Re: [time-nuts] Long Wave Radio-Frequency standard testing

[Dave Daniel]

Answers inline

> On Jan 19, 2021, at 16:27, Bob kb8tq  wrote:
> 
> Hi
> 
> Assuming the goal is a normal ADEV or xDEV sort of calculation:
> 
> If you replace the raw phase values with zero that can mess things up
> 
> 0 seconds +20 ns
> 1 seconds +22 ns
> 2 seconds +23 ns
> 3 seconds +25 ns
> 4 seconds +27ns
> 5 seconds +29 ns
> 
> If you “loose” one of those 20 to 30 ns values and replace it with zero, you 
> have significantly
> changed the data set.

Ok

> 
> Even if you are looking at deltas,

Nope. It doesn’t work for deltas. 

> zero stuffing would be problematic with that
> (contrived) phase data set. 
> 
> 1 seconds +2
> 2 seconds +1
> 3 seconds +2
> 4 seconds +2
> 5 seconds +2
> 
> If the objective is something like a PLL then “hold at the last value” is the 
> only practical
> answer to the question. You don’t *have* the next value and you need to stuff 
> something
> into the control loop computation.

For a control loop, certainly. For just generating the waveform after DAC using 
interploation, it works well.

So, my suggestion doesn’t cover all use cases, and I learned something. That 
makes it a good day.

DaveD
> 
> Bob
> 
> 
>> On Jan 19, 2021, at 1:37 PM, Dave Daniel  wrote:
>> 
>> Or one can replace those values with zero. That eliminates them; averaging 
>> then proceeds without those values altering the most probable correct 
>> average.
>> 
>> DaveD
>> 
>>> On Jan 19, 2021, at 08:49, Bob kb8tq  wrote:
>>> 
>>> Hi
>>> 
>>> The normal approach to filling a gap is to put in a point that is the 
>>> average
>>> of the two adjacent points. The assumption is that this is a “safe” value 
>>> that
>>> will not blow up the result. That’s probably ok if it is done rarely. The 
>>> risk is
>>> that you are running a filter process (averaging is a low pass filter). 
>>> 
>>> If you pull out a *lot* of outliers and replace them, you are doing a lot 
>>> of filtering.
>>> Since you are measuring noise, filtering is very likely to improve the 
>>> result. 
>>> The question becomes - how representative is the result after a lot of this 
>>> or 
>>> that has been done? 
>>> 
>>> Obviously the answer to all this depends on what you are trying to do. If 
>>> you
>>> are running a control loop and the output improves, that’s fine. If you are 
>>> trying to provide an accurate measure of noise …. maybe not so much :) 
>>> 
>>> Bob
>>> 
 On Jan 19, 2021, at 2:15 AM, Gilles Clement  wrote:
 
 Hi, 
 Yes outliers removal creates gap in Stable32.
 The « fill »  function can fills gaps with interpolated values. 
 It does not change much the graphs, except in the low Tau area (see 
 attached). 
 Do you know a discussion of impact of outliers removal ? 
 Gilles. 
 
 
 
> Le 18 janv. 2021 à 22:06, Bob kb8tq  a écrit :
> 
> Hi
> 
> As you throw away samples that are far off the mean, you reduce the sample
> rate ( or at least create gaps in the record). Dealing with that could be 
> difficult.
> 
> Bob
> 
>>> On Jan 18, 2021, at 1:33 PM, Gilles Clement  wrote:
>>> 
>>> Hi
>>> 
>>> Very cool !!!
>>> 
>>> The red trace is obviously the one to focus on. Some sort of digital 
>>> loop that
>>> only operates under the “known good” conditions would seem to make 
>>> sense. 
>>> 
>>> Thanks for sharing 
>>> 
>>> Bob
>> 
>> Hi,
>> I tried something with the idea to consider night records fluctuations 
>> as « outliers » as compared to day records. 
>> Indeed the 3 days record mean value is flat and the histogram quite 
>> gaussian. 
>> So I processed the 3 days record (green trace) with Stable32’s « Check 
>> Function »,
>> while removing outliers with decreasing values of the Sigma Factor. The 
>> graph below shows the outcome. 
>> The graph with Sigma=0.8 (blue trace) connects rather well with the 1Day 
>> record (red trace). 
>> Would this be a workable approach ? 
>> Best, 
>> Gilles. 
>> 
>> 
>> 
>> 
>> 
>> ___
>> time-nuts mailing list -- time-nuts@lists.febo.com
>> To unsubscribe, go to 
>> http://lists.febo.com/mailman/listinfo/time-nuts_lists.febo.com
>> and follow the instructions there.
> 
> 
> ___
> time-nuts mailing list -- time-nuts@lists.febo.com
> To unsubscribe, go to 
> http://lists.febo.com/mailman/listinfo/time-nuts_lists.febo.com
> and follow the instructions there.
 
