Author Topic: DP832A output  (Read 12746 times)

0 Members and 1 Guest are viewing this topic.

Offline blueskullTopic starter

  • Supporter
  • ****
  • !
  • Posts: 367
  • Country: cn
  • BA7LKP
DP832A output
« on: January 15, 2019, 07:48:24 am »
Are there any DP832A owners out there? I'm considering a power supply, and this has had my attention.

My requirements:

1. Test devices made for 5V input (test up to 5.5V, 2A)
2. Power analog circuits (up to +-15V 0.5A)
3. Power power circuits (up to 12V 2A, 2ch, independent)
4. Current logger as a simple SMU (no 4-wire needed)

So far, the contestants are E36312A, DP832A and DP832.

E36311A and DP831/DP831A are eliminated for the lack of independent triple outputs. The high voltage outputs on those units are tracking-only.
E36313A is eliminated for the higher rated RMS noise.

So before I pull the trigger on DP832A, I need to ask the few questions:

1. What is its 5V output range? It has 5.5V OVP, so can I force it to output 5.5V? I know some BK power supplies rate lower voltage than OVP, and if I set OVP higher, their output voltage can be stretched by a bit.
2. What is its difference than its older brother besides the nice color display and hence the graphing capability? Is there anything in particular that is worth the $350 after resolution hack and new revision power FET upgrade?




Thanks

//Edit: new question added: can DP832A be hacked using the DP832 Riglol tool?
//Edit2: I see it comes with all options enabled, so no hacking needed.
« Last Edit: January 15, 2019, 08:10:53 am by blueskull »
 

Offline Jeroen3

  • Super Contributor
  • ***
  • Posts: 4077
  • Country: nl
  • Embedded Engineer
    • jeroen3.nl
Re: DP832A output
« Reply #1 on: January 15, 2019, 08:24:29 am »
E36311A and DP831/DP831A are eliminated for the lack of independent triple outputs. The high voltage outputs on those units are tracking-only.
Channel 3 shares common with the channel 2.

1. What is its 5V output range? It has 5.5V OVP, so can I force it to output 5.5V? I know some BK power supplies rate lower voltage than OVP, and if I set OVP higher, their output voltage can be stretched by a bit.
2. What is its difference than its older brother besides the nice color display and hence the graphing capability? Is there anything in particular that is worth the $350 after resolution hack and new revision power FET upgrade?
1. It does not go up to 5.5 I think. Must check.
2. Color screen

//Edit: new question added: can DP832A be hacked using the DP832 Riglol tool?
Yes. I bought the non-A and still have LAN and mV resolution.

LAN might cause the unit to spontaneously reboot, see Rigol DP832 bugs topic

Edit: Maybe look at the Rhode and Schwarz power supplies. Those are true 3 channel. For example: NGE103 or NGE103B (can't tell the difference). (€800)
« Last Edit: January 15, 2019, 09:03:35 am by Jeroen3 »
 

Online 2N3055

  • Super Contributor
  • ***
  • Posts: 6545
  • Country: hr
Re: DP832A output
« Reply #2 on: January 15, 2019, 10:17:55 am »
DP831/DP831A are eliminated for the lack of independent triple outputs. The high voltage outputs on those units are tracking-only.
Just to clarify, +/-30V channels have common ground. But they are full independent in control and settings and measurements (or you can set them to tracking too if needed).

Please note that OVP on those is software based. Meaning, readback will measure too high voltage and will send command to shut down channel. It is slow, it can take 200 msec for it to react.
For critical circuits I use external crowbar circuit..
 

Offline Gandalf_Sr

  • Super Contributor
  • ***
  • Posts: 1729
  • Country: us
Re: DP832A output
« Reply #3 on: January 15, 2019, 10:53:13 am »
I have the DP832 (not A) and I like it.  It's 0-32V 0-32V 0-5.3V for CH 1-3 respectively. Limitation is that Ch 2 and Ch 3 share the same ground.  It's a very well made PSU and can be 'upgraded' to everything the DP832A has except for the (weird) multi-color graphical display.  If you relaxed your 5.5V max to 5.3V you might be there.

