Author Topic: SIGLENT SDS1102CML oscilloscope problem  (Read 10809 times)

0 Members and 1 Guest are viewing this topic.

Offline Rick

  • Frequent Contributor
  • **
  • Posts: 309
  • Country: tr
SIGLENT SDS1102CML oscilloscope problem
« on: May 17, 2013, 04:45:07 pm »
Somebody from this forum mentioned on youtube that he had problems with a Rigol DS1052E due to saturated input amplifiers.
My Siglent SDS1102CML has the same kind of problem.
Although the video quality is poor I think it is quite clear.
 

Offline mswhin63

  • Frequent Contributor
  • **
  • Posts: 299
  • Country: au
Re: SIGLENT SDS1102CML oscilloscope problem
« Reply #1 on: May 18, 2013, 01:36:57 am »
Where are you taking the reference from?
.
 

Offline Rick

  • Frequent Contributor
  • **
  • Posts: 309
  • Country: tr
Re: SIGLENT SDS1102CML oscilloscope problem
« Reply #2 on: May 18, 2013, 03:09:25 pm »
Where are you taking the reference from?

From the scope itself. It's the square wave of the scope.
 

Offline mswhin63

  • Frequent Contributor
  • **
  • Posts: 299
  • Country: au
Re: SIGLENT SDS1102CML oscilloscope problem
« Reply #3 on: May 19, 2013, 03:20:02 am »
It could be the reference itself, try another reference if you can
.
 

Offline Rick

  • Frequent Contributor
  • **
  • Posts: 309
  • Country: tr
Re: SIGLENT SDS1102CML oscilloscope problem
« Reply #4 on: May 19, 2013, 10:25:58 am »
It could be the reference itself, try another reference if you can

You may be right. Thank you.
I have just tried to use another square wave, I couldn't see the same problem with that signal.
Still, there is a problem with the reference.
 

Offline rf-loop

  • Super Contributor
  • ***
  • Posts: 3069
  • Country: fi
  • Starting with DLL21
Re: SIGLENT SDS1102CML oscilloscope problem
« Reply #5 on: May 19, 2013, 11:21:18 am »
It could be the reference itself, try another reference if you can

Still, there is a problem with the reference.

Probe compensation signal is for probe compensation adjustment using probe 10x
It can not do with this signal?
You adjust vertical setting so that top and bottom of signal is visible on screen and adjust probe comp. If this can do, there is no problem with this reference.

If practice and theory is not equal it tells that used application of theory  is wrong or the theory itself is wrong.
It is much easier to think an apple fall to the ground than to think that the earth and the apple will begin to move toward each other and collide.
 

Offline Rick

  • Frequent Contributor
  • **
  • Posts: 309
  • Country: tr
Re: SIGLENT SDS1102CML oscilloscope problem
« Reply #6 on: May 19, 2013, 12:27:33 pm »
It could be the reference itself, try another reference if you can

Still, there is a problem with the reference.

Probe compensation signal is for probe compensation adjustment using probe 10x
It can not do with this signal?
You adjust vertical setting so that top and bottom of signal is visible on screen and adjust probe comp. If this can do, there is no problem with this reference.

This problem only happens when the signal top and bottom are outside the screen and I shift the signal to bring the top and bottom inside the screen, in 1x probe setting. That shouldn't happen, should it?
I do not see that problem with the 10x setting as I already compensated. It only affects the 1x mode. As I shift the signal up and down on the screen that effect is accentuated or attenuated.
 

Offline rf-loop

  • Super Contributor
  • ***
  • Posts: 3069
  • Country: fi
  • Starting with DLL21
Re: SIGLENT SDS1102CML oscilloscope problem
« Reply #7 on: May 20, 2013, 06:56:00 am »
It could be the reference itself, try another reference if you can

Still, there is a problem with the reference.

Probe compensation signal is for probe compensation adjustment using probe 10x
It can not do with this signal?
You adjust vertical setting so that top and bottom of signal is visible on screen and adjust probe comp. If this can do, there is no problem with this reference.

This problem only happens when the signal top and bottom are outside the screen and I shift the signal to bring the top and bottom inside the screen, in 1x probe setting. That shouldn't happen, should it?
I do not see that problem with the 10x setting as I already compensated. It only affects the 1x mode. As I shift the signal up and down on the screen that effect is accentuated or attenuated.