 ___
 time-nuts mailing list -- time-nuts@lists.febo.com
 To unsubscribe, go to 
 http://lists.febo.com/mailman/listinfo/time-nuts_lists.febo.com
 and follow the instructions there.
>>> 
>>> 
>>> ___
>>> time-nuts mailing list -- time-nuts@lists.febo.com
>>> To 

Re: [time-nuts] Long Wave Radio-Frequency standard testing

[Michael Wouters]

Hello Gilles

There's a reasonable way of treating data with gaps in it:

https://iopscience.iop.org/article/10.1088/0026-1394/45/6/S19

Essentially, any averaging interval with missing data is dropped from
the ADEV summation.
This reduces the number of intervals averaged over and increases the
uncertainty but is better than faking data.

Stable32 provides an  implementation of the algorithm. See p18 of the manual.

Cheers
Michael




On Tue, Jan 19, 2021 at 6:18 PM Gilles Clement  wrote:
>
> Hi,
> Yes outliers removal creates gap in Stable32.
> The « fill »  function can fills gaps with interpolated values.
> It does not change much the graphs, except in the low Tau area (see attached).
> Do you know a discussion of impact of outliers removal ?
> Gilles.
>
>
>
> > Le 18 janv. 2021 à 22:06, Bob kb8tq  a écrit :
> >
> > Hi
> >
> > As you throw away samples that are far off the mean, you reduce the sample
> > rate ( or at least create gaps in the record). Dealing with that could be 
> > difficult.
> >
> > Bob
> >
> >> On Jan 18, 2021, at 1:33 PM, Gilles Clement  wrote:
> >>
> >>> Hi
> >>>
> >>> Very cool !!!
> >>>
> >>> The red trace is obviously the one to focus on. Some sort of digital loop 
> >>> that
> >>> only operates under the “known good” conditions would seem to make sense.
> >>>
> >>> Thanks for sharing
> >>>
> >>> Bob
> >>
> >> Hi,
> >> I tried something with the idea to consider night records fluctuations as 
> >> « outliers » as compared to day records.
> >> Indeed the 3 days record mean value is flat and the histogram quite 
> >> gaussian.
> >> So I processed the 3 days record (green trace) with Stable32’s « Check 
> >> Function »,
> >> while removing outliers with decreasing values of the Sigma Factor. The 
> >> graph below shows the outcome.
> >> The graph with Sigma=0.8 (blue trace) connects rather well with the 1Day 
> >> record (red trace).
> >> Would this be a workable approach ?
> >> Best,
> >> Gilles.
> >>
> >>
> >>
> >>
> >>
> >> ___
> >> time-nuts mailing list -- time-nuts@lists.febo.com
> >> To unsubscribe, go to 
> >> http://lists.febo.com/mailman/listinfo/time-nuts_lists.febo.com
> >> and follow the instructions there.
> >
> >
> > ___
> > time-nuts mailing list -- time-nuts@lists.febo.com
> > To unsubscribe, go to 
> > http://lists.febo.com/mailman/listinfo/time-nuts_lists.febo.com
> > and follow the instructions there.
>
> ___
> time-nuts mailing list -- time-nuts@lists.febo.com
> To unsubscribe, go to 
> http://lists.febo.com/mailman/listinfo/time-nuts_lists.febo.com
> and follow the instructions there.

