[time-nuts] Re: Question about GPS 1PPS ADEV
On 4/16/22 15:53, Matthias Welwarsky wrote: John, what constellations did you have enabled during the test? Just GPS, or also others? I usually run with GPS and Galileo enabled (and I avoid GLONASS, it messes everything up). Can this make a difference? I'm afraid I'm not certain. Normally, I do all my work with GPS only, as for timing purposes adding in the other constellations can result in clock-jumping and other issues. But at the beginning of the paper I said that all receivers were set to default configuration except for setting to 0D timing mode when available. So I don't know if that means I left all the default constellations were enabled. My gut feeling, though, is that I probably turned everything off but GPS. John ___ time-nuts mailing list -- time-nuts@lists.febo.com -- To unsubscribe send an email to time-nuts-le...@lists.febo.com To unsubscribe, go to and follow the instructions there.
[time-nuts] Re: Question about GPS 1PPS ADEV
Hi Any time you look inside a control loop, there will be bumps and the like that are a function of the control loop plus the dynamics of the device being controlled plus the environment. If you have a straight line … something is wrong. Bob > On Apr 16, 2022, at 1:20 PM, Matthias Welwarsky > wrote: > > On Samstag, 16. April 2022 17:55:17 CEST John Ackermann N8UR wrote: >> Finally, it looks like you're comparing the raw PPS with an oscillator >> that is steered by that same PPS. I have to think their correlation >> could lead to possible and unpredictable errors. It would be better to >> have the OCXO remain unsteered during the measurement. > > That's correct. But I'm not trying to characterize the OCXO or the GNSS > receiver. I was just wondering about this perceived discrepancy. > >> >> Best, >> John >> >> >> >>> Best regards, >>> Matthias >>> >>> https://hamsci.org/sites/default/files/publications/2020_TAPR_DCC/ >>> N8UR_GPS_Evaluation_August2020.pdf >>> >>> >>> ___ >>> time-nuts mailing list -- time-nuts@lists.febo.com -- To unsubscribe send >>> an email to time-nuts-le...@lists.febo.com > To unsubscribe, go to and >>> follow the instructions there. >> >> ___ >> time-nuts mailing list -- time-nuts@lists.febo.com -- To unsubscribe send an >> email to time-nuts-le...@lists.febo.com To unsubscribe, go to and follow >> the instructions there. > > > > ___ > time-nuts mailing list -- time-nuts@lists.febo.com -- To unsubscribe send an > email to time-nuts-le...@lists.febo.com > To unsubscribe, go to and follow the instructions there. ___ time-nuts mailing list -- time-nuts@lists.febo.com -- To unsubscribe send an email to time-nuts-le...@lists.febo.com To unsubscribe, go to and follow the instructions there.
[time-nuts] Re: Question about GPS 1PPS ADEV
Hi > On Apr 16, 2022, at 11:55 AM, John Ackermann N8UR wrote: > > > > On 4/16/22 09:52, Matthias Welwarsky wrote: >> Dear list members, >> in 2020 John Ackermann published an evaluative survey of current day GPS and >> GNSS receivers (URL below). I have a question about figure 26, which shows, >> among others, the ADEV of a NEO-M8T against a Cesium reference, with >> quantization correction applied. The curve shows a significant "bulge" above >> 1e-10 between about 10s and 100s tau. >> I'm using a LEA-M8T in my DIY GPSDO, which I think is the same chipset in a >> slightly different package. I have attached an image worth about 3000 seconds >> of data, raw 1PPS phase difference against the LO used in the GPSDO. The >> GPSDO >> is locked, not in hold-over mode. >> The bulge between 10s and 100s is not really visible here. There is a slight >> bend, but not as pronounced. My explanation is that this is due to the LO >> being pulled by the GNSS receiver so that it is no longer fully visible. I >> reason that, were the LO more