Author Topic: Big error in Siglent CRO  (Read 2834 times)

0 Members and 1 Guest are viewing this topic.

Offline VK3DRBTopic starter

  • Super Contributor
  • ***
  • Posts: 2252
  • Country: au
Big error in Siglent CRO
« on: March 22, 2017, 12:12:07 pm »
Hi all.

Today I saw a strange thing. I was helping a customer cure his inrush current issue. The test house measured about 100A inrush current into DC circuit from an external power supply. The external power supply would crowbar a couple of times before coming up when the highly capacitive load was connected. I reduced the inrush current at switch on to about 10A using an inush current limiter (an NTC thermistor), which worked fine.

Before the fix, my 300MHz, 2GS/s Tektronix CRO showed a peak inrush current of about 100 to 120A. The test house reported a similar inrush current. But the customer's Siglent Digital Oscilloscope Model Sds1102cml, 100mhz CRO showed a peak current of 350A. In all cases the bandwidth was not attentuated. Measurement was voltage across a 10 milliohm load ground return path, non differential as one side is earthed. So the voltage spike was 3X the other CRO.

I suspect the Siglent CRO has a problem. It seems odd that my Tektronix CRO and the NATA approved test house's CRO has similar results, but the Siglent is way, way out. I did not spend time testing the rest of his Siglent CRO, as I am just too busy.

I have seem this sort of thing before with a low end Rigol CRO where the measurements was about 20% out from that of a high quality calibrated instrument. The same Tek probe was used in the Siglent  and vice versa - the 200% error stayed with the Siglent. Happens with X1 and X10 probe settings.

Is this a case of getting what you pay for? Does anyone have any clues on why the Siglent CRO could be so out?

 

Offline Zbig

  • Frequent Contributor
  • **
  • Posts: 927
  • Country: pl
Re: Big error in Siglent CRO
« Reply #1 on: March 22, 2017, 12:48:09 pm »
CRO - Cathode Ray Oscilloscope
DSO - Digital Storage Oscilloscope
There's no such thing as "Digital CRO" The way you call modern DSOs "CROs" may confuse some newbies ;)
 

Offline PA0PBZ

  • Super Contributor
  • ***
  • Posts: 5121
  • Country: nl
Re: Big error in Siglent CRO
« Reply #2 on: March 22, 2017, 12:53:03 pm »
I happen to have a few DSCRO's  ;D
Keyboard error: Press F1 to continue.
 

Offline voltsandjolts

  • Supporter
  • ****
  • Posts: 2281
  • Country: gb
Re: Big error in Siglent CRO
« Reply #3 on: March 22, 2017, 02:48:32 pm »
I think it's very unfair to call-out any test equipment like this without doing some proper investigation yourself  :--
Surely it wouldn't take more than 30 minutes to perform a comparison test on such a simple setup.
BTW I have never used or owned any Siglent equipment.
 

Offline coppice

  • Super Contributor
  • ***
  • Posts: 8605
  • Country: gb
Re: Big error in Siglent CRO
« Reply #4 on: March 22, 2017, 04:05:50 pm »
Measurement was voltage across a 10 milliohm load ground return path, non differential as one side is earthed.
Try reading that a few times, and see if it makes any sense. It is seems sensible, then maybe you need to read it some more.
I suspect the Siglent CRO has a problem. It seems odd that my Tektronix CRO and the NATA approved test house's CRO has similar results, but the Siglent is way, way out. I did not spend time testing the rest of his Siglent CRO, as I am just too busy.
I think you need to trust instruments until you are really really sure the problem is not your fault, because most of the time it is.
 

Online tautech

  • Super Contributor
  • ***
  • Posts: 28139
  • Country: nz
  • Taupaki Technologies Ltd. Siglent Distributor NZ.
    • Taupaki Technologies Ltd.
Re: Big error in Siglent CRO
« Reply #5 on: March 22, 2017, 07:26:51 pm »
Can I ask how the measurements were taken ?
From the OSD or reading from graticules ?

Did you try using a Single shot setup in the DSO ?

 :-//

Edit
Please provide a screenshot if you can.
« Last Edit: March 22, 2017, 07:51:15 pm by tautech »
Avid Rabid Hobbyist
Siglent Youtube channel: https://www.youtube.com/@SiglentVideo/videos
 

Offline VK3DRBTopic starter

  • Super Contributor
  • ***
  • Posts: 2252
  • Country: au
Re: Big error in Siglent CRO
« Reply #6 on: March 24, 2017, 12:47:54 pm »
The CRO was at a client's site. I will ask him to give me a screen shot. I am not saying the Siglent is at fault. I just don't know what would cause it. My labour is charged out at $175 per hour. Hence spending time to investigate this might end up costing more than the CRO, and time was very tight when I was there. Hence the question here. I was sent for one task only - to investigate and reduce inrush current in a device, not to diagnose a possible issue with a client's instrument - unfortunately. The peaks where shown on the graticules, not the measurements.