___
time-nuts mailing list -- time-nuts@lists.febo.com
To unsubscribe, go to 
http://lists.febo.com/mailman/listinfo/time-nuts_lists.febo.com
and follow the instructions there.

Re: [time-nuts] Long Wave Radio-Frequency standard testing

[Bob kb8tq]

Hi

Assuming the goal is a normal ADEV or xDEV sort of calculation:

If you replace the raw phase values with zero that can mess things up

0 seconds +20 ns
1 seconds +22 ns
2 seconds +23 ns
3 seconds +25 ns
4 seconds +27ns
5 seconds +29 ns

If you “loose” one of those 20 to 30 ns values and replace it with zero, you 
have significantly
changed the data set.

Even if you are looking at deltas, zero stuffing would be problematic with that
(contrived) phase data set. 

1 seconds +2
2 seconds +1
3 seconds +2
4 seconds +2
5 seconds +2

If the objective is something like a PLL then “hold at the last value” is the 
only practical
answer to the question. You don’t *have* the next value and you need to stuff 
something
into the control loop computation.

Bob


> On Jan 19, 2021, at 1:37 PM, Dave Daniel  wrote:
> 
> Or one can replace those values with zero. That eliminates them; averaging 
> then proceeds without those values altering the most probable correct average.
> 
> DaveD
> 
>> On Jan 19, 2021, at 08:49, Bob kb8tq  wrote:
>> 
>> Hi
>> 
>> The normal approach to filling a gap is to put in a point that is the average
>> of the two adjacent points. The assumption is that this is a “safe” value 
>> that
>> will not blow up the result. That’s probably ok if it is done rarely. The 
>> risk is
>> that you are running a filter process (averaging is a low pass filter). 
>> 
>> If you pull out a *lot* of outliers and replace them, you are doing a lot of 
>> filtering.
>> Since you are measuring noise, filtering is very likely to improve the 
>> result. 
>> The question becomes - how representative is the result after a lot of this 
>> or 
>> that has been done? 
>> 
>> Obviously the answer to all this depends on what you are trying to do. If you
>> are running a control loop and the output improves, that’s fine. If you are 
>> trying to provide an accurate measure of noise …. maybe not so much :) 
>> 
>> Bob
>> 
>>> On Jan 19, 2021, at 2:15 AM, Gilles Clement  wrote:
>>> 
>>> Hi, 
>>> Yes outliers removal creates gap in Stable32.
>>> The « fill »  function can fills gaps with interpolated values. 
>>> It does not change much the graphs, except in the low Tau area (see 
>>> attached). 
>>> Do you know a discussion of impact of outliers removal ? 
>>> Gilles. 
>>> 
>>> 
>>> 
 Le 18 janv. 2021 à 22:06, Bob kb8tq  a écrit :
 
 Hi
 
 As you throw away samples that are far off the mean, you reduce the sample
 rate ( or at least create gaps in the record). Dealing with that could be 
 difficult.
 
 Bob
 
>> On Jan 18, 2021, at 1:33 PM, Gilles Clement  wrote:
>> 
>> Hi
>> 
>> Very cool !!!
>> 
>> The red trace is obviously the one to focus on. Some sort of digital 
>> loop that
>> only operates under the “known good” conditions would seem to make 
>> sense. 
>> 
>> Thanks for sharing 
>> 
>> Bob
> 
> Hi,
> I tried something with the idea to consider night records fluctuations as 
> « outliers » as compared to day records. 
> Indeed the 3 days record mean value is flat and the histogram quite 
> gaussian. 
> So I processed the 3 days record (green trace) with Stable32’s « Check 
> Function »,
> while removing outliers with decreasing values of the Sigma Factor. The 
> graph below shows the outcome. 
> The graph with Sigma=0.8 (blue trace) connects rather well with the 1Day 
> record (red trace). 
> Would this be a workable approach ? 
> Best, 
> Gilles. 
> 
> 
> 
> 
> 
> ___
> time-nuts mailing list -- time-nuts@lists.febo.com
> To unsubscribe, go to 
> http://lists.febo.com/mailman/listinfo/time-nuts_lists.febo.com
> and follow the instructions there.
 