stable, more loosely coupled to the GNSS, I >> should see the bulge from figure 26. Would you agree? > > My theory of the ADEV flat spot when the M8T is qErr corrected is that > multiple things contribute to noise on the PPS output, and the qErr is only > one of them, and it consists of a noise source (clock granularity) that is > unrelated to any external analog process. > > Since Qerr has a fixed limit of ± XX nanoseconds (half the receiver clock > granularity), its contribution to the PPS noise decreases with longer > averaging times. Meanwhile, the other sources of noise such as ionosphere, > etc., are slower and become more pronounced as tau increases. At around 30 > seconds, the external noise factors become larger than the qErr. > > So at short tau cancelling out the qErr gets rid of a major noise source and > improves ADEV but as tau increases beyond ~30 seconds the qErr contribution > is gradually outweighed by the other noise sources and disappears, returning > the ADEV slope to its normal -1. > > Also note that in Fig. 26, at tau below 3 seconds it's possible that both the > M8T and F9T corrected plots are limited by the TICC resolution, which is > around 8e-11 @ 1 second on a good day. That's below these traces, but may > still be high enough to impact the measurement. > > Finally, it looks like you're comparing the raw PPS with an oscillator that > is steered by that same PPS. I have to think their correlation could lead to > possible and unpredictable errors. It would be better to have the OCXO > remain unsteered during the measurement. The only way to make meaningful single comparison measurements of what’s going on is to compare to an external “free running” standard that has a stability adequate to the task. Normally this means an ADEV 5 to 10X better than what you expect to see on the device under test. Bob > > Best, > John > > >> Best regards, >> Matthias >> https://hamsci.org/sites/default/files/publications/2020_TAPR_DCC/ >> N8UR_GPS_Evaluation_August2020.pdf >> ___ >> time-nuts mailing list -- time-nuts@lists.febo.com -- To unsubscribe send an >> email to time-nuts-le...@lists.febo.com >> To unsubscribe, go to and follow the instructions there. > ___ > time-nuts mailing list -- time-nuts@lists.febo.com -- To unsubscribe send an > email to time-nuts-le...@lists.febo.com > To unsubscribe, go to and follow the instructions there. ___ time-nuts mailing list -- time-nuts@lists.febo.com -- To unsubscribe send an email to time-nuts-le...@lists.febo.com To unsubscribe, go to and follow the instructions there.
[time-nuts] Re: Question about GPS 1PPS ADEV
On Samstag, 16. April 2022 17:55:17 CEST John Ackermann N8UR wrote: > Finally, it looks like you're comparing the raw PPS with an oscillator > that is steered by that same PPS. I have to think their correlation > could lead to possible and unpredictable errors. It would be better to > have the OCXO remain unsteered during the measurement. That's correct. But I'm not trying to characterize the OCXO or the GNSS receiver. I was just wondering about this perceived discrepancy. > > Best, > John > > > > > Best regards, > > Matthias > > > > https://hamsci.org/sites/default/files/publications/2020_TAPR_DCC/ > > N8UR_GPS_Evaluation_August2020.pdf > > > > > > ___ > > time-nuts mailing list -- time-nuts@lists.febo.com -- To unsubscribe send > > an email to time-nuts-le...@lists.febo.com To unsubscribe, go to and > > follow the instructions there. > > ___ > time-nuts mailing list -- time-nuts@lists.febo.com -- To unsubscribe send an > email to time-nuts-le...@lists.febo.com To unsubscribe, go to and follow > the instructions there. ___ time-nuts mailing list -- time-nuts@lists.febo.com -- To unsubscribe send an email to time-nuts-le...@lists.febo.com To unsubscribe, go to and follow the instructions there.
