Author Topic: Rigol DS2072 vertical problem  (Read 1804 times)

0 Members and 1 Guest are viewing this topic.

Offline battlecoderTopic starter

  • Regular Contributor
  • *
  • Posts: 159
  • Country: cl
    • A BIT of Mystery
Rigol DS2072 vertical problem
« on: November 10, 2023, 03:53:27 am »
Hello all,
I bought a DS2072 oscilloscope in a "bundle" of assorted instruments from a fellow enthusiast. Everything was in working condition except a Rigol DS2072 scope that is having some issues.
The scope seems to be working mostly fine in almost every way, and the horizontal axis doesn't seem to present any problem. The vertical system, however, is where it's failing.


Both CH1 and CH2 have a "voltage" offset (this is both channels with no signal applied):
1923777-0

The offset changes with the vertical sensitivity setting, and with the position of the traces on the screen. The offset is exactly the same in AC and DC coupling modes.
If I move the traces to the center of the screen, however, the offset disappears. This happens in both channels:
1923783-1


The traces follow the screen position "correctly" when GND-coupled. They are drawn exactly at the trace level:
1923789-2

The vertical sensitivity is also off. This is supposed to be a 1Vpp wave, at 100mV/DIV (should be displayed with clipping):
1923795-3

The same 1Vpp wave, but at 500mV/DIV (should cover 2 divisions):
1923801-4


More details: I updated the firmware to the last available version (00.03.06) in case this was a software issue. Hardware version is 2.0. That did not help.
I have performed the self-calibration several times. With and without warm-up, and before and after the firmware update.
I also noticed that the scope had every "option" unlocked. In the System Info panel, for "model" it says "2202", and I assume this is due to the fact that it has the 200Mhz option too.
I don't need this scope to be 200Mhz, so if this, or any of the unlocked features is causing calibration issues I will more than happy to undo that, assuming it's possible. I have no idea what method was followed to unlock those options.

I also disassembled the scope to see if there any blown up component in either channels or the surrounding circuitry, but nothing seems burnt or damaged, "visually".

Any help would be greatly appreciated.
« Last Edit: November 10, 2023, 05:54:49 pm by battlecoder »
 

Offline ebastler

  • Super Contributor
  • ***
  • Posts: 7375
  • Country: de
Re: Rigol DS2072 vertical problem
« Reply #1 on: November 10, 2023, 06:17:31 am »
Have you tried a "factory reset"?
https://gotroot.ca/rigol/ds2000_factory_reset.pdf
 
The following users thanked this post: battlecoder

Offline battlecoderTopic starter

  • Regular Contributor
  • *
  • Posts: 159
  • Country: cl
    • A BIT of Mystery
Re: Rigol DS2072 vertical problem
« Reply #2 on: November 10, 2023, 07:23:47 am »
Have you tried a "factory reset"?
https://gotroot.ca/rigol/ds2000_factory_reset.pdf
Interesting. Just tried it. It beeped several times, and it definitely started in a different config than I left it on, but that unfortunately didn't fix the issue.
Thanks for the suggestion, though.
 

Offline BlownUpCapacitor

  • Frequent Contributor
  • **
  • Posts: 342
  • Country: us
  • I kinda suck at the art of electronics tbh
Re: Rigol DS2072 vertical problem
« Reply #3 on: November 11, 2023, 07:36:23 am »
Capacitors. I had a similar problem in my Hantek and it just came down to some faulty capacitors filtering the supply voltage for the ADCs. The capacitors weren't allowing the proper voltage to power the ADCs. I knew this because I had a good Hantek scope that had different voltages. Replaced some caps and all was good after a self-cal. The capacitors had become conductive, which I found quite strange as they were MLCC types. But anything can happen with cheap parts I guess. I also found it odd that the PSU didn't trip.

However, this might not be the case for you but I think it's worth a try to go poking around with a probe to check the supply voltages and some capacitors.
Hehe, spooked my friends with an exploding electrolytic capacitor the other day 😁.
 
The following users thanked this post: battlecoder

Offline ebastler

  • Super Contributor
  • ***
  • Posts: 7375
  • Country: de
Re: Rigol DS2072 vertical problem
« Reply #4 on: November 11, 2023, 07:45:58 am »
Capacitors.

Your username suggests some perception bias, but yes, that is a plausible cause.  ;)

Without a second scope to compare with, it might be difficult to figure out which voltages are correct though. There is a "service guide" from Rigol, but it does not go beyond disassembly instructions: https://www.manualslib.com/manual/1543054/Rigol-Ds2000-Series.html. So, as you suggested, probing around and looking for implausible voltages or capacitance values is probably the only way to go.
 
