Re: [time-nuts] temperature
Interestingly, HP for a long time sold"quartz thermometers" based around a probe with a quartz crystal with a well characterized linear temperature coefficient. They called the crystal cut "LC" (Linear Coefficient): http://www.hpl.hp.com/hpjournal/pdfs/IssuePDFs/1965-03.pdf http://en.wikipedia.org/wiki/Quartz_thermometer On Sun, Dec 8, 2013 at 10:55 PM, Neville Michie wrote: > Hi, > I use a HP3468A multimeter to measure a PT100 platinum resistance > thermometer. It gives me resolution of one mK, but calibration is another > matter. > It is best to use a 4 terminal device, but 2 terminal into the 4 terminal > input works well. Thermoelectric effects and the requirement for 1 > microvolt stability > makes wiring them into your own circuit difficult. One of the great > technical difficulties is to get a resistor to compare them against, it > must be very stable, > have no thermoelectric effects and have a temperature coefficient in the > order of one PPM. I always admire the way HP designed their ohm meters. > There are other issues, however. Whereas a volt meter can connect > perfectly to a reference, a PRT can only report its own temperature. > That is no problem when you are working in a well stirred water bath, that > will have the PRT at the same temperature as the object in the same bath. > When you get to measure air temperature you are into serious sampling > errors, the PRT has some self heating and so is air velocity sensitive, and > the air > you are measuring may not be the same air as is over your OCXO or item of > interest. There is a personal plume of warm air rising from an observer, so > you must be careful with your measurement technique. > The same problems occur with quartz crystal thermometers, which is why > they are not more commonly found in surplus. > A PT100 sensor is quite cheap, and their calibration is little short of > brilliant. However a they would cost much more if their calibration is > traceable. > For my use, I use an ice-point cell as a calibration check, with care you > get 10mK accuracy. You only need the knowledge how to set it up, a blender > to make ice slush, > and a picnic vacuum flask, to make your own calibration reference. > I use thermistors for air measurement, and calibrate them against the > PT100 in a thermostatic water bath. Thermistors can be run with a very low > level of self heating and they are very sensitive, their resistance > changes 4% per Centigrade degree, and they come in high values like 100K > ohm. You read > them in a bridge circuit with a voltmeter, so they are many orders of > magnitude easier to use than a 100 ohm PRT. > They are made small enough to get them in close contact > with the object to be measured. > If you want to know about humidity measurement I can tell you much about > that, > cheers, > Neville Michie > > On 08/12/2013, at 12:40 PM, Mark Spencer wrote: > > > Sorry if this is somewhat off topic, but I'd be interested in more > details re precision temperature measurement devices. Have been using an > inexpensive USB temperature sensor for the last year or so to monitor the > temperature in my lab and have been looking at the correlation between > frequency shifts in some ocxo's vs temperature changes. I should also > start taking humidity measurements as well at some point. > > > > > > Any pointers re suitable instruments to accomplish this that can be > sourced via the usual surplus sources would be welcome. > > > > Thanks in advance > > Mark Spencer > > > > Sent from my iPad > > ___ > > 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. > ___ 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] temperature
Hi, I use a HP3468A multimeter to measure a PT100 platinum resistance thermometer. It gives me resolution of one mK, but calibration is another matter. It is best to use a 4 terminal device, but 2 terminal into the 4 terminal input works well. Thermoelectric effects and the requirement for 1 microvolt stability makes wiring them into your own circuit difficult. One of the great technical difficulties is to get a resistor to compare them against, it must be very stable, have no thermoelectric effects and have a temperature coefficient in the order of one PPM. I always admire the way HP designed their ohm meters. There are other issues, however. Whereas a volt meter can connect perfectly to a reference, a PRT can only report its own temperature. That is no problem when you are working in a well stirred water bath, that will have the PRT at the same temperature as the object in the same bath. When you get to measure air temperature you are into serious sampling errors, the PRT has some self heating and so is air velocity sensitive, and the air you are measuring may not be the same air as is over your OCXO or item of interest. There is a personal plume of warm air rising from an observer, so you must be careful with your measurement technique. The same problems occur with quartz crystal thermometers, which is why they are not more commonly found in surplus. A PT100 sensor is quite cheap, and their calibration is little short of brilliant. However a they would cost much more if their calibration is traceable. For my use, I use an ice-point cell as a calibration check, with care you get 10mK accuracy. You only need the knowledge how to set it up, a blender to make ice slush, and a picnic vacuum flask, to make your own calibration reference. I use thermistors for air measurement, and calibrate them against the PT100 in a thermostatic water bath. Thermistors can be run with a very low level of self heating and they are very sensitive, their resistance changes 4% per Centigrade degree, and they come in high values like 100K ohm. You read them in a bridge circuit with a voltmeter, so they are many orders of magnitude easier to use than a 100 ohm PRT. They are made small enough to get them in close contact with the object to be measured. If you want to know about humidity measurement I can tell you much about that, cheers, Neville Michie On 08/12/2013, at 12:40 PM, Mark Spencer wrote: > Sorry if this is somewhat off topic, but I'd be interested in more details re > precision temperature measurement devices. Have been using an inexpensive > USB temperature sensor for the last year or so to monitor the temperature in > my lab and have been looking at the correlation between frequency shifts in > some ocxo's vs temperature changes. I should also start taking humidity > measurements as well at some point. > > > Any pointers re suitable instruments to accomplish this that can be sourced > via the usual surplus sources would be welcome. > > Thanks in advance > Mark Spencer > > Sent from my iPad > ___ > 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.