[time-nuts] Re: Question about GPS 1PPS ADEV
On 4/16/22 09:52, Matthias Welwarsky wrote: Dear list members, in 2020 John Ackermann published an evaluative survey of current day GPS and GNSS receivers (URL below). I have a question about figure 26, which shows, among others, the ADEV of a NEO-M8T against a Cesium reference, with quantization correction applied. The curve shows a significant "bulge" above 1e-10 between about 10s and 100s tau. I'm using a LEA-M8T in my DIY GPSDO, which I think is the same chipset in a slightly different package. I have attached an image worth about 3000 seconds of data, raw 1PPS phase difference against the LO used in the GPSDO. The GPSDO is locked, not in hold-over mode. The bulge between 10s and 100s is not really visible here. There is a slight bend, but not as pronounced. My explanation is that this is due to the LO being pulled by the GNSS receiver so that it is no longer fully visible. I reason that, were the LO more stable, more loosely coupled to the GNSS, I should see the bulge from figure 26. Would you agree? My theory of the ADEV flat spot when the M8T is qErr corrected is that multiple things contribute to noise on the PPS output, and the qErr is only one of them, and it consists of a noise source (clock granularity) that is unrelated to any external analog process. Since Qerr has a fixed limit of ± XX nanoseconds (half the receiver clock granularity), its contribution to the PPS noise decreases with longer averaging times. Meanwhile, the other sources of noise such as ionosphere, etc., are slower and become more pronounced as tau increases. At around 30 seconds, the external noise factors become larger than the qErr. So at short tau cancelling out the qErr gets rid of a major noise source and improves ADEV but as tau increases beyond ~30 seconds the qErr contribution is gradually outweighed by the other noise sources and disappears, returning the ADEV slope to its normal -1. Also note that in Fig. 26, at tau below 3 seconds it's possible that both the M8T and F9T corrected plots are limited by the TICC resolution, which is around 8e-11 @ 1 second on a good day. That's below these traces, but may still be high enough to impact the measurement. Finally, it looks like you're comparing the raw PPS with an oscillator that is steered by that same PPS. I have to think their correlation could lead to possible and unpredictable errors. It would be better to have the OCXO remain unsteered during the measurement. Best, John Best regards, Matthias https://hamsci.org/sites/default/files/publications/2020_TAPR_DCC/ N8UR_GPS_Evaluation_August2020.pdf ___ time-nuts mailing list -- time-nuts@lists.febo.com -- To unsubscribe send an email to time-nuts-le...@lists.febo.com To unsubscribe, go to and follow the instructions there. ___ time-nuts mailing list -- time-nuts@lists.febo.com -- To unsubscribe send an email to time-nuts-le...@lists.febo.com To unsubscribe, go to and follow the instructions there.
[time-nuts] Re: Question about GPS 1PPS ADEV
Hello Matthias, provided your OCXO is sufficiently stable, i.e., better stability than the 1PPS, I agree that you should see said "bulge". Judging from the DAC plot, the EFC voltage is being adjusted within the tau range in which you expect the NEO-M8T bulge (10~100 s), so I'd expect that to have an effect on the TIC ADEV. Could you measure in holdover mode with fixed EFC voltage? Best regards, Carsten On 16.04.22 15:52, Matthias Welwarsky wrote: I reason that, were the LO more stable, more loosely coupled to the GNSS, I should see the bulge from figure 26. Would you agree? ___ time-nuts mailing list -- time-nuts@lists.febo.com -- To unsubscribe send an email to time-nuts-le...@lists.febo.com To unsubscribe, go to and follow the instructions there.
[time-nuts] Re: Looking for a new GNSS-receiver for a disciplined oscillator - Which one to buy?
Hello, On Mon, Mar 14, 2022 at 05:59:58PM +, Elad Wind wrote: > How about the OCP time card? > There should be a few suppliers today. I couldn't find any, as of now. https://github.com/opencomputeproject/Time-Appliance-Project/tree/master/Time-Card still reads: 'However, we are currently working with several suppliers and will have their contact info soon available to allow you to purchase an out-of-the-box ready Time Card.' > If you need a NIC and a GNSS receiver in one look at NVIDIA’s > https://www.opencompute.org/products/441/nvidia-connectx-6-dx-dual-port-100gbe-nic-and-time-card FWIW, the MCX623106GN-CDAT Nvida card is already marked as legacy and EOL: https://docs.nvidia.com/networking/display/ConnectX6DxEN The MCX623106GC-CDAT might be a similar successor (100 GBE with PPS In/Out, Enhanced-SyncE & PTP GM support and GNSS, PCIe 4.0 x16), but I don't even find a dedicated product page for it - let a lone a shop listing. Certainly, it doesn't look cheap. Better availabilty should have the Cisco Nexus NIC GM (the GM formerly known as Exablaze ExaNIC GM), which can be ordered with an OCXO or CSAC: https://exablaze.com/docs/exanic/user-guide/exanic-gm/ https://www.cisco.com/c/en/us/products/collateral/interfaces-modules/nexus-smartnic/datasheet-c78-743824.pdf Also not cheap, especially after the Cisco acquisition. Intel also lists a NIC with GPS onboard option, but it's pretty new, i.e. probably still at the sample/pre-order stage. Best regards, Georg -- '[..] they shall speak with new tongues; They shall take up serpents; and if they drink any deadly thing, it shall not hurt them; [..]' Mark 16:17-18 ;) ___ time-nuts mailing list -- time-nuts@lists.febo.com -- To unsubscribe send an email to time-nuts-le...@lists.febo.com To unsubscribe, go to and follow the instructions there.
