Re: [time-nuts] Long Wave Radio-Frequency standard testing
Hi, Yes outliers removal creates gap in Stable32. The « fill » function can fills gaps with interpolated values. It does not change much the graphs, except in the low Tau area (see attached). Do you know a discussion of impact of outliers removal ? Gilles. > Le 18 janv. 2021 à 22:06, Bob kb8tq a écrit : > > Hi > > As you throw away samples that are far off the mean, you reduce the sample > rate ( or at least create gaps in the record). Dealing with that could be > difficult. > > Bob > >> On Jan 18, 2021, at 1:33 PM, Gilles Clement wrote: >> >>> Hi >>> >>> Very cool !!! >>> >>> The red trace is obviously the one to focus on. Some sort of digital loop >>> that >>> only operates under the “known good” conditions would seem to make sense. >>> >>> Thanks for sharing >>> >>> Bob >> >> Hi, >> I tried something with the idea to consider night records fluctuations as « >> outliers » as compared to day records. >> Indeed the 3 days record mean value is flat and the histogram quite >> gaussian. >> So I processed the 3 days record (green trace) with Stable32’s « Check >> Function », >> while removing outliers with decreasing values of the Sigma Factor. The >> graph below shows the outcome. >> The graph with Sigma=0.8 (blue trace) connects rather well with the 1Day >> record (red trace). >> Would this be a workable approach ? >> Best, >> Gilles. >> >> >> >> >> >> ___ >> time-nuts mailing list -- time-nuts@lists.febo.com >> To unsubscribe, go to >> http://lists.febo.com/mailman/listinfo/time-nuts_lists.febo.com >> and follow the instructions there. > > > ___ > time-nuts mailing list -- time-nuts@lists.febo.com > To unsubscribe, go to > http://lists.febo.com/mailman/listinfo/time-nuts_lists.febo.com > and follow the instructions there. ___ time-nuts mailing list -- time-nuts@lists.febo.com To unsubscribe, go to http://lists.febo.com/mailman/listinfo/time-nuts_lists.febo.com and follow the instructions there.
Re: [time-nuts] Leakage, tinySA
Dana Whitlow said: > Hal, my 10 MHz birdie is at a level of about -95 dBm, which puts it a little > more than 10 db above the noise floor in 3 kHz BW. My noise floor is fuzzy. The bottom is -110, even when the display goes down to -120. The fuzz is ~3 dB. My 10 MHz birdie looks like it is below the -100 dB line, but the text for the marker often prints out something in the high 90s. I think that's a narrow spike that is hard to see sneaking over the line. So I'd call the height of my birdie to be a bit below 10 dB. -- These are my opinions. I hate spam. ___ time-nuts mailing list -- time-nuts@lists.febo.com To unsubscribe, go to http://lists.febo.com/mailman/listinfo/time-nuts_lists.febo.com and follow the instructions there.
Re: [time-nuts] Leakage, tinySA
Hal, my 10 MHz birdie is at a level of about -95 dBm, which puts it a little more than 10 db above the noise floor in 3 kHz BW. I've found that the "spur removal" function sometimes works, but not always, if I change the IF frequency setting to 433.9 MHz. It's mostly bimodal behavior- sometimes the spur remover kills it entirely, but on some sweeps it only reduces the spur level by about 10 dB. And on occasional sweeps the improvement is by some other amount. Dana On Mon, Jan 18, 2021 at 2:47 PM Hal Murray wrote: > > notfad...@gmail.com said: > > FWIW I don't have spurs at 10MHz. > > Have you looked carefully? I have to zoom in on the bandwidth in order to > see > it. Somebody else called it a birdie. > > My unit came from R I'm reasonably sure it's genuine. > > If I look at 10 MHz center and 25 or 100 k span, there is a faint > something > down in the noise. Most of the time, the marker that automagically ends > up on > top of the highest peak is in the middle. At a quick glance, there is > nothing > there. If I look carefully, there is usually a small bump about 3 kHz > wide. > It's easier to see at 25 kHz since it's wider. It's faintly visible on a > waterfall. > > That's with the antenna terminated or when I wander down the street. > There is > a similar thing at 50 MHz. > > > -- > These are my opinions. I hate spam. > > > > > ___ > time-nuts mailing list -- time-nuts@lists.febo.com > To unsubscribe, go to > http://lists.febo.com/mailman/listinfo/time-nuts_lists.febo.com > and follow the instructions there. > ___ time-nuts mailing list -- time-nuts@lists.febo.com To unsubscribe, go to http://lists.febo.com/mailman/listinfo/time-nuts_lists.febo.com and follow the instructions there.
