Author Topic: EEVblog #683 - Rigol DS1000Z & DS2000 Oscilloscope Jitter Problems  (Read 505483 times)

0 Members and 1 Guest are viewing this topic.

Offline HexfeT

  • Newbie
  • Posts: 8
  • Country: tr
Re: EEVblog #683 - Rigol DS1000Z & DS2000 Oscilloscope Jitter Problems
« Reply #225 on: November 16, 2014, 04:53:55 pm »
Playing around some more.  Notice how memory depth/averaging change the artifact behaviour.

if you change memory depth, changes how much waveform captured and displayed. Long memory depth mean less waveform capturing.

Also notice when scope is stopped, horizontal position changed, the artifacts disappear.

This is normal. When you stopped, only one waveform trace displaying.

if different trigger offset value no affect displayed waveform, jitter is not periodycally come and go (like dave's) probably your signal source is bad.
 

Offline dr.diesel

  • Super Contributor
  • ***
  • Posts: 2214
  • Country: us
  • Cramming the magic smoke back in...
Re: EEVblog #683 - Rigol DS1000Z & DS2000 Oscilloscope Jitter Problems
« Reply #226 on: November 16, 2014, 05:03:07 pm »
No problem, 10Mhz OCXO:
thank you. but it seems I wasn't clear enough:

Please feed the 10 MHz from the OCXO into the external reference input on the back of the scope. You then have to select Reference Clock Input in the utilty menu.
Then measure e.g. the same 16 MHz TCXO from the video before.

This way we can see if the reason for the jitter is jitter from the internal oscillator of the scope.

Gotcha, Scope is using external GPSDO already, and was in place in all of my videos.  Scope internal oscillator or external GPSDO does not appear to have any effect.

Offline dr.diesel

  • Super Contributor
  • ***
  • Posts: 2214
  • Country: us
  • Cramming the magic smoke back in...
Re: EEVblog #683 - Rigol DS1000Z & DS2000 Oscilloscope Jitter Problems
« Reply #227 on: November 16, 2014, 05:05:18 pm »
This is normal. When you stopped, only one waveform trace displaying.

I don't  think so, the artifact does not disappear when stopped.  Only disappears when I change the horizontal position when stopped.

Offline HexfeT

  • Newbie
  • Posts: 8
  • Country: tr
Re: EEVblog #683 - Rigol DS1000Z & DS2000 Oscilloscope Jitter Problems
« Reply #228 on: November 16, 2014, 05:11:14 pm »
This is normal. When you stopped, only one waveform trace displaying.

I don't  think so, the artifact does not disappear when stopped.  Only disappears when I change the horizontal position when stopped.

Because when you stopped, displaying last multiple traces same time. When you change horizontal position, displaying last one.
« Last Edit: November 16, 2014, 05:23:48 pm by HexfeT »
 

Offline miguelvp

  • Super Contributor
  • ***
  • Posts: 5550
  • Country: us
Re: EEVblog #683 - Rigol DS1000Z & DS2000 Oscilloscope Jitter Problems
« Reply #229 on: November 16, 2014, 07:04:24 pm »
So on the first problem of moving the trigger point 5 us in the past, it is clear that the signal source has a lot to do with the jitter and extremely stable signals might not show the problem.

How does that explain that other scopes don't show that problem? and is the signal really that bad that jitters by that much?

Furthermore why will a bad input signal correct itself at exactly 5 us extra? I can see that happen one in a million sources that has some harmonic at 10 KHz (if my math is right on the frequency for 10 us), Or maybe the if the same function generator was used and it has that harmonic in their design. But that will show up in other scopes as well.

Edit: I hope Rigol finds and fixes the problem. I also hope Rigol don't mask the problem by just making it look like it works, then we won't be able to capture real jitter, but that should be easy to test when/if they do fix the problem so I doubt they will take that approach.
« Last Edit: November 16, 2014, 07:11:09 pm by miguelvp »
 

Offline marmad

  • Super Contributor
  • ***
  • Posts: 2979
  • Country: aq
    • DaysAlive
Re: EEVblog #683 - Rigol DS1000Z & DS2000 Oscilloscope Jitter Problems
« Reply #230 on: November 16, 2014, 07:07:02 pm »
Interesting to note that the AC Trigger coupling issue was never really noticed for two years on the well loved Rigol 2000 series.  As there is an option for AC trigger coupling I would have thought that Rigol would have a test case for it.
Again, as has already been mentioned more than once on the forum: the AC-coupled trigger jitter on the DS2000 was already noticed, reported, and talked about here about a year and a half ago. We also discussed the trigger offset problem (which Dave didn't mention) with the AC-coupled trigger.

