Re: [time-nuts] Timepod Phase Noise Measurements and 3 corner hat
On Sunday, August 30, 2015 05:06:38 PM James Flynn wrote: Martyn Smith martyn@... writes: MEASUREMENT TECHNIQUE We have three sources close in frequency (within 1 x 10E-8 of each other). Two sources are connected to the Ch0 and Ch 2 inputs of the timepod, the unit under test to the ref input. According to John, the result will be the actual phase noise of the ref input, even if it is lower than the two other sources. No further calculations need to be made. The is a publication by the NPL that talks about a similar technique, but comes to a somewhat different conclusion. http://publications.npl.co.uk/npl_web/pdf/mgpg68.pdf Bottom line: it holds that the phase noise of all three sources can be determined if they are taken two at a time; one as ref and the other as the DUT. ___ time-nuts mailing list -- time-nuts@febo.com To unsubscribe, go to https://www.febo.com/cgi-bin/mailman/listinfo/time-nuts and follow the instructions there. How can one have any faith in a paper that cant even get the fundamentals of resistor noise correct? It also perpetuates the myth that Ulrich Rodhe cofounded Rodhe and Schwarz. Lothar Rodhe (Ulrich's father I believe) cofounded Rodhe and Schwarz. Bruce ___ time-nuts mailing list -- time-nuts@febo.com To unsubscribe, go to https://www.febo.com/cgi-bin/mailman/listinfo/time-nuts and follow the instructions there.
Re: [time-nuts] Timepod Phase Noise Measurements and 3 corner hat
The is a publication by the NPL that talks about a similar technique, but comes to a somewhat different conclusion. http://publications.npl.co.uk/npl_web/pdf/mgpg68.pdf Bottom line: it holds that the phase noise of all three sources can be determined if they are taken two at a time; one as ref and the other as the DUT. Yes, that's the three-cornered hat technique. The document is a bit dated in that respect -- it was published in 2004, just as Timing Solutions was starting to work on cross-correlated direct digital measurements. The dual-reference method always converges on the DUT noise if set up properly, but it doesn't give you any insight into the two independent sources being used as references. You have to swap the DUT with each of the references and repeat the measurement if you want to characterize all three sources, while the N-cornered hat returns separated variances for all sources at once (at least ideally.) -- john, KE5FX Miles Design LLC ___ time-nuts mailing list -- time-nuts@febo.com To unsubscribe, go to https://www.febo.com/cgi-bin/mailman/listinfo/time-nuts and follow the instructions there.
Re: [time-nuts] Timepod Phase Noise Measurements and 3 corner hat
John, On 08/31/2015 12:37 AM, John Miles wrote: The is a publication by the NPL that talks about a similar technique, but comes to a somewhat different conclusion. http://publications.npl.co.uk/npl_web/pdf/mgpg68.pdf Bottom line: it holds that the phase noise of all three sources can be determined if they are taken two at a time; one as ref and the other as the DUT. Yes, that's the three-cornered hat technique. The document is a bit dated in that respect -- it was published in 2004, just as Timing Solutions was starting to work on cross-correlated direct digital measurements. The dual-reference method always converges on the DUT noise if set up properly, but it doesn't give you any insight into the two independent sources being used as references. You have to swap the DUT with each of the references and repeat the measurement if you want to characterize all three sources, while the N-cornered hat returns separated variances for all sources at once (at least ideally.) Indeed. On that note, did you look more closely on the NIST analysis of cross-correlation and a possible cancellation and thus overly optimistic results? Did it have any consequence on your code? What did you take away from it? Cheers, Magnus ___ time-nuts mailing list -- time-nuts@febo.com To unsubscribe, go to https://www.febo.com/cgi-bin/mailman/listinfo/time-nuts and follow the instructions there.
Re: [time-nuts] Timepod Phase Noise Measurements and 3 corner hat
On that note, did you look more closely on the NIST analysis of cross-correlation and a possible cancellation and thus overly optimistic results? Did it have any consequence on your code? What did you take away from it? Yes, you can definitely get divots in the PN trace, especially in multiple-hour runs needed to reach very low noise levels. I've seen that on occasion when making measurements with independent downconverters. If the DUT signal experiences significant phase shift in one downconverter path relative to the other, I imagine that's a good way to provoke this behavior. Fortunately I haven't run into any instances of the worst-case scenario described in the Nelson paper, where the whole noise floor collapses without any exhibiting any other weird artifacts. The phenomenon is certainly worth keeping in mind but most people are not going to run into it, especially with the Ch0 and Ch2 jumpers in place. -- john, KE5FX Miles Design LLC ___ time-nuts mailing list -- time-nuts@febo.com To unsubscribe, go to https://www.febo.com/cgi-bin/mailman/listinfo/time-nuts and follow the instructions there.
Re: [time-nuts] Timepod Phase Noise Measurements and 3 corner hat
Hi John, On 08/31/2015 05:12 AM, John Miles wrote: On that note, did you look more closely on the NIST analysis of cross-correlation and a possible cancellation and thus overly optimistic results? Did it have any consequence on your code? What did you take away from it? Yes, you can definitely get divots in the PN trace, especially in multiple-hour runs needed to reach very low noise levels. I've seen that on occasion when making measurements with independent downconverters. If the DUT signal experiences significant phase shift in one downconverter path relative to the other, I imagine that's a good way to provoke this behavior. Fortunately I haven't run into any instances of the worst-case scenario described in the Nelson paper, where the whole noise floor collapses without any exhibiting any other weird artifacts. The phenomenon is certainly worth keeping in mind but most people are not going to run into it, especially with the Ch0 and Ch2 jumpers in place. I'm the kind of guy that keeps those jumpers off and feeding three signals, so that's why I am asking. :) Cheers, Magnus ___ time-nuts mailing list -- time-nuts@febo.com To unsubscribe, go to https://www.febo.com/cgi-bin/mailman/listinfo/time-nuts and follow the instructions there.