The following users thanked this post: battlecoder

Offline battlecoderTopic starter

  • Regular Contributor
  • *
  • Posts: 159
  • Country: cl
    • A BIT of Mystery
Re: Rigol DS2072 vertical problem
« Reply #5 on: November 11, 2023, 04:48:59 pm »
Capacitors. I had a similar problem in my Hantek and it just came down to some faulty capacitors filtering the supply voltage for the ADCs. The capacitors weren't allowing the proper voltage to power the ADCs. I knew this because I had a good Hantek scope that had different voltages. Replaced some caps and all was good after a self-cal. The capacitors had become conductive, which I found quite strange as they were MLCC types. But anything can happen with cheap parts I guess. I also found it odd that the PSU didn't trip.

However, this might not be the case for you but I think it's worth a try to go poking around with a probe to check the supply voltages and some capacitors.
Thank you. Will definitely try this as soon as I'm back to my lab. I have some LCR tweezers that I've successfully used in the past to measure caps in circuit that might come in handy to detect faulty capacitors, but without an schematic finding the right value for them might be a problem.


Capacitors.

Your username suggests some perception bias, but yes, that is a plausible cause.  ;)

Without a second scope to compare with, it might be difficult to figure out which voltages are correct though. There is a "service guide" from Rigol, but it does not go beyond disassembly instructions: https://www.manualslib.com/manual/1543054/Rigol-Ds2000-Series.html. So, as you suggested, probing around and looking for implausible voltages or capacitance values is probably the only way to go.
Yeah, I would truly appreciate if they had more info about the internals, but I guess manufacturers no longer offer schematics or detailed service guides.
« Last Edit: November 11, 2023, 04:51:04 pm by battlecoder »
 

Offline tautech

  • Super Contributor
  • ***
  • Posts: 29810
  • Country: nz
  • Taupaki Technologies Ltd. Siglent Distributor NZ.
    • Taupaki Technologies Ltd.
Re: Rigol DS2072 vertical problem
« Reply #6 on: November 11, 2023, 04:52:45 pm »
There is much investigated about DS2000 issues in this old thread by member Bud:
https://www.eevblog.com/forum/projects/project-yaigol-fixing-rigol-scope-design-problems/
Avid Rabid Hobbyist.
Some stuff seen @ Siglent HQ cannot be shared.
 
The following users thanked this post: battlecoder

Offline battlecoderTopic starter

  • Regular Contributor
  • *
  • Posts: 159
  • Country: cl
    • A BIT of Mystery
Re: Rigol DS2072 vertical problem
« Reply #7 on: November 11, 2023, 05:12:45 pm »
There is much investigated about DS2000 issues in this old thread by member Bud:
https://www.eevblog.com/forum/projects/project-yaigol-fixing-rigol-scope-design-problems/
Thanks! That thread seems to cover the 2000 series!. It even includes some analysis of the frontend. This could definitely come in handy. Thank you SO much.
 

Offline battlecoderTopic starter

  • Regular Contributor
  • *
  • Posts: 159
  • Country: cl
    • A BIT of Mystery
Re: Rigol DS2072 vertical problem
« Reply #8 on: November 12, 2023, 03:41:46 am »
Oh, another thing I noticed, is that the offset applies to both channels equally. I know that in the picture it looks to be a different offset for CH1 and CH2, but that's because they are at different heights on the screen (is that a  hardware thing ? Does moving the trace up and down inject a DC offset on the ADC? or is it just "drawing" position? I suspect that it has to be setting an offset on the circuitry somehow... considering that moving a trace up and down also affect ADC clipping, so the trace height on the screen has to be setting a "middle" voltage for the ADC or something).

If I set both channels to the same sensitivity, and move both traces to the same height, they will have the same offset. Even if I only enable one channel, and I set it to a specific height, it will exhibit the same offset as the other channel if set at the same height on the screen.

This leads me to believe that it's either something that's affecting the circuitry of both channels at the same time (a bad reference voltage for example) OR  is affecting a sampling stage AFTER both channels have produced their signals. Considering that the vertical offset is set separately for each channels (but shows the same behavior), I'm inclined to believe it's more of something that is affecting both channels equally, not something that happens afterwards.
 

