EEVblog Electronics Community Forum
Electronics => Metrology => Topic started by: tkamiya on September 30, 2019, 07:33:12 pm
-
Most GPSDO has Crystal based oscillators: plain Xtal, temperature compensated, ovened, double ovened, etc. Some comes with Rubidium oscillator.
EXCEPT FOR HOLDOVER case, are there any advantage for having Rubidium? In case of GPSDO+Rb, loop constant is set so long, it won't be able to deal with short fluctuation. Plus, all Rb modules I know has XO in it. I've never seen just plain Rb generating the output.
Can anyone help me out?
-
"Plain Rb" does not generate anything, it responds to external signal sweeping over. This signal can be generated by XO, OCXO, VCO or anything else that oscillates and can be frequency shifted. It's, basically, an ocsillator slaved or disciplined to Rb cell response.
Your average telecom grade Rb source has poor short-term stability and awful phase noise because they use cheap convenient oscillator running at seemingly random (chosen out of convenience) frequency and then synthesize required signal digitally, say, using DDS.
If you take good quality Rb from HP, Efratom or Stanford Research - they use high quality OCXO already operating at required output frequency - say 10MHz. This is why they produce clean pure sinewave with low phase noise and good short term frequency stability (same thing, different chart.)
Leo
Plus, all Rb modules I know has XO in it. I've never seen just plain Rb generating the output.
-
Leo, nice to meet you, and thank you for your response! (you must be the famous pulse generator person)
I have Rb from Standard Research, PRS10. I also have Efratom FRS-C. As far as I know, they came out of telecom installations. I know PRS10 has SC cut crystal and FRS-C has AT cut. I don't recall if they are single or double oven'ed. Xtal frequency is 10MHz.
These seem to be very common. These are also ex-telecom.
What do you consider these? Cheap telecom stuff or good quality and proper stuff? Do you know if manufacturers select units post production and send lower grade stuff for telecom use?
When you say cheap telecom, are you referring to a type with user-configurable output frequency?
-
Hi Taka,
You are well equipped then!
I am not 100% sure about FRS, but FRK has 10MHz OCXO that is slaved by the servo http://bama.edebris.com/download/efratom/frk/Efratom_FRK.pdf (http://bama.edebris.com/download/efratom/frk/Efratom_FRK.pdf)
By "cheap telecom units" I mean FE-5650A and FE-5680A that have flooded the market. They are really a good budget design but not the best for metrology use.
If you are interested in Rb operation fundamentals, have a look at Bill Wriley's excellent primer. It's worth a read:
http://www.wriley.com/Rubidium%20Frequency%20Standard%20Primer%20102211.pdf (http://www.wriley.com/Rubidium%20Frequency%20Standard%20Primer%20102211.pdf)
Cheers
Leo
-
It all depends what you're looking for in a GPSDO. There have been a number of Rb units, including the Nortel unit you described previously. Austron OT-20, Odetics 365/565, FEI-Zyfer 385, and others have units that are Rb based or have Rb as an option. Note that the photo of the FEI-Zyfer unit shows it does use an FE-5650A Rb oscillator. The OT-20 works with either a cell receiver card or a Motorola GPS receiver.
-
GPSDO is for long term correct the center frequency error, for the jitter, RB oscillator (by zeeman effect control) is very weak; till the phase noise, I checked several test scheme, I believe many of them do not know what is frequency reference, use a TCXO spectrum analyzer (hign dynamic) to test a Rubidium or Cesium oscillator's 1Hz offset peak value, interesting.
-
I went ahead and built such a unit. Actually, I've been working on it for the past 6 month or more.
GPS receiver feeds 1 pps into PRS-10. User output comes from PRS-10 through buffer/distribution amplifiers. Today, I am doing preliminary tests using TICC + LH setup. Of course, effects of gps control won't be apparent for months but as of now, FTS4040/A (Cs) and my unit are neck to neck. Adev at 100 sec are 1.x E-11. At 500 sec, both 5 or 6 10E-12. I'm rather impressed....
I believe the default loop constant on GPS steering is something like 24 hours. So it's a very gentle pull. It has been turned on for about 2 weeks. My Cs is, unfortunately, quite old. Other than it locks effortlessly, I don't know it's accuracy or stability. I'm going to have to trust it's at least stable, since it can maintain lock for as long as I keep it on.
-
Yesterday I have found and opened my old Spectracom NetClock/GPS 9183 with Rubidium option and found they are doing exactly that.
It has PRS10 inside which is disciplined from Oncore GPS receiver.
Cheers
Leo
-
I'm working on something similar, a plug-on PCB that you can mount onto a LPRO-101. It'll have a pin socket for a GPS board which can feed a 1PPS signal, but if there's space enough I'll provide an SMA jack to feed an external 1PPS signal from another timing receiver.
If you want to follow along, I've started a thread here:
https://www.eevblog.com/forum/projects/diy-gpsdo-project-w-stm32-tdc7200/msg2801922/#msg2801922 (https://www.eevblog.com/forum/projects/diy-gpsdo-project-w-stm32-tdc7200/msg2801922/#msg2801922)
-
Good luck with your project!
My project is complete, hardware wise and running. Trying to test it is another story altogether. My Cesium standard is old and that stability and accuracy of it is in question. To make it worse, room temperature fluctuate quite a bit. I am still contemplating, how am I going to validate this thing I created.
-
Good luck with your project!
My project is complete, hardware wise and running. Trying to test it is another story altogether. My Cesium standard is old and that stability and accuracy of it is in question. To make it worse, room temperature fluctuate quite a bit. I am still contemplating, how am I going to validate this thing I created.
use a young Rubidium oscillator, the tolerance will be lower, then you can compare with your old cesium clock, as usual, 4040a also exists the aging rate, if no calibration, there is no meaning for the cesium clock if more than 20 years. I will release several Spectratime clock within one week.