Voltage range is divided to two band. Low band is 200mV/div and less and high band is (206mV) 500mV/div and more.  (Edit: Mv -> mV)

Using  200mV/div there is "worst case" (and then agen with end of upper band)

With 200mV/div (input DC coupled):
Signal 2.5V p-p (ecample square -1.25V - +1.25) with zero dc offset. Now top and bottom is 2.5div over display area.  If now adjust oscilloscope vertical shift up, so that signal bottom move to display top just visible, signal bottom shape is still acceptable ok.  If adjust vertical shift (down) so that signal top is bottom of display just visible signal shape is acceptable ok. 
Where ever is signal (not in display vertical area) it need know that ADC range is around visible vertical range)
If 200mV/div range signal DC level is more than +/-1.25V (2.5Vpp with zero offset) and adjust offset so that signal bottom or top  is moved to display area this signal is distorted. More level and more distortion.
 

This limit need know for avoid problems.  With 200mV/div this signal level is 12.5div (display area is 8div)  But agen, ADC range is around display range, so if you stop scope and vertically shift or zoom. signal top and bottom is clipped. This is natural. Of course it can do different but, who want it becouse reality is that ADC is 8 bit and it is good that this all resolution is used for displayed area where also measurements are made.



If oscilloscope vertical shift is zero but do it with signal itself offset, just same.
(these test images I have loosed but later also designed better tests for characterize input stage  limits)

It can roughly tell that with 200mV/div maximum signal level is 2.5Vp-p.


(this 200mV/div range is perhaps most narrow proportionally. (also highest voltages in upper voltage band but I  do not have now so high p-p level + offset level signal generator for reliable test it. Tek scope calibrator have high voltages of course but then, no V offset.  But good practice is keep signal p-p level less that displayed area.

It need remember that ADC range is near same as visible display, idependent of vertical offset.

Low voltage band, oscilloscope vertical  offset range is  +/- 1.6V
High voltage band it is +/-40V. This I can not test now becouse I do not have reliable signal source for this offset voltage.


Edit: Mistake in level thinking or better say, I did not understand my own data.
Edit:  added this new test A


A>>>>>>>>>>>>>>>

I made now some amount better test with this "worst case" voltage range what is just this 200mV/div.

200mV/div display vertical area is 1.6V.
I have tested now with separate injected DC offset. Test freq 2.5kHz.

Scope 200mV/div and DC
Signal -800mV to +800mV square.
Signal top and bottom just visible.


Adjusted oscilloscope vertical shift up so that signal bottom is just top of display barely visible.
Signal shape ok.

Adjusted oscilloscope vertical shift down so that signal top is just bottom of display barely visible.
Signal shape ok.

Oscilloscope vertical shift (offset) -1.6V
Add DC offset to signal +1.6V so that signal is just center.
Signal shape ok.

Add more (double) offset so that signal bottom moves to display top.
Signal shape ok.


Oscilloscope vertical shift (offset) +1.6V
Add DC offset to signal -1.6V so that signal is just center.
Signal shape ok.

Add more (double) offset so that signal bottom moves to display top.
Signal shape ok.


If rise signal p-p value over this 1.6V in 200mV DC range, one or more (A) tests start fail... more and more as signal peak peak level rise.

I recomment to think 1.6V is maximum signal peak peak level if use 200mV/div settings.
Also this is nearly full range for ADC

>>>>>>>>>>>>>A







Later I will do better tests for input amplifier limits and offset. If there is too high level signal there can adjust offset so that signal bottom or top is visible but distorted/clipped. This is in my opinion more like mis use oscilloscope than error in design. Of course there can limit maximum offset range so that user can not adjust it so that there can see this,  but then we loose something (as can see these test images where is used full offset range but signal p-p level is inside display (and ADC) range. I think better is "know your equipment".   Of course if design it for 3 voltage bands it is more easy (with cheap way) handle this input amplifier-signal max level-offset-display range.  Example Owon have made this selection, there is three main bands for whole V range. In Siglent this model and  if I remember right also Rigol and Hantek have this same two band input circuit principle.

Low voltage band offset range test with 5mV/div  and with 200mV/div
5mV/div signal is 30mVpp and 200mV/div signal is 1.2Vpp.




