schematic of the channel C board with >> parts list and may assist in getting you started. > > I see that they have a factory select resistor that unbalances > the input slightly to keep the prescaler from oscillating with > no input. This a simple, obvious "solution" to this problem. > It is also non-optimum, for three reasons. It desensitizes the > prescaler, it uses a factory select part that may not work > consistently over temperature, and it still results in a range > of input power where the counter displays garbage. > > In the 5334B C channel, I used a diode detector that turned on > the counter at a repeatable input power. Thus you never got garbage > on the display due to low input power. The detector used the > well known circuit where you bias the diode on with a few 100 > microamps of DC and have a compensating diode that is not excited > by RF. A difference amp subtracts the compensating diode voltage. > Simple, but effective. > > It is also interesting to note that in 1987 I used the MB506 prescaler. > Here they have simply upgraded to the MB510. Not likely to be > any better on noisy sources. > > Rick Karlquist N6RK > > > _______________________________________________ > time-nuts mailing list -- time-nuts at febo.com > To unsubscribe, go to > https://www.febo.com/cgi-bin/mailman/listinfo/time-nuts > and follow the instructions there. From brooke at pacific.net Sun Feb 15 20:49:23 2009 From: brooke at pacific.net (Brooke Clarke) Date: Sun, 15 Feb 2009 12:49:23 -0800 Subject: [time-nuts] DIY Frequency extension for HP Agilent 53181A, 53131A or 53181A In-Reply-To: <002101c98f9c$af9d94e0$0ed8bea0$@com> References: <002101c98f9c$af9d94e0$0ed8bea0$@com> Message-ID: <49987FD3.4030306@pacific.net> Hi Samuel: The 2x5 pin connector pin out is not at all clear. Your Channel 3 photo seems to show: # Chan 3 ---------------- 1 DCOM 2 CH3 Out 3 4 5 6 +12 in 7 8 DCOM 9 10 Fun gnd Your 12.4 Ghz photo has the ribbon cable covering the PCB at the connector so can not be used for pinout info. But the manuals show: # 12.4 G Chan 3 Main Board pin# NG pin# NG Pin# OK ------------------------------------------ 1 DC com PreSsl Out DCOM 2 DC com DC com CH3 in 3 +12 in +12 in DCOM 4 64/65 in A com CH3_DET_EN out 5 Fun Gnd Fun Gnd SW+12V 6 PreSsl Out n.c. SW+12V 7 Opt +5 Opt +5 SW+5V 8 +12 in +12 in SW-12V 9 Fun Opn Fun gnd CH3CODE1 10 Fun gnd Fun gnd CH3CODE0 The manual pin out info for Chan 3 and 12.4 GHz seems bogus. Probably is OK if the connector pin numbering is changed. But haven't figured out the key to that. The photo matches the main board pin out. nx2 IDC connectors are numbered: 1 3 5 7 9 2 4 6 8 1 0 but the cable is numbered (band)1 2 3 4 5 6 7 8 9 10 IC packages are numbered: 1 0 9 8 7 6 1 2 3 4 5 Have Fun, Brooke Clarke http://www.prc68.com Samuel D. [x86/CPC] wrote: >> If you download the 53132A component level information manual from the >> Agilent site you'll find a schematic of the channel 3 board with values > and J2 >> pinout. The 53131A version of this manual doesn't seem to include this. > > Excellent ! I only checked the 53131A datasheet without success. > > I'm checking the schematics... > > > > > -----Message d'origine----- > De : time-nuts-bounces at febo.com [mailto:time-nuts-bounces at febo.com] De la > part de GandalfG8 at aol.com > Envoy? : dimanche 15 f?vrier 2009 18:42 > ? : time-nuts at febo.com > Objet : Re: [time-nuts] DIY Frequency extension for HP Agilent 53181A, > 53131A or 53181A > > > In a message dated 15/02/2009 17:05:21 GMT Standard Time, sam at canardpc.com > writes: > > - What's the divider ratio used ? Perhaps /16 for 3 GHz, /32 for 5 GHz, and > /64 for 12.4 GHz, but we really need to be sure. > - What's the gain of each one of the four amplifiers present ? > - How does the counter aware about the option used ? There is no > "intelligence" on the option board, so, the counter may detect the type of > prescaler with some pull-up or pull-down resistors somewhere. > - What is the exact pinout for the J2 connector ? > > In order to get an answer, we need to find somebody with a genuine option > board willing to spend an hour for the project... > > > > ------------------- > If you download the 53132A component level information manual from the > Agilent site you'll find a schematic of the channel 3 board with values and > J2 > pinout. > The 53131A version of this manual doesn't seem to include this. > > regards > > Nigel > GM8PZR > _______________________________________________ > time-nuts mailing list -- time-nuts at febo.com > To unsubscribe, go to > https://www.febo.com/cgi-bin/mailman/listinfo/time-nuts > and follow the instructions there. > > > _______________________________________________ > time-nuts mailing list -- time-nuts at febo.com > To unsubscribe, go to https://www.febo.com/cgi-bin/mailman/listinfo/time-nuts > and follow the instructions there. > > From sam at canardpc.com Sun Feb 15 21:04:59 2009 From: sam at canardpc.com (Samuel D. [x86/CPC]) Date: Sun, 15 Feb 2009 22:04:59 +0100 Subject: [time-nuts] DIY Frequency extension for HP Agilent 53181A, 53131A or 53181A In-Reply-To: <69EF7180556A462F819309A4A8CEF07F@xps630> References: <003901c98eda$a1128510$e3378f30$@com><1872EF473D184CE5A01C929E443207D0@p4253><43fc33a655d8644ca3c013b6be3b892c.squirrel@webmail.sonic.net><001901c98f8f$76f4cda0$64de68e0$@com><16805.206.174.20.67.1234724519.squirrel@mymail.acsalaska.net> <10521b8d00d5278f3bf56d329e34fd84.squirrel@webmail.sonic.net> <69EF7180556A462F819309A4A8CEF07F@xps630> Message-ID: <003001c98fb1$105cf280$3116d780$@com> You're right. On the J2 connector, there is : * Frequency output : Where the divided input frequency goes. - Pin 2 on counter - Pin 6 on EXT_PCB * CH3CODE0 & CH3CODE1 : Used to detect the type of prescaler - Pin 9&10 on counter - Pin 5&10 on EXT_PCB - 1/1 on 12.4 GHz board - 0/0 on 3 GHz board - 0 = GND / 1 = +5V * CH3_DET_EN : Presence detection. Active low. - Pin 4 on counter - Pin 7 on EXT_PCB - If grounded, CH3 is enabled * 5V line - Pin 7 on counter - Pin 4 on EXT_PCB * 12V line - Pin 5&6 on counter - Pin 3&8 on EXT_PCB * Ground - Pin 1&3 on counter - Pin 1&2 on EXT_PCB The 3 GHz board is built with 4-stage of Avago MSA-0986 (7 dB gain each, rated 0.1 to 3 Ghz) and a fujitsu MB510-PF divider, rated up to 2.7 GHz. Pin 6 & 3 of the MB510-PF are high, so the divide ratio is set to 128. Reproduce this will not be a problem. MSA-0986 costs $4.2 each and MB510-PF is sold for no more than $7. Keep in mind that Agilent sold the board for $2000 ! BTW, I wasn't able to determinate the divider ratio of the 12.4 GHz board : U1 is not on the schematics and U2 is only a 2 GHz divider ML12034-5P from Lansdale Semi (set to /64). I think I will focus on the 3 GHz ext board first. -----Message d'origine----- De?: time-nuts-bounces at febo.com [mailto:time-nuts-bounces at febo.com] De la part de RFSPACE Envoy??: dimanche 15 f?vrier 2009 21:09 ??: richard at karlquist.com; Discussion of precise time and frequency measurement; mccorkle at ptialaska.net Objet?: Re: [time-nuts] DIY Frequency extension for HP Agilent 53181A, 53131A or 53181A Samuel, The prescaler option is detected by pin 9 and 10 of the 10 pin header. If you look at the page3 of 4 of the 53132A, you see a pair of pins called CH3CODE1 and CH3CODE0. Those tell the processor that the board is plugged in. If you go to the page for the 12.4 GHz prescaler, you can see that the pins 5 and 10 can be pulled high or low to tell the counter what board it is. The pinout on the main board and the prescaler board is different to confuse the hacker. The pinout on the prescaler board is: 6-1 7-2 8-3 9-4 10-5 the pinout on the main board is: 2-1 4-3 6-5 8-7 10-9 This is why the pins 10 and 5 that are used to ID the board, match the schematic pins 9 and 10. It looks really simple to build this board. I would like to add 3GHz to my HP53131A. Currently I use a programmable divider from Hittite. I have it set to /10 and do the math in my head. Pieter, N4IP ----- Original Message ----- From: "Rick Karlquist" To: ; "Discussion of precise time and frequency measurement" Sent: Sunday, February 15, 2009 2:53 PM Subject: Re: [time-nuts] DIY Frequency extension for HP Agilent 53181A, 53131A or 53181A > Richard H McCorkle wrote: >> Hi Samuel, >> >> I suggest starting by downloading the component level information for >> the 53131A and 53132A from tha Agilent site at the following links. >> >> http://cp.literature.agilent.com/litweb/pdf/5989-6308EN.pdf >> http://cp.literature.agilent.com/litweb/pdf/5989-6307EN.pdf >> >> The 53132A document includes a schematic of the channel C board with >> parts list and may assist in getting you started. > > I see that they have a factory select resistor that unbalances > the input slightly to keep the prescaler from oscillating with > no input. This a simple, obvious "solution" to this problem. > It is also non-optimum, for three reasons. It desensitizes the > prescaler, it uses a factory select part that may not work > consistently over temperature, and it still results in a range > of input power where the counter displays garbage. > > In the 5334B C channel, I used a diode detector that turned on > the counter at a repeatable input power. Thus you never got garbage > on the display due to low input power. The detector used the > well known circuit where you bias the diode on with a few 100 > microamps of DC and have a compensating diode that is not excited > by RF. A difference amp subtracts the compensating diode voltage. > Simple, but effective. > > It is also interesting to note that in 1987 I used the MB506 prescaler. > Here they have simply upgraded to the MB510. Not likely to be > any better on noisy sources. > > Rick Karlquist N6RK > > > _______________________________________________ > time-nuts mailing list -- time-nuts at febo.com > To unsubscribe, go to > https://www.febo.com/cgi-bin/mailman/listinfo/time-nuts > and follow the instructions there. _______________________________________________ time-nuts mailing list -- time-nuts at febo.com To unsubscribe, go to https://www.febo.com/cgi-bin/mailman/listinfo/time-nuts and follow the instructions there. From flash452 at yahoo.com Sun Feb 15 21:26:09 2009 From: flash452 at yahoo.com (Flash Crews) Date: Sun, 15 Feb 2009 13:26:09 -0800 (PST) Subject: [time-nuts] buyers freewave -ranger -superstar ll -thales B12 -cmc allstar -micro pulse ant Message-ID: <367893.93088.qm@web51601.mail.re2.yahoo.com> hi looking for buyers and maybe price range - used not new freewave p/n DGRC9 p/n ranger-RS p/n DGRC9ras I HAVE OVER 160+ OF THIS ITEM CMC ALLSTAR 12 GPS BASE CARD I HAVE OVER 250+ NOT SURE WHERE TO LOOK FOR PART # SUPERSTAR 2 I HAVE OVER 50+ NOT SURE WHERE TO LOOK FOR PART # THALES B12 BOARD P/N 711823-01 I HAVE OVER 200+ MICRO PULSE GPS ANTENNA P/N 2371LZ I HAVE OVER 150+ I ALSO HAVE RF CARDS 6INCH ROUND GYRO BOARDS -LOOKS LIKE WATER INSIDE PLASTIC BOX From g4hup at btinternet.com Sun Feb 15 21:50:41 2009 From: g4hup at btinternet.com (dave powis) Date: Sun, 15 Feb 2009 21:50:41 +0000 (GMT) Subject: [time-nuts] New Distribution Amplifier Kit Message-ID: <557062.46271.qm@web86304.mail.ird.yahoo.com> Hello all, Just to let you know that I've added a 4 way distribution amplifier kit to my products at http://g4hup.com/DA/DA1_4.htm. I'll also openly tell you that is not intended as a high quality lab instrument, but is a simple general purpose tool intended for distribution of signals anywhere up to 1800MHz, where phase and output level accuracy variations are not very critical. There is (deliberately) no frequency tailoring or transformer isolation provided in the unit, but details of how to achieve this externally are given in the documentation for the unit. It is ideal for sending 10MHz to your test equipment stack for general purpose measurements, or for distributing an rf signal to a number of destinations, such as sharing an IF signal between conventional and SDR receivers. Having run it down in the last paragraph, when driven from a nominal +10dBm source the measured phase variation between the outputs at 10MHz is less than 3 degrees, and it provides 4 +10dBm outputs within +/- 0.25dB. There is also a variant that will provide distribution of GPS antenna signals, including providing power feed for the active GPS antenna itself. For those who may not want to know about this, my apologies for the bandwidth! regards, Dave, G4HUP http://g4hup.com From bruce.griffiths at xtra.co.nz Sun Feb 15 22:41:30 2009 From: bruce.griffiths at xtra.co.nz (Bruce Griffiths) Date: Mon, 16 Feb 2009 11:41:30 +1300 Subject: [time-nuts] buyers freewave -ranger -superstar ll -thales B12 -cmc allstar -micro pulse ant In-Reply-To: <367893.93088.qm@web51601.mail.re2.yahoo.com> References: <367893.93088.qm@web51601.mail.re2.yahoo.com> Message-ID: <49989A1A.7060100@xtra.co.nz> For the Superstar II: Probably the only way to tell is to 1) determine if the model is a 3.3V or a 5V version Maybe the ARM chip part No. or the 10MHz on board TCVXO part No. will help with this. 2) Power it up and query using Starview. Message 45 can be used to tell how much memory it has. These come in SBAS versions and non SBAS versions. Some models have carrier phase output data available, some do not. The PPS timing performance isnt stellar, but if the carrier phase data is used a high performance GPSDO is possible (requires replacing the on board TCVXO output with 10MHz from the OCXO being disciplined). Bruce Flash Crews wrote: > hi looking for buyers and maybe price range - used not new > > > freewave p/n DGRC9 p/n ranger-RS p/n DGRC9ras I HAVE OVER 160+ OF THIS ITEM > > > > CMC ALLSTAR 12 GPS BASE CARD I HAVE OVER 250+ > NOT SURE WHERE TO LOOK FOR PART # > > SUPERSTAR 2 I HAVE OVER 50+ > NOT SURE WHERE TO LOOK FOR PART # > > > THALES B12 BOARD P/N 711823-01 I HAVE OVER 200+ > > MICRO PULSE GPS ANTENNA P/N 2371LZ I HAVE OVER 150+ > > I ALSO HAVE > > RF CARDS > 6INCH ROUND GYRO BOARDS -LOOKS LIKE WATER INSIDE PLASTIC BOX > > > > > > > _______________________________________________ > time-nuts mailing list -- time-nuts at febo.com > To unsubscribe, go to https://www.febo.com/cgi-bin/mailman/listinfo/time-nuts > and follow the instructions there. > > From SAIDJACK at aol.com Sun Feb 15 23:24:22 2009 From: SAIDJACK at aol.com (SAIDJACK at aol.com) Date: Sun, 15 Feb 2009 18:24:22 EST Subject: [time-nuts] Tbolt temperature sensor Message-ID: Hello Tom, over the last week we've been doing a test on a double oven Fury GPSDO in response to our discussion. The first part of the test was running the unit on the lab bench without modifications. A couple of days ago, we removed the OCXO Supply regulator feedback sense resistor, and put the sense point right at the VCC pin of the OCXO, therefore Kelvin-sensing the supply voltage behind the current sense resistor, and thus having an extremely well regulated voltage directly on the OCXO, without any VCC current-drop related changes. This was to see if there is any effect on the thermal stability caused by the voltage drop across the Power supply sense resistor, traces, filters, etc. This drop is over 50mV worst case on an un-modified unit. You can see the before and after effects on the plot here: _http://www.jackson-labs.com/images/gpsstat.htm_ (http://www.jackson-labs.com/images/gpsstat.htm) You can see the point where we switched the sense resistor as a discontinuity in the EFC voltage. The measured tempco of the unit before and after did not change, it is about 6.0 in both cases (which is a very good number on this particular unit). The unit's performance as measured by GPSCon before and after the modification is also identical, it measures an average of 0.05ns, and 2ns standard deviation. The worst-case deviation is +9ns, and -8ns. Thus on this particular unit there is no measurable effect from the voltage drop between the power supply regulator and the OCXO power supply pin, and in both cases the measured tempco is around 6, which is very good, and means there is very little thermally-induced EFC voltage change required. Keeping in mind that our double oven OCXOs have about 0.12A to 0.15A current range (a 30mA+ delta), and the single oven OCXO's have only about 4mA current change (about 10x less), then we can deduce that the 10x smaller current change in our single-oven OCXO's will have even less VCC/Ground current-loop induced error. bye, Said In a message dated 2/4/2009 17:51:17 Pacific Standard Time, tvb at LeapSecond.com writes: > To make this work somewhat smoothly, one would need at least 0.001C or > better 0.0001C resolution. Hi Said, Can you provide some real data, to help me believe this claim? When you get below 1C or 0.1C resolution it starts to matter from which direction the temperature gradient is headed, or from which angle air is flowing. Or which side is up. Or what the humidity is, etc. This, because temperature gradients break any steady state model you have based on a point source (e.g., temp sensor) or average temperature measurement (e.g., oven current). Then there is the matter of the rate at which temperature changes; slow temp changes and rapid temp changes affect an OCXO quite differently. This, due to different thermal time constants of all the metal and insulation materials in and around the OCXO or GPSDO. > To give you an example: a typical single-oven OCXO has about 1ppb per degree > C change. This would mean the unit could only be adjusted by 2.5E-010 steps > with the Dallas Temp sensor as a reference. So the average error would be > 1.25E-010, which results in a massive 12.5 microseconds average drift error in > 1000s intervals due to the temperature quantization error! Are you assuming a design where one takes a Dallas reading each second and stuffs it into the EFC every second? No one would actually do that. Instead if one averages the temperature sensor over 10, 100 or more seconds you avoid large EFC steps. Everything else in a GPSDO is all about slow averaging; there's no reason temperature adaptation cannot be treated the same. I suppose I should add a random temperature cycle hold-over test to the abuse I inflict on each GPSDO here. Are you saying the Fury would do much better than others in this regard? Again, given this is time-nuts, some real data would be nice. /tvb _______________________________________________ time-nuts mailing list -- time-nuts at febo.com To unsubscribe, go to https://www.febo.com/cgi-bin/mailman/listinfo/time-nuts and follow the instructions there. From had at to-way.com Mon Feb 16 00:01:56 2009 From: had at to-way.com (Had) Date: Sun, 15 Feb 2009 16:01:56 -0800 Subject: [time-nuts] Kinemetrics Truetime 60-DC In-Reply-To: <00cf01c98fb3$a074d8e0$90eef804@pavilion> References: <00b001c98fa9$5825ff60$90eef804@pavilion> <20090215180556.25D18E0480D@mail-in03.adhost.com> <00cf01c98fb3$a074d8e0$90eef804@pavilion> Message-ID: <20090216000200.52F618ADF27@mail-in07.adhost.com> John, What version of Adobe reader are you using. Should be 7.0 or later. The pdf was created with Adobe Pro 7.something and opens just fine in Reader 8 or 9. Had At 01:23 PM 2/15/2009, you wrote: >Hello, For some reason I cannot open the Kinemetrics Truetime 60-DC page. >I can open any of the others listed but not the one that I want. Any >advice?? Thanks, John in WV----- Original Message ----- >From: "Had" >To: "Discussion of precise time and frequency measurement" > >Sent: Sunday, February 15, 2009 10:05 AM >Subject: Re: [time-nuts] Kinemetrics Truetime 60-DC > > > > > > John, > > From sam at canardpc.com Mon Feb 16 00:33:44 2009 From: sam at canardpc.com (Samuel D. [x86/CPC]) Date: Mon, 16 Feb 2009 01:33:44 +0100 Subject: [time-nuts] DIY Frequency extension for HP Agilent 53181A, 53131A or 53181A In-Reply-To: <003001c98fb1$105cf280$3116d780$@com> References: <003901c98eda$a1128510$e3378f30$@com><1872EF473D184CE5A01C929E443207D0@p4253><43fc33a655d8644ca3c013b6be3b892c.squirrel@webmail.sonic.net><001901c98f8f$76f4cda0$64de68e0$@com><16805.206.174.20.67.1234724519.squirrel@mymail.acsalaska.net> <10521b8d00d5278f3bf56d329e34fd84.squirrel@webmail.sonic.net> <69EF7180556A462F819309A4A8CEF07F@xps630> <003001c98fb1$105cf280$3116d780$@com> Message-ID: <003c01c98fce$399f0e90$acdd2bb0$@com> About prescaler : Well, that's not as simple as I first thought. First, I need a prescaler able to ramp from 100 MHz to 3+ GHz. And all prescalers I saw come with horrible sensitivity in the 0.1-0.4 GHz range. More, they are all obsoletes and not produced anymore. Now, I understand the choice of Agilent to use a massive 4-stage amplifier in order to feed the prescaler with a usable signal in the 0.1-0.4 GHz range. So, there is two choices : - Rebuild the exact replica of the original 3-GHz option. The amplifiers are still produced by Avago and I can source 10- or 20- MB510-PF on a old-parts broker. - Build an entirely new design with a 2-stage divider : First Hittite 13 GHz /8 divider, then any 2 GHz divider. With some good shielding and a bit of amp, it may work fine as a 13 GHz prescaler. I think I will try the first option first, and, if all goes well, switching to the 13 GHz part. Now, I need to source some MB510-PF and design a PCB. I think I will sold the extension as a kit. I just can't solder all the SMD for everybody, but the kit will include all the components, including the passives components and will be sold at a VERY reasonable price. -----Message d'origine----- De?: time-nuts-bounces at febo.com [mailto:time-nuts-bounces at febo.com] De la part de Samuel D. [x86/CPC] Envoy??: dimanche 15 f?vrier 2009 22:05 ??: 'Discussion of precise time and frequency measurement' Objet?: Re: [time-nuts] DIY Frequency extension for HP Agilent 53181A, 53131A or 53181A You're right. On the J2 connector, there is : * Frequency output : Where the divided input frequency goes. - Pin 2 on counter - Pin 6 on EXT_PCB * CH3CODE0 & CH3CODE1 : Used to detect the type of prescaler - Pin 9&10 on counter - Pin 5&10 on EXT_PCB - 1/1 on 12.4 GHz board - 0/0 on 3 GHz board - 0 = GND / 1 = +5V * CH3_DET_EN : Presence detection. Active low. - Pin 4 on counter - Pin 7 on EXT_PCB - If grounded, CH3 is enabled * 5V line - Pin 7 on counter - Pin 4 on EXT_PCB * 12V line - Pin 5&6 on counter - Pin 3&8 on EXT_PCB * Ground - Pin 1&3 on counter - Pin 1&2 on EXT_PCB The 3 GHz board is built with 4-stage of Avago MSA-0986 (7 dB gain each, rated 0.1 to 3 Ghz) and a fujitsu MB510-PF divider, rated up to 2.7 GHz. Pin 6 & 3 of the MB510-PF are high, so the divide ratio is set to 128. Reproduce this will not be a problem. MSA-0986 costs $4.2 each and MB510-PF is sold for no more than $7. Keep in mind that Agilent sold the board for $2000 ! BTW, I wasn't able to determinate the divider ratio of the 12.4 GHz board : U1 is not on the schematics and U2 is only a 2 GHz divider ML12034-5P from Lansdale Semi (set to /64). I think I will focus on the 3 GHz ext board first. -----Message d'origine----- De?: time-nuts-bounces at febo.com [mailto:time-nuts-bounces at febo.com] De la part de RFSPACE Envoy??: dimanche 15 f?vrier 2009 21:09 ??: richard at karlquist.com; Discussion of precise time and frequency measurement; mccorkle at ptialaska.net Objet?: Re: [time-nuts] DIY Frequency extension for HP Agilent 53181A, 53131A or 53181A Samuel, The prescaler option is detected by pin 9 and 10 of the 10 pin header. If you look at the page3 of 4 of the 53132A, you see a pair of pins called CH3CODE1 and CH3CODE0. Those tell the processor that the board is plugged in. If you go to the page for the 12.4 GHz prescaler, you can see that the pins 5 and 10 can be pulled high or low to tell the counter what board it is. The pinout on the main board and the prescaler board is different to confuse the hacker. The pinout on the prescaler board is: 6-1 7-2 8-3 9-4 10-5 the pinout on the main board is: 2-1 4-3 6-5 8-7 10-9 This is why the pins 10 and 5 that are used to ID the board, match the schematic pins 9 and 10. It looks really simple to build this board. I would like to add 3GHz to my HP53131A. Currently I use a programmable divider from Hittite. I have it set to /10 and do the math in my head. Pieter, N4IP ----- Original Message ----- From: "Rick Karlquist" To: ; "Discussion of precise time and frequency measurement" Sent: Sunday, February 15, 2009 2:53 PM Subject: Re: [time-nuts] DIY Frequency extension for HP Agilent 53181A, 53131A or 53181A > Richard H McCorkle wrote: >> Hi Samuel, >> >> I suggest starting by downloading the component level information for >> the 53131A and 53132A from tha Agilent site at the following links. >> >> http://cp.literature.agilent.com/litweb/pdf/5989-6308EN.pdf >> http://cp.literature.agilent.com/litweb/pdf/5989-6307EN.pdf >> >> The 53132A document includes a schematic of the channel C board with >> parts list and may assist in getting you started. > > I see that they have a factory select resistor that unbalances > the input slightly to keep the prescaler from oscillating with > no input. This a simple, obvious "solution" to this problem. > It is also non-optimum, for three reasons. It desensitizes the > prescaler, it uses a factory select part that may not work > consistently over temperature, and it still results in a range > of input power where the counter displays garbage. > > In the 5334B C channel, I used a diode detector that turned on > the counter at a repeatable input power. Thus you never got garbage > on the display due to low input power. The detector used the > well known circuit where you bias the diode on with a few 100 > microamps of DC and have a compensating diode that is not excited > by RF. A difference amp subtracts the compensating diode voltage. > Simple, but effective. > > It is also interesting to note that in 1987 I used the MB506 prescaler. > Here they have simply upgraded to the MB510. Not likely to be > any better on noisy sources. > > Rick Karlquist N6RK > > > _______________________________________________ > time-nuts mailing list -- time-nuts at febo.com > To unsubscribe, go to > https://www.febo.com/cgi-bin/mailman/listinfo/time-nuts > and follow the instructions there. _______________________________________________ time-nuts mailing list -- time-nuts at febo.com To unsubscribe, go to https://www.febo.com/cgi-bin/mailman/listinfo/time-nuts and follow the instructions there. _______________________________________________ time-nuts mailing list -- time-nuts at febo.com To unsubscribe, go to https://www.febo.com/cgi-bin/mailman/listinfo/time-nuts and follow the instructions there. From jltran at worldnet.att.net Mon Feb 16 01:18:53 2009 From: jltran at worldnet.att.net (J. L. Trantham) Date: Sun, 15 Feb 2009 19:18:53 -0600 Subject: [time-nuts] DIY Frequency extension for HP Agilent 53181A, 53131A or 53181A In-Reply-To: <10521b8d00d5278f3bf56d329e34fd84.squirrel@webmail.sonic.net> Message-ID: <6F4BE70DC59F4758A71F2AC9342617CA@S0028384766> Rick, I have been 'reading the mail' on this discussion and have a somewhat off topic question. I acquired a 5334B that has a 'noisy' channel C. It originally did not read when fed a signal. I then adjusted R328, the peak detector, and then it correctly read the signal. However, when I disconnect the signal, it continues to 'gate' and read some numbers. Any thoughts? Also, is it possible to add a 'channel C' to a unit that does not have it. I have noticed that the unit I have without a channel C has most (if not all) of the parts populated on the board except for the 8 pin IC that plugs into the socket, U302, which appears to be an MB506. Is all that is required is to plug in an MB506 and add a connector to the front panel, or is there a firmware issue as well? If a firmware issue, is it possible to 'rewrite' an EPROM or some such to deal with this? Thanks, Joe -----Original Message----- From: time-nuts-bounces at febo.com [mailto:time-nuts-bounces at febo.com] On Behalf Of Rick Karlquist Sent: Sunday, February 15, 2009 1:54 PM To: mccorkle at ptialaska.net; Discussion of precise time and frequency measurement Subject: Re: [time-nuts] DIY Frequency extension for HP Agilent 53181A, 53131A or 53181A Richard H McCorkle wrote: > Hi Samuel, > > I suggest starting by downloading the component level information for > the 53131A and 53132A from tha Agilent site at the following links. > > http://cp.literature.agilent.com/litweb/pdf/5989-6308EN.pdf > http://cp.literature.agilent.com/litweb/pdf/5989-6307EN.pdf > > The 53132A document includes a schematic of the channel C board with > parts list and may assist in getting you started. I see that they have a factory select resistor that unbalances the input slightly to keep the prescaler from oscillating with no input. This a simple, obvious "solution" to this problem. It is also non-optimum, for three reasons. It desensitizes the prescaler, it uses a factory select part that may not work consistently over temperature, and it still results in a range of input power where the counter displays garbage. In the 5334B C channel, I used a diode detector that turned on the counter at a repeatable input power. Thus you never got garbage on the display due to low input power. The detector used the well known circuit where you bias the diode on with a few 100 microamps of DC and have a compensating diode that is not excited by RF. A difference amp subtracts the compensating diode voltage. Simple, but effective. It is also interesting to note that in 1987 I used the MB506 prescaler. Here they have simply upgraded to the MB510. Not likely to be any better on noisy sources. Rick Karlquist N6RK _______________________________________________ time-nuts mailing list -- time-nuts at febo.com To unsubscribe, go to https://www.febo.com/cgi-bin/mailman/listinfo/time-nuts and follow the instructions there. From grant at ghengineering.co.uk Mon Feb 16 12:54:27 2009 From: grant at ghengineering.co.uk (Grant Hodgson) Date: Mon, 16 Feb 2009 12:54:27 +0000 Subject: [time-nuts] DIY Frequency extension for HP Agilent In-Reply-To: References: Message-ID: <49996203.80106@ghengineering.co.uk> Samuel The NLB-300 and NBB-300 series MMICs from RFMD work very at up to 13GHz or so, certainly far better than a Mini-circuits ERA- whose gain will start to roll off at higher frequencies. Designing a broadband gain stage to work up to 12GHz+ may not be as easy as it seems. However it is made somewhat easier by the use of broadband chokes from the likes of Mini-Circuits, wideband inductors from Coilcraft and broadband capacitors from ATC - to name but three - there may be others as well. The 4-stage amplifier used by HP not only increases the signal level, it increases the slew rate (V/uSec) at lower frequencies. It's not just about signal level, it's about getting fast edges. The Hittite HMC363 /8 is a great IC, using static flip-flops, and will work all the way down to DC - IF the edges are fast enough. Ironically it may be the case that the lower frequency limit presents more of a challenge than the higher frequency limit. regards Grant From david.partridge at dsl.pipex.com Mon Feb 16 14:17:04 2009 From: david.partridge at dsl.pipex.com (David C. Partridge) Date: Mon, 16 Feb 2009 14:17:04 -0000 Subject: [time-nuts] DIY Frequency extension for HP Agilent In-Reply-To: <49996203.80106@ghengineering.co.uk> References: <49996203.80106@ghengineering.co.uk> Message-ID: <3C656C7CC2B348F9A5CA10B01D5D4AFE@APOLLO> An option might be to use a comparator somewhere in the circuit to "square up" the input signal. The only question would be to find one that will work happily at that sort of frequency without bursting into song, or sitting there like a lemon (i.e. doing not a lot). Certainly the ADCMP600 (or MAX999) won't hack it. The other alternative I can think of is to over-amplify the signal and then use a (very) high speed Schottky diode limiter to clip it. Dave -----Original Message----- From: time-nuts-bounces at febo.com [mailto:time-nuts-bounces at febo.com] On Behalf Of Grant Hodgson Sent: 16 February 2009 12:54 To: Discussion of precise time and frequency measurement Subject: Re: [time-nuts] DIY Frequency extension for HP Agilent Samuel The NLB-300 and NBB-300 series MMICs from RFMD work very at up to 13GHz or so, certainly far better than a Mini-circuits ERA- whose gain will start to roll off at higher frequencies. Designing a broadband gain stage to work up to 12GHz+ may not be as easy as it seems. However it is made somewhat easier by the use of broadband chokes from the likes of Mini-Circuits, wideband inductors from Coilcraft and broadband capacitors from ATC - to name but three - there may be others as well. The 4-stage amplifier used by HP not only increases the signal level, it increases the slew rate (V/uSec) at lower frequencies. It's not just about signal level, it's about getting fast edges. The Hittite HMC363 /8 is a great IC, using static flip-flops, and will work all the way down to DC - IF the edges are fast enough. Ironically it may be the case that the lower frequency limit presents more of a challenge than the higher frequency limit. regards Grant _______________________________________________ time-nuts mailing list -- time-nuts at febo.com To unsubscribe, go to https://www.febo.com/cgi-bin/mailman/listinfo/time-nuts and follow the instructions there. From richard at karlquist.com Mon Feb 16 14:47:51 2009 From: richard at karlquist.com (Rick Karlquist) Date: Mon, 16 Feb 2009 06:47:51 -0800 (PST) Subject: [time-nuts] DIY Frequency extension for HP Agilent In-Reply-To: <49996203.80106@ghengineering.co.uk> References: <49996203.80106@ghengineering.co.uk> Message-ID: <6656fc48964cefad9953b3947186a020.squirrel@webmail.sonic.net> Grant Hodgson wrote: Ironically it may be the > case that the lower frequency limit presents more of a challenge than > the higher frequency limit. > > regards > > Grant Yes, as a C-channel designer at HP, and also having worked with other C-channel designers, I can verify that the lower frequency limit is a big challenge. Rick Karlquist N6RK From holrum at hotmail.com Mon Feb 16 18:56:08 2009 From: holrum at hotmail.com (Mark Sims) Date: Mon, 16 Feb 2009 18:56:08 +0000 Subject: [time-nuts] Tbolt temperature sensor In-Reply-To: References: Message-ID: I found this bit 'o code for reading the DS1620 temperature sensor chip. The comments at the beginning explain what is probably going on with the newer Thunderbolts. Trimble is apparently doing the "old style" register reads and cannot access the high resolution counter register in the newer DS1620 chips. http://www.hawo.stw.uni-erlangen.de/misc/temperatur/dstempd.c Also the older chips returned a (temperature dependent) divisor value that could apparently yield around 0.01C resolution. The newer chips return a divisor that is fixed at 16... 0.0625C resolution. _________________________________________________________________ See how Windows Mobile brings your life together?at home, work, or on the go. http://clk.atdmt.com/MRT/go/msnnkwxp1020093182mrt/direct/01/ From bruce.griffiths at xtra.co.nz Mon Feb 16 18:58:44 2009 From: bruce.griffiths at xtra.co.nz (Bruce Griffiths) Date: Tue, 17 Feb 2009 07:58:44 +1300 Subject: [time-nuts] DIY Frequency extension for HP Agilent In-Reply-To: <49996203.80106@ghengineering.co.uk> References: <49996203.80106@ghengineering.co.uk> Message-ID: <4999B764.1090308@xtra.co.nz> Grant Hodgson wrote: > Samuel > > The NLB-300 and NBB-300 series MMICs from RFMD work very at up to 13GHz > or so, certainly far better than a Mini-circuits ERA- whose gain will > start to roll off at higher frequencies. Designing a broadband gain > stage to work up to 12GHz+ may not be as easy as it seems. However it > is made somewhat easier by the use of broadband chokes from the likes of > Mini-Circuits, wideband inductors from Coilcraft and broadband > capacitors from ATC - to name but three - there may be others as well. > > The 4-stage amplifier used by HP not only increases the signal level, it > increases the slew rate (V/uSec) at lower frequencies. It's not just > about signal level, it's about getting fast edges. The Hittite HMC363 > /8 is a great IC, using static flip-flops, and will work all the way > down to DC - IF the edges are fast enough. Ironically it may be the > case that the lower frequency limit presents more of a challenge than > the higher frequency limit. > > regards > > Grant > > > The Piconics broadband iron powder filled conical inductors supposedly work from 10MHz or so to 40GHz. Bruce From jmiles at pop.net Mon Feb 16 19:09:47 2009 From: jmiles at pop.net (John Miles) Date: Mon, 16 Feb 2009 11:09:47 -0800 Subject: [time-nuts] DIY Frequency extension for HP Agilent In-Reply-To: <49996203.80106@ghengineering.co.uk> Message-ID: > > The NLB-300 and NBB-300 series MMICs from RFMD work very at up to 13GHz > or so, certainly far better than a Mini-circuits ERA- whose gain will > start to roll off at higher frequencies. Designing a broadband gain > stage to work up to 12GHz+ may not be as easy as it seems. However it > is made somewhat easier by the use of broadband chokes from the likes of > Mini-Circuits, wideband inductors from Coilcraft and broadband > capacitors from ATC - to name but three - there may be others as well. > > The 4-stage amplifier used by HP not only increases the signal level, it > increases the slew rate (V/uSec) at lower frequencies. It's not just > about signal level, it's about getting fast edges. The Hittite HMC363 > /8 is a great IC, using static flip-flops, and will work all the way > down to DC - IF the edges are fast enough. Ironically it may be the > case that the lower frequency limit presents more of a challenge than > the higher frequency limit. True. The Hittite parts can be used all the way down to DC, if the edges are fast enough. :) The last time I wanted to build a prescaler capable of DC operation, I ended up using a tunnel diode to speed up the edges of low-frequency input signals. I used some crude LC filtering to bypass the TD for faster signals, routing them directly to the prescaler chip. The circuit worked well enough, but it was pretty twitchy. -- john, KE5FX From mpb45 at clanbaker.org Mon Feb 16 19:27:46 2009 From: mpb45 at clanbaker.org (Michael Baker) Date: Mon, 16 Feb 2009 14:27:46 -0500 Subject: [time-nuts] Lady Heather display is GREAT!!! Message-ID: <4999BE32.50100@clanbaker.org> Hello, John and Mark-- The display is beautiful and tells a great deal more from a quick glance than the Trimble factory supplied software. You guys did a great job with it!! The display looks really good when I do an ALT / ENTER to run the display full screen. This display is the "cat's meow" and I will likely dedicate an old laptop I have sitting around to running it and leave it running full time on the shelf in my shop. Thanks again for all your work on this-- You and Mark have done all us T-Bolt users a great favor!! I am wondering about the big empty space on the lower left center of the display.... Is this for some sort of future embellishment? Hmmmmmm..... How about a very large clock ticking away in that space or some other sort of running information....? Mike Baker -------------------------------- From jmiles at pop.net Mon Feb 16 19:31:56 2009 From: jmiles at pop.net (John Miles) Date: Mon, 16 Feb 2009 11:31:56 -0800 Subject: [time-nuts] Lady Heather display is GREAT!!! In-Reply-To: <4999BE32.50100@clanbaker.org> Message-ID: Hmm... in /vl mode there might be room for a large-format digital clock display. Not quite enough free space in the normal or /vs modes, I don't think. -- john, KE5FX > -----Original Message----- > From: time-nuts-bounces at febo.com [mailto:time-nuts-bounces at febo.com]On > Behalf Of Michael Baker > Sent: Monday, February 16, 2009 11:28 AM > To: time-nuts at febo.com > Subject: [time-nuts] Lady Heather display is GREAT!!! > > > Hello, John and Mark-- > > The display is beautiful and tells a great deal more > from a quick glance than the Trimble factory supplied > software. You guys did a great job with it!! > > The display looks really good when I do an ALT / ENTER > to run the display full screen. This display is the > "cat's meow" and I will likely dedicate an old laptop > I have sitting around to running it and leave it running > full time on the shelf in my shop. > > Thanks again for all your work on this-- You and Mark have > done all us T-Bolt users a great favor!! > > I am wondering about the big empty space on the lower > left center of the display.... Is this for some sort > of future embellishment? Hmmmmmm..... How about a > very large clock ticking away in that space or some other > sort of running information....? > > Mike Baker > -------------------------------- > > > _______________________________________________ > time-nuts mailing list -- time-nuts at febo.com > To unsubscribe, go to > https://www.febo.com/cgi-bin/mailman/listinfo/time-nuts > and follow the instructions there. > From holrum at hotmail.com Mon Feb 16 20:05:03 2009 From: holrum at hotmail.com (Mark Sims) Date: Mon, 16 Feb 2009 20:05:03 +0000 Subject: [time-nuts] Lady Heather display is GREAT!!! In-Reply-To: References: Message-ID: If you happen to have a Thunderbolt-E four of the lines in that space would be taken up by the four additional satellites that the Thunderbolt-E can track... ---------------------------------------- _________________________________________________________________ See how Windows Mobile brings your life together?at home, work, or on the go. http://clk.atdmt.com/MRT/go/msnnkwxp1020093182mrt/direct/01/ From alf.pousar at surffi.net Mon Feb 16 20:59:45 2009 From: alf.pousar at surffi.net (Alf Pousar) Date: Mon, 16 Feb 2009 22:59:45 +0200 Subject: [time-nuts] The best timing GPS receiver now on the market? Message-ID: Hello folks, can anyone tell me what or whom is producing, making the most accurate 1pps gps receiver now available. I have a Navman Jupiter-T TU60-D120 . It was about the best you could get in 2004. My GSPDO is ticking nicely in the Brooks Shera design/configuration. I just got one calibrated Rubidium oscillator made by (?) Datum model LPRO. Once you understand the physics behind it, its fascinating to watch it lock. All that quantum physics in such a small package!! 