Offline tautech

  • Super Contributor
  • ***
  • Posts: 29810
  • Country: nz
  • Taupaki Technologies Ltd. Siglent Distributor NZ.
    • Taupaki Technologies Ltd.
Re: Rigol DS2072 vertical problem
« Reply #9 on: November 12, 2023, 04:16:31 am »
Don't discount someone hasn't hurt both inputs = oh that doesn't look right, let's look at it on the other channel = both channels zapped.  :bullshit:
Avid Rabid Hobbyist.
Some stuff seen @ Siglent HQ cannot be shared.
 

Offline battlecoderTopic starter

  • Regular Contributor
  • *
  • Posts: 159
  • Country: cl
    • A BIT of Mystery
Re: Rigol DS2072 vertical problem
« Reply #10 on: November 12, 2023, 04:36:41 am »
Don't discount someone hasn't hurt both inputs = oh that doesn't look right, let's look at it on the other channel = both channels zapped.  :bullshit:
That's...true  :-\ Hopefully it's not the case. I'm at my lab now and will see if I can find faulty caps or suspicious voltages. With nothing to compare against it will be hard, but since the other thread contains the part numbers for several components I can at least check if the voltages make sense.

As much as I love tech equipment I've mostly repaired old stuff (for which schematics are available) and simple supply issues (where the device isn't even powering ON, and the supply is not delivering power), so this kind of issue is a first for me. Will take me a while to fix it, IF I'm ever able to.
 

Offline battlecoderTopic starter

  • Regular Contributor
  • *
  • Posts: 159
  • Country: cl
    • A BIT of Mystery
Re: Rigol DS2072 vertical problem
« Reply #11 on: November 12, 2023, 06:01:08 am »
Alright, so I couldn't check much today because it's suddenly very late, but I checked some voltage regulators right outside the CH1/CH2 "cans", next to Rigol's branded chip (FPGA? custom IC?), and they were reading 3.26V instead of 3.3V. A similar voltage regulator in a different area of the board was reading 3.3V spot on, so I'm not sure if this is an actual problem or if that's within reasonable accuracy.

Anyway, I did manage to find two test points on the board (right outside the cans, near the corner of the scope) where the vertical offset of the traces on screen manifest as a voltage:


With the trace at the center of the screen my multimeter was reading approximately -635mV, with the trace at the bottom -550mV, and with the trace at the top around -750mV. All "negative" (relative to global GND, but maybe I should test against the VGND_1 pin that's in the corner). I'm not an expert but I would have expected a swing that has 0V at either extreme or at the center. Not sure if this is where the problem lies, but this area is where I'll probably be poking around a bit more tomorrow.

EDIT: Additionally, the testpoint labeled "OS_2.5V" does indeed read 2.5V on my multimeter.
« Last Edit: November 12, 2023, 06:02:42 am by battlecoder »
 

Offline battlecoderTopic starter

  • Regular Contributor
  • *
  • Posts: 159
  • Country: cl
    • A BIT of Mystery
Re: Rigol DS2072 vertical problem
« Reply #12 on: December 18, 2023, 02:35:01 pm »
I know this is a bit of a long shot, but anyone here owns this oscilloscope and is willing to give me some reference values for the test points?

OR, alternative, is there a chance that this is just a matter of some missing calibration values and it's just software and not hardware?
 

Offline TurboTom

  • Super Contributor
  • ***
  • Posts: 1471
  • Country: de
Re: Rigol DS2072 vertical problem
« Reply #13 on: December 18, 2023, 03:51:59 pm »
The OS_CH1 and OS_CH2 are the offset (bias) voltages for the two channels. The circuitry is a smaple-hold with a single channel 16 bit DAC (Texas DAC8560) and an analog multiplexer (HC4051) to distribute analog signals to several hold circuits consisting of the two quad CMOS OPAMPs TLC274 and the RC "chickenfood" around. This provides all the analog control voltages that are required in different places in the trigger / analog input circuitry.

I've still got an MSO 2000A that's currently not in use --  I can take some measurements if you like, but the festive season will still keep me busy for a few days, so if you're not in a hurry, I'll be prepared to tear mine down to do some comparisons...
 
The following users thanked this post: egonotto, battlecoder

Offline battlecoderTopic starter

  • Regular Contributor
  • *
  • Posts: 159
  • Country: cl
    • A BIT of Mystery
Re: Rigol DS2072 vertical problem
« Reply #14 on: December 18, 2023, 09:30:41 pm »
I would really appreciate that. Thanks for the insights on the overall circuit structure btw, I'd be a pity to throw this scope away or sell it "for repairs", when the issue could be (potentially) easy to repair. Everything else works fine so it could be just a bad voltage reference (fingers crossed).
 


Share me

Digg  Facebook  SlashDot  Delicious  Technorati  Twitter  Google  Yahoo
Smf