By the way, one place I worked at had an early Rigol CRO where the measurements of DC voltages were about 10% out from that measured by a calibrated Fluke meter, which surprisingly also deviated from the displaced trace. In that case, the CRO's measurement (firmware maybe?) was out of whack.

Sorry about using the word "CRO" if it offends or confuses anyone. In Australia almost every electronics person uses the term CRO for an oscilloscope even if it has no cathode ray tube. typing in CRO is a lot easier than oscilloscope. I guess it is similar to the English calling any vacuum clean a Hoover. Or when we say "shipping" when a product is being delivered by road transport rather than by boat.     
 

Offline voltsandjolts

  • Supporter
  • ****
  • Posts: 2281
  • Country: gb
Re: Big error in Siglent CRO
« Reply #7 on: March 24, 2017, 12:58:53 pm »
I am not saying the Siglent is at fault.

Really?
You gave this thread the title "Big error in Siglent CRO".
 
The following users thanked this post: Zbig

Offline EEVblog

  • Administrator
  • *****
  • Posts: 37661
  • Country: au
    • EEVblog
Re: Big error in Siglent CRO
« Reply #8 on: March 24, 2017, 01:02:26 pm »
Could maybe be the pulse response of the front end amplifier. Modern digital scopes do seen to vary a bit in this regard.
Also, it could be a test lead and input capacitive loading issue maybe. Not all scopes have the same input capacitance. Are you using correctly compensated scope probes?
e.g. perfectly compensated probes on one scope could be out by a large amount if you connect those probes to another scope.
 

Offline Zbig

  • Frequent Contributor
  • **
  • Posts: 927
  • Country: pl
Re: Big error in Siglent CRO
« Reply #9 on: March 24, 2017, 01:20:28 pm »
The CRO was at a client's site. I will ask him to give me a screen shot. I am not saying the Siglent is at fault. I just don't know what would cause it. My labour is charged out at $175 per hour. Hence spending time to investigate this might end up costing more than the CRO, and time was very tight when I was there.

Yet you create a topic titled "Big error in Siglent CRO"? :-// You say you charge people $175 per hour (or your employer does, whatever). Why would you even mention that? Thinking that's of any relevance here says a lot about you, already. Anyhow, it doesn't make any statements you pull out of thin air like that an immediate truth.

Hence the question here. I was sent for one task only - to investigate and reduce inrush current in a device, not to diagnose a possible issue with a client's instrument - unfortunately. The peaks where shown on the graticules, not the measurements.

By the way, one place I worked at had an early Rigol CRO where the measurements of DC voltages were about 10% out from that measured by a calibrated Fluke meter, which surprisingly also deviated from the displaced trace. In that case, the CRO's measurement (firmware maybe?) was out of whack.
What does it have to do with the topic or your $175/hour job where you think you probably heard that some other instrument, which you don't know how to call properly, potentially showed some bad readings? Is that another "analog for the win" oldtimer rant?

Sorry about using the word "CRO" if it offends or confuses anyone. In Australia almost every electronics person uses the term CRO for an oscilloscope even if it has no cathode ray tube. typing in CRO is a lot easier than oscilloscope. I guess it is similar to the English calling any vacuum clean a Hoover. Or when we say "shipping" when a product is being delivered by road transport rather than by boat.

Most posts by Australians here I see call things by their proper names. They call the things "DSOs" or just "scopes". You can call it a "squigglescreen" for all I care - it's not me sounding silly.
« Last Edit: March 24, 2017, 01:57:49 pm by Zbig »
 

Offline coppice

  • Super Contributor
  • ***
  • Posts: 8605
  • Country: gb
Re: Big error in Siglent CRO
« Reply #10 on: March 24, 2017, 03:04:35 pm »
Could maybe be the pulse response of the front end amplifier. Modern digital scopes do seen to vary a bit in this regard.
Also, it could be a test lead and input capacitive loading issue maybe. Not all scopes have the same input capacitance. Are you using correctly compensated scope probes?
e.g. perfectly compensated probes on one scope could be out by a large amount if you connect those probes to another scope.
Yeah, those things will make some difference, but not differentially probing a 10 milliohm resistor will make the results a complete farce.
 

Offline free_electron

  • Super Contributor
  • ***
  • Posts: 8515
  • Country: us
    • SiliconValleyGarage
Re: Big error in Siglent CRO
« Reply #11 on: March 24, 2017, 03:19:52 pm »
are you sure your probe is not ringing ?
Professional Electron Wrangler.
Any comments, or points of view expressed, are my own and not endorsed , induced or compensated by my employer(s).
 


Share me

Digg  Facebook  SlashDot  Delicious  Technorati  Twitter  Google  Yahoo
Smf