[Edit Sep 2019] The DP832 can be changed to a DP832A using a 'magic' USB drive, the full details are in this thread and, as is explained there, it also applies to other Rigol equipment e.g. the DG1022Z can be converted into a DG1062Z.

The only thing I changed was to replace the fan with a quieter one - I have a full thread here on that which includes fan model number and resistor value to make the fan running detection system work properly
« Last Edit: September 23, 2019, 08:31:46 am by Gandalf_Sr »
If at first you don't succeed, get a bigger hammer
 

Offline PartialDischarge

  • Super Contributor
  • ***
  • Posts: 1611
  • Country: 00
Re: DP832A output
« Reply #4 on: January 15, 2019, 11:22:35 am »
I have a dp832 hacked to A, ch3 goes to 5.3V, overall very sturdy, but be prepared to change the fan, it's noisy.
 

Offline nctnico

  • Super Contributor
  • ***
  • Posts: 26842
  • Country: nl
    • NCT Developments
Re: DP832A output
« Reply #5 on: January 15, 2019, 11:31:02 am »
Are there any DP832A owners out there? I'm considering a power supply, and this has had my attention.

My requirements:

1. Test devices made for 5V input (test up to 5.5V, 2A)
2. Power analog circuits (up to +-15V 0.5A)
3. Power power circuits (up to 12V 2A, 2ch, independent)
4. Current logger as a simple SMU (no 4-wire needed)

So far, the contestants are E36312A, DP832A and DP832.

E36311A and DP831/DP831A are eliminated for the lack of independent triple outputs. The high voltage outputs on those units are tracking-only.
E36313A is eliminated for the higher rated RMS noise.

So before I pull the trigger on DP832A, I need to ask the few questions:

1. What is its 5V output range? It has 5.5V OVP, so can I force it to output 5.5V? I know some BK power supplies rate lower voltage than OVP, and if I set OVP higher, their output voltage can be stretched by a bit.
2. What is its difference than its older brother besides the nice color display and hence the graphing capability? Is there anything in particular that is worth the $350 after resolution hack and new revision power FET upgrade?
Get the E36312A or the E36313A. They are whisper quiet and accurate.

AFAIK the DP823 doesn't have true isolated outputs so be sure to check that.
There are small lies, big lies and then there is what is on the screen of your oscilloscope.
 

Offline tautech

  • Super Contributor
  • ***
  • Posts: 28246
  • Country: nz
  • Taupaki Technologies Ltd. Siglent Distributor NZ.
    • Taupaki Technologies Ltd.
Re: DP832A output
« Reply #6 on: January 15, 2019, 11:44:07 am »
A SPD3303X-E hacked to SPD3303X might be worth a look, whisper quiet, isolated outputs, graphing and a great SET/Load display.

5 min overview:
https://youtu.be/H-BiphMqp5g
Avid Rabid Hobbyist
Siglent Youtube channel: https://www.youtube.com/@SiglentVideo/videos
 

Offline TurboTom

  • Super Contributor
  • ***
  • Posts: 1389
  • Country: de
Re: DP832A output
« Reply #7 on: January 15, 2019, 11:53:25 am »
Vs. the SPD3303X (E) I like the option to be able to enter the voltages/currents via the keypad on the DP8xx series (despite the funny layout...). This makes dialing in parameters very quick and grants low probability for errors. A digital PSU witout a decimal keypad wouldn't be an option for me. But that's just my personal preferance (just like I won't ever get used to analog PSUs with coarse/fine adjustments, I'ld prefer one with a decent multi-turn pot anytime). YMMV
 

Offline RoGeorge

  • Super Contributor
  • ***
  • Posts: 6162
  • Country: ro
Re: DP832A output
« Reply #8 on: January 15, 2019, 11:53:51 am »
DP832 have 3 channels (I have a DP832 for about a few years)
- CH1 is a 32V/3.2A, isolated (floating), can be wired externally in series or parallel with CH2 (+/-32V or +64V) but there is NO sink current capability (all 3 channels are working in FIRST QUADRANT ONLY, positive voltage, positive current)
- CH2 is a 32V/3.2A, have an internal common ground (by design) with CH3
- CH3 is a 5.3V/3.2A, common ground with CH2, by design
- Each channel can be operated independently
- CH1 and CH2 can be set in tracking mode or independent