I'm sorry Dave was having a brain aneurysm, but he wasn't reporting anything about the DS2000 that most of us older owners of the DSO didn't already know.

It would be nice if Rigol fixed it - but in the list of bugs that need(ed) fixing on the DSO it didn't/doesn't rank high. Just use DC coupling and manually adjust the level.
 

Online tszaboo

  • Super Contributor
  • ***
  • Posts: 7307
  • Country: nl
  • Current job: ATEX product design
Re: EEVblog #683 - Rigol DS1000Z & DS2000 Oscilloscope Jitter Problems
« Reply #231 on: November 16, 2014, 08:14:21 pm »
I believe I seen very similar DC triggering issues with TEK TDS2000 scopes a few years back. When the trigger wasnt on the screen it had more jitter than when it was on the screen. I remember, because we based design decisions on this, and we almost got burned.
And also, I believe that problem was fixed with a firmware update.

I'm curious, this case: Is there any SMPS in the scope working at 100Khz? Is it possible that the internal 10Mhz reference signal has this jitter? This is where your universal counter would come handy, Dave.
 

Offline alank2

  • Super Contributor
  • ***
  • Posts: 2183
Re: EEVblog #683 - Rigol DS1000Z & DS2000 Oscilloscope Jitter Problems
« Reply #232 on: November 16, 2014, 09:06:21 pm »
Could it be anything to do with the intensity graded display, what happens when the intensity graded display is switched off?

This is what I wonder.  I'm not sure if the intensity grading can be turned off, can it?
 

Offline David Hess

  • Super Contributor
  • ***
  • Posts: 16545
  • Country: us
  • DavidH
Re: EEVblog #683 - Rigol DS1000Z & DS2000 Oscilloscope Jitter Problems
« Reply #233 on: November 16, 2014, 09:10:17 pm »
So on the first problem of moving the trigger point 5 us in the past, it is clear that the signal source has a lot to do with the jitter and extremely stable signals might not show the problem.

The signal source itself needs to have low long period jitter (at 5 us, 10 us, etc.) for a problem with the oscilloscope itself to show up.  If the signal source jitter is high, then a problem with the oscilloscope will be masked.

Quote
How does that explain that other scopes don't show that problem? and is the signal really that bad that jitters by that much?

The jitter at 5 us and at intervals of 10 us after that appears to be frequency modulation of the DSO sampling clock at 100 kHz and apparently does not affect all units.

Quote
Furthermore why will a bad input signal correct itself at exactly 5 us extra? I can see that happen one in a million sources that has some harmonic at 10 KHz (if my math is right on the frequency for 10 us), Or maybe the if the same function generator was used and it has that harmonic in their design. But that will show up in other scopes as well.

It could be a problem in the signal source but the other oscilloscopes did not show it.

If the signal source or the oscilloscope sampling clock is frequency modulated at 100 kHz, what happens is that for half of the 100 kHz cycle or 5 us, the average frequency may be off by x amount.  During the next half cycle, the average frequency will be off by -x amount.  Over both halves of the complete cycle of modulation, the errors cancel out so there is no error at 10 us and multiples of 10 us.  The error at 5 us and 10 us increments after that is from the odd number of half cycles of modulation which do not cancel out.

Now the above might seem contrived but some clock sources are deliberately designed to be modulated for EMC (electromagnetic compliance) reasons and 100 kHz is a common modulation frequency in this application.  This is usually referred to as spread spectrum clocking:

http://en.wikipedia.org/wiki/Spread_spectrum#Spread-spectrum_clock_signal_generation

Rigol might accidentally have used a clock oscillator which includes clock spreading or it could have happened in other ways.
« Last Edit: November 16, 2014, 09:13:41 pm by David Hess »
 

Offline dr.diesel

  • Super Contributor
  • ***
  • Posts: 2214
  • Country: us
  • Cramming the magic smoke back in...
Re: EEVblog #683 - Rigol DS1000Z & DS2000 Oscilloscope Jitter Problems
« Reply #234 on: November 16, 2014, 09:13:04 pm »
Could it be anything to do with the intensity graded display, what happens when the intensity graded display is switched off?

This is what I wonder.  I'm not sure if the intensity grading can be turned off, can it?

From 0-100% it doesn't appear to alter the quantity/size/shape of the artifacts.  No off switch that I can find.

Offline David Hess

  • Super Contributor
  • ***
  • Posts: 16545
  • Country: us
  • DavidH
Re: EEVblog #683 - Rigol DS1000Z & DS2000 Oscilloscope Jitter Problems
« Reply #235 on: November 16, 2014, 09:17:11 pm »
Could it be anything to do with the intensity graded display, what happens when the intensity graded display is switched off?