« Last Edit: May 20, 2013, 04:56:11 pm by rf-loop »
If practice and theory is not equal it tells that used application of theory  is wrong or the theory itself is wrong.
It is much easier to think an apple fall to the ground than to think that the earth and the apple will begin to move toward each other and collide.
 

Offline Rick

  • Frequent Contributor
  • **
  • Posts: 309
  • Country: tr
Re: SIGLENT SDS1102CML oscilloscope problem
« Reply #8 on: May 20, 2013, 10:56:07 am »

Voltage range is divided to two band. Low band is 200Mv/div and less and high band is (206mV) 500mV/div and more.

Using  200mV/div there is "worst case" (and then agen with end of upper band)

With 200mV/div (input DC coupled):
Signal 2.5V p-p (ecample square -1.25V - +1.25) with zero dc offset. Now top and bottom is 2.5div over display area.  If now adjust oscilloscope vertical shift up, so that signal bottom move to display top just visible, signal bottom shape is still acceptable ok.  If adjust vertical shift (down) so that signal top is bottom of display just visible signal shape is acceptable ok. 
Where ever is signal (not in display vertical area) it need know that ADC range is around visible vertical range)
If 200mV/div range signal DC level is more than +/-1.25V (2.5Vpp with zero offset) and adjust offset so that signal bottom or top  is moved to display area this signal is distorted. More level and more distortion.  This limit need know for avoid problems.  With 200mV/div this signal level is 12.5div (display area is 8div)  But agen, ADC range is around display range, so if you stop scope and vertically shift or zoom. signal top and bottom is clipped. This is natural. Of course it can do different but, who want it becouse reality is that ADC is 8 bit and it is good that this all resolution is used for displayed area where also measurements are made.

If oscilloscope vertical shift is zero but do it with signal itself offset, just same.
(these test images I have loosed but later also designed better tests for characterize input stage  limits)

It can roughly tell that with 200mV/div maximum signal level is 2.5Vp-p. (this 200mV/div range is perhaps most narrow proportionally. (also highest voltages in upper voltage band but I  do not have now so high p-p level + offset level signal generator for reliable test it. Tek scope calibrator have high voltages of course but then, no V offset.  But good practice is keep signal p-p level less that displayed area.

It need remember that ADC range is near same as visible display, idependent of vertical offset.

Low voltage band, oscilloscope vertical  offset range is  +/- 1.6V
High voltage band it is +/-40V. This I can not test now becouse I do not have reliable signal source for this offset voltage.

Low voltage band offset range test with 5mV/div  and with 200mV/div
5mV/div signal is 30mVpp and 200mV/div signal is 1.2Vpp.

Thank you for the info.
 

Offline Rick

  • Frequent Contributor
  • **
  • Posts: 309
  • Country: tr
Re: SIGLENT SDS1102CML oscilloscope problem
« Reply #9 on: May 20, 2013, 02:48:51 pm »
By the way the screen update rate seems to be slow. I think the screen is updated every second. Is this to be expected? How does it compare to Rigol 1052E or equivalent Owons?
 

Offline tinhead

  • Super Contributor
  • ***
  • Posts: 1903
  • Country: 00
    • If you like my hacks, send me a donation
Re: SIGLENT SDS1102CML oscilloscope problem
« Reply #10 on: May 20, 2013, 03:24:16 pm »
By the way the screen update rate seems to be slow. I think the screen is updated every second. Is this to be expected? How does it compare to Rigol 1052E or equivalent Owons?

it seems you mixing things, screen and waveform update rate are two complettly different things.

Example Hantek/Tekway DST/DSO/MSO is update waveforms up to 2500 times per second, but screen up to 50
times per second, where Owon SDS is upgating screen up to 30 times per second and waveforms as well up to
30 times per second as the firmware is "locked" to screen update rate.

Typical Rigol E/Siglent CNL/CML waveform update rate is ~ 800 times per second (wfms/s), and 2000 waveforms
per second for Siglent CFL and Rigol CA (and as well for the Agilent/LeCroy versions of these DSOs).
« Last Edit: May 20, 2013, 03:30:08 pm by tinhead »
I don't want to be human! I want to see gamma rays, I want to hear X-rays, and I want to smell dark matter ...
I want to reach out with something other than these prehensile paws and feel the solar wind of a supernova flowing over me.
 