- can be unlocked for internal voltmeters to display mV and mA resolution (without upgrade the resolution is 10mV/10mA) and other fancy software features
- has CC/CV mode
- linear power source (no SMPS inside)
- fan is a little noisy for a quiet room at home, OK for a lab at work
- very cumbersome software menu/submenu for more complex features
- very good and very handy physical buttons, easy to control frequently tasks like setting voltage/current, etc., has digital keypad too, just that the number are placed in circles, which is funny but easy to use.  I appreciate a lot that it has distinct buttons for V, mV, A, mA (the arrows around the main knob)
- SCPI over LXI or USB
- a few digital i/o for trigger/automating testing
- loaded with SW features when unlocked

!!! It happened to me once to see it reset and reboot out of nowhere, at a total load of about only 60W, just because I was constantly reading the voltage and current by LAN, using SCPI, for logging purposes

!!! It's big and heavy (almost half a meter deep, so plan for the workbench real estate)

!!! It has some problem with the common ground between channels, a voltage drop somewhere on the ground path between CH2 and CH3 IIRC, can be a problem at high currents, can be fixed by simply putting a thicker wire inside the DP832 case, near the front connectors, never bothered me but the issue is there, some design/manufacture flaw.  Easy to fix but needs to remove the seal in order to do it.

Hard to beat DP832 at the price/features ratio.  Good quality overall (for the money), but has some childish bugs and design flaws: i.e. mine has a few 10s of mA or mV erroneous display only when close to zero, otherwise is more or less OK (but the accuracy and precision are NOT 1mV/mA, only the resolution is that good)

TL;DR
Overall:
- Great for the money in an average lab, horrible and unreliable if you look for metrology grade equipment.
- Don't know about the Voltage noise/stabilization, didn't really needed that thing at home tinkering.
- Will I buy that again?  Yes, for my needs, in a home lab, is great.  (In fact, for the next one I will look for a 4 quadrant model, in order to avoid buying an extra instrument for a controlled load, but that's just me)
« Last Edit: January 15, 2019, 12:54:52 pm by RoGeorge »
 

Offline nctnico

  • Super Contributor
  • ***
  • Posts: 26842
  • Country: nl
    • NCT Developments
Re: DP832A output
« Reply #9 on: January 15, 2019, 12:09:30 pm »
A digital PSU witout a decimal keypad wouldn't be an option for me.
I agree. I have a couple of PSUs with a keypad and it is a very good way to set a PSU to the right voltage/current quickly.
There are small lies, big lies and then there is what is on the screen of your oscilloscope.
 

Offline Wolfgang

  • Super Contributor
  • ***
  • Posts: 1774
  • Country: de
  • Its great if it finally works !
    • Electronic Projects for Fun
Re: DP832A output
« Reply #10 on: January 15, 2019, 12:34:19 pm »
Are there any DP832A owners out there? I'm considering a power supply, and this has had my attention.

My requirements:

1. Test devices made for 5V input (test up to 5.5V, 2A)
2. Power analog circuits (up to +-15V 0.5A)
3. Power power circuits (up to 12V 2A, 2ch, independent)
4. Current logger as a simple SMU (no 4-wire needed)

So far, the contestants are E36312A, DP832A and DP832.

E36311A and DP831/DP831A are eliminated for the lack of independent triple outputs. The high voltage outputs on those units are tracking-only.
E36313A is eliminated for the higher rated RMS noise.

So before I pull the trigger on DP832A, I need to ask the few questions:

1. What is its 5V output range? It has 5.5V OVP, so can I force it to output 5.5V? I know some BK power supplies rate lower voltage than OVP, and if I set OVP higher, their output voltage can be stretched by a bit.
2. What is its difference than its older brother besides the nice color display and hence the graphing capability? Is there anything in particular that is worth the $350 after resolution hack and new revision power FET upgrade?




