Pump stops when CPU heats

[Geonaute]

Hello everyone,

If you think that this thread's title doesn't make any sens, then we agree : my issue doesn't make any sens.

I have a 'Eisbaer 120 CPU - Black' since 2019, because I have a small case ; it is a decent AiO for its purpose.

A few months ago, the pump stopped for no reason ; I was changing another fan that had nothing to do with the AiO, but the pump stopped anyway and I assumed that it just died ('DC-LT 2600 Ultra low noise' has a pretty bad reputation online...), so I replaced it with another basic AiO before I found the time to order a new pump to fix my Eisbaer. The basic AiO worked just ok, so the problem was not my mother board.

Long story short : fixing my Eisbaer yesterday (fallowing this tutorial :

), I found out that the pump I wanted to change was working just fine (and that the new pump I ordered never worked, but that's another story...) so I put it back into the AiO, checking at every step that the pump was still working, and it was until I was done.

Anyhow, this is where this gets very weird : when I set up the Eisbaer onto my CPU, I started my computer, quite confidently, and I noticed that the pump stopped working once again... So I just took the block off the CPU, started my computer (I know this is stupid dangerous), and the pump was working just fine... I checked the wires and everything was fine... So I decided to place the block on the CPU while the computer was booting, and I was speachless : the pump stops working when the block gets too hot... I mean, that's the whole purpose of the block, how is it even possible ???

As soon as the CPU's temperature rises and needs