73, Alf OH2QM From bruce.griffiths at xtra.co.nz Mon Feb 16 21:36:38 2009 From: bruce.griffiths at xtra.co.nz (Bruce Griffiths) Date: Tue, 17 Feb 2009 10:36:38 +1300 Subject: [time-nuts] The best timing GPS receiver now on the market? In-Reply-To: References: Message-ID: <4999DC66.9070300@xtra.co.nz> Alf Pousar wrote: > Hello folks, > > can anyone tell me what or whom is producing, making the most accurate 1pps gps receiver now available. > I have a Navman Jupiter-T TU60-D120 . It was about the best you could get in 2004. > > My GSPDO is ticking nicely in the Brooks Shera design/configuration. > I just got one calibrated Rubidium oscillator made by (?) Datum model LPRO. > Once you understand the physics behind it, its fascinating to watch it lock. All that quantum physics in such a small package!! > > 73, Alf OH2QM > The iLotus M12M Timing receiver?? http://www.synergy-gps.com/content/view/20/34/ Bruce From mpb45 at clanbaker.org Tue Feb 17 01:05:47 2009 From: mpb45 at clanbaker.org (Michael Baker) Date: Mon, 16 Feb 2009 20:05:47 -0500 Subject: [time-nuts] Thunderbolt E vs Lady Heather open space.... Message-ID: <499A0D6B.5000805@clanbaker.org> Time-Nutters-- RE: Utilizing the large open space in the Lady Heather display for a (large) running clock or other application... Mark said: If you happen to have a Thunderbolt-E four of the lines in that space would be taken up by the four additional satellites that the Thunderbolt-E can track... ---------------------------------------- True, but it appears that that there would still be a rather large open space under the Cursor GPS Time indicator... Mike Baker WA4HFR Gainesville (Micanopy) Florida, USA From d.seiter at comcast.net Tue Feb 17 02:28:06 2009 From: d.seiter at comcast.net (d.seiter at comcast.net) Date: Tue, 17 Feb 2009 02:28:06 +0000 (UTC) Subject: [time-nuts] Lady Heather display is GREAT!!! In-Reply-To: <1399882505.646001234837380352.JavaMail.root@sz0108a.emeryville.ca.mail.comcast.net> Message-ID: <1205179890.648641234837686163.JavaMail.root@sz0108a.emeryville.ca.mail.comcast.net> Lady Heather is great; I finally had a chance to try her out. However, about 2 hours into my test it started loosing sats (indoor antenna, 3 to 6 sats at any given time; it lost 5 over the course of about 2 minutes) and now the tbolt won't acquire anything. This, of course has nothing to do with LH, but I'm wondering if it went belly-up. I'm going to let it cool down completely and try restarting. The only alarms it displays are discipline and satellite. When it died I was away for a few minutes, and nothing touched it. I tried a warm restart, and I saw it try to grab a sat once over a 20 minute span. I haven't used it in months, but it has always been reliable. -Dave From holrum at hotmail.com Tue Feb 17 04:04:37 2009 From: holrum at hotmail.com (Mark Sims) Date: Tue, 17 Feb 2009 04:04:37 +0000 Subject: [time-nuts] Lady Heather display is GREAT!!! In-Reply-To: References: Message-ID: That sounds like your indoor antenna is the problem. I have had very poor luck with indoor antennas. Even my outdoor antenna is rather bad... too many very big trees. I get constant satellite constellation changes (sometimes several times a minute). My first self survey took three days to complete. Where my antenna is now, I can complete a survey in a couple of hours (but at only a couple of times each day). Also many GPS receivers tend to have a hard time starting up when they have not been used for a while. It takes a long time for the oscillator to stabilize enough to work properly. I had a tbolt that had not been used for a couple of years. It took about a week before it settled down enough to be really usable. ---------------------------------------- However, about 2 hours into my test it started loosing sats (indoor antenna, 3 to 6 sats at any given time; _________________________________________________________________ See how Windows Mobile brings your life together?at home, work, or on the go. http://clk.atdmt.com/MRT/go/msnnkwxp1020093182mrt/direct/01/ From jltran at worldnet.att.net Tue Feb 17 05:10:44 2009 From: jltran at worldnet.att.net (J. L. Trantham) Date: Mon, 16 Feb 2009 23:10:44 -0600 Subject: [time-nuts] Ovenaire OSC 49-38B, EIP Part Number 2030010-03 Message-ID: I found a pin out of this part but does anyone have a circuit diagram? It is a 10 MHz OCXO with a daily aging rate of <+/- 1 ppb after 72 hours of operation. Thanks in advance. Joe From jim77742 at gmail.com Tue Feb 17 10:13:24 2009 From: jim77742 at gmail.com (Jim Palfreyman) Date: Tue, 17 Feb 2009 21:13:24 +1100 Subject: [time-nuts] Thunderbolt power supply open heart surgery Message-ID: Hi folks, Well I woke up the other morning at 2am to find all the power off in the house. I was about to ring the power company and thought I'd better check my meter box. Turns out one of the RCDs had tripped. I reset it and all was fine. What puzzled me was what tripped it. It took out my workshop, speaking clock, various GPSDOs and an HP rubidium I've borrowed. However all was working upon reset. Today I downloaded Lady Heather and went to test it and noticed the should-be-running Thunderbolt was cold. I checked the voltages from the power supply and - nothing. Ignoring the life-or-death dire warnings on the power supply case I opened it. It had *that* smell. You know the smell of that special smoke that keeps all electronic components working because when it's released they stop. R1 is black and charred. Can anyone tell me what value R1 is and/or where I might get a new power supply (just in case it took other components with it)? Has anyone else had this problem? Regards, Jim From david.partridge at dsl.pipex.com Tue Feb 17 10:28:52 2009 From: david.partridge at dsl.pipex.com (David C. Partridge) Date: Tue, 17 Feb 2009 10:28:52 -0000 Subject: [time-nuts] Thunderbolt power supply open heart surgery In-Reply-To: References: Message-ID: I don't know the value of that resistor, but suspect that its cremation was collateral damage as something else failed short circuit. Dave -----Original Message----- From: time-nuts-bounces at febo.com [mailto:time-nuts-bounces at febo.com] On Behalf Of Jim Palfreyman Sent: 17 February 2009 10:13 To: Discussion of precise time and frequency measurement Subject: [time-nuts] Thunderbolt power supply open heart surgery Hi folks, Well I woke up the other morning at 2am to find all the power off in the house. I was about to ring the power company and thought I'd better check my meter box. Turns out one of the RCDs had tripped. I reset it and all was fine. What puzzled me was what tripped it. It took out my workshop, speaking clock, various GPSDOs and an HP rubidium I've borrowed. However all was working upon reset. Today I downloaded Lady Heather and went to test it and noticed the should-be-running Thunderbolt was cold. I checked the voltages from the power supply and - nothing. Ignoring the life-or-death dire warnings on the power supply case I opened it. It had *that* smell. You know the smell of that special smoke that keeps all electronic components working because when it's released they stop. R1 is black and charred. Can anyone tell me what value R1 is and/or where I might get a new power supply (just in case it took other components with it)? Has anyone else had this problem? Regards, Jim _______________________________________________ time-nuts mailing list -- time-nuts at febo.com To unsubscribe, go to https://www.febo.com/cgi-bin/mailman/listinfo/time-nuts and follow the instructions there. From namichie at gmail.com Tue Feb 17 10:38:31 2009 From: namichie at gmail.com (Neville Michie) Date: Tue, 17 Feb 2009 21:38:31 +1100 Subject: [time-nuts] Thunderbolt power supply open heart surgery In-Reply-To: References: Message-ID: Hi Jim, you may have a problem similar to mine. In Sydney, with a nominal supply voltage of 240 Volts, the supply at my place is above 250 volts. The supply authority says that is in the limits of their specification. I do not know why they run it so high, maybe they think they will sell more power. The problem is that many appliances have a 240 or 250 maximum, dual power supplies for 110 V switch to take 220 V. My solution is to run my lab through a step down transformer to about 210 volts. It puts less pressure on the capacitors in the power supplies. I have not had a power supply problem for about 4 years now. cheers, Neville Michie On 17/02/2009, at 9:13 PM, Jim Palfreyman wrote: > Hi folks, > > Well I woke up the other morning at 2am to find all the power off > in the > house. I was about to ring the power company and thought I'd better > check my > meter box. Turns out one of the RCDs had tripped. I reset it and > all was > fine. > > What puzzled me was what tripped it. It took out my workshop, speaking > clock, various GPSDOs and an HP rubidium I've borrowed. However all > was > working upon reset. > > Today I downloaded Lady Heather and went to test it and noticed the > should-be-running Thunderbolt was cold. I checked the voltages from > the > power supply and - nothing. Ignoring the life-or-death dire > warnings on the > power supply case I opened it. It had *that* smell. You know the > smell of > that special smoke that keeps all electronic components working > because when > it's released they stop. > > R1 is black and charred. > > Can anyone tell me what value R1 is and/or where I might get a new > power > supply (just in case it took other components with it)? > > Has anyone else had this problem? > > Regards, > > Jim > _______________________________________________ > time-nuts mailing list -- time-nuts at febo.com > To unsubscribe, go to https://www.febo.com/cgi-bin/mailman/listinfo/ > time-nuts > and follow the instructions there. From dave.g0dja at tiscali.co.uk Tue Feb 17 11:02:52 2009 From: dave.g0dja at tiscali.co.uk (Dave Ackrill) Date: Tue, 17 Feb 2009 11:02:52 +0000 Subject: [time-nuts] Thunderbolt power supply open heart surgery In-Reply-To: References: Message-ID: <499A995C.2090705@tiscali.co.uk> Jim Palfreyman wrote: > Has anyone else had this problem? The PSU that came with my Thunderbolt was DOA as well, like yours there was the characteristic smell of burnt electronics when I opened it up. I was lucky, I managed to get an almost exact replacement at a local electronics shop and, having looked at my dead PSU, I would suggest that repairing the dead one might be a bit of a problem. Even if you work out the resistor value you have to wonder what caused that amount of current to flow though it? My guess would be that something else went short circuit and replacing whatever it was might be more trouble than buying a new PSU... Dave From martinrh45 at googlemail.com Tue Feb 17 11:15:40 2009 From: martinrh45 at googlemail.com (Martin Richmond-Hardy) Date: Tue, 17 Feb 2009 11:15:40 +0000 Subject: [time-nuts] Thunderbolt power supply open heart surgery In-Reply-To: <499A995C.2090705@tiscali.co.uk> References: <499A995C.2090705@tiscali.co.uk> Message-ID: <8a3357d10902170315j40aa37cm822b2b807d747125@mail.gmail.com> Swollen capacitors or ones with a little volcano hole are a clue, but there may be other components that have been stressed but are not yet dead and now have a shortened life-expectancy. Had this problem with my iMac. Died 3 days after the smoke alert.Best of luck! Martin G8BHC 2009/2/17 Dave Ackrill > Jim Palfreyman wrote: > > > Has anyone else had this problem? > > The PSU that came with my Thunderbolt was DOA as well, like yours there > was the characteristic smell of burnt electronics when I opened it up. > > I was lucky, I managed to get an almost exact replacement at a local > electronics shop and, having looked at my dead PSU, I would suggest that > repairing the dead one might be a bit of a problem. Even if you work > out the resistor value you have to wonder what caused that amount of > current to flow though it? My guess would be that something else went > short circuit and replacing whatever it was might be more trouble than > buying a new PSU... > > Dave > > _______________________________________________ > time-nuts mailing list -- time-nuts at febo.com > To unsubscribe, go to > https://www.febo.com/cgi-bin/mailman/listinfo/time-nuts > and follow the instructions there. > -- Martin Richmond-Hardy From francesco.messineo at gmail.com Tue Feb 17 11:33:36 2009 From: francesco.messineo at gmail.com (francesco messineo) Date: Tue, 17 Feb 2009 12:33:36 +0100 Subject: [time-nuts] rockwell gps chipset datasheet? Message-ID: Hello all, I have given a GPS receiver board from an old anti-theft system, it's free for me to play with (and more are available if I ever figure out if they can be useful). The board has a rockwell chipset for the GPS part and two antenna connectors (selectable with an rf switch). Both antennas have independent power supply, I think it's +5V, I've not yet powered it up, I'm waiting to figure out more things about it. Rockwell chips are R6732-13 (the one where RF signal goes) and the bigger 11577-11. Now I haven't been able to dig any pinout at least for the 11577-1. Anyone? Thanks in advance Frank IZ8DWF From rbarrioss at msn.com Tue Feb 17 12:22:18 2009 From: rbarrioss at msn.com (Roberto Barrios) Date: Tue, 17 Feb 2009 13:22:18 +0100 Subject: [time-nuts] rockwell gps chipset datasheet? In-Reply-To: References: Message-ID: Hi Frank, The Rockwell 11577-11 is found in most TU-DXXX Rockwell-Jupiter GPS receivers. I've never seen anything like a RF switch on them, but you could have a custom or exotic one :). What they do have most times is a model number etched or printed on the PCB. Doesn't yours have one? You can find a few Rockwell GPS datasheets here, where you can compare yours with the pictures: http://www.gpskit.nl/downloads-nl.htm Regards, Roberto EB4EQA ---------------------------------------- Message: 1 Date: Tue, 17 Feb 2009 12:33:36 +0100 From: francesco messineo Subject: [time-nuts] rockwell gps chipset datasheet? To: time-nuts at febo.com Message-ID: Content-Type: text/plain; charset=ISO-8859-1 Hello all, I have given a GPS receiver board from an old anti-theft system, it's free for me to play with (and more are available if I ever figure out if they can be useful). The board has a rockwell chipset for the GPS part and two antenna connectors (selectable with an rf switch). Both antennas have independent power supply, I think it's +5V, I've not yet powered it up, I'm waiting to figure out more things about it. Rockwell chips are R6732-13 (the one where RF signal goes) and the bigger 11577-11. Now I haven't been able to dig any pinout at least for the 11577-1. Anyone? Thanks in advance Frank IZ8DWF From francesco.messineo at gmail.com Tue Feb 17 12:42:26 2009 From: francesco.messineo at gmail.com (francesco messineo) Date: Tue, 17 Feb 2009 13:42:26 +0100 Subject: [time-nuts] rockwell gps chipset datasheet? In-Reply-To: References: Message-ID: Hi Roberto, On 2/17/09, Roberto Barrios wrote: > Hi Frank, > > The Rockwell 11577-11 is found in most TU-DXXX Rockwell-Jupiter GPS > receivers. I've never seen anything like a RF switch on them, but you could > have a custom or exotic one :). What they do have most times is a model > number etched or printed on the PCB. Doesn't yours have one? > the PCB is custom made from an italian company, I could try to ask them if they still exists, but it's very difficult that they release any info about it anyway. The RF switch is there to software select one of two antennas, probably for redundancy, given the particular application for which it has been made. > You can find a few Rockwell GPS datasheets here, where you can compare yours > with the pictures: http://www.gpskit.nl/downloads-nl.htm I already seen this site, but there's no component level datasheet, on my board only the two rockwell chips have been used, and probably in a custom design. Best regards Frank IZ8DWF From jltran at worldnet.att.net Tue Feb 17 12:46:45 2009 From: jltran at worldnet.att.net (J. L. Trantham) Date: Tue, 17 Feb 2009 06:46:45 -0600 Subject: [time-nuts] Thunderbolt power supply open heart surgery In-Reply-To: Message-ID: <3EAFFD29DF60433B88FD09759AD40313@S0028384766> Jim, I had a TAPR Thunderbolt connected to the power supply that came with it and left it on for several days. My location is bothered by various power spikes and outages. After one power outage that tripped the circuit breaker to the work shop, I discovered the Thunderbolt injured (no communication, no 1 PPS, but 10 MHz out was present) but the power supply was OK, at least as measured by a voltmeter. I have since powered up a second unit using a nice linear power supply and no problem leaving it on for weeks at a time. I have yet to dig in to the Thunderbolt to chase down the problem. Joe -----Original Message----- From: time-nuts-bounces at febo.com [mailto:time-nuts-bounces at febo.com] On Behalf Of Jim Palfreyman Sent: Tuesday, February 17, 2009 4:13 AM To: Discussion of precise time and frequency measurement Subject: [time-nuts] Thunderbolt power supply open heart surgery Hi folks, Well I woke up the other morning at 2am to find all the power off in the house. I was about to ring the power company and thought I'd better check my meter box. Turns out one of the RCDs had tripped. I reset it and all was fine. What puzzled me was what tripped it. It took out my workshop, speaking clock, various GPSDOs and an HP rubidium I've borrowed. However all was working upon reset. Today I downloaded Lady Heather and went to test it and noticed the should-be-running Thunderbolt was cold. I checked the voltages from the power supply and - nothing. Ignoring the life-or-death dire warnings on the power supply case I opened it. It had *that* smell. You know the smell of that special smoke that keeps all electronic components working because when it's released they stop. R1 is black and charred. Can anyone tell me what value R1 is and/or where I might get a new power supply (just in case it took other components with it)? Has anyone else had this problem? Regards, Jim _______________________________________________ time-nuts mailing list -- time-nuts at febo.com To unsubscribe, go to https://www.febo.com/cgi-bin/mailman/listinfo/time-nuts and follow the instructions there. From james.p.lux at jpl.nasa.gov Tue Feb 17 14:14:28 2009 From: james.p.lux at jpl.nasa.gov (Lux, James P) Date: Tue, 17 Feb 2009 06:14:28 -0800 Subject: [time-nuts] Thunderbolt power supply open heart surgery In-Reply-To: Message-ID: On 2/17/09 2:38 AM, "Neville Michie" wrote: > Hi Jim, > you may have a problem similar to mine. > In Sydney, with a nominal supply voltage of 240 Volts, > the supply at my place is above 250 volts. The supply authority > says that is in the limits of their specification. Typically, limits are +/- 5% or +/- 10%. Either way, 250 is within 5% of 240. As another matter, at light loads in your house/neighborhood, the voltages will rise, since the distribution voltage is usually set up so that at nominal load, it's correct, and that allows for some IR drop in the lines. > I do not know why they run it so high, maybe they think they will > sell more power. > The problem is that many appliances have a 240 or 250 maximum, dual > power supplies for 110 V > switch to take 220 V. That's actually pretty unlikely. 110V might be what it says on the label, but in reality, nominal line voltage in the US is 120V, at least at the meter. There's a difference in the "service voltage" and the "utilization voltage" in most cases, for instance, 230V electric motors are operated on a 240V service. From dave.g0dja at tiscali.co.uk Tue Feb 17 15:12:21 2009 From: dave.g0dja at tiscali.co.uk (Dave Ackrill) Date: Tue, 17 Feb 2009 15:12:21 +0000 Subject: [time-nuts] Thunderbolt power supply open heart surgery In-Reply-To: References: Message-ID: <499AD3D5.6040708@tiscali.co.uk> Lux, James P wrote: >> I do not know why they run it so high, maybe they think they will >> sell more power. >> The problem is that many appliances have a 240 or 250 maximum, dual >> power supplies for 110 V >> switch to take 220 V. > > That's actually pretty unlikely. 110V might be what it says on the label, > but in reality, nominal line voltage in the US is 120V, at least at the > meter. There's a difference in the "service voltage" and the "utilization > voltage" in most cases, for instance, 230V electric motors are operated on a > 240V service. It can make a difference where you are on the network as well. Having been an Engineer with an electricity company here in the UK, and worked for a time in the planning office, the designs are based upon ensuring that the people at the far end of the feeders from a substation get their electricity within the stated minimum and the ones nearest the substation within the stated maximum. In the UK we now have a +/-10% limit from a nominated 230V (single phase) supply. So, nearest the substation you may well see up to 253V and the farthest away could see 207V. As has already been said, at times of high demand the voltage will tend to be lower than at low demand times. In order for the supply to be deemed to be out-of-limits the minimum or maximum must be reached for a significant period of time. So, if it dips below 207V for only a brief time every now-and-then it doesn't count as being out-of-limits either. Same applies to over limit as well. Dave From hmurray at megapathdsl.net Tue Feb 17 17:00:33 2009 From: hmurray at megapathdsl.net (Hal Murray) Date: Tue, 17 Feb 2009 09:00:33 -0800 Subject: [time-nuts] Thunderbolt power supply open heart surgery In-Reply-To: Message from "Lux, James P" of "Tue, 17 Feb 2009 06:14:28 PST." Message-ID: <20090217170034.C9649BCDD@ip-64-139-1-69.sjc.megapath.net> > As another matter, at light loads in your house/neighborhood, the > voltages will rise, since the distribution voltage is usually set up > so that at nominal load, it's correct, and that allows for some IR > drop in the lines. I got interested in this area a while ago. PG&E replaced the transformer on the pole right outside my house. It seems as though my light bulbs don't last as long as they used to, but I don't have any real data to back that up. One day, I got lucky and found their web page: http://www.pge.com/tariffs/tm2/pdf/ELEC_RULES_2.pdf That refers to NMEA/ANSI: American National Standard C84.1 They want $60, but PG&E provides enough info to read between some of the lines. Page 4 says: If the nominal voltage is 120, PG&E tries for a min of 114. They have a max of 120 for Class A or 126 for class B. I think that is at the entrance to your house. Page 6 says: You should expect a min of 110 and a max of 125. I think that means that C84 allows 4 volts drop in your house. The next page says 10% drop (108 V) on circuits not supplying lighting loads. There is lots of weasel wording. Page 15 has a note about harmful wave forms. I assume that covers switching power supplies. Page 20 says street lights have to meet 90% power factor. Several years ago, when I inquired about how to monitor line voltage, somebody suggested using an APC UPS. Many thanks. That's what I've been doing. The voltage in my house usually runs between 120 and 125. I occasionally see it go above. I'd have to dig some to find a sample. I also see various glitches where it drops below 110. Most of them are short enough so that my PCs that are not on the UPS ride through it. -- These are my opinions, not necessarily my employer's. I hate spam. From stanley_reynolds at yahoo.com Tue Feb 17 17:36:16 2009 From: stanley_reynolds at yahoo.com (Stanley Reynolds) Date: Tue, 17 Feb 2009 09:36:16 -0800 (PST) Subject: [time-nuts] Thunderbolt power supply open heart surgery References: <499AD3D5.6040708@tiscali.co.uk> Message-ID: <132400.76251.qm@web30306.mail.mud.yahoo.com> We had a thread a while back about Dranetz power monitors and UPS built in monitors. All nice to have, another older technology for power problems is ferroresonant UPS. More copper and weight as well as a little less efficient but I like the rugged and simple design. You will also see power conditioners which have the flywheel transformer and capacitor but no batteries or UPS less expensive to maintain but will clean up "dirty" power. ? Stanley From kc0vsj at yahoo.com Tue Feb 17 18:53:04 2009 From: kc0vsj at yahoo.com (Tom Clifton) Date: Tue, 17 Feb 2009 10:53:04 -0800 (PST) Subject: [time-nuts] Rockwell GPS Chipset Message-ID: <707035.5881.qm@web37002.mail.mud.yahoo.com> Any chance it is a Jupiter? http://www.gpskit.nl/gps-readme.html If so, they are fine receivers. THe header pins are 2mm spcing - same as used on laptop hard drive adaptors From SAIDJACK at aol.com Tue Feb 17 18:57:31 2009 From: SAIDJACK at aol.com (SAIDJACK at aol.com) Date: Tue, 17 Feb 2009 13:57:31 EST Subject: [time-nuts] Thunderbolt power supply open heart surgery Message-ID: Hi there, my Watts Up? power meter showed a reading of 130V+ one day coming from PG&E here in NorCal. I confirmed this with another meter. _https://www.wattsupmeters.com/secure/products.php_ (https://www.wattsupmeters.com/secure/products.php) I called PG&E, and they were surprised that I had this capability, and said one other customer in the entire town complained as well. Turns out the auto-transformer voltage regulator in the main power central hub was busted. They said it takes months to get a new unit from Germany, so they manually set the tap to -5V. We are now getting 120 - 125V from the grid, but PG&E has to manually adjust for load changes now (air-conditioning loading during the summer etc). The technician also said that if anything blows that PG&E would pay for repairs or replacement, I would just have to reference my service call. Remember that a 10% increase in voltage will reduce bulb lifetime by 50% typically... bye, Said In a message dated 2/17/2009 09:01:24 Pacific Standard Time, hmurray at megapathdsl.net writes: > As another matter, at light loads in your house/neighborhood, the > voltages will rise, since the distribution voltage is usually set up > so that at nominal load, it's correct, and that allows for some IR > drop in the lines. I got interested in this area a while ago. PG&E replaced the transformer on the pole right outside my house. It seems as though my light bulbs don't last as long as they used to, but I don't have any real data to back that up. From francesco.messineo at gmail.com Tue Feb 17 19:00:32 2009 From: francesco.messineo at gmail.com (francesco messineo) Date: Tue, 17 Feb 2009 20:00:32 +0100 Subject: [time-nuts] Rockwell GPS Chipset In-Reply-To: <707035.5881.qm@web37002.mail.mud.yahoo.com> References: <707035.5881.qm@web37002.mail.mud.yahoo.com> Message-ID: On 2/17/09, Tom Clifton wrote: > Any chance it is a Jupiter? > > http://www.gpskit.nl/gps-readme.html > > If so, they are fine receivers. THe header pins are 2mm spcing - same as used on laptop hard drive adaptors > It uses the same two chips (at least in one photo of a jupiter I have seen the 11577-11, but the board I have is custom made with these chips. To identify the right signals I'd have to find the pinout of the rockwell chips. Thanks Frank IZ8DWF From d.seiter at comcast.net Tue Feb 17 21:22:42 2009 From: d.seiter at comcast.net (d.seiter at comcast.net) Date: Tue, 17 Feb 2009 21:22:42 +0000 (UTC) Subject: [time-nuts] Lady Heather display is GREAT!!! In-Reply-To: Message-ID: <1946768303.985081234905762939.JavaMail.root@sz0108a.emeryville.ca.mail.comcast.net> Hi Mark- I hope your right!?? The thing is that this Tbolt has always behaved itself w/ the indoor antenna and has never taken long to stablize.? When it stops raining, I'll stick it outside and see what happens.? I really need to put in a feed through to the roof because my Z3801A is struggling with an indoor antenna too; but the roof needs to get replaced first... -Dave ----- Original Message ----- From: "Mark Sims" To: time-nuts at febo.com Sent: Monday, February 16, 2009 8:04:37 PM GMT -08:00 US/Canada Pacific Subject: [time-nuts] Lady Heather display is GREAT!!! That sounds like your indoor antenna is the problem. ? I have had very poor luck with indoor antennas. ?Even my outdoor antenna is rather bad... ?too many very big trees. ? I get constant satellite constellation changes (sometimes several times a minute). ? My first self survey took three days to complete. ?Where my antenna is now, ?I can complete a survey in a couple of hours (but at only a couple of times each day). Also many GPS receivers tend to ?have a hard time starting up when they have not been used for a while. ?It takes a long time for the oscillator to stabilize enough to work properly. ?I had a tbolt that had not been used for a couple of years. ?It took about a week before it settled down enough to be really usable. ? ---------------------------------------- However, about 2 hours into my test it started loosing sats (indoor antenna, 3 to 6 sats at any given time; _________________________________________________________________ See how Windows Mobile brings your life together?at home, work, or on the go. http://clk.atdmt.com/MRT/go/msnnkwxp1020093182mrt/direct/01/ _______________________________________________ time-nuts mailing list -- time-nuts at febo.com To unsubscribe, go to https://www.febo.com/cgi-bin/mailman/listinfo/time-nuts and follow the instructions there. From alf.pousar at surffi.net Tue Feb 17 21:50:02 2009 From: alf.pousar at surffi.net (Alf Pousar) Date: Tue, 17 Feb 2009 23:50:02 +0200 Subject: [time-nuts] Best GPS timing Message-ID: Thanks Bruce for your answer, it was the only one! 73, Alf From SAIDJACK at aol.com Tue Feb 17 21:59:13 2009 From: SAIDJACK at aol.com (SAIDJACK at aol.com) Date: Tue, 17 Feb 2009 16:59:13 EST Subject: [time-nuts] Best GPS timing Message-ID: Hi Alf, if you can get an M12+ from iLotus, that's not too shabby either. M12M is the king of the crop though as Bruce said. Now you got two answers :) bye, Said In a message dated 2/17/2009 13:50:54 Pacific Standard Time, alf.pousar at surffi.net writes: Thanks Bruce for your answer, it was the only one! 73, Alf From holrum at hotmail.com Tue Feb 17 22:58:06 2009 From: holrum at hotmail.com (Mark Sims) Date: Tue, 17 Feb 2009 22:58:06 +0000 Subject: [time-nuts] Best GPS timing In-Reply-To: References: Message-ID: Well, actually there are some geodetic quality carrier phase based units out there... $25,000+ a pop... ----------------- M12M is the king of the crop though as Bruce said. _________________________________________________________________ See how Windows Mobile brings your life together?at home, work, or on the go. http://clk.atdmt.com/MRT/go/msnnkwxp1020093182mrt/direct/01/ From SAIDJACK at aol.com Tue Feb 17 23:21:57 2009 From: SAIDJACK at aol.com (SAIDJACK at aol.com) Date: Tue, 17 Feb 2009 18:21:57 EST Subject: [time-nuts] Best GPS timing Message-ID: Hi there, I did see that in the recent GPS World receiver survey, but Javad for example only claims 3ns rms accuracy... The m12+ claim 2ns rms in the old Motorola literature, but that only with a properly surveyed antenna, sawtooth compensation, and good antenna-delay calibration of course. Does anyone know if there is anything below say $200 that is better than the M12 series? bye, Said In a message dated 2/17/2009 14:58:45 Pacific Standard Time, holrum at hotmail.com writes: Well, actually there are some geodetic quality carrier phase based units out there... $25,000+ a pop... From bg at lysator.liu.se Tue Feb 17 23:28:11 2009 From: bg at lysator.liu.se (bg at lysator.liu.se) Date: Wed, 18 Feb 2009 00:28:11 +0100 (CET) Subject: [time-nuts] Best GPS timing In-Reply-To: References: Message-ID: <53831.87.227.52.225.1234913291.squirrel@webmail.lysator.liu.se> > > Well, actually there are some geodetic quality carrier phase based units > out there... $25,000+ a pop... There are dual/triple frequency geodetic receivers with external frequency input available in the $3k-4k range. The CMC/Novatel SuperstarII used to be available for ca $300 with carrier phase. Removal of the 10MHz onboard TCXO, and feeding with your own frequency standard of choice is widely documented. -- Bj?rn From bruce.griffiths at xtra.co.nz Tue Feb 17 23:47:12 2009 From: bruce.griffiths at xtra.co.nz (Bruce Griffiths) Date: Wed, 18 Feb 2009 12:47:12 +1300 Subject: [time-nuts] Best GPS timing In-Reply-To: <53831.87.227.52.225.1234913291.squirrel@webmail.lysator.liu.se> References: <53831.87.227.52.225.1234913291.squirrel@webmail.lysator.liu.se> Message-ID: <499B4C80.1060606@xtra.co.nz> bg at lysator.liu.se wrote: >> Well, actually there are some geodetic quality carrier phase based units >> out there... $25,000+ a pop... >> > > There are dual/triple frequency geodetic receivers with external frequency > input available in the $3k-4k range. > > The CMC/Novatel SuperstarII used to be available for ca $300 with carrier > phase. Removal of the 10MHz onboard TCXO, and feeding with your own > frequency standard of choice is widely documented. > > -- > > Bj?rn > > Bj?rn The Superstar II is now obsolete. Last orders were taken last year. Bruce From bg at lysator.liu.se Tue Feb 17 23:50:37 2009 From: bg at lysator.liu.se (bg at lysator.liu.se) Date: Wed, 18 Feb 2009 00:50:37 +0100 (CET) Subject: [time-nuts] Best GPS timing In-Reply-To: References: Message-ID: <54012.87.227.52.225.1234914637.squirrel@webmail.lysator.liu.se> > Hi there, > > I did see that in the recent GPS World receiver survey, but Javad for > example only claims 3ns rms accuracy... > Where do Javad claim 3ns accuracy? +++++ [ZA], [ZB] PPS Offset struct PPSOffset {5} { f4 offs; // PPS offset in nanoseconds u1 cs; // Checksum }; Due to a hardware limitation, PPS signals are discrete with a resolution of 25 ns. JAVAD GNSS receiver allows you to compensate for this discreteness error by means of utilizing this message. It contains the offset between the scheduled PPS time and the actual pulse edge's arrival time. ++++++++ They do claim 10cm pseudo-range accuracy and 1 mm carrier phase accuracy. Why would not the "PPS Offset" message above be an accurate "sawtooth" measurement down to the PVT solution accuracy? Where the accuracy might be several meters in stand alone mode going down to cm level with post processing. -- Bj?rn From bg at lysator.liu.se Tue Feb 17 23:53:31 2009 From: bg at lysator.liu.se (bg at lysator.liu.se) Date: Wed, 18 Feb 2009 00:53:31 +0100 (CET) Subject: [time-nuts] Best GPS timing In-Reply-To: <499B4C80.1060606@xtra.co.nz> References: <53831.87.227.52.225.1234913291.squirrel@webmail.lysator.liu.se> <499B4C80.1060606@xtra.co.nz> Message-ID: <54017.87.227.52.225.1234914811.squirrel@webmail.lysator.liu.se> > Bj?rn > > The Superstar II is now obsolete. > Last orders were taken last year. > > Bruce > Bruce, Yes, I know. >> The CMC/Novatel SuperstarII used to ****** Sorry for not beeing clear enough. -- Bj?rn From bruce.griffiths at xtra.co.nz Tue Feb 17 23:58:56 2009 From: bruce.griffiths at xtra.co.nz (Bruce Griffiths) Date: Wed, 18 Feb 2009 12:58:56 +1300 Subject: [time-nuts] Best GPS timing In-Reply-To: <54017.87.227.52.225.1234914811.squirrel@webmail.lysator.liu.se> References: <53831.87.227.52.225.1234913291.squirrel@webmail.lysator.liu.se> <499B4C80.1060606@xtra.co.nz> <54017.87.227.52.225.1234914811.squirrel@webmail.lysator.liu.se> Message-ID: <499B4F40.5000605@xtra.co.nz> bg at lysator.liu.se wrote: >> Bj?rn >> >> The Superstar II is now obsolete. >> Last orders were taken last year. >> >> Bruce >> >> > > Bruce, > > Yes, I know. > > >>> The CMC/Novatel SuperstarII used to >>> > ****** > > Sorry for not beeing clear enough. > > -- > > Bj?rn > > > Is it possible to reprogram Superstar II's to produce carrier phase outputs? This may be of interest if some non carrier phase versions were available at the right price. Or would it be much easier just to build a custom GPS receiver based on a standard RF chipset with a convenient LO frequency? Bruce From hmurray at megapathdsl.net Wed Feb 18 00:18:26 2009 From: hmurray at megapathdsl.net (Hal Murray) Date: Tue, 17 Feb 2009 16:18:26 -0800 Subject: [time-nuts] Antennas, roofs In-Reply-To: Message from d.seiter@comcast.net of "Tue, 17 Feb 2009 21:22:42 GMT." <1946768303.985081234905762939.JavaMail.root@sz0108a.emeryville.ca.mail.comcast.net> Message-ID: <20090218001827.29C49BCDD@ip-64-139-1-69.sjc.megapath.net> > I really need to put in a feed through to the roof because my Z3801A > is struggling with an indoor antenna too; but the roof needs to get > replaced first... Speaking of antennas and roofs.... Currently, my antennas are inside. That's good enough most of the time. It's also good for providing nasty test cases to software. I also need a new roof. I'm scheming to poke a hole in the roof so I can get some antennas in a better position. If I have more than one antenna, does it matter how near eachother they are located? I'm picturing a plastic pipe that sticks up a few feet and a bracket at the base that has the right magic angle to match the pitch of my roof. The pipe would screw or glue into the bracket. The bracket would get screwed to the roof over a hole. The cables would go through the hole and up inside the pipe. I haven't worked out the details for the top of the pipe yet. My (handwave) straw man is a U turn to keep the rain out, and mount the antennas on the main pipe. Maybe a T to get them out to the side. Do brackets like that exist? If so, what term or brand do I google for? I have a typical not-very-steep sloped roof. Is there a standard angle? ... Plan B would be to stick the pipe through the roof and attach it inside to the side of a rafter. I assume the roofers can treat it like a plumbing vent pipe. A slightly crazy idea... Has anybody poked antennas up inside a skylight? I'm thinking of the setup which has a hole in the ceiling of a room, a box from that hole through the attic space up to and through the roof, and a plastic dome on top. A shelf or bracket on the inside of the box would get the antennas almost on the outside. I've seen ads for metal roofs/shingles, the claimed advantage being long life which is attractive to me. I assume they would be a disaster for antennas inside. What about outside, slightly above the roof? I'd expect bad things, but maybe there is some way to turn it into an advantage. -- These are my opinions, not necessarily my employer's. I hate spam. From bg at lysator.liu.se Wed Feb 18 00:35:15 2009 From: bg at lysator.liu.se (bg at lysator.liu.se) Date: Wed, 18 Feb 2009 01:35:15 +0100 (CET) Subject: [time-nuts] Best GPS timing In-Reply-To: <499B4F40.5000605@xtra.co.nz> References: <53831.87.227.52.225.1234913291.squirrel@webmail.lysator.liu.se> <499B4C80.1060606@xtra.co.nz> <54017.87.227.52.225.1234914811.squirrel@webmail.lysator.liu.se> <499B4F40.5000605@xtra.co.nz> Message-ID: <54328.87.227.52.225.1234917315.squirrel@webmail.lysator.liu.se> > Is it possible to reprogram Superstar II's to produce carrier phase > outputs? > This may be of interest if some non carrier phase versions were > available at the right price. Yes, there is a DOS-program. But you need some codes from CMC/Novatel to unlock extra features you want. > Or would it be much easier just to build a custom GPS receiver based on > a standard RF chipset with a convenient LO frequency? If the unlock codes are still for purchase, that would obviously be cheaper by most peoples standards, than building from scratch. There used to be other receivers based on the now obsolete Zarlink chipset. Signav and GPS Creations comes to mind. > Bruce -- Bj?rn From wb6bnq at cox.net Wed Feb 18 00:49:26 2009 From: wb6bnq at cox.net (WB6BNQ) Date: Tue, 17 Feb 2009 16:49:26 -0800 Subject: [time-nuts] Antennas, roofs References: <20090218001827.29C49BCDD@ip-64-139-1-69.sjc.megapath.net> Message-ID: <499B5B16.1E512924@cox.net> Hi Hal, A metal roof is generally used for those areas that have heavy snow. Such a roof would be disastrous for the GPS if the antenna is low on top of the roof because of reflections. A metal roof would be harder to seal, I would think, for holes drilled through it. Also a metal roof can be quite slippery and dangerous for someone climbing around on it. The composition roof, at least, provides a less slippery surface and may provide less reflections. Although I would suspect that one would want to mount the antenna at least 10 feet above the roof surface, if not higher based upon surrounding objects like trees and so forth. If there is a header section above the top of the wall line (i.e., in the roof