Thanks

//Edit: new question added: can DP832A be hacked using the DP832 Riglol tool?
//Edit2: I see it comes with all options enabled, so no hacking needed.

Hi,

I have both DP832 and DP832A and I like them. Both are running trouble-free for some years now.  What I also like is stable USBTMC/SCPI programmability so they can be remote controlled and automated for tests.
 

Online 2N3055

  • Super Contributor
  • ***
  • Posts: 6545
  • Country: hr
Re: DP832A output
« Reply #11 on: January 15, 2019, 12:40:51 pm »
Well, no PSU is metrology grade equipment.  1 mV precision on 30V ch would be 33 ppm. 
There are SMU-s out there with that kind of specs, but prices are huge.

Rigol specify 0.05%+20mV for set and 0.05%+10mV for readout. My 831 was much better than that.

So to speak colloquially, if you want to use readout for measurement, you will get same results like if you were to use Fluke 87V for measurements. For which you would need 6 instruments.
If that is good enough for you it's a fine piece of equipment.

If you need better than something else must be considered. E36313A is comparable in power, fully independent channels , and has slightly better accuracy specs but not much. And 3x the price.
 

Offline RoGeorge

  • Super Contributor
  • ***
  • Posts: 6162
  • Country: ro
Re: DP832A output
« Reply #12 on: January 15, 2019, 12:47:25 pm »
- CH3 is a 6.5V/3.2A, common ground with CH2, by design
mine goes to 5.3V only

My bad, sorry.
It's max 5.3V/3.2A

Offline nctnico

  • Super Contributor
  • ***
  • Posts: 26842
  • Country: nl
    • NCT Developments
Re: DP832A output
« Reply #13 on: January 15, 2019, 01:14:07 pm »
If you need better than something else must be considered. E36313A is comparable in power, fully independent channels , and has slightly better accuracy specs but not much. And 3x the price.
And let's not forget the low current measurement range which gives uA resolution up to 20mA on the E36312A and E36313A.
There are small lies, big lies and then there is what is on the screen of your oscilloscope.
 

Online 2N3055

  • Super Contributor
  • ***
  • Posts: 6545
  • Country: hr
Re: DP832A output
« Reply #14 on: January 15, 2019, 01:16:40 pm »
If you need better than something else must be considered. E36313A is comparable in power, fully independent channels , and has slightly better accuracy specs but not much. And 3x the price.
And let's not forget the low current measurement range which gives uA resolution up to 20mA on the E36312A and E36313A.
That is correct. That is actually the only significant plus that you would see in use, if you need such measurements.
 

Offline RoGeorge

  • Super Contributor
  • ***
  • Posts: 6162
  • Country: ro
Re: DP832A output
« Reply #15 on: January 15, 2019, 02:06:16 pm »
A few major drawback to consider at DP832:
- strange things like load constantly drawing 5mA, but CH1 meter shows 15mA.  This offset happens in my DP832 only for CH1, and only for something in the range of 10mA or so.
- the internal V/A/P meter indicator is bogus at very low voltages or currents.  The A sensor for the panel meter is placed somewhere before the big filter capacitors, so you won't see the current flowing through the load, but the current flowing through the filtering caps + load, which may be very misleading in the mA/mV range.
- the protection itself is reading the current BEFORE the output caps, so if you set the CC mode at 10mA current, let's say for a LED, but forget the voltage at some 10-20V big, when you connect the LED to the wires, it will pop like popcorn, because there was enough energy (stored in the filtering caps) to make the LED explode.

It can happen to set for 20mA, but get 18A instead, about a thousand times more! :palm:
https://hackaday.io/project/7590-retardo-davinci/log/25376-rigol-dp832-power-supply-set-for-20-ma-can-kill-a-led



Fig 1: Voltage drop at connecting a 1 ohm load, DP832 in constant current (CC mode) set to 20mA ^

CC mode is bogus.  Same for very low voltage CV mode (think FPGA here), double check with another instruments.  I don't have numbers about how bad the stabilization is at very low voltages and/or currents, but my conclusion was I'd better wont use DP832 for that.