-
My project is complete, hardware wise and running. Trying to test it is another story altogether. My Cesium standard is old and that stability and accuracy of it is in question. To make it worse, room temperature fluctuate quite a bit. I am still contemplating, how am I going to validate this thing I created.
One option is to use a good rubidium like an Efratom LPRO and reduce the effects of the environment on it.
The temperature could be controlled or compensated for, and air pressure changes can also be compensated for. I've had more success with temperature control than compensation, and although the air pressure compensation worked quite well with the Datum and Efratom LPROs, it did not seem to work so well with the Temex LPFRS and FEI 5680A.
Some Rb oscillators also have 'features' like digital temperature compensation, which can cause steps in the frequency output.
Another option would be to use a suitable dual frequency GPS and post process the data from it.
-
Whatever it may be, it will have to be well characterized. So far, all I have is datasheet written 20+ years ago and equipment just as old. I'm sure it is no longer meeting the same spec it left the factory. How far it has drifted is quite unknown.
-
Whatever you do, you'll likely have to characterize it yourself - or spend a lot of money.
If you log a reference rubidium against your DUT and GPS, and also log air pressure and the temperature (sometimes even the temperature at each piece of equipment involved) for a few weeks you should start to get a picture of how they behave.
In general, the more references you test, the more you learn, and the more references you can test against each other at the one time the better.
A good quartz oscillator can be useful for short term measurements like ADEV and phase noise, but you also need gear capable of measuring that too.
-
That's the problem. I don't have a good "reference". Each device at this point is DUT. How do I solve this dilemma? I have more than 10 Rb. Slightly less GPSDO with OCXO, one GPSDO + Rb, and one OLD Cesium.
In fact, first few try, two DUT with one "reference" (that was Cs), I got two identical trace. That's impossible. Even the initial lockup Servo movement (up down up down) is identical and in sync. Obviously, my issue there is measuring method or Cs.
If I can have one trustable reference, my job will be a lot easier.
I guess they don't call this Time NUTS for nothing.
-
That's the problem. I don't have a good "reference". Each device at this point is DUT. How do I solve this dilemma? I have more than 10 Rb. Slightly less GPSDO with OCXO, one GPSDO + Rb, and one OLD Cesium.
All GPSDO/OCXO have a similar ADev curve. Here's an example:
http://www.leapsecond.com/pages/gpsdo/free-lock-gpsdo4.gif (http://www.leapsecond.com/pages/gpsdo/free-lock-gpsdo4.gif)
The graphs with the dots show four different GPSDOs. Below about 1000 sec. you're seeing the OCXO. Then they all merge with GPS and look more or less identical. The hump in the 10 - 1000 sec. range can often be tuned out by optimizing the GPSDO time constant and damping factor.
I've never seen a graph of a GPSDO/Rb unit, but I expect it to look similar. Eventually, the performance will be determined by GPS.
In fact, first few try, two DUT with one "reference" (that was Cs), I got two identical trace. That's impossible. Even the initial lockup Servo movement (up down up down) is identical and in sync. Obviously, my issue there is measuring method or Cs.
If I can have one trustable reference, my job will be a lot easier.
I guess they don't call this Time NUTS for nothing.
At every value of Tau, an ADEV curve shows you the combined results of the ADEV values of three items: your measurement system, your reference, and your DUT. If one of those is significantly worse than the others, it will dominate the results. It's also typical that you need to use different references for different values of Tau. A good OCXO is the best reference for low values of Tau. Medium values of Tau match up with a REALLY good OCXO or a Rb. At high values of Tau, you might say that Cs is necessary, but ADEV isn't really the right measurement for high values of Tau. Just use aging instead.
Chasing better references and better measurement systems is a never-ending activity.
If you really need to determine the ADEV of your various devices, there are two ways to do it. Send one out to be measured by someone who has the knowledge and equipment to do it. I don't think a cal lab would do it. The second way is to gather enough measurement equipment and do a Three-Corner Hat measurement.
-
That's the problem. I don't have a good "reference". Each device at this point is DUT. How do I solve this dilemma? I have more than 10 Rb. Slightly less GPSDO with OCXO, one GPSDO + Rb, and one OLD Cesium.
In fact, first few try, two DUT with one "reference" (that was Cs), I got two identical trace. That's impossible. Even the initial lockup Servo movement (up down up down) is identical and in sync. Obviously, my issue there is measuring method or Cs.
In that case it might be worth just measuring some of the rubidium oscillators against each other for starters. With the PRS10 you can adjust the frequency by precise amounts and make sure that everything measures as it should. The 1PPS can be compared with GPS too.
If the Cesium works at all you can replace on of the rubidiums with it and see how it goes, knowing that your measurement setup has been tested.
The Reference/DUT distinction is a bit irrelevant without a true master reference. That was my point about logging several oscillators and GPS at the same time - they all act as references for each other. It's not perfect, but it's a start. They may not behave perfectly, but other than for environmental changes, they are unlikely to misbehave in the same way at the same time.
If I can have one trustable reference, my job will be a lot easier.
I guess they don't call this Time NUTS for nothing.
That was why I started looking at dual frequency GPS - I think it's probably the way to go now.
-
Lets say I compare two PRS10. Connect one as timing source and measure another. Then, the data I get belongs to which one? We are back to square one.
Maybe I'll do time-interval measurement? (I have TICC)
-
Lets say I compare two PRS10. Connect one as timing source and measure another. Then, the data I get belongs to which one? We are back to square one.