area), you may want to consider passing the pipe horizontally through the header instead of making yet another hole for leakage in your roof. Radio Shack use to carry brackets for side mounting a pole along side the wall. Two such brackets mounted ten feet apart would provide plenty of stability for small antennas, even including small VHF/UHF bean antennas. Radio Shack also use to have metal antenna poles that would stack end to end and give the ability to have any given length (within reason). Just some thoughts. Bill....WB6BNQ Hal Murray wrote: > > I really need to put in a feed through to the roof because my Z3801A > > is struggling with an indoor antenna too; but the roof needs to get > > replaced first... > > Speaking of antennas and roofs.... > > Currently, my antennas are inside. That's good enough most of the time. > It's also good for providing nasty test cases to software. > > I also need a new roof. > > I'm scheming to poke a hole in the roof so I can get some antennas in a > better position. > > If I have more than one antenna, does it matter how near eachother they are > located? > > I'm picturing a plastic pipe that sticks up a few feet and a bracket at the > base that has the right magic angle to match the pitch of my roof. The pipe > would screw or glue into the bracket. The bracket would get screwed to the > roof over a hole. The cables would go through the hole and up inside the > pipe. > > I haven't worked out the details for the top of the pipe yet. My (handwave) > straw man is a U turn to keep the rain out, and mount the antennas on the > main pipe. Maybe a T to get them out to the side. > > Do brackets like that exist? If so, what term or brand do I google for? I > have a typical not-very-steep sloped roof. Is there a standard angle? ... > > Plan B would be to stick the pipe through the roof and attach it inside to > the side of a rafter. I assume the roofers can treat it like a plumbing vent > pipe. > > A slightly crazy idea... Has anybody poked antennas up inside a skylight? > I'm thinking of the setup which has a hole in the ceiling of a room, a box > from that hole through the attic space up to and through the roof, and a > plastic dome on top. A shelf or bracket on the inside of the box would get > the antennas almost on the outside. > > I've seen ads for metal roofs/shingles, the claimed advantage being long life > which is attractive to me. I assume they would be a disaster for antennas > inside. What about outside, slightly above the roof? I'd expect bad things, > but maybe there is some way to turn it into an advantage. > > -- > These are my opinions, not necessarily my employer's. I hate spam. > > _______________________________________________ > time-nuts mailing list -- time-nuts at febo.com > To unsubscribe, go to https://www.febo.com/cgi-bin/mailman/listinfo/time-nuts > and follow the instructions there. From bruce.griffiths at xtra.co.nz Wed Feb 18 00:50:26 2009 From: bruce.griffiths at xtra.co.nz (Bruce Griffiths) Date: Wed, 18 Feb 2009 13:50:26 +1300 Subject: [time-nuts] Antennas, roofs In-Reply-To: <20090218001827.29C49BCDD@ip-64-139-1-69.sjc.megapath.net> References: <20090218001827.29C49BCDD@ip-64-139-1-69.sjc.megapath.net> Message-ID: <499B5B52.4030202@xtra.co.nz> Hal Murray wrote: >> I really need to put in a feed through to the roof because my Z3801A >> is struggling with an indoor antenna too; but the roof needs to get >> replaced first... >> > > Speaking of antennas and roofs.... > > Currently, my antennas are inside. That's good enough most of the time. > It's also good for providing nasty test cases to software. > > I also need a new roof. > > > I'm scheming to poke a hole in the roof so I can get some antennas in a > better position. > > If I have more than one antenna, does it matter how near eachother they are > located? > > > I'm picturing a plastic pipe that sticks up a few feet and a bracket at the > base that has the right magic angle to match the pitch of my roof. The pipe > would screw or glue into the bracket. The bracket would get screwed to the > roof over a hole. The cables would go through the hole and up inside the > pipe. > > I haven't worked out the details for the top of the pipe yet. My (handwave) > straw man is a U turn to keep the rain out, and mount the antennas on the > main pipe. Maybe a T to get them out to the side. > > > Do brackets like that exist? If so, what term or brand do I google for? I > have a typical not-very-steep sloped roof. Is there a standard angle? ... > > Plan B would be to stick the pipe through the roof and attach it inside to > the side of a rafter. I assume the roofers can treat it like a plumbing vent > pipe. > > > A slightly crazy idea... Has anybody poked antennas up inside a skylight? > I'm thinking of the setup which has a hole in the ceiling of a room, a box > from that hole through the attic space up to and through the roof, and a > plastic dome on top. A shelf or bracket on the inside of the box would get > the antennas almost on the outside. > > > I've seen ads for metal roofs/shingles, the claimed advantage being long life > which is attractive to me. I assume they would be a disaster for antennas > inside. What about outside, slightly above the roof? I'd expect bad things, > but maybe there is some way to turn it into an advantage. > > > > Instead of making a custom mating flange for the particular roof angle, you could mount the pipe perpendicular to the roof and mount the antenna on a bracket on a bracket attached to the pipe. Just drill the bracket so that the antenna is vertical. See attached images for possible details. Bruce -------------- next part -------------- A non-text attachment was scrubbed... Name: TQ9140.png Type: image/png Size: 26698 bytes Desc: not available Url : http://www.febo.com/pipermail/time-nuts/attachments/20090218/86c71911/attachment-0002.png -------------- next part -------------- A non-text attachment was scrubbed... Name: TQ9141.png Type: image/png Size: 27784 bytes Desc: not available Url : http://www.febo.com/pipermail/time-nuts/attachments/20090218/86c71911/attachment-0003.png From wd6cmu at earthlink.net Wed Feb 18 01:17:29 2009 From: wd6cmu at earthlink.net (Eric Williams) Date: Tue, 17 Feb 2009 17:17:29 -0800 Subject: [time-nuts] Antennas, roofs In-Reply-To: <20090218001827.29C49BCDD@ip-64-139-1-69.sjc.megapath.net> References: <1946768303.985081234905762939.JavaMail.root@sz0108a.emeryville.ca.mail.comcast.net> <20090218001827.29C49BCDD@ip-64-139-1-69.sjc.megapath.net> Message-ID: Yup, running my T-bolt on an indoor patch antenna hoisted to the apex of a skylight. Works fine. My GPS NTP server uses a bullet antenna mounted on a short (~1ft) pole bolted to a block of wood on the side of the house, feedthrough is PVC pipe going through the wall with a 45-degree coupler on the outside pointing down to keep out rain. On Tue, Feb 17, 2009 at 4:18 PM, Hal Murray wrote: > > > A slightly crazy idea... Has anybody poked antennas up inside a skylight? > I'm thinking of the setup which has a hole in the ceiling of a room, a box > from that hole through the attic space up to and through the roof, and a > plastic dome on top. A shelf or bracket on the inside of the box would get > the antennas almost on the outside. From Murray.Greenman at rakon.com Wed Feb 18 01:24:06 2009 From: Murray.Greenman at rakon.com (Murray Greenman) Date: Wed, 18 Feb 2009 14:24:06 +1300 Subject: [time-nuts] Indoor GPS for GPSDO In-Reply-To: References: Message-ID: You might as well forget using an indoor antenna for GPSDO use. For best performance, you need an antenna which sees the top of the sky (say down to 30? elevation) really clearly, and ideally you want an antenna designed for timing applications, which favours high angle satellites. This is quite different to navigation use, where low angle satellites give better angular calculations. I'd say if you can't achieve 24 hours without a holdover event, you need to improve the antenna. How long it takes to perform the survey is another matter, and seems to depend on the GPSDO. I've experience with some of the hot new receivers (such as the ublox LEA-5) with good indoor navigation performance, but even with this type, I would not expect the timing to be perfect without a good antenna in a good location. You should also be able to achieve 7 - 8 satellites in the fix most of the time, with at least 40dB average CNR. My Samsung GCRU-D achieves these figures with no difficulty using a Novatel GPS-600 antenna and 30m of cable. Performance of my Trimble NTGS50AA is also similar, on a commercial hockey-puck designed for timing reception. The HP Z3801 has only a 6-ch receiver, and on an old Furuno marine quadrifilar antenna (designed for navigation) performs less well, and I have short holdover events every few days. All antennas have a clear view of the top of the sky. 73, Murray ZL1BPU From wa2lbi at frontiernet.net Wed Feb 18 01:33:40 2009 From: wa2lbi at frontiernet.net (Ken Winterling) Date: Tue, 17 Feb 2009 20:33:40 -0500 Subject: [time-nuts] Antennas, roofs In-Reply-To: <20090218001827.29C49BCDD@ip-64-139-1-69.sjc.megapath.net> References: <1946768303.985081234905762939.JavaMail.root@sz0108a.emeryville.ca.mail.comcast.net> <20090218001827.29C49BCDD@ip-64-139-1-69.sjc.megapath.net> Message-ID: <608ed23c0902171733s3e03920djf4c2013abe2d989b@mail.gmail.com> Hal, If your new roof will have a ridge vent you can poke the antenna cables through it to the outside leaving enough slack to create a drip loop before continuing to the antenna(s). Since the ridge vent follows the slope of the roof the cable will exit on the down side so water will run off. Ken, WA2LBI On Tue, Feb 17, 2009 at 19:18, Hal Murray wrote: > > > I really need to put in a feed through to the roof because my Z3801A > > is struggling with an indoor antenna too; but the roof needs to get > > replaced first... > > Speaking of antennas and roofs.... > > Currently, my antennas are inside. That's good enough most of the time. > It's also good for providing nasty test cases to software. > > I also need a new roof. > > > I'm scheming to poke a hole in the roof so I can get some antennas in a > better position. > > If I have more than one antenna, does it matter how near eachother they are > located? > > > I'm picturing a plastic pipe that sticks up a few feet and a bracket at the > base that has the right magic angle to match the pitch of my roof. The > pipe > would screw or glue into the bracket. The bracket would get screwed to the > roof over a hole. The cables would go through the hole and up inside the > pipe. > > I haven't worked out the details for the top of the pipe yet. My > (handwave) > straw man is a U turn to keep the rain out, and mount the antennas on the > main pipe. Maybe a T to get them out to the side. > > > Do brackets like that exist? If so, what term or brand do I google for? I > have a typical not-very-steep sloped roof. Is there a standard angle? ... > > Plan B would be to stick the pipe through the roof and attach it inside to > the side of a rafter. I assume the roofers can treat it like a plumbing > vent > pipe. > > > A slightly crazy idea... Has anybody poked antennas up inside a skylight? > I'm thinking of the setup which has a hole in the ceiling of a room, a box > from that hole through the attic space up to and through the roof, and a > plastic dome on top. A shelf or bracket on the inside of the box would get > the antennas almost on the outside. > > > I've seen ads for metal roofs/shingles, the claimed advantage being long > life > which is attractive to me. I assume they would be a disaster for antennas > inside. What about outside, slightly above the roof? I'd expect bad > things, > but maybe there is some way to turn it into an advantage. > > > > -- > These are my opinions, not necessarily my employer's. I hate spam. > > > > > _______________________________________________ > time-nuts mailing list -- time-nuts at febo.com > To unsubscribe, go to > https://www.febo.com/cgi-bin/mailman/listinfo/time-nuts > and follow the instructions there. > From EWKehren at aol.com Wed Feb 18 01:34:32 2009 From: EWKehren at aol.com (EWKehren at aol.com) Date: Tue, 17 Feb 2009 20:34:32 EST Subject: [time-nuts] Antennas, roofs Message-ID: Depending where you are most homes have sewer vents. I have used this repeatedly in the US and in Germany. This is a PVC pipe going through the roof and normally well sealed by the roofer, but open on the top. In the past Radio Shack had pipe to pipe clamps the pair for less than $10. You choose the pipe closest to your GPS location and drill a hole in the pipe inside the house, above the highest toilet and bring the cable into the house. Seal it with RTV to prevent odor. Water or sewer will never rise to that level, it would overflow the toilet if stopped up. This way you have a solid set up, the pipe you connect to the sewer vent can be long enough to clear the roof! Worked for me. Bert Kehren Miami WB5MZJ **************You can't always choose whom you love, but you can choose how to find them. Start with AOL Personals. (http://personals.aol.com/?ncid=emlcntuslove00000002) From stanw1le at verizon.net Wed Feb 18 01:55:30 2009 From: stanw1le at verizon.net (Stan W1LE) Date: Tue, 17 Feb 2009 20:55:30 -0500 Subject: [time-nuts] Antennas, roofs In-Reply-To: <20090218001827.29C49BCDD@ip-64-139-1-69.sjc.megapath.net> References: <20090218001827.29C49BCDD@ip-64-139-1-69.sjc.megapath.net> Message-ID: <499B6A92.1020105@verizon.net> Hello Hal, You can find roof penetrations in the plumbing dept of Home Depot or Lowes. Some will handle from a 4" (ID) PVC pipe used as a drain vent, down to a 1" trade size pipe. I prefer the rubber gland on sheet metal, vs. the rubber gland on plastic. Using the white DWV (drain waste vent) PVC pipe will work but is a bright white eyesore. Conside the toned down, grey electrical conduit, also schedule 40. Galvi pipe works too. These glands are common to a residential roofer. Talk to multiple roofing contractors and get their ideas. My plan is to have the GPS antenna on a metal mast (pipe) slightly higher than the ridge line, penetrating the roof about 1-2' below the ridge line. I will secure the metal mast to the rafter inside and run the coaxial cable thru the mast to the antenna which screws onto the top of the mast. I will use standard roofing shingles, non metal. A metal roof will allow reflections off of the roof into the antenna. Keep the antenna above the highest point of the roof. But, a standing seam metal roof will allow a cost effective way to secure PV solar panels. Model GPS system performance at the exact location with a temporary antenna before any permanent holes are made. Right now I am using a cheap active patch antenna , ~ 5$ on ebay, outside on a temporary tripod at gutter height. Lady Heather shows me with 8 satellites, usually under reception. Try the various inside locations for a GPS antenna. If you can live with the results, avoid the leak potential of a hole in the roof. Ultimately, find the best location with a temporary antenna. make the roof penetration, then add a HP multicoupler inside and feed other GPS receivers. All from a single antenna Your RF electromagnetic environment may be unique. Test at temporary locations. Hopefully, from the final location you can see clean sky from horizon to horizon. I like the radomed (bubble) on the roof, inside installation. Only gotcha is the non standard product and you would have to engineer it. Talk to you roofing contractor for ideas, he does this stuph all day long. What does your Mrs. have to say about the installation ?? Metal roofs are noisey with precipitation, like hale, rain, or snow, as well as creaky with thermal expansion/contraction. I would expect a metal roof, properly installed, to last a lifetime, Can not say that about asphault/fiberglass roofing shingles. Stan, W1LE FN41sr Cape Cod Hal Murray wrote: >> I really need to put in a feed through to the roof because my Z3801A >> is struggling with an indoor antenna too; but the roof needs to get >> replaced first... >> > > Speaking of antennas and roofs.... > > Currently, my antennas are inside. That's good enough most of the time. > It's also good for providing nasty test cases to software. > > I also need a new roof. > > > I'm scheming to poke a hole in the roof so I can get some antennas in a > better position. > > If I have more than one antenna, does it matter how near eachother they are > located? > > > I'm picturing a plastic pipe that sticks up a few feet and a bracket at the > base that has the right magic angle to match the pitch of my roof. The pipe > would screw or glue into the bracket. The bracket would get screwed to the > roof over a hole. The cables would go through the hole and up inside the > pipe. > > I haven't worked out the details for the top of the pipe yet. My (handwave) > straw man is a U turn to keep the rain out, and mount the antennas on the > main pipe. Maybe a T to get them out to the side. > > > Do brackets like that exist? If so, what term or brand do I google for? I > have a typical not-very-steep sloped roof. Is there a standard angle? ... > > Plan B would be to stick the pipe through the roof and attach it inside to > the side of a rafter. I assume the roofers can treat it like a plumbing vent > pipe. > > > A slightly crazy idea... Has anybody poked antennas up inside a skylight? > I'm thinking of the setup which has a hole in the ceiling of a room, a box > from that hole through the attic space up to and through the roof, and a > plastic dome on top. A shelf or bracket on the inside of the box would get > the antennas almost on the outside. > > > I've seen ads for metal roofs/shingles, the claimed advantage being long life > which is attractive to me. I assume they would be a disaster for antennas > inside. What about outside, slightly above the roof? I'd expect bad things, > but maybe there is some way to turn it into an advantage. > > > > From Murray.Greenman at rakon.com Wed Feb 18 02:04:06 2009 From: Murray.Greenman at rakon.com (Murray Greenman) Date: Wed, 18 Feb 2009 15:04:06 +1300 Subject: [time-nuts] Roof-mount GPS antenna In-Reply-To: References: Message-ID: Down here in NZ, a common form of TV antenna mount is the 'hockey stick', which is a J-shaped pole with a plate for barge-board or gable mount (on a vertical surface) and a vertical pipe about 1m long and 20mm diameter. A typical pipe-mount GPS antenna is easy to mount on this, either directly threaded on, or clamped to the side with a U-bolt. The threaded on connection is really tidy if you have the right antenna, as the cable can go down through the pipe. Another way to do pipe-mount is to strap a short length of suitable pipe to a roof vent or chimney, but you'll have to be creative with where to bring the cable in - perhaps over the edge of the roof, provide a drip curve, then under the roof through the soffit or gable end and across the attic and through the ceiling. Sewer pipe vents are good candidates. Chimneys less so, unless they are unused. I once used a hockey-puck magnetic mount type antenna on a small metal plate, and chocked the plate horizontal on the sloping (metal) roof using a block of wood. It worked fine, but had the bad habit of blowing away in a storm, even when I weighted the plate down with bricks. Look for TV hardware in your nearest DSE/Jaycar/Tandy/hardware store. 73, Murray ZL1BPU From SAIDJACK at aol.com Wed Feb 18 02:03:04 2009 From: SAIDJACK at aol.com (SAIDJACK at aol.com) Date: Tue, 17 Feb 2009 21:03:04 EST Subject: [time-nuts] Best GPS timing Message-ID: Hi, in the previous edition of the magazine GPS WORLD there is a GPS receiver survey, in the table they state 3ns rms timing, but there is no clear definition what the measurement details are. bye, Said In a message dated 2/17/2009 15:51:15 Pacific Standard Time, bg at lysator.liu.se writes: > Hi there, > > I did see that in the recent GPS World receiver survey, but Javad for > example only claims 3ns rms accuracy... > Where do Javad claim 3ns accuracy? From SAIDJACK at aol.com Wed Feb 18 05:20:26 2009 From: SAIDJACK at aol.com (SAIDJACK at aol.com) Date: Wed, 18 Feb 2009 00:20:26 EST Subject: [time-nuts] Best GPS timing Message-ID: FYI: here is an older timing brochure from Motorola stating 2ns rms performance for the M12+: bye, Said In a message dated 2/17/2009 18:04:41 Pacific Standard Time, SAIDJACK at aol.com writes: Hi, in the previous edition of the magazine GPS WORLD there is a GPS receiver survey, in the table they state 3ns rms timing, but there is no clear definition what the measurement details are. bye, Said In a message dated 2/17/2009 15:51:15 Pacific Standard Time, bg at lysator.liu.se writes: > Hi there, > > I did see that in the recent GPS World receiver survey, but Javad for > example only claims 3ns rms accuracy... > Where do Javad claim 3ns accuracy? _______________________________________________ time-nuts mailing list -- time-nuts at febo.com To unsubscribe, go to https://www.febo.com/cgi-bin/mailman/listinfo/time-nuts and follow the instructions there. -------------- next part -------------- A non-text attachment was scrubbed... Name: M12Timingbrochure.jpg Type: image/jpeg Size: 82479 bytes Desc: not available Url : http://www.febo.com/pipermail/time-nuts/attachments/20090218/51e036dd/attachment-0001.jpg From d.seiter at comcast.net Wed Feb 18 08:45:39 2009 From: d.seiter at comcast.net (d.seiter at comcast.net) Date: Wed, 18 Feb 2009 08:45:39 +0000 (UTC) Subject: [time-nuts] Antennas, roofs In-Reply-To: <608ed23c0902171733s3e03920djf4c2013abe2d989b@mail.gmail.com> Message-ID: <619502179.1235901234946739463.JavaMail.root@sz0108a.emeryville.ca.mail.comcast.net> In my case, I have a large number of low angle peaks and valleys (hence the leakage issue); my sky view is really good, and I'm planning on putting a set of parallel horizontal bars about 18" apart between my 4" sewer riser and another point about 15' away. All cables will be going through a 5" roof T vent which happens to be attached to nothing in particular (I had to insert a screen to keep the roof rats out- they were not happy!). The big issue is that I can't get inside the crawl space of the area in question, so I need to wait for drier weather to rip up the existing roof and put everything in place. Sigh... ----- Original Message ----- From: "Ken Winterling" To: "Discussion of precise time and frequency measurement" Sent: Tuesday, February 17, 2009 6:33:40 PM GMT -07:00 US/Canada Mountain Subject: Re: [time-nuts] Antennas, roofs Hal, If your new roof will have a ridge vent you can poke the antenna cables through it to the outside leaving enough slack to create a drip loop before continuing to the antenna(s). Since the ridge vent follows the slope of the roof the cable will exit on the down side so water will run off. Ken, WA2LBI On Tue, Feb 17, 2009 at 19:18, Hal Murray wrote: > > > I really need to put in a feed through to the roof because my Z3801A > > is struggling with an indoor antenna too; but the roof needs to get > > replaced first... > > Speaking of antennas and roofs.... > > Currently, my antennas are inside. That's good enough most of the time. > It's also good for providing nasty test cases to software. > > I also need a new roof. > > > I'm scheming to poke a hole in the roof so I can get some antennas in a > better position. > > If I have more than one antenna, does it matter how near eachother they are > located? > > > I'm picturing a plastic pipe that sticks up a few feet and a bracket at the > base that has the right magic angle to match the pitch of my roof. The > pipe > would screw or glue into the bracket. The bracket would get screwed to the > roof over a hole. The cables would go through the hole and up inside the > pipe. > > I haven't worked out the details for the top of the pipe yet. My > (handwave) > straw man is a U turn to keep the rain out, and mount the antennas on the > main pipe. Maybe a T to get them out to the side. > > > Do brackets like that exist? If so, what term or brand do I google for? I > have a typical not-very-steep sloped roof. Is there a standard angle? ... > > Plan B would be to stick the pipe through the roof and attach it inside to > the side of a rafter. I assume the roofers can treat it like a plumbing > vent > pipe. > > > A slightly crazy idea... Has anybody poked antennas up inside a skylight? > I'm thinking of the setup which has a hole in the ceiling of a room, a box > from that hole through the attic space up to and through the roof, and a > plastic dome on top. A shelf or bracket on the inside of the box would get > the antennas almost on the outside. > > > I've seen ads for metal roofs/shingles, the claimed advantage being long > life > which is attractive to me. I assume they would be a disaster for antennas > inside. What about outside, slightly above the roof? I'd expect bad > things, > but maybe there is some way to turn it into an advantage. > > > > -- > These are my opinions, not necessarily my employer's. I hate spam. > > > > > _______________________________________________ > time-nuts mailing list -- time-nuts at febo.com > To unsubscribe, go to > https://www.febo.com/cgi-bin/mailman/listinfo/time-nuts > and follow the instructions there. > _______________________________________________ time-nuts mailing list -- time-nuts at febo.com To unsubscribe, go to https://www.febo.com/cgi-bin/mailman/listinfo/time-nuts and follow the instructions there. From phk at phk.freebsd.dk Wed Feb 18 09:00:40 2009 From: phk at phk.freebsd.dk (Poul-Henning Kamp) Date: Wed, 18 Feb 2009 09:00:40 +0000 Subject: [time-nuts] Antennas, roofs In-Reply-To: Your message of "Wed, 18 Feb 2009 08:45:39 GMT." <619502179.1235901234946739463.JavaMail.root@sz0108a.emeryville.ca.mail.comcast.net> Message-ID: <18959.1234947640@critter.freebsd.dk> In message <619502179.1235901234946739463.JavaMail.root at sz0108a.emeryville.ca.m ail.comcast.net>, d.seiter at comcast.net writes: >> A slightly crazy idea... Has anybody poked antennas up inside a skylight? Yes, it works, I have a M12T running that way. It is an acrylic or polycarbonate (not sure) dome type skylight. When I replaced the roof on my house, I had three 1/2" wirepipes installed over one of the attic windows and had three irons mounted under the tiles for attachment of antennas: http://phk.freebsd.dk/misc/dscf0180.jpg So now I can just open the window and fiddle with my antennas as much as I like. The wirepipes are wide enough that a BNC can be squeezed through. Poul-Henning -- Poul-Henning Kamp | UNIX since Zilog Zeus 3.20 phk at FreeBSD.ORG | TCP/IP since RFC 956 FreeBSD committer | BSD since 4.3-tahoe Never attribute to malice what can adequately be explained by incompetence. From magnus at rubidium.dyndns.org Wed Feb 18 21:35:55 2009 From: magnus at rubidium.dyndns.org (Magnus Danielson) Date: Wed, 18 Feb 2009 22:35:55 +0100 Subject: [time-nuts] Book announcement Message-ID: <499C7F3B.4090103@rubidium.dyndns.org> Dear fellow time-nuts! Enrico asked me to forward this to you, and I am happy to. I received my copy monday and is very excited about getting into this volume! It sums up much of his writing and presentations on oscillators and phase noise. There's even a nice phase-noise plot of various distribution amplifiers that John Ackermann is properly credited for. (I hit onto the ceiling twice, so no fancy graphic this time) Best Regards, Magnus 8<---- Dear time-nuts, I am pleased to inform you that my book on oscillator noise and stability is finally available. Cheers, Enrico LINK E. Rubiola, Phase noise and frequency stability in oscillators, Cambridge University Press http://www.cambridge.org/uk/catalogue/catalogue.asp?isbn=9780521886772 http://rubiola.org (click on "Oscillator noise" on the grey left column "Contents") Enrico Rubiola, Ph.D, Sc.D. professor of electronics web: http://rubiola.org e-mail: rubiola at femto-st.fr FEMTO-ST Institute 32 av. de l'Observatoire 25044 Besancon, FRANCE voice: +33 (0) 381.853940 (E.Rubiola) voice: +33 (0) 381.853999 (switchboard) fax: +33 (0) 381.853998 ---->8 From mpb45 at clanbaker.org Thu Feb 19 00:45:39 2009 From: mpb45 at clanbaker.org (Michael Baker) Date: Wed, 18 Feb 2009 19:45:39 -0500 Subject: [time-nuts] Lady Heather command line modification question... Message-ID: <499CABB3.2050609@clanbaker.org> Hello, John-- Every now and then, LH says my T-bolt PPS and DAC are experiencing wide excursions. The PPS OADEV and the OSC OADEV both stay down around 6 X e-13 but PPS and DAC both get hinky for an hour or two and then things settle back down for a few hours. This process seems to repeat itself 2X or 3X per day. It may be due to all the tall trees around my place and my less than ideal sky view. In any event, every few hours the violet PPS line and the green DAC line go off chart and it would be nice to be able to expand the scale just a bit to better see what is happening. I am trying to change the scale for the PPS graph from 2ns/div to 5ns/div and the DAC from 50uv/div to 100uv/div I have been unsuccessful at entering the appropriate command line modifiers-- I right-click on the Lady Heather desktop icon and select PROPERTIES. In the PROPERTIES box I see the Target address which is: "C:\Program Files\Heather\heather.exe" I put a space after the .exe and enter /mp[=5] and close the quote I enter OK and then get an error message that says: The name 'C:Program Files\Heather\heather.exe /mp[=5]' specified in the Target box is not valid. Make sure the path and file name are correct. What am I doing wrong? Maybe I am not clear where to enter spaces? I have tried different combinations, but no joy... Is this not the correct place to enter the /mp=5 suffix? Thanks, Mike Baker ------------------------- From wa2lbi at frontiernet.net Thu Feb 19 01:02:19 2009 From: wa2lbi at frontiernet.net (Ken Winterling) Date: Wed, 18 Feb 2009 20:02:19 -0500 Subject: [time-nuts] Lady Heather command line modification question... In-Reply-To: <499CABB3.2050609@clanbaker.org> References: <499CABB3.2050609@clanbaker.org> Message-ID: <608ed23c0902181702x2554b405x18964193b8a9db7e@mail.gmail.com> Mike, Put the qualifier goes after the last quote. Here is how I specify that COM2 is to be used: "C:\Program Files\Heather\heather.exe" /2 Ken, WA2LBI On Wed, Feb 18, 2009 at 19:45, Michael Baker wrote: . . . > I am trying to change the scale for the PPS graph from 2ns/div > to 5ns/div and the DAC from 50uv/div to 100uv/div > > I have been unsuccessful at entering the appropriate command > line modifiers-- I right-click on the Lady Heather desktop icon > and select PROPERTIES. In the PROPERTIES box I see the Target > address which is: > > "C:\Program Files\Heather\heather.exe" > > I put a space after the .exe and enter /mp[=5] and close the quote > > I enter OK and then get an error message that says: > > The name 'C:Program Files\Heather\heather.exe /mp[=5]' specified > in the Target box is not valid. Make sure the path and > file name are correct. > > What am I doing wrong? Maybe I am not clear where to enter > spaces? I have tried different combinations, but no joy... > > Is this not the correct place to enter the /mp=5 suffix? > > Thanks, > > Mike Baker From jmiles at pop.net Thu Feb 19 01:30:27 2009 From: jmiles at pop.net (John Miles) Date: Wed, 18 Feb 2009 17:30:27 -0800 Subject: [time-nuts] Lady Heather command line modification question... In-Reply-To: <499CABB3.2050609@clanbaker.org> Message-ID: The parameters need to be outside the quotes, or they're considered part of the executable filename itself. (When I posted the latest version the other day, I added an example -- see http://www.ke5fx.com/heather/readme.htm and click on the small image of the shortcut dialog.) -- john, KE5FX > The name 'C:Program Files\Heather\heather.exe /mp[=5]' specified > in the Target box is not valid. Make sure the path and > file name are correct. > > What am I doing wrong? Maybe I am not clear where to enter > spaces? I have tried different combinations, but no joy... > > Is this not the correct place to enter the /mp=5 suffix? > From mikes at flatsurface.com Thu Feb 19 02:01:09 2009 From: mikes at flatsurface.com (Mike S) Date: Wed, 18 Feb 2009 21:01:09 -0500 Subject: [time-nuts] Lady Heather command line modification question... In-Reply-To: References: <499CABB3.2050609@clanbaker.org> Message-ID: <20090219020117.28DA51165C1@hamburg.alientech.net> John, I just started playing around with Lady Heather (for some reason, that sounds kinky :-) ) Any chance of adding the ability to read from the log file when starting, so the graphs/stats can be recovered from a log? And optionally, continue collecting/adding data to that log, as might be desired if a stop/start is necessary. Also, it's not documented that will quit the program. :-( From jmiles at pop.net Thu Feb 19 02:32:43 2009 From: jmiles at pop.net (John Miles) Date: Wed, 18 Feb 2009 18:32:43 -0800 Subject: [time-nuts] Lady Heather command line modification question... In-Reply-To: <20090219020117.28DA51165C1@hamburg.alientech.net> Message-ID: Correct (although it only takes one to exit if you're not in review mode.) I may add the ability to lock/unlock the keyboard in an upcoming version, so that it'll ignore any accidental user input. I agree that it would be cool if the initial graph contents could be obtained by reading the log file, so that you could continue logging data across execution sessions and scroll back as far as you want. That'd be Mark's call, though, as I only did the Windows conversion. -- john, KE5FX > -----Original Message----- > From: Mike S [mailto:mikes at flatsurface.com] > Sent: Wednesday, February 18, 2009 6:01 PM > To: Discussion of precise time and frequency measurement > Cc: John Miles > Subject: Re: [time-nuts] Lady Heather command line modification > question... > > > John, > > I just started playing around with Lady Heather (for some reason, that > sounds kinky :-) ) > > Any chance of adding the ability to read from the log file when > starting, so the graphs/stats can be recovered from a log? And > optionally, continue collecting/adding data to that log, as might be > desired if a stop/start is necessary. > > Also, it's not documented that will quit the program. :-( > > From wd6cmu at earthlink.net Thu Feb 19 03:15:03 2009 From: wd6cmu at earthlink.net (Eric Williams) Date: Wed, 18 Feb 2009 19:15:03 -0800 Subject: [time-nuts] Lady Heather command line modification question... In-Reply-To: <20090219020117.28DA51165C1@hamburg.alientech.net> References: <499CABB3.2050609@clanbaker.org> <20090219020117.28DA51165C1@hamburg.alientech.net> Message-ID: If someone is taking suggestions, I just thought it would be nice if I could get 9 values for the accumulated amount of time spent receiving N satellites, where N goes from 0 to 8. (Or more, for newer T-Bolts.) Seems like it would be useful for evaluating and comparing the quality of antenna sites. Thanks! From holrum at hotmail.com Thu Feb 19 04:40:27 2009 From: holrum at hotmail.com (Mark Sims) Date: Thu, 19 Feb 2009 04:40:27 +0000 Subject: [time-nuts] Lady Heather command line modification question... In-Reply-To: References: Message-ID: Yes, it would be nice to be able to load the log file and resume from that position, but what about the gap between when you stopped logging and restarted? That would destroy the integrity of the OADEV/ADEV numbers. You could just ignore the gap in the graphs, but the OADEV numbers would be utterly meaningless. The log file is opened in APPEND mode where the new data is appended to the log each time the log is started (assuming you did not delete the log before starting the program). It should be fairly easy to import the log into a spreadsheet to produce graphs over multiple runs. Note hat each hour in the log is flagged with a few lines of header beginning with a '#'. Same for comments that flag potentially important state changes. ESC is used to exit the program. If you happen to be scrolling back the plot data ESC (or just about any function key, I use DEL) will end that mode. The next ESC will exit the program. If you did not intend to exit the program, consult your Dictionary of Russian Sailor Curse Words and Phrases. Note that on some systems, the DOS version of the program may occasionally appear to lock up when you press a key. You can usually recover by pressing any key 16-20 times until the keyboard buffer overflows. You should hear a beep from the system BIOS. The next keystroke or two should unlock the program. If it does not, consult your Dictionary of Drunken Russian Sailor Curse Words and Phrases... mine is rather worn out in that area... I recomend a key that is rather meaningless to the program (like SPACE). Any suggestions for additions/improvements to the program are greatly appreciated... ********************** _________________________________________________________________ Want to do more with Windows Live? Learn ?10 hidden secrets? from Jamie. http://windowslive.com/connect/post/jamiethomson.spaces.live.com-Blog-cns!550F681DAD532637!5295.entry?ocid=TXT_TAGLM_WL_domore_092008 From stanw1le at verizon.net Thu Feb 19 13:00:01 2009 From: stanw1le at verizon.net (Stan W1LE) Date: Thu, 19 Feb 2009 08:00:01 -0500 Subject: [time-nuts] Lady Heather command line modification question... In-Reply-To: <499CABB3.2050609@clanbaker.org> References: <499CABB3.2050609@clanbaker.org> Message-ID: <499D57D1.9080902@verizon.net> You tried 'C:Program Files\Heather\heather.exe /mp[=5]' Try "C:\Heather\heather.exe" /mp5 But, I have been unable to get more than one switch to work, the first in line.. We need clarification on the exact syntax to use for multiple command line switches. Stan, W1LE Michael Baker wrote: > Hello, John-- > > Every now and then, LH says my T-bolt PPS and DAC are experiencing > wide excursions. The PPS OADEV and the OSC OADEV both stay down > around 6 X e-13 but PPS and DAC both get hinky for an hour or two > and then things settle back down for a few hours. This process > seems to repeat itself 2X or 3X per day. It may be due to all the > tall trees around my place and my less than ideal sky view. > > In any event, every few hours the violet PPS line and the green > DAC line go off chart and it would be nice to be able to expand > the scale just a bit to better see what is happening. > > I am trying to change the scale for the PPS graph from 2ns/div > to 5ns/div and the DAC from 50uv/div to 100uv/div > > I have been unsuccessful at entering the appropriate command > line modifiers-- I right-click on the Lady Heather desktop icon > and select PROPERTIES. In the PROPERTIES box I see the Target > address which is: > > "C:\Program Files\Heather\heather.exe" > > I put a space after the .exe and enter /mp[=5] and close the quote > > I enter OK and then get an error message that says: > > The name 'C:Program Files\Heather\heather.exe /mp[=5]' specified > in the Target box is not valid. Make sure the path and > file name are correct. > > What am I doing wrong? Maybe I am not clear where to enter > spaces? I have tried different combinations, but no joy... > > Is this not the correct place to enter the /mp=5 suffix? > > Thanks, > > Mike Baker > ------------------------- > > > > > > > _______________________________________________ > time-nuts mailing list -- time-nuts at febo.com > To unsubscribe, go to https://www.febo.com/cgi-bin/mailman/listinfo/time-nuts > and follow the instructions there. > > From stanw1le at verizon.net Thu Feb 19 13:14:41 2009 From: stanw1le at verizon.net (Stan W1LE) Date: Thu, 19 Feb 2009 08:14:41 -0500 Subject: [time-nuts] Trimble T'Bolt Schematic for troubleshooting In-Reply-To: <499D57D1.9080902@verizon.net> References: <499CABB3.2050609@clanbaker.org> <499D57D1.9080902@verizon.net> Message-ID: <499D5B41.3070608@verizon.net> Hello The Net: My T'Bolt died and it was not the external power supply. Any pointers to troubleshooting techniques and a schematic ? I should first start ohming out the on board DC power distribution, then trace the various voltages, after DC power is applied. I am only finding one test point, TP423 in the GPS receiver section. Any idea of a good wave shape to be expected here ? Thanks Stan, W1LE Cape Cod FN41sr ZZZZZ From jmiles at pop.net Thu Feb 19 13:12:30 2009 From: jmiles at pop.net (John Miles) Date: Thu, 19 Feb 2009 05:12:30 -0800 Subject: [time-nuts] Lady Heather command line modification question... In-Reply-To: <499D57D1.9080902@verizon.net> Message-ID: Hmm. Are you following the example shown in the readme file? What is the exact command line you're using where multiple options aren't working? I normally use "C:\Heather\heather.exe" /i=10 /mt=130 ... myself. -- john, KE5FX > But, I have been unable to get more than one switch to work, the > first in line.. > > We need clarification on the exact syntax to use for multiple > command line switches. > > Stan, W1LE > From stanw1le at verizon.net Thu Feb 19 13:18:17 2009 From: stanw1le at verizon.net (Stan W1LE) Date: Thu, 19 Feb 2009 08:18:17 -0500 Subject: [time-nuts] Lady Heather command line modification question... In-Reply-To: <499D57D1.9080902@verizon.net> References: <499CABB3.2050609@clanbaker.org> <499D57D1.9080902@verizon.net> Message-ID: <499D5C19.50506@verizon.net> Got up too early this morning, no caffeine yet.... note correction. Stan W1LE wrote: > You tried > > 'C:Program Files\Heather\heather.exe /mp[=5]' > > Try > > "C:\Heather\heather.exe" /mp5 > Need to add the directory for the Program Files try "C:\Program Files\Heather\heather.exe" /mp5 Thanks for your patience Stan, W1LE From stanw1le at verizon.net Thu Feb 19 13:23:27 2009 From: stanw1le at verizon.net (Stan W1LE) Date: Thu, 19 Feb 2009 08:23:27 -0500 Subject: [time-nuts] Lady Heather command line modification question... In-Reply-To: References: Message-ID: <499D5D4F.7070807@verizon.net> Hello John, Thanks for the tip, I did not add the space between the switches, and I did not add the = sign. The example you posted on the readme file was difficult for me to read. I shall try the correct syntax later this morning after a caffeine break. Suggestion for Mark, How about some indication on the main screen what the interval (i) is ?? Stan, W1LE John Miles wrote: > Hmm. Are you following the example shown in the readme file? What is the > exact command line you're using where multiple options aren't working? > > I normally use > > "C:\Heather\heather.exe" /i=10 /mt=130 > > ... myself. > > -- john, KE5FX > > > >> But, I have been unable to get more than one switch to work, the >> first in line.. >> >> We need clarification on the exact syntax to use for multiple >> command line switches. >> >> Stan, W1LE >> >> > > > _______________________________________________ > time-nuts mailing list -- time-nuts at febo.com > To unsubscribe, go to https://www.febo.com/cgi-bin/mailman/listinfo/time-nuts > and follow the instructions there. > > From w1ksz at earthlink.net Thu Feb 19 22:27:45 2009 From: w1ksz at earthlink.net (Richard W. Solomon) Date: Thu, 19 Feb 2009 15:27:45 -0700 (GMT-07:00) Subject: [time-nuts] LUCENT RFTGm-II-XO Info ?? Message-ID: <25362754.1235082465503.JavaMail.root@elwamui-hound.atl.sa.earthlink.net> I snagged one of these "over there" and am looking for any documentation on it (connector pin-outs, schematics, manuals, etc). Can anyone point me to a source of this data ? Thanks, Dick, W1KSZ From stanley_reynolds at yahoo.com Fri Feb 20 01:21:40 2009 From: stanley_reynolds at yahoo.com (Stanley Reynolds) Date: Thu, 19 Feb 2009 17:21:40 -0800 (PST) Subject: [time-nuts] LUCENT RFTGm-II-XO Info ?? References: <25362754.1235082465503.JavaMail.root@elwamui-hound.atl.sa.earthlink.net> Message-ID: <615302.15663.qm@web30306.mail.mud.yahoo.com> I have some info on a RFG-M-XO not a "II" I'm not sure it is what you have. I also have some info on the GPS version. Link www.n4iqt.com/rfgmxo Stanley ________________________________ From: Richard W. Solomon To: time-nuts at febo.com Sent: Thursday, February 19, 2009 4:27:45 PM Subject: [time-nuts] LUCENT RFTGm-II-XO Info ?? I snagged one of these "over there" and am looking for any documentation on it (connector pin-outs, schematics, manuals, etc). Can anyone point me to a source of this data ? Thanks, Dick, W1KSZ _______________________________________________ time-nuts mailing list -- time-nuts at febo.com To unsubscribe, go to https://www.febo.com/cgi-bin/mailman/listinfo/time-nuts and follow the instructions there. From stanley_reynolds at yahoo.com Fri Feb 20 01:47:44 2009 From: stanley_reynolds at yahoo.com (Stanley Reynolds) Date: Thu, 19 Feb 2009 17:47:44 -0800 (PST) Subject: [time-nuts] Fw: LUCENT RFTGm-II-XO Info ?? Message-ID: <575437.271.qm@web30305.mail.mud.yahoo.com> Link to GPS stuff www.n4iqt.com/lucentgps . Stanley From bjones0 at mindspring.com Fri Feb 20 03:03:46 2009 From: bjones0 at mindspring.com (bjones0 at mindspring.com) Date: Thu, 19 Feb 2009 22:03:46 -0500 Subject: [time-nuts] Thunderbolt serial port In-Reply-To: <6.2.5.6.2.20080605163443.02d5c258@jpl.nasa.gov> References: <6F62D5A392B04D15853529C4BEDFCC85@athlon> <4848727E.60108@smiffytech.com> <6.2.5.6.2.20080605163443.02d5c258@jpl.nasa.gov> Message-ID: <1235099026.27417.13.camel@skuld> I'm tired of banging my head against the wall so I think it is time to ask.. Is there a way to test if the serial port on the Thunderbolt works? I know the Thunderbolt itself powers up and works (I get a 9.9999978 Mhz signal out of the 10Mhz output when it is on) but I have been unable to talk/listen to it in Windows (tboltmon or Lady Heather) or in Linux (even looking at the raw serial port state). I stuck my breakout box on the line but I get no lights on the Thunderbolt side. Thanks, Brian From jltran at worldnet.att.net Fri Feb 20 03:22:49 2009 From: jltran at worldnet.att.net (J. L. Trantham) Date: Thu, 19 Feb 2009 21:22:49 -0600 Subject: [time-nuts] Thunderbolt serial port In-Reply-To: <1235099026.27417.13.camel@skuld> Message-ID: <4F11911843594155A3267537DBAE5E82@S0028384766> Brian, I had a failure of a TAPR Thunderbolt where the 10 MHz out was present (though not disciplined and, therefore, not 10.0000000000 MHz on my 5370B when 'clocked' by another disciplined Thunderbolt) but there was no 1 PPS and there was no ability to communicate via the serial port. I have not had a chance to explore the origin of the failure but plan to at some point. I suspect you have a similar failure to what I had. The 10 MHz was close, like yours, but not exact. I would be very interested in knowing what you might find as the 'failed component'. Thanks, Joe -----Original Message----- From: time-nuts-bounces at febo.com [mailto:time-nuts-bounces at febo.com] On Behalf Of bjones0 at mindspring.com Sent: Thursday, February 19, 2009 9:04 PM To: time-nuts at febo.com Subject: [time-nuts] Thunderbolt serial port I'm tired of banging my head against the wall so I think it is time to ask.. Is there a way to test if the serial port on the Thunderbolt works? I know the Thunderbolt itself powers up and works (I get a 9.9999978 Mhz signal out of the 10Mhz output when it is on) but I have been unable to talk/listen to it in Windows (tboltmon or Lady Heather) or in Linux (even looking at the raw serial port state). I stuck my breakout box on the line but I get no lights on the Thunderbolt side. Thanks, Brian _______________________________________________ time-nuts mailing list -- time-nuts at febo.com To unsubscribe, go to https://www.febo.com/cgi-bin/mailman/listinfo/time-nuts and follow the instructions there. From brooke at pacific.net Fri Feb 20 03:35:25 2009 From: brooke at pacific.net (Brooke Clarke) Date: Thu, 19 Feb 2009 19:35:25 -0800 Subject: [time-nuts] Thunderbolt serial port In-Reply-To: <1235099026.27417.13.camel@skuld> References: <6F62D5A392B04D15853529C4BEDFCC85@athlon> <4848727E.60108@smiffytech.com> <6.2.5.6.2.20080605163443.02d5c258@jpl.nasa.gov> <1235099026.27417.13.camel@skuld> Message-ID: <499E24FD.2030608@pacific.net> Hi Brian: There have been a small number of problems related to poor solder joints on U9 (RS-232 chip) or the serial connector. see: http://www.prc68.com/I/ThunderBolt.shtml#Maint Have Fun, Brooke Clarke http://www.prc68.com bjones0 at mindspring.com wrote: > I'm tired of banging my head against the wall so I think it is time to > ask.. Is there a way to test if the serial port on the Thunderbolt > works? I know the Thunderbolt itself powers up and works (I get a > 9.9999978 Mhz signal out of the 10Mhz output when it is on) but I have > been unable to talk/listen to it in Windows (tboltmon or Lady Heather) > or in Linux (even looking at the raw serial port state). I stuck my > breakout box on the line but I get no lights on the Thunderbolt side. > > Thanks, > Brian > > > _______________________________________________ > time-nuts mailing list -- time-nuts at febo.com > To unsubscribe, go to https://www.febo.com/cgi-bin/mailman/listinfo/time-nuts > and follow the instructions there. > > From mcleann at bigpond.com Fri Feb 20 04:33:10 2009 From: mcleann at bigpond.com (Nic McLean) Date: Fri, 20 Feb 2009 15:33:10 +1100 Subject: [time-nuts] Thunderbolt serial port In-Reply-To: <499E24FD.2030608@pacific.net> References: <6F62D5A392B04D15853529C4BEDFCC85@athlon> <4848727E.60108@smiffytech.com> <6.2.5.6.2.20080605163443.02d5c258@jpl.nasa.gov><1235099026.27417.13.camel@skuld> <499E24FD.2030608@pacific.net> Message-ID: Are you using a straight through or crossover serial cable? Nic -----Original Message----- From: time-nuts-bounces at febo.com [mailto:time-nuts-bounces at febo.com] On Behalf Of Brooke Clarke Sent: Friday, 20 February 2009 2:35 PM To: Discussion of precise time and frequency measurement Subject: Re: [time-nuts] Thunderbolt serial port Hi Brian: There have been a small number of problems related to poor solder joints on U9 (RS-232 chip) or the serial connector. see: http://www.prc68.com/I/ThunderBolt.shtml#Maint Have Fun, Brooke Clarke http://www.prc68.com bjones0 at mindspring.com wrote: > I'm tired of banging my head against the wall so I think it is time to > ask.. Is there a way to test if the serial port on the Thunderbolt > works? I know the Thunderbolt itself powers up and works (I get a > 9.9999978 Mhz signal out of the 10Mhz output when it is on) but I have > been unable to talk/listen to it in Windows (tboltmon or Lady Heather) > or in Linux (even looking at the raw serial port state). I stuck my > breakout box on the line but I get no lights on the Thunderbolt side. > > Thanks, > Brian > > > _______________________________________________ > time-nuts mailing list -- time-nuts at febo.com > To unsubscribe, go to https://www.febo.com/cgi-bin/mailman/listinfo/time-nuts > and follow the instructions there. > > _______________________________________________ time-nuts mailing list -- time-nuts at febo.com To unsubscribe, go to https://www.febo.com/cgi-bin/mailman/listinfo/time-nuts and follow the instructions there. From mcleann at bigpond.com Fri Feb 20 04:35:08 2009 From: mcleann at bigpond.com (Nic McLean) Date: Fri, 20 Feb 2009 15:35:08 +1100 Subject: [time-nuts] FW: Thunderbolt serial port Message-ID: <1E225DFC3BC541328E2B57992FCE27EC@PC755913417801> Are you using a straight or crossover serial cable? Nic -----Original Message----- From: time-nuts-bounces at febo.com [mailto:time-nuts-bounces at febo.com] On Behalf Of bjones0 at mindspring.com Sent: Friday, 20 February 2009 2:04 PM To: time-nuts at febo.com Subject: [time-nuts] Thunderbolt serial port I'm tired of banging my head against the wall so I think it is time to ask.. Is there a way to test if the serial port on the Thunderbolt works? I know the Thunderbolt itself powers up and works (I get a 9.9999978 Mhz signal out of the 10Mhz output when it is on) but I have been unable to talk/listen to it in Windows (tboltmon or Lady Heather) or in Linux (even looking at the raw serial port state). I stuck my breakout box on the line but I get no lights on the Thunderbolt side. Thanks, Brian _______________________________________________ time-nuts mailing list -- time-nuts at febo.com To unsubscribe, go to https://www.febo.com/cgi-bin/mailman/listinfo/time-nuts and follow the instructions there. From eseacord at cloud9.net Fri Feb 20 04:48:47 2009 From: eseacord at cloud9.net (ESeacord) Date: Thu, 19 Feb 2009 23:48:47 -0500 Subject: [time-nuts] Lady Heather question serial error In-Reply-To: Message-ID: Hi all After finally getting the right serial cable, only uses tx-rx and ground. Everytime I start Lady Heather, I get the error message "serial receive error, code 0x2" The only other process running is outlook. The tbolt monitor program dspmon and tboltmon both appear to work properly. My system is 2.6ghz intel quad core with 4 gig of mem and plenty of hard drive space. Any ideas anyone?? Thanks much Rix Seacord 22 Austin Rd Mahopac, NY 10541 WA2TGF (Extra) Home 845-628-0892 Cell 914-262-9186 ??? -----Original Message----- From: time-nuts-bounces at febo.com [mailto:time-nuts-bounces at febo.com] On Behalf Of Mark Sims Sent: Wednesday, February 18, 2009 11:40 PM To: time-nuts at febo.com Subject: [time-nuts] Lady Heather command line modification question... Yes, it would be nice to be able to load the log file and resume from that position, but what about the gap between when you stopped logging and restarted? That would destroy the integrity of the OADEV/ADEV numbers. You could just ignore the gap in the graphs, but the OADEV numbers would be utterly meaningless. The log file is opened in APPEND mode where the new data is appended to the log each time the log is started (assuming you did not delete the log before starting the program). It should be fairly easy to import the log into a spreadsheet to produce graphs over multiple runs. Note hat each hour in the log is flagged with a few lines of header beginning with a '#'. Same for comments that flag potentially important state changes. ESC is used to exit the program. If you happen to be scrolling back the plot data ESC (or just about any function key, I use DEL) will end that mode. The next ESC will exit the program. If you did not intend to exit the program, consult your Dictionary of Russian Sailor Curse Words and Phrases. Note that on some systems, the DOS version of the program may occasionally appear to lock up when you press a key. You can usually recover by pressing any key 16-20 times until the keyboard buffer overflows. You should hear a beep from the system BIOS. The next keystroke or two should unlock the program. If it does not, consult your Dictionary of Drunken Russian Sailor Curse Words and Phrases... mine is rather worn out in that area... I recomend a key that is rather meaningless to the program (like SPACE). Any suggestions for additions/improvements to the program are greatly appreciated... ********************** _________________________________________________________________ Want to do more with Windows Live? Learn ?10 hidden secrets? from Jamie. http://windowslive.com/connect/post/jamiethomson.spaces.live.com-Blog-cns!55 0F681DAD532637!5295.entry?ocid=TXT_TAGLM_WL_domore_092008 _______________________________________________ time-nuts mailing list -- time-nuts at febo.com To unsubscribe, go to https://www.febo.com/cgi-bin/mailman/listinfo/time-nuts and follow the instructions there. From jmiles at pop.net Fri Feb 20 04:56:45 2009 From: jmiles at pop.net (John Miles) Date: Thu, 19 Feb 2009 20:56:45 -0800 Subject: [time-nuts] Lady Heather question serial error In-Reply-To: Message-ID: Hmm. No, I have no idea what would do that. :( There's some sort of handshaking incompatibility, it sounds like. Try a fully-populated serial cable, just to see if it makes a difference. -- john, KE5FX > -----Original Message----- > From: time-nuts-bounces at febo.com [mailto:time-nuts-bounces at febo.com]On > Behalf Of ESeacord > Sent: Thursday, February 19, 2009 8:49 PM > To: 'Discussion of precise time and frequency measurement' > Subject: Re: [time-nuts] Lady Heather question serial error > > > Hi all > After finally getting the right serial cable, only uses tx-rx and ground. > Everytime I start Lady Heather, I get the error message "serial receive > error, code 0x2" > The only other process running is outlook. > The tbolt monitor program dspmon and tboltmon both appear to work > properly. > My system is 2.6ghz intel quad core with 4 gig of mem and plenty of hard > drive space. > Any ideas anyone?? > Thanks much > > > Rix Seacord > 22 Austin Rd > Mahopac, NY 10541 > WA2TGF (Extra) > Home 845-628-0892 > Cell 914-262-9186 > ??? > > From bjones0 at mindspring.com Fri Feb 20 11:01:17 2009 From: bjones0 at mindspring.com (bjones0 at mindspring.com) Date: Fri, 20 Feb 2009 06:01:17 -0500 Subject: [time-nuts] Thunderbolt serial port In-Reply-To: References: <6F62D5A392B04D15853529C4BEDFCC85@athlon> <4848727E.60108@smiffytech.com> <6.2.5.6.2.20080605163443.02d5c258@jpl.nasa.gov> <1235099026.27417.13.camel@skuld> <499E24FD.2030608@pacific.net> Message-ID: <1235127677.30329.3.camel@skuld> Nic - I have tried several of each... and even tried using the breakout box to create a crossover/null modem but no luck. Brooke - Thanks for the insight. I will tear in to it this weekend and see what I can find. Thanks. -Brian On Fri, 2009-02-20 at 15:33 +1100, Nic McLean wrote: > Are you using a straight through or crossover serial cable? > Nic > > -----Original Message----- > From: time-nuts-bounces at febo.com [mailto:time-nuts-bounces at febo.com] On > Behalf Of Brooke Clarke > Sent: Friday, 20 February 2009 2:35 PM > To: Discussion of precise time and frequency measurement > Subject: Re: [time-nuts] Thunderbolt serial port > > Hi Brian: > > There have been a small number of problems related to poor solder joints on > U9 > (RS-232 chip) or the serial connector. see: > http://www.prc68.com/I/ThunderBolt.shtml#Maint > > Have Fun, > > Brooke Clarke > http://www.prc68.com > > bjones0 at mindspring.com wrote: > > I'm tired of banging my head against the wall so I think it is time to > > ask.. Is there a way to test if the serial port on the Thunderbolt > > works? I know the Thunderbolt itself powers up and works (I get a > > 9.9999978 Mhz signal out of the 10Mhz output when it is on) but I have > > been unable to talk/listen to it in Windows (tboltmon or Lady Heather) > > or in Linux (even looking at the raw serial port state). I stuck my > > breakout box on the line but I get no lights on the Thunderbolt side. > > > > Thanks, > > Brian > > > > > > _______________________________________________ > > time-nuts mailing list -- time-nuts at febo.com > > To unsubscribe, go to > https://www.febo.com/cgi-bin/mailman/listinfo/time-nuts > > and follow the instructions there. > > > > > > > _______________________________________________ > time-nuts mailing list -- time-nuts at febo.com > To unsubscribe, go to > https://www.febo.com/cgi-bin/mailman/listinfo/time-nuts > and follow the instructions there. > > > > _______________________________________________ > time-nuts mailing list -- time-nuts at febo.com > To unsubscribe, go to https://www.febo.com/cgi-bin/mailman/listinfo/time-nuts > and follow the instructions there. From wb6bnq at cox.net Fri Feb 20 11:21:55 2009 From: wb6bnq at cox.net (WB6BNQ) Date: Fri, 20 Feb 2009 03:21:55 -0800 Subject: [time-nuts] Thunderbolt serial port References: <6F62D5A392B04D15853529C4BEDFCC85@athlon> <4848727E.60108@smiffytech.com> <6.2.5.6.2.20080605163443.02d5c258@jpl.nasa.gov> <1235099026.27417.13.camel@skuld> <499E24FD.2030608@pacific.net> <1235127677.30329.3.camel@skuld> Message-ID: <499E9253.41986FA9@cox.net> Brian, Have you measured the voltages with a scope to see if it may not be a compliance problem ? Bill....WB6BNQ "bjones0 at mindspring.com" wrote: > Nic - I have tried several of each... and even tried using the breakout > box to create a crossover/null modem but no luck. > > Brooke - > Thanks for the insight. I will tear in to it this weekend and see what > I can find. > > Thanks. > > -Brian > > On Fri, 2009-02-20 at 15:33 +1100, Nic McLean wrote: > > Are you using a straight through or crossover serial cable? > > Nic > > > > -----Original Message----- > > From: time-nuts-bounces at febo.com [mailto:time-nuts-bounces at febo.com] On > > Behalf Of Brooke Clarke > > Sent: Friday, 20 February 2009 2:35 PM > > To: Discussion of precise time and frequency measurement > > Subject: Re: [time-nuts] Thunderbolt serial port > > > > Hi Brian: > > > > There have been a small number of problems related to poor solder joints on > > U9 > > (RS-232 chip) or the serial connector. see: > > http://www.prc68.com/I/ThunderBolt.shtml#Maint > > > > Have Fun, > > > > Brooke Clarke > > http://www.prc68.com > > > > bjones0 at mindspring.com wrote: > > > I'm tired of banging my head against the wall so I think it is time to > > > ask.. Is there a way to test if the serial port on the Thunderbolt > > > works? I know the Thunderbolt itself powers up and works (I get a > > > 9.9999978 Mhz signal out of the 10Mhz output when it is on) but I have > > > been unable to talk/listen to it in Windows (tboltmon or Lady Heather) > > > or in Linux (even looking at the raw serial port state). I stuck my > > > breakout box on the line but I get no lights on the Thunderbolt side. > > > > > > Thanks, > > > Brian > > > > > > > > > _______________________________________________ > > > time-nuts mailing list -- time-nuts at febo.com > > > To unsubscribe, go to > > https://www.febo.com/cgi-bin/mailman/listinfo/time-nuts > > > and follow the instructions there. > > > > > > > > > > > > _______________________________________________ > > time-nuts mailing list -- time-nuts at febo.com > > To unsubscribe, go to > > https://www.febo.com/cgi-bin/mailman/listinfo/time-nuts > > and follow the instructions there. > > > > > > > > _______________________________________________ > > time-nuts mailing list -- time-nuts at febo.com > > To unsubscribe, go to https://www.febo.com/cgi-bin/mailman/listinfo/time-nuts > > and follow the instructions there. > > _______________________________________________ > time-nuts mailing list -- time-nuts at febo.com > To unsubscribe, go to https://www.febo.com/cgi-bin/mailman/listinfo/time-nuts > and follow the instructions there. From daun at yeagley.net Fri Feb 20 14:02:21 2009 From: daun at yeagley.net (Daun Yeagley) Date: Fri, 20 Feb 2009 09:02:21 -0500 Subject: [time-nuts] FW: [Accutrons] clock Message-ID: <5B7632C9D1A7480F82312FF88AC4A8A0@daundell> Here's an interesting item that one of the guys on the Accutron group posted. A great Geek item, especially for time nuts. Daun ________________________________ From: accutrons at yahoogroups.com Sent: Friday, February 20, 2009 7:16 AM To: Accutrons Group Subject: [Accutrons] clock Good morning all, This is a little off topic but it is still about time. One of my kids found this site. This T-shirt is for the ultimate time person. http://www.x-tremegeek.com/templates/TshirtStoreDetail.asp?StoreID=1&StyleID =202 a little pricy, but funny. Dick Here's a TinyURL version of the link: http://tinyurl.com/c6dxqj Daun From gretchendenisebaxter at gmail.com Fri Feb 20 14:43:16 2009 From: gretchendenisebaxter at gmail.com (Gretchen Baxter) Date: Fri, 20 Feb 2009 09:43:16 -0500 Subject: [time-nuts] new white paper on network time sync. Message-ID: <4fdb424f0902200643y77e85cd3h16f61e4e883a61@mail.gmail.com> Hello, just saw this new white paper on network time sync. I think it is really for mgmt., lets them know this time synchronization is important. http://www.ins.com/WorkArea/showcontent.aspx?id=2926 /gdb From ch at murgatroid.com Fri Feb 20 18:16:30 2009 From: ch at murgatroid.com (christopher hoover) Date: Fri, 20 Feb 2009 10:16:30 -0800 Subject: [time-nuts] Initial release of Lady Heather for Windows In-Reply-To: References: Message-ID: <003f01c99387$5a8c6c30$0fa54490$@com> John Miles wrote: > Thanks for the confirmation -- I don't believe anyone else has tried it > on > Windows 7 before. Did you have to run it as admin? > > Christopher, I imagine there's either a WoW64 compatibility problem or > (more > likely) the unit you're using isn't 100% compatible with Trimble's TSIP > implementation for the Thunderbolt. It would be interesting to hear if > you > can try 64-bit Vista with a real Thunderbolt; John, With a real Thunderbolt, it works perfectly fine on real COM1: under Vista SP1 x64 without elevation (UAC on). It just doesn't like the Trimble GPSTM's TSIP. (BTW, anyone else have one of these?) Thanks guys. Cheers, -ch From steve65 at suddenlink.net Fri Feb 20 19:35:29 2009 From: steve65 at suddenlink.net (Steve) Date: Fri, 20 Feb 2009 14:35:29 -0500 Subject: [time-nuts] Does the Thunderbolt have to be connected to a real serial port? Message-ID: <499F0601.9030508@suddenlink.net> Does the Thunderbolt have to be connected to a real serial port to operate properly with tboltmon.exe and Lady Heather? Or will it run through a USB to serial adapter such as those manufactured by Belkin and Byterunner? Thanks. Steve K8JQ From pe1rks at xs4all.nl Fri Feb 20 19:52:05 2009 From: pe1rks at xs4all.nl (S. Nestra) Date: Fri, 20 Feb 2009 20:52:05 +0100 Subject: [time-nuts] Does the Thunderbolt have to be connected to a real serial port? In-Reply-To: <499F0601.9030508@suddenlink.net> References: <499F0601.9030508@suddenlink.net> Message-ID: <499F09E5.4020300@xs4all.nl> I am using the Thunderbolt and Lady Heather on WinXP through a USB -> serial adapter with a prolific chipset without any problems. So it is possible. Stijn PE1RKS Steve schreef: > Does the Thunderbolt have to be connected to a real serial port to > operate properly with tboltmon.exe and Lady Heather? Or will it run > through a USB to serial adapter such as those manufactured by Belkin and > Byterunner? > Thanks. > > Steve K8JQ > > > _______________________________________________ > time-nuts mailing list -- time-nuts at febo.com > To unsubscribe, go to https://www.febo.com/cgi-bin/mailman/listinfo/time-nuts > and follow the instructions there. > -- Real radios have motors. From phk at phk.freebsd.dk Fri Feb 20 19:55:29 2009 From: phk at phk.freebsd.dk (Poul-Henning Kamp) Date: Fri, 20 Feb 2009 19:55:29 +0000 Subject: [time-nuts] Does the Thunderbolt have to be connected to a real serial port? In-Reply-To: Your message of "Fri, 20 Feb 2009 20:52:05 +0100." <499F09E5.4020300@xs4all.nl> Message-ID: <35837.1235159729@critter.freebsd.dk> In message <499F09E5.4020300 at xs4all.nl>, "S. Nestra" writes: Quite a lot of USB-serial adapters cheat at various levels. The worst output 0/+5V or +/-2.5V either of which are totally outside the spec at the receiving end. The acceptable output +/-5V, which i out of spec for a transmitter, but inside spec for a receiver, so a short or no cable, it works fine. The good output +/-9V or more as the spec requires. -- Poul-Henning Kamp | UNIX since Zilog Zeus 3.20 phk at FreeBSD.ORG | TCP/IP since RFC 956 FreeBSD committer | BSD since 4.3-tahoe Never attribute to malice what can adequately be explained by incompetence. From sam at canardpc.com Fri Feb 20 20:20:32 2009 From: sam at canardpc.com (Samuel D. [x86/CPC]) Date: Fri, 20 Feb 2009 21:20:32 +0100 Subject: [time-nuts] Does the Thunderbolt have to be connected to a real serial port? In-Reply-To: <35837.1235159729@critter.freebsd.dk> References: Your message of "Fri, 20 Feb 2009 20:52:05 +0100." <499F09E5.4020300@xs4all.nl> <35837.1235159729@critter.freebsd.dk> Message-ID: <00b901c99398$ae9ac360$0bd04a20$@com> > The good output +/-9V or more as the spec requires. According to the official RS232C specifications, the valid voltage range must be between +/-3V and +/-15V (+/-25V for the first draft). A device that absolutly require +/-9V is not RS232 compliant. BTW, as you said, some cheap USB<->Serial adapters output 0/+5V TTL levels and don't work with many devices. -----Message d'origine----- De?: time-nuts-bounces at febo.com [mailto:time-nuts-bounces at febo.com] De la part de Poul-Henning Kamp Envoy??: vendredi 20 f?vrier 2009 20:55 ??: Discussion of precise time and frequency measurement Objet?: Re: [time-nuts] Does the Thunderbolt have to be connected to a real serial port? In message <499F09E5.4020300 at xs4all.nl>, "S. Nestra" writes: Quite a lot of USB-serial adapters cheat at various levels. The worst output 0/+5V or +/-2.5V either of which are totally outside the spec at the receiving end. The acceptable output +/-5V, which i out of spec for a transmitter, but inside spec for a receiver, so a short or no cable, it works fine. The good output +/-9V or more as the spec requires. -- Poul-Henning Kamp | UNIX since Zilog Zeus 3.20 phk at FreeBSD.ORG | TCP/IP since RFC 956 FreeBSD committer | BSD since 4.3-tahoe Never attribute to malice what can adequately be explained by incompetence. _______________________________________________ time-nuts mailing list -- time-nuts at febo.com To unsubscribe, go to https://www.febo.com/cgi-bin/mailman/listinfo/time-nuts and follow the instructions there. From Arnold.Tibus at gmx.de Fri Feb 20 20:21:30 2009 From: Arnold.Tibus at gmx.de (Arnold Tibus) Date: Fri, 20 Feb 2009 21:21:30 +0100 Subject: [time-nuts] Does the Thunderbolt have to be connected to a real serial port? In-Reply-To: <499F0601.9030508@suddenlink.net> Message-ID: 'Lady Heather's' Thunderbolt control software (where does the name come from?) does work quite well under Windows XP even via Ethernet cable and an Ethernet to RS232 converter and a virtual com port! Both ways are running, DOS Window and graphical Windows. It is important to set the right com port number option and having Thunderbolt connected and running. If not I get: CreateFile error: Port unavailable or in use (i) Access denied [ok] Many thanks to the developers! regards Arnold, DK2WT On Fri, 20 Feb 2009 14:35:29 -0500, Steve wrote: >Does the Thunderbolt have to be connected to a real serial port to >operate properly with tboltmon.exe and Lady Heather? Or will it run >through a USB to serial adapter such as those manufactured by Belkin and >Byterunner? >Thanks. > >Steve K8JQ > > >_______________________________________________ >time-nuts mailing list -- time-nuts at febo.com >To unsubscribe, go to https://www.febo.com/cgi-bin/mailman/listinfo/time-nuts >and follow the instructions there. > From eseacord at cloud9.net Sat Feb 21 03:20:31 2009 From: eseacord at cloud9.net (ESeacord) Date: Fri, 20 Feb 2009 22:20:31 -0500 Subject: [time-nuts] Lady Heather question serial error In-Reply-To: Message-ID: John Thanks much for getting my head on straight. I bought(ugh) a standard rs232 9 pin cable and part of the problem left me. I was using a Belkin usb to serial adapter(f5u109) but it still gave me a problem. Switching to a Hawking(h-uc232s) adapter and reloading the drivers caused all the problems to go in hiding. Initially the hawking made my system think it was a mouse when the thunderbolt was attached. Its been running for a couple of hours now and seems to have settled down. Thanks again Rix Seacord 22 Austin Rd Mahopac, NY 10541 WA2TGF (Extra) Home 845-628-0892 Cell 914-262-9186 ??? -----Original Message----- From: time-nuts-bounces at febo.com [mailto:time-nuts-bounces at febo.com] On Behalf Of John Miles Sent: Thursday, February 19, 2009 11:57 PM To: Discussion of precise time and frequency measurement Subject: Re: [time-nuts] Lady Heather question serial error Hmm. No, I have no idea what would do that. :( There's some sort of handshaking incompatibility, it sounds like. Try a fully-populated serial cable, just to see if it makes a difference. -- john, KE5FX > -----Original Message----- > From: time-nuts-bounces at febo.com [mailto:time-nuts-bounces at febo.com]On > Behalf Of ESeacord > Sent: Thursday, February 19, 2009 8:49 PM > To: 'Discussion of precise time and frequency measurement' > Subject: Re: [time-nuts] Lady Heather question serial error > > > Hi all > After finally getting the right serial cable, only uses tx-rx and ground. > Everytime I start Lady Heather, I get the error message "serial receive > error, code 0x2" > The only other process running is outlook. > The tbolt monitor program dspmon and tboltmon both appear to work > properly. > My system is 2.6ghz intel quad core with 4 gig of mem and plenty of hard > drive space. > Any ideas anyone?? > Thanks much > > > Rix Seacord > 22 Austin Rd > Mahopac, NY 10541 > WA2TGF (Extra) > Home 845-628-0892 > Cell 914-262-9186 > ??? > > _______________________________________________ time-nuts mailing list -- time-nuts at febo.com To unsubscribe, go to https://www.febo.com/cgi-bin/mailman/listinfo/time-nuts and follow the instructions there. From david.partridge at dsl.pipex.com Sat Feb 21 06:40:43 2009 From: david.partridge at dsl.pipex.com (David C. Partridge) Date: Sat, 21 Feb 2009 06:40:43 -0000 Subject: [time-nuts] Lady Heather question serial error In-Reply-To: References: Message-ID: Yes, Windows always seems to think a Thunderbolt is an MS mouse. Look back in the archives about using /Fastdetect in boot.ini (BTW this doesn't work with WinXP 64 - so I had to disable the serial mouse in the devices). Dave -----Original Message----- From: time-nuts-bounces at febo.com [mailto:time-nuts-bounces at febo.com] On Behalf Of ESeacord Sent: 21 February 2009 03:21 To: 'Discussion of precise time and frequency measurement' Subject: Re: [time-nuts] Lady Heather question serial error John Thanks much for getting my head on straight. I bought(ugh) a standard rs232 9 pin cable and part of the problem left me. I was using a Belkin usb to serial adapter(f5u109) but it still gave me a problem. Switching to a Hawking(h-uc232s) adapter and reloading the drivers caused all the problems to go in hiding. Initially the hawking made my system think it was a mouse when the thunderbolt was attached. Its been running for a couple of hours now and seems to have settled down. Thanks again Rix Seacord 22 Austin Rd Mahopac, NY 10541 WA2TGF (Extra) Home 845-628-0892 Cell 914-262-9186 ??? -----Original Message----- From: time-nuts-bounces at febo.com [mailto:time-nuts-bounces at febo.com] On Behalf Of John Miles Sent: Thursday, February 19, 2009 11:57 PM To: Discussion of precise time and frequency measurement Subject: Re: [time-nuts] Lady Heather question serial error Hmm. No, I have no idea what would do that. :( There's some sort of handshaking incompatibility, it sounds like. Try a fully-populated serial cable, just to see if it makes a difference. -- john, KE5FX > -----Original Message----- > From: time-nuts-bounces at febo.com [mailto:time-nuts-bounces at febo.com]On > Behalf Of ESeacord > Sent: Thursday, February 19, 2009 8:49 PM > To: 'Discussion of precise time and frequency measurement' > Subject: Re: [time-nuts] Lady Heather question serial error > > > Hi all > After finally getting the right serial cable, only uses tx-rx and ground. > Everytime I start Lady Heather, I get the error message "serial > receive error, code 0x2" > The only other process running is outlook. > The tbolt monitor program dspmon and tboltmon both appear to work > properly. > My system is 2.6ghz intel quad core with 4 gig of mem and plenty of > hard drive space. > Any ideas anyone?? > Thanks much > > > Rix Seacord > 22 Austin Rd > Mahopac, NY 10541 > WA2TGF (Extra) > Home 845-628-0892 > Cell 914-262-9186 > ??? > > _______________________________________________ time-nuts mailing list -- time-nuts at febo.com To unsubscribe, go to https://www.febo.com/cgi-bin/mailman/listinfo/time-nuts and follow the instructions there. _______________________________________________ time-nuts mailing list -- time-nuts at febo.com To unsubscribe, go to https://www.febo.com/cgi-bin/mailman/listinfo/time-nuts and follow the instructions there. From mpb45 at clanbaker.org Sat Feb 21 07:04:22 2009 From: mpb45 at clanbaker.org (Michael Baker) Date: Sat, 21 Feb 2009 02:04:22 -0500 Subject: [time-nuts] Weird Lady Heather behavior observed...?? Message-ID: <499FA776.7090805@clanbaker.org> Hello, Time-Nutters-- Is there any possibility that opening and closing the Lady Heather display inflicts some sort of temporary disturbance to the T-bolt it is monitoring? If I run the Lady Heather display on my Dell Dimension 9200 desktop screen and let it run undisturbed for 8 or 10 hours, it tells me that my T-bolt has settled down nicely; PPS staying within +/- 2ns to 4ns and Osc never straying outside of roughly +0.01ppb to -0.01ppb. The PPS OADEV gets down to the 4e-13 category and the OSC OADEV to around the 3e-13 category. The green DAC graph stays well centered within one division over or under (@ 100uV/div) and the violet PPS graph also stays well centered within +/- one division (100uV/div). The yellow temp line stays centered inside one division. Everything looks great.... right...?? As long as I do not minimize or close and then re-open the LH display, this situation will continue for many hours. HOWEVER-- if I close the LH display and open it back up shortly thereafter, all my T-bolt running parameters start fluctuating wildly and going off-scale. Even the yellow temp line goes off scale and PPS OADEV and OSC OADEV both drop back to a high e-12. It takes about 90 minutes for everything to settle back down, after which everything runs tightly for many hours until the next time I minimize or close and re-open the LH display. If this is just a coincidence, it sure has happened with consistent regularity over the last week or so since I first noticed this behavior. Is this just a coincidence? Has anyone else noticed anything like this? I can send anyone a before and after screen capture of what I am describing. Mike Baker WA4HFR Gainesville (Micanopy) Florida USA --------------------------------------------- From mkern at fastmail.fm Sat Feb 21 14:23:24 2009 From: mkern at fastmail.fm (Markus Kern) Date: Sat, 21 Feb 2009 15:23:24 +0100 Subject: [time-nuts] Different Thunderbolt versions Message-ID: <19210986468.20090221152324@fastmail.fm> Hello everyone, I am new to the list and currently looking for a timing standard for a VLBI application. After some searching I have decided to use two thunderbolts for a first attempt and now have some questions regarding the different available versions. There seem to be two types of Thunderbolts available on eBay right now. One is the "flat" board mount version without PSU and which requires a (+12V, -12V, +5V) power supply. The other is the "boxed" version with integrated PSU and which requires 24V. These are all old Thunderbolts with 8 channel GPS. The new Thunderbolt E with 12 channels obviously doesn't show up on eBay yet. (See [1] for comparison) My first question concerns the power supply. [2] shows that the output phase noise is significantly dependent on the PSU. How does the integrated PSU of the boxed version compare to the PSUs tested in [2]? Is it worth getting the boxed (somewhat more expensive on eBay) version just for the PSU? My second question is about the Thunderbolt's OCXO. [3] shows how to replace the built in OCXO with a more stable one and suggest that not all Thunderbolts have the same OCXOs. The bad OCXO in [3] appears to be from PIEZO and has S/N 1384. The better Thunderbolts apparently have "10811-class" OCXOs (like the replacement). Neither, unfortunately, match the labeling of the OCXOs seen on eBay (e.g. [4]). Is there an easy way to distinguish these different Thunderbolts? Thanks, Markus [1] http://www.febo.com/pipermail/time-nuts/2008-May/031370.html [2] http://www.leapsecond.com/pages/tbolt/noise.htm [3] http://www.thegleam.com/ke5fx/tbolt.htm [4] http://cgi.ebay.de/ws/eBayISAPI.dll?ViewItem&ssPageName=STRK:MEWAX:IT&item=170300146215 From sr75pro at free.fr Sat Feb 21 15:44:37 2009 From: sr75pro at free.fr (Sylvain RICHARD) Date: Sat, 21 Feb 2009 16:44:37 +0100 Subject: [time-nuts] Different Thunderbolt versions In-Reply-To: <19210986468.20090221152324@fastmail.fm> References: <19210986468.20090221152324@fastmail.fm> Message-ID: <49A02165.6040704@free.fr> Markus Kern wrote: > Hello everyone, > Hi, > I am new to the list and currently looking for a timing standard for a > VLBI application. After some searching I have decided to use two > thunderbolts for a first attempt and now have some questions regarding > the different available versions. > > There seem to be two types of Thunderbolts available on eBay right now. > One is the "flat" board mount version without PSU and which requires > a (+12V, -12V, +5V) power supply. The other is the "boxed" version > with integrated PSU and which requires 24V. These are all old > Thunderbolts with 8 channel GPS. The new Thunderbolt E with 12 > channels obviously doesn't show up on eBay yet. (See [1] for > comparison) > > My first question concerns the power supply. [2] shows that the output > phase noise is significantly dependent on the PSU. How does the > integrated PSU of the boxed version compare to the PSUs tested in [2]? > Is it worth getting the boxed (somewhat more expensive on eBay) version > just for the PSU? > I believe that the blue trace (#3) on the last figure from John Miles' page [3] is with the integrated 24 to -12, +12, +5 V PSU. So you are better off buying a cheap barebones Thunderbolt and a cheap linear PSU. > My second question is about the Thunderbolt's OCXO. [3] shows how to > replace the built in OCXO with a more stable one and suggest that not > all Thunderbolts have the same OCXOs. The bad OCXO in [3] appears to > be from PIEZO and has S/N 1384. The better Thunderbolts apparently > have "10811-class" OCXOs (like the replacement). Neither, > unfortunately, match the labeling of the OCXOs seen on eBay (e.g. [4]). > Is there an easy way to distinguish these different Thunderbolts? > > Thanks, > Markus > > [1] http://www.febo.com/pipermail/time-nuts/2008-May/031370.html > [2] http://www.leapsecond.com/pages/tbolt/noise.htm > [3] http://www.thegleam.com/ke5fx/tbolt.htm > [4] http://cgi.ebay.de/ws/eBayISAPI.dll?ViewItem&ssPageName=STRK:MEWAX:IT&item=170300146215 > > Good luck, -- Sylvain RICHARD From magnus at rubidium.dyndns.org Sat Feb 21 16:21:01 2009 From: magnus at rubidium.dyndns.org (Magnus Danielson) Date: Sat, 21 Feb 2009 17:21:01 +0100 Subject: [time-nuts] Lady Heather question serial error In-Reply-To: References: Message-ID: <49A029ED.1090706@rubidium.dyndns.org> David C. Partridge skrev: > Yes, Windows always seems to think a Thunderbolt is an MS mouse. Look back > in the archives about using /Fastdetect in boot.ini (BTW this doesn't work > with WinXP 64 - so I had to disable the serial mouse in the devices). The cure for that is very old - don't have the serial port connected during windows booting. It happend to me too, but recalling the old workaround and I had no problem what so ever. I'm considering a Linux port since I want to reduce my dependency on Windows at all times. Cheers, Magnus From jmiles at pop.net Sat Feb 21 16:28:22 2009 From: jmiles at pop.net (John Miles) Date: Sat, 21 Feb 2009 08:28:22 -0800 Subject: [time-nuts] Weird Lady Heather behavior observed...?? In-Reply-To: <499FA776.7090805@clanbaker.org> Message-ID: > HOWEVER-- if I close the LH display and open it back > up shortly thereafter, all my T-bolt running parameters > start fluctuating wildly and going off-scale. Even the > yellow temp line goes off scale and PPS OADEV and OSC OADEV > both drop back to a high e-12. At what tau values, though? The idea behind ADEV is that it goes lower the longer you run. When you first launch the program, you only have a short-term history to observe. > > It takes about 90 minutes for everything to settle back down, > after which everything runs tightly for many hours until > the next time I minimize or close and re-open the LH display. The graph-recentering behavior can be extremely confusing because it gives the appearance of jumps that aren't really happening. Try running with /m=4 to compress the graph height 4x and see if you still observe unexpected initial behavior. I haven't seen anything happening myself that I couldn't eventually put down to recentering. -- john, KE5FX From cfharris at erols.com Sat Feb 21 16:42:33 2009 From: cfharris at erols.com (Chuck Harris) Date: Sat, 21 Feb 2009 11:42:33 -0500 Subject: [time-nuts] Lady Heather question serial error In-Reply-To: <49A029ED.1090706@rubidium.dyndns.org> References: <49A029ED.1090706@rubidium.dyndns.org> Message-ID: <49A02EF9.9070008@erols.com> Magnus Danielson wrote: > I'm considering a Linux port since I want to reduce my dependency on > Windows at all times. If you do, welcome! Please take a trip to: http://counter.li.org/ And add to the count of linux users. Because there is no organization that takes money from the linux users, there is no other way of knowing how many of us are out there. Estimates put us at around 29 million. -Chuck Harris From jmiles at pop.net Sat Feb 21 16:46:31 2009 From: jmiles at pop.net (John Miles) Date: Sat, 21 Feb 2009 08:46:31 -0800 Subject: [time-nuts] Different Thunderbolt versions In-Reply-To: <49A02165.6040704@free.fr> Message-ID: > I believe that the blue trace (#3) on the last figure from John Miles' > page [3] is with the integrated 24 to -12, +12, +5 V PSU. So you are > better off buying a cheap barebones Thunderbolt and a cheap linear PSU. That's my thinking, assuming the currently-available surplus ones all have good-quality rocks like the TAPR ones did. I don't have any basis to speculate on that, or to assume that all of the older Thundebolts with PIEZO OCXOs are as unstable (and spur-ridden) as mine was. My original unit had been in continuous service for over 5 years at the time I measured it. Until the TAPR group buy happened, it was the only one I'd ever used. -- john, KE5FX From francesco.messineo at gmail.com Sat Feb 21 17:10:18 2009 From: francesco.messineo at gmail.com (francesco messineo) Date: Sat, 21 Feb 2009 18:10:18 +0100 Subject: [time-nuts] Lady Heather question serial error In-Reply-To: <49A029ED.1090706@rubidium.dyndns.org> References: <49A029ED.1090706@rubidium.dyndns.org> Message-ID: On 2/21/09, Magnus Danielson wrote: > I'm considering a Linux port since I want to reduce my dependency on > Windows at all times. > before knowing about the existence of lady heather I had started coding a very simple Linux/Unix program that would work much like tboltmon.exe (ncurses based). Then I tried both Dos (under dosemu) and windows (under wine) versions of lady heather and they work just fine, so I abandoned (for now) that effort. I don't have any dependance on windows, I actually don't have any windows box around. Regards Frank IZ8DWF From cdelect at juno.com Sat Feb 21 20:56:38 2009 From: cdelect at juno.com (Corby Dawson) Date: Sat, 21 Feb 2009 12:56:38 -0800 Subject: [time-nuts] FTS 1200 and 9110 oscillator short term stabilities Message-ID: <20090221.125639.3004.2.cdelect@juno.com> Hi, Here are some results of the FTS 1200 and FTS 9110 oscillators I have recently tested. 