Author Topic: Rigol DP832 OVP overshooting with huge delay  (Read 1164 times)

0 Members and 1 Guest are viewing this topic.

Offline Professzore

  • Contributor
  • Posts: 8
  • Country: hu
Rigol DP832 OVP overshooting with huge delay
« on: August 23, 2017, 03:10:54 am »
Hi,
I've been into "Rigol DS832 experience" since a few days only. The most important of the list of aspects was a trusty overvoltage and overcurrent protection with a fine tuning possibility. Theoretically, DP832 knows it well.
Here I come with my very recent experience with it. Basic setup was a Rigol DS832, a Rigol MSO1074Z and a SiLabs EFM32 Pearl Gecko kit externally powered through the "button cell jumper" with 3.0 V (from this source, only the MCU works, no debugger and other features are available, overall consumption is around 5 mA or less in normal mode). Because I couldn't find any source for a trustable schematics for this evaluation board, I was trying to set-up an overvoltage protection. To test my equipment, I disconnected the Gecko and used a 1k ohm resistor to maintain an approximate 3 mA load at 3V. Only for testing, I set up 3.1 V OVP, 10 mA OCP, and turned the drive voltage to 30 V (the max) with 5 mA current. The first attachment shows the result.
The second is almost the same setting with 3A current set without OCP.
It is clearly visible that Rigol seriously overshoots the OVP setting for an unacceptable long period (500 ms). In the worst case, +27V, which is -- of course -- a killer to a board like this.
My questions:
1. Is there any software solution to "hack" this problem (in a "perfect world" if set voltage is over the limit of OVP, the output can not be turned on in any way)?
2. Is there any hint to avoid any accident regarding the above-mentioned problem?

Thanks a lot for your help,
(BTW, it's a waaaaaay better programmable PSU than my previous Axoiomet.)
Yours sincerely,
Tamás
 

Offline exe

  • Supporter
  • ****
  • Posts: 484
  • Country: nl
  • self-educated hobbist
Re: Rigol DP832 OVP overshooting with huge delay
« Reply #1 on: August 24, 2017, 06:08:35 am »
Sorry, I might not fully understand the issue, but might it happens because of large output caps? Afaik, they are 1000uF, so OCP might not be very fast. That's why I have a PSU with only 10uF caps (yet it was enough to fry an opamp's clamping diodes).

Actually, you can find some similar tests on the forum so you can compare your results with others. But it should be a pretty good device.
 

Offline IAmBack

  • Regular Contributor
  • *
  • Posts: 81
  • Country: pl
Re: Rigol DP832 OVP overshooting with huge delay
« Reply #2 on: August 24, 2017, 07:28:53 am »
OCP protection works with delay of several hundreds miliseconds (in my case it was something like 400 - http://www.eevblog.com/forum/testgear/rigol-dp832-firmware-updates-and-bug-list/msg1128301/#msg1128301 ).
According to Rigol's support - it's O'k. So, having this "feature" in mind, we're going to love our PSUs even more :)
 

Offline nctnico

  • Super Contributor
  • ***
  • Posts: 14285
  • Country: nl
    • NCT Developments
Re: Rigol DP832 OVP overshooting with huge delay
« Reply #3 on: August 24, 2017, 07:35:49 am »
I don't fully understand what you did. You set the OVP (over voltage protection) to 3.1V and the supply still puts 27V to appear on the output when set to 30V? If yes, then that is clearly a bug. OVP should prevent setting a voltage beyond the OVP limit!

As a generic sidenote: Some (older) PSUs even have a crowbar or (limited current sink) which kicks in and shorts (discharges) the output in case of an overvoltage event.
There are small lies, big lies and then there is what is on the screen of your oscilloscope.
 

Offline IAmBack

  • Regular Contributor
  • *
  • Posts: 81
  • Country: pl
Re: Rigol DP832 OVP overshooting with huge delay
« Reply #4 on: August 24, 2017, 07:49:17 am »
Just for test, I've checked OVP behavior of Elektro-Automatik EA-PS 2042-10B.
I've set OVP to 3V, output voltage to 42V, and during the test PSU "charged" it's output cap to 8V in 17.8ms, then switched off.
Better than Rigol, but "fragile" circuit would be fried anyway...
 