Maybe I'll do time-interval measurement? (I have TICC)
The noise is the sum (or rather the square of the noise is the sum of the squares) of both if they are uncorrelated. This is where the three cornered hat approach comes in - measure 3 devices then you have 3 equations (the measurement of each against each other) and 3 unknowns (the noise of each) which is enough to solve for each separately.
The measurements should be taken at the same time to minimize environment variations (if they are all dependent on temperature then hopefully they all vary in roughly the same way and it cancels out).
Such an approach isn't going to be as accurate as a good reference but it is the only approach really available to hobbyists who don't have half a million $ or £ available to buy a hydrogen maser! :)
EDIT: half a million is probably a bit of an exaggeration for the cost of an hydrogen maser but they are not cheap. An alternative such as a super low noise OCXO such as the Oscilloquartz BVA which would be a good reference for times up to say 100 seconds are about £30k new and even on ebay are over £1k. (The BVA is no longer manufactured anyway I think.)
https://www.eevblog.com/forum/metrology/oscilloquartz-8600-3-gpsdo-internal-photos/ (https://www.eevblog.com/forum/metrology/oscilloquartz-8600-3-gpsdo-internal-photos/)
-
I think Hydrogen Maser starts around 250K USD. Plus a temperature stable environment, running cost, etc, etc, etc.... It is certainly not in my wallet... I know about the OQ BVA but going eBay route will put me back to the very problem I want to solve. Yet, another unknown.
I have 3 T-bolt and more than 3 PRS10. I also have 4 HP5335A which 3 of them are from same generation. 3 cornered hat is certainly possible. I can at least pick "the best one" of each type. I can run them at the same time but I have to think about recording the running results.
I thought about putting OCXO in an thermos. I was told it was a bad idea, but BVA is a dewar design. I could possibly buy a really big dewar to give more thermal mass and try that. (yup, another unknown!)
I would love to find a traceable metrology lab that can get involved. Shipping being a big problem, I don't think that's realistic either. I'll research the "hat" method.
-
To do a Three-Cornered-Hat measurement, you must do the measurements simultaneously. If you don't, the data isn't correlated and the calculations usually fail with zeros or infinities popping up.
Timelab has the calculations built-in so the processing becomes trivial.
It's okay to have an oscillator in a Dewar if it's designed to be in a Dewar. If you put any old oscillator in a Dewar or just add insulation of any type to it, it can overheat.
-
The alternative to the 3 cornered hat approach is to have two identical oscillators as you can then make the approximation that the noise is divided between them so that the ADEV is divided by sqrt(2). Of course this is only an approximation.
-
Have you talked to the Cal labs in your area?
Britest or Custom Cal seem to have fairly decent pricing for locals.
As you say you only need one unit calibrated.
-
Tkamiya,
I try to understand what problem you are trying to solve.
For me, the following holds:
The Planck time (tP) is the unit of time in the system of natural units known as Planck units. A Planck time unit is the time required for light to travel a distance of 1 Planck length in a vacuum, which is a time interval of approximately 5.39 × 10−44 s.
Since NIST is improving one decade per 10 years, we have quite some distance to cover.
Classical estimates put it to approx 300 years.
Or there is something we do not fully understand yet. And remember: there are no absolute measurements.
So, in the mean time we have to rely on NIST. Why not average all your equipment readings?
And monitor it's statistics, mean, average, standard deviation, highest and lowest values?
Hydrogen maser will not bring you nearer to the truth.
Frans
-
I went ahead and built such a unit. Actually, I've been working on it for the past 6 month or more.
GPS receiver feeds 1 pps into PRS-10. User output comes from PRS-10 through buffer/distribution amplifiers. Today, I am doing preliminary tests using TICC + LH setup. Of course, effects of gps control won't be apparent for months but as of now, FTS4040/A (Cs) and my unit are neck to neck. Adev at 100 sec are 1.x E-11. At 500 sec, both 5 or 6 10E-12. I'm rather impressed....
I believe the default loop constant on GPS steering is something like 24 hours. So it's a very gentle pull. It has been turned on for about 2 weeks. My Cs is, unfortunately, quite old. Other than it locks effortlessly, I don't know it's accuracy or stability. I'm going to have to trust it's at least stable, since it can maintain lock for as long as I keep it on.
That's probably the way to go - just get familiar with logging to TimeLab and LH, and using the results.
Since the TICC has 2 channels, you could use the 4040 as the reference and measure it against a rubidium and GPS to get some idea of its shorter and longer term stability as well as its accuracy. This also effectively logs the rubidium against GPS.
To reduce the GPS noise you could use a GPSDO instead, but it might be best not to use a long time constant in the GPSDO as that can lead its oscillator affecting the frequency.
Even when you do not have a very good reference like a hydrogen maser, you can get a good idea of performance by measuring against several different references like quartz, rubidium and GPS.
If you can then also log the temperatures of the oscillators that you are using, as well as the air pressure and humidity, you will then be able to work out their environmental sensitivity.
In the case of a Rubidium oscillator the ambient conditions often dominate the performance. Ageing may be about 1E-13/day, but air pressure sensitivity could be up to 1E-13/mbar, and the TC could be over 10E-13/°C.
-
I guess I’m not quite as concerned as some others about relying on the long term measurements I get by using the GPS satellite system. I haven’t had any problems with the level of accuracy I see, at least for quality of the standards I now own. One of the GPSDOs that I mainly rely on is a modified EBSCTM (NTPB15AA) with one of the typical Tbolt displays added and a SMPS built into the unit. Lady Heather treats it pretty much like a Tbolt and uses almost all of the same commands. I’ve had this running for a few years and have a lot of confidence in it. Here is a graph from it that covers almost a day.