 
 ___
 time-nuts mailing list -- time-nuts@lists.febo.com
 To unsubscribe, go to 
 http://lists.febo.com/mailman/listinfo/time-nuts_lists.febo.com
 and follow the instructions there.
>>> 
>>> ___
>>> time-nuts mailing list -- time-nuts@lists.febo.com
>>> To unsubscribe, go to 
>>> http://lists.febo.com/mailman/listinfo/time-nuts_lists.febo.com
>>> and follow the instructions there.
>> 
>> 
>> ___
>> time-nuts mailing list -- time-nuts@lists.febo.com
>> To unsubscribe, go to 
>> http://lists.febo.com/mailman/listinfo/time-nuts_lists.febo.com
>> and follow the instructions there.
> 
> ___
> time-nuts mailing list -- time-nuts@lists.febo.com
> To unsubscribe, go to 
> http://lists.febo.com/mailman/listinfo/time-nuts_lists.febo.com
> and follow the instructions there.


___
time-nuts mailing list -- time-nuts@lists.febo.com
To unsubscribe, go to 

Re: [time-nuts] Ebay Huawei Ublox M8T Modules Warning

[Bob kb8tq]

Hi

I don’t know if anybody else bought any of these or not. 

> On Jan 9, 2021, at 10:20 PM, Bob kb8tq  wrote:
> 
> Hi
> 
> Just for the sake of listing all the variations:
> 
> https://www.ebay.com/itm/ON-SALE-U-BLOX-ublox-LEA-M8T-0-10-HUAWEI-GPS-Timing-Module-Board/333778776570?ssPageName=STRK%3AMEBIDX%3AIT&_trksid=p2057872.m2749.l2649
>  
> 
> 
> Is the M8T on what I would *guess* is the original board. Simply from looking 
> at the picture I think 
> you can guess pretty well what’s what on the board.  
> 
> Hint: If you buy quite a few …. errr … 10 … there s a pretty good chance that 
> something more than 
> 20% might get knocked off the price …. I bought a pretty large pile of stuff 
> so it’s not clear if that 
> had some impact on what offers got accepted …..
> 
> If that’s not your favorite RF connector on the board, there are SMA’s that 
> likely fit in the same footprint.
> 
> Bob

If you did, the link now goes to a new auction (at a higher price). 
That auction shows the board in it’s proper enclosure and provides a bit
more ( but not quite all) information on the module. Since they are now 
free shipping a much heavier gizmo, that might explain some of the 
price increase. 

If you look at the box, the connector on that box most certainly is not an 
HDMI connector. However it *does* tell you what signals are running around.
You have the PPS out on RS-422. You also get the serial out of the module
on another RS-422 pair. 

The clock in / clock out stuff … no idea. The EXT-INT pin on the M8T is driven 
by the 8051 CPU on the board. It’s a good bet that’s what those clocks are 
getting to.

The TI switcher chip on the board has it’s input clamped at 20V. It puts out 
6.2V. No
idea what the correct input is. It seemed to be very happy with the 15V I put 
on it. 
Pin 1 on J1 is power in. Pin 6 on J2 is ground. Ground also shows up on one of 
the mounting holes. The other pins on J2 appear to be 3 RS-422 pairs. 

All of the I/O lines are protected with clamp diodes. The antenna has multiple 
layers of protection. The debug pins on the C8051F320 come out to a connector
that may or may not be populated on this or that board. There is a flash chip 
on 
the back side and an EEPROM on the front. If somebody was more ambitious 
than I am, reprogramming the MCU to do fancy stuff might be possible. 

Fun !!!

Bob
___
time-nuts mailing list -- time-nuts@lists.febo.com
To unsubscribe, go to 
http://lists.febo.com/mailman/listinfo/time-nuts_lists.febo.com
and follow the instructions there.

