Rigol DP832 measures "ghost" current

[Milentije]

Hello to everyone!

My DP832 finally arrived 3 day ago. And, immediately after opening I had a problem. The fan was dead on arrival and I couldn't turn on any of the outputs. Because I didn't want to wait for another month, or even more, for RMA I opened the unit and replaced the fan. After that the power supply allowed me to turn the outputs. NICE, I thought.

BUT, now there is a new problem.



Problem is in channel 1 and 2, channel 3 works fine.

Problem appears when there is NO load on the channels 1 and 2. As you can see from the picture that I linked above, power supply reads the current of 50mA on channel 1 and 150mA on the channel 2. Problem stays there even if I set the current limit at any level higher than 50mA for CH1 and 150mA for CH2.
If I try to set the OCP for CH1 lower than 50mA, power supply turns that channel off because of OCP.
Same is happening if I set OCP at the value lower than 150mA for CH2.

Problem disappears when there is any kind of load. Even a single green LED with 1k resistor solves the problem (@ >3V). With that load readings are as expected on both channels.

What could be a reason for this?

Thanks in advance,
M.

EDIT: I apologize to moderators for opening a new thread but I couldn't find the right one to put this in.

EDIT 2: I forgot to mention this:



Problem disapperas on the CH 1 when the voltage is set to 20V or more, but there is no change on the CH 2 even if the voltage is set to 30V.

[dadler]

I would send it back and get a new, fresh one.

[XFDDesign]

He decided to crack it open and void the warranty. He's rooted.

In all probability, from very limited information, I would guess this never go through QA/QC and Cal. The "ghost current" is likely an offset error that needs to be trimmed. Either way, it's probably a paperweight now, without more extensive work being performed.

[Skimask]

 
Or...search for that particular 'issue' here and find the solution.

[Milentije]

He decided to crack it open and void the warranty. He's rooted.

"Warranty void if broken" sticker is in one piece, and there is no evidence of opening (except this thread). So, there is no problem about that.

In all probability, from very limited information, I would guess this never go through QA/QC and Cal.

I think the same. I highly doubt about validity of cal. certificat.

The "ghost current" is likely an offset error that needs to be trimmed. Either way, it's probably a paperweight now, without more extensive work being performed.

It will be sent back.

Thank you all for your answers and help. It is going back.