I also use a couple of modified Thunderbolt with no internal OCXOs that I use to check the drift and stability of 10Mhz oscillators that I want to test. I can connect both the 10Mhz and the EFC from the Tbolt to the oscillator or to just check drift, only connect the 10Mhz, and let the oscillator free run. Below is a graph of one of my Rb standards I was adjusting using this method. The white trace basically displays the jitter and that trace stays very close to the reference line while the purple PPS trace displays the accumulated drift. Initially the frequency was low and after about 3 hours I made an adjustment and overcorrected a little and after about another 3 hours I decreased the frequency and my guess was pretty close.
Some of the GPSDOs I own don’t output all the information that the Tbolts do so I can run both the GPSDO under test and the modified Tbolt, feeding the 10Mhz from the GPSDO under test to the modified Tbolt and let the Tbolt graph the output to get almost all of the same information except for the DAC data which in this case would be meaningless.
-
I also use a couple of modified Thunderbolt with no internal OCXOs that I use to check the drift and stability of 10Mhz oscillators that I want to test. I can connect both the 10Mhz and the EFC from the Tbolt to the oscillator or to just check drift, only connect the 10Mhz, and let the oscillator free run.
I do sometimes regret not stocking up on a few thunderbolts - they're useful as test gear as well as GPSDOs.
These plots are from a test that I did with a temperature controlled and air pressure compensated Efratom LPRO against a sawtooth corrected M12+T GPS. The diurnal variations on the GPS show up quite well against the LPRO. Red/pink traces are the LPRO frequency.
1000s averages were used for these plots, so the stats plots are a bit different/lower than they would be with the 1s data.
-
I guess I’m not quite as concerned as some others
Let's NOT get too reasonable here. :-DD
Happy Holidays!
-
I have a quite a bit of PRS-10 and just recently, bought some T-bolt. One with older OCXO and 3 with Trimble OCXO. Looks like surplus is starting to dwindle, so got some for myself. My main standard is basically the same with Arthur's. GPSTM. (Nortel OEM t-bolt)
Regrettably, I made few mistakes along the way. I was so pleased with GPSTM, I bought 5 more at very very reasonable prices. Unfortunately, these are made by different OEM. It has some crazy jitters. So they are now sitting in storage bin.
I have Cs but I don't have very much trust in it. It's over 20 year old with an original tube. It locks but a reliable source told me, 4040A is particularly prone to temperature induced drift.
-
I have a quite a bit of PRS-10 and just recently, bought some T-bolt. One with older OCXO and 3 with Trimble OCXO. Looks like surplus is starting to dwindle, so got some for myself. My main standard is basically the same with Arthur's. GPSTM. (Nortel OEM t-bolt)
Regrettably, I made few mistakes along the way. I was so pleased with GPSTM, I bought 5 more at very very reasonable prices. Unfortunately, these are made by different OEM. It has some crazy jitters. So they are now sitting in storage bin.
I have Cs but I don't have very much trust in it. It's over 20 year old with an original tube. It locks but a reliable source told me, 4040A is particularly prone to temperature induced drift.
for Cesium, only 5071A
-
If I could afford a working 5071A, I'd have one. All I have is WORKING 4040A and DEAD 5071A.
I have been running a 14 day test of PRS-10 being steered by a GPSDO. It's only half way through but it is showing some promise. Sadly though, this won't be definitive as house standard is yet another GPSDO. Once this is done, I'll do the same test with Cs being the standard.
I may actually run a 30 day test next time. I still haven't run into a clear point where Adev starts to go up.
-
If I could afford a working 5071A, I'd have one. All I have is WORKING 4040A and DEAD 5071A.
Datum/FTS made drop-in replacements for many of the HP Cs tubes. I don't know how much interfacing would be required to move the tube from your 4040 to the 5071A or if it's even practical, but it might be worth exploring.
I have been running a 14 day test of PRS-10 being steered by a GPSDO. It's only half way through but it is showing some promise. Sadly though, this won't be definitive as house standard is yet another GPSDO.
What? A properly working GPSDO is the best standard available to mere mortals. The USAF continuously monitors the satellites and corrects their frequencies as necessary. The result is a system that's more stable than Cs. So, if you have multiple GPSDOs of different types that agree with each other, you can be sure that they are on frequency.
I still haven't run into a clear point where Adev starts to go up.
Of course not! A rising Adev suggests aging. Once you get past the OCXO range and into the GPS range of the graph, you should never see a rising Adev on a GPSDO. The only thing you'll see is a flat Adev which represents whatever white noise exists in the system. Typically, this will be in the e-13 range or lower.
You might see things like diurnal effects that look like aging if you don't run the graph long enough.
-
GPSDO is accurate to only about E-08. Only when one integrates over long time, it will reach somewhere between E-12 to E-13 depending on the quality of OCXO. So I have two GPSDO, one house standard and one DUT. Every second, comparison is made. That comparison is accurate only to E-08. (instantaneously) But that's not exactly what I'm doing.... That's just my house standard.
Take a look at this release note: https://www.thinksrs.com/downloads/pdfs/catalog/PRS10c.pdf (https://www.thinksrs.com/downloads/pdfs/catalog/PRS10c.pdf)
Page 2, top left. That's what I'm doing. PRS10 is supposed to drift/age minutely and adev will head up. It is supposed to, then, get hammered down by GPSDO. I haven't seen that point, yet. I am seeing gradual up trend but confidence range is too large to interpret it. It looks like settling down to 4xE-12.