[time-nuts] A Crystal & Oscillator resource

[John Allen]

Hi nuts, FYI I stumbled on this list of white papers on crystals and
oscillators etc.

 

Application Notes and Technical Briefs:

https://www.ctscorp.com/resource-center/application-notes/

 

Regards, John

 

John Allen

PC Support Solutions

Bolton, MA 01740

  Mailto:j...@pcsupportsolutions.com

M: 508 361-6229

 

___
time-nuts mailing list -- time-nuts@lists.febo.com
To unsubscribe, go to 
http://lists.febo.com/mailman/listinfo/time-nuts_lists.febo.com
and follow the instructions there.

Re: [time-nuts] Long Wave Radio-Frequency standard testing

[Dave Daniel]

Or one can replace those values with zero. That eliminates them; averaging then 
proceeds without those values altering the most probable correct average.

DaveD

> On Jan 19, 2021, at 08:49, Bob kb8tq  wrote:
> 
> Hi
> 
> The normal approach to filling a gap is to put in a point that is the average
> of the two adjacent points. The assumption is that this is a “safe” value that
> will not blow up the result. That’s probably ok if it is done rarely. The 
> risk is
> that you are running a filter process (averaging is a low pass filter). 
> 
> If you pull out a *lot* of outliers and replace them, you are doing a lot of 
> filtering.
> Since you are measuring noise, filtering is very likely to improve the 
> result. 
> The question becomes - how representative is the result after a lot of this 
> or 
> that has been done? 
> 
> Obviously the answer to all this depends on what you are trying to do. If you
> are running a control loop and the output improves, that’s fine. If you are 
> trying to provide an accurate measure of noise …. maybe not so much :) 
> 
> Bob
> 
>> On Jan 19, 2021, at 2:15 AM, Gilles Clement  wrote:
>> 
>> Hi, 
>> Yes outliers removal creates gap in Stable32.
>> The « fill »  function can fills gaps with interpolated values. 
>> It does not change much the graphs, except in the low Tau area (see 
>> attached). 
>> Do you know a discussion of impact of outliers removal ? 
>> Gilles. 
>> 
>> 
>> 
>>> Le 18 janv. 2021 à 22:06, Bob kb8tq  a écrit :
>>> 
>>> Hi
>>> 
>>> As you throw away samples that are far off the mean, you reduce the sample
>>> rate ( or at least create gaps in the record). Dealing with that could be 
>>> difficult.
>>> 
>>> Bob
>>> 
> On Jan 18, 2021, at 1:33 PM, Gilles Clement  wrote:
> 
> Hi
> 
> Very cool !!!
> 
> The red trace is obviously the one to focus on. Some sort of digital loop 
> that
> only operates under the “known good” conditions would seem to make sense. 
> 
> Thanks for sharing 
> 
> Bob
 
 Hi,
 I tried something with the idea to consider night records fluctuations as 
 « outliers » as compared to day records. 
 Indeed the 3 days record mean value is flat and the histogram quite 
 gaussian. 
 So I processed the 3 days record (green trace) with Stable32’s « Check 
 Function »,
 while removing outliers with decreasing values of the Sigma Factor. The 
 graph below shows the outcome. 
 The graph with Sigma=0.8 (blue trace) connects rather well with the 1Day 
 record (red trace). 
 Would this be a workable approach ? 
 Best, 
 Gilles. 
 
 
 
 
 
 ___
 time-nuts mailing list -- time-nuts@lists.febo.com
 To unsubscribe, go to 
 http://lists.febo.com/mailman/listinfo/time-nuts_lists.febo.com
 and follow the instructions there.
>>> 
>>> 
>>> ___
>>> time-nuts mailing list -- time-nuts@lists.febo.com
>>> To unsubscribe, go to 
>>> http://lists.febo.com/mailman/listinfo/time-nuts_lists.febo.com
>>> and follow the instructions there.
>> 
>> ___
>> time-nuts mailing list -- time-nuts@lists.febo.com
>> To unsubscribe, go to 
>> http://lists.febo.com/mailman/listinfo/time-nuts_lists.febo.com
>> and follow the instructions there.
> 
> 
> ___
> time-nuts mailing list -- time-nuts@lists.febo.com
> To unsubscribe, go to 
> http://lists.febo.com/mailman/listinfo/time-nuts_lists.febo.com
> and follow the instructions there.