Re: [time-nuts] (no subject)
On 1/18/21 10:45 AM, Bob kb8tq wrote: Hi That’s pretty neat !!! At least as I read the posters ( and I could have easily missed something …): Their pulse technique (using the giant telescopes) gives them an instant reading on the distance / delay to the cubesat. Even if there are propagation issues, they are measured (and eliminated) in that pulse process. With GPS, we don’t get that sort of live data. In terms of setting up something to use it in my back yard ….yikes … that’s one mighty big “antenna” on their ground station. Since you have to deliver enough power to the sat *and* back via reflection to “trigger photo diodes” the “minimum” required optics could be pretty big …. They don’t go into that so who knows …. In the current implementation, this looks like a “one user at a time” sort of system. I suspect that’s fine if the minimum optics are anywhere near as big as what’s in that picture. If this all *can* be done with a backyard sized telescope, scheduling could be interesting …. Bob I'll bet they used those facilities because they were available, not because they needed that much aperture. As an example, the TBIRD mission sent Gbps+ data rates from LEO to Earth with a few cm aperture on the spacecraft and <1 meter on Earth. You can visually see a laser pointer on the Earth's surface from ISS with no additional optics (you can also see a 800+W metal halide spotlight) - there was an experiment where they did this from some fairly dark place in the Southwest desert. The lack of easy multi-user access with two way systems is why GPS is a one way system. ___ time-nuts mailing list -- time-nuts@lists.febo.com To unsubscribe, go to http://lists.febo.com/mailman/listinfo/time-nuts_lists.febo.com and follow the instructions there.
Re: [time-nuts] Long Wave Radio-Frequency standard testing
Hi As you throw away samples that are far off the mean, you reduce the sample rate ( or at least create gaps in the record). Dealing with that could be difficult. Bob > On Jan 18, 2021, at 1:33 PM, Gilles Clement wrote: > >> Hi >> >> Very cool !!! >> >> The red trace is obviously the one to focus on. Some sort of digital loop >> that >> only operates under the “known good” conditions would seem to make sense. >> >> Thanks for sharing >> >> Bob > > Hi, > I tried something with the idea to consider night records fluctuations as « > outliers » as compared to day records. > Indeed the 3 days record mean value is flat and the histogram quite gaussian. > So I processed the 3 days record (green trace) with Stable32’s « Check > Function », > while removing outliers with decreasing values of the Sigma Factor. The graph > below shows the outcome. > The graph with Sigma=0.8 (blue trace) connects rather well with the 1Day > record (red trace). > Would this be a workable approach ? > Best, > Gilles. > > > > > > ___ > time-nuts mailing list -- time-nuts@lists.febo.com > To unsubscribe, go to > http://lists.febo.com/mailman/listinfo/time-nuts_lists.febo.com > and follow the instructions there. ___ time-nuts mailing list -- time-nuts@lists.febo.com To unsubscribe, go to http://lists.febo.com/mailman/listinfo/time-nuts_lists.febo.com and follow the instructions there.
Re: [time-nuts] Leakage, tinySA
notfad...@gmail.com said: > FWIW I don't have spurs at 10MHz. Have you looked carefully? I have to zoom in on the bandwidth in order to see it. Somebody else called it a birdie. My unit came from R I'm reasonably sure it's genuine. If I look at 10 MHz center and 25 or 100 k span, there is a faint something down in the noise. Most of the time, the marker that automagically ends up on top of the highest peak is in the middle. At a quick glance, there is nothing there. If I look carefully, there is usually a small bump about 3 kHz wide. It's easier to see at 25 kHz since it's wider. It's faintly visible on a waterfall. That's with the antenna terminated or when I wander down the street. There is a similar thing at 50 MHz. -- These are my opinions. I hate spam. ___ time-nuts mailing list -- time-nuts@lists.febo.com To unsubscribe, go to http://lists.febo.com/mailman/listinfo/time-nuts_lists.febo.com and follow the instructions there.