[time-nuts] Timepod Phase Noise Measurements and 3 corner hat
Hello, HISTORY I make GPS/GNSS frequency standards (ultra low phase noise types) and I always have to prove my results to the customer. John Miles told me how to make absolute phase noise measurements using three sources with the timepod. MEASUREMENT TECHNIQUE We have three sources close in frequency (within 1 x 10E-8 of each other). Two sources are connected to the Ch0 and Ch 2 inputs of the timepod, the unit under test to the ref input. According to John, the result will be the actual phase noise of the ref input, even if it is lower than the two other sources. No further calculations need to be made. John asked me for results. First of all I tried three sources very close in frequency (1 x 10E-12). I did measurements at bandwidths of 0.5Hz, 5 Hz and 50 Hz. The reason I did this is because the actual two reference sources I wanted to use, weren't going to be exactly the same frequency, but 1 x 10E-8 apart. My results showed me that the bandwidth made no difference to the measurements. So I now made three sets of measurements with unit A, unit B and using references units C and D. As they weren't exactly on frequency I chose a 50 Hz bandwidth, so the frequency differences didn't come into play. So I measured unit A and unit B using the above technique. Then I used the normal timepod technique and connected unit A to the main input, unit B to the ref input and recoupled the two SMA jumpers. So I have three results, unit A's phase noise, unit B's phase noise and the phase noise of both unit A and B together. RESULTS I got the following at offsets of 1/10/100/1k/10k/100kHz: Unit A made -112.6 / -141.3/ -160.0 / -165.4/ -166.9 / -167.1 Unit B made -113.7 / -138.9 / -160.0 / -167.2 / -168.7 / -168.9 Both together made: -110.2 / -137.0 / -156.6 / -163.1 / -164.6 / -164.8 CHECKING RESULTS So if we combine units A's and unit B's individual phase noise, then compare the combined result phase noise, is the combined result 3 dB higher, as it should be. I did the maths and the combined measurements were exactly 3 dB higher (plus/minus 0.1 dB) except at 100 Hz offset which differed by 0.4 dB. But 100 Hz is always difficult because I made these measurements on a 50 Hz system so there is always a 100 Hz spike to contend with CONCLUSION John Miles technique works well. I now get immediate results of my frequency standards as I will always use the 3 source method of measurement on the timepod. John have saved me about 10 hours a week of measurements!!! Regards Martyn Smith ___ time-nuts mailing list -- time-nuts@febo.com To unsubscribe, go to https://www.febo.com/cgi-bin/mailman/listinfo/time-nuts and follow the instructions there.
Re: [time-nuts] Timepod Phase Noise Measurements and 3 corner hat
Martyn, Thanks for this report. This is exactly why I invested in the time-pod and why I use 2 reference sources for measurements. It would be interesting to see papers on the validation of these measurements. Cheers, Magnus On 08/29/2015 01:54 PM, Martyn Smith wrote: Hello, HISTORY I make GPS/GNSS frequency standards (ultra low phase noise types) and I always have to prove my results to the customer. John Miles told me how to make absolute phase noise measurements using three sources with the timepod. MEASUREMENT TECHNIQUE We have three sources close in frequency (within 1 x 10E-8 of each other). Two sources are connected to the Ch0 and Ch 2 inputs of the timepod, the unit under test to the ref input. According to John, the result will be the actual phase noise of the ref input, even if it is lower than the two other sources. No further calculations need to be made. John asked me for results. First of all I tried three sources very close in frequency (1 x 10E-12). I did measurements at bandwidths of 0.5Hz, 5 Hz and 50 Hz. The reason I did this is because the actual two reference sources I wanted to use, weren't going to be exactly the same frequency, but 1 x 10E-8 apart. My results showed me that the bandwidth made no difference to the measurements. So I now made three sets of measurements with unit A, unit B and using references units C and D. As they weren't exactly on frequency I chose a 50 Hz bandwidth, so the frequency differences didn't come into play. So I measured unit A and unit B using the above technique. Then I used the normal timepod technique and connected unit A to the main input, unit B to the ref input and recoupled the two SMA jumpers. So I have three results, unit A's phase noise, unit B's phase noise and the phase noise of both unit A and B together. RESULTS I got the following at offsets of 1/10/100/1k/10k/100kHz: Unit A made -112.6 / -141.3/ -160.0 / -165.4/ -166.9 / -167.1 Unit B made -113.7 / -138.9 / -160.0 / -167.2 / -168.7 / -168.9 Both together made: -110.2 / -137.0 / -156.6 / -163.1 / -164.6 / -164.8 CHECKING RESULTS So if we combine units A's and unit B's individual phase noise, then compare the combined result phase noise, is the combined result 3 dB higher, as it should be. I did the maths and the combined measurements were exactly 3 dB higher (plus/minus 0.1 dB) except at 100 Hz offset which differed by 0.4 dB. But 100 Hz is always difficult because I made these measurements on a 50 Hz system so there is always a 100 Hz spike to contend with CONCLUSION John Miles technique works well. I now get immediate results of my frequency standards as I will always use the 3 source method of measurement on the timepod. John have saved me about 10 hours a week of measurements!!! Regards Martyn Smith ___ time-nuts mailing list -- time-nuts@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@febo.com To unsubscribe, go to https://www.febo.com/cgi-bin/mailman/listinfo/time-nuts and follow the instructions there.