Offline SparkMark

  • Newbie
  • Posts: 1
  • Country: gb
Re: Rigol DP832 OVP overshooting with huge delay
« Reply #5 on: August 24, 2017, 11:38:42 pm »
Out of interest tried this out on my new E36313A (+free U1242C at the moment which sweetens the ouch very slightly).

With 3V into 1K, OVP set to 3.1, switching to 25V results in a peak voltage of 6V at ~ 5mS and back to 3V at 18mS (pic 1). Reducing the load to12R doesn't change the peak very much, but obviously decays are much faster (pic 2).

Beggars the question why the instrument allows you to set the voltage above the OVP limit, surely even if you want constant current operation, the voltage should never be allowed to go above the OVP point.
 

Offline zucca

  • Supporter
  • ****
  • Posts: 1209
  • Country: it
Re: Rigol DP832 OVP overshooting with huge delay
« Reply #6 on: August 24, 2017, 11:47:01 pm »
My personal experience: a power supply delivers just power not protection.
If you think the PS will save your board you will cook it someday and everybody will laugh because the PS is in spec.
Can't know what you don't love. St. Augustine
Can't love what you don't know. Zucca
 

Offline Ice-Tea

  • Frequent Contributor
  • **
  • Posts: 910
  • Country: be
    • Freelance Hardware Engineer
Re: Rigol DP832 OVP overshooting with huge delay
« Reply #7 on: August 24, 2017, 11:54:11 pm »
With 3V into 1K, OVP set to 3.1, switching to 25V results in a peak voltage of 6V at ~ 5mS and back to 3V at 18mS (pic 1). Reducing the load to12R doesn't change the peak very much, but obviously decays are much faster (pic 2).

And this, folks, is why you pay premium bucks for premium brands.
An engineer never has a problem. He just needs more time.

FS: TTi TSX1820P, TDS754A (w SPC fail), Agilent Infinium 54815A, 54825A, R&S CMU 200 (multiple units, various options) UPL, SMIQ03,Tektronix CSA8000B, HP 8594E, 8595E, Marconi 6201B (8GHz), IFR 2390A (22GHz), 2383 (4GHz)
 

Offline 2N3055

  • Frequent Contributor
  • **
  • Posts: 990
  • Country: hr
Re: Rigol DP832 OVP overshooting with huge delay
« Reply #8 on: August 25, 2017, 05:20:09 am »
With 3V into 1K, OVP set to 3.1, switching to 25V results in a peak voltage of 6V at ~ 5mS and back to 3V at 18mS (pic 1). Reducing the load to12R doesn't change the peak very much, but obviously decays are much faster (pic 2).

And this, folks, is why you pay premium bucks for premium brands.

They both have equally stupid software bug (omission in user interface).. The fact that expensive one reacts faster is small consolation...
Neither should allow user to set voltage to more than OPV setting....
 

Offline Ice-Tea

  • Frequent Contributor
  • **
  • Posts: 910
  • Country: be
    • Freelance Hardware Engineer
Re: Rigol DP832 OVP overshooting with huge delay
« Reply #9 on: August 25, 2017, 06:25:42 am »
Depends on what you consider "small". If one sets the board on fire and the other does not.. Or even worse, if one introduces one of those lovely "now you see me, now you don't" malfunctions that sets and engineer on a two week goose chase and the other does not...
An engineer never has a problem. He just needs more time.