This is what I wonder.  I'm not sure if the intensity grading can be turned off, can it?

From 0-100% it doesn't appear to alter the quantity/size/shape of the artifacts.  No off switch that I can find.

Increasing the trigger holdoff may have the effect of disabling intensity grading.
 

Offline miguelvp

  • Super Contributor
  • ***
  • Posts: 5550
  • Country: us
Re: EEVblog #683 - Rigol DS1000Z & DS2000 Oscilloscope Jitter Problems
« Reply #236 on: November 16, 2014, 09:47:12 pm »
Thanks David,

Yeah the inverse of 0.00001 seconds it's indeed 100KHz  back to school for me :)
 

Offline Loboscope

  • Regular Contributor
  • *
  • Posts: 62
  • Country: de
Re: EEVblog #683 - Rigol DS1000Z & DS2000 Oscilloscope Jitter Problems
« Reply #237 on: November 16, 2014, 10:18:19 pm »
My DS1104Z (bought some month ago at Batronix, calibratetd before testing, Firmware 00.04.01.SP2, Board Rev. 0.1.1) shows the AC-Trigger issue, but doesen´t show the 5µs Delay-Jitter, tested with different frequencies (Signal source: Siglent SGD1025, connected directly with BNC-BNC coax-cable). The AC-Trigger is only relevant for freqencies higher than 1 MHz and gets worst with shorter horizontal settings as expected. LPF reduces the jitter, averaging supresses it perfectly.
Also by moving the AC-Trigger Point upwards to the maximum level of the signal on the screen, just before trigger gets lost (for example the top of the overshoot of a square wave) I can eliminate the jitter entirely. This indeed is a behavior I that is common on my older analogue scope (Philips PM 3055), there I often have to adjust the trigger level to achieve a stable picture. With trigger set to DC the test-signals will always be stable on my DS1104Z.
Frequency Counter works allways correct.

The AC-Triggering-issue obviously semms to be a common problem spread over different Rigol-scope-models, but the delay-issue seems not to be common. If my observation is correct, concerning the DS1kZ-Series, all of the DS1054Z mentioned in this thread show this issue, but only few to none DS1074Z and DS1104Z (like mine).

This lets me raise a theory: Could this issue be caused by spreading of specs of some parts on the boards? Therewith I mean, that when Rigol does a final testing of the boards as quality check at the end of the manufactoring process there maybe some boards rejected due to not reaching the specs to a margin of lets say 100%. But imagine (most of these) these boards are not defect, but only not good enough to be sold as 70 or 100 MHz boards. If there was a appreciable number of rejected but not defective boards maybe Rigol decided to sell them as 50 Mhz boards, the DS1054Z.
They will work up to this frequency but will not be accurate enough to met the specs of his predecessors. Who of the normal users of the DS1054Z can really carry out measurements of this precision to find out these certainly only very small differences - and from a practical view they would certainly never matter in normal use. It matters only Rigol to classify the different models.

BUT could this theory explain this observed phenomenon?
What will say the experts here in this forum whose knowledge i really admire?
« Last Edit: November 16, 2014, 11:25:01 pm by Loboscope »
 

Offline rs20

  • Super Contributor
  • ***
  • Posts: 2317
  • Country: au
Re: EEVblog #683 - Rigol DS1000Z & DS2000 Oscilloscope Jitter Problems
« Reply #238 on: November 16, 2014, 10:29:10 pm »
Increasing the trigger holdoff may have the effect of disabling intensity grading.

On my DS2202, the ac trigger issue persists identically in both these scenarios:
- Holdoff set to 1s
- Manually mashing the single shot button
 

Offline Teneyes

  • Frequent Contributor
  • **
  • Posts: 498
  • Country: ca
Re: EEVblog #683 - Rigol DS1000Z & DS2000 Oscilloscope Jitter Problems
« Reply #239 on: November 16, 2014, 11:42:47 pm »
Ok   with early DS2000,   SN  < 00350

Back on July 3 ,2013 I sent Rigol North American this
----------------------------------------------------------------------------------------------
Hello Jason and Joel
 
Here is a report of a Small bug on the Rigol DS2000 with latest Firmware 00.01.01.00.02
I think this Bug is in previous FW also.
 
FW 00.01.01.00.02
A small Bug also associated with the Trigger setting of AC Coupling
Note  Wim13 has already reported that there is large Trigger Jitter when this mode is selected

I appears that also in this mode of AC coupled Trigger,
the Display of the waveform is NOT in the  correct position.
The waveform position is 50nSec to the left (early)
The offset in a fixed amount of time 50nSec
The Offset position changes at each timebase setting to be = 50nSec!!

