Author Topic: Rigol DS1052E (an)other nasty surprise?  (Read 11318 times)

0 Members and 1 Guest are viewing this topic.

Offline goober

  • Newbie
  • Posts: 3
  • Country: au
Rigol DS1052E (an)other nasty surprise?
« on: July 06, 2011, 10:51:29 am »
I have just unboxed a Rigol 1052E and proceeded to do the thing you do when you get a new scope, put a probe on the square wave calibrator. As I set it to auto everything went fine at 500mV/div - perfect square wave. As I proceeded to bump up the gain to 200mV/div to look at the noise floor it introduced a nasty 1 div tilt (200mV) to the bottom of the square wave that seemed to change as I positioned the trace up or down.

I know the trace now goes off scale, but is this typical of these scopes?

F/W is 2.05 SP2.
Things checked:
- Probe is on 1x.
- Tried both probes.
- Recalibrated.

Any help appreciated.
 

Offline saturation

  • Super Contributor
  • ***
  • Posts: 4788
  • Country: us
  • Doveryai, no proveryai
    • NIST
Re: Rigol DS1052E (an)other nasty surprise?
« Reply #1 on: July 06, 2011, 11:12:25 am »
Post a picture of the problem, please.
Best Wishes,

 Saturation
 

Offline goober

  • Newbie
  • Posts: 3
  • Country: au
Re: Rigol DS1052E (an)other nasty surprise?
« Reply #2 on: July 07, 2011, 07:58:18 am »
Hi, as requested here are the pics.
First, at 500mV/div all looks fine.
Next, "Click", (literally, as the attentuator clicks to another scale) to 200Mv/div.
Is the input stage being overloaded? Are we at the limits of the A/D? What gives?
 

Online Mechatrommer

  • Super Contributor
  • ***
  • Posts: 9206
  • Country: my
  • reassessing directives...
Re: Rigol DS1052E (an)other nasty surprise?
« Reply #3 on: July 07, 2011, 09:53:08 am »
i confirmed this on my scope and even worst.
pls note my rigol is older batch with fw of 2.02
« Last Edit: July 07, 2011, 09:55:08 am by Mechatrommer »
if something can select, how cant it be intelligent? if something is intelligent, how cant it exist?
 

Offline marmad

  • Super Contributor
  • ***
  • Posts: 2979
  • Country: aq
    • DaysAlive
Re: Rigol DS1052E (an)other nasty surprise?
« Reply #4 on: July 07, 2011, 12:03:39 pm »
The Rigol Quick Start says probes should be calibrated at 10x (as in attached .png)... so I assume it's an overload problem.
 

Offline saturation

  • Super Contributor
  • ***
  • Posts: 4788
  • Country: us
  • Doveryai, no proveryai
    • NIST
Re: Rigol DS1052E (an)other nasty surprise?
« Reply #5 on: July 07, 2011, 12:29:02 pm »
There is no compensation on the x1 setting of scope probes and the 'surprise' appears to be just poorly tuned probes.
« Last Edit: July 08, 2011, 11:23:42 am by saturation »
Best Wishes,

 Saturation
 

Offline Semantics

  • Contributor
  • Posts: 48
  • Country: us
Re: Rigol DS1052E (an)other nasty surprise?
« Reply #6 on: July 07, 2011, 12:47:02 pm »
While, sure, you calibrate on 10X, it sure looks like the device is showing a waveform that's not true.

I don't have a 1052, but I do have an 1102 (before the 100MHz hack was widespread knowledge) and it's happening pretty strongly with mine, Firmware 2.02.SP2

Here's something odd, too, if I push STOP and use the vertical adjustment, the waveform moves as expected, but as soon as I hit RUN, the falling edge continues to go below the screen while the next pulse's rising edge seems to have gone up according to the offset / 3 (see the attached params, shifted up 200mV on the display but the params show position at 600mV). Waveform looks and behaves fine at 500mV/div

 

Offline saturation

  • Super Contributor
  • ***
  • Posts: 4788
  • Country: us
  • Doveryai, no proveryai
    • NIST
Re: Rigol DS1052E (an)other nasty surprise?
« Reply #7 on: July 07, 2011, 01:34:21 pm »
What appears to be 'compensated' after coarse adjustment at 500mV show its faults at a higher resolution; if the sync is still stable, you can continue adjusting the compensation from there too.