Re: [time-nuts] Is this measurement for real?
I've been on the FMT forum, I hate those yahoo groups and refuse to re-join them, nearly everytime I would go there, I had to renew my membership or change my password. Not the people, just the interface. I don't believe in yahoo any more and have blocked it from my computer except the few people I get email from at yahoo, but nearly all the bad spam I get comes from yahoo. I tried the TS-2000 group there too, for some reason they have 2 groups, but I found them pretty useless. Not much past CB. Rant over. Dave ___ 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] Is this measurement for real?
On 12/09/2013 03:43 AM, wb6bnq wrote: > Hi Magnus, > > If the TAPR module you are referring to is the ICS-501 thing, that is > really not suitable for this purpose. Equally, neither are the DDS > chips. Besides never being right where you want them you have the > spurii to deal with. Way too much filtering required. These two > items aside from being surface mount have the added requirement of > needing a damn cpu to tell the DDS what to do, not too mention the > additional circuit board space. > > For a fixed frequency requirement it is much easier and cheaper to do > it with a proper PLL and a good but inexpensive VCXO. Which, by the > way, I found a small stash of the VCXO's on the right frequency for > $25 each from a highly reputable US manufacturer (an over stocked > run). And, TOO BOOT, I have the layout all done in through hole > intended to fit in a $15 ABS EMI/RFI shielded box about 6.7 inches by > 3.3 inches. When you can do propper PLL with a good VCO and have the layout, then it is a no-brainer. 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] Is this measurement for real?
Dave - Let me call your attention to the FMT nuts list if you are not already on it (I didn't find an entry from you in the November ARRL FMT). I'm not sure how many of the gurus there are on this one, although the HP 3586B is a tool used by some. I myself depend on some HP gear, a 105B and 8640B. There you can inquire and learn about AM BC stations locked or not to gps, as well as intentional carrier offset applied to mitigate slow deep fading in zones of overlapping coverage. http://groups.yahoo.com/neo/groups/FMT-nuts/info 73, Ed, k1ggi -Original Message- From: time-nuts-boun...@febo.com [mailto:time-nuts-boun...@febo.com] On Behalf Of quartz55 Sent: Sunday, December 08, 2013 21:06 PM To: Discussion of precise time and frequency measurement Subject: Re: [time-nuts] Is this measurement for real? Well, thanks for the comments. I really didn't know what to expect from the SoundBlaster card. I thought AM stations were better than that. I started everything cold today and the 1K from the SM started at around 999.998600 and drifted up to 999.999800 in about an hour and half, stayed there around 4 hours but then started back down to 999.999290 in another 4 hours. It's going back up a bit tonight, it's around 999.999420 now. I do see in SpecLab there is a way to lock the sound card to a 1pps GPSDO. I may try that, but my Nortel only puts out even second pulses and I really haven't seen them yet. Anyhow, after warm up, it looks like I should be able to detect 1 mHz differences given other things are stable, like the DSP in the TS2000, I'm thinking about running the fan full time in it to try to stabalize that. How about the UHF TV stations, are they locked to something, they all seem to be on even .025 KHz on my service monitor which can only resolve to 1 Hz? I'm looking for frequency standards I can receive better than the 2M/440 beacons (N4MW) I can hardly hear from central VA. WWV is about useless, I get 2 Hz spreads from them with SpecLab at 20 MHz, it may be WWVH interefering. I have to say, I've gone from hardly being able to measure 1 Hz differences with the service moniter to nearly mHz differences, that's what 3 decimel places? It's funny how the closer you get the closer you want to get. And now to figure out how accurate it is. I've got a FT897D I'm not using, been thinking about selling it and get Jackson Labs unit, but I'm not sure what that will get me past the Nortel unless the extra satellites/bands will make it more accurate? Dave N3DT ___ 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.
Re: [time-nuts] Is this measurement for real?