But when the trigger is DC coupling the Position is correct.at the center of the Display

------------------------------------------------------------------------------------------------------------------

The FW is now 00.03.01..... 

So I guess It was very low on the bug fix list at Rigol,  with the MSO, DS1000, 2000A models coming out


--------------------------------------

PS   using a "T" connector to the External Trigger on AC coupled triggering I see a more stable display
« Last Edit: November 25, 2014, 01:04:09 pm by Teneyes »
IiIiIiIiIi  --  curiosity killed the cat but, satisfaction brought it back
 

Offline David Hess

  • Super Contributor
  • ***
  • Posts: 16545
  • Country: us
  • DavidH
Re: EEVblog #683 - Rigol DS1000Z & DS2000 Oscilloscope Jitter Problems
« Reply #240 on: November 17, 2014, 03:42:26 am »
Thanks David,

Yeah the inverse of 0.00001 seconds it's indeed 100KHz  back to school for me :)

I have run across this problem before where a clock source is inadvertently frequency or phase modulated by some other signal.  Looking for the nulls or shortest period of time over which the clock returns to the correct position reveals the frequency of the interference.  An oscilloscope is one of the more common tools which can be used to directly observe this but spectrum analyzers and some frequency counters are better in a lot of cases.  The FFT display from one of the bad Rigols might reveal something.

I broke out my clipboard to make sure that the math matched the observations.

The AC-Triggering-issue obviously semms to be a common problem spread over different Rigol-scope-models, but the delay-issue seems not to be common. If my observation is correct, concerning the DS1kZ-Series, all of the DS1054Z mentioned in this thread show this issue, but only few to none DS1074Z and DS1104Z (like mine).

This lets me raise a theory: Could this issue be caused by spreading of specs of some parts on the boards? Therewith I mean, that when Rigol does a final testing of the boards as quality check at the end of the manufactoring process there maybe some boards rejected due to not reaching the specs to a margin of lets say 100%. But imagine (most of these) these boards are not defect, but only not good enough to be sold as 70 or 100 MHz boards. If there was a appreciable number of rejected but not defective boards maybe Rigol decided to sell them as 50 Mhz boards, the DS1054Z.

I would consider this more of a functional flaw than a deviation from the specifications.  The problem might be as simple as a part substitution or a configuration mistake.

Quote
They will work up to this frequency but will not be accurate enough to met the specs of his predecessors. Who of the normal users of the DS1054Z can really carry out measurements of this precision to find out these certainly only very small differences - and from a practical view they would certainly never matter in normal use. It matters only Rigol to classify the different models.

I cannot see DSOs being graded for this sort of specification.  Low jitter clocking is well understood and the magnitude of the problem here is relatively large.

Measuring long period jitter is one of the things DSOs should do well and even a 50 MHz one with a 7 nanosecond rise time should be able to resolve timing to better than a nanosecond.  Good analog oscilloscopes could have a jitter of about 250 picoseconds at a delay time of 5 microseconds which is significantly better than that displayed on the Rigols having problems.
 

Offline leppie

  • Frequent Contributor
  • **
  • Posts: 269
  • Country: za
Re: EEVblog #683 - Rigol DS1000Z & DS2000 Oscilloscope Jitter Problems
« Reply #241 on: November 17, 2014, 05:30:37 am »
An oscilloscope is one of the more common tools which can be used to directly observe this but spectrum analyzers and some frequency counters are better in a lot of cases.  The FFT display from one of the bad Rigols might reveal something.

I had a look just using the scopes own FFT, but could not see any difference.
 

Offline poida_pie

  • Regular Contributor
  • *
  • Posts: 119
  • Country: au
Re: EEVblog #683 - Rigol DS1000Z & DS2000 Oscilloscope Jitter Problems
« Reply #242 on: November 17, 2014, 06:19:23 am »
the DS1054Z arrived today.

One more data point:
Software version: 00.04.02.SP3
Board version: 0.1.1

Got teh jitters.
But also got >100MHz bandwidth too after applying some codes.


What a sensational DSO for the up and coming EE kids who are playing with Arduino/R-Pi/Pics/etc/etc
and inventing a new world as fast as they can!



 

Offline dr.diesel

  • Super Contributor
  • ***
  • Posts: 2214
  • Country: us
  • Cramming the magic smoke back in...
Re: EEVblog #683 - Rigol DS1000Z & DS2000 Oscilloscope Jitter Problems
« Reply #243 on: November 17, 2014, 11:15:55 am »
Back on July 3 ,2013 I sent Rigol North American this
----------------------------------------------------------------------------------------------
Hello Jason and Joel

FYI, Joel no longer works for Rigol.