If you hit stop, you are looking at an image in memory.  Run is real time.  Since the calibration waveform is fixed at ? 1Vpp, I forget,  if you increase the V/div the image is off the screen, which is what the photos show.


While, sure, you calibrate on 10X, it sure looks like the device is showing a waveform that's not true.

I don't have a 1052, but I do have an 1102 (before the 100MHz hack was widespread knowledge) and it's happening pretty strongly with mine, Firmware 2.02.SP2

Here's something odd, too, if I push STOP and use the vertical adjustment, the waveform moves as expected, but as soon as I hit RUN, the falling edge continues to go below the screen while the next pulse's rising edge seems to have gone up according to the offset / 3 (see the attached params, shifted up 200mV on the display but the params show position at 600mV). Waveform looks and behaves fine at 500mV/div


Best Wishes,

 Saturation
 

Online Mechatrommer

  • Super Contributor
  • ***
  • Posts: 9206
  • Country: my
  • reassessing directives...
Re: Rigol DS1052E (an)other nasty surprise?
« Reply #8 on: July 07, 2011, 01:43:21 pm »
There is no compensation on the x1 setting of scope probes and the 'surprise' appears to be just poorly tuned probes.
i dont think so. 1X at 500mV/div is doing fine. when we step 1 down to 200mV/div this thing suddenly happen (with relay click inside). i think some problem internal of scope at that or lower 200mV/div setting.

further investigation:
1) to be more accurate, this thing happened at 245mV/div (with fine vertical tuning and relay clicked), 250mV/div is doing total 100% fine.
2) at 250mV/div or greater, we can offset vertically to |30V| and greater, but at 245mV/div and lower, we only can do vertical offset max to |2V|, strange.
3) recompensating (turn the probe screw) do not change anything (marmad, please explain further?)
4) using home-made bnc cable, result is the same as using rigol probe.

conclusion: probe fault is ruled out. its internal (hw or fw) of scope fault!
« Last Edit: July 07, 2011, 02:02:35 pm by Mechatrommer »
if something can select, how cant it be intelligent? if something is intelligent, how cant it exist?
 

Offline ejeffrey

  • Super Contributor
  • ***
  • Posts: 1989
  • Country: us
Re: Rigol DS1052E (an)other nasty surprise?
« Reply #9 on: July 07, 2011, 03:35:07 pm »
It certainly sounds like an overloading input amplifier to me.  Nothing to do with the probe should depend on the internal attenuator setting.
 

Offline marmad

  • Super Contributor
  • ***
  • Posts: 2979
  • Country: aq
    • DaysAlive
Re: Rigol DS1052E (an)other nasty surprise?
« Reply #10 on: July 07, 2011, 05:01:12 pm »
Quote
3) recompensating (turn the probe screw) do not change anything (marmad, please explain further?)

As saturation said, the compensation cap only effects the 10x probe setting - no compensation on the 1x.

The probe compensator output voltage is 3Vpp.  The offset range of the scope (2mV to 100 250mV) is 4Vpp - when you try to measure at 1x with the vertical position too offset, you overshoot the limits of the scope and have distortion.  You have the same problem if you use the 10x setting at the 20mV range.

Here is the same signal again (the compensator output) measured at 1x with the offset adjusted lower (so as not to overshoot).  The probe compensator output on my scope measures at +3.1Vpp - and the distortion vanishes when I have my vertical offset adjusted to about -800mv - giving me ~3.2V of headroom.

Quote
2) at 250mV/div or greater, we can offset vertically to |30V| and greater, but at 245mV/div and lower, we only can do vertical offset max to |2V|, strange.

Yes, Mechatrommer, the manual says "Offset Range ±40V(200mV-10V), ±2V(2mV-100mV)" - but in fact, the relay that changes the range switches when you go from 500mV to 200mV (on 1X - or 5V to 2V on 10X) - not 200mV to 100mV.  Can someone with an older model confirm that this is the same for theirs as well?