 
The following users thanked this post: plurn

Offline Jeroen3

  • Super Contributor
  • ***
  • Posts: 4077
  • Country: nl
  • Embedded Engineer
    • jeroen3.nl
Re: DP832A output
« Reply #16 on: January 15, 2019, 02:11:13 pm »
The Rigol also gives a glitch on Channel 1 when you turn it on with the mains switch.
 

Offline nctnico

  • Super Contributor
  • ***
  • Posts: 26842
  • Country: nl
    • NCT Developments
Re: DP832A output
« Reply #17 on: January 15, 2019, 02:23:31 pm »
- the internal V/A/P meter indicator is bogus at very low voltages or currents.  The A sensor for the panel meter is placed somewhere before the big filter capacitors, so you won't see the current flowing through the load, but the current flowing through the filtering caps + load, which may be very misleading in the mA/mV range.
- the protection itself is reading the current BEFORE the output caps, so if you set the CC mode at 10mA current, let's say for a LED, but forget the voltage at some 10-20V big, when you connect the LED to the wires, it will pop like popcorn, because there was enough energy (stored in the filtering caps) to make the LED explode.
This is normal for most power supplies. However the size of the output capacitor determines how badly a load can be damaged if you expect it to act as a current source.
There are small lies, big lies and then there is what is on the screen of your oscilloscope.
 

Online 2N3055

  • Super Contributor
  • ***
  • Posts: 6545
  • Country: hr
Re: DP832A output
« Reply #18 on: January 15, 2019, 02:42:46 pm »


PSU is not proper CC source.  None of them, even 1600€ Keysight mentioned. They all have large capacitor on very output (all of them , not only Rigol) to ensure stability of voltage under various complex loads.
There are some PSU topologies that are stable without large capacitor on output, and they will have less of a pulse but even them will have a discontinuity when switching from CV to CC mode.
If you want to test LEDS you need proper CC source that is constantly in CC mode, and also has max voltage limit so you don't damage LED if you reverse it and exceed max reverse voltage.
Diode testers for sensitive RF diodes actually have inputs shorted, then you connect diode, and then you remove short and ramp up voltage and current slowly to avoid any kind of transient.

In this case proper procedure to test a LED would be to set output voltage to some voltage larger than Froward voltage for that LED, but smaller than maximum reverse voltage. Then you set some current in a range you want to check. Disable output, connect LED, and enable output. PSU will start from 0V and will not produce transient.

The Rigol also gives a glitch on Channel 1 when you turn it on with the mains switch.

They all do that too. It has been discussed in detail here in forum. It was investigated and found to be induced voltage in connected grounds (scope and PSU) from switch on transient. It is not really on output, but induced in probe ground.
 

Offline Jeroen3

  • Super Contributor
  • ***
  • Posts: 4077
  • Country: nl
  • Embedded Engineer
    • jeroen3.nl
Re: DP832A output
« Reply #19 on: January 15, 2019, 03:28:03 pm »
The Rigol also gives a glitch on Channel 1 when you turn it on with the mains switch.

They all do that too. It has been discussed in detail here in forum. It was investigated and found to be induced voltage in connected grounds (scope and PSU) from switch on transient. It is not really on output, but induced in probe ground.
It is not a measurement effect. I found out with an isolated current probe and measured current when switching on.
 

Offline nctnico

  • Super Contributor
  • ***
  • Posts: 26842
  • Country: nl
    • NCT Developments
Re: DP832A output
« Reply #20 on: January 15, 2019, 03:48:22 pm »
They all do that too. It has been discussed in detail here in forum. It was investigated and found to be induced voltage in connected grounds (scope and PSU) from switch on transient. It is not really on output, but induced in probe ground.
AFAIK this was discussed for the Keysight E3631xA series. Perhaps Jeroen can show some oscilloscope pictures from his measurements. If the spikes are in the double digit us (micro second) territory (or shorter) I wouldn't worry.
There are small lies, big lies and then there is what is on the screen of your oscilloscope.
 