Hi Magnus, If the TAPR module you are referring to is the ICS-501 thing, that is really not suitable for this purpose. Equally, neither are the DDS chips. Besides never being right where you want them you have the spurii to deal with. Way too much filtering required. These two items aside from being surface mount have the added requirement of needing a damn cpu to tell the DDS what to do, not too mention the additional circuit board space. For a fixed frequency requirement it is much easier and cheaper to do it with a proper PLL and a good but inexpensive VCXO. Which, by the way, I found a small stash of the VCXO's on the right frequency for $25 each from a highly reputable US manufacturer (an over stocked run). And, TOO BOOT, I have the layout all done in through hole intended to fit in a $15 ABS EMI/RFI shielded box about 6.7 inches by 3.3 inches. But, thanks for the suggestions, BillWB6BNQ Magnus Danielson wrote: Bill, On 12/09/2013 02:30 AM, wb6bnq wrote: Hi Dave, My question was more centered on determining your expectations. I ran into an even worse condition with the cheap sound card, in my shop computer, I used for the Frequency Monitoring Tests (FMT) ran by Connie (K5CM). Being in a space with no temperature control at all, the sound card had a 7.0 Hertz variation over a few minutes. Clearly, it was a crystal going wild. As an experiment, I decided to rip out the crystal and replace it with the output of a HF synthesizer dialed to the proper frequency. My synthesizer, and other LAB equipment, is locked to my house standard which is monitored (not controlled by) with GPS. As expected, the results were spectacular ! I ended up with a measurement process that had a resolution of 120ns, and maybe somewhat less. At 1000 Hertz that is an uncertainty of +/- 1.2e-10. I did not try to account for ground loops or other anomalies; and the sound card was some cheap $18 item with no spectacular ratings in and of itself. I am now finishing up on a project to replace my expensive commercial synthesizer so it can return to test equipment duty. If you are interested in what I am doing in that regard, email me about it off list. As for the TS-2000 radio, I have not studied it, per se. But like a lot of these modern radios there are several possible error points within their design that could cause offsets and drift that may affect the outcome depending upon your application. However, if you are using a common analog detection type radio in the "AM" mode, then the radio does not matter to the outcome. The radio only serves as a mixer, albeit an expensive one. For example, when comparing an approximate 10 MHz unknown signal, the mixing action provides four (4) more decades of resolution if the output of the mix is 1000 Hertz. If using a modern DSP radio in the "AM" mode you may have to account for possible slight errors in the internal codecs (A>D/D>A). The only two error points that matter (using "AM" in the above example) is the local signal generator used to beat against the unknown incoming signal and the computer's sound card stability. If both are tied to your "House Standard," then it is totally up to the quality of your local standard's stability and accuracy. There are dirt cheap DDS-modules to buy from China, based on AD DDS-chip. If you need help with the reference frequency, you might want to use that TAPR module for clock-stepup. I don't think the TAPR module does the frequency you want straight away. Just to give you some ideas to follow up. It will be much cheaper and compact than your RF-generator today, and considering you already is listening to a stepped down signal, the purity is good enough for the purpose. 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 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] Is this measurement for real?
Bill, On 12/09/2013 02:30 AM, wb6bnq wrote: > Hi Dave, > > My question was more centered on determining your expectations. I ran > into an even worse condition with the cheap sound card, in my shop > computer, I used for the Frequency Monitoring Tests (FMT) ran by > Connie (K5CM). Being in a space with no temperature control at all, > the sound card had a 7.0 Hertz variation over a few minutes. Clearly, > it was a crystal going wild. > > As an experiment, I decided to rip out the crystal and replace it with > the output of a HF synthesizer dialed to the proper frequency. My > synthesizer, and other LAB equipment, is locked to my house standard > which is monitored (not controlled by) with GPS. > > As expected, the results were spectacular ! I ended up with a > measurement process that had a resolution of 120ns, and maybe somewhat > less. At 1000 Hertz that is an uncertainty of +/- 1.2e-10. I did not > try to account for ground loops or other anomalies; and the sound card > was some cheap $18 item with no spectacular ratings in and of itself. > > I am now finishing up on a project to replace my expensive commercial > synthesizer so it can return to test equipment duty. If you are > interested in what I am doing in that regard, email me about it off list. > > As for the TS-2000 radio, I have not studied it, per se. But like a > lot of these modern radios there are several possible error points > within their design that could cause offsets and drift that may affect > the outcome depending upon your application. > > However, if you are using a common analog detection type radio in the > "AM" mode, then the radio does not matter to the outcome. The radio > only serves as a mixer, albeit an expensive one. For example, when > comparing an approximate 10 MHz unknown signal, the mixing action > provides four (4) more decades of resolution if the output of the mix > is 1000 Hertz. If using a modern DSP radio in the "AM" mode you may > have to account for possible slight errors in the internal codecs > (A>D/D>A). > > The only two error points that matter (using "AM" in the above > example) is the local signal generator used to beat against the > unknown incoming signal and the computer's sound card stability. If > both are tied to your "House Standard," then it is totally up to the > quality of your local standard's stability and accuracy. There are dirt cheap DDS-modules to buy from China, based on AD DDS-chip. If you need help with the reference frequency, you might want to use that TAPR module for clock-stepup. I don't think the TAPR module does the frequency you want straight away. Just to give you some ideas to follow up. It will be much cheaper and compact than your RF-generator today, and considering you already is listening to a stepped down signal, the purity is good enough for the purpose. 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] Is this measurement for real?
