[time-nuts] A new take on the all-hardware GPSDO concept

Lars Walenius lars.walenius at hotmail.com
Thu Oct 6 16:06:48 EDT 2016

Hi Nick

The answer to your question is that I have only tested the 1PPS. I enclose two non saw-tooth corrected charts
from the same run. Normally I run my LPRO Rb in hold mode for such a test but this time I run it as GPSDO with 8000secs time constant as I use it for other tests. The TIC was a HP5370B. The Venus 838 was enclosed in a towel.

As you see it is often excursions of around 50ns pp but the 20 seconds moving average I show is not moving a lot due to the ”50ns excursions”. I have one possible explanation why my Venus 838 (that I got from you) show these excursions and not other timing receivers I have like the M12 and LEA6T. I Think the TCXO in the Venus 838 might be digitally temperature compensated and that it makes jumps in the 1 to 2E-8 range. That is 10-20ns per second. After just a few seconds the NCO (?) inside the Venus corrects the error. Today I tried just to breathe (blow) on the Venus and I could see more than 100ns pp excursions. Always around the mean.

I really encourage you to test your Venus 1PPS with your TIC (53220A?) if you haven’t already done it. As reference the Tbolt or if you have a Rb is very useful. Of course testing a GPS module against a GPSDO might give wrong results for Taus longer than the time constant but still you will learn a lot. Best for you is to use a free running Rb if you still have one running.

At the same time you can check your GPSDO´s TIC (ADC) if it shows the same as your counter.

As I see that you use almost a copy of the TIC in my Arduino GPSDO I think you shouldn´t have more than a few ns pp ripple. The ripple shall also average out if your design works like mine. Probably it also good to test the temperature stability of the GPSDO TIC. Mine have about 0.3ns/C that seems ok for me. Your FET might give worse results?
For these tests I think you should use two stable oscillators and for example a PICDIV down to 1PPS. I really like the PD26 as you can step the output so that you get time marks every 200ns for the GPSDO TIC (with 1us range) if you use just one oscillator for both the 10MHz and 1PPS.


Från: Nick Sayer via time-nuts<mailto:time-nuts at febo.com>
Skickat: den 27 september 2016 06:25
Till: Discussion of precise time and frequency measurement<mailto:time-nuts at febo.com>
Ämne: Re: [time-nuts] A new take on the all-hardware GPSDO concept

> On Sep 16, 2016, at 1:13 PM, Lars Walenius wrote:
> My experience with the Venus838-T is only 2 weeks but disappointing. This can also be guessed from the datasheet ADEV curve, that I guess is sawtooth corrected values as it starts at 3E-9 at 1s, but is only 1E-11 at 1000s a factor 10 worse than I get with the LEA-6T with the same antenna and setup. If anyone have ADEV-MDEV curves to share I would be glad to see what can be achieved with the venus838-T. My conclusion is also that sawtooth correction is useless on my 838-T.

Are you talking about the PPS output or the frequency output (10 MHz by default)?

I haven’t attempted to get ADEV plots of the PPS output mainly because I’m not sure the best experimental setup.

I could (try to) capture time differences between the PPS output of a thunderbolt and the PPS output of the Venus, but would taking the ADEV of that give correct results?

> Lars
>> Nick wrote:
>> Jim Miller's 10 kHz GPSDO that’s been referenced here has either solved this problem, or the 10 kHz output of the >Jupiter is substantially better than the Venus’ 10 MHz output, or the design doesn’t give the results time-nuts expect >from a GPSDO. Which of those applies?

-------------- next part --------------
A non-text attachment was scrubbed...
Name: Venus 838T 200secs.PNG
Type: image/png
Size: 38357 bytes
Desc: Venus 838T 200secs.PNG
URL: <http://www.febo.com/pipermail/time-nuts/attachments/20161006/4e8d0cc3/attachment-0002.png>
-------------- next part --------------
A non-text attachment was scrubbed...
Name: Venus 838T 32000secs.PNG
Type: image/png
Size: 97959 bytes
Desc: Venus 838T 32000secs.PNG
URL: <http://www.febo.com/pipermail/time-nuts/attachments/20161006/4e8d0cc3/attachment-0003.png>

More information about the time-nuts mailing list