Online 2N3055

  • Super Contributor
  • ***
  • Posts: 6545
  • Country: hr
Re: DP832A output
« Reply #21 on: January 15, 2019, 04:27:41 pm »
The Rigol also gives a glitch on Channel 1 when you turn it on with the mains switch.

They all do that too. It has been discussed in detail here in forum. It was investigated and found to be induced voltage in connected grounds (scope and PSU) from switch on transient. It is not really on output, but induced in probe ground.
It is not a measurement effect. I found out with an isolated current probe and measured current when switching on.
They all do that too. It has been discussed in detail here in forum. It was investigated and found to be induced voltage in connected grounds (scope and PSU) from switch on transient. It is not really on output, but induced in probe ground.
AFAIK this was discussed for the Keysight E3631xA series. Perhaps Jeroen can show some oscilloscope pictures from his measurements. If the spikes are in the double digit us (micro second) territory (or shorter) I wouldn't worry.

Yes absolutely, I would like to see that and explanation of measurement setup. I would like to reproduce it if possible. It would be great to know exactly what is happening here..
 

Offline RoGeorge

  • Super Contributor
  • ***
  • Posts: 6162
  • Country: ro
Re: DP832A output
« Reply #22 on: January 15, 2019, 06:28:17 pm »
For my Rigol DP832, the current spike (at the DC banana output plugs) when the mains switch is turned on is for real.  To my surprise, the voltage is negative.  This can be clearly seen with an analog passive microampermeter (just a galvanometer, with an indicator needle and no electronics).

- The needle violently hits the out of range side of the scale when the mains switch is turned on.
- Looking at the needle, it is more than perfectly clear that there is real current flowing through the wires and coming from the power source.  It is not some induced switching noise.
- A smaller move happens when the mains is turned off, too.
- CH1's spike looks much stronger than CH2.
- CH3's spike seems to beat them all.

I didn't make any controlled tests, just connected a uA-meter.  The galvanometer is from an old Russian cassette tape recorder, or something, 100uA full scale.

While the spike is for real, it doesn't seem to have enough energy to do any damage.
 
The following users thanked this post: 2N3055

Offline Jeroen3

  • Super Contributor
  • ***
  • Posts: 4077
  • Country: nl
  • Embedded Engineer
    • jeroen3.nl
Re: DP832A output
« Reply #23 on: January 15, 2019, 06:39:00 pm »
I will redo my measurement and will report back. (24 hours)

Edit (still getting back tomorrow, but could do this now):
While the spike is for real, it doesn't seem to have enough energy to do any damage.
You'd think. Or maybe my unit is faulty. But look:
« Last Edit: January 15, 2019, 06:55:29 pm by Jeroen3 »
 
The following users thanked this post: RoGeorge

Online 2N3055

  • Super Contributor
  • ***
  • Posts: 6545
  • Country: hr
Re: DP832A output
« Reply #24 on: January 15, 2019, 06:43:28 pm »
For my Rigol DP832, the current spike (at the DC banana output plugs) when the mains switch is turned on is for real.  To my surprise, the voltage is negative.  This can be clearly seen with an analog passive microampermeter (just a galvanometer, with an indicator needle and no electronics).

- The needle violently hits the out of range side of the scale when the mains switch is turned on.
- Looking at the needle, it is more than perfectly clear that there is real current flowing through the wires and coming from the power source.  It is not some induced switching noise.
- A smaller move happens when the mains is turned off, too.
- CH1's spike looks much stronger than CH2.
- CH3's spike seems to beat them all.

I didn't make any controlled tests, just connected a uA-meter.  The galvanometer is from an old Russian cassette tape recorder, or something, 100uA full scale.

While the spike is for real, it doesn't seem to have enough energy to do any damage.

Thanks for the detail. I will check it.
 


Share me

Digg  Facebook  SlashDot  Delicious  Technorati  Twitter  Google  Yahoo
Smf