Well, thanks for the comments. I really didn't know what to expect from the SoundBlaster card. I thought AM stations were better than that. I started everything cold today and the 1K from the SM started at around 999.998600 and drifted up to 999.999800 in about an hour and half, stayed there around 4 hours but then started back down to 999.999290 in another 4 hours. It's going back up a bit tonight, it's around 999.999420 now. I do see in SpecLab there is a way to lock the sound card to a 1pps GPSDO. I may try that, but my Nortel only puts out even second pulses and I really haven't seen them yet. Anyhow, after warm up, it looks like I should be able to detect 1 mHz differences given other things are stable, like the DSP in the TS2000, I'm thinking about running the fan full time in it to try to stabalize that. How about the UHF TV stations, are they locked to something, they all seem to be on even .025 KHz on my service monitor which can only resolve to 1 Hz? I'm looking for frequency standards I can receive better than the 2M/440 beacons (N4MW) I can hardly hear from central VA. WWV is about useless, I get 2 Hz spreads from them with SpecLab at 20 MHz, it may be WWVH interefering. I have to say, I've gone from hardly being able to measure 1 Hz differences with the service moniter to nearly mHz differences, that's what 3 decimel places? It's funny how the closer you get the closer you want to get. And now to figure out how accurate it is. I've got a FT897D I'm not using, been thinking about selling it and get Jackson Labs unit, but I'm not sure what that will get me past the Nortel unless the extra satellites/bands will make it more accurate? Dave N3DT ___ 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] Is this measurement for real?
Hi Dave, My question was more centered on determining your expectations. I ran into an even worse condition with the cheap sound card, in my shop computer, I used for the Frequency Monitoring Tests (FMT) ran by Connie (K5CM). Being in a space with no temperature control at all, the sound card had a 7.0 Hertz variation over a few minutes. Clearly, it was a crystal going wild. As an experiment, I decided to rip out the crystal and replace it with the output of a HF synthesizer dialed to the proper frequency. My synthesizer, and other LAB equipment, is locked to my house standard which is monitored (not controlled by) with GPS. As expected, the results were spectacular ! I ended up with a measurement process that had a resolution of 120ns, and maybe somewhat less. At 1000 Hertz that is an uncertainty of +/- 1.2e-10. I did not try to account for ground loops or other anomalies; and the sound card was some cheap $18 item with no spectacular ratings in and of itself. I am now finishing up on a project to replace my expensive commercial synthesizer so it can return to test equipment duty. If you are interested in what I am doing in that regard, email me about it off list. As for the TS-2000 radio, I have not studied it, per se. But like a lot of these modern radios there are several possible error points within their design that could cause offsets and drift that may affect the outcome depending upon your application. However, if you are using a common analog detection type radio in the "AM" mode, then the radio does not matter to the outcome. The radio only serves as a mixer, albeit an expensive one. For example, when comparing an approximate 10 MHz unknown signal, the mixing action provides four (4) more decades of resolution if the output of the mix is 1000 Hertz. If using a modern DSP radio in the "AM" mode you may have to account for possible slight errors in the internal codecs (A>D/D>A). The only two error points that matter (using "AM" in the above example) is the local signal generator used to beat against the unknown incoming signal and the computer's sound card stability. If both are tied to your "House Standard," then it is totally up to the quality of your local standard's stability and accuracy. BillWB6BNQ quartz55 wrote: Well, Bill, I just don't know. I think that's why I asked. However, I did measure it again over 2.5 hrs and I got a roughly 500uHz drift pretty much all in one direction. I'm wondering if this is telling me I'm cabable of measuring 1mHz with some amount of success, I realize I will have to keep repeating this to see what happens in the long term. I've never tried to measure these small increments before and was wondering if others have had success doing this or am I chasing a rabbit down a hole? I did try using the WSPR software to measure freqs and that seemed to be all over the place, plus there's no way to record it over time except manually. I tried recording the WWV 500/600 Hz AM audio freqs with Speclab and it's so noisy it's hard to get a good plot, but from looking at the plots I did make it seems to be in the range of 20mHz variation, but pretty much centered on the freqs. I also measured some AM stations and I had drifts of 56 mHz for a 1030 KHz station, and I have measured the DSP drift as around 14 mHz. Do the AM stations really drift around that much? I thought they were pretty stable. When I measure the service monitor locked to the Rb, I get much less drifts than actual stations, like I said, around 15 mHz and that seems to relate to the fan in the TS2000 going on and off. Dave N3DT ___ 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.