1SEC 10SEC 100SEC 1200 1 7.14-13 1.48-12 1.92-12 1200 2 3.29-13 1.80-13 3.28-13 1 and 2 above previously tested 1200 1B 4.00-13 5.50-13 1.00-12 1200 2B 4.29-13 3.24-13 4.70-13 1200 5B 9.27-13 8.14-13 1.29-12 9110 1.44-12 4.96-13 3.49-13 9110 B 7.59-13 1.10-12 3.36-13 They all have four 5Mhz outputs, 0 to +10VDC control voltage, no mechanical trim, and run on +24VDC. Oscillator section is built into a vacuum dewar. The B designates units I am testing for a friend, I believe he wants to keep the best one and sell the rest. If that turns out to be the case I will post a heads up when I list them on eBay for him. Cheers! Corby Dawson in California ____________________________________________________________ Find home loans here. Complete financial solutions. http://thirdpartyoffers.juno.com/TGL2141/fc/BLSrjpTIsOd0BmYzrdgGfUUEmBc2ujqDwJ70Qn0uVfMvOdRSQT9DOTfOKiF/ From bruce.griffiths at xtra.co.nz Sat Feb 21 21:22:09 2009 From: bruce.griffiths at xtra.co.nz (Bruce Griffiths) Date: Sun, 22 Feb 2009 10:22:09 +1300 Subject: [time-nuts] Different Thunderbolt versions In-Reply-To: <19210986468.20090221152324@fastmail.fm> References: <19210986468.20090221152324@fastmail.fm> Message-ID: <49A07081.1040204@xtra.co.nz> Markus Kern wrote: > Hello everyone, > > I am new to the list and currently looking for a timing standard for a > VLBI application. After some searching I have decided to use two > thunderbolts for a first attempt and now have some questions regarding > the different available versions. > > There seem to be two types of Thunderbolts available on eBay right now. > One is the "flat" board mount version without PSU and which requires > a (+12V, -12V, +5V) power supply. The other is the "boxed" version > with integrated PSU and which requires 24V. These are all old > Thunderbolts with 8 channel GPS. The new Thunderbolt E with 12 > channels obviously doesn't show up on eBay yet. (See [1] for > comparison) > > My first question concerns the power supply. [2] shows that the output > phase noise is significantly dependent on the PSU. How does the > integrated PSU of the boxed version compare to the PSUs tested in [2]? > Is it worth getting the boxed (somewhat more expensive on eBay) version > just for the PSU? > > My second question is about the Thunderbolt's OCXO. [3] shows how to > replace the built in OCXO with a more stable one and suggest that not > all Thunderbolts have the same OCXOs. The bad OCXO in [3] appears to > be from PIEZO and has S/N 1384. The better Thunderbolts apparently > have "10811-class" OCXOs (like the replacement). Neither, > unfortunately, match the labeling of the OCXOs seen on eBay (e.g. [4]). > Is there an easy way to distinguish these different Thunderbolts? > > Thanks, > Markus > > [1] http://www.febo.com/pipermail/time-nuts/2008-May/031370.html > [2] http://www.leapsecond.com/pages/tbolt/noise.htm > [3] http://www.thegleam.com/ke5fx/tbolt.htm > [4] http://cgi.ebay.de/ws/eBayISAPI.dll?ViewItem&ssPageName=STRK:MEWAX:IT&item=170300146215 > > Markus The Thunderbolt GPSDO is too noisy to be particularly useful for generating the sampling clock for VLBI, unless your observation frequency is relatively low. To achieve useful results the carrier phase error between any pair of VLBI sites needs to be less than 50 degrees or so over the signal integration period. For observations at 1420MHz, and short integration times (~ 100sec) the phase error needs to be 100ps or less. For longer integration times (10,000 sec is desirable) and higher frequencies the timing constraints are far more stringent. Whilst clock modeling can be used to improve performance it can only correct systematic effects. Bruce From bjones0 at mindspring.com Sun Feb 22 00:57:34 2009 From: bjones0 at mindspring.com (bjones0 at mindspring.com) Date: Sat, 21 Feb 2009 19:57:34 -0500 Subject: [time-nuts] Thunderbolt serial port In-Reply-To: References: <6F62D5A392B04D15853529C4BEDFCC85@athlon> <4848727E.60108@smiffytech.com> <6.2.5.6.2.20080605163443.02d5c258@jpl.nasa.gov> <1235099026.27417.13.camel@skuld> <499E24FD.2030608@pacific.net> Message-ID: <1235264254.3447.16.camel@skuld> I got around to tearing into the Thunderbolt this afternoon and after testing the serial port and traces in every way I could I found the problem... The power supply (which got with it in the TAPR buy) went south apparently. I got a 24v difference between pins 1 & 6 but when you checked them against the common the +12v line was @ +6v and the -12v line was -18v. The 5v line did not even register. I pulled a computer power supply and modified it slightly and once I was providing proper voltage it works like a champ. I never thought to suspect the power supply because when I added the plug it tested (voltages) okay and once I hooked up the Thunderbolt the first time it exhibited the same behavior (no serial and the 9.9999978 Mhz on the 10Mhz output). Is there anything wrong with running the Thunderbolt from a Computer PSU? Thanks for all the help! Brian And just passing some unrelated info along... While looking for some weather related info I ran across someone who was selling new Trimble active GPS antennas w/ a nice long cable (PN 40767-40) for $15 and now that my Thunderbolt works I was able to test it and it seems to work fine (even indoors!). If anyone want to contact him: Eric Wolf eric.n5ebw at gmail.com . Also he emailed me this after I told him it was working: "I wanted to pass along something that I have had in my inventory for about 6 months but cannot find the use for. It's an AVCOM Microwave Distribution Module that basically splits a 1575 MHz GPS signal, and amplifies it to 8 different antennas. The exact model is MDM-1575-8 and the description reads: "The MDM-1575-8 is an 8-way, amplified, 1575 MHz power divider. The INPUT port has +5VDC on connector and the 8 OUTPUT ports are DC Blocked." " > -----Original Message----- > From: time-nuts-bounces at febo.com [mailto:time-nuts-bounces at febo.com] On > Behalf Of Brooke Clarke > Sent: Friday, 20 February 2009 2:35 PM > To: Discussion of precise time and frequency measurement > Subject: Re: [time-nuts] Thunderbolt serial port > > Hi Brian: > > There have been a small number of problems related to poor solder joints on > U9 > (RS-232 chip) or the serial connector. see: > http://www.prc68.com/I/ThunderBolt.shtml#Maint > > Have Fun, > > Brooke Clarke > http://www.prc68.com > > bjones0 at mindspring.com wrote: > > I'm tired of banging my head against the wall so I think it is time to > > ask.. Is there a way to test if the serial port on the Thunderbolt > > works? I know the Thunderbolt itself powers up and works (I get a > > 9.9999978 Mhz signal out of the 10Mhz output when it is on) but I have > > been unable to talk/listen to it in Windows (tboltmon or Lady Heather) > > or in Linux (even looking at the raw serial port state). I stuck my > > breakout box on the line but I get no lights on the Thunderbolt side. > > > > Thanks, > > Brian > > > > > > _______________________________________________ > > time-nuts mailing list -- time-nuts at febo.com > > To unsubscribe, go to > https://www.febo.com/cgi-bin/mailman/listinfo/time-nuts > > and follow the instructions there. > > > > > > > _______________________________________________ > time-nuts mailing list -- time-nuts at febo.com > To unsubscribe, go to > https://www.febo.com/cgi-bin/mailman/listinfo/time-nuts > and follow the instructions there. > > > > _______________________________________________ > time-nuts mailing list -- time-nuts at febo.com > To unsubscribe, go to https://www.febo.com/cgi-bin/mailman/listinfo/time-nuts > and follow the instructions there. From tvb at LeapSecond.com Sun Feb 22 01:05:59 2009 From: tvb at LeapSecond.com (Tom Van Baak) Date: Sat, 21 Feb 2009 17:05:59 -0800 Subject: [time-nuts] Different Thunderbolt versions References: <19210986468.20090221152324@fastmail.fm> <49A07081.1040204@xtra.co.nz> Message-ID: <6C8E7ADA880040799C1477C5C1C32AD5@pc52> > For observations at 1420MHz, and short integration times (~ 100sec) the > phase error needs to be 100ps or less. > For longer integration times (10,000 sec is desirable) and higher > frequencies the timing constraints are far more stringent. > Whilst clock modeling can be used to improve performance it can only > correct systematic effects. Bruce, Approximately what range of tau and ADEV(tau) would this VLBI requirement correspond to? A couple of the TBolts I've tested are at or just under 1e-12 at 1 second and stay almost flat out to 100 or even 1000 seconds. /tvb From masondg44 at comcast.net Sun Feb 22 01:13:09 2009 From: masondg44 at comcast.net (Dave M) Date: Sat, 21 Feb 2009 20:13:09 -0500 Subject: [time-nuts] Different Thunderbolt versions References: Message-ID: <32C2665C055244E08122B8081642ECC3@DELL2350> I currently have an HP Z3801A, which seems to be running well. I've been looking at the T-Bolts on Ebay for a while, but have been holding off buying. This post just gave me an opportunity to ask some questions. I've seen references to the TAPR group buy in previous posts. It obviously happened before I found this group, but am interested in the details. Things like: When did the buy happen? Which model T-Bolt was bought? What was the price? Was PSU included? Are any units still available? Is another buy planned for the near future? Thanks for the info. Dave M MasonDG44 at comcast dot net (Just substitute the appropriate characters in the address) Life is like a roll of toilet paper; the closer it gets to the end, the faster it goes. ------------------------------ Message: 6 Date: Sat, 21 Feb 2009 08:46:31 -0800 From: "John Miles" Subject: Re: [time-nuts] Different Thunderbolt versions To: "Discussion of precise time and frequency measurement" Message-ID: Content-Type: text/plain; charset="us-ascii" > I believe that the blue trace (#3) on the last figure from John Miles' > page [3] is with the integrated 24 to -12, +12, +5 V PSU. So you are > better off buying a cheap barebones Thunderbolt and a cheap linear PSU. That's my thinking, assuming the currently-available surplus ones all have good-quality rocks like the TAPR ones did. I don't have any basis to speculate on that, or to assume that all of the older Thundebolts with PIEZO OCXOs are as unstable (and spur-ridden) as mine was. My original unit had been in continuous service for over 5 years at the time I measured it. Until the TAPR group buy happened, it was the only one I'd ever used. -- john, KE5FX From bruce.griffiths at xtra.co.nz Sun Feb 22 01:20:01 2009 From: bruce.griffiths at xtra.co.nz (Bruce Griffiths) Date: Sun, 22 Feb 2009 14:20:01 +1300 Subject: [time-nuts] Different Thunderbolt versions In-Reply-To: <6C8E7ADA880040799C1477C5C1C32AD5@pc52> References: <19210986468.20090221152324@fastmail.fm> <49A07081.1040204@xtra.co.nz> <6C8E7ADA880040799C1477C5C1C32AD5@pc52> Message-ID: <49A0A841.8000702@xtra.co.nz> Tom Van Baak wrote: >> For observations at 1420MHz, and short integration times (~ 100sec) the >> phase error needs to be 100ps or less. >> For longer integration times (10,000 sec is desirable) and higher >> frequencies the timing constraints are far more stringent. >> Whilst clock modeling can be used to improve performance it can only >> correct systematic effects. >> > > Bruce, > > Approximately what range of tau and ADEV(tau) would this > VLBI requirement correspond to? A couple of the TBolts I've > tested are at or just under 1e-12 at 1 second and stay almost > flat out to 100 or even 1000 seconds. > > /tvb > > > Tom Really should use something like TDEV but roughly (within a factor of 2 or so at 1420MHz): ADEV < 1E-10 or better @ tau = 1sec for 1 second integration time. (integration time to short for anything but very bright objects) ADEV < 1E-11 or better @ tau = 1sec for 10 second integration time. (integration time to short for anything but very bright objects) ADEV < 1E-12 or better @ tau = 1sec for 100 second integration time. (integration time somewhat marginal for most objects) ADEV < 1E-13 or better @ tau = 1sec for 1000 second integration time. ADEV < 1E-14 or better @ tau = 1sec for 10,000 second integration time. Also need to take into account that a similar LO would presumably be used at each antenna. For integration times of up to a few hundred seconds VLBI has been done with a hydrogen maser at one antenna and a PRS10A at the other. However the performance isnt Earth shattering and there are some issues with long term drift and transport induced frequency shifts of the PRS10A. For very short integration periods a good OCXO is adequate (at 1420MHz). When an integration time of 1000sec or more is required the LO requirents are more stringent. Bruce From james.p.lux at jpl.nasa.gov Sun Feb 22 01:45:54 2009 From: james.p.lux at jpl.nasa.gov (Lux, James P) Date: Sat, 21 Feb 2009 17:45:54 -0800 Subject: [time-nuts] Different Thunderbolt versions In-Reply-To: <49A0A841.8000702@xtra.co.nz> Message-ID: On 2/21/09 5:20 PM, "Bruce Griffiths" wrote: > Tom Van Baak wrote: >>> For observations at 1420MHz, and short integration times (~ 100sec) the >>> phase error needs to be 100ps or less. >>> For longer integration times (10,000 sec is desirable) and higher >>> frequencies the timing constraints are far more stringent. >>> Whilst clock modeling can be used to improve performance it can only >>> correct systematic effects. >>> >> >> Bruce, >> >> Approximately what range of tau and ADEV(tau) would this >> VLBI requirement correspond to? A couple of the TBolts I've >> tested are at or just under 1e-12 at 1 second and stay almost >> flat out to 100 or even 1000 seconds. >> >> /tvb >> >> >> > Tom > > Really should use something like TDEV but roughly (within a factor of 2 > or so at 1420MHz): > > ADEV < 1E-10 or better @ tau = 1sec for 1 second integration time. > (integration time to short for anything but very bright objects) > ADEV < 1E-11 or better @ tau = 1sec for 10 second integration time. > (integration time to short for anything but very bright objects) > ADEV < 1E-12 or better @ tau = 1sec for 100 second integration time. > (integration time somewhat marginal for most objects) > ADEV < 1E-13 or better @ tau = 1sec for 1000 second integration time. > ADEV < 1E-14 or better @ tau = 1sec for 10,000 second integration time. > > > > Also need to take into account that a similar LO would presumably be > used at each antenna. > > For integration times of up to a few hundred seconds VLBI has been done > with a hydrogen maser at one antenna and a PRS10A at the other. > However the performance isnt Earth shattering and there are some issues > with long term drift and transport induced frequency shifts of the PRS10A. > In the Deep Space Network, we do Delta DOR, which is a form of VLBI, using a convenient quasar as the common view timing reference. (and, of course, they have hydrogen masers at the stations as well) http://trs-new.jpl.nasa.gov/dspace/handle/2014/37674 From stanley_reynolds at yahoo.com Sun Feb 22 01:48:06 2009 From: stanley_reynolds at yahoo.com (Stanley Reynolds) Date: Sat, 21 Feb 2009 17:48:06 -0800 (PST) Subject: [time-nuts] Different Thunderbolt versions References: <32C2665C055244E08122B8081642ECC3@DELL2350> Message-ID: <478447.45660.qm@web30307.mail.mud.yahoo.com> ________________________________ From: Dave M To: time-nuts at febo.com Sent: Saturday, February 21, 2009 7:13:09 PM Subject: Re: [time-nuts] Different Thunderbolt versions I currently have an HP Z3801A, which seems to be running well. I've been looking at the T-Bolts on Ebay for a while, but have been holding off buying. This post just gave me an opportunity to ask some questions. I've seen references to the TAPR group buy in previous posts. It obviously happened before I found this group, but am interested in the details. Things like: When did the buy happen? Which model T-Bolt was bought? What was the price? Was PSU included? Are any units still available? Is another buy planned for the near future? Probably best to contact Tom Van Baak direct at tvb at LeapSecond.com C/P of some past posts: [time-nuts] Thunderbolt Group Buy, part 1... Tom Van Baak Tom Van Baak ...AddMonday, May 5, 2008 4:55:21 AM To:Discussion of precise time and frequency measurement BACKGROUND: About a year ago, when surplus Thunderbolt GPS receivers first appeared on eBay, several of us pursued the idea of a time-nuts group buy. We kept it fairly quiet at the time while we probed the source, verified actual performance, and negotiated quantity & pricing. Recall that the first Thunderbolts on eBay fetched prices on the order of $300 (I know, I bought the first one). Later the price floated down to the $200 level. Even this price seemed like a good deal for such a name-brand GPSDO. These almost new unused Thunderbolts are quite nice; with performance much better than shown on Trimble's web site. Last year about 50 of you put in "pre-orders" for a total of about 80 units. With that leverage we negotiated a deal with the original California supplier. THE DEAL: I'm very pleased to announce that TAPR is going to stock these Thunderbolts for us and handle ordering and payment issues. Note that TAPR is non-profit and our time-nuts list has a non-commercial, amateur and hobby focus. So you will get these Thunderbolts at cost. The group price is just $115 each. You get a Trimble Thunderbolt AND a nice universal AC input +5/+12/-12 volt DC power supply AND the mating 6-pin DC Molex power connector AND a BNC-F antenna adapter. I will advise you when the official TAPR.org Thunderbolt ordering web page is ready. For an example of what you will receive, see: http://www.leapsecond.com/pages/tapr-tbolt/ ORDERING: 1) Those of you who "pre-ordered" last year will get email this week with instructions on how to submit payment for the Thunderbolt(s) you previously said you would like to buy. 2) Those of you newcomers to Time-Nuts should expect to wait until the first batch has been shipped and TAPR is ready for you. There should be enough for everyone; so don't panic. Flat rate shipping will be about $9 (US), $23 (Canada), or $37 (international). Same shipping for 1, 2, or 3 units per box, I'm told. I hope you agree the wait was worth it. Whether it's your first Thunderbolt or not -- this is an amazing deal for any of us. A lot of work went into making it happen. /tvb [time-nuts] Thunderbolt Group Buy, part 3... Tom Van Baak Tom Van Baak ...AddTo: Discussion of precise time and frequency measurement Discussion of precise time and frequency measurement ----- Forwarded Message ---- From: Stanley Reynolds To: Discussion of precise time and frequency measurement Sent: Saturday, February 21, 2009 7:48:06 PM Subject: Re: [time-nuts] Different Thunderbolt versions ________________________________ From: Dave M To: time-nuts at febo.com Sent: Saturday, February 21, 2009 7:13:09 PM Subject: Re: [time-nuts] Different Thunderbolt versions I currently have an HP Z3801A, which seems to be running well. I've been looking at the T-Bolts on Ebay for a while, but have been holding off buying. This post just gave me an opportunity to ask some questions. I've seen references to the TAPR group buy in previous posts. It obviously happened before I found this group, but am interested in the details. Things like: When did the buy happen? Which model T-Bolt was bought? What was the price? Was PSU included? Are any units still available? Is another buy planned for the near future? Probably best to contact Tom Van Baak direct at tvb at LeapSecond.com C/P of some past posts: [time-nuts] Thunderbolt Group Buy, part 1... Tom Van Baak Tom Van Baak ...AddMonday, May 5, 2008 4:55:21 AM To:Discussion of precise time and frequency measurement BACKGROUND: About a year ago, when surplus Thunderbolt GPS receivers first appeared on eBay, several of us pursued the idea of a time-nuts group buy. We kept it fairly quiet at the time while we probed the source, verified actual performance, and negotiated quantity & pricing. Recall that the first Thunderbolts on eBay fetched prices on the order of $300 (I know, I bought the first one). Later the price floated down to the $200 level. Even this price seemed like a good deal for such a name-brand GPSDO. These almost new unused Thunderbolts are quite nice; with performance much better than shown on Trimble's web site. Last year about 50 of you put in "pre-orders" for a total of about 80 units. With that leverage we negotiated a deal with the original California supplier. THE DEAL: I'm very pleased to announce that TAPR is going to stock these Thunderbolts for us and handle ordering and payment issues. Note that TAPR is non-profit and our time-nuts list has a non-commercial, amateur and hobby focus. So you will get these Thunderbolts at cost. The group price is just $115 each. You get a Trimble Thunderbolt AND a nice universal AC input +5/+12/-12 volt DC power supply AND the mating 6-pin DC Molex power connector AND a BNC-F antenna adapter. I will advise you when the official TAPR.org Thunderbolt ordering web page is ready. For an example of what you will receive, see: http://www.leapsecond.com/pages/tapr-tbolt/ ORDERING: 1) Those of you who "pre-ordered" last year will get email this week with instructions on how to submit payment for the Thunderbolt(s) you previously said you would like to buy. 2) Those of you newcomers to Time-Nuts should expect to wait until the first batch has been shipped and TAPR is ready for you. There should be enough for everyone; so don't panic. Flat rate shipping will be about $9 (US), $23 (Canada), or $37 (international). Same shipping for 1, 2, or 3 units per box, I'm told. I hope you agree the wait was worth it. Whether it's your first Thunderbolt or not -- this is an amazing deal for any of us. A lot of work went into making it happen. /tvb [time-nuts] Thunderbolt Group Buy, part 3... Tom Van Baak Tom Van Baak ...AddTo: Discussion of precise time and frequency measurement Discussion of precise time and frequency measurement <32C2665C055244E08122B8081642ECC3@DELL2350> Message-ID: >I currently have an HP Z3801A, which seems to be running > well. I've been looking at the T-Bolts on Ebay for a while, but > have been holding off buying. This post just gave me an > opportunity to ask some questions. > > I've seen references to the TAPR group buy in previous posts. > It obviously happened before I found this group, but am interested > in the details. > Things like: > When did the buy happen? > Which model T-Bolt was bought? > What was the price? > Was PSU included? > Are any units still available? > Is another buy planned for the near future? > > Thanks for the info. > Hi Dave, Yes, there have been 3 or 4 large batches of TBolts sold through TAPR over the past year or two. The price ranged from about 95 to 125. Most were sold with power supply; some without (when the source of cheap switching supplies ran out). For more info see: http://www.leapsecond.com/tbolt-faq.htm http://www.leapsecond.com/pages/tapr-tbolt/ http://www.leapsecond.com/pages/tbolt/noise.htm Yes, there will be another group buy or two. However, this time I'm testing all of them first and then announcing the sale (expect mid-March) so that no one has that long wait between ordering and paying/receiving. So hang on a bit longer and I'll send the details next month. Note a number of people have asked privately (you know who you are) and those requests will be handled first. > Dave M > MasonDG44 at comcast dot net (Just substitute the appropriate characters in the address) > > Life is like a roll of toilet paper; the closer it gets to the end, the faster it goes. This is an interesting T&F problem, actually. It depends what you mean by "it". If "it goes" means the paper itself -- if you pull at a constant linear rate then it goes neither slower nor faster. It's constant and then suddenly ends, that's all. If "it goes" means the rotation (frequency) of the roll -- if you pull the paper at a constant linear rate then, yes, a predictable drift in frequency occurs as the radius of the roll decreases. For an interesting mental exercise (perhaps while you're sitting down near said roll of TP) see if the frequency drift is linear or log or exponential. For extra credit, when the roll is slightly bent (as often happens) rather than perfectly circular see what effect that modulation has on the short-term ADEV of the frequency. Alternatively, if you unroll the TP by attaching a stone to the end and letting it drop over a cliff then you have a really interesting physics problem considering the acceleration of gravity, varying mass, inertia, and radius of the roll, nonlinear friction in both air and roller, etc. /tvb From wb6bnq at cox.net Sun Feb 22 03:00:23 2009 From: wb6bnq at cox.net (WB6BNQ) Date: Sat, 21 Feb 2009 19:00:23 -0800 Subject: [time-nuts] Toilet paper speed References: <32C2665C055244E08122B8081642ECC3@DELL2350> Message-ID: <49A0BFC7.324F8FAD@cox.net> Tom, Don't forget to save a few squares ! Bill....WB6BNQ Tom Van Baak wrote: > This is an interesting T&F problem, actually. It depends what > you mean by "it". If "it goes" means the paper itself -- if you > pull at a constant linear rate then it goes neither slower nor > faster. It's constant and then suddenly ends, that's all. > > If "it goes" means the rotation (frequency) of the roll -- if you > pull the paper at a constant linear rate then, yes, a predictable > drift in frequency occurs as the radius of the roll decreases. > > For an interesting mental exercise (perhaps while you're sitting > down near said roll of TP) see if the frequency drift is linear or > log or exponential. For extra credit, when the roll is slightly bent > (as often happens) rather than perfectly circular see what effect > that modulation has on the short-term ADEV of the frequency. > > Alternatively, if you unroll the TP by attaching a stone to the end > and letting it drop over a cliff then you have a really interesting > physics problem considering the acceleration of gravity, varying > mass, inertia, and radius of the roll, nonlinear friction in both air > and roller, etc. > > /tvb > > _______________________________________________ > time-nuts mailing list -- time-nuts at febo.com > To unsubscribe, go to https://www.febo.com/cgi-bin/mailman/listinfo/time-nuts > and follow the instructions there. From mikes at flatsurface.com Sun Feb 22 03:17:18 2009 From: mikes at flatsurface.com (Mike S) Date: Sat, 21 Feb 2009 22:17:18 -0500 Subject: [time-nuts] Different Thunderbolt versions, group buy In-Reply-To: References: <32C2665C055244E08122B8081642ECC3@DELL2350> Message-ID: <20090222031720.8C3941165C1@hamburg.alientech.net> At 09:51 PM 2/21/2009, Tom Van Baak wrote... >So hang on a bit longer and I'll send the details next month. > >Note a number of people have asked privately (you know who >you are) and those requests will be handled first. Since you've opened a can-o-worms here, I'll put my name in the hat for one of the next batch. From jgd at johngsbbq.com Sun Feb 22 04:25:23 2009 From: jgd at johngsbbq.com (John DeArmond) Date: Sat, 21 Feb 2009 23:25:23 -0500 Subject: [time-nuts] Lady Heather question serial error In-Reply-To: <49A02EF9.9070008@erols.com> References: <49A029ED.1090706@rubidium.dyndns.org> <49A02EF9.9070008@erols.com> Message-ID: <1235276723.7596.44.camel@neonjohn> On Sat, 2009-02-21 at 11:42 -0500, Chuck Harris wrote: > Magnus Danielson wrote: > > > I'm considering a Linux port since I want to reduce my dependency on > > Windows at all times. > > If you do, welcome! Please take a trip to: > > http://counter.li.org/ You might also be interested in my account of going cold-turkey in converting my laptop to Ubuntu about 2 months ago. I am almost windows-free. there are still a few things that I must have XP for so I installed VMware and have XP booting as a task under Ubuntu. I show you how to do it for free. http://www.johndearmond.com/category/computing/page/3/ BTW, a couple of weeks ago I read a Gardner Group report that puts the number of desktop users (not counting the servers that run the net) at about the same number as Mac users. 8-10% of the market. That's wonderful news. John -- John DeArmond Tellico Plains, Occupied TN http://www.neon-john.com <-- email from here http://www.johndearmond.com <-- Best damned Blog on the net From dave.g0dja at tiscali.co.uk Sun Feb 22 09:01:36 2009 From: dave.g0dja at tiscali.co.uk (Dave Ackrill) Date: Sun, 22 Feb 2009 09:01:36 +0000 Subject: [time-nuts] (OT) Frequency and duration of roll out In-Reply-To: References: <32C2665C055244E08122B8081642ECC3@DELL2350> Message-ID: <49A11470.6060002@tiscali.co.uk> > Alternatively, if you unroll the TP by attaching a stone to the end > and letting it drop over a cliff then you have a really interesting > physics problem considering the acceleration of gravity, varying > mass, inertia, and radius of the roll, nonlinear friction in both air > and roller, etc. I remember that, when I was a trainee, we (the engineering trainees) spent some time learning some of the skills that the people that would be working for us had to master. One such was jointing 3 phase cables together. If you wanted to cheese off the instructor, you went to the roll of thin cotton tape, hanging on a roll on the wall, and pulled hard. The weight of the tape coming off the roll was enough to spin the disk, thus bringing more tape off, which was heavy enough to spin the roll some more, so more tape came off, and so on. Thinking back, I don't remember timing the spin rate, more like nipping back to our places so as to be able to deny having carried out the act. :-) Dave (G0DJA) From phk at phk.freebsd.dk Sun Feb 22 09:09:16 2009 From: phk at phk.freebsd.dk (Poul-Henning Kamp) Date: Sun, 22 Feb 2009 09:09:16 +0000 Subject: [time-nuts] (OT) Frequency and duration of roll out In-Reply-To: Your message of "Sun, 22 Feb 2009 09:01:36 GMT." <49A11470.6060002@tiscali.co.uk> Message-ID: <11721.1235293756@critter.freebsd.dk> In message <49A11470.6060002 at tiscali.co.uk>, Dave Ackrill writes: >If you wanted to cheese off the instructor, you went to the roll of thin >cotton tape, hanging on a roll on the wall, and pulled hard. The weight >of the tape coming off the roll was enough to spin the disk, thus >bringing more tape off, which was heavy enough to spin the roll some >more, so more tape came off, and so on. This was a well know issue when manipulating the roll while splicing audiotapes. The funny thing was, 2" tapes are really strong. We tied an old studio worktape to the trailer-hook of a friends car when he got married. The tape unrolled flawlessly, the roll clattered wonderfully on the cobbles, and when they turned out of the drive-way it snagged a lamppost and brought the car to a halt until the groom went out and cut the tape :-) -- Poul-Henning Kamp | UNIX since Zilog Zeus 3.20 phk at FreeBSD.ORG | TCP/IP since RFC 956 FreeBSD committer | BSD since 4.3-tahoe Never attribute to malice what can adequately be explained by incompetence. From mkern at fastmail.fm Sun Feb 22 16:32:05 2009 From: mkern at fastmail.fm (Markus Kern) Date: Sun, 22 Feb 2009 17:32:05 +0100 Subject: [time-nuts] Different Thunderbolt versions In-Reply-To: <49A0A841.8000702@xtra.co.nz> References: <19210986468.20090221152324@fastmail.fm> <49A07081.1040204@xtra.co.nz> <6C8E7ADA880040799C1477C5C1C32AD5@pc52> <49A0A841.8000702@xtra.co.nz> Message-ID: <12619539593.20090222173205@fastmail.fm> Thanks everyone for all the input. I actually want to work at frequencies below 50Mhz where a 50 degree phase error corresponds to about 3ns. So for 100 seconds integration time an ADEV of 1E-11 or better at tau = 1s should be fine. I want to keep the system as simple and cheap as possible which is why I want to give the Thunderbolt a try. Also at these low frequencies the required accuracy approaches that claimed by some GPS receivers (4ns for the M12+ ?) so might it be possible to build a system which is accurate for arbitrary integration times? Markus On 22.02.2009, 02:20 Bruce Griffiths wrote: > Tom Van Baak wrote: >>> For observations at 1420MHz, and short integration times (~ 100sec) the >>> phase error needs to be 100ps or less. >>> For longer integration times (10,000 sec is desirable) and higher >>> frequencies the timing constraints are far more stringent. >>> Whilst clock modeling can be used to improve performance it can only >>> correct systematic effects. >>> >> >> Bruce, >> >> Approximately what range of tau and ADEV(tau) would this >> VLBI requirement correspond to? A couple of the TBolts I've >> tested are at or just under 1e-12 at 1 second and stay almost >> flat out to 100 or even 1000 seconds. >> >> /tvb >> >> >> > Tom > Really should use something like TDEV but roughly (within a factor of 2 > or so at 1420MHz): > ADEV < 1E-10 or better @ tau = 1sec for 1 second integration time. > (integration time to short for anything but very bright objects) > ADEV < 1E-11 or better @ tau = 1sec for 10 second integration time. > (integration time to short for anything but very bright objects) > ADEV < 1E-12 or better @ tau = 1sec for 100 second integration time. > (integration time somewhat marginal for most objects) > ADEV < 1E-13 or better @ tau = 1sec for 1000 second integration time. > ADEV < 1E-14 or better @ tau = 1sec for 10,000 second integration time. > Also need to take into account that a similar LO would presumably be > used at each antenna. > For integration times of up to a few hundred seconds VLBI has been done > with a hydrogen maser at one antenna and a PRS10A at the other. > However the performance isnt Earth shattering and there are some issues > with long term drift and transport induced frequency shifts of the PRS10A. > For very short integration periods a good OCXO is adequate (at 1420MHz). > When an integration time of 1000sec or more is required the LO > requirents are more stringent. > Bruce From imp at bsdimp.com Sun Feb 22 16:58:09 2009 From: imp at bsdimp.com (M. Warner Losh) Date: Sun, 22 Feb 2009 09:58:09 -0700 (MST) Subject: [time-nuts] (OT) Frequency and duration of roll out In-Reply-To: <11721.1235293756@critter.freebsd.dk> References: <49A11470.6060002@tiscali.co.uk> <11721.1235293756@critter.freebsd.dk> Message-ID: <20090222.095809.1286498905.imp@bsdimp.com> In message: <11721.1235293756 at critter.freebsd.dk> "Poul-Henning Kamp" writes: : In message <49A11470.6060002 at tiscali.co.uk>, Dave Ackrill writes: : : >If you wanted to cheese off the instructor, you went to the roll of thin : >cotton tape, hanging on a roll on the wall, and pulled hard. The weight : >of the tape coming off the roll was enough to spin the disk, thus : >bringing more tape off, which was heavy enough to spin the roll some : >more, so more tape came off, and so on. : : This was a well know issue when manipulating the roll while splicing : audiotapes. : : The funny thing was, 2" tapes are really strong. : : We tied an old studio worktape to the trailer-hook of a friends car : when he got married. The tape unrolled flawlessly, the roll : clattered wonderfully on the cobbles, and when they turned out of : the drive-way it snagged a lamppost and brought the car to a halt : until the groom went out and cut the tape :-) Try doing that with CDs :-) Warner From hmurray at megapathdsl.net Sun Feb 22 17:33:16 2009 From: hmurray at megapathdsl.net (Hal Murray) Date: Sun, 22 Feb 2009 09:33:16 -0800 Subject: [time-nuts] Different Thunderbolt versions In-Reply-To: Message from Markus Kern of "Sun, 22 Feb 2009 17:32:05 +0100." <12619539593.20090222173205@fastmail.fm> Message-ID: <20090222173317.BBA46BCDC@ip-64-139-1-69.sjc.megapath.net> > I actually want to work at frequencies below 50Mhz where a 50 degree > phase error corresponds to about 3ns. What is interesting to listen to below 50 MHz? Are there any handy strong signals you can use for debugging/calibration? > Also at these low frequencies the required accuracy approaches that > claimed by some GPS receivers (4ns for the M12+ ?) so might it be > possible to build a system which is accurate for arbitrary integration > times? If you haven't seen it yet, this is wonderful background: http://www.gpstime.com/files/PTTI/PTTI_2006.pdf In particular, be sure to check out the hanging bridges. How far apart are your antennas? Things might get a lot simpler if you can see the same GPS satellites. If your runs are short enough, you might get better results by letting the oscillator coast during the run. That is unplug the antennas or turn the feedback off in software. This could be all wrong.... There are two things you need. One is frequency, the other is time (PPS). If the time is slightly off, that's equivalent to pointing your antenna in a different direction. You get the same sort of error if the location of an antenna is slightly off. I think the Earth's rotation introduces an error that's equivalent to the frequency being slightly off. -- These are my opinions, not necessarily my employer's. I hate spam. From Murray.Greenman at