GPSDO module

  • 2
  • Question
  • Updated 3 years ago
  • Answered
  • (Edited)
I got my GPSDO module in the other day and installed it (very easy install by the way) and set everything up.  Took quite a while for the initial lock of sats to happen but after that initial setup and lock it's been much quicker since.  So my question is, for those that have them, where does your frequency error usually run?  Mine is around 1.770000e-11 with fluctuation but that is about average.  Second question, what does this number really mean?

Thanks

David
KK4QOE
Photo of David Warnberg

David Warnberg

  • 538 Posts
  • 56 Reply Likes

Posted 3 years ago

  • 2
Photo of EA4GLI - 8P9EH - Salvador

EA4GLI - 8P9EH - Salvador

  • 1758 Posts
  • 536 Reply Likes
Photo of EA4GLI - 8P9EH - Salvador

EA4GLI - 8P9EH - Salvador

  • 1758 Posts
  • 536 Reply Likes
Curious about the numbers myself. I purchased an external antenna on eBay, for $10 and I get faster lock and more sats.
Photo of Reg

Reg

  • 120 Posts
  • 73 Reply Likes
David:

Mine usually runs approximately 3.400000e-12.  I am using a Tram marine GPS antenna mounted high enough to have a horizon-to-horizon view of the satellites.  My experience with the provided indoor antenna was that it provided a significantly lower accuracy number.  Steve, N5AC, once posted that we multiple this number times the operating frequency to yield the error in Hz.  At 14.2 MHz this yields an error of 251 microHertz.

Reg
Photo of Reg

Reg

  • 120 Posts
  • 73 Reply Likes
Ernest:

I'm using the Tram 1655.  I have 2 of them.  One serves the GPSDO in the FLEX-6700 and the second one (different room) serves a Jackson Labs LC-XO-PLUS GPSDO which feeds a Stanford Research Systems SF730/1 10 MHz distribution amp which connects to the instruments on the test bench (Keysight 53522B 2-channel waveform generator, Keysight 53230A frequency counter/timer, Keysight DSOX4034A oscillosope, Keysight N9320B spectrum analyzer, etc.)

I once had a Tram GPS antenna which was branded as RayMarine and it was mounted about 12" below a 12 KW marine RADAR.  It worked fine in that environment so I trusted them for the ham shack and lab.

Reg
Photo of Ernest - W4EG

Ernest - W4EG

  • 206 Posts
  • 38 Reply Likes
Thank you Reg... 73
Photo of David Warnberg

David Warnberg

  • 538 Posts
  • 56 Reply Likes
Reg... how did you connect the Tram GPS to the Radio.. That GPS looks to have a BNC connector and the Radio has a SMA connection.. did you just use an adapter?? or did you change out the cable?

Thanks
Photo of Reg

Reg

  • 120 Posts
  • 73 Reply Likes
David:

Your guess is correct.  On Amazon I purchased a short pre-assembled coax with an SMA male on one end and a BNC female on the other.  I used the coax in order to relieve strain on the rear of the FLEX-6700.

The Jackson Labs LC-XO-PLUS GPSDO that I use on the test bench came with a MMCX to BNC female adapter cable.

Reg
Photo of Jim, KJ3P

Jim, KJ3P

  • 88 Posts
  • 12 Reply Likes

Hi Reg,

Using your first post's numbers,  3.400000e-12 and 14.2 MHz, I get an error of 48.28 microHz.  Never was good at math...where did I go wrong?

In any event, what's a few microHertz among friends?

  --jim KJ3P

Photo of Ken - NM9P

Ken - NM9P, Elmer

  • 4025 Posts
  • 1248 Reply Likes
For those interested...

The "e-11" is engineering/mathematics shorthand for "multiply this number times ten to the -11 power". Which essentially means shift the decimal point to the left 11 places (for negative exponents). That is a pretty small number! But when multiplied by a frequency in megahertz, you can get a number that shows the approximate frequency error, as Reg posted.

Shorthand exponential code like this is used when dealing with extremely large or extremely small numbers. Either way, it saves on paper and ink and eliminates many errors because of misaligned decimal columns!

Simple examples....
I could say my station is running 1.257e3 Watts, which means that I am running 1.257 X 10 to the third power. = 1.257 X 1000 Watts. = 1257 Watts.

If I am running 3.5e-4 Watts, then I am running .00035 Watts, which is 0.35 milliwatts....less than a thousandth of a watt!

But at these powers, most engineers use a different scale, referencing deciBels above or below a milliwatt. But that is another post. I still get lost in those numbers without my calculator and its log functions!

"Math is fun!" My Physics and Calculus professor used to say..."Learn it, love it, live it!"