Please note, SRS's note is Allan VARIANCE and the tool I'm using shows Allen Deviation. I know the difference is square root. I'm trying to wrap head around this now. But overall, I'm very pleased.
Yes, I looked at that replacement tube..... Symmetricom makes them now. I saw somewhere, the price is about USD40K for standard tube. I don't know about transplanting. I'd rather leave that tube as-is, where is.
-
"Root Allan Variance" is the same as Allan deviation, it just sounds a bit more posh.
Leo
Please note, SRS's note is Allan VARIANCE and the tool I'm using shows Allen Deviation. I know the difference is square root. I'm trying to wrap head around this now. But overall, I'm very pleased.
-
Woooo.... That sounds NICE!
So that IS the same thing then.
-
hi Tkamiya, this article of Ulrich Bangert
http://www.ulrich-bangert.de/AMSAT-Journal.pdf (http://www.ulrich-bangert.de/AMSAT-Journal.pdf)
provides an excellent introduction to the stability of oscillators and frequency standards.
But sorry it is available only in German but very welll worth any efforts to try a translation.
Dual mixer method for measuring stability is explained as well as understanding Allan deviation and the usefulness
of coupling OCXOs and GPS.
have fun Kutte
-
In a quick unscientific test, I measured the frequency of my Tbolt. The connection was: Tbolt -> 10 MHz low pass filter -> Fluke PM6681 counter. 1 sec. gate time, referenced to HP 5065A. Variations in frequency were around +- 1 millihertz, i.e. 1e-10. Changing the gate time to 5 sec. reduced the variations to around 0.2 millihertz or 2e-11.
As you stated, integrating the GPSDO signal over long periods improves the results. That's exactly what an Allan Deviation graph is showing. The only way you will see a rising graph is if the long-term average frequency is changing. That won't happen with GPS. So since your reference is also a GPSDO, the only thing left that could be changing is your measurement system.
The PRS10 document wanders back and forth between Variance and Deviation. Very confusing. There's no date on it. I wonder if it dates back to before Selective Availability was disabled? The price it quotes is the same as the price in June 2000. The same graph was also used at that time. Selective Availability was disabled in May 2000. Selective Availability would significantly increase the noise on of any GPS-derived measurement. Long duration measurements wouldn't be affected, but short-term measurements would suffer.
-
If you are diving much below 10-13, remember, that GPS time is steered towards UTC (modulo 1s) but can be as far away as 10ns at times. I don't have a clear idea of what ADEV plot of UTC-GPS looks like, only have standard deviation figures from https://www.gps.gov/governance/advisory/meetings/2017-11/powers.pdf (https://www.gps.gov/governance/advisory/meetings/2017-11/powers.pdf)
This might be only of academic importance but interesting anyhow.
Notice, that both GPS time and UTC derived from GPS transmission are 1ns RMS away from UTC(USNO.)
Leo
-
Hm. If you get 1e-8 only with a GPSDO, for small tau (< TC), that would hint that the oscillator of the GPSDO is rubbish. With a good OCXO you should get into the 1e-11 range.
-
I misspoke. 1 pps out of GPS is 10^-8 range at low tau. You are absolutely correct that good OCXO should give ^-11 ish.
-
There is something quite addictive about these stuff.... oh oh.... am I becoming a time-nuts?
-
Edpalmer42,
Actually, I'm measuring an Adev of PRS10 output steered by GPSTM (Nortel version of T-bolt) using a reference of another GPSTM. What I can summarize from that confusing SRS release note is that once aging takes over and the graph starts to head UP, GPS should bring it down. So aging will be non-issue. So, this is NOT quite GPS vs GPS.
I found something interesting. My graph indeed turns UP at one point but I noticed it NEVER touches 1E-11. Then I turned on the error bar. Aha! Margin of error (not enough samples) is affecting the graph. I have this running for 14 days and this is the 8th day. It has been like this for 3 days or so. (almost touches 1E-11) So if I were to run this forever, I expect it will settle somewhere around 4E-12.
Sorry about bad quality photographs. The measurement machine is in different subnet.
-
I misspoke. 1 pps out of GPS is 10^-8 range at low tau. You are absolutely correct that good OCXO should give ^-11 ish.
Okay, now I see what you're saying. Yes, if you measure the Allan Deviation of a raw GPS receiver, rather than a GPSDO, you can expect an ADEV value of around 1e-8 @ Tau = 1sec. Newer designs do somewhat better. From there, it improves at a slope of -1, i.e. one decade improvement of ADEV for every decade of time until it reaches the white noise limit. Beyond that, the value is a constant.
But remember that your reference GPSDO has already integrated all that out. It has climbed down the ADEV curve and is giving you an OCXO-smoothed signal that is on frequency. The short-term noise is due to whatever oscillator is in the GPSDO. In my case, the 1e-10 value is the noise of the OCXO in the Tbolt. So, if your primary concern is frequency, you're there. You have to set the gate time to average out the short-term noise to a level that's appropriate for your needs.
Since you're using a GPSTM to feed your PRS10, you're using the PRS10 as a clean-up oscillator to try to improve the output of the GPSTM and keep the PRS10 on frequency. I don't know if there's enough info in the PRS10 manual to determine if this will give you a cleaner signal, but it should keep it on frequency. The shape of the ADEV curve will depend on the PRS10's internal disciplining algorithm.
By the way, you're not 'becoming' a time-nut. You passed that threshold a long time ago! ;)
-
My experiment has nothing to do with Planck Time, or anything of (such a tiny) magnitude. According to SRS PRS10 manual, PRS10 has an AllenDev that bottoms then goes up due to drift. The same manual says adding PPS from GPS steers Rb and eliminate the drift part of Adev.
I'm trying to do just that.
