[time-nuts] Test WWV timecube against Cesium, Rubidium, MASER or other precision time (UT-1) metrology
kb8tq at n1k.org
Thu Dec 7 16:55:26 EST 2017
If you happen to pick 5 or 10 MHz as your target frequency, all sorts of interesting
things come into play. The first is that any 5 or 10 MHz local standard distribution
may well get into your receiver. That can create all sorts of odd effects as the signal
fades. Next is that you aren’t just on a channel for two stations. There are a number of
standard broadcasts that pop up. Depending on propagation, they may be strong
enough to notice.
*If* you can get WWV to give you sub 1 ms timing accuracy (or even low ms), that’s better
accuracy (as opposed to jitter) than NTP can do with a normal home internet connection.
Internet modems have some pretty nasty timing asymmetries built into them. On top of that
you may have asymmetric routing. Up stream and downstream routing is a “two pipes”
process in a lot of head end installations. Again, we’re talking about internet and not
some sort of GPS based system.
To get the accuracy into the 1 ms range on WWV, you would need a pretty good idea of the path
length between you and WWV. If you are looking at ground wave, that is a bit easier than
if you are bouncing off the ionosphere. Just as with WWVB, there probably is a “best time
of day” to run your comparison.
Since you get a lot of ticks, there is no real need to just use one. Feed the audio into a computer and
let it sum them up. Monitoring for 10 minutes does not seem to be overly crazy. That’s still a lot
of ticks. If you dig into the archives, there are some good posts from PHK about doing this
sort of thing with Loran-C.
> On Dec 7, 2017, at 4:29 PM, Bill Hawkins <bill.iaxs at pobox.com> wrote:
> One way to compare any WWV receiver to a local standard is to use the
> PPS output of a standard against the PPS tick modulated on WWV. The tick
> is five cycles of a one KHz signal derived from the master frequency.
> It will be a bit tricky to determine the onset of the first cycle amid
> the noise on shortwave radio. A computation that determined that there
> were just 5 cycles and worked backwards to determine the timestamp of
> the beginning or middle of the tick could then allow calculation of the
> offset between the standard PPS and the tick. Limit of accuracy might be
> 100 microseconds.
> Years ago, I had a standard calibrator made by Lavoie that had a vacuum
> tube WWV receiver. IIRC, the WWV carrier caused a circular sweep on a 2
> inch CRT. The sine wave from a standard modulated the intensity of the
> circular trace, so that a bright half moon appeared on the CRT and
> rotated at the error rate between the two frequencies. On several
> evenings the dominant signal varied between WWV and WWVH (identified by
> the voice broadcasts). Here in Minneapolis the phase difference between
> the two stations was about 180 degrees, causing the bright arc on the
> CRT to change sides.
> So yes, this could be interesting for a hobbyist, but it won't add
> anything to Science.
> A MASER is overkill. Heck, so are Rubidium and Caesium.
> A naked crystal will be rock solid compared to received WWV.
> OTOH, NTP has marvelous mathematical tricks to reduce Internet
> propagation delay.
> A scheme to reduce varying atmospheric delay would be useful, if there
> weren't much better ways to get a standard frequency.
> Bill Hawkins
> -----Original Message-----
> From: time-nuts [mailto:time-nuts-bounces at febo.com] On Behalf Of Patrick
> Sent: Thursday, December 07, 2017 7:48 AM
> Hello Friends,
> I am picking up locally a couple of vintage analog Radio Shack SW time
> cube radios, 70s vintage, 3 switchable SW frequencies. Two types, the
> one pictured and a Radio Shack model also that has WWV and Weather
> channel VHF frequencies.
> I am interested in an accurate bench test to compare the analog
> shortwave radios time reporting hopefully UT-1 against other available
> references. For accuracy, and
> repeatability. Could eventually add an SDR to the mix, too.
> The 5,10,15 mhz radios obviously are subject to the WWV Ft Collins site,
> propagation distance delays, somewhat calculable, and the vagaries of
> Ionospheric propagation, and, propagation delays between the antenna and
> the measured tap point to the seconds ticks of WWV. I have some
> microwave professionals, who are also hams here in Auburn who may enjoy
> doing a bench test, with published results, etc. But wonder if anyone
> else would be interested in borrowing a RS Timecube radio (and/or use an
> SDR) and designing an accurate bench test against available modern
> We are talking probably HUGE UT-1 errors compared to what this group
> plays with, and that is OK but I think still a worthwhile test,
> especially if the errors using available and cheap equipment are
> predictable, and repeatable.
> ----- %< ----- [snip of microwave stuff]
> Best, 73, Pat Barthelow AA6EG
> apol <apolloeme at gmail.com>loeme at gmail.com
> 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.
More information about the time-nuts