Last edit: Changed line to to reflect Mechatrommer's find.
« Last Edit: July 07, 2011, 05:40:30 pm by marmad »
 

Offline A Hellene

  • Frequent Contributor
  • **
  • Posts: 601
  • Country: gr
Re: Rigol DS1052E (an)other nasty surprise?
« Reply #11 on: July 07, 2011, 07:56:21 pm »
It certainly sounds like an overloading input amplifier to me. Nothing to do with the probe should depend on the internal attenuator setting.
Exactly!

In the schematics below, it can be seen the trimmer capacitor for compensation of the input attenuator, which is engaged only when the range is set between 500mV/Div to 10V/Div. Of course, the trimmer cannot be accurately calibrated using the oscilloscope's calibration output as a reference. By the way, the calibration output comes directly from an FPGA output through a voltage divider (3.37V_FPGA_IO buffered by 1K50 and shunted to GND via 15K0) and a weak low pass filter to smooth the edges.



These are the schematics (drafts) of the Channel 1 front-end circuitry existing within the shielding that I have drawn to troubleshoot my unit.

Please, note that the values of the chip capacitors are estimated, since they have no markings and were measured while on board.


-George



[EDIT]: Schematics updated.
« Last Edit: July 09, 2011, 12:11:09 am by A Hellene »
Hi! This is George; and I am three and a half years old!
(This was one of my latest realisations, now in my early fifties!...)
 

Offline marmad

  • Super Contributor
  • ***
  • Posts: 2979
  • Country: aq
    • DaysAlive
Re: Rigol DS1052E (an)other nasty surprise?
« Reply #12 on: July 07, 2011, 08:00:41 pm »
Quote
which is engaged only when the range is set between 500mV/Div to 10V/Div.

Actually, on my scope (and Mechatrommer's too it appears), the range seems to be between 200mV to 10V (250mV when using the fine adjust) - at least that's when the offset changes from 4V to 40V (or vice-versa).  Yep.... the manual is wrong stating 200mV to 10V.
« Last Edit: July 08, 2011, 11:45:29 am by marmad »
 

Offline A Hellene

  • Frequent Contributor
  • **
  • Posts: 601
  • Country: gr
Re: Rigol DS1052E (an)other nasty surprise?
« Reply #13 on: July 07, 2011, 08:18:04 pm »
I guess I could be wrong but this is what I have registered during my analysis, since figuring out of the scope is a work still in progress; and the time I have in my hands is very limited these days...

Thank you for pointing out this possible mistake I may have done.
I will have to put my dismantled scope together for confirmation and I will came back to that.


-George
Hi! This is George; and I am three and a half years old!
(This was one of my latest realisations, now in my early fifties!...)
 

Offline marmad

  • Super Contributor
  • ***
  • Posts: 2979
  • Country: aq
    • DaysAlive
Re: Rigol DS1052E (an)other nasty surprise?
« Reply #14 on: July 07, 2011, 08:38:40 pm »
George,

What's strange is that the manual for the DS1052E reflects what you reported - but on my scope I can hear the relay that switches between the 4V and 40V ranges thrown exactly when I cross 250mV using the fine adjust (as Mechatrommer reported).   See correction above.
 
Mark
« Last Edit: July 08, 2011, 11:47:34 am by marmad »
 

Offline A Hellene

  • Frequent Contributor
  • **
  • Posts: 601
  • Country: gr
Re: Rigol DS1052E (an)other nasty surprise?
« Reply #15 on: July 07, 2011, 09:13:30 pm »
Mark,

It seems that the DS1000E series instruments are smarter than what we thought they were, since it did not occur to me to test the attenuator engagement (the Relay#1x clicking) using the range fine-adjust feature.

For what it's worth, this is the truth table of the attenuator engagement relay(s):
Code: [Select]
---------------------------------------------------------
| Range: |   1      2     3     4     5     6      7    |
|--------|----------------------------------------------|
|  High  |   -      -    0.5V  1.0V  2.0V  5.0V    10V  |
|  Low   | 2.0mV  5.0mV  10mV  20mV  50mV  100mV  200mV |
---------------------------------------------------------

The truth table above also confirms the calculated attenuation ratio of 1:45 mentioned in the previous message, since the fifth range needs 1:40 attenuation and all the other ones 1:50; fine tuning can be done by trimming the VGA gain, in software or using both. So, the calculated attenuation ratio of the schematics posted must not be so wrong, even if its value seems to be odd.


