New Rigol DP832 Problem

[belzrebuth]

Hi,
I've just received a new DP832 power supply from Batronix.
Assume the supply has been turned off with all the outputs off.
The problem is that with a multimeter connected to CH2 ,after powering on the unit and until it completes its boot sequence I have my multimeter displaying 32V on the output.
When the unit is boot up the output returns to 0V as expected.
Ch1 and 3 do not have this.
Is this "normal" somehow or I got a defective unit?
Luckilly I didn't have any circuit connected when I was initially testing it othewise it'd have blown for sure..

[belzrebuth]

Yeah, I figured.
I've contacted Batronix and I'm now waiting for their answer.
I will post the outcome of this here.

It really is fortunate that I didn't have anything connected to this.
I wonder who pays the damage if something gets destroyed by a faulty piece of test equipment.
The sole reason for anyone to buy such a piece is to somehow ensure proper powering of a DUT and here we are

[belzrebuth]

The main issue here (aside from the faulty part) is that such a behavior makes one question if there's any QC happening at all with Rigol products.

I don't have anything against Rigol (so far I didn't) but what kind of QC pass this device bears when it reaches a customer with this kind of grave issue.

We're not talking about a difficult to discover kind of glitch or whatever that's just dangerous even for the operator; I could have had a led or a capacitor connected to the PSU and my kid close to the bench.

What could have happened then?

[RoGeorge]

I guess it's not normal.

Just measured mine, and it doesn't do that.  Thought, my DP832 is not a DP832 any more since its software was tricked to think it's a DP832A now (same hardware but different software that shows a multi-colored screen and a much readable character set).

Upgrade method and firmware versions can be found in this topic
https://www.eevblog.com/forum/testgear/need-help-hacking-dp832-for-multicolour-option/msg2323122/#msg2323122

[rstofer]

DP832 is known to have that power on jolt, but 32V is way too high than others have reported, and the duration is way too long that you can snap a picture of it.

Normally you would have a 1.xV jolt lasting for a blink of an eye, and the only way to prolong it enough to photo it would be connecting it to an incandescent lamp and photo the cooldown of the filament.

Constant 32V is WRONG. You need to return it.

That's the result I get.  I suppose I could try to trigger a scope on the startup but using just a DMM, there's no reading of a startup pulse on my DP832.

If the pulse is on the order of 1V with a duration of some milliseconds, there's not a lot of energy in the pulse.  But, like I said, I don't see it.

I wonder if a firmware upgrade will solve the problem.  Given that the PS is new, maybe a firmware downgrade...

[belzrebuth]

I don't think this is a software issue since it's happening before the unit is fully booted.
Quite the opposite actually since when it's all booted up the output immediately descends to zero.

[belzrebuth]

I've got a replacement from Batronix.
Now I get 0.98 volts during power up on both channels.
This again is not an overshoot, it lasts until the entire boot procedure is performed.

Again I don't think this is normal...
I'd expect a brief overshoot at most.

Should I return that one too?

If I'm never going to get 0v during power up I might as well keep it and just be careful what I leave connected to it.
I'm really dissapointed with Rigol at this point..
Luckily the voltage is not enough to fry most things like 32V would but still, it's annoying.

[RoGeorge]

Is it just voltage, or it generates current, too?
Maybe the small voltage bump at start charges the output capacitors to 0.98V.   

Try measuring with a load connected, e.g. an auto light bulb or a resistor of a few ohms.

[belzrebuth]

No it can supply current.
I've connected a 10ohm resistor to the output and a multimeter in parallel to that and I still get 0.9V.
So it's able to deliver at least 100mA.
My guess is that I could also connect a 1ohm and get 1A

[belzrebuth]

Returning it for a second time is a great hassle for me.
I'm stunned by the lack of QC on Rigol's part.
Should I ask a partial refund or whatever?
Maybe try to repair it myself or at least find out what's going on?
I couldn't care less for warranty if a brand new unit is problematic (twice in a row).
I mean the return and replacement thing could go on forever and this could also impact the reseller if they have to pay for double shipping back and forth.
I really don't know what to write to them.
And it's certainly not a user error.
It's a really straightforward issue and 100% reproducible.

[RoGeorge]

My DP832 doesn't have one volt during boot.  There is a very small bump (with current too, not only voltage), but it's a negative voltage and it decays fast, doesn't stay during the whole boot.

What firmware version do you have?  Mine is 00.01.16.

[belzrebuth]

Same version as you 00.01.16

I've contacted Batronix for providing a solution.
I really am unlucky with this one.
From what I've seen around many people use these happily.
I got 2/2 defective

[Jeroen3]

Yeah, cheap stuff weird problems. i know.
https://www.eevblog.com/forum/testgear/dp832a-output/msg2122648/#msg2122648
https://www.eevblog.com/forum/testgear/dp832a-output/msg2125255/#msg2125255

From this topic.
https://www.eevblog.com/forum/testgear/dp832a-output/

[belzrebuth]

7/7 ?!?
I'd give up at 3 I think, really.

We're short for options though at certain price ranges so you either cough a lot more than you initially wanted or just do the return & replace game until you score.(which is sad for an instrument of its grade)

Weird thing though most scopes on the field nowadays are Rigol and never seen any complaints..
A scope should be a lot more complex than a PSU.

How come they did that right and there's a design flaw on their most sold power supply?
I can't be that unlucky.
There's definitely something going hardware-wise..

[belzrebuth]

Here's Batronix's reply for this issue:
.............................................................................

Hello,

I have talked to our contact from Rigol about it.

Following, the answer from Rigol.