I did have a go at it, and results are rather disappointing. With PPS, the Allen dev of PRS-10 is 10 times worse at all interval! Oddly enough, partial graph collates fairly well. I did not tune anything and left everything as default. This is the first try. By default, PRS-10 comes standard with loop gain and loop constant suitable for GPS, so I went with it.
Now that I have this result, I am at a phase to think through what might have gone wrong. Ground loop and noise on PPS line comes first to my mind. I was rather sloppy in this area. I made the voltage where it's needed willy-nilly. I should be more careful. I didn't appreciate the level of scale I was dealing with. Also, GPS I used was of unknown nature. I now have few T-bolt, so I'll use that next.
I am not going to upload the graph.... It's so SO disappointing....... :palm:
-
My experiment has nothing to do with Planck Time, or anything of (such a tiny) magnitude. According to SRS PRS10 manual, PRS10 has an AllenDev that bottoms then goes up due to drift. The same manual says adding PPS from GPS steers Rb and eliminate the drift part of Adev.
I'm trying to do just that.
I did have a go at it, and results are rather disappointing. With PPS, the Allen dev of PRS-10 is 10 times worse at all interval! Oddly enough, partial graph collates fairly well. I did not tune anything and left everything as default. This is the first try. By default, PRS-10 comes standard with loop gain and loop constant suitable for GPS, so I went with it.
Now that I have this result, I am at a phase to think through what might have gone wrong. Ground loop and noise on PPS line comes first to my mind. I was rather sloppy in this area. I made the voltage where it's needed willy-nilly. I should be more careful. I didn't appreciate the level of scale I was dealing with. Also, GPS I used was of unknown nature. I now have few T-bolt, so I'll use that next.
I am not going to upload the graph.... It's so SO disappointing....... :palm:
Hm. It's difficult to imagine noise on the 1PPS signal being a factor. You'd need quite a lot of energy to disturb it.
A standard navigation type GPS would have quite a lot of jitter on the 1PPS and since the PRS-10 is not communicating with it, it can not apply quantization correction (sawtooth correction). That will require a lot of averaging to compensate for. I can imagine the loop constant on the PRS-10 will need tweaking.
-
Firstly, any result is a good result - post the graphs please.
Secondly, please explain in details what are you using as a DUT, reference source and measurement setup?
If you were to use the same source to produce 1PPS and to measure ADEV/MDEV against you can probably get better insight into short/mid term behaviour of the control loops.
Leo
-
Interesting articles:
NIST: https://www.nist.gov/pml/time-and-frequency-division/primary-standard-nist-f1 (https://www.nist.gov/pml/time-and-frequency-division/primary-standard-nist-f1)
with picture below
Getting Better All the Time: JILA Strontium Atomic Clock Sets New Records
https://www.nist.gov/news-events/news/2015/04/getting-better-all-time-jila-strontium-atomic-clock-sets-new-records (https://www.nist.gov/news-events/news/2015/04/getting-better-all-time-jila-strontium-atomic-clock-sets-new-records)
and: Released February 4, 2010, Updated March 21, 2018
https://www.nist.gov/news-events/news/2010/02/nists-second-quantum-logic-clock-based-aluminum-ion-now-worlds-most-precise (https://www.nist.gov/news-events/news/2010/02/nists-second-quantum-logic-clock-based-aluminum-ion-now-worlds-most-precise)
Hackaday: CONFESSIONS OF A REFORMED FREQUENCY STANDARD NUT >>
https://hackaday.com/2018/01/17/confessions-of-a-reformed-frequency-standard-nut/ (https://hackaday.com/2018/01/17/confessions-of-a-reformed-frequency-standard-nut/)
Mark the progress in the last 50 years.
Quote: There is a seduction to measurement, something that draws you in until it becomes an obsession.
-
By popular demand, here's the graph and the details.
The graph was made by measuring each DUT, one at a time, by HP53132A counter. Gate time was set to 1s so that I will get 12 digit resolution. HP53132A had GPSTM by Nortel (basically a T-bolt inside) as sync source. Between the counter and the GPSDO are two distribution amp and RG58. Antenna is 28dbi gain type and goes through a series of distribution amps, 58536A. It feeds both the timing source and the below GPS.
HP53122A was connected to TimeLab (latest edition) via RS232C and recorded as "Talk only" device.
DUT #1 is a PRS-10 at default setting connected to 1 pps by GPS. GPS is this one:
https://www.tindie.com/products/nsayer/gps-clock/ (https://www.tindie.com/products/nsayer/gps-clock/) Version 3, I believe. It uses a Skytraq Venus838LPx-T chip, which is a timing grade chip. I confirmed PPS is valid and PRS10 recognizes it.
DUT #2 is just a PRS-10 (same one).
Common to both is an internal distribution module for PPS _OUTPUT_ which consists of 74AC07 and 10MHz output distribution amp, which is an unmodified video distribution unit.
Power supply situation is a mess.
24V comes from linear regulator and feeds PRS10
That 24V goes through 7505 3 terminal regulator, becomes 5V and feeds 74AC07
There is a separate 12V switching supply that feeds the video amp. At the same time, it goes through a buck converter and feeds the GPS module (the one that feeds PRS10).
Now the graph....
Blue line is DUT #1 (14 days)
Magenta line is DUT #2 (3 days)
Please note, there is a waviness around 10s. If I do modified Allen, these are obvious impact from mains. It kind of averaged out now but it was obvious (In modified Allen) in first few days.
What I am realizing now is that I didn't pay any attention to ground loops and ground points. There is a path for power which has multiple (duplicate) connections to the cabinet, and there is a signal ground which is connected willy-nilly. I definitely revisit this.