-George
« Last Edit: July 07, 2011, 09:16:17 pm by A Hellene »
Hi! This is George; and I am three and a half years old!
(This was one of my latest realisations, now in my early fifties!...)
 

Online Mechatrommer

  • Super Contributor
  • ***
  • Posts: 9206
  • Country: my
  • reassessing directives...
Re: Rigol DS1052E (an)other nasty surprise?
« Reply #16 on: July 08, 2011, 03:05:57 am »
thanx George for providing the schematic. @marmad the relay click range (high side?) is 500mV - 10V on coarse adj. (250mV - 10V on fine), not 200mV - 10V as bolded in your quote below.

Quote
which is engaged only when the range is set between 500mV/Div to 10V/Div.
Actually, on my scope (and Mechatrommer's too it appears), the range seems to be between 200mV to 10V (250mV when using the fine adjust) - at least that's when the offset changes from 4V to 40V (or vice-versa).
if something can select, how cant it be intelligent? if something is intelligent, how cant it exist?
 

Offline A Hellene

  • Frequent Contributor
  • **
  • Posts: 601
  • Country: gr
Re: Rigol DS1052E (an)other nasty surprise?
« Reply #17 on: July 08, 2011, 03:42:52 am »
You are welcome, Mechatrommer.

As I wrote at tinhead's thread on the Tekway/Hantek hack, I will publish the schematics when they will become decently error-free.


-George
Hi! This is George; and I am three and a half years old!
(This was one of my latest realisations, now in my early fifties!...)
 

Offline saturation

  • Super Contributor
  • ***
  • Posts: 4788
  • Country: us
  • Doveryai, no proveryai
    • NIST
Re: Rigol DS1052E (an)other nasty surprise?
« Reply #18 on: July 08, 2011, 11:12:28 am »
I stand corrected folks, I finally had a chance to test the scope, firmware 2.04, unmodded 50 MHz.  Hellene's detailed schematics are invaluable, mecha's and marmads test made it very clear, so I am in error. 

It certainly sounds like an overloading input amplifier to me.  Nothing to do with the probe should depend on the internal attenuator setting.
« Last Edit: July 08, 2011, 11:27:47 am by saturation »
Best Wishes,

 Saturation
 

Offline marmad

  • Super Contributor
  • ***
  • Posts: 2979
  • Country: aq
    • DaysAlive
Re: Rigol DS1052E (an)other nasty surprise?
« Reply #19 on: July 08, 2011, 11:43:30 am »
Quote
In the schematics below, it can be seen the trimmer capacitor for compensation of the input attenuator, which is engaged only when the range is set between 500mV/Div to 10V/Div.

Quote
The truth table above also confirms the calculated attenuation ratio of 1:45 mentioned in the previous message, since the fifth range needs 1:40 attenuation and all the other ones 1:50; fine tuning can be done by trimming the VGA gain, in software or using both. So, the calculated attenuation ratio of the schematics posted must not be so wrong, even if its value seems to be odd.

George,

Yes, so sorry... late bleary-eyed night last night... somehow engaged my brain into a reversed settings range  ;)  Your reported findings were exactly correct from the beginning and completely consistent with ours... it's the DS1052E manual that is incorrectly stating that the range is set between 200mV/Div to 10V/Div.  It seems they've rounded down from the actual 250mV [fine adjust] for the specs - unless that applies to the DS1152E - but I was under the impression that they were electronically identical inside.

And let me add my thanks as well for the invaluable schematics... those will be extremely useful.