Ken - NM9P
Photo of Martin Ewing AA6E

Martin Ewing AA6E

  • 293 Posts
  • 58 Reply Likes
Oscillator stability and accuracy are complicated issues. Specifying one number like ppm over 24 hrs is interesting, but it's not the whole story. Ideally, you want to know the stability over all time scales. At short timescales, this is more or less the same as LO phase noise that we're familiar with. The "Allan Variance" gives you the variance at every timescale. See https://en.wikipedia.org/wiki/Allan_v... .

The kind of performance you get with the GPSDO is hardly necessary for normal HF operations, but it's helpful if you want to do JT9 in microwave bands! And it's great for radio science (VLBI) or general bragging rights.
Photo of Peter K1PGV

Peter K1PGV, Elmer

  • 541 Posts
  • 315 Reply Likes
That's very interesting.  I would imagine the period of time that the oscillator is running matters, as well.  The longer the oscillator is powered-up, the more accurate it gets I assume. I know this was asked before, but: As long as the radio is plugged-in, is the GPSDO running?  Flex claims an accuracy of the internal GPSDO option as being "5 x 10-12 over 24 hours."  Which is super good for the price and for the purposes of keeping you on the right frequency.

I have a Fury external GPSDO (because I'm nuts about time and frequency accuracy -- you can see the current of it's performance plot here) and over the past 1340 hours I'm getting an average jitter of 1.4903e-15 (assuming I'm calculating that correctly) -- You can also see from the plot that the frequency was disciplined to less than +41ns / -36ns at all times over the past 1340+ hours.  I've got a crappy antenna from ebay stuck outside next to my house, on top of my central air conditioning unit.  It can usually see 6 or 7 satellites, so I figure that's good enough.

I'd be curious to hear what other frequency references people use, and what their average wander (jitter) is.

Peter
K1PGV
Photo of David Warnberg

David Warnberg

  • 538 Posts
  • 56 Reply Likes
Peter..
           I would assume the GPSDO run's all the time even with the radio powered off... why you ask?  The light on the front stays lit as yellow which it did not do before.  Also I'm just using the stock GPS antenna now and every time I've looked I have no less then 8 sats locked with usually 10 to 12 in view and the antenna is sitting in a window on the south side of the house (not sure it that matters)


Photo of EA4GLI - 8P9EH - Salvador

EA4GLI - 8P9EH - Salvador

  • 1758 Posts
  • 536 Reply Likes
That is all correct. If you were to stop power from PSU to the Flex radio you would lose that lock. As long as there is 12v to the radio the GPS stays on.

Another piece of info that you might have missed. The sticker that comes with the antenna should go in the side with the engraved letters. Given the performance you are getting I am sure you placed it correctly.

Also, the supplied antenna is an indoors antenna. Water will get inside and break it, I placed mine outside..... because I didn't read the instructions... which led to me to buy the eBay one.
Photo of EA4GLI - 8P9EH - Salvador

EA4GLI - 8P9EH - Salvador

  • 1758 Posts
  • 536 Reply Likes
I bought this one and it works great.
http://www.ebay.com/itm/External-Amplified-tomtom-GPS-Antenna-for-GlobalSat-AT-65-SMA-28dbi-high-gai...



Less that $9, Free Shipping and it works outdoors.
It comes with 5m (15ft) of cable.

With the internal one, I could get to see at most 8 to 9 Sats. With this antenna placed outside I get consistently 17 sats visible with 10 tracked.
Photo of Ken - NM9P

Ken - NM9P, Elmer

  • 4025 Posts
  • 1248 Reply Likes
I have found that those antennas, for GPS and similar for XM/Sirius Radio, work even better when mounted on a flat metal surface at least 4 or more inches in diameter. It seems to have a reflector effect. I discovered this while placing one in the overhead compartment of my Roadtrek RV. I figured it would go through the fiberglass roof just fine. But after experiencing many dropouts, I mounted it to a coffee can lid and placed it in exactly the same place and the signal was much stronger.

I can't explain it exactly, but you might try it if you are currently mounting it on a non-conductive surface.

Has anyone else experienced this?
(Edited)
Photo of EA4GLI - 8P9EH - Salvador

EA4GLI - 8P9EH - Salvador

  • 1758 Posts
  • 536 Reply Likes
That is a great point, I will put a metal plate underneath mine to test your findings.
These eBay antennas come with a magnet for the purpose of attaching to car/vehicles' roofs.
Photo of EA4GLI - 8P9EH - Salvador

EA4GLI - 8P9EH - Salvador

  • 1758 Posts
  • 536 Reply Likes
More info about GPS antennas can be found on this post by Tim>
https://helpdesk.flexradio.com/hc/en-us/articles/202118638-External-GPS-Antennas