rakon.com Sun Feb 22 17:51:58 2009 From: Murray.Greenman at rakon.com (Murray Greenman) Date: Mon, 23 Feb 2009 06:51:58 +1300 Subject: [time-nuts] Trimble GPSTM In-Reply-To: References: Message-ID: Chris - Which model Trimble do you mean? I have a Trimble NTGS50AA, built for the Nortel Metrocell system. It talks TSIP, although its native language is (poorly understood) SCPI. The NTGS50AA works fine with TBOLTMON. Regards, Murray Greenman ZL1BPU From magnus at rubidium.dyndns.org Sun Feb 22 19:00:36 2009 From: magnus at rubidium.dyndns.org (Magnus Danielson) Date: Sun, 22 Feb 2009 20:00:36 +0100 Subject: [time-nuts] Different Thunderbolt versions In-Reply-To: <20090222173317.BBA46BCDC@ip-64-139-1-69.sjc.megapath.net> References: <20090222173317.BBA46BCDC@ip-64-139-1-69.sjc.megapath.net> Message-ID: <49A1A0D4.9010800@rubidium.dyndns.org> > There are two things you need. One is frequency, the other is time (PPS). > > If the time is slightly off, that's equivalent to pointing your antenna in a > different direction. You get the same sort of error if the location of an > antenna is slightly off. > > I think the Earth's rotation introduces an error that's equivalent to the > frequency being slightly off. There is two effects... the two antennas will experience different doppler shifts (assuming object in the sky), which can be canceled out fairly easy. There is also the Sagnac effect which causes an apparent time difference, which also can be canceled out fairly easy. Cheers, Magnus From bruce.griffiths at xtra.co.nz Sun Feb 22 20:12:14 2009 From: bruce.griffiths at xtra.co.nz (Bruce Griffiths) Date: Mon, 23 Feb 2009 09:12:14 +1300 Subject: [time-nuts] Different Thunderbolt versions In-Reply-To: <12619539593.20090222173205@fastmail.fm> References: <19210986468.20090221152324@fastmail.fm> <49A07081.1040204@xtra.co.nz> <6C8E7ADA880040799C1477C5C1C32AD5@pc52> <49A0A841.8000702@xtra.co.nz> <12619539593.20090222173205@fastmail.fm> Message-ID: <49A1B19E.9020908@xtra.co.nz> Markus Even with sawtooth correction the performance of the M12+T was found inadequate for the LOFAR array. They use SRS FS725 rubidium sources disciplined by M12+T GPS timing receivers. They also state that the ionosphere contribution to ADEV is about 8E-12 @10s. Bruce Markus Kern wrote: > Thanks everyone for all the input. > > I actually want to work at frequencies below 50Mhz where a 50 degree > phase error corresponds to about 3ns. So for 100 seconds integration > time an ADEV of 1E-11 or better at tau = 1s should be fine. I want to > keep the system as simple and cheap as possible which is why I want to > give the Thunderbolt a try. > > Also at these low frequencies the required accuracy approaches that > claimed by some GPS receivers (4ns for the M12+ ?) so might it be > possible to build a system which is accurate for arbitrary integration > times? > > > Markus > > > On 22.02.2009, 02:20 Bruce Griffiths wrote: > > >> Tom Van Baak wrote: >> >>>> For observations at 1420MHz, and short integration times (~ 100sec) the >>>> phase error needs to be 100ps or less. >>>> For longer integration times (10,000 sec is desirable) and higher >>>> frequencies the timing constraints are far more stringent. >>>> Whilst clock modeling can be used to improve performance it can only >>>> correct systematic effects. >>>> >>>> >>> Bruce, >>> >>> Approximately what range of tau and ADEV(tau) would this >>> VLBI requirement correspond to? A couple of the TBolts I've >>> tested are at or just under 1e-12 at 1 second and stay almost >>> flat out to 100 or even 1000 seconds. >>> >>> /tvb >>> >>> >>> >>> >> Tom >> > > >> Really should use something like TDEV but roughly (within a factor of 2 >> or so at 1420MHz): >> > > >> ADEV < 1E-10 or better @ tau = 1sec for 1 second integration time. >> (integration time to short for anything but very bright objects) >> ADEV < 1E-11 or better @ tau = 1sec for 10 second integration time. >> (integration time to short for anything but very bright objects) >> ADEV < 1E-12 or better @ tau = 1sec for 100 second integration time. >> (integration time somewhat marginal for most objects) >> ADEV < 1E-13 or better @ tau = 1sec for 1000 second integration time. >> ADEV < 1E-14 or better @ tau = 1sec for 10,000 second integration time. >> > > > > >> Also need to take into account that a similar LO would presumably be >> used at each antenna. >> > > >> For integration times of up to a few hundred seconds VLBI has been done >> with a hydrogen maser at one antenna and a PRS10A at the other. >> However the performance isnt Earth shattering and there are some issues >> with long term drift and transport induced frequency shifts of the PRS10A. >> > > >> For very short integration periods a good OCXO is adequate (at 1420MHz). >> When an integration time of 1000sec or more is required the LO >> requirents are more stringent. >> > > >> Bruce >> > > > > _______________________________________________ > time-nuts mailing list -- time-nuts at febo.com > To unsubscribe, go to https://www.febo.com/cgi-bin/mailman/listinfo/time-nuts > and follow the instructions there. > > From dave.mallery at gmail.com Sun Feb 22 23:35:25 2009 From: dave.mallery at gmail.com (Dave Mallery) Date: Sun, 22 Feb 2009 16:35:25 -0700 Subject: [time-nuts] Lady Heather question serial error In-Reply-To: <49A029ED.1090706@rubidium.dyndns.org> References: <49A029ED.1090706@rubidium.dyndns.org> Message-ID: <3010c9470902221535q29e02e28yb8aa55f4664775cb@mail.gmail.com> oh yes please!!! dave On Sat, Feb 21, 2009 at 9:21 AM, Magnus Danielson < magnus at rubidium.dyndns.org> wrote: > David C. Partridge skrev: > > Yes, Windows always seems to think a Thunderbolt is an MS mouse. Look > back > > in the archives about using /Fastdetect in boot.ini (BTW this doesn't > work > > with WinXP 64 - so I had to disable the serial mouse in the devices). > > The cure for that is very old - don't have the serial port connected > during windows booting. It happend to me too, but recalling the old > workaround and I had no problem what so ever. > > I'm considering a Linux port since I want to reduce my dependency on > Windows at all times. > > Cheers, > Magnus > > _______________________________________________ > time-nuts mailing list -- time-nuts at febo.com > To unsubscribe, go to > https://www.febo.com/cgi-bin/mailman/listinfo/time-nuts > and follow the instructions there. > -- Dave Mallery, K5EN (ubuntu linux 8.10) PO Box 3519; T or C, NM 87901 no gates... no windows! free at last! linux counter #64628 (since 1997) "People aren't as dumb as Microsoft needs them to be." --PJ, May 2007 From masondg44 at comcast.net Mon Feb 23 02:25:24 2009 From: masondg44 at comcast.net (Dave M) Date: Sun, 22 Feb 2009 21:25:24 -0500 Subject: [time-nuts] Different Thunderbolt versions, group buy References: Message-ID: <2CE1F5643085456E84026FDDED7B74C5@DELL2350> LMAO @ the new TP discussion!!!! Thanks for the description of the TBolt buy. Yes, please add my name to the list of buyers for one of the next batch of T_Bolts. I've been putting off buying one to the Ebay offerings, but this opportunity will push me over the edge. I'll definitely buy one. Power supplies aren't a problem for me... I have a few linears that should work well, or I could build whatever I need. If finding a reliable PSU becomes a problem, I can take the T-Bolt without a PSU. Dave M MasonDG44 at comcast dot net (Just substitute the appropriate characters in the address) Life is like a roll of toilet paper; the closer it gets to the end, the faster it goes. ---------------------------------------------------------------------- Message: 1 Date: Sat, 21 Feb 2009 18:51:38 -0800 From: "Tom Van Baak" Subject: Re: [time-nuts] Different Thunderbolt versions, group buy To: "Discussion of precise time and frequency measurement" Message-ID: Content-Type: text/plain; format=flowed; charset="Windows-1252"; reply-type=original >I currently have an HP Z3801A, which seems to be running > well. I've been looking at the T-Bolts on Ebay for a while, but > have been holding off buying. This post just gave me an > opportunity to ask some questions. > > I've seen references to the TAPR group buy in previous posts. > It obviously happened before I found this group, but am interested > in the details. > Things like: > When did the buy happen? > Which model T-Bolt was bought? > What was the price? > Was PSU included? > Are any units still available? > Is another buy planned for the near future? > > Thanks for the info. > Hi Dave, Yes, there have been 3 or 4 large batches of TBolts sold through TAPR over the past year or two. The price ranged from about 95 to 125. Most were sold with power supply; some without (when the source of cheap switching supplies ran out). For more info see: http://www.leapsecond.com/tbolt-faq.htm http://www.leapsecond.com/pages/tapr-tbolt/ http://www.leapsecond.com/pages/tbolt/noise.htm Yes, there will be another group buy or two. However, this time I'm testing all of them first and then announcing the sale (expect mid-March) so that no one has that long wait between ordering and paying/receiving. So hang on a bit longer and I'll send the details next month. Note a number of people have asked privately (you know who you are) and those requests will be handled first. From stijena at tapko.de Mon Feb 23 08:35:44 2009 From: stijena at tapko.de (Predrag Dukic) Date: Mon, 23 Feb 2009 09:35:44 +0100 Subject: [time-nuts] Different Thunderbolt versions, group buy In-Reply-To: <2CE1F5643085456E84026FDDED7B74C5@DELL2350> References: <2CE1F5643085456E84026FDDED7B74C5@DELL2350> Message-ID: <7.0.1.0.1.20090223093226.01f97e98@tapko.de> Well, since people are already announcing their intentions to buy from the next batch, maybe I would do vise to do the same. Of course, if I am eligible, as I already bought one from the previous batch. So, depending on it and on the size of the batch I want to buy one or two units. Predrag Dukic At 03:25 23.2.2009, you wrote: >LMAO @ the new TP discussion!!!! >Thanks for the description of the TBolt buy. > > >Yes, please add my name to the list of buyers for one of the next >batch of T_Bolts. I've been putting off buying one to the Ebay >offerings, but this opportunity will push me over the edge. I'll >definitely buy one. >Power supplies aren't a problem for me... I have a few linears that >should work well, or I could build whatever I need. If finding a >reliable PSU becomes a problem, I can take the T-Bolt without a PSU. > >Dave M >MasonDG44 at comcast dot net (Just substitute the appropriate >characters in the address) > >Life is like a roll of toilet paper; the closer it gets to the end, >the faster it goes. > > ---------------------------------------------------------------------- > > Message: 1 > Date: Sat, 21 Feb 2009 18:51:38 -0800 > From: "Tom Van Baak" > Subject: Re: [time-nuts] Different Thunderbolt versions, group buy > To: "Discussion of precise time and frequency measurement" > > Message-ID: > Content-Type: text/plain; format=flowed; charset="Windows-1252"; > reply-type=original > > >I currently have an HP Z3801A, which seems to be running > > well. I've been looking at the T-Bolts on Ebay for a while, but > > have been holding off buying. This post just gave me an > > opportunity to ask some questions. > > > > I've seen references to the TAPR group buy in previous posts. > > It obviously happened before I found this group, but am interested > > in the details. > > Things like: > > When did the buy happen? > > Which model T-Bolt was bought? > > What was the price? > > Was PSU included? > > Are any units still available? > > Is another buy planned for the near future? > > > > Thanks for the info. > > > > Hi Dave, > > Yes, there have been 3 or 4 large batches of TBolts sold through > TAPR over the past year or two. The price ranged from about 95 > to 125. Most were sold with power supply; some without (when the > source of cheap switching supplies ran out). For more info see: > http://www.leapsecond.com/tbolt-faq.htm > http://www.leapsecond.com/pages/tapr-tbolt/ > http://www.leapsecond.com/pages/tbolt/noise.htm > > Yes, there will be another group buy or two. However, this time > I'm testing all of them first and then announcing the sale (expect > mid-March) so that no one has that long wait between ordering > and paying/receiving. > > So hang on a bit longer and I'll send the details next month. > > Note a number of people have asked privately (you know who > you are) and those requests will be handled first. > > >_______________________________________________ >time-nuts mailing list -- time-nuts at febo.com >To unsubscribe, go to https://www.febo.com/cgi-bin/mailman/listinfo/time-nuts >and follow the instructions there. From XDE-L2G3 at myamail.com Mon Feb 23 09:06:16 2009 From: XDE-L2G3 at myamail.com (Mike Monett) Date: Mon, 23 Feb 2009 04:06:16 -0500 Subject: [time-nuts] WWVB Wall and Travel Clocks On Sale Message-ID: To All, For those who might be interested, here's a good selection of WWVB (60KHz) wall clocks on sale at $14.00 and up. I have no idea of the quality or the vendor, but I just ordered two for comparison with GPS. http://www.giftngadget.com/analog-atomic-wall-clocks.html Regards, Mike From XDE-L2G3 at myamail.com Mon Feb 23 12:55:26 2009 From: XDE-L2G3 at myamail.com (Mike Monett) Date: Mon, 23 Feb 2009 07:55:26 -0500 Subject: [time-nuts] WWVB Wall and Travel Clocks On Sale Message-ID: Update on my previous post - tvb has a nice page on WWVB receivers at http://leapsecond.com/pages/sony-wwvb/ Looks like these receivers might make good wall clocks. But for checking GPSDO's - not so good:) Regards, Mike From martinrh45 at googlemail.com Mon Feb 23 13:24:13 2009 From: martinrh45 at googlemail.com (Martin Richmond-Hardy) Date: Mon, 23 Feb 2009 13:24:13 +0000 Subject: [time-nuts] Different Thunderbolt versions, group buy In-Reply-To: <20090222031720.8C3941165C1@hamburg.alientech.net> References: <32C2665C055244E08122B8081642ECC3@DELL2350> <20090222031720.8C3941165C1@hamburg.alientech.net> Message-ID: Please put me down for a TBolt in the next bulk purchase. 73 de Martin Richmond-Hardy G8BHC QTH: JO02pa Skype: callto://martinrh45 or callto://g8bhc-Suffolk Truphone on my iPhone From jltran at worldnet.att.net Mon Feb 23 13:56:09 2009 From: jltran at worldnet.att.net (J. L. Trantham) Date: Mon, 23 Feb 2009 07:56:09 -0600 Subject: [time-nuts] Different Thunderbolt versions, group buy In-Reply-To: <7.0.1.0.1.20090223093226.01f97e98@tapko.de> Message-ID: Same for me. I, also, would like to have another Thunderbolt. I have been very pleased with the Thunderbolt from the prior group buys. Joe -----Original Message----- From: time-nuts-bounces at febo.com [mailto:time-nuts-bounces at febo.com] On Behalf Of Predrag Dukic Sent: Monday, February 23, 2009 2:36 AM To: Discussion of precise time and frequency measurement Subject: Re: [time-nuts] Different Thunderbolt versions, group buy Well, since people are already announcing their intentions to buy from the next batch, maybe I would do vise to do the same. Of course, if I am eligible, as I already bought one from the previous batch. So, depending on it and on the size of the batch I want to buy one or two units. Predrag Dukic At 03:25 23.2.2009, you wrote: >LMAO @ the new TP discussion!!!! >Thanks for the description of the TBolt buy. > > >Yes, please add my name to the list of buyers for one of the next >batch of T_Bolts. I've been putting off buying one to the Ebay >offerings, but this opportunity will push me over the edge. I'll >definitely buy one. >Power supplies aren't a problem for me... I have a few linears that >should work well, or I could build whatever I need. If finding a >reliable PSU becomes a problem, I can take the T-Bolt without a PSU. > >Dave M >MasonDG44 at comcast dot net (Just substitute the appropriate >characters in the address) > >Life is like a roll of toilet paper; the closer it gets to the end, >the faster it goes. > > ---------------------------------------------------------------------- > > Message: 1 > Date: Sat, 21 Feb 2009 18:51:38 -0800 > From: "Tom Van Baak" > Subject: Re: [time-nuts] Different Thunderbolt versions, group buy > To: "Discussion of precise time and frequency measurement" > > Message-ID: > Content-Type: text/plain; format=flowed; charset="Windows-1252"; > reply-type=original > > >I currently have an HP Z3801A, which seems to be running > > well. I've been looking at the T-Bolts on Ebay for a while, but > > have been holding off buying. This post just gave me an > > opportunity to ask some questions. > > > > I've seen references to the TAPR group buy in previous posts. > > It obviously happened before I found this group, but am interested > > in the details. > > Things like: > > When did the buy happen? > > Which model T-Bolt was bought? > > What was the price? > > Was PSU included? > > Are any units still available? > > Is another buy planned for the near future? > > > > Thanks for the info. > > > > Hi Dave, > > Yes, there have been 3 or 4 large batches of TBolts sold through > TAPR over the past year or two. The price ranged from about 95 > to 125. Most were sold with power supply; some without (when the > source of cheap switching supplies ran out). For more info see: > http://www.leapsecond.com/tbolt-faq.htm > http://www.leapsecond.com/pages/tapr-tbolt/ > http://www.leapsecond.com/pages/tbolt/noise.htm > > Yes, there will be another group buy or two. However, this time > I'm testing all of them first and then announcing the sale (expect > mid-March) so that no one has that long wait between ordering > and paying/receiving. > > So hang on a bit longer and I'll send the details next month. > > Note a number of people have asked privately (you know who > you are) and those requests will be handled first. > > >_______________________________________________ >time-nuts mailing list -- time-nuts at febo.com >To unsubscribe, go to https://www.febo.com/cgi-bin/mailman/listinfo/time-nuts >and follow the instructions there. _______________________________________________ time-nuts mailing list -- time-nuts at febo.com To unsubscribe, go to https://www.febo.com/cgi-bin/mailman/listinfo/time-nuts and follow the instructions there. From EWKehren at aol.com Mon Feb 23 14:06:19 2009 From: EWKehren at aol.com (EWKehren at aol.com) Date: Mon, 23 Feb 2009 09:06:19 EST Subject: [time-nuts] Different Thunderbolt versions, group buy Message-ID: Please add me to the list. Bert Kehren Miami WB5MZJ **************You're invited to Hollywood's biggest party: Get Oscars updates, red carpet pics and more at Moviefone. (http://movies.aol.com/oscars-academy-awards?ncid=emlcntusmovi00000001) From phase3 at wcc.net Mon Feb 23 15:32:56 2009 From: phase3 at wcc.net (Bill Mason) Date: Mon, 23 Feb 2009 09:32:56 -0600 Subject: [time-nuts] Different Thunderbolt versions, group buy In-Reply-To: Message-ID: I have a 3801 and would like to have one of the Tbolts....how do I get on the list? Thanks... Bill Mason W5STP -----Original Message----- From: time-nuts-bounces at febo.com [mailto:time-nuts-bounces at febo.com] On Behalf Of EWKehren at aol.com Sent: Monday, February 23, 2009 8:06 AM To: time-nuts at febo.com Subject: Re: [time-nuts] Different Thunderbolt versions, group buy Please add me to the list. Bert Kehren Miami WB5MZJ **************You're invited to Hollywood's biggest party: Get Oscars updates, red carpet pics and more at Moviefone. (http://movies.aol.com/oscars-academy-awards?ncid=emlcntusmovi00000001) _______________________________________________ time-nuts mailing list -- time-nuts at febo.com To unsubscribe, go to https://www.febo.com/cgi-bin/mailman/listinfo/time-nuts and follow the instructions there. From masondg44 at comcast.net Mon Feb 23 15:53:25 2009 From: masondg44 at comcast.net (Dave M) Date: Mon, 23 Feb 2009 10:53:25 -0500 Subject: [time-nuts] WWVB Wall and Travel Clocks On Sale References: Message-ID: Thanks for the alert. I already have several WWVB clocks in my house and shop. Two are the SkyScan brand (one analog, the other digital); the other is one of the LaCrosse projection clocks. All three are almost always in agreement; never differing by more than a fraction of a second. I think I bought all except the LaCrosse clock from WalMart. I don't think I paid more than $20 US for any of the clocks. Dave M MasonDG44 at comcast dot net (Just substitute the appropriate characters in the address) Life is like a roll of toilet paper; the closer it gets to the end, the faster it goes. ------------------------------ Message: 2 Date: Mon, 23 Feb 2009 04:06:16 -0500 From: "Mike Monett" Subject: [time-nuts] WWVB Wall and Travel Clocks On Sale To: "" Message-ID: Content-Type: text/plain; charset="iso-8859-1" To All, For those who might be interested, here's a good selection of WWVB (60KHz) wall clocks on sale at $14.00 and up. I have no idea of the quality or the vendor, but I just ordered two for comparison with GPS. http://www.giftngadget.com/analog-atomic-wall-clocks.html Regards, Mike From w1ksz at earthlink.net Mon Feb 23 16:34:19 2009 From: w1ksz at earthlink.net (Richard W. Solomon) Date: Mon, 23 Feb 2009 09:34:19 -0700 (GMT-07:00) Subject: [time-nuts] RFTGm-II-XO Data Message-ID: <29633427.1235406860253.JavaMail.root@elwamui-rubis.atl.sa.earthlink.net> I messed up and deleted the link to the Lucent RFTGm-II-XO. Can someone send it to me again. I tried a Google search to no avail. Thanks, Dick, W1KSZ From biwa at att.net Mon Feb 23 17:00:22 2009 From: biwa at att.net (Burt I. Weiner) Date: Mon, 23 Feb 2009 09:00:22 -0800 Subject: [time-nuts] T-Bolt Group Buy... Message-ID: <7.0.0.16.2.20090223085925.0365ba58@att.net> I'd also like to get on the list. Burt, K6OQK Burt I. Weiner Associates Broadcast Technical Services Glendale, California U.S.A. biwa at att.net K6OQK From stanley_reynolds at yahoo.com Mon Feb 23 17:01:06 2009 From: stanley_reynolds at yahoo.com (Stanley Reynolds) Date: Mon, 23 Feb 2009 09:01:06 -0800 (PST) Subject: [time-nuts] RFTGm-II-XO Data References: <29633427.1235406860253.JavaMail.root@elwamui-rubis.atl.sa.earthlink.net> Message-ID: <773164.11757.qm@web30301.mail.mud.yahoo.com> Dick, The RFTG-m-XO is at http://www.n4iqt.com/lucentgps/?I'm looking for pictures now to help id any differences to what you have with the "II". Do you have pictures of your unit ? I think the "RFTG" is GPSDO and "RFG" has no GPS built in link at http://www.n4iqt.com/rfgmxo/?, not sure what the "II" is you have. Stanley ________________________________ From: Richard W. Solomon To: time-nuts at febo.com Sent: Monday, February 23, 2009 10:34:19 AM Subject: [time-nuts] RFTGm-II-XO Data I messed up and deleted the link to the Lucent RFTGm-II-XO. Can someone send it to me again. I tried a Google search to no avail. Thanks, Dick, W1KSZ _______________________________________________ time-nuts mailing list -- time-nuts at febo.com To unsubscribe, go to https://www.febo.com/cgi-bin/mailman/listinfo/time-nuts and follow the instructions there. From w1ksz at earthlink.net Mon Feb 23 17:06:45 2009 From: w1ksz at earthlink.net (Richard W. Solomon) Date: Mon, 23 Feb 2009 10:06:45 -0700 (GMT-07:00) Subject: [time-nuts] RFTGm-II-XO Data Message-ID: <21906096.1235408805964.JavaMail.root@elwamui-rubis.atl.sa.earthlink.net> The link I am looking for has the pin-outs for the connectors. Not nice to hook +24 to a 5v line !! Thanks, Dick, W1KSZ -----Original Message----- >From: Stanley Reynolds >Sent: Feb 23, 2009 10:01 AM >To: time-nuts at febo.com >Subject: Re: [time-nuts] RFTGm-II-XO Data > >Dick, > >The RFTG-m-XO is at http://www.n4iqt.com/lucentgps/?I'm looking for pictures now to help id any differences to what you have with the "II". Do you have pictures of your unit ? I think the "RFTG" is GPSDO and "RFG" has no GPS built in link at http://www.n4iqt.com/rfgmxo/?, not sure what the "II" is you have. > >Stanley > > > >________________________________ >From: Richard W. Solomon >To: time-nuts at febo.com >Sent: Monday, February 23, 2009 10:34:19 AM >Subject: [time-nuts] RFTGm-II-XO Data > >I messed up and deleted the link to the Lucent RFTGm-II-XO. >Can someone send it to me again. I tried a Google search >to no avail. > >Thanks, Dick, W1KSZ > >_______________________________________________ >time-nuts mailing list -- time-nuts at febo.com >To unsubscribe, go to https://www.febo.com/cgi-bin/mailman/listinfo/time-nuts >and follow the instructions there. >_______________________________________________ >time-nuts mailing list -- time-nuts at febo.com >To unsubscribe, go to https://www.febo.com/cgi-bin/mailman/listinfo/time-nuts >and follow the instructions there. From bell.d.a at comcast.net Mon Feb 23 17:18:31 2009 From: bell.d.a at comcast.net (bell.d.a at comcast.net) Date: Mon, 23 Feb 2009 17:18:31 +0000 (UTC) Subject: [time-nuts] Different Thunderbolt versions, group buy In-Reply-To: Message-ID: <1127061799.1545481235409511379.JavaMail.root@sz0121a.emeryville.ca.mail.comcast.net> Please add me to the next group buy list. Sincere thanks ----- Original Message ----- From: EWKehren at aol.com To: time-nuts at febo.com Sent: Monday, February 23, 2009 6:06:19 AM GMT -08:00 US/Canada Pacific Subject: Re: [time-nuts] Different Thunderbolt versions, group buy Please add me to the list. Bert Kehren Miami ? WB5MZJ **************You're invited to Hollywood's biggest party: Get Oscars updates, red carpet pics and more at Moviefone. (http://movies.aol.com/oscars-academy-awards?ncid=emlcntusmovi00000001) _______________________________________________ time-nuts mailing list -- time-nuts at febo.com To unsubscribe, go to https://www.febo.com/cgi-bin/mailman/listinfo/time-nuts and follow the instructions there. From djl at montana.com Mon Feb 23 17:23:15 2009 From: djl at montana.com (Don Latham) Date: Mon, 23 Feb 2009 10:23:15 -0700 (MST) Subject: [time-nuts] T-Bolt Group Buy... In-Reply-To: <7.0.0.16.2.20090223085925.0365ba58@att.net> References: <7.0.0.16.2.20090223085925.0365ba58@att.net> Message-ID: <2972.216.14.241.110.1235409795.squirrel@webmail.montana.com> I would like one as well. BTW, I drove an ordinary "quartz" wall clock with the 1 pps output of a Motorola gps time unit as follows: use a low pwr ttl Dtype f/f. Clock to 1 pps, data to +5v. Put 180 ohm or so (as I remember) in series with the q and -q outputs. Carefully find the clock pulse output, remove the coil connections, and hook 1 of the 180 ohm r's to each end of the coil. This gives one reversal per second which drives the second hand. If it amuses you, the clock will drive perfectly well backwards... It's been a while, so I don't remember any more details, so have fun with it. Don Dr. Don Latham AJ7LL Six Mile Systems LLP 17850 Six Mile Road POB 134 Huson, MT, 59846 VOX 406-626-4304 www.lightningforensics.com www.sixmilesystems.com From tvb at LeapSecond.com Mon Feb 23 17:33:28 2009 From: tvb at LeapSecond.com (Tom Van Baak) Date: Mon, 23 Feb 2009 09:33:28 -0800 Subject: [time-nuts] Different Thunderbolt versions, group buy References: <1127061799.1545481235409511379.JavaMail.root@sz0121a.emeryville.ca.mail.comcast.net> Message-ID: Sorry, maybe I wasn't clear. I will announce details of next group buy in mid-March. Until then, please wait quietly. You do not need to send email to me, or to the group at this time. Please don't reply to this message either. Thanks, /tvb From tvb at LeapSecond.com Mon Feb 23 17:33:25 2009 From: tvb at LeapSecond.com (Tom Van Baak) Date: Mon, 23 Feb 2009 09:33:25 -0800 Subject: [time-nuts] T-Bolt Group Buy... References: <7.0.0.16.2.20090223085925.0365ba58@att.net> Message-ID: <55511E5A079E44A88280DC2B97C641AC@pc52> Sorry, maybe I wasn't clear. I will announce details of next group buy in mid-March. Until then, please wait quietly. You do not need to send email to me, or to the group at this time. Please don't reply to this message either. Thanks, /tvb From stanley_reynolds at yahoo.com Mon Feb 23 17:48:25 2009 From: stanley_reynolds at yahoo.com (Stanley Reynolds) Date: Mon, 23 Feb 2009 09:48:25 -0800 (PST) Subject: [time-nuts] RFTGm-II-XO Data References: <21906096.1235408805964.JavaMail.root@elwamui-rubis.atl.sa.earthlink.net> Message-ID: <122386.13644.qm@web30303.mail.mud.yahoo.com> Is this it ? http://n4iqt.com/rfgmxo/RFGMXO.txt I think they all receive power on J1 pin 1?+24vdc, Ground pin 2. The power may normally loop thru one box to another as described in the text document above. My older non "II" unit needs a load on the GPS antenna connector to be happy I use a resistor as the antenna preamp is powered by another receiver thru a?splitter. Stanley ________________________________ From: Richard W. Solomon To: Discussion of precise time and frequency measurement Sent: Monday, February 23, 2009 11:06:45 AM Subject: Re: [time-nuts] RFTGm-II-XO Data The link I am looking for has the pin-outs for the connectors. Not nice to hook +24 to a 5v line !! Thanks, Dick, W1KSZ -----Original Message----- >From: Stanley Reynolds >Sent: Feb 23, 2009 10:01 AM >To: time-nuts at febo.com >Subject: Re: [time-nuts] RFTGm-II-XO Data > >Dick, > >The RFTG-m-XO is at http://www.n4iqt.com/lucentgps/?I'm looking for pictures now to help id any differences to what you have with the "II". Do you have pictures of your unit ? I think the "RFTG" is GPSDO and "RFG" has no GPS built in link at http://www.n4iqt.com/rfgmxo/?, not sure what the "II" is you have. > >Stanley > > > >________________________________ >From: Richard W. Solomon >To: time-nuts at febo.com >Sent: Monday, February 23, 2009 10:34:19 AM >Subject: [time-nuts] RFTGm-II-XO Data > >I messed up and deleted the link to the Lucent RFTGm-II-XO. >Can someone send it to me again. I tried a Google search >to no avail. > >Thanks, Dick, W1KSZ > >_______________________________________________ >time-nuts mailing list -- time-nuts at febo.com >To unsubscribe, go to https://www.febo.com/cgi-bin/mailman/listinfo/time-nuts >and follow the instructions there. >_______________________________________________ >time-nuts mailing list -- time-nuts at febo.com >To unsubscribe, go to https://www.febo.com/cgi-bin/mailman/listinfo/time-nuts >and follow the instructions there. _______________________________________________ time-nuts mailing list -- time-nuts at febo.com To unsubscribe, go to https://www.febo.com/cgi-bin/mailman/listinfo/time-nuts and follow the instructions there. From cdelect at juno.com Mon Feb 23 18:00:18 2009 From: cdelect at juno.com (Corby Dawson) Date: Mon, 23 Feb 2009 10:00:18 -0800 Subject: [time-nuts] next thunderbolt buy Message-ID: <20090223.100018.1320.5.cdelect@juno.com> Hi, Put me on the list also. Thanks! Corby Dawson ____________________________________________________________ 30 Seconds can save a lifetime. Get it done. Its never been easier. http://thirdpartyoffers.juno.com/TGL2141/fc/BLSrjpTQcxImVtsp1I4ud0g3sh8lLJAlhqMAJGwZKZEbcnSlH6aQmV1jvlW/ From brooke at pacific.net Mon Feb 23 18:02:47 2009 From: brooke at pacific.net (Brooke Clarke) Date: Mon, 23 Feb 2009 10:02:47 -0800 Subject: [time-nuts] WWVB Wall and Travel Clocks On Sale In-Reply-To: References: Message-ID: <49A2E4C7.1040700@pacific.net> Hi Tom: I've looked at dozens of Sony "Dream Machine" clocks on eBay and none of them have WWVB capability. Some are called preset or self setting, but these may just be using a button cell back up clock. What is the model number of your Sony WWVB clock? Have Fun, Brooke Clarke http://www.prc68.com Mike Monett wrote: > Update on my previous post - tvb has a nice page on WWVB receivers at > > http://leapsecond.com/pages/sony-wwvb/ > > Looks like these receivers might make good wall clocks. > > But for checking GPSDO's - not so good:) > > Regards, > > Mike > > _______________________________________________ > time-nuts mailing list -- time-nuts at febo.com > To unsubscribe, go to https://www.febo.com/cgi-bin/mailman/listinfo/time-nuts > and follow the instructions there. > > From w1ksz at earthlink.net Mon Feb 23 18:10:15 2009 From: w1ksz at earthlink.net (Richard W. Solomon) Date: Mon, 23 Feb 2009 11:10:15 -0700 (GMT-07:00) Subject: [time-nuts] RFTGm-II-XO Data Message-ID: <19238928.1235412615986.JavaMail.root@elwamui-rubis.atl.sa.earthlink.net> OK, that should do it. Now, hopefully I will not let all the magic smoke out !! Thanks, Dick, W1KSZ -----Original Message----- >From: Stanley Reynolds >Sent: Feb 23, 2009 10:48 AM >To: time-nuts at febo.com >Subject: Re: [time-nuts] RFTGm-II-XO Data > >Is this it ? http://n4iqt.com/rfgmxo/RFGMXO.txt >I think they all receive power on J1 pin 1?+24vdc, Ground pin 2. The power may normally loop thru one box to another as described in the text document above. My older non "II" unit needs a load on the GPS antenna connector to be happy I use a resistor as the antenna preamp is powered by another receiver thru a?splitter. > >Stanley > > > > >________________________________ >From: Richard W. Solomon >To: Discussion of precise time and frequency measurement >Sent: Monday, February 23, 2009 11:06:45 AM >Subject: Re: [time-nuts] RFTGm-II-XO Data > >The link I am looking for has the pin-outs for the connectors. Not nice to >hook +24 to a 5v line !! > >Thanks, Dick, W1KSZ > >-----Original Message----- >>From: Stanley Reynolds >>Sent: Feb 23, 2009 10:01 AM >>To: time-nuts at febo.com >>Subject: Re: [time-nuts] RFTGm-II-XO Data >> >>Dick, >> >>The RFTG-m-XO is at http://www.n4iqt.com/lucentgps/?I'm looking for pictures now to help id any differences to what you have with the "II". Do you have pictures of your unit ? I think the "RFTG" is GPSDO and "RFG" has no GPS built in link at http://www.n4iqt.com/rfgmxo/?, not sure what the "II" is you have. >> >>Stanley >> >> >> >>________________________________ >>From: Richard W. Solomon >>To: time-nuts at febo.com >>Sent: Monday, February 23, 2009 10:34:19 AM >>Subject: [time-nuts] RFTGm-II-XO Data >> >>I messed up and deleted the link to the Lucent RFTGm-II-XO. >>Can someone send it to me again. I tried a Google search >>to no avail. >> >>Thanks, Dick, W1KSZ >> >>_______________________________________________ >>time-nuts mailing list -- time-nuts at febo.com >>To unsubscribe, go to https://www.febo.com/cgi-bin/mailman/listinfo/time-nuts >>and follow the instructions there. >>_______________________________________________ >>time-nuts mailing list -- time-nuts at febo.com >>To unsubscribe, go to https://www.febo.com/cgi-bin/mailman/listinfo/time-nuts >>and follow the instructions there. > > >_______________________________________________ >time-nuts mailing list -- time-nuts at febo.com >To unsubscribe, go to https://www.febo.com/cgi-bin/mailman/listinfo/time-nuts >and follow the instructions there. >_______________________________________________ >time-nuts mailing list -- time-nuts at febo.com >To unsubscribe, go to https://www.febo.com/cgi-bin/mailman/listinfo/time-nuts >and follow the instructions there. From tvb at LeapSecond.com Mon Feb 23 18:32:49 2009 From: tvb at LeapSecond.com (Tom Van Baak) Date: Mon, 23 Feb 2009 10:32:49 -0800 Subject: [time-nuts] WWVB Wall and Travel Clocks On Sale References: <49A2E4C7.1040700@pacific.net> Message-ID: <1FF7D3FA910B4509B28BB8B796653D23@pc52> Brooke, The first set of WWVB clocks that showed up in the 90's had external antennas, among them Oregon Scientific and Sony. Within a few years they all went with internal antennas and got ever more compact. RC wristwatches are now common. I can't make out the model number from my own web photos so I'll dig around for the original clock and let you know. /tvb ----- Original Message ----- From: "Brooke Clarke" To: "Discussion of precise time and frequency measurement" Sent: Monday, February 23, 2009 10:02 AM Subject: Re: [time-nuts] WWVB Wall and Travel Clocks On Sale > Hi Tom: > > I've looked at dozens of Sony "Dream Machine" clocks on eBay and none of them > have WWVB capability. Some are called preset or self setting, but these may > just be using a button cell back up clock. > > What is the model number of your Sony WWVB clock? > > Have Fun, > > Brooke Clarke > http://www.prc68.com > > Mike Monett wrote: >> Update on my previous post - tvb has a nice page on WWVB receivers at >> >> http://leapsecond.com/pages/sony-wwvb/ >> >> Looks like these receivers might make good wall clocks. >> >> But for checking GPSDO's - not so good:) >> >> Regards, >> >> Mike From w1ksz at earthlink.net Mon Feb 23 19:05:14 2009 From: w1ksz at earthlink.net (Richard W. Solomon) Date: Mon, 23 Feb 2009 12:05:14 -0700 (GMT-07:00) Subject: [time-nuts] RFTGm-II-XO Data Message-ID: <20685356.1235415915263.JavaMail.root@elwamui-rubis.atl.sa.earthlink.net> One success this day, the Lucent Box shows Green and 15 MHz comes out and now to check how accurate it is (only joking !!). Thanks for the help. 73, Dick, W1KSZ -----Original Message----- >From: Stanley Reynolds >Sent: Feb 23, 2009 10:48 AM >To: time-nuts at febo.com >Subject: Re: [time-nuts] RFTGm-II-XO Data > >Is this it ? http://n4iqt.com/rfgmxo/RFGMXO.txt >I think they all receive power on J1 pin 1?