___
time-nuts mailing list -- time-nuts@lists.febo.com
To unsubscribe, go to 
http://lists.febo.com/mailman/listinfo/time-nuts_lists.febo.com
and follow the instructions there.

Re: [time-nuts] Long Wave Radio-Frequency standard testing

[Bob kb8tq]

Hi

The normal approach to filling a gap is to put in a point that is the average
of the two adjacent points. The assumption is that this is a “safe” value that
will not blow up the result. That’s probably ok if it is done rarely. The risk 
is
that you are running a filter process (averaging is a low pass filter). 

If you pull out a *lot* of outliers and replace them, you are doing a lot of 
filtering.
Since you are measuring noise, filtering is very likely to improve the result. 
The question becomes - how representative is the result after a lot of this or 
that has been done? 

Obviously the answer to all this depends on what you are trying to do. If you
are running a control loop and the output improves, that’s fine. If you are 
trying to provide an accurate measure of noise …. maybe not so much :) 

Bob

> On Jan 19, 2021, at 2:15 AM, Gilles Clement  wrote:
> 
> Hi, 
> Yes outliers removal creates gap in Stable32.
> The « fill »  function can fills gaps with interpolated values. 
> It does not change much the graphs, except in the low Tau area (see 
> attached). 
> Do you know a discussion of impact of outliers removal ? 
> Gilles. 
> 
> 
> 
>> Le 18 janv. 2021 à 22:06, Bob kb8tq  a écrit :
>> 
>> Hi
>> 
>> As you throw away samples that are far off the mean, you reduce the sample
>> rate ( or at least create gaps in the record). Dealing with that could be 
>> difficult.
>> 
>> Bob
>> 
>>> On Jan 18, 2021, at 1:33 PM, Gilles Clement  wrote:
>>> 
 Hi
 
 Very cool !!!
 
 The red trace is obviously the one to focus on. Some sort of digital loop 
 that
 only operates under the “known good” conditions would seem to make sense. 
 
 Thanks for sharing 
 
 Bob
>>> 
>>> Hi,
>>> I tried something with the idea to consider night records fluctuations as « 
>>> outliers » as compared to day records. 
>>> Indeed the 3 days record mean value is flat and the histogram quite 
>>> gaussian. 
>>> So I processed the 3 days record (green trace) with Stable32’s « Check 
>>> Function »,
>>> while removing outliers with decreasing values of the Sigma Factor. The 
>>> graph below shows the outcome. 
>>> The graph with Sigma=0.8 (blue trace) connects rather well with the 1Day 
>>> record (red trace). 
>>> Would this be a workable approach ? 
>>> Best, 
>>> Gilles. 
>>> 
>>> 
>>> 
>>> 
>>> 
>>> ___
>>> time-nuts mailing list -- time-nuts@lists.febo.com
>>> To unsubscribe, go to 
>>> http://lists.febo.com/mailman/listinfo/time-nuts_lists.febo.com
>>> and follow the instructions there.
>> 
>> 
>> ___
>> time-nuts mailing list -- time-nuts@lists.febo.com
>> To unsubscribe, go to 
>> http://lists.febo.com/mailman/listinfo/time-nuts_lists.febo.com
>> and follow the instructions there.
> 
> ___
> time-nuts mailing list -- time-nuts@lists.febo.com
> To unsubscribe, go to 
> http://lists.febo.com/mailman/listinfo/time-nuts_lists.febo.com
> and follow the instructions there.


___
time-nuts mailing list -- time-nuts@lists.febo.com
To unsubscribe, go to 
http://lists.febo.com/mailman/listinfo/time-nuts_lists.febo.com
and follow the instructions there.