[time-nuts] Re: General discussion of PID algorithms applied to GPSDO control loops (continued 1)
Bob kb8tq writes: >>> You mean the FLL and PLL are exclusive of each other ? I guess you are >> right, but I am trying to think "outside the box" and see if there are any >> alternatives. > >You will have two people driving the car at the same time. One hits the >accelerator and the other hits the brakes at the same time. They both can’t >be active *and* feed the EFC at the same time. The practical answer is to >run each during the warmup phase that it makes sense to do so. You _can_ have hybrid steering, but you must assign "weight" to the different contributions, so that they will never oscillate. I normally have found it better to have a big switch which decides who gets to control, based on the (external) circumstances. As a general rule of thumb, FLL's only make sense if the product of the rate at which you measure phase difference, and the jitter-noise when you do, ends up way to the right and above the allan-intercept. Prof. Dave's infamous "Call NIST once a day with a modem" mode in NTPD is a good example: Forget about tracking temperature, XO drift or anything else: Just try to get the average frequency right on a timescale measured in weeks. Poul-Henning PS: When you implement your PLL: The way to void "wind-up" durign startup, is to short the integrator, until the phase error has reached its proper sign. It is surprising how hard it is to write code to spot that, compared to deciding it manually :-) -- Poul-Henning Kamp | UNIX since Zilog Zeus 3.20 p...@freebsd.org | TCP/IP since RFC 956 FreeBSD committer | BSD since 4.3-tahoe Never attribute to malice what can adequately be explained by incompetence. ___ time-nuts mailing list -- time-nuts@lists.febo.com -- To unsubscribe send an email to time-nuts-le...@lists.febo.com To unsubscribe, go to and follow the instructions there.
[time-nuts] Re: General discussion of PID algorithms applied to GPSDO control loops (continued 1)
Hi Even if you *do* have a number for your OCXO EFC, what is that number? The nominal sensitivity is fine. The next layer is the ratio of maximum slope to minimum slope. This is a very different thing than the “linearity” that is normally specified. Even the linearity is something you need to look carefully at the Mil-O-55310 definition. It’s not quite what you would expect it to be. Bottom line: the sensitivity can easily vary 2:1 or even 4:1 on a pretty good part. Bob > On Apr 15, 2022, at 5:13 PM, Skip Withrow wrote: > > One of the things that cannot be ignored are the statements made in > item #10 regarding the DAC and OCXO EFC. If the DAC is 16 bits that > is 2^16=65,536 steps. If the DAC outputs 0-4 volts that is 61uV/step. > > Now you have to take this number and multiply it by your OCXO > sensitivity (not given), and you have the minimum frequency step that > can be made. > > This is the best that you can ever do. So, your TIC/1pps resolution > does not need to be any better than this. Waiting a long time to get > higher resolution may (probably) buys you nothing. > > This is where all the magic comes in - higher resolution DACs, more > stable oscillators, more stable GPS PPS, etc. > > To a first order, overlaying the line from item #6 with the ADEV of > the oscillator used and setting the loop time constant at the > crossover tau works pretty good. > > As has been said many times, if it was easy everybody would be doing it. > > Regards, > Skip Withrow > ___ > time-nuts mailing list -- time-nuts@lists.febo.com -- To unsubscribe send an > email to time-nuts-le...@lists.febo.com > To unsubscribe, go to and follow the instructions there. ___ time-nuts mailing list -- time-nuts@lists.febo.com -- To unsubscribe send an email to time-nuts-le...@lists.febo.com To unsubscribe, go to and follow the instructions there.