+24vdc, Ground pin 2. The power may normally loop thru one box to another as described in the text document above. My older non "II" unit needs a load on the GPS antenna connector to be happy I use a resistor as the antenna preamp is powered by another receiver thru a?splitter. > >Stanley > > > > >________________________________ >From: Richard W. Solomon >To: Discussion of precise time and frequency measurement >Sent: Monday, February 23, 2009 11:06:45 AM >Subject: Re: [time-nuts] RFTGm-II-XO Data > >The link I am looking for has the pin-outs for the connectors. Not nice to >hook +24 to a 5v line !! > >Thanks, Dick, W1KSZ > >-----Original Message----- >>From: Stanley Reynolds >>Sent: Feb 23, 2009 10:01 AM >>To: time-nuts at febo.com >>Subject: Re: [time-nuts] RFTGm-II-XO Data >> >>Dick, >> >>The RFTG-m-XO is at http://www.n4iqt.com/lucentgps/?I'm looking for pictures now to help id any differences to what you have with the "II". Do you have pictures of your unit ? I think the "RFTG" is GPSDO and "RFG" has no GPS built in link at http://www.n4iqt.com/rfgmxo/?, not sure what the "II" is you have. >> >>Stanley >> >> >> >>________________________________ >>From: Richard W. Solomon >>To: time-nuts at febo.com >>Sent: Monday, February 23, 2009 10:34:19 AM >>Subject: [time-nuts] RFTGm-II-XO Data >> >>I messed up and deleted the link to the Lucent RFTGm-II-XO. >>Can someone send it to me again. I tried a Google search >>to no avail. >> >>Thanks, Dick, W1KSZ >> >>_______________________________________________ >>time-nuts mailing list -- time-nuts at febo.com >>To unsubscribe, go to https://www.febo.com/cgi-bin/mailman/listinfo/time-nuts >>and follow the instructions there. >>_______________________________________________ >>time-nuts mailing list -- time-nuts at febo.com >>To unsubscribe, go to https://www.febo.com/cgi-bin/mailman/listinfo/time-nuts >>and follow the instructions there. > > >_______________________________________________ >time-nuts mailing list -- time-nuts at febo.com >To unsubscribe, go to https://www.febo.com/cgi-bin/mailman/listinfo/time-nuts >and follow the instructions there. >_______________________________________________ >time-nuts mailing list -- time-nuts at febo.com >To unsubscribe, go to https://www.febo.com/cgi-bin/mailman/listinfo/time-nuts >and follow the instructions there. From ed_palmer at sasktel.net Mon Feb 23 19:41:13 2009 From: ed_palmer at sasktel.net (Ed Palmer) Date: Mon, 23 Feb 2009 13:41:13 -0600 Subject: [time-nuts] WWVB Wall and Travel Clocks On Sale In-Reply-To: <49A2E4C7.1040700@pacific.net> References: <49A2E4C7.1040700@pacific.net> Message-ID: <49A2FBD9.5000908@sasktel.net> Hi Brooke, Is there something special about the Sony clock that you want? Before I found out about Time Nuts & GPSDOs I was looking at using one of these "atomic clocks" as a source of time and /or frequency. Most of them have a point where you can get to the time code. Getting to the 60 KHz carrier was a bit trickier. e.g. I picked up a few Hummer Atomic Clocks on clearance for about $5. The radio is a seperate unit that sends out the time code continuously. The clock takes about 5 minutes to lock in. I'm about 680 miles from Fort Collins. Ed Brooke Clarke wrote: > Hi Tom: > > I've looked at dozens of Sony "Dream Machine" clocks on eBay and none of them > have WWVB capability. Some are called preset or self setting, but these may > just be using a button cell back up clock. > > What is the model number of your Sony WWVB clock? > > Have Fun, > > Brooke Clarke > http://www.prc68.com > > Mike Monett wrote: > >> Update on my previous post - tvb has a nice page on WWVB receivers at >> >> http://leapsecond.com/pages/sony-wwvb/ >> >> Looks like these receivers might make good wall clocks. >> >> But for checking GPSDO's - not so good:) >> >> Regards, >> >> Mike >> From brooke at pacific.net Mon Feb 23 20:12:20 2009 From: brooke at pacific.net (Brooke Clarke) Date: Mon, 23 Feb 2009 12:12:20 -0800 Subject: [time-nuts] WWVB Wall and Travel Clocks On Sale In-Reply-To: <49A2FBD9.5000908@sasktel.net> References: <49A2E4C7.1040700@pacific.net> <49A2FBD9.5000908@sasktel.net> Message-ID: <49A30324.7040005@pacific.net> Hi Ed: Yes, the one shown on Tom's web page: http://leapsecond.com/pages/sony-wwvb/ uses two 60 kHz crystals in the filter rather than the single crystal like in the C-max CMMR-6P-60 (Digi-Key p/n: 561-1014-ND) that uses a single crystal. C-Max is the new name for Temic that made the chip in Tom's clock. http://www.prc68.com/I/Loop.shtml#CMMR6P60 I connected the CMMR-6P-60 to a couple of AA batteries and it's been blinking for the last month or two. But that data is very noisy. Have Fun, Brooke Clarke http://www.prc68.com Ed Palmer wrote: > Hi Brooke, > > Is there something special about the Sony clock that you want? Before I > found out about Time Nuts & GPSDOs I was looking at using one of these > "atomic clocks" as a source of time and /or frequency. Most of them > have a point where you can get to the time code. Getting to the 60 KHz > carrier was a bit trickier. > > e.g. I picked up a few Hummer Atomic Clocks on clearance for about $5. > The radio is a seperate unit that sends out the time code continuously. > The clock takes about 5 minutes to lock in. I'm about 680 miles from > Fort Collins. > > Ed > > Brooke Clarke wrote: >> Hi Tom: >> >> I've looked at dozens of Sony "Dream Machine" clocks on eBay and none >> of them have WWVB capability. Some are called preset or self >> setting, but these may just be using a button cell back up clock. >> >> What is the model number of your Sony WWVB clock? >> >> Have Fun, >> >> Brooke Clarke >> http://www.prc68.com >> >> Mike Monett wrote: >> >>> Update on my previous post - tvb has a nice page on WWVB receivers at >>> >>> http://leapsecond.com/pages/sony-wwvb/ >>> >>> Looks like these receivers might make good wall clocks. >>> >>> But for checking GPSDO's - not so good:) >>> >>> Regards, >>> >>> Mike >>> > > From ed_palmer at sasktel.net Mon Feb 23 21:41:24 2009 From: ed_palmer at sasktel.net (Ed Palmer) Date: Mon, 23 Feb 2009 15:41:24 -0600 Subject: [time-nuts] WWVB Wall and Travel Clocks On Sale In-Reply-To: <49A30324.7040005@pacific.net> References: <49A2E4C7.1040700@pacific.net> <49A2FBD9.5000908@sasktel.net> <49A30324.7040005@pacific.net> Message-ID: <49A31804.4080609@sasktel.net> I see. The Hummer clock has a single xtal circuit. When I tried to monitor it with a scope just now, even in A-B mode with x10 probes, the data was trash and the clock gives up after about 5 minutes. But if I leave it alone, it only takes a few minutes to lock in. Same with a couple of other cheap clocks. I remember that later in the evening when the signal is stronger, I could monitor it with an analog scope with no problem. You could decode it just by looking at it. Ed Brooke Clarke wrote: > Hi Ed: > > Yes, the one shown on Tom's web page: > http://leapsecond.com/pages/sony-wwvb/ > uses two 60 kHz crystals in the filter rather than the single crystal > like in the C-max CMMR-6P-60 (Digi-Key p/n: 561-1014-ND) that uses a > single crystal. C-Max is the new name for Temic that made the chip in > Tom's clock. > > http://www.prc68.com/I/Loop.shtml#CMMR6P60 > > I connected the CMMR-6P-60 to a couple of AA batteries and it's been > blinking for the last month or two. But that data is very noisy. > > Have Fun, > > Brooke Clarke > http://www.prc68.com > > Ed Palmer wrote: >> Hi Brooke, >> >> Is there something special about the Sony clock that you want? >> Before I found out about Time Nuts & GPSDOs I was looking at using >> one of these "atomic clocks" as a source of time and /or frequency. >> Most of them have a point where you can get to the time code. >> Getting to the 60 KHz carrier was a bit trickier. >> >> e.g. I picked up a few Hummer Atomic Clocks on clearance for about >> $5. The radio is a seperate unit that sends out the time code >> continuously. The clock takes about 5 minutes to lock in. I'm about >> 680 miles from Fort Collins. >> >> Ed >> >> Brooke Clarke wrote: >>> Hi Tom: >>> >>> I've looked at dozens of Sony "Dream Machine" clocks on eBay and >>> none of them have WWVB capability. Some are called preset or self >>> setting, but these may just be using a button cell back up clock. >>> >>> What is the model number of your Sony WWVB clock? >>> >>> Have Fun, >>> >>> Brooke Clarke >>> http://www.prc68.com >>> >>> Mike Monett wrote: >>> >>>> Update on my previous post - tvb has a nice page on WWVB receivers at >>>> >>>> http://leapsecond.com/pages/sony-wwvb/ >>>> >>>> Looks like these receivers might make good wall clocks. >>>> >>>> But for checking GPSDO's - not so good:) >>>> >>>> Regards, >>>> >>>> Mike >>>> >> >> > From jmiles at pop.net Tue Feb 24 00:59:41 2009 From: jmiles at pop.net (John Miles) Date: Mon, 23 Feb 2009 16:59:41 -0800 Subject: [time-nuts] Never mind Thunderbolts... Message-ID: Can TAPR put together a group buy on these? http://www.teachspin.com/instruments/optical_pumping/index.shtml -- john, KE5FX From johnday at wordsnimages.com Tue Feb 24 02:22:30 2009 From: johnday at wordsnimages.com (John Day) Date: Mon, 23 Feb 2009 21:22:30 -0500 Subject: [time-nuts] Never mind Thunderbolts... In-Reply-To: References: Message-ID: At 07:59 PM 2/23/2009, you wrote: >Can TAPR put together a group buy on these? >http://www.teachspin.com/instruments/optical_pumping/index.shtml > >-- john, KE5FX An absolute bargain at just $16,000! Wonderful, do I get a bulk deal for two? LOL John >_______________________________________________ >time-nuts mailing list -- time-nuts at febo.com >To unsubscribe, go to https://www.febo.com/cgi-bin/mailman/listinfo/time-nuts >and follow the instructions there. From bill at iaxs.net Tue Feb 24 02:34:52 2009 From: bill at iaxs.net (Bill Hawkins) Date: Mon, 23 Feb 2009 20:34:52 -0600 Subject: [time-nuts] Never mind Thunderbolts... In-Reply-To: References: Message-ID: <3F7A9F59E1F344E0BA17E9B796A3482D@cyrus> Well then, how about kits, or DIY instructions? Something in "The Boy Electrician?" There's only so much you can learn with eBay bargains and deviation plots. Bill Hawkins -----Original Message----- From: John Day Sent: Monday, February 23, 2009 8:23 PM At 07:59 PM 2/23/2009, you wrote: >Can TAPR put together a group buy on these? >http://www.teachspin.com/instruments/optical_pumping/index.shtml > >-- john, KE5FX An absolute bargain at just $16,000! Wonderful, do I get a bulk deal for two? LOL John From bruce.griffiths at xtra.co.nz Tue Feb 24 02:53:06 2009 From: bruce.griffiths at xtra.co.nz (Bruce Griffiths) Date: Tue, 24 Feb 2009 15:53:06 +1300 Subject: [time-nuts] Never mind Thunderbolts... In-Reply-To: <3F7A9F59E1F344E0BA17E9B796A3482D@cyrus> References: <3F7A9F59E1F344E0BA17E9B796A3482D@cyrus> Message-ID: <49A36112.2060009@xtra.co.nz> Bill Hawkins wrote: > Well then, how about kits, or DIY instructions? Something in "The Boy > Electrician?" > > There's only so much you can learn with eBay bargains and deviation > plots. > > Bill Hawkins > > > -----Original Message----- > From: John Day > Sent: Monday, February 23, 2009 8:23 PM > > At 07:59 PM 2/23/2009, you wrote: > >> Can TAPR put together a group buy on these? >> http://www.teachspin.com/instruments/optical_pumping/index.shtml >> >> -- john, KE5FX >> > > An absolute bargain at just $16,000! Wonderful, do I get a bulk deal for > two? > > LOL > > John > > > > _______________________________________________ > time-nuts mailing list -- time-nuts at febo.com > To unsubscribe, go to https://www.febo.com/cgi-bin/mailman/listinfo/time-nuts > and follow the instructions there. > > Bill The setup is reasonably straight forward to duplicate. However a pair of Helmholtz coils isnt optimum. 3 pairs would be better as it may not be convenient/possible to align the apparatus with the Earth's magnetic field, There are coil arrangements that produce more uniform fields than Helmholtz coils. It may be even more convenient to build ones own 795nm ECDL to replace the lamp. Bruce From wb6bnq at cox.net Tue Feb 24 03:09:55 2009 From: wb6bnq at cox.net (WB6BNQ) Date: Mon, 23 Feb 2009 19:09:55 -0800 Subject: [time-nuts] Never mind Thunderbolts... References: <3F7A9F59E1F344E0BA17E9B796A3482D@cyrus> <49A36112.2060009@xtra.co.nz> Message-ID: <49A36503.A3F496DA@cox.net> Why do all that when the entire instrument is all ready available for cheap ? Any Rubidium frequency oscillator from Efratrom, Frequency electronics, Symmetricom, or the old Datum units, etc., off ebay for $200 or less is essentially the same thing as this highly over priced school lab device. If you carefully disassemble RB frequency device you will have all the same components, albeit on a smaller scale. Just add a magnifying glass and you got it ! ! ! Bill....WB6BNQ From bruce.griffiths at xtra.co.nz Tue Feb 24 03:20:03 2009 From: bruce.griffiths at xtra.co.nz (Bruce Griffiths) Date: Tue, 24 Feb 2009 16:20:03 +1300 Subject: [time-nuts] Never mind Thunderbolts... In-Reply-To: <49A36503.A3F496DA@cox.net> References: <3F7A9F59E1F344E0BA17E9B796A3482D@cyrus> <49A36112.2060009@xtra.co.nz> <49A36503.A3F496DA@cox.net> Message-ID: <49A36763.4050201@xtra.co.nz> WB6BNQ wrote: > Why do all that when the entire instrument is all ready available for cheap ? > > Any Rubidium frequency oscillator from Efratrom, Frequency electronics, > Symmetricom, or the old Datum units, etc., off ebay for $200 or less is essentially > the same thing as this highly over priced school lab device. > > If you carefully disassemble RB frequency device you will have all the same > components, albeit on a smaller scale. Just add a magnifying glass and you got it > ! ! ! > > Bill....WB6BNQ > > > To achieve a higher SNR (and hence smaller Allan deviation) and allow testing of other optically pumped alkali vapour standards? Bruce From dforbes at dakotacom.net Tue Feb 24 17:34:49 2009 From: dforbes at dakotacom.net (David Forbes) Date: Tue, 24 Feb 2009 10:34:49 -0700 Subject: [time-nuts] Never mind Thunderbolts... In-Reply-To: <49A36112.2060009@xtra.co.nz> References: <3F7A9F59E1F344E0BA17E9B796A3482D@cyrus> <49A36112.2060009@xtra.co.nz> Message-ID: <49A42FB9.4000802@dakotacom.net> Bruce Griffiths wrote: > > The setup is reasonably straight forward to duplicate. > However a pair of Helmholtz coils isnt optimum. > 3 pairs would be better as it may not be convenient/possible to align > the apparatus with the Earth's magnetic field, > > There are coil arrangements that produce more uniform fields than > Helmholtz coils. > > It may be even more convenient to build ones own 795nm ECDL to replace > the lamp. > > Bruce I look at this thing as a means of demonstrating the effects rather than as an actual timekeeping instrument. After all, it's aimed at student use. But it does lead one to dream about building a real Rb vapor cell timekeeping device that's all exposed. Sure, the vapor cell would have to be temperature stabilized, and various other improvements would have to be made, and it wouldn't work really well. But think of the geegaw appeal! --David F From mkern at fastmail.fm Tue Feb 24 18:48:50 2009 From: mkern at fastmail.fm (Markus Kern) Date: Tue, 24 Feb 2009 19:48:50 +0100 Subject: [time-nuts] Different Thunderbolt versions In-Reply-To: <20090222173317.BBA46BCDC@ip-64-139-1-69.sjc.megapath.net> References: Message from Markus Kern of "Sun, 22 Feb 2009 17:32:05 +0100." <12619539593.20090222173205@fastmail.fm> <20090222173317.BBA46BCDC@ip-64-139-1-69.sjc.megapath.net> Message-ID: <13819286937.20090224194850@fastmail.fm> On 22.02.2009, 18:33 Hal Murray wrote: >> I actually want to work at frequencies below 50Mhz where a 50 degree >> phase error corresponds to about 3ns. > What is interesting to listen to below 50 MHz? Are there any handy strong > signals you can use for debugging/calibration? There are several bright sources which should be relatively easy to observe: Cygnus A, Virgo A, Centaurus A, Taurus A, Fornax A and of course the sun. >> Also at these low frequencies the required accuracy approaches that >> claimed by some GPS receivers (4ns for the M12+ ?) so might it be >> possible to build a system which is accurate for arbitrary integration >> times? > If you haven't seen it yet, this is wonderful background: > http://www.gpstime.com/files/PTTI/PTTI_2006.pdf > In particular, be sure to check out the hanging bridges. > How far apart are your antennas? Things might get a lot simpler if you can > see the same GPS satellites. I haven't decided on a fixed baseline. I plan to use a semi-mobile station to increase the baseline once it works at closer distances. So using the same satellite might be possible but I guess there are additional complications of synchronizing the handover in that case. > If your runs are short enough, you might get better results by letting the > oscillator coast during the run. That is unplug the antennas or turn the > feedback off in software. Yes, that might be something worth trying. > This could be all wrong.... > There are two things you need. One is frequency, the other is time (PPS). > If the time is slightly off, that's equivalent to pointing your antenna in a > different direction. You get the same sort of error if the location of an > antenna is slightly off. > I think the Earth's rotation introduces an error that's equivalent to the > frequency being slightly off. I think there are two requirements. One is having the sampling synchronized between stations (I want to use direct sampling without any analog mixing) to be able to correlate the signals at all. The other is tagging the samples with the current time in order to keep the search space for the correlator sufficiently low. I believe the later can be easily achieved by synchronously sampling the pps signal together with the data. Markus From mkern at fastmail.fm Tue Feb 24 19:29:48 2009 From: mkern at fastmail.fm (Markus Kern) Date: Tue, 24 Feb 2009 20:29:48 +0100 Subject: [time-nuts] Different Thunderbolt versions In-Reply-To: <49A1B19E.9020908@xtra.co.nz> References: <19210986468.20090221152324@fastmail.fm> <49A07081.1040204@xtra.co.nz> <6C8E7ADA880040799C1477C5C1C32AD5@pc52> <49A0A841.8000702@xtra.co.nz> <12619539593.20090222173205@fastmail.fm> <49A1B19E.9020908@xtra.co.nz> Message-ID: <10121744609.20090224202948@fastmail.fm> On 22.02.2009, 21:12 Bruce Griffiths wrote: > Markus > Even with sawtooth correction the performance of the M12+T was found > inadequate for the LOFAR array. > They use SRS FS725 rubidium sources disciplined by M12+T GPS timing > receivers. I didn't mean using the M12 by itself, obviously a clock stable enough over the time the M12 pps must be integrated has to be used. If we are using the ADEV limits you proposed then at 50 MHz (= 3ns acceptable error) the timing requirement is an ADEV of 3*1E-(8+x) at tau = x seconds. From the measurements at http://www.leapsecond.com/pages/gpsdo/ it seems the Thunderbolt gets pretty close to that. The LOFAR clock system is described at [1]. In section 3.1.3.3 they say: "Some Crystal Oscillators have the advantage that they have a better Allan variance for periods of up to 10s and therefore it can be claimed that they have a better performance than the SRS-FS725 Rb-reference standard. The performance for time periods above 10s, the SRS-FS725 performs better. Therefore choosing an OCXO would require a maximum calibration interval of 10s and it would require a significantly better GPS (or GALILEO) receiver because de Rb-reference is used to average the PPS signal from the GPS receiver thereby making it possible to identify the time difference between stations at receive frequencies above 10MHz." I think this means that they are using pps integration times above 10 seconds. I couldn't find any reference to the actual value though. LOFAR is also working at frequencies up to 240MHz so the timing requirements are definitely higher. They say that a station time offset of 200ps does not affect performance as long as it remains stable over time. > They also state that the ionosphere contribution to ADEV is about 8E-12 > @10s. Yes, from which they infer that "the reference clock shall have an Allan variance of 1e-11 or less over 10s." I am not sure if this has to do with the propagation of the GPS signal or if they mean that they need a clock stable enough to later compensate for the different delays of the observed signal through the ionosphere. Markus [1] http://www.lofar.org/operations/lib/exe/fetch.php?id=public%3Adocuments%3Alofar_documents&cache=cache&media=public:documents:19_detailed_description_of_clock_sync.pdf From bruce.griffiths at xtra.co.nz Tue Feb 24 20:37:58 2009 From: bruce.griffiths at xtra.co.nz (Bruce Griffiths) Date: Wed, 25 Feb 2009 09:37:58 +1300 Subject: [time-nuts] Different Thunderbolt versions In-Reply-To: <10121744609.20090224202948@fastmail.fm> References: <19210986468.20090221152324@fastmail.fm> <49A07081.1040204@xtra.co.nz> <6C8E7ADA880040799C1477C5C1C32AD5@pc52> <49A0A841.8000702@xtra.co.nz> <12619539593.20090222173205@fastmail.fm> <49A1B19E.9020908@xtra.co.nz> <10121744609.20090224202948@fastmail.fm> Message-ID: <49A45AA6.30304@xtra.co.nz> Markus Kern wrote: > On 22.02.2009, 21:12 Bruce Griffiths wrote: > > >> Markus >> > > >> Even with sawtooth correction the performance of the M12+T was found >> inadequate for the LOFAR array. >> They use SRS FS725 rubidium sources disciplined by M12+T GPS timing >> receivers. >> > > I didn't mean using the M12 by itself, obviously a clock stable enough > over the time the M12 pps must be integrated has to be used. > > If we are using the ADEV limits you proposed then at 50 MHz (= 3ns > acceptable error) the timing requirement is an ADEV of 3*1E-(8+x) at > tau = x seconds. From the measurements at > http://www.leapsecond.com/pages/gpsdo/ it seems the Thunderbolt gets > pretty close to that. > > The LOFAR clock system is described at [1]. In section 3.1.3.3 they > say: > > "Some Crystal Oscillators have the advantage that they have a better > Allan variance for periods of up to 10s and therefore it can be claimed > that they have a better performance than the SRS-FS725 Rb-reference > standard. The performance for time periods above 10s, the SRS-FS725 > performs better. Therefore choosing an OCXO would require a maximum > calibration interval of 10s and it would require a significantly better > GPS (or GALILEO) receiver because de Rb-reference is used to average > the PPS signal from the GPS receiver thereby making it possible to > identify the time difference between stations at receive frequencies > above 10MHz." > > I think this means that they are using pps integration times above 10 > seconds. I couldn't find any reference to the actual value though. > > LOFAR is also working at frequencies up to 240MHz so the timing > requirements are definitely higher. They say that a station time > offset of 200ps does not affect performance as long as it remains > stable over time. > > >> They also state that the ionosphere contribution to ADEV is about 8E-12 >> @10s. >> > > Yes, from which they infer that "the reference clock shall have an > Allan variance of 1e-11 or less over 10s." I am not sure if this has > to do with the propagation of the GPS signal or if they mean that they > need a clock stable enough to later compensate for the different > delays of the observed signal through the ionosphere. > > Markus > > [1] http://www.lofar.org/operations/lib/exe/fetch.php?id=public%3Adocuments%3Alofar_documents&cache=cache&media=public:documents:19_detailed_description_of_clock_sync.pdf > > > Markus The ionosphere contribution to the Allan deviation at GPS frequencies is much smaller (by a factor of 10-100 or so) than that, as is evident from carrier phase measurements. At 50MHz the ionospheric phase shift, dispersion and instability will be much greater than at GPS frequencies. They are merely ensuring that the LO contribution to Allan deviation is much smaller than that of the ionosphere. If you look at the Allan deviation plot on the PRS10 page: http://www.thinksrs.com/products/PRS10.htm This indicates that the likely disciplining loop time constant will be several thousand seconds. Close isn't good enough: the phase differences between pairs of stations is significant, the Allan deviation needs to be at least 30% lower per station. If the errors at station pairs have significant correlation the requirement can be relaxed somewhat. Bruce From gatoalfa at gmail.com Wed Feb 25 06:48:53 2009 From: gatoalfa at gmail.com (German Alvarez) Date: Tue, 24 Feb 2009 22:48:53 -0800 Subject: [time-nuts] Tektronix 603/604, looking for any documentation Message-ID: <49A4E9D5.1010407@gmail.com> To use with my Dutchtronix AVR Oscilloscope Clock I just got a Tektronix 604 X&Y&Z display. It is working fine but I would like to get any type of documentation, in particular it has a DB-25 on the back that I want to explore. Thanks in advance for any help about this. --ga From david.partridge at dsl.pipex.com Wed Feb 25 07:23:48 2009 From: david.partridge at dsl.pipex.com (David C. Partridge) Date: Wed, 25 Feb 2009 07:23:48 -0000 Subject: [time-nuts] Tektronix 603/604, looking for any documentation In-Reply-To: <49A4E9D5.1010407@gmail.com> References: <49A4E9D5.1010407@gmail.com> Message-ID: <61ABD6CFB3904C29B55FE6DD65B2EE17@APOLLO> Aretk Media : http://artekmed.startlogic.com/page2.html Dave -----Original Message----- From: time-nuts-bounces at febo.com [mailto:time-nuts-bounces at febo.com] On Behalf Of German Alvarez Sent: 25 February 2009 06:49 To: time-nuts at febo.com Subject: [time-nuts] Tektronix 603/604, looking for any documentation To use with my Dutchtronix AVR Oscilloscope Clock I just got a Tektronix 604 X&Y&Z display. It is working fine but I would like to get any type of documentation, in particular it has a DB-25 on the back that I want to explore. Thanks in advance for any help about this. --ga _______________________________________________ time-nuts mailing list -- time-nuts at febo.com To unsubscribe, go to https://www.febo.com/cgi-bin/mailman/listinfo/time-nuts and follow the instructions there. From df6jb at ulrich-bangert.de Wed Feb 25 08:24:51 2009 From: df6jb at ulrich-bangert.de (Ulrich Bangert) Date: Wed, 25 Feb 2009 09:24:51 +0100 Subject: [time-nuts] Different Thunderbolt versions In-Reply-To: <49A45AA6.30304@xtra.co.nz> Message-ID: Bruce, > If you look at the Allan deviation plot on the PRS10 page: > http://www.thinksrs.com/products/PRS10.htm > > This indicates that the likely disciplining loop time > constant will be several thousand seconds. It should be noted that this plot shows a VERY PESSIMISTIC (almost two decades) behaviour of the GPS 1 pps, perhaps a remainder from SA and/or a not so well timing receiver. A M12+ would have its sawtooth corrected 1 pps at 2-4E-12 @ 1000 s, giving a crosspoint to the AD of the rubidium alone anywhere between 1000s and 10000 s. Best regards Ulrich > -----Ursprungliche Nachricht----- > Von: time-nuts-bounces at febo.com > [mailto:time-nuts-bounces at febo.com] Im Auftrag von Bruce Griffiths > Gesendet: Dienstag, 24. Februar 2009 21:38 > An: Markus Kern; Discussion of precise time and frequency measurement > Betreff: Re: [time-nuts] Different Thunderbolt versions > > > > Markus Kern wrote: > > On 22.02.2009, 21:12 Bruce Griffiths > > wrote: > > > > > >> Markus > >> > > > > > >> Even with sawtooth correction the performance of the M12+T > was found > >> inadequate for the LOFAR > array. > >> They use SRS FS725 rubidium sources disciplined by M12+T > GPS timing > >> receivers. > >> > > > > I didn't mean using the M12 by itself, obviously a clock > stable enough > > over the time the M12 pps must be integrated has to be used. > > > > If we are using the ADEV limits you proposed then at 50 MHz (= 3ns > > acceptable error) the timing requirement is an ADEV of > 3*1E-(8+x) at > > tau = x seconds. From the measurements at > > http://www.leapsecond.com/pages/gpsdo/ it seems the > Thunderbolt gets > > pretty close to that. > > > > The LOFAR clock system is described at [1]. In section 3.1.3.3 they > > say: > > > > "Some Crystal Oscillators have the advantage that they have > a better > > Allan variance for periods of up to 10s and therefore it can be > > claimed that they have a better performance than the SRS-FS725 > > Rb-reference standard. The performance for time periods > above 10s, the > > SRS-FS725 performs better. Therefore choosing an OCXO would > require a > > maximum calibration interval of 10s and it would require a > > significantly better GPS (or GALILEO) receiver because de > Rb-reference > > is used to average the PPS signal from the GPS receiver > thereby making > > it possible to identify the time difference between stations at > > receive frequencies above 10MHz." > > > > I think this means that they are using pps integration > times above 10 > > seconds. I couldn't find any reference to the actual value though. > > > > LOFAR is also working at frequencies up to 240MHz so the timing > > requirements are definitely higher. They say that a station time > > offset of 200ps does not affect performance as long as it remains > > stable over time. > > > > > >> They also state that the ionosphere contribution to ADEV is about > >> 8E-12 @10s. > >> > > > > Yes, from which they infer that "the reference clock shall have an > > Allan variance of 1e-11 or less over 10s." I am not sure if > this has > > to do with the propagation of the GPS signal or if they > mean that they > > need a clock stable enough to later compensate for the different > > delays of the observed signal through the ionosphere. > > > > Markus > > > > [1] > > > http://www.lofar.org/operations/lib/exe/fetch.php?id=public%3Adocument > > > s%3Alofar_documents&cache=cache&media=public:documents:19_detailed_des > > cription_of_clock_sync.pdf > > > > > > > > Markus > > The ionosphere contribution to the Allan deviation at GPS > frequencies is much smaller (by a factor of 10-100 or so) > than that, as is evident from carrier phase measurements. At > 50MHz the ionospheric phase shift, dispersion and instability > will be much greater than at GPS frequencies. They are merely > ensuring that the LO contribution to Allan deviation is much > smaller than that of the ionosphere. > > If you look at the Allan deviation plot on the PRS10 page: > http://www.thinksrs.com/products/PRS10.htm > > This indicates that the likely disciplining loop time > constant will be several thousand seconds. > > Close isn't good enough: the phase differences between pairs > of stations is significant, the Allan deviation needs to be > at least 30% lower per station. If the errors at station > pairs have significant correlation the requirement can be > relaxed somewhat. > > Bruce > > _______________________________________________ > time-nuts mailing list -- time-nuts at febo.com > To unsubscribe, go to > https://www.febo.com/cgi-bin/mailman/listinfo/time-nuts > and follow the instructions there. From dforbes at dakotacom.net Wed Feb 25 16:16:14 2009 From: dforbes at dakotacom.net (David Forbes) Date: Wed, 25 Feb 2009 09:16:14 -0700 Subject: [time-nuts] Tektronix 603/604, looking for any documentation In-Reply-To: <49A4E9D5.1010407@gmail.com> References: <49A4E9D5.1010407@gmail.com> Message-ID: At 10:48 PM -0800 2/24/09, German Alvarez wrote: >To use with my Dutchtronix AVR Oscilloscope Clock I just got a Tektronix >604 X&Y&Z display. It is working fine but I would like to get any type >of documentation, in particular it has a DB-25 on the back that I want >to explore. > >Thanks in advance for any help about this. > >--ga German, I have a 603 manual at the office. I could mail it to you if you wanted to pay postage, or I could scan a few pages of interest. -- --David Forbes, Tucson, AZ http://www.cathodecorner.com/ From had at to-way.com Wed Feb 25 17:57:28 2009 From: had at to-way.com (Had) Date: Wed, 25 Feb 2009 09:57:28 -0800 Subject: [time-nuts] Oscillator Article Message-ID: <20090225175733.C45ABE04815@mail-in03.adhost.com> Nice article on oscillators in the Feb. 19th issue of EDN Had From rdarlington at gmail.com Wed Feb 25 18:16:25 2009 From: rdarlington at gmail.com (Robert Darlington) Date: Wed, 25 Feb 2009 11:16:25 -0700 Subject: [time-nuts] Oscillator Article In-Reply-To: <20090225175733.C45ABE04815@mail-in03.adhost.com> References: <20090225175733.C45ABE04815@mail-in03.adhost.com> Message-ID: Had, correct me if I'm wrong, but I think the article you are talking about is this one: http://www.edn.com/article/CA6636507.html?text=oscillators I don't have my copy sitting around but found this article with the same data and subject on the website. -Bob On Wed, Feb 25, 2009 at 10:57 AM, Had wrote: > > Nice article on oscillators in the Feb. 19th issue of EDN > > Had > > > _______________________________________________ > time-nuts mailing list -- time-nuts at febo.com > To unsubscribe, go to > https://www.febo.com/cgi-bin/mailman/listinfo/time-nuts > and follow the instructions there. > From tvb at LeapSecond.com Wed Feb 25 18:29:32 2009 From: tvb at LeapSecond.com (Tom Van Baak) Date: Wed, 25 Feb 2009 10:29:32 -0800 Subject: [time-nuts] Oscillator Article References: <20090225175733.C45ABE04815@mail-in03.adhost.com> Message-ID: http://www.edn.com/contents/images/6636507.pdf From SAIDJACK at aol.com Wed Feb 25 19:14:41 2009 From: SAIDJACK at aol.com (SAIDJACK at aol.com) Date: Wed, 25 Feb 2009 14:14:41 EST Subject: [time-nuts] Oscillator Article Message-ID: Good stuff coming down the pipe, I looked at the Mems/Si technology and the phase noise, thermal stability, and jitter specs are orders of magnitude higher than what we would like to see here on Time Nuts. I think Quartz is going to be the king of the hill for some time to come due to it's Q (>1E+06)... Having programmability and vibration insensitivity is nice though, and for most high-volume applications these parts will do just fine and reduce cost, sensitivity, and size. Then again for high-volume consumer electronics you always want to have an alternate vendor, and these are not entrenched and proven companies yet.. What get's me very excited is the fact that CSACs may be coming down the pipe soon. bye, Said In a message dated 2/25/2009 10:30:33 Pacific Standard Time, tvb at LeapSecond.com writes: http://www.edn.com/contents/images/6636507.pdf From SAIDJACK at aol.com Wed Feb 25 19:23:27 2009 From: SAIDJACK at aol.com (SAIDJACK at aol.com) Date: Wed, 25 Feb 2009 14:23:27 EST Subject: [time-nuts] Different Thunderbolt versions Message-ID: Hi Ulrich, notice that that plot also conveniently cuts-off the ADEV below 10s, where we can see it go up substantially at around 10s... Guess they did not want anyone seeing their <10Hz performance; this is called good specmanship. bye, Said In a message dated 2/25/2009 00:25:27 Pacific Standard Time, df6jb at ulrich-bangert.de writes: It should be noted that this plot shows a VERY PESSIMISTIC (almost two decades) behaviour of the GPS 1 pps, perhaps a remainder from SA and/or a not so well timing receiver. A M12+ would have its sawtooth corrected 1 pps at 2-4E-12 @ 1000 s, giving a crosspoint to the AD of the rubidium alone anywhere between 1000s and 10000 s. Best regards Ulrich From had at to-way.com Wed Feb 25 20:14:01 2009 From: had at to-way.com (Had) Date: Wed, 25 Feb 2009 12:14:01 -0800 Subject: [time-nuts] Oscillator Article In-Reply-To: References: <20090225175733.C45ABE04815@mail-in03.adhost.com> Message-ID: <20090225201405.982942D74CD@mail-in01.adhost.com> Yup, that is it. I should have put the title in the original post. Had At 10:16 AM 2/25/2009, you wrote: >Had, correct me if I'm wrong, but I think the article you are talking about >is this one: > >http://www.edn.com/article/CA6636507.html?text=oscillators > >I don't have my copy sitting around but found this article with the same >data and subject on the website. From bruce.griffiths at xtra.co.nz Wed Feb 25 20:29:22 2009 From: bruce.griffiths at xtra.co.nz (Bruce Griffiths) Date: Thu, 26 Feb 2009 09:29:22 +1300 Subject: [time-nuts] Different Thunderbolt versions In-Reply-To: References: Message-ID: <49A5AA22.9080008@xtra.co.nz> Ulrich Another clue as to the likely disciplining loop time constant can be found on page 6 of the FS725 manual: http://www.thinksrs.com/downloads/PDFs/Manuals/FS725m.pdf Where it states that the available range for the loop time constant is 480 sec to 64800 sec Bruce Ulrich Bangert wrote: > Bruce, > > >> If you look at the Allan deviation plot on the PRS10 page: >> http://www.thinksrs.com/products/PRS10.htm >> >> This indicates that the likely disciplining loop time >> constant will be several thousand seconds. >> > > It should be noted that this plot shows a VERY PESSIMISTIC (almost two > decades) behaviour of the GPS 1 pps, perhaps a remainder from SA and/or a > not so well timing receiver. > > A M12+ would have its sawtooth corrected 1 pps at 2-4E-12 @ 1000 s, giving a > crosspoint to the AD of the rubidium alone anywhere between 1000s and 10000 > s. > > Best regards > Ulrich > > > >> -----Ursprungliche Nachricht----- >> Von: time-nuts-bounces at febo.com >> [mailto:time-nuts-bounces at febo.com] Im Auftrag von Bruce Griffiths >> Gesendet: Dienstag, 24. Februar 2009 21:38 >> An: Markus Kern; Discussion of precise time and frequency measurement >> Betreff: Re: [time-nuts] Different Thunderbolt versions >> >> >> >> Markus Kern wrote: >> >>> On 22.02.2009, 21:12 Bruce Griffiths >>> wrote: >>> >>> >>> >>>> Markus >>>> >>>> >>> >>> >>>> Even with sawtooth correction the performance of the M12+T >>>> >> was found >> >>>> inadequate for the LOFAR >>>> >> array. >> >>>> They use SRS FS725 rubidium sources disciplined by M12+T >>>> >> GPS timing >> >>>> receivers. >>>> >>>> >>> I didn't mean using the M12 by itself, obviously a clock >>> >> stable enough >> >>> over the time the M12 pps must be integrated has to be used. >>> >>> If we are using the ADEV limits you proposed then at 50 MHz (= 3ns >>> acceptable error) the timing requirement is an ADEV of >>> >> 3*1E-(8+x) at >> >>> tau = x seconds. From the measurements at >>> http://www.leapsecond.com/pages/gpsdo/ it seems the >>> >> Thunderbolt gets >> >>> pretty close to that. >>> >>> The LOFAR clock system is described at [1]. In section 3.1.3.3 they >>> say: >>> >>> "Some Crystal Oscillators have the advantage that they have >>> >> a better >> >>> Allan variance for periods of up to 10s and therefore it can be >>> claimed that they have a better performance than the SRS-FS725 >>> Rb-reference standard. The performance for time periods >>> >> above 10s, the >> >>> SRS-FS725 performs better. Therefore choosing an OCXO would >>> >> require a >> >>> maximum calibration interval of 10s and it would require a >>> significantly better GPS (or GALILEO) receiver because de >>> >> Rb-reference >> >>> is used to average the PPS signal from the GPS receiver >>> >> thereby making >> >>> it possible to identify the time difference between stations at >>> receive frequencies above 10MHz." >>> >>> I think this means that they are using pps integration >>> >> times above 10 >> >>> seconds. I couldn't find any reference to the actual value though. >>> >>> LOFAR is also working at frequencies up to 240MHz so the timing >>> requirements are definitely higher. They say that a station time >>> offset of 200ps does not affect performance as long as it remains >>> stable over time. >>> >>> >>> >>>> They also state that the ionosphere contribution to ADEV is about >>>> 8E-12 @10s. >>>> >>>> >>> Yes, from which they infer that "the reference clock shall have an >>> Allan variance of 1e-11 or less over 10s." I am not sure if >>> >> this has >> >>> to do with the propagation of the GPS signal or if they >>> >> mean that they >> >>> need a clock stable enough to later compensate for the different >>> delays of the observed signal through the ionosphere. >>> >>> Markus >>> >>> [1] >>> >>> >> http://www.lofar.org/operations/lib/exe/fetch.php?id=public%3Adocument >> >> s%3Alofar_documents&cache=cache&media=public:documents:19_detailed_des >> >>> cription_of_clock_sync.pdf >>> >>> >>> >>> >> Markus >> >> The ionosphere contribution to the Allan deviation at GPS >> frequencies is much smaller (by a factor of 10-100 or so) >> than that, as is evident from carrier phase measurements. At >> 50MHz the ionospheric phase shift, dispersion and instability >> will be much greater than at GPS frequencies. They are merely >> ensuring that the LO contribution to Allan deviation is much >> smaller than that of the ionosphere. >> >> If you look at the Allan deviation plot on the PRS10 page: >> http://www.thinksrs.com/products/PRS10.htm >> >> This indicates that the likely disciplining loop time >> constant will be several thousand seconds. >> >> Close isn't good enough: the phase differences between pairs >> of stations is significant, the Allan deviation needs to be >> at least 30% lower per station. If the errors at station >> pairs have significant correlation the requirement can be >> relaxed somewhat. >> >> Bruce >> >> From bruce.griffiths at xtra.co.nz Wed Feb 25 20:32:40 2009 From: bruce.griffiths at xtra.co.nz (Bruce Griffiths) Date: Thu, 26 Feb 2009 09:32:40 +1300 Subject: [time-nuts] Different Thunderbolt versions In-Reply-To: References: Message-ID: <49A5AAE8.4010007@xtra.co.nz> Said Specified Allan deviation is 2E-11 @1s. Bruce SAIDJACK at aol.com wrote: > Hi Ulrich, > > notice that that plot also conveniently cuts-off the ADEV below 10s, where > we can see it go up substantially at around 10s... > > Guess they did not want anyone seeing their <10Hz performance; this is > called good specmanship. > > bye, > Said > > > In a message dated 2/25/2009 00:25:27 Pacific Standard Time, > df6jb at ulrich-bangert.de writes: > > It should be noted that this plot shows a VERY PESSIMISTIC (almost two > decades) behaviour of the GPS 1 pps, perhaps a remainder from SA and/or a > not so well timing receiver. > > A M12+ would have its sawtooth corrected 1 pps at 2-4E-12 @ 1000 s, giving a > crosspoint to the AD of the rubidium alone anywhere between 1000s and 10000 > s. > > Best regards > Ulrich > > _______________________________________________ > time-nuts mailing list -- time-nuts at febo.com > To unsubscribe, go to https://www.febo.com/cgi-bin/mailman/listinfo/time-nuts > and follow the instructions there. > > From holrum at hotmail.com Wed Feb 25 21:26:00 2009 From: holrum at hotmail.com (Mark Sims) Date: Wed, 25 Feb 2009 21:26:00 +0000 Subject: [time-nuts] Oscillator Article In-Reply-To: References: Message-ID: Dallas Semi has a cute little programmable dual oscillator chip, the DS1077. 