Re: [time-nuts] (no subject)
On 1/18/21 7:53 AM, Skip Withrow wrote: Hello time-nuts, Came across this poster of the CHOMPTT cubesat mission. https://digitalcommons.usu.edu/cgi/viewcontent.cgi?article=3494=smallsat Obviously, they are doing measurements on a per orbit basis. But if you had an optical ground station and the clock difference information (correction) you could calibrate your home clock. Basically, this is what we already do with GPS with CORS correction information to get much better than 20ns accuracy. One question that I have is the uncertainty with optical paths lower than for GPS (RF)? I suspect maybe. Almost certainly better - at L-band, the ionization of the ionosphere is a big uncertainty, both because the propagation velocity changes as the ion concentration, but because that effect can cause refraction, so the "ray path" from satellite to you isn't straight. Dual frequency receivers can solve for the velocity difference (since it's inversely proportional to frequency squared) but they can't solve for the non-straight-path. (https://apps.dtic.mil/dtic/tr/fulltext/u2/a142725.pdf, Klobuchar 1983) For optical paths, there are some refractive effects close to the horizon but at elevation angles that are higher, they are much reduced. A system such as this might also provide an independent source of time to GPS. But don't we already have that with GLONASS, Beidou, and Galileo? Yes. However, there's always interest in time transfer that doesn't use GNSS, particularly in space applications. more on CHOMPTT https://digitalcommons.usu.edu/smallsat/2017/all2017/49/ is from 2017 (pre launch) https://digitalcommons.usu.edu/smallsat/2019/all2019/108/ is from 2019 (post launch) They used the same "narrow temperature range" CSAC that I did on my satellites (one of which was on the same launch) Regards, Skip Withrow ___ time-nuts mailing list -- time-nuts@lists.febo.com To unsubscribe, go to http://lists.febo.com/mailman/listinfo/time-nuts_lists.febo.com and follow the instructions there. ___ time-nuts mailing list -- time-nuts@lists.febo.com To unsubscribe, go to http://lists.febo.com/mailman/listinfo/time-nuts_lists.febo.com and follow the instructions there.
Re: [time-nuts] (no subject)
Hi That’s pretty neat !!! At least as I read the posters ( and I could have easily missed something …): Their pulse technique (using the giant telescopes) gives them an instant reading on the distance / delay to the cubesat. Even if there are propagation issues, they are measured (and eliminated) in that pulse process. With GPS, we don’t get that sort of live data. In terms of setting up something to use it in my back yard ….yikes … that’s one mighty big “antenna” on their ground station. Since you have to deliver enough power to the sat *and* back via reflection to “trigger photo diodes” the “minimum” required optics could be pretty big …. They don’t go into that so who knows …. In the current implementation, this looks like a “one user at a time” sort of system. I suspect that’s fine if the minimum optics are anywhere near as big as what’s in that picture. If this all *can* be done with a backyard sized telescope, scheduling could be interesting …. Bob > On Jan 18, 2021, at 10:53 AM, Skip Withrow wrote: > > Hello time-nuts, > > Came across this poster of the CHOMPTT cubesat mission. > https://digitalcommons.usu.edu/cgi/viewcontent.cgi?article=3494=smallsat > > Obviously, they are doing measurements on a per orbit basis. But if > you had an optical ground station and the clock difference information > (correction) you could calibrate your home clock. Basically, this is > what we already do with GPS with CORS correction information to get > much better than 20ns accuracy. One question that I have is the > uncertainty with optical paths lower than for GPS (RF)? I suspect > maybe. > > A system such as this might also provide an independent source of time > to GPS. But don't we already have that with GLONASS, Beidou, and > Galileo? > > Regards, > Skip Withrow > > ___ > time-nuts mailing list -- time-nuts@lists.febo.com > To unsubscribe, go to > http://lists.febo.com/mailman/listinfo/time-nuts_lists.febo.com > and follow the instructions there. ___ time-nuts mailing list -- time-nuts@lists.febo.com To unsubscribe, go to http://lists.febo.com/mailman/listinfo/time-nuts_lists.febo.com and follow the instructions there.
Re: [time-nuts] Long Wave Radio-Frequency standard testing
> Hi > > Very cool !!! > > The red trace is obviously the one to focus on. Some sort of digital loop that > only operates under the “known good” conditions would seem to make sense. > > Thanks for sharing > > Bob Hi, I tried something with the idea to consider night records fluctuations as « outliers » as compared to day records. Indeed the 3 days record mean value is flat and the histogram quite gaussian. So I processed the 3 days record (green trace) with Stable32’s « Check Function », while removing outliers with decreasing values of the Sigma Factor. The graph below shows the outcome. The graph with Sigma=0.8 (blue trace) connects rather well with the 1Day record (red trace). Would this be a workable approach ? Best, Gilles. ___ time-nuts mailing list -- time-nuts@lists.febo.com To unsubscribe, go to http://lists.febo.com/mailman/listinfo/time-nuts_lists.febo.com and follow the instructions there.