FS: TTi TSX1820P, TDS754A (w SPC fail), Agilent Infinium 54815A, 54825A, R&S CMU 200 (multiple units, various options) UPL, SMIQ03,Tektronix CSA8000B, HP 8594E, 8595E, Marconi 6201B (8GHz), IFR 2390A (22GHz), 2383 (4GHz)
 

Offline IAmBack

  • Regular Contributor
  • *
  • Posts: 81
  • Country: pl
Re: Rigol DP832 OVP overshooting with huge delay
« Reply #10 on: August 25, 2017, 07:24:57 am »
Just checked my BK Precision 9185, and it definitely has software protection disabling the output when the OVP is lower than the set voltage.
But...
Output can't be enabled even thou the terminals are shorted (and PSU should work in CC mode)...
« Last Edit: August 25, 2017, 07:47:09 am by IAmBack »
 

Offline 2N3055

  • Frequent Contributor
  • **
  • Posts: 990
  • Country: hr
Re: Rigol DP832 OVP overshooting with huge delay
« Reply #11 on: August 25, 2017, 08:04:49 am »
Depends on what you consider "small". If one sets the board on fire and the other does not.. Or even worse, if one introduces one of those lovely "now you see me, now you don't" malfunctions that sets and engineer on a two week goose chase and the other does not...

I'm not saying it's not better.. But both are making stupid software mistake to allow you to actually enable the channel with obviously wrong setting..
I know OVP is not hardware crowbar protection.. It should be there to mainly prevent user errors.. It is more of user settings limiter... But it is obviously not implemented that way..


Just checked my BK Precision 9185, and it definitely has software protection disabling the output when the OVP is lower than the set voltage.
But...
Output can't be enabled even thou the terminals are shorted (and PSU should work in CC mode)...


That is actually well implemented software OVP.  You cannot trust it to be fast enough to replicate protection of real hardware crowbar... As I said, software OVP is sort of user settings limiter.. (or should be done that way to be useful).  If I know I will power something with Vmax 5.5V, I set OVP to 5.5V so I cannot absentmindedly rotate encoder or type in voltage to more than that....

If I need real hardware overvoltage protection when working on something really important, I will connect real crowbar limiter in parallel with PSU output to be sure...
 

Offline RoGeorge

  • Frequent Contributor
  • **
  • Posts: 818
  • Country: ro
Re: Rigol DP832 OVP overshooting with huge delay
« Reply #12 on: August 25, 2017, 08:31:25 am »
DP832 can only generate current, it can not sink current, so any voltage that charged the output filtering capacitors of the DP832 while there was no load attached will then burst into whatever load you decide to attach no matter what protection current you set.

What you encounter was normal behavior, and expected. I understood that only after I literally popped a LED with the current set at 20mA:
https://hackaday.io/project/7590-retardo-davinci/log/25376-rigol-dp832-power-supply-set-for-20-ma-can-kill-a-led

Offline colorado.rob

  • Regular Contributor
  • *
  • Posts: 140
  • Country: us
Re: Rigol DP832 OVP overshooting with huge delay
« Reply #13 on: August 27, 2017, 01:30:27 am »
I'm not saying it's not better.. But both are making stupid software mistake to allow you to actually enable the channel with obviously wrong setting..
I know OVP is not hardware crowbar protection.. It should be there to mainly prevent user errors.. It is more of user settings limiter... But it is obviously not implemented that way..

Just checked my BK Precision 9185, and it definitely has software protection disabling the output when the OVP is lower than the set voltage.
But...
Output can't be enabled even thou the terminals are shorted (and PSU should work in CC mode)...


That is actually well implemented software OVP.  You cannot trust it to be fast enough to replicate protection of real hardware crowbar... As I said, software OVP is sort of user settings limiter.. (or should be done that way to be useful).  If I know I will power something with Vmax 5.5V, I set OVP to 5.5V so I cannot absentmindedly rotate encoder or type in voltage to more than that....

If I need real hardware overvoltage protection when working on something really important, I will connect real crowbar limiter in parallel with PSU output to be sure...
My TTi MX100TP allows one to set the output voltage well above the OVP limit, but clamps the voltage quickly when it exceeds the OVP threshold, allowing it to go over by about 1V for only a short period of time.

The attached screenshot is with the PSU set to 3.3V OVP with the output set at 20V into 1k ohm.  With the current limited to 100mA into 10 ohms, it jumps to 2V before CC mode sets in, dropping to 1V.  Overshoot in both cases is little more than 1V.

 


Share me

Digg  Facebook  SlashDot  Delicious  Technorati  Twitter  Google  Yahoo
Smf