8KHz to 133 Mhz in a SOIC-8 package. Available from Sparkfun for 3 bucks a pop: http://www.sparkfun.com/commerce/product_info.php?products_id=9089 Definitely not Nut grade, but useful... too bad it doesn't have an M/N or PLL based freq divider. ---------------------------------------- _________________________________________________________________ It?s the same Hotmail?. If by ?same? you mean up to 70% faster. http://windowslive.com/online/hotmail?ocid=TXT_TAGLM_WL_HM_AE_Same_022009 From rdarlington at gmail.com Wed Feb 25 21:43:27 2009 From: rdarlington at gmail.com (Robert Darlington) Date: Wed, 25 Feb 2009 14:43:27 -0700 Subject: [time-nuts] Oscillator Article In-Reply-To: <20090225201405.982942D74CD@mail-in01.adhost.com> References: <20090225175733.C45ABE04815@mail-in03.adhost.com> <20090225201405.982942D74CD@mail-in01.adhost.com> Message-ID: TVB's version is better. PDF for the win. On Wed, Feb 25, 2009 at 1:14 PM, Had wrote: > > Yup, that is it. I should have put the title in the original post. > > Had > > > At 10:16 AM 2/25/2009, you wrote: > >Had, correct me if I'm wrong, but I think the article you are talking > about > >is this one: > > > >http://www.edn.com/article/CA6636507.html?text=oscillators > > > >I don't have my copy sitting around but found this article with the same > >data and subject on the website. > _______________________________________________ > time-nuts mailing list -- time-nuts at febo.com > To unsubscribe, go to > https://www.febo.com/cgi-bin/mailman/listinfo/time-nuts > and follow the instructions there. > From gatoalfa at gmail.com Wed Feb 25 22:35:19 2009 From: gatoalfa at gmail.com (German Alvarez) Date: Wed, 25 Feb 2009 14:35:19 -0800 Subject: [time-nuts] Tektronix 603/604, looking for any documentation In-Reply-To: References: Message-ID: <49A5C7A7.7040906@gmail.com> Thanks to everyone in the list. I purchased the operators & service manual from www.ArtekMedia.com and after some cleaning and minor adjustments, it is performing as it was 35 years ago. Hats off to the designers of the unit, it is incredible how nice it performs with the duchtronics clock. Now I have to convince my SO that it makes a beautifully mantle clock, specially since the 604 is 20" depth. --ga From SAIDJACK at aol.com Thu Feb 26 06:30:11 2009 From: SAIDJACK at aol.com (SAIDJACK at aol.com) Date: Thu, 26 Feb 2009 01:30:11 EST Subject: [time-nuts] Different Thunderbolt versions Message-ID: Hi Bruce, seems that would be in line with extending the 10s plot down to 1s in a linear fashion. Not that great. Tom's GPSDO measurements show some GPSDO's go down to ~E-013 at 1s. bye, Said In a message dated 2/25/2009 12:33:22 Pacific Standard Time, bruce.griffiths at xtra.co.nz writes: Said Specified Allan deviation is 2E-11 @1s. Bruce From bruce.griffiths at xtra.co.nz Thu Feb 26 07:04:16 2009 From: bruce.griffiths at xtra.co.nz (Bruce Griffiths) Date: Thu, 26 Feb 2009 20:04:16 +1300 Subject: [time-nuts] Different Thunderbolt versions In-Reply-To: References: Message-ID: <49A63EF0.6070907@xtra.co.nz> SAIDJACK at aol.com wrote: > Hi Bruce, > > seems that would be in line with extending the 10s plot down to 1s in a > linear fashion. > > Not that great. Tom's GPSDO measurements show some GPSDO's go down to ~E-013 > at 1s. > > bye, > Said > > > In a message dated 2/25/2009 12:33:22 Pacific Standard Time, > bruce.griffiths at xtra.co.nz writes: > > Said > > Specified Allan deviation is 2E-11 @1s. > > > Bruce > > That isnt a great advantage in this application given that the contribution of the ionosphere to the VLBI receiver output signal phase noise is around 8E-11 @ 10s at 50MHz. Typically in VLBI integration times op up to 10,000 seconds or more are used. Integration times shorter than 100 sec are not usually of much interest. Bruce From mkern at fastmail.fm Thu Feb 26 13:51:48 2009 From: mkern at fastmail.fm (Markus Kern) Date: Thu, 26 Feb 2009 14:51:48 +0100 Subject: [time-nuts] Different Thunderbolt versions In-Reply-To: <49A45AA6.30304@xtra.co.nz> References: <19210986468.20090221152324@fastmail.fm> <49A07081.1040204@xtra.co.nz> <6C8E7ADA880040799C1477C5C1C32AD5@pc52> <49A0A841.8000702@xtra.co.nz> <12619539593.20090222173205@fastmail.fm> <49A1B19E.9020908@xtra.co.nz> <10121744609.20090224202948@fastmail.fm> <49A45AA6.30304@xtra.co.nz> Message-ID: <863535593.20090226145148@fastmail.fm> On 24.02.2009, 21:37 Bruce Griffiths wrote: > Markus Kern wrote: >> On 22.02.2009, 21:12 Bruce Griffiths >> wrote: >> >> >>> Markus >>> >> >> >>> Even with sawtooth correction the performance of the M12+T was found >>> inadequate for the LOFAR array. >>> They use SRS FS725 rubidium sources disciplined by M12+T GPS timing >>> receivers. >>> >> >> I didn't mean using the M12 by itself, obviously a clock stable enough >> over the time the M12 pps must be integrated has to be used. >> >> If we are using the ADEV limits you proposed then at 50 MHz (= 3ns >> acceptable error) the timing requirement is an ADEV of 3*1E-(8+x) at >> tau = x seconds. From the measurements at >> http://www.leapsecond.com/pages/gpsdo/ it seems the Thunderbolt gets >> pretty close to that. >> >> The LOFAR clock system is described at [1]. In section 3.1.3.3 they >> say: >> >> "Some Crystal Oscillators have the advantage that they have a better >> Allan variance for periods of up to 10s and therefore it can be claimed >> that they have a better performance than the SRS-FS725 Rb-reference >> standard. The performance for time periods above 10s, the SRS-FS725 >> performs better. Therefore choosing an OCXO would require a maximum >> calibration interval of 10s and it would require a significantly better >> GPS (or GALILEO) receiver because de Rb-reference is used to average >> the PPS signal from the GPS receiver thereby making it possible to >> identify the time difference between stations at receive frequencies >> above 10MHz." >> >> I think this means that they are using pps integration times above 10 >> seconds. I couldn't find any reference to the actual value though. >> >> LOFAR is also working at frequencies up to 240MHz so the timing >> requirements are definitely higher. They say that a station time >> offset of 200ps does not affect performance as long as it remains >> stable over time. >> >> >>> They also state that the ionosphere contribution to ADEV is about 8E-12 >>> @10s. >>> >> >> Yes, from which they infer that "the reference clock shall have an >> Allan variance of 1e-11 or less over 10s." I am not sure if this has >> to do with the propagation of the GPS signal or if they mean that they >> need a clock stable enough to later compensate for the different >> delays of the observed signal through the ionosphere. >> >> Markus >> >> [1] >> http://www.lofar.org/operations/lib/exe/fetch.php?id=public%3Adocuments%3Alofar_documents&cache=cache&media=public:documents:19_detailed_description_of_clock_sync.pdf >> >> >> > Markus > The ionosphere contribution to the Allan deviation at GPS frequencies is > much smaller (by a factor of 10-100 or so) than that, as is evident from > carrier phase measurements. > At 50MHz the ionospheric phase shift, dispersion and instability will be > much greater than at GPS frequencies. > They are merely ensuring that the LO contribution to Allan deviation is > much smaller than that of the ionosphere. Yes, that's what I thought. > If you look at the Allan deviation plot on the PRS10 page: > http://www.thinksrs.com/products/PRS10.htm > This indicates that the likely disciplining loop time constant will be > several thousand seconds. Ok, so it may indeed be necessary to use a rubidium oscillator which has the required stability over that time frame. I suppose a GPS disciplined Rb-clock will be much more expensive than a Thunderbolt. However there are relatively cheap rubidium oscillators like the LPRO 101 out there. Are they suitable and has anyone tried to slave them to GPS? From reading the mailing list archives it seems Brooks Shera's circuit won't be suitable for this. > Close isn't good enough: the phase differences between pairs of stations > is significant, the Allan deviation needs to be at least 30% lower per > station. > If the errors at station pairs have significant correlation the > requirement can be relaxed somewhat. I realize that the phase difference is important and that things won't work if it's not low enough :) > Bruce From gsteinba52 at aol.com Thu Feb 26 16:15:56 2009 From: gsteinba52 at aol.com (gsteinba52 at aol.com) Date: Thu, 26 Feb 2009 11:15:56 -0500 Subject: [time-nuts] Tektronix 603/604, looking for any documentation Message-ID: <8CB6644A3994DFB-540-1087@WEBMAIL-DZ01.sysops.aol.com> Perhaps you could up-end the unit and reflect the display off of an adjustable mirror (front surface, of course...). Jerry Date: Wed, 25 Feb 2009 14:35:19 -0800 From: German Alvarez Subject: Re: [time-nuts] Tektronix 603/604, looking for any documentation To: time-nuts at febo.com Message-ID: <49A5C7A7.7040906 at gmail.com> Content-Type: text/plain; charset=ISO-8859-1; format=flowed Thanks to everyone in the list. I purchased the operators & service manual from www.ArtekMedia.com and after some cleaning and minor adjustments, it is performing as it was 35 years ago. Hats off to the designers of the unit, it is incredible how nice it performs with the duchtronics clock. Now I have to convince my SO that it makes a beautifully mantle clock, specially since the 604 is 20" depth. --ga From SAIDJACK at aol.com Thu Feb 26 17:19:31 2009 From: SAIDJACK at aol.com (SAIDJACK at aol.com) Date: Thu, 26 Feb 2009 12:19:31 EST Subject: [time-nuts] Different Thunderbolt versions Message-ID: Hi Bruce, the PRS-10 datasheet is pretty smart in not showing the data below 10s through, it would make their >10s performance not shine as much. There certainly could not be any measurement reason for not showing <10s ADEV. As someone has mentioned, they also used a very bad GPS receiver as a reference which makes their ADEV look spectacular against the GPS ADEV. It is possible that this plot is pre-SA-disabled, but then should they not update their website with more recent data? I guess one could also argue that even if you use a bad GPS the unit will still generate a good ADEV above 10s intervals and that is what they are trying to show. It's all in the specmanship. bye, Said In a message dated 2/25/2009 23:04:54 Pacific Standard Time, bruce.griffiths at xtra.co.nz writes: That isnt a great advantage in this application given that the contribution of the ionosphere to the VLBI receiver output signal phase noise is around 8E-11 @ 10s at 50MHz. Typically in VLBI integration times op up to 10,000 seconds or more are used. Integration times shorter than 100 sec are not usually of much interest. Bruce From had at to-way.com Thu Feb 26 17:32:22 2009 From: had at to-way.com (Had) Date: Thu, 26 Feb 2009 09:32:22 -0800 Subject: [time-nuts] Austron 6014 ? Message-ID: <20090226173223.4D32D98DA0E@mail-in05.adhost.com> Hi Gang, Does anyone have any documentation or manuals for the Austron 6014 Frequency Multiplier? Thanks, Had K7MLR From bruce.griffiths at xtra.co.nz Thu Feb 26 19:45:11 2009 From: bruce.griffiths at xtra.co.nz (Bruce Griffiths) Date: Fri, 27 Feb 2009 08:45:11 +1300 Subject: [time-nuts] Different Thunderbolt versions In-Reply-To: References: Message-ID: <49A6F147.6000600@xtra.co.nz> Said The other thing to consider is the noise in the built in FS725 PPS error measuring system. Whilst the resolution is 1ns it isn't that accurate and its instability may also be worse than this. Internal resolution may be as high as 200ps. Bruce SAIDJACK at aol.com wrote: > Hi Bruce, > > the PRS-10 datasheet is pretty smart in not showing the data below 10s > through, it would make their >10s performance not shine as much. There certainly > could not be any measurement reason for not showing <10s ADEV. > > As someone has mentioned, they also used a very bad GPS receiver as a > reference which makes their ADEV look spectacular against the GPS ADEV. > > It is possible that this plot is pre-SA-disabled, but then should they not > update their website with more recent data? I guess one could also argue that > even if you use a bad GPS the unit will still generate a good ADEV above 10s > intervals and that is what they are trying to show. It's all in the > specmanship. > > bye, > Said > > > In a message dated 2/25/2009 23:04:54 Pacific Standard Time, > bruce.griffiths at xtra.co.nz writes: > > That isnt a great advantage in this application given that the > contribution of the ionosphere to the VLBI receiver output signal phase > noise is around 8E-11 @ 10s at 50MHz. > Typically in VLBI integration times op up to 10,000 seconds or more are > used. Integration times shorter than 100 sec are not usually of much > interest. > > Bruce > > _______________________________________________ > time-nuts mailing list -- time-nuts at febo.com > To unsubscribe, go to https://www.febo.com/cgi-bin/mailman/listinfo/time-nuts > and follow the instructions there. > > From bruce.griffiths at xtra.co.nz Thu Feb 26 19:53:28 2009 From: bruce.griffiths at xtra.co.nz (Bruce Griffiths) Date: Fri, 27 Feb 2009 08:53:28 +1300 Subject: [time-nuts] Different Thunderbolt versions In-Reply-To: <863535593.20090226145148@fastmail.fm> References: <19210986468.20090221152324@fastmail.fm> <49A07081.1040204@xtra.co.nz> <6C8E7ADA880040799C1477C5C1C32AD5@pc52> <49A0A841.8000702@xtra.co.nz> <12619539593.20090222173205@fastmail.fm> <49A1B19E.9020908@xtra.co.nz> <10121744609.20090224202948@fastmail.fm> <49A45AA6.30304@xtra.co.nz> <863535593.20090226145148@fastmail.fm> Message-ID: <49A6F338.4010802@xtra.co.nz> Markus If one used a GPS timing receiver like the M12M + T or equivalent a single shot PPS phase error measurement resolution of 1ns or better is desirable. This is easily achieved using a microprocessor with built in charge redistribution ADC and a simple interpolation circuit. One can either devise a means of adjusting the rubidium frequency via the C field or use a high resolution, low noise, low spur, synthesiser with a small tuning range to produce an output frequency that is locked to the GPS receiver PPS output. With an appropriate receiver one could even use carrier phase data to reduce the measurement noise potentially allowing shorter loop time constants to be used. Bruce Markus Kern wrote: > On 24.02.2009, 21:37 Bruce Griffiths wrote: > > >> Markus Kern wrote: >> >>> On 22.02.2009, 21:12 Bruce Griffiths >>> wrote: >>> >>> >>> >>>> Markus >>>> >>>> >>> >>> >>>> Even with sawtooth correction the performance of the M12+T was found >>>> inadequate for the LOFAR array. >>>> They use SRS FS725 rubidium sources disciplined by M12+T GPS timing >>>> receivers. >>>> >>>> >>> I didn't mean using the M12 by itself, obviously a clock stable enough >>> over the time the M12 pps must be integrated has to be used. >>> >>> If we are using the ADEV limits you proposed then at 50 MHz (= 3ns >>> acceptable error) the timing requirement is an ADEV of 3*1E-(8+x) at >>> tau = x seconds. From the measurements at >>> http://www.leapsecond.com/pages/gpsdo/ it seems the Thunderbolt gets >>> pretty close to that. >>> >>> The LOFAR clock system is described at [1]. In section 3.1.3.3 they >>> say: >>> >>> "Some Crystal Oscillators have the advantage that they have a better >>> Allan variance for periods of up to 10s and therefore it can be claimed >>> that they have a better performance than the SRS-FS725 Rb-reference >>> standard. The performance for time periods above 10s, the SRS-FS725 >>> performs better. Therefore choosing an OCXO would require a maximum >>> calibration interval of 10s and it would require a significantly better >>> GPS (or GALILEO) receiver because de Rb-reference is used to average >>> the PPS signal from the GPS receiver thereby making it possible to >>> identify the time difference between stations at receive frequencies >>> above 10MHz." >>> >>> I think this means that they are using pps integration times above 10 >>> seconds. I couldn't find any reference to the actual value though. >>> >>> LOFAR is also working at frequencies up to 240MHz so the timing >>> requirements are definitely higher. They say that a station time >>> offset of 200ps does not affect performance as long as it remains >>> stable over time. >>> >>> >>> >>>> They also state that the ionosphere contribution to ADEV is about 8E-12 >>>> @10s. >>>> >>>> >>> Yes, from which they infer that "the reference clock shall have an >>> Allan variance of 1e-11 or less over 10s." I am not sure if this has >>> to do with the propagation of the GPS signal or if they mean that they >>> need a clock stable enough to later compensate for the different >>> delays of the observed signal through the ionosphere. >>> >>> Markus >>> >>> [1] >>> http://www.lofar.org/operations/lib/exe/fetch.php?id=public%3Adocuments%3Alofar_documents&cache=cache&media=public:documents:19_detailed_description_of_clock_sync.pdf >>> >>> >>> >>> > > >> Markus >> > > >> The ionosphere contribution to the Allan deviation at GPS frequencies is >> much smaller (by a factor of 10-100 or so) than that, as is evident from >> carrier phase measurements. >> At 50MHz the ionospheric phase shift, dispersion and instability will be >> much greater than at GPS frequencies. >> They are merely ensuring that the LO contribution to Allan deviation is >> much smaller than that of the ionosphere. >> > > Yes, that's what I thought. > > >> If you look at the Allan deviation plot on the PRS10 page: >> http://www.thinksrs.com/products/PRS10.htm >> > > >> This indicates that the likely disciplining loop time constant will be >> several thousand seconds. >> > > Ok, so it may indeed be necessary to use a rubidium oscillator which > has the required stability over that time frame. > > I suppose a GPS disciplined Rb-clock will be much more expensive than > a Thunderbolt. However there are relatively cheap rubidium oscillators > like the LPRO 101 out there. Are they suitable and has anyone tried to > slave them to GPS? From reading the mailing list archives it seems > Brooks Shera's circuit won't be suitable for this. > > >> Close isn't good enough: the phase differences between pairs of stations >> is significant, the Allan deviation needs to be at least 30% lower per >> station. >> If the errors at station pairs have significant correlation the >> requirement can be relaxed somewhat. >> > > I realize that the phase difference is important and that things won't > work if it's not low enough :) > > >> Bruce >> > > > > _______________________________________________ > time-nuts mailing list -- time-nuts at febo.com > To unsubscribe, go to https://www.febo.com/cgi-bin/mailman/listinfo/time-nuts > and follow the instructions there. > > From gwinn at raytheon.com Thu Feb 26 20:38:54 2009 From: gwinn at raytheon.com (Joseph M Gwinn) Date: Thu, 26 Feb 2009 15:38:54 -0500 Subject: [time-nuts] Rubiola on Mixers Message-ID: A while back, there was a discussion on how best to use double balanced diode ring mixers for delay measurements. I came upon a very useful paper by Enrico Rubiola: "Tutorial on the double balanced mixer", Enrico Rubiola, 21 Aug 2006 , 52 pages, arXiv:physics/0608211v1 [physics.ins-det]. Joe From peterawson at earthlink.net Thu Feb 26 22:47:46 2009 From: peterawson at earthlink.net (Pete) Date: Thu, 26 Feb 2009 15:47:46 -0700 Subject: [time-nuts] Shameless commerce Message-ID: I have a 59992A J06 that needs a home. Before trying *Bay, it seemed prudent to let the time-nuts list know, as it's of limited interest & there are not many available. If you're using a 5370A or B, it's a handy accessory. Please eMail me off list peterawson at earthlink.net if you can use tihs item. I'm not giving it away, but I'm confident of finding a reasonable arrangement. Pete Rawson From tvb at LeapSecond.com Fri Feb 27 05:10:55 2009 From: tvb at LeapSecond.com (Tom Van Baak) Date: Thu, 26 Feb 2009 21:10:55 -0800 Subject: [time-nuts] GPSDO for a week Message-ID: <1E08F65BA795431CBC391BDC3197C981@pc52> Here are some fresh, detailed plots of a TBolt for 8 days. http://www.leapsecond.com/pages/tbolt-8d/ Perhaps this helps with recent VLBI questions. Note the plots above are against a local standard and should not be confused with plots people are making through TSIP or TboltMon or LH, which are essentially just a TBolt comparing against itself. If this doesn't make sense I can explain more. /tvb From magnus at rubidium.dyndns.org Fri Feb 27 07:37:32 2009 From: magnus at rubidium.dyndns.org (Magnus Danielson) Date: Fri, 27 Feb 2009 08:37:32 +0100 Subject: [time-nuts] GPSDO for a week In-Reply-To: <1E08F65BA795431CBC391BDC3197C981@pc52> References: <1E08F65BA795431CBC391BDC3197C981@pc52> Message-ID: <49A7983C.2010709@rubidium.dyndns.org> Tom Van Baak skrev: > Here are some fresh, detailed plots of a TBolt for 8 days. > http://www.leapsecond.com/pages/tbolt-8d/ > > Perhaps this helps with recent VLBI questions. I think this is a very nice contribution to the discussion. I think the phase lot and TDEV (not ADEV as written under the TDEV plot) display interesting deviations. Notice the 86400 s sawtooth shapes burried in noise, notice how they can have different directions. However, I am wondering if this is not a 12h property since sometimes the halfs seems to have opposite direction, but the auto-correlator plot speaks against this with its very clear spikes which suggests a more stable 24h pattern. Nice to see the 1-1,5 ns time-stability in that range and slopes at both sides. Notice how it goes about 1 ns all the way up to 10000 s. > Note the plots above are against a local standard and should > not be confused with plots people are making through TSIP or > TboltMon or LH, which are essentially just a TBolt comparing > against itself. If this doesn't make sense I can explain more. Good point. Cheers, Magnus From bruce.griffiths at xtra.co.nz Fri Feb 27 12:38:22 2009 From: bruce.griffiths at xtra.co.nz (Bruce Griffiths) Date: Sat, 28 Feb 2009 01:38:22 +1300 Subject: [time-nuts] Different Thunderbolt versions In-Reply-To: <863535593.20090226145148@fastmail.fm> References: <19210986468.20090221152324@fastmail.fm> <49A07081.1040204@xtra.co.nz> <6C8E7ADA880040799C1477C5C1C32AD5@pc52> <49A0A841.8000702@xtra.co.nz> <12619539593.20090222173205@fastmail.fm> <49A1B19E.9020908@xtra.co.nz> <10121744609.20090224202948@fastmail.fm> <49A45AA6.30304@xtra.co.nz> <863535593.20090226145148@fastmail.fm> Message-ID: <49A7DEBE.4080709@xtra.co.nz> Markus Article discussing modelling effects of the ionosphere at low frequencies: http://ens.ewi.tudelft.nl/pubs/tol07isscs.pdf I presume you intend to make the VLBI observations at night when the ionosphere has the least effect on the propagation delay at 50MHz. Bruce Markus Kern wrote: > On 24.02.2009, 21:37 Bruce Griffiths wrote: > > >> Markus Kern wrote: >> >>> On 22.02.2009, 21:12 Bruce Griffiths >>> wrote: >>> >>> >>> >>>> Markus >>>> >>>> >>> >>> >>>> Even with sawtooth correction the performance of the M12+T was found >>>> inadequate for the LOFAR array. >>>> They use SRS FS725 rubidium sources disciplined by M12+T GPS timing >>>> receivers. >>>> >>>> >>> I didn't mean using the M12 by itself, obviously a clock stable enough >>> over the time the M12 pps must be integrated has to be used. >>> >>> If we are using the ADEV limits you proposed then at 50 MHz (= 3ns >>> acceptable error) the timing requirement is an ADEV of 3*1E-(8+x) at >>> tau = x seconds. From the measurements at >>> http://www.leapsecond.com/pages/gpsdo/ it seems the Thunderbolt gets >>> pretty close to that. >>> >>> The LOFAR clock system is described at [1]. In section 3.1.3.3 they >>> say: >>> >>> "Some Crystal Oscillators have the advantage that they have a better >>> Allan variance for periods of up to 10s and therefore it can be claimed >>> that they have a better performance than the SRS-FS725 Rb-reference >>> standard. The performance for time periods above 10s, the SRS-FS725 >>> performs better. Therefore choosing an OCXO would require a maximum >>> calibration interval of 10s and it would require a significantly better >>> GPS (or GALILEO) receiver because de Rb-reference is used to average >>> the PPS signal from the GPS receiver thereby making it possible to >>> identify the time difference between stations at receive frequencies >>> above 10MHz." >>> >>> I think this means that they are using pps integration times above 10 >>> seconds. I couldn't find any reference to the actual value though. >>> >>> LOFAR is also working at frequencies up to 240MHz so the timing >>> requirements are definitely higher. They say that a station time >>> offset of 200ps does not affect performance as long as it remains >>> stable over time. >>> >>> >>> >>>> They also state that the ionosphere contribution to ADEV is about 8E-12 >>>> @10s. >>>> >>>> >>> Yes, from which they infer that "the reference clock shall have an >>> Allan variance of 1e-11 or less over 10s." I am not sure if this has >>> to do with the propagation of the GPS signal or if they mean that they >>> need a clock stable enough to later compensate for the different >>> delays of the observed signal through the ionosphere. >>> >>> Markus >>> >>> [1] >>> http://www.lofar.org/operations/lib/exe/fetch.php?id=public%3Adocuments%3Alofar_documents&cache=cache&media=public:documents:19_detailed_description_of_clock_sync.pdf >>> >>> >>> >>> > > >> Markus >> > > >> The ionosphere contribution to the Allan deviation at GPS frequencies is >> much smaller (by a factor of 10-100 or so) than that, as is evident from >> carrier phase measurements. >> At 50MHz the ionospheric phase shift, dispersion and instability will be >> much greater than at GPS frequencies. >> They are merely ensuring that the LO contribution to Allan deviation is >> much smaller than that of the ionosphere. >> > > Yes, that's what I thought. > > >> If you look at the Allan deviation plot on the PRS10 page: >> http://www.thinksrs.com/products/PRS10.htm >> > > >> This indicates that the likely disciplining loop time constant will be >> several thousand seconds. >> > > Ok, so it may indeed be necessary to use a rubidium oscillator which > has the required stability over that time frame. > > I suppose a GPS disciplined Rb-clock will be much more expensive than > a Thunderbolt. However there are relatively cheap rubidium oscillators > like the LPRO 101 out there. Are they suitable and has anyone tried to > slave them to GPS? From reading the mailing list archives it seems > Brooks Shera's circuit won't be suitable for this. > > >> Close isn't good enough: the phase differences between pairs of stations >> is significant, the Allan deviation needs to be at least 30% lower per >> station. >> If the errors at station pairs have significant correlation the >> requirement can be relaxed somewhat. >> > > I realize that the phase difference is important and that things won't > work if it's not low enough :) > > >> Bruce >> > > > > _______________________________________________ > time-nuts mailing list -- time-nuts at febo.com > To unsubscribe, go to https://www.febo.com/cgi-bin/mailman/listinfo/time-nuts > and follow the instructions there. > > From mkern at fastmail.fm Fri Feb 27 22:51:25 2009 From: mkern at fastmail.fm (Markus Kern) Date: Fri, 27 Feb 2009 23:51:25 +0100 Subject: [time-nuts] Different Thunderbolt versions In-Reply-To: <49A7DEBE.4080709@xtra.co.nz> References: <19210986468.20090221152324@fastmail.fm> <49A07081.1040204@xtra.co.nz> <6C8E7ADA880040799C1477C5C1C32AD5@pc52> <49A0A841.8000702@xtra.co.nz> <12619539593.20090222173205@fastmail.fm> <49A1B19E.9020908@xtra.co.nz> <10121744609.20090224202948@fastmail.fm> <49A45AA6.30304@xtra.co.nz> <863535593.20090226145148@fastmail.fm> <49A7DEBE.4080709@xtra.co.nz> Message-ID: <13838268046.20090227235125@fastmail.fm> Thanks for the article link Bruce. I haven't investigated how to best compensate for ionospheric delays yet. Some ionosphere model will surely be necessary. While observing at night will help due to lower TEC I hope to have a good enough model to also get useful data during the day. I have also recently stumbled over AIPY, a Python package for interferometry which should help a lot with the data processing: http://setiathome.berkeley.edu/~aparsons/aipy/aipy.cgi http://setiathome.berkeley.edu/~aparsons/papers/2008-08-10_LFSW_AIPY_Presentation.pdf Markus On 27.02.2009, 13:38 Bruce Griffiths wrote: > Markus > Article discussing modelling effects of the ionosphere at low frequencies: > http://ens.ewi.tudelft.nl/pubs/tol07isscs.pdf > I presume you intend to make the VLBI observations at night when the > ionosphere has the least effect on the propagation delay at 50MHz. > Bruce > Markus Kern wrote: >> On 24.02.2009, 21:37 Bruce Griffiths >> wrote: >> >> >>> Markus Kern wrote: >>> >>>> On 22.02.2009, 21:12 Bruce Griffiths >>>> wrote: >>>> >>>> >>>> >>>>> Markus >>>>> >>>>> >>>> >>>> >>>>> Even with sawtooth correction the performance of the M12+T was found >>>>> inadequate for the LOFAR >>>>> array. >>>>> They use SRS FS725 rubidium sources disciplined by M12+T GPS timing >>>>> receivers. >>>>> >>>>> >>>> I didn't mean using the M12 by itself, obviously a clock stable enough >>>> over the time the M12 pps must be integrated has to be used. >>>> >>>> If we are using the ADEV limits you proposed then at 50 MHz (= 3ns >>>> acceptable error) the timing requirement is an ADEV of 3*1E-(8+x) at >>>> tau = x seconds. From the measurements at >>>> http://www.leapsecond.com/pages/gpsdo/ it seems the Thunderbolt gets >>>> pretty close to that. >>>> >>>> The LOFAR clock system is described at [1]. In section 3.1.3.3 they >>>> say: >>>> >>>> "Some Crystal Oscillators have the advantage that they have a better >>>> Allan variance for periods of up to 10s and therefore it can be claimed >>>> that they have a better performance than the SRS-FS725 Rb-reference >>>> standard. The performance for time periods above 10s, the SRS-FS725 >>>> performs better. Therefore choosing an OCXO would require a maximum >>>> calibration interval of 10s and it would require a significantly better >>>> GPS (or GALILEO) receiver because de Rb-reference is used to average >>>> the PPS signal from the GPS receiver thereby making it possible to >>>> identify the time difference between stations at receive frequencies >>>> above 10MHz." >>>> >>>> I think this means that they are using pps integration times above 10 >>>> seconds. I couldn't find any reference to the actual value though. >>>> >>>> LOFAR is also working at frequencies up to 240MHz so the timing >>>> requirements are definitely higher. They say that a station time >>>> offset of 200ps does not affect performance as long as it remains >>>> stable over time. >>>> >>>> >>>> >>>>> They also state that the ionosphere contribution to ADEV is about 8E-12 >>>>> @10s. >>>>> >>>>> >>>> Yes, from which they infer that "the reference clock shall have an >>>> Allan variance of 1e-11 or less over 10s." I am not sure if this has >>>> to do with the propagation of the GPS signal or if they mean that they >>>> need a clock stable enough to later compensate for the different >>>> delays of the observed signal through the ionosphere. >>>> >>>> Markus >>>> >>>> [1] >>>> http://www.lofar.org/operations/lib/exe/fetch.php?id=public%3Adocuments%3Alofar_documents&cache=cache&media=public:documents:19_detailed_description_of_clock_sync.pdf >>>> >>>> >>>> >>>> >> >> >>> Markus >>> >> >> >>> The ionosphere contribution to the Allan deviation at GPS frequencies is >>> much smaller (by a factor of 10-100 or so) than that, as is evident from >>> carrier phase measurements. >>> At 50MHz the ionospheric phase shift, dispersion and instability will be >>> much greater than at GPS frequencies. >>> They are merely ensuring that the LO contribution to Allan deviation is >>> much smaller than that of the ionosphere. >>> >> >> Yes, that's what I thought. >> >> >>> If you look at the Allan deviation plot on the PRS10 page: >>> http://www.thinksrs.com/products/PRS10.htm >>> >> >> >>> This indicates that the likely disciplining loop time constant will be >>> several thousand seconds. >>> >> >> Ok, so it may indeed be necessary to use a rubidium oscillator which >> has the required stability over that time frame. >> >> I suppose a GPS disciplined Rb-clock will be much more expensive than >> a Thunderbolt. However there are relatively cheap rubidium oscillators >> like the LPRO 101 out there. Are they suitable and has anyone tried to >> slave them to GPS? From reading the mailing list archives it seems >> Brooks Shera's circuit won't be suitable for this. >> >> >>> Close isn't good enough: the phase differences between pairs of stations >>> is significant, the Allan deviation needs to be at least 30% lower per >>> station. >>> If the errors at station pairs have significant correlation the >>> requirement can be relaxed somewhat. >>> >> >> I realize that the phase difference is important and that things won't >> work if it's not low enough :) >> >> >>> Bruce >>> From mkern at fastmail.fm Fri Feb 27 23:13:15 2009 From: mkern at fastmail.fm (Markus Kern) Date: Sat, 28 Feb 2009 00:13:15 +0100 Subject: [time-nuts] Different Thunderbolt versions In-Reply-To: <49A6F338.4010802@xtra.co.nz> References: <19210986468.20090221152324@fastmail.fm> <49A07081.1040204@xtra.co.nz> <6C8E7ADA880040799C1477C5C1C32AD5@pc52> <49A0A841.8000702@xtra.co.nz> <12619539593.20090222173205@fastmail.fm> <49A1B19E.9020908@xtra.co.nz> <10121744609.20090224202948@fastmail.fm> <49A45AA6.30304@xtra.co.nz> <863535593.20090226145148@fastmail.fm> <49A6F338.4010802@xtra.co.nz> Message-ID: <11739577828.20090228001315@fastmail.fm> On 26.02.2009, 20:53 Bruce Griffiths wrote: > Markus > If one used a GPS timing receiver like the M12M + T or equivalent a > single shot PPS phase error measurement resolution of 1ns or better is > desirable. > This is easily achieved using a microprocessor with built in charge > redistribution ADC and a simple interpolation circuit. Can you describe what you mean by interpolation circuit? I assume you are talking about a system which is able to get the 1ns resolution without requiring the 1 GHz clock necessary with direct sampling? > One can either devise a means of adjusting the rubidium frequency via > the C field or use a high resolution, low noise, low spur, synthesiser > with a small tuning range to produce an output frequency that is locked > to the GPS receiver PPS output. > With an appropriate receiver one could even use carrier phase data to > reduce the measurement noise potentially allowing shorter loop time > constants to be used. > Bruce > Markus Kern wrote: >> On 24.02.2009, 21:37 Bruce Griffiths >> wrote: >> >> >>> Markus Kern wrote: >>> >>>> On 22.02.2009, 21:12 Bruce Griffiths >>>> wrote: >>>> >>>> >>>> >>>>> Markus >>>>> >>>>> >>>> >>>> >>>>> Even with sawtooth correction the performance of the M12+T was found >>>>> inadequate for the LOFAR >>>>> array. >>>>> They use SRS FS725 rubidium sources disciplined by M12+T GPS timing >>>>> receivers. >>>>> >>>>> >>>> I didn't mean using the M12 by itself, obviously a clock stable enough >>>> over the time the M12 pps must be integrated has to be used. >>>> >>>> If we are using the ADEV limits you proposed then at 50 MHz (= 3ns >>>> acceptable error) the timing requirement is an ADEV of 3*1E-(8+x) at >>>> tau = x seconds. From the measurements at >>>> http://www.leapsecond.com/pages/gpsdo/ it seems the Thunderbolt gets >>>> pretty close to that. >>>> >>>> The LOFAR clock system is described at [1]. In section 3.1.3.3 they >>>> say: >>>> >>>> "Some Crystal Oscillators have the advantage that they have a better >>>> Allan variance for periods of up to 10s and therefore it can be claimed >>>> that they have a better performance than the SRS-FS725 Rb-reference >>>> standard. The performance for time periods above 10s, the SRS-FS725 >>>> performs better. Therefore choosing an OCXO would require a maximum >>>> calibration interval of 10s and it would require a significantly better >>>> GPS (or GALILEO) receiver because de Rb-reference is used to average >>>> the PPS signal from the GPS receiver thereby making it possible to >>>> identify the time difference between stations at receive frequencies >>>> above 10MHz." >>>> >>>> I think this means that they are using pps integration times above 10 >>>> seconds. I couldn't find any reference to the actual value though. >>>> >>>> LOFAR is also working at frequencies up to 240MHz so the timing >>>> requirements are definitely higher. They say that a station time >>>> offset of 200ps does not affect performance as long as it remains >>>> stable over time. >>>> >>>> >>>> >>>>> They also state that the ionosphere contribution to ADEV is about 8E-12 >>>>> @10s. >>>>> >>>>> >>>> Yes, from which they infer that "the reference clock shall have an >>>> Allan variance of 1e-11 or less over 10s." I am not sure if this has >>>> to do with the propagation of the GPS signal or if they mean that they >>>> need a clock stable enough to later compensate for the different >>>> delays of the observed signal through the ionosphere. >>>> >>>> Markus >>>> >>>> [1] >>>> http://www.lofar.org/operations/lib/exe/fetch.php?id=public%3Adocuments%3Alofar_documents&cache=cache&media=public:documents:19_detailed_description_of_clock_sync.pdf >>>> >>>> >>>> >>>> >> >> >>> Markus >>> >> >> >>> The ionosphere contribution to the Allan deviation at GPS frequencies is >>> much smaller (by a factor of 10-100 or so) than that, as is evident from >>> carrier phase measurements. >>> At 50MHz the ionospheric phase shift, dispersion and instability will be >>> much greater than at GPS frequencies. >>> They are merely ensuring that the LO contribution to Allan deviation is >>> much smaller than that of the ionosphere. >>> >> >> Yes, that's what I thought. >> >> >>> If you look at the Allan deviation plot on the PRS10 page: >>> http://www.thinksrs.com/products/PRS10.htm >>> >> >> >>> This indicates that the likely disciplining loop time constant will be >>> several thousand seconds. >>> >> >> Ok, so it may indeed be necessary to use a rubidium oscillator which >> has the required stability over that time frame. >> >> I suppose a GPS disciplined Rb-clock will be much more expensive than >> a Thunderbolt. However there are relatively cheap rubidium oscillators >> like the LPRO 101 out there. Are they suitable and has anyone tried to >> slave them to GPS? From reading the mailing list archives it seems >> Brooks Shera's circuit won't be suitable for this. >> >> >>> Close isn't good enough: the phase differences between pairs of stations >>> is significant, the Allan deviation needs to be at least 30% lower per >>> station. >>> If the errors at station pairs have significant correlation the >>> requirement can be relaxed somewhat. >>> >> >> I realize that the phase difference is important and that things won't >> work if it's not low enough :) >> >> >>> Bruce >>> From mkern at fastmail.fm Fri Feb 27 23:31:24 2009 From: mkern at fastmail.fm (Markus Kern) Date: Sat, 28 Feb 2009 00:31:24 +0100 Subject: [time-nuts] GPSDO for a week In-Reply-To: <1E08F65BA795431CBC391BDC3197C981@pc52> References: <1E08F65BA795431CBC391BDC3197C981@pc52> Message-ID: <6840667234.20090228003124@fastmail.fm> On 27.02.2009, 06:10 Tom Van Baak wrote: > Here are some fresh, detailed plots of a TBolt for 8 days. > http://www.leapsecond.com/pages/tbolt-8d/ > Perhaps this helps with recent VLBI questions. Thanks a lot for taking the time to do these measurements Tom. If I am interpreting the data correctly the ADEV stays below 3E-(8+tau) out to tau=1000s and is above that for tau=10000s. So the Thunderbold is indeed not suitable for VLBI at 40-50MHz with integration times longer than 1000s. > Note the plots above are against a local standard and should > not be confused with plots people are making through TSIP or > TboltMon or LH, which are essentially just a TBolt comparing > against itself. If this doesn't make sense I can explain more. > /tvb From tvb at LeapSecond.com Fri Feb 27 23:47:52 2009 From: tvb at LeapSecond.com (Tom Van Baak) Date: Fri, 27 Feb 2009 15:47:52 -0800 Subject: [time-nuts] GPSDO for a week References: <1E08F65BA795431CBC391BDC3197C981@pc52> <6840667234.20090228003124@fastmail.fm> Message-ID: > Thanks a lot for taking the time to do these measurements Tom. > > If I am interpreting the data correctly the ADEV stays below > 3E-(8+tau) out to tau=1000s and is above that for tau=10000s. So the > Thunderbold is indeed not suitable for VLBI at 40-50MHz with > integration times longer than 1000s. I'm not sure what you mean by the expression: 3E-(8+tau). For example, if you plug in tau = 1000s into that equation, you get 3e-1008, which isn't right. Also how does this compare with your earlier requirement of 3 ns at 1 second (which would be 3e-9). I'll help, but I'm a bit confused what your performance goal is and how that relates to 50 MHz. /tvb From bruce.griffiths at xtra.co.nz Sat Feb 28 00:02:01 2009 From: bruce.griffiths at xtra.co.nz (Bruce Griffiths) Date: Sat, 28 Feb 2009 13:02:01 +1300 Subject: [time-nuts] GPSDO for a week In-Reply-To: References: <1E08F65BA795431CBC391BDC3197C981@pc52> <6840667234.20090228003124@fastmail.fm> Message-ID: <49A87EF9.4080501@xtra.co.nz> Tom Van Baak wrote: >> Thanks a lot for taking the time to do these measurements Tom. >> >> If I am interpreting the data correctly the ADEV stays below >> 3E-(8+tau) out to tau=1000s and is above that for tau=10000s. So the >> Thunderbold is indeed not suitable for VLBI at 40-50MHz with >> integration times longer than 1000s. >> > > I'm not sure what you mean by the expression: 3E-(8+tau). For > example, if you plug in tau = 1000s into that equation, you get > 3e-1008, which isn't right. > > Also how does this compare with your earlier requirement of > 3 ns at 1 second (which would be 3e-9). I'll help, but I'm a bit > confused what your performance goal is and how that relates > to 50 MHz. > > /tvb > > Tom Try 3E-(8 + log_10 (Tau)). The requirement should be something like: 1) The rms phase error between independent LO pairs should be < 1 radian for the duration of the signal integration. This requirement is akin to avoiding carrier collapse of signal when the sideband noise power exceeds the carrier power. 2) Ideally the LO phase noise should be less than the ionospheric contribution (however only have one data point for the ionosphere ADEV ~8E-11@ 10s). Presumably this is the contribution at night as the ionosphere TEC rises significantly during the day as does the ionospheric delay and its fluctuations). If the delay through the ionosphere is plotted as a function of time, the dawn wave where the TEC starts to rise can clearly be seen. Periodic disturbances of TEC can also occur at this time. Bruce From mkern at fastmail.fm Sat Feb 28 00:20:40 2009 From: mkern at fastmail.fm (Markus Kern) Date: Sat, 28 Feb 2009 01:20:40 +0100 Subject: [time-nuts] GPSDO for a week In-Reply-To: References: <1E08F65BA795431CBC391BDC3197C981@pc52> <6840667234.20090228003124@fastmail.fm> Message-ID: <8043622625.20090228012040@fastmail.fm> On 28.02.2009, 00:47 Tom Van Baak wrote: >> Thanks a lot for taking the time to do these measurements Tom. >> >> If I am interpreting the data correctly the ADEV stays below >> 3E-(8+tau) out to tau=1000s and is above that for tau=10000s. So the >> Thunderbold is indeed not suitable for VLBI at 40-50MHz with >> integration times longer than 1000s. > I'm not sure what you mean by the expression: 3E-(8+tau). For > example, if you plug in tau = 1000s into that equation, you get > 3e-1008, which isn't right. Your are right, of course. I just realized I made the same mistake in my previous post. The ADEV requirement is 3E-9 * tau at tau or equivalently 3E-(9+x) at tau = 1Ex. > Also how does this compare with your earlier requirement of > 3 ns at 1 second (which would be 3e-9). I'll help, but I'm a bit > confused what your performance goal is and how that relates > to 50 MHz. At f = 50MHz the period is T = 1/f = 20ns and a 50 degree phase error corresponds to T*50/360 = 2.7ns = 2.7E-9s (or 3ns at 46 Mhz). Since the phase error must be below 50 degrees over the entire signal integration time the required ADEV decreases with the integration time d as 2.7E-9 * tau at tau = d. > /tvb