[time-nuts] (no subject)
Hello time-nuts, Came across this poster of the CHOMPTT cubesat mission. https://digitalcommons.usu.edu/cgi/viewcontent.cgi?article=3494=smallsat Obviously, they are doing measurements on a per orbit basis. But if you had an optical ground station and the clock difference information (correction) you could calibrate your home clock. Basically, this is what we already do with GPS with CORS correction information to get much better than 20ns accuracy. One question that I have is the uncertainty with optical paths lower than for GPS (RF)? I suspect maybe. A system such as this might also provide an independent source of time to GPS. But don't we already have that with GLONASS, Beidou, and Galileo? Regards, Skip Withrow ___ time-nuts mailing list -- time-nuts@lists.febo.com To unsubscribe, go to http://lists.febo.com/mailman/listinfo/time-nuts_lists.febo.com and follow the instructions there.
Re: [time-nuts] Leakage, tinySA
BTW, I did purchase my 10MHz-"leaky" TinySA through R & L. Had I not had the clue that R & L was a reliable source, I would never have ordered one ot these things at all. Tom, I don't walk around- I pedal around on a recumbent trike. Dana On Mon, Jan 18, 2021 at 5:20 AM Bill Notfaded wrote: > Knock offs is part of the reason I posted the main US distributer earlier. > The software is basically open source so anyone could theoretically make > one. There's only one original manufacturer though. R Electronics is > the only sanctioned US distributer I'm aware of. Also they have the best > price in US you can find for obvious reasons. FWIW I don't have spurs at > 10MHz. > > Bill in AZ > > On Sun, Jan 17, 2021, 5:35 PM Dana Whitlow wrote: > > > Hi Dan, > > > > I remember seeing the list of checks to make regarding knockoff clones, > > and even remember a few of them: to wit: > > > > 1. Comes in molded packaging, not poly foam. Check! > > 2. Has the internal shielding. Check! > > 3. Passes all the self tests, including 7 & 8. Check! > > > > I had high hopes of moving the birdie off 10 MHz by changing the IF > > frequency > > setting, but this did no good. I'd sure like to see a detailed block > > diagram, but > > do not know where to look. > > > > I also tried turning on the "spur eliminator", which did help by several > > dB, but > > my birdie still sits a few dB able the displayed noise floor in the > > narrowest > > available resolution BW of ~ 3 kHz. > > > > So, for the moment, I'm out of ideas. > > > > Thanks, > > > > Dana > > > > > > On Sun, Jan 17, 2021 at 5:21 PM Dan Kemppainen > > wrote: > > > > > Dana, > > > > > > Just an FYI. Apparently, there are knockoff clone copies of the TinySA > > > out there. (Yes, making a knockoff clone of a $50 spectrum analyzer is > a > > > thing, go figure!) > > > > > > Apparently, these have substandard performance, and a host of > > > issues/concerns. I ran into this on the TinySA web site. They have some > > > hints on how to detect if you have a clone or not. Might be worth > > > looking into to see if you have a sub-par copy. > > > > > > Dan > > > > > > > > > > > > > > > On 1/16/2021 12:00 PM, time-nuts-requ...@lists.febo.com wrote: > > > > Date: Sat, 16 Jan 2021 05:45:19 -0600 > > > > From: Dana Whitlow > > > > To: Discussion of precise time and frequency measurement > > > > > > > > Subject: Re: [time-nuts] Leakage, tinySA > > > > Message-ID: > > > >> > qrjlg-bw68wgdahmlhe9qulpupijuqedapz...@mail.gmail.com> > > > > Content-Type: text/plain; charset="UTF-8" > > > > > > > > Another data point: My newly-received TinySA has a big birdie at 10 > > MHz, > > > > which is a > > > > disappointment since I had hoped to use the unit as a sniffer to > track > > > down > > > > leaks in my > > > > 10 MHz distribution system. This birdie in the TinySA persists even > > when > > > > nothing is > > > > connected to the input and does not vary in amplitude when I move > > around > > > > the house, > > > > or with time while I'm sitting still with the unit. > > > > > > > > I, too, received no manual. > > > > > > > > Dana > > > > > > ___ > > > time-nuts mailing list -- time-nuts@lists.febo.com > > > To unsubscribe, go to > > > http://lists.febo.com/mailman/listinfo/time-nuts_lists.febo.com > > > and follow the instructions there. > > > > > ___ > > time-nuts mailing list -- time-nuts@lists.febo.com > > To unsubscribe, go to > > http://lists.febo.com/mailman/listinfo/time-nuts_lists.febo.com > > and follow the instructions there. > > > ___ > time-nuts mailing list -- time-nuts@lists.febo.com > To unsubscribe, go to > http://lists.febo.com/mailman/listinfo/time-nuts_lists.febo.com > and follow the instructions there. > ___ time-nuts mailing list -- time-nuts@lists.febo.com To unsubscribe, go to http://lists.febo.com/mailman/listinfo/time-nuts_lists.febo.com and follow the instructions there.