> It is normal. This is caused by the electromagnetic interference signal generated by the power supply when power on.
> Because the input impedance of the oscilloscope is very high, the generated voltage is high, but the energy is very small, which will not affect the device.
>
> This applies to the entire DP800 series. There is no specification which describes or specifies output currents or output voltages during the boot process.
> It is therefore advisable to separate any external wiring from the output sockets during the boot process.

I would say that it depends on the kind of DUT if this would effect it or not but regarding to Rigol this should still be okay.

Best regards,

Jan-Steffen Schwiemann

.............................................................................

From what I gather from this message is that they want me to keep a defective supply; stating some nonsense about EMI or whatever.

Calling this "normal"..
If it really is normal what would be the correct procedure to test it and what results should I be expecting?
Constant 0.9V on the outputs with infinite current capability is okay?
Energy is low?
I've sent them a picture of a 10ohm resistor sinking 0.1mA without voltage drop..
Is this low?

[RoGeorge]

Don't expect them to investigate pictures and such.

Give them a schematic diagram of the circuit, clear steps to reproduce
1. unplug the device
2. connect a 10 ohms load to channel 2
3. ...

Say it to them again, in clear words (even if something seems trivial to you, keep explaining to them all the details) write that you are not measuring with a high impedance instrument, calculate the time/energy for them, kindly ask to forward the issue to an engineer, mention you already read the standard Q and A but you need an engineering response and not a PR/marketing response, etc.  Stay focused on the technical aspect, don't rant about it, just explain your problem and ask how to solve.

Most important, ask them to open a ticket, and ask for a tracing number, so you can follow how the issue is solved.

Alternatively, you can just ignore the issue (most probably you will not let loads connected during boot anyway, therefore the only bad situation can appear if an unintended reset happens while you are working), or just return it and buy something else.

[belzrebuth]

@RoGeorge Thanks mate.
I will try to let them at least aknowledge what's going on since this seems like a generic response from Rigol for those who contacted them about the well known power-on spike issue.

I know the PSU probably won't destroy anything since despite the current capability voltage is pretty low.
But it's still annoying to get it out of the box and be presented with such an issue, for them it's just another help ticket; for me it's earned money going into their product.
And what's more annoying is the way Rigol (and maybe Batronix also) handles it.
They essentially want me to call it normal because their "Rigol contact" says so..Lol

I sadly can't find another supply in this price range, so I will probably keep it.

[belzrebuth]

The happy side of this is that I've still got around 1.5 year of warranty on my DG1022z left, and I've been waiting for it to end in order to replace the horribly noisy fan which makes the instrument very annoying to operate.

With all that happening I've literaly lost *any* faith in Rigol so I can proceed with the fan replacement

[belzrebuth]

Okay looks like Batronix accepts the problem.
They were helpful enough to give me the options of :
a full refund (and send the unit back)
10% refund (and keep the unit)
send it to repair
or replacement with another unit that they check before sending.
I chose the later so we'll see!

I really wonder what might really be the culprit for this.
On both the first and second unit this constant voltage goes away very quickly when the machine boots.
Could this merely be a firmware issue of some sort?

[belzrebuth]

I've tried configuring the analog boards by pressing the HELP-HELP-M4-M2-M1 and HELP-HELP-M4-M2-M2 with a multimeter connected at the outputs.
I get the constant 0.98V during normal operation after doing that.
I need to enable and disable the output to bring it back to 0V.
Interestingly since I've done that CH1 has a permanent positive offset of 0.0047V with the output disabled.
I don't know what to make of this really.
Since like simply by software "poking" the analog boards the constant voltage manifests.
Does anyone know that the analog board configuration does?

[RoGeorge]

I don't know.  Recalibrate, maybe.

Search for 'DP832 calibration' and see if you can run the procedure.  There is a table with many points to correct the output at various voltage levels, and another table for the measured values displayed when a channel is on.  Look for the official docs, or read more than one example about how to do it.  I've seen a few wrong ones online, so don't rush following the first you see.  My DP832 is a little off for very small values, too, but I never did a recalibration because I never needed mV accuracy.

Okay looks like Batronix accepts the problem.
They were helpful enough to give me the options of :
a full refund (and send the unit back)
10% refund (and keep the unit)
send it to repair
or replacement with another unit that they check before sending.

That's the most outstanding attitude I ever heard from a seller.
Thumb up for Batronix!   

[bryman79]

Just discovered the same issue (32V unloaded on Channel 2 for 3 or 4 seconds after power on and capable of supplying up to 56mA under load) on my DP832.  I purchased back in early October 2020, and Tequipment.net told me over the phone that they won't return the item for an exchange.  They also said they weren't aware of the issue, which is clearly not true since they have exchanged units with the same issue (according to posts on this forum). 

I emailed Rigol and they were aware of the issue - apparently it can be fixed by a "software update" and calibration.  Given that it's at startup and 32V (not just a few mV), doesn't seem like a calibration issue to me. 

So buyer beware.  The distributors know of this long-standing problem as does Rigol.

[riccardo.pittini]

same issues:  noticed that at power on a 12V fan that was connected to ch2 went on for about 3-4sec.

Power on no load : 31V :O
Power on with 12V 2W fan: 15V ...that's a lot of energy!!!

:S I think I will send the unit back for warranty.

[riccardo.pittini]

I wanted to inform more people about the startup spike.

So I had the 30V startup spike for 3-4sec on Ch2 (it kept running a 12V fan I had and it also killed it...as the loaded voltage was about 15V :O).

I contacted the supplier (Batronix, great service). Note that my unit is already 1year old. They provided me an unofficial update (I was already running v16 and they provided me a new v16 which apparently is not officially released yet on the rigol website).

The new firmware FIXED the problem!!! I don't have anymore the 30V spike 

[nemail2]

hmm these kinds of issues is why I like relays at the output of a PSU. nothing connected ain't damaging anything...