Also, from datasheet and schematic, I can see this is a timing grade implementation but I have no other data. I used this because I wanted a time display. It may be worthwhile to substitute a known unit, such as T-bolt, for if nothing else, verification. (or as suggested, use the same timing source as the counter)
Another thing is phase shifting through dist amps and coax. I could use a dedicated setup just for this, which will eliminate the possibility. Also, I need to eliminate dist amps for outputs for purpose of measurement as these are unknown entity.
-
By the way....
I am aware of issues using GPSDO to measure something supposedly better than GPSDO alone. But this is the best I can do. I do have an older Cesium but its stability and accuracy is very much in doubt. In fact, I have no way to measure that conclusively, either.
I was expecting something similar to what is described in PRS10 literature. It came out nowhere close.
(please see a graph on page 3)
-
There's more that's right with your graphs than you might think.
Remember that at any value of Tau, an Allan Deviation graph shows you an RMS combination of the performance of your measurement system, your reference, and your DUT (Device Under Test). In this case, you could say we have two DUTs, but that's okay. For us hobbyists, our measurement system is typically the worst part of the system so it shows up first. If GPS is involved, you can usually pick out its contributions.
Let's start at Tau = 1 sec. The red graph shows ~ 3e-11. The 53132A is spec'ed as having a 150 ps resolution. Your measurement of 30 ps means that the counter is doing averaging. That may be okay. I never use frequency for measurements like this because I'm always suspicious of things like that. They tend to make things look better than they really are. I prefer a nice, clean, simple time interval. But, even so, it's the worst performer so from Tau = 1 sec. up to Tau = 10 sec. the 53132A is all you're seeing because if you measure its noise floor, that's what you will see. [1] So we know that the PRS10 must have a lower (better) ADEV. But why is the blue graph 10 times worse? We know it's not the counter or the PRS10, so it has to be the GPS or perhaps your test setup. Check the settings on the counter. If you're triggering on a sine wave, you probably want AC coupled triggering at, or maybe slightly offset from, zero volts. If you're triggering on a 1 PPS signal, you want DC triggering at about 1/2 of the pulse's amplitude. Adjust these as necessary for your exact situation. Inspecting the signals with a 'scope may be beneficial.
Now let's move to Tau = 10 sec. through to 1000 sec. First, regarding the wiggles - they're trivial. If you can get rid of them by grooming your test setup, great! If not, don't worry about it. Even though the blue graph is high, it's starting to show GPSDO behaviour. It first tries to level out. This shows the performance of the internal oscillator. It then starts to curve down exactly as it should. As it cruises through 1000 sec. it's right on target. The red graph also looks good. Since the graph is basically straight from 1 sec. you know that's the 53132A you're seeing. So the PRS10 must be better than that. Then, at 100 sec. it switches and starts rising. That's the PRS10 rising past the counter and becoming visible. A rising curve like that suggests aging. Was the PRS10 - or some other part of the system - still warming up? Even a locked Rb needs time to settle down. Also, look at the value. We're down in the e-13 range!!!
Now let's look around 3000 - 4000 sec. Here the blue graph goes off the rails. It looks like it's NOT being disciplined anymore. I don't know what to make of that. More investigation would be required. But the red graph looks good. Despite the apparent aging of the PRS10, the graph turns and follows GPS. Why? Because you used the GPSTM as the reference for your 53132A so you have to expect the GPS 'signature' to appear in your data! That's why I NEVER use a GPSDO as my reference. A GPSDO is a perfect frequency reference, but it can color your data as we're seeing here. My old reference was an Efratom FRK Rubidium. My current one is an HP 5065A Rubidium. Here's my favorite picture that shows the GPS signature. Look at the graphs with the bold colors and dots. Four different types of GPSDO and they all end up running down, to the right, in parallel.
(http://www.leapsecond.com/pages/gpsdo/free-lock-gpsdo4.gif)
Bottom line here is that your measurements are pretty much what's expected. The only outlier is the high-Tau performance of your 1 PPS source as shown in the blue graph.
FYI, I've attached a recent data run I did where I compared my FTS 4065A to my HP 5065A. I included the specified limits for the 4065A. First you see my measurement system. Then you see the 4065A. It's only when you get past 10K seconds that the 5065A starts to appear.
By the way, you don't have to run Timelab in "Talk only" mode. It natively supports the 53132A. I don't know if that gives you any additional capabilities or not.
Ed
[1] I'm really not sure how to measure the frequency noise floor. Extend the gate time?
You can measure the time interval noise floor by starting with a good quality signal. If your PRS10 is configured for 1 PPS out, use that, otherwise use the 10 MHz output. Split the signal in two, preferably with a good 50 ohm splitter, but a BNC T-connector will do. Feed the two signals through different length coax cables to the Ch 1 and Ch2 inputs of the 53132A. Measure Time Interval from 1 to 2. Don't let the cables move or vibrate or change temperature. Let Timelab collect the data for a few hours. Your ADEV should show a straight line dropping down with a slope of -1. This will be a factor in every measurement you take with that counter.
Edit: I mixed up Time Interval & Frequency measurements/limits. Hopefully, this straightens things out. :palm:
-
Ed,
Thank you very much for your detailed analysis. You are right - I am having trouble interpreting the graph. I am pretty much a newbie-time-semi-nuts. Your discussion of the graph really helps. I will take time to read section-by-section tomorrow as it is getting late here in Eastern US. (Florida)
JUST few hours ago, I finally managed to set up GPIB with TimeLab to talk to HP52132A, so I don't have to rely on serial. I don't yet know if that will mean anything in terms of measurement. I does allow me to use other equipment with TimeLab though.