Re: [time-nuts] Leakage, tinySA
Knock offs is part of the reason I posted the main US distributer earlier. The software is basically open source so anyone could theoretically make one. There's only one original manufacturer though. R Electronics is the only sanctioned US distributer I'm aware of. Also they have the best price in US you can find for obvious reasons. FWIW I don't have spurs at 10MHz. Bill in AZ On Sun, Jan 17, 2021, 5:35 PM Dana Whitlow wrote: > Hi Dan, > > I remember seeing the list of checks to make regarding knockoff clones, > and even remember a few of them: to wit: > > 1. Comes in molded packaging, not poly foam. Check! > 2. Has the internal shielding. Check! > 3. Passes all the self tests, including 7 & 8. Check! > > I had high hopes of moving the birdie off 10 MHz by changing the IF > frequency > setting, but this did no good. I'd sure like to see a detailed block > diagram, but > do not know where to look. > > I also tried turning on the "spur eliminator", which did help by several > dB, but > my birdie still sits a few dB able the displayed noise floor in the > narrowest > available resolution BW of ~ 3 kHz. > > So, for the moment, I'm out of ideas. > > Thanks, > > Dana > > > On Sun, Jan 17, 2021 at 5:21 PM Dan Kemppainen > wrote: > > > Dana, > > > > Just an FYI. Apparently, there are knockoff clone copies of the TinySA > > out there. (Yes, making a knockoff clone of a $50 spectrum analyzer is a > > thing, go figure!) > > > > Apparently, these have substandard performance, and a host of > > issues/concerns. I ran into this on the TinySA web site. They have some > > hints on how to detect if you have a clone or not. Might be worth > > looking into to see if you have a sub-par copy. > > > > Dan > > > > > > > > > > On 1/16/2021 12:00 PM, time-nuts-requ...@lists.febo.com wrote: > > > Date: Sat, 16 Jan 2021 05:45:19 -0600 > > > From: Dana Whitlow > > > To: Discussion of precise time and frequency measurement > > > > > > Subject: Re: [time-nuts] Leakage, tinySA > > > Message-ID: > > >> qrjlg-bw68wgdahmlhe9qulpupijuqedapz...@mail.gmail.com> > > > Content-Type: text/plain; charset="UTF-8" > > > > > > Another data point: My newly-received TinySA has a big birdie at 10 > MHz, > > > which is a > > > disappointment since I had hoped to use the unit as a sniffer to track > > down > > > leaks in my > > > 10 MHz distribution system. This birdie in the TinySA persists even > when > > > nothing is > > > connected to the input and does not vary in amplitude when I move > around > > > the house, > > > or with time while I'm sitting still with the unit. > > > > > > I, too, received no manual. > > > > > > Dana > > > > ___ > > time-nuts mailing list -- time-nuts@lists.febo.com > > To unsubscribe, go to > > http://lists.febo.com/mailman/listinfo/time-nuts_lists.febo.com > > and follow the instructions there. > > > ___ > time-nuts mailing list -- time-nuts@lists.febo.com > To unsubscribe, go to > http://lists.febo.com/mailman/listinfo/time-nuts_lists.febo.com > and follow the instructions there. > ___ time-nuts mailing list -- time-nuts@lists.febo.com To unsubscribe, go to http://lists.febo.com/mailman/listinfo/time-nuts_lists.febo.com and follow the instructions there.