Offline rf-loop

  • Super Contributor
  • ***
  • Posts: 4061
  • Country: fi
  • Born in Finland with DLL21 in hand
Re: EEVblog #683 - Rigol DS1000Z & DS2000 Oscilloscope Jitter Problems
« Reply #244 on: November 17, 2014, 11:49:51 am »
the DS1054Z arrived today.

One more data point:
Software version: 00.04.02.SP3
Board version: 0.1.1

Got teh jitters.
But also got >100MHz bandwidth too after applying some codes.


What a sensational DSO for the up and coming EE kids who are playing with Arduino/R-Pi/Pics/etc/etc
and inventing a new world as fast as they can!

And what is your sinewave  quality (specifications)  in picture 1 and 2.
Do you know how much it have jitter measured over 5us time period.
I drive a LEC (low el. consumption) BEV car. Smoke exhaust pipes - go to museum. In Finland quite all electric power is made using nuclear, wind, solar and water.

Wises must compel the mad barbarians to stop their crimes against humanity. Where have the wises gone?
 

Offline i4004

  • Regular Contributor
  • *
  • Posts: 71
  • Country: hr
Re: EEVblog #683 - Rigol DS1000Z & DS2000 Oscilloscope Jitter Problems
« Reply #245 on: November 17, 2014, 12:14:11 pm »
there was some discussion about is this a hardware or software issue:
but how would you achieve this in software alone?
are those spread-spectrum clocks configurable by software?
 

Offline han

  • Frequent Contributor
  • **
  • Posts: 311
  • Country: 00
Re: EEVblog #683 - Rigol DS1000Z & DS2000 Oscilloscope Jitter Problems
« Reply #246 on: November 17, 2014, 01:12:42 pm »
did ac coupling trigger jitter intensity change if the trigger level changed?
 

Offline tequipment

  • Regular Contributor
  • *
  • Posts: 207
  • Country: us
    • TEquipment
Re: EEVblog #683 - Rigol DS1000Z & DS2000 Oscilloscope Jitter Problems
« Reply #247 on: November 17, 2014, 02:39:00 pm »
Hello everyone.  I am not sure if you saw the Rigol response back.  If not here it is:

https://www.tequipment.net/assets/1/7/EEVBlog_message1.pdf

Thanks
Evan Cirelli

TEquipment.NET

 

Offline David Hess

  • Super Contributor
  • ***
  • Posts: 16545
  • Country: us
  • DavidH
Re: EEVblog #683 - Rigol DS1000Z & DS2000 Oscilloscope Jitter Problems
« Reply #248 on: November 17, 2014, 02:53:02 pm »
Got teh jitters.
But also got >100MHz bandwidth too after applying some codes.

The jitter at 10us matters as well because if it is high there, then either the test source is noisy or there is a different problem.

[/quote]
there was some discussion about is this a hardware or software issue:
but how would you achieve this in software alone?
are those spread-spectrum clocks configurable by software?

Some of them are.  The oscillators often have an input which sets the mode and the same applies to PLL clock generators which support this function.
 

Offline jlmoon

  • Supporter
  • ****
  • Posts: 609
  • Country: us
  • If you fail the first time, keep trying!
Re: EEVblog #683 - Rigol DS1000Z & DS2000 Oscilloscope Jitter Problems
« Reply #249 on: November 17, 2014, 03:10:13 pm »
Interesting to note that the AC Trigger coupling issue was never really noticed for two years on the well loved Rigol 2000 series.  As there is an option for AC trigger coupling I would have thought that Rigol would have a test case for it.
Again, as has already been mentioned more than once on the forum: the AC-coupled trigger jitter on the DS2000 was already noticed, reported, and talked about here about a year and a half ago. We also discussed the trigger offset problem (which Dave didn't mention) with the AC-coupled trigger.

I'm sorry Dave was having a brain aneurysm, but he wasn't reporting anything about the DS2000 that most of us older owners of the DSO didn't already know.

It would be nice if Rigol fixed it - but in the list of bugs that need(ed) fixing on the DSO it didn't/doesn't rank high. Just use DC coupling and manually adjust the level.

As a a old DS2202 owner I remember the previous post as well.  One reason I just threw my 2202 in my field kit and chose to keep my Agilent stuff cozy and warm in my lab environment.  The Chinese equipment is not ready for the rigors of serious lab grade analysis and testing ... yet!!
Recharged Volt-Nut
 


Share me

Digg  Facebook  SlashDot  Delicious  Technorati  Twitter  Google  Yahoo
Smf