Re: [time-nuts] Is this measurement for real?
I missed the start of this thread but it you need it many computer audio interfaces can be run off an external clock. It is really common to do this in a studio setup. You think "does it matter? Can humans hear a 10ppm difference in pitch?" No but if you make two recordings each using a different clock then the error is cumulative and you might get a phase error near the end. So most audio interface have a clock input. Even my cheap $200 Presonus unit will take a clock input via it's fiber optic s/pdiff input. I still doubt it matters much for normal recording studio work. On Sun, Dec 8, 2013 at 12:12 PM, Hal Murray wrote: > > Well, the question is do you think that is good or bad for the sound > card ? > > Low cost crystals make reasonable thermometers. Ballpark is 1 PPM/C. > > If your system has a good ntp setup, you can use it to measure the actual > sound card clock. With luck (or skill) you can track that over time and > compare it with temperature. It will probably help to get the temperature > probe right on the crystal rather than someplace nearby. > > > > > -- > These are my opinions. I hate spam. > > > > ___ > 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. > -- Chris Albertson Redondo Beach, California ___ 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] Is this measurement for real?
Dave wrote: I also measured some AM stations and I had drifts of 56 mHz for a 1030 KHz station, and I have measured the DSP drift as around 14 mHz. Do the AM stations really drift around that much? I thought they were pretty stable. The FCC requirement for AM stations is +/- 20 Hz. It is not uncommon to find stations 10 Hz or more off frequency. I just checked one of the major stations here, and it is more than 16 Hz high with hundreds of mHz variation over 15 minutes. Some stations are locked to GPS these days, but far fewer than I expect. And some that are "locked" to GPS are rather loosely coupled, so they have offsets of as much as a Hz or even more (synthesizer steps, I presume) and/or drift of +/- tens to even hundreds of mHz. Most AM stations still use crystals in bang-bang ovens. It is often very easy to watch the oven heater cycle if you are using something with enough resolution (I generally use a spectrum analyzer with resolution in the few mHz range for such work). This afternoon, I can see 5 carriers on 1030 +/- 20 Hz. The station closest to you (WWGB) has been about 1 Hz low over the last few hours, with about 300 mHz variation (peak to peak). The variation shows the familiar asymmetrical ramp of an oven cycling with a period of ~29 minutes. It takes ~6.8 minutes to fall from its positive peak to its negative peak (heater on), and 21.9 minutes to climb back to its positive peak (heater off). Best regards, Charles ___ 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] Simulation of oscillator noise
On 12/08/2013 11:46 AM, Attila Kinali wrote: > Hi Magnus, > > On Fri, 29 Nov 2013 17:42:25 +0100 > Magnus Danielson wrote: > When I look in the data sheets of oscillator that I find on the internet, they only have precision estimates like 1ppm or 1ppb, but no detailed allan variance graphs. >>> Yes. Because in the class of cheap AT cut oscillators, you dont worry about >>> allan variance. The instability due to temperature dependence of your >>> system is much higher than the temperature-free (in)stability. The ADEV >>> becomes "relevant" only after you do at least a temperature compensation >>> or temperature control. >> The specification for temperature variations is a poor excuse too. Some >> vendors have learned that the hard way. > Could you explain a little bit what you mean here? I don't think > i get exactly what you are hinting at. Well. While staying within +/- 10 ppm over the temperature range may be one way of specifying the temperature dependence, it does not give you a good sense how it behaves at some particular temperature. Also, these are long-term dependence, but what happens when there is a quick change in temperature, what happens to frequency... and phase. Then we have drift properties. Let's put it another way. There is a reason cheap oscillators are cheap. 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] Is this measurement for real?
