[time-nuts] ACE-III GPS receivers
didier at cox.net
Sun Jan 28 01:27:22 EST 2007
Dr Bruce Griffiths wrote:
> Didier Juges wrote:
>> Nice Symmetricom antenna there, Jason. I need to find a similar mount. I
>> have made an S shaped support with PVC tube, but I have not found a
>> clean way to secure it to the wall. The metal bracket looks much cleaner.
>> I will look for laptop hard drive hardware. Even though I have yet to
>> find one where there actually was a cable (as someone else pointed out,
>> they tend to be mounted rigidly), it may be easier to find than the bulk
>> cable. I'll check the local computer store, they may have cables. The
>> Jupiter receiver that is currently running is soldered directly to the
>> header, and it sure looks ugly. By comparison, the soldered antenna
>> connection looks much better. I looked for MCX connectors or pigtails
>> and they cost as much as the GPS receiver :-(
>> Only problem with all these different antennas and receivers, they all
>> have different connectors: MCX, SMA, F, N, SMB, and my "good" coax
>> cables have BNC!
>> Speaking of antenna cable: when I got the Thunderbolt, I locally bought
>> 50 feet of relatively expensive (may have been $18 or so), high grade
>> (at least advertised as such) and good looking 75 ohm coax cable with F
>> connectors (I say good looking because the outside jacket is clear, and
>> you could see the braid is thick and has good coverage, and the
>> connectors look well mounted). Then I found the antenna/cable did not
>> work too well with the Thunderbolt and I switched to the puck antenna.
>> Then, independently, for my ham radio hobby, I also bought 10 rolls of
>> 50' 75 ohm coax cable with molded BNCs for $1 each (I bought 10 to
>> amortize the $10 shipping)!! on eBay, and the $1 cables ($2 if you
>> include shipping) are better than the expensive cable by about 1.5 dB at
>> 1.8 GHz for a 50' cut. This was a good measurement (signal generator and
>> power meter). I also compared leakage by spreading the cables upstairs
>> in my house (I have a big house...) with a proper termination at one end
>> and the other end hooked to the spectrum analyzer and looking for local
>> FM stations (not very scientific, I'll admit, feel free to propose a
>> better method). The cheap cable was better (lower signal) by about 10dB.
>> For what I paid for the expensive cable, I could have gotten a bunch
>> more of the $1 cables... Made in China of course :-) Oh well...
>> Regarding the Thunderbolt, it simply works much better with the puck
>> than with the Bullet, so it's not the receiver setting. It could be
>> location, as even in the same approximate location (unless that location
>> is perfectly in the clear, which is not the case), different antennas
>> may have different performance by being differently affected by local
>> reflections. My puck antenna went from marginal to great by raising it 3
>> feet in the shack.
>> Didier KO4BB
>> Jason Rabel wrote:
>>> As Joseph said, notebook hard drive cables are 2mm. You can just peel off
>>> the number of wires that you need from the ribbon. If I remember I'll try to
>>> find all the proper pages in Mouser for the various bits.
>>> I had to mount an antenna outside to get reception because the building is
>>> all metal. My office is on the north-side and setting a receiver in the
>>> window I *might* be able to get 2 or 3 satellites for brief moments, but not
>>> long enough to do any good.
>>> The mount was already there from an existing antenna (which reminds me I
>>> need to take pictures of it, maybe one of you ham guys would have a use for
>>> it). Having a scissor lift made it extremely easy to mount that sucker up
>>> there. :) I *think* I paid like $30-$40 for mine, it was NIB. I had to run
>>> about 100ft of RG-58 from the antenna across the building and finally down
>>> to my office.
>>> The Jupiter receivers has a MCX connector, but it sounds like your
>>> direct-connect method works great (and much cheaper).
>>> You know I was reading some GPS manual (don't ask me which), and you could
>>> actually set the minimum signal strength for the receiver to use a
>>> satellite. Perhaps the Thunderbolt has a similar setting and it is turned
>>> higher than a normal receiver?
>>> I totally can sympathize about the lack of time. I need about a dozen clones
>>> of myself!
>> time-nuts mailing list
>> time-nuts at febo.com
> Most (except for Trimble,..) GPS receivers and antennas are designed to
> use 50 ohm cable.
> The HP58512A, HP58536A GPS distribution amplifiers use 50 ohm RF connectors.
> Trimble Bullet GPS antennas have a 50 ohm output impedance.
> Trimble literature however is ambiguous in that in the Resolution T
> receiver datasheets talk about using RG59 to connect to the antenna.
> Rockwell Jupiter receivers are specified for 50 ohm systems.
> To confuse the issue the M12+ M12M receivers don't appear to specify the
> antenna cable impedance.
> Both MCX 50 ohm and 75 ohm connectors are intermateable as are MMCX 50
> ohm and 75 ohm connectors.
> Datum (Micropulse) 26dB, 40dB and 50dB gain quadrifilar helix timing
> antennas are 50 ohm.
> Patch antennas can be marginal for timing receivers particularly if the
> antenna isn't optimally sited.
> Timing antennas incorporating 1575.42MHz bandpass filters help a lot
> with interference.
> Quadrifilar helix antennas work very well for timing applications.
> Maybe someone needs to systematically collect/collate GPS receiver and
> antenna impedance specs.
> time-nuts mailing list
> time-nuts at febo.com
In my experience, the mismatch loss by using 75 ohm cable if often more
than made up by lower insertion loss of 75 ohm cable versus 50 ohm cable
of the same diameter (and cost), particularly when having long runs at
high frequency. Impedance mismatch may cause other problems in addition
to mismatch loss, particularly in broad band systems, but considering
the high insertion loss of common coax cable at GPS frequencies, the
mismatch should not be a problem. To some extend, the high insertion
loss helps mask the effects of VSWR. Trimble also recommends RG-59 in
the Thunderbolt manual, and even provides such cable with the kit.
Regarding the antenna, I am sure that my Symmetricom 53532A, with its
-70dB filter at 50 MHz, will do great and I am anxious to have it installed.
The 53532A antenna responds to circular polarization. If I could make a
suitable test antenna (maybe a small helix) and hook it up to the HP
8657B signal generator, it would be possible to plot the frequency
response of the antenna/cable off the air on the spectrum analyzer and
see if VSWR causes a lot of passband ripple, but I doubt it. Considering
the relatively narrow bandwidth of the L1 signal and the insertion loss
in the cable (I forgot how much, but it's in the order of 15dB IIRC), I
would bet it's negligible.
More information about the time-nuts