I am thinking, use of GPSDO as clock was a bad idea. As they are completely different GPSDO, I have two oscillators (standard and DUT) that's yanked all over the place. Maybe that's the missing piece in this puzzle. I am thinking maybe I'll just use double oven Xtal. Initial idea was, this combination of GPS and Rb will give me the stability and accuracy that I wanted.
I have a FTS4040A Cesium. Should I use THAT instead?? By the way, the signal being compared is a sine 10MHz, and the trigger is AC coupled. Actually, I didn't even think about that, so I will review it in depth.
Thanks again,
Taka
ps. In a perfect world, everyone has a Hydrogen Maser or two. Hello, tvb...
-
Unfortunately, I kind of messed up my analysis. I always use Time Interval instead of frequency for my measurements and I got mixed up because you're using frequency. |O I've edited my message to correct the mixup.
Ed
-
I wouldn't recommend the 4040A as your reference for a couple of reasons. First, if you look at the graph I just posted for my 4065A, it's okay, but not stellar when it comes to ADEV - no Cesium is. They, like GPS are frequency references. Second, there's still the problem of using up your Cesium. Once it's gone, you've got a really effective boat anchor! Using it as a reference means that it will be running a lot!
Searching for a good reference is a never-ending quest. Almost every time-nut moves from reference to reference as they find a better one. It's like mining for diamonds - digging through lots of ore, looking for a gem or two. They also use different references for different measurements. OCXO or DOCXO for short time intervals. Maybe Rb for middle intervals. Maybe GPS or Cesium for long term measurements - i.e. make a measurement every few days to check for drift, not ADEV. Yes, there are a few crazy people that have their own hydrogen masers, but let's remember ...."everything in moderation".... (says the guy with a Cesium reference ) ::) ;)
I don't know what you've got for OCXOs or DOCXOs. The best item you've mentioned is the PRS10. It's got decent low-Tau stability and good aging and noise. If you look here: http://www.ke5fx.com/rb.htm (http://www.ke5fx.com/rb.htm) you'll see a good comparison between some of the more popular Rb standards. The PRS10 looks quite good. I'd just run it without trying to discipline it to GPS. My current reference is an HP 5065A. I can't tell you how good it really is, but it's good enough for me. I have a few OCXOs and DOCXOs that might beat it at low Tau values, but I can't measure that low. :-//
Ed
-
Using a GPSDO as a reference for these kinds of measurements is really going to "color" your data a lot. While experimenting with my own GPSDO platform I was using one of these UCCM GPSDO modules as the 10MHz input and 1PPS from a different GPS timing receiver. You could really see the GPSDO at work, large cycles of 4500s long up- and down movement as it steered its internal OCXO around. A bit of context can be found here:
https://www.eevblog.com/forum/projects/diy-gpsdo-project-w-stm32-tdc7200/msg2858734/#msg2858734 (https://www.eevblog.com/forum/projects/diy-gpsdo-project-w-stm32-tdc7200/msg2858734/#msg2858734)
For this type of measurements you definitely need a more stable reference.
-
I have an LPRO-101 as well. Also half a dozen of HP10811-60111, a dozen or so of oscilloquartz 8663-01.
Moderation.... got it!
-
..... moderation .....
I just found some forgotten Efratom FRS stuff.... (um....)
Here's what I am doing now. Since we kind of think GPS or GPS related portion is at fault, I've decided to use something else to steer PRS10. I have some genuine Trimble Tbolt, so I took one, used its 10MHz out to feed the counter sync port, and took its (Tbolt) PPS to feed the PPS IN of PRS10. They are now synced up and I'm seeing an incredible result. (kind of expected) The PRS10 is still warming up, so I'll have to let it run for some time before I can say anything meaningful.
After this test proves GPS is at fault, I'll use an independent GPS to feed PPS IN and retest.
-
I am finding out, there is a major problem with my measuring setup. Corby is helping me with it right now (different thread). Until this is resolved, the project itself is on hold.
-
Thanks to many folks on this forum, my measuring system problem has been resolved. It was the parameter I gave to the TimeLab software. It was MY falut, not the equipment or software. It was reading incoming data as Hz when it should be MHz. Huge scaling issue. I'm currently doing a verification run on HP53132A noise floor. I'll have to do the same to my standard. Sadly, that also means every measurement I took in the past contain the same error. Luckily, data from this point on will be correct.
-
Thanks to many folks on this forum, my measuring system problem has been resolved. It was the parameter I gave to the TimeLab software. It was MY falut, not the equipment or software. It was reading incoming data as Hz when it should be MHz. Huge scaling issue. I'm currently doing a verification run on HP53132A noise floor. I'll have to do the same to my standard. Sadly, that also means every measurement I took in the past contain the same error. Luckily, data from this point on will be correct.
Easily done. I made a similar mistake when first using the Stable32 software - I entered files in Hz when that software expects frequencies to be an error ratio (f-f0)/f0 - this also made a huge difference! :)
-
My day job is a software developer / programmer at one of the major software company. I also spent some time as a support rep. I've learned always approach cases with major doubt in users' report, and never trust what's being said as fact.
In THIS case, the user was ME. Sure enough, I was the cause of problem. Doesn't surprise me, really. But I've been using the same setup for some time. It took me 6 month to realize my own mistake when corner case alerted me of possible issues.
It has been a good learning experience.
By the way, this project started first, because I wanted to do it. The major motivation was, a lot of people said I cannot do it, and SRS (who made PRS10) said the feature was really meant for those who make GPDSO and Rb for living. Well.... they are currently winning but the never counted those who are majorly obsessive! Let's see what happens in next 12 month. This is going to be a long project.