> Well, the question is do you think that is good or bad for the sound card ? Low cost crystals make reasonable thermometers. Ballpark is 1 PPM/C. If your system has a good ntp setup, you can use it to measure the actual sound card clock. With luck (or skill) you can track that over time and compare it with temperature. It will probably help to get the temperature probe right on the crystal rather than someplace nearby. -- These are my opinions. I hate spam. ___ 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] Efratom Rubidium SPTB-100/LN-001 info wanted
Is the Tektronix p/n on the unit ? it may be 119--xx On Thu, Dec 5, 2013 at 7:34 PM, Rex wrote: > I have an SPTB-100. It seems to be a custom version of a standard FRK(LN) > rubidium for Tektronics. > > I got mine around 2000. (Mine is an 88 date-code vs. your ancient 86 > version.) It didn't work initially because the internal Xtal oscillator > wouldn't sweep through lock frequency. Changed a cap in the oscillator and > it worked again. I've only powered it up a few times; I assume mine is > still working. I was told it is low-noise for a rubidium. > > Somewhere locally, I got a copy of a manual for it. Other than the cover > page, it seems to be a standard FRK manual, which the KO4BB pages already > have. Today, I made a short PDF with a scan of the cover page and a few > notes I wrote up. > > You can get it here: > http://www.xertech.net/pub/SPTB-100%20notes.pdf > > I also uploaded it to KO4BB for future references. > > If anyone knows more or sees any problems with what I wrote up, please let > me know and I'll correct or update. > > -Rex KK6MK > > > > On 12/4/2013 7:08 PM, Jerome Peters wrote: > >> Hello, >> >> Can anybody shed some light on this particular model: >> Model - SPTB-100/LN-001 >> Efratom part No. 703-200-11 >> Date Code 8609 >> >> (I have looked at K04BB's website, it has lots of other models, but not >> this one). >> >> Thank You, >> Jerome >> AF6UX >> >> >> --- >> This email message is for the sole use of the intended recipient(s) and >> may contain >> confidential information. Any unauthorized review, use, disclosure or >> distribution >> is prohibited. If you are not the intended recipient, please contact the >> sender by >> reply email and destroy all copies of the original message. >> >> --- >> ___ >> 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. > ___ 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] Simulation of oscillator noise
Hi Magnus, On Fri, 29 Nov 2013 17:42:25 +0100 Magnus Danielson wrote: > >> When I look in the data sheets of oscillator that I find on the internet, > >> they only have precision estimates like 1ppm or 1ppb, but no detailed allan > >> variance graphs. > > Yes. Because in the class of cheap AT cut oscillators, you dont worry about > > allan variance. The instability due to temperature dependence of your > > system is much higher than the temperature-free (in)stability. The ADEV > > becomes "relevant" only after you do at least a temperature compensation > > or temperature control. > The specification for temperature variations is a poor excuse too. Some > vendors have learned that the hard way. Could you explain a little bit what you mean here? I don't think i get exactly what you are hinting at. Attila Kinali -- 1.) Write everything down. 2.) Reduce to the essential. 3.) Stop and question. -- The Habits of Highly Boring People, Chris Sauve ___ 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] Simulation of oscillator noise
Servus Wolfgang, On Sat, 30 Nov 2013 10:32:42 +0100 Wolfgang Wallner wrote: > At my institute (TU Vienna, Computer Engineering) there has been a > bachelor thesis which dealt with simulation of IEEE 1588 in OMNeT++ (a > discrete event simulator). > But the assumptions where rather simple (both of the clock model and the > implemented version of IEEE 1588). > > For my master thesis I would like to enhance both aspects. > I would like to do a full implementation of IEEE 1588 and to use a more > realistic clock model. I think you could easily do a PhD on this topic alone. Thus I would highly recommend to focus on one aspect only for your master thesis. For simplicity, i'd first use some numbers on a good OCXO. These are much better specified and measured than the cheaper ones. E.g. you can use the Oscilloquartz 8607 as reference. If what data is freely available online is not enough for you, try contacting the manufacturer. They always have better data available, but do not publish it (don't ask me why). But still, they are usually quite generous with handing this data out for specific projects. Using a good oscillator will also give you a chance to verify your model. It should still be close to what the simulation with an ideal oscillator. Check for any deviation and try to explain it from the model. If you cannot explain it, it might be a simulation artefact. >From there, you can then start to degrade the oscillator model until it matches those of the oscillators you actually want to study. Attila Kinali -- 1.) Write everything down. 2.) Reduce to the essential. 3.) Stop and question. -- The Habits of Highly Boring People, Chris Sauve ___ 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] simulation of interconnected clocks
On Sat, 30 Nov 2013 06:31:01 -0800 Jim Lux wrote: > Recently, I've been looking at the variations of some human clocks which > are millenia old: Galileo used his pulse as a timer for his famous "roll > balls down a ramp" experimenet". I thought that some time-nuts might be > interested in working with a clock that's a bit different than one > depending on atomic vibrations, or motion within a crystal lattice. I don't know whether this is of any help to you, but some time ago i stumbled about some old lectures by Charles Peskin on the heart and to its chaotic self-synchronization [1]. If you are interested in the synchronisation phenomena in biological oscillators, i can recommend you [2]. Also a good read is [3] which gives a quite lengthy analysis on Kuramotos model [4]. Also a nice review paper is [5], which starts from Kuramoto and explains the current unsolved problems with coupled oscillators and their mathematical description. Attila Kinali [1] "Mathematical aspects of heart physiology", by Peskin, 1975 http://math.nyu.edu/faculty/peskin/heartnotes/index.html [2] "Synchronization of Pulse-Coupled Biological Oscillators" by Mirollo and Strogatz, 1990 http://math.bd.psu.edu/faculty/stevens/MATH497K/Papers/Syncrhonization.pdf [3] "The Kuramoto model: A simple paradigm for synchronization phenomena", by Acbron, Bonilla, Vincente, Ritort, Spigler, 2005 http://rmp.aps.org/abstract/RMP/v77/i1/p137_1 [4] "Self-entrainment of a population of coupled non-linear oscillators" by Kuramoto, 1975 http://www.springerlink.com/content/71073361941277h8/ [5] "From Kuramoto to Crawford: exploring the onset of synchronization in populations of coupled oscillators", by Strogatz, 2000 http://www.sciencedirect.com/science/article/pii/S016727890944 -- 1.) Write everything down. 2.) Reduce to the essential. 3.) Stop and question. -- The Habits of Highly Boring People, Chris Sauve ___ 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] Is this measurement for real?