Mark
 

Offline A Hellene

  • Frequent Contributor
  • **
  • Posts: 601
  • Country: gr
Re: Rigol DS1052E (an)other nasty surprise?
« Reply #20 on: July 08, 2011, 11:52:56 am »
saturation,

This overloading malfunction might also shed some light to the odd observation in my problematic unit, according to which the x100 probes were producing a rectangular square waveform reading the 1.00KHz probe calibration output but a slightly trapezoidal waveform reading an 1.038MHz perfect square signal:


8: Ch1 x100 waveform using ground lead, unfiltered, posted here, and:


8. Ch2 x100 waveform using ground lead, unfiltered, posted here.

Please, note that the x100 probe was properly compensated using the 1.00KHz probe compensator connector.
_____


Mark,

It is alright, do not worry!
And, thank you!
_____


-George
« Last Edit: July 08, 2011, 11:56:06 am by A Hellene »
Hi! This is George; and I am three and a half years old!
(This was one of my latest realisations, now in my early fifties!...)
 

Offline saturation

  • Super Contributor
  • ***
  • Posts: 4788
  • Country: us
  • Doveryai, no proveryai
    • NIST
Re: Rigol DS1052E (an)other nasty surprise?
« Reply #21 on: July 10, 2011, 12:13:05 pm »
Hello Hellene and others,

Its very subtle on your photos.  Its interesting that the 1052e has been discussed on the forum for ~ 2 years, and its impressive what blemishes people find, the good news is the more we discover, good and bad, the more we can act on it when viewing strange output on waveforms.

I don't thing the other cheapo scopes that look like it and compete in its market, such as Atten, or Owon, get as much investigation.
Best Wishes,

 Saturation
 

Offline A Hellene

  • Frequent Contributor
  • **
  • Posts: 601
  • Country: gr
Re: Rigol DS1052E (an)other nasty surprise?
« Reply #22 on: July 10, 2011, 02:10:38 pm »
Ah, I see, dear saturation!

Well, look, if this specific device was the first one I laid my hands on I would think that this would be the average quality of those instruments and --probably-- I would not even bother complaining about it. BUT this is the second DS1052E and the first one was perfect in comparison. If you recall my very first message, where I posted the 2.05 SP2 hack, I was happy with the instrument, even though I called it a "quite noisy oscilloscope."

The truth is that I was able to read almost perfect waveforms of a few millivolts p-p amplitude on shunt resistors of a switching power supply I was working on; and I was pretty happy with it. Now, I only have to touch the ground clip to the probe tip to literally fill the display with a ~100MHz noise garbage that prevents me from reading even the 3.0V calibration output...

What can I say...


-George
« Last Edit: July 10, 2011, 02:16:09 pm by A Hellene »
Hi! This is George; and I am three and a half years old!
(This was one of my latest realisations, now in my early fifties!...)
 

Offline goober

  • Newbie
  • Posts: 3
  • Country: au
Re: Rigol DS1052E (an)other nasty surprise?
« Reply #23 on: July 12, 2011, 09:31:25 am »
Well, here are some more pics for the forum to digest.
I guess 2 pictures are worth 2 KWords ...

Top image 1x probe setting
Bottom image 10x probe setting

In short - don't ever use 1x if you want to trust your reading at moderately high frequency bursts.
C'mon Rigol what gives? - I've got a PIC with a 10bit ADC that can do better. Is this a dud unit?
These are only 400Khz pulse trains, this is not an overload problem  :-\
I really don't expect a lot for the price I paid, but really, a decent response at 400Khz - is that too much to ask?  :o

Source : Couple of address lines from serial ram, burst of pulses approx 200mS apart. Same response with a plain short bit of cat 5 and BNC adaptor.
« Last Edit: July 12, 2011, 09:40:42 am by goober »
 

Offline saturation

  • Super Contributor
  • ***
  • Posts: 4788
  • Country: us
  • Doveryai, no proveryai
    • NIST
Re: Rigol DS1052E (an)other nasty surprise?
« Reply #24 on: July 12, 2011, 10:38:07 am »
Most x1 ratings on probes are rated to ~ 6MHz, x10 is rated to 100-220 MHz, depending on the model supplied.
The edge of the pulse trains represent the high harmonics of the fundamental, about x9, or 3.6 MHz,about were a roll off should begin.
Best Wishes,

 Saturation
 


Share me

Digg  Facebook  SlashDot  Delicious  Technorati  Twitter  Google  Yahoo
Smf