Offline Rick

  • Frequent Contributor
  • **
  • Posts: 309
  • Country: tr
Re: SIGLENT SDS1102CML oscilloscope problem
« Reply #11 on: May 20, 2013, 05:41:33 pm »
By the way the screen update rate seems to be slow. I think the screen is updated every second. Is this to be expected? How does it compare to Rigol 1052E or equivalent Owons?

it seems you mixing things, screen and waveform update rate are two complettly different things.

Example Hantek/Tekway DST/DSO/MSO is update waveforms up to 2500 times per second, but screen up to 50
times per second, where Owon SDS is upgating screen up to 30 times per second and waveforms as well up to
30 times per second as the firmware is "locked" to screen update rate.

Typical Rigol E/Siglent CNL/CML waveform update rate is ~ 800 times per second (wfms/s), and 2000 waveforms
per second for Siglent CFL and Rigol CA (and as well for the Agilent/LeCroy versions of these DSOs).

I realize the difference.
However I was referring to the screen update rate. I guess that for the SDS1102CML the screen update rate should be of the same order of magnitude as the values you mention (30, 50...). So my screen update rate is too slow. When I connect or disconnect  the probes, my SDS1102CML shows the new state one second later on the screen. This is related with the screen update rate isn't it? The seller told me it must be adjustable... But I cannot find any setting for that. Is it possible to adjust that screen update rate or ...?

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

May be I am wrong again, when I disconnect the probe it just freezes the display, otherwise as soon as the probe is connected the signal shows up on the screen.
« Last Edit: May 20, 2013, 07:25:11 pm by Rick »
 

Offline Deckert

  • Regular Contributor
  • *
  • Posts: 137
  • Country: za
    • TechBench
Re: SIGLENT SDS1102CML oscilloscope problem
« Reply #12 on: May 20, 2013, 09:22:45 pm »
When I connect or disconnect  the probes, my SDS1102CML shows the new state one second later on the screen. This is related with the screen update rate isn't it?

Disconnecting the probes is not a reliable way of testing screen update rate - best get a signal generator to do that. You must realize how the scope does it's updates - and will only do so while it has a reliable trigger

For example, if you have the scope set to auto-trigger and you have it connected to a 1kHz reference input, the update will happen continually while it is triggering. When you disconnect the probes, there may not be any trigger available any more, so the scope takes a second to auto-trigger and show 0V line.

Switch it to Normal trigger mode, disconnect and you'll see it's as if the display freezes - that's simply because there is no trigger for it to act upon and it shows the last triggered wave. When I connect my CML to an adjustable signal source (e.g. wave-gen) and adjust frequency and toggle waveform type, the update on the display is instantaneous.

--deckert
 

Offline Rick

  • Frequent Contributor
  • **
  • Posts: 309
  • Country: tr
Re: SIGLENT SDS1102CML oscilloscope problem
« Reply #13 on: May 20, 2013, 09:56:57 pm »
When I connect or disconnect  the probes, my SDS1102CML shows the new state one second later on the screen. This is related with the screen update rate isn't it?

Disconnecting the probes is not a reliable way of testing screen update rate - best get a signal generator to do that. You must realize how the scope does it's updates - and will only do so while it has a reliable trigger

For example, if you have the scope set to auto-trigger and you have it connected to a 1kHz reference input, the update will happen continually while it is triggering. When you disconnect the probes, there may not be any trigger available any more, so the scope takes a second to auto-trigger and show 0V line.

Switch it to Normal trigger mode, disconnect and you'll see it's as if the display freezes - that's simply because there is no trigger for it to act upon and it shows the last triggered wave. When I connect my CML to an adjustable signal source (e.g. wave-gen) and adjust frequency and toggle waveform type, the update on the display is instantaneous.

--deckert

Thank you. It works fine then. It shows the frequency changes immediately.
 

Offline rf-loop

  • Super Contributor
  • ***
  • Posts: 3069
  • Country: fi
  • Starting with DLL21
Re: SIGLENT SDS1102CML oscilloscope problem
« Reply #14 on: May 21, 2013, 07:36:27 pm »
« Last Edit: May 21, 2013, 07:38:40 pm by rf-loop »
If practice and theory is not equal it tells that used application of theory  is wrong or the theory itself is wrong.
It is much easier to think an apple fall to the ground than to think that the earth and the apple will begin to move toward each other and collide.
 


Share me

Digg  Facebook  SlashDot  Delicious  Technorati  Twitter  Google  Yahoo
Smf