Well, Bill, I just don't know. I think that's why I asked. However, I did measure it again over 2.5 hrs and I got a roughly 500uHz drift pretty much all in one direction. I'm wondering if this is telling me I'm cabable of measuring 1mHz with some amount of success, I realize I will have to keep repeating this to see what happens in the long term. I've never tried to measure these small increments before and was wondering if others have had success doing this or am I chasing a rabbit down a hole? I did try using the WSPR software to measure freqs and that seemed to be all over the place, plus there's no way to record it over time except manually. I tried recording the WWV 500/600 Hz AM audio freqs with Speclab and it's so noisy it's hard to get a good plot, but from looking at the plots I did make it seems to be in the range of 20mHz variation, but pretty much centered on the freqs. I also measured some AM stations and I had drifts of 56 mHz for a 1030 KHz station, and I have measured the DSP drift as around 14 mHz. Do the AM stations really drift around that much? I thought they were pretty stable. When I measure the service monitor locked to the Rb, I get much less drifts than actual stations, like I said, around 15 mHz and that seems to relate to the fan in the TS2000 going on and off. Dave N3DT ___ 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] Is this measurement for real?
Dave: What is it that you "almost don't believe" ? That it is this good? Or that it is this bad? Remember that the sound card has a sampling clock, that also contributes to any errors. If it is a regular computer sound card, then it is using the same $0.33 crystal reference as the rest of the computer. --- Graham / KE9H == On 12/7/2013 2:24 PM, quartz55 wrote: I've been testing my soundcard measurement capability with Speclab, so I hooked my LPRO Rb to lock my Moto Service Monitor, fed the 1K audio out into the sound card and I measured the freq over about a half hour. I got a spread of about 107uHz. I've tried a few other things through my TS2000 using a 1500Hz tone from the USB and I can see when the fan goes on and off although I have the RX main osc locked to the GPSDO with the XRef from VK3HZ. I can only assume the drifting in the RX is due to the DSP processing crystal drifting with temp. I get about 13mh drift on that. Anyhow, here's a shot of the 1000Hz directly into the soundcard with Speclab. http://i251.photobucket.com/albums/gg287/DogTi/time/1k30m_zpsc4173981.jpg I almost don't believe it, does anyone else? Dave N3DT ___ 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.
Re: [time-nuts] Is this measurement for real?
Hi Dave, Well, the question is do you think that is good or bad for the sound card ? BillWB6BNQ quartz55 wrote: I've been testing my soundcard measurement capability with Speclab, so I hooked my LPRO Rb to lock my Moto Service Monitor, fed the 1K audio out into the sound card and I measured the freq over about a half hour. I got a spread of about 107uHz. I've tried a few other things through my TS2000 using a 1500Hz tone from the USB and I can see when the fan goes on and off although I have the RX main osc locked to the GPSDO with the XRef from VK3HZ. I can only assume the drifting in the RX is due to the DSP processing crystal drifting with temp. I get about 13mh drift on that. Anyhow, here's a shot of the 1000Hz directly into the soundcard with Speclab. http://i251.photobucket.com/albums/gg287/DogTi/time/1k30m_zpsc4173981.jpg I almost don't believe it, does anyone else? Dave N3DT ___ 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.
Re: [time-nuts] Is this measurement for real?
Sounds about right. Sent from mobile > On Dec 7, 2013, at 2:24 PM, "quartz55" wrote: > > I've been testing my soundcard measurement capability with Speclab, so I > hooked my LPRO Rb to lock my Moto Service Monitor, fed the 1K audio out into > the sound card and I measured the freq over about a half hour. I got a > spread of about 107uHz. I've tried a few other things through my TS2000 > using a 1500Hz tone from the USB and I can see when the fan goes on and off > although I have the RX main osc locked to the GPSDO with the XRef from VK3HZ. > I can only assume the drifting in the RX is due to the DSP processing > crystal drifting with temp. I get about 13mh drift on that. Anyhow, here's > a shot of the 1000Hz directly into the soundcard with Speclab. > http://i251.photobucket.com/albums/gg287/DogTi/time/1k30m_zpsc4173981.jpg I > almost don't believe it, does anyone else? > > Dave > N3DT > ___ > 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.
Re: [time-nuts] OT : different Rx and Tx baud rate on same port
You could also consider making a simple gadget with your microcontroller of choice, that presents an interface to the PC of a fixed baud rate (perhaps 115200) and handles the weirdo split rates for the GPS gadget on one or two of its other ports. Henry On Sat, Dec 7, 2013 at 12:18 PM, Hal Murray wrote: > > b...@lysator.liu.se said: >> Looking at the "stty" unix command. It seems clear that split baud rates has >> been supported at one time. > > On Linux, man termios gives lots of API details. > > On NetBSD and FreeBSD, that gets an overview. man tcsetattr gets the API > details. > > If you use stty to change things, the new info is sticky. So if you have a > program that is all set to go except that the baud rates aren't right and > that program doesn't smash the baud rates, you can set them with stty and > your program should work. I haven't tried it with split baud rates, but the > normal (non-split) case works. > > If your gizmo uses simple ASCII, you can test things with cat /dev/wherever > and things like this on another terminal > echo "blah blah..." > /dev/wherever > > > -- > These are my opinions. I hate spam. > > > > ___ > 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.
Re: [time-nuts] OT : different Rx and Tx baud rate on same port
An alternative to writing your own software is using 'socat' which can interface between a tcp port and serial port (perhaps using a raspberry pi or the like as the interface). It apparently has settings for separate ispeed and ospeed. Its competitor, ser2net, doesn't seem to have that ability. This way you can create a virtual serial port with the baud rate details hidden. As for whether all this would work on current hardware -- I haven't tested because I haven't had reason to try, nor the hardware to test it on. I can tell you, as I noted above, that changing baud rates on the fly works, even with telnet connections using RFC2217 commands, since I use that with 1-wire work. The only problems were some of the stranger settings, like 6-bit words. Paul Alfille On Sat, Dec 7, 2013 at 3:18 PM, Hal Murray wrote: > > b...@lysator.liu.se said: > > Looking at the "stty" unix command. It seems clear that split baud rates > has > > been supported at one time. > > On Linux, man termios gives lots of API details. > > On NetBSD and FreeBSD, that gets an overview. man tcsetattr gets the API > details. > > If you use stty to change things, the new info is sticky. So if you have a > program that is all set to go except that the baud rates aren't right and > that program doesn't smash the baud rates, you can set them with stty and > your program should work. I haven't tried it with split baud rates, but > the > normal (non-split) case works. > > If your gizmo uses simple ASCII, you can test things with cat /dev/wherever > and things like this on another terminal > echo "blah blah..." > /dev/wherever > > > -- > These are my opinions. I hate spam. > > > > ___ > 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.
Re: [time-nuts] temperature
Sorry if this is somewhat off topic, but I'd be interested in more details re precision temperature measurement devices. Have been using an inexpensive USB temperature sensor for the last year or so to monitor the temperature in my lab and have been looking at the correlation between frequency shifts in some ocxo's vs temperature changes. I should also start taking humidity measurements as well at some point. Any pointers re suitable instruments to accomplish this that can be sourced via the usual surplus sources would be welcome. Thanks in advance Mark Spencer Sent from my iPad ___ 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] HP5370B for sale
Hi Guys, I have a fully functional HP5370B for sale in the UK - Price is UKĀ£550 International shipping will be at cost, shipping within the UK is included Anyone interested can contact me theough the list Cheers, Pete ___ 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] Efratom Rubidium SPTB-100/LN-001 info wanted
I have three of these units, and have been looking for the right manual for years. Thanks for this part, but I believe the SPTB is different from the FRK in the schematics and some of the construction detail - for example, since it's 5 MHz vs 10, that circuit is different, not just the crystal frequency. I have found other things in the PLL too, that probably relate to the "LN" performance. I have managed to use the *FRK manual, which is close enough to get by - the circuit operation is the same, just the design details differ. I also had to replace a capacitor in one unit that would not lock - it was a leaky ceramic one in the PLL. I had to do a bit of guesswork and shotgunning, since the circuit topology and parts layout were quite a bit different in that area. *I may be confusing it with another model - the M-100, I think, that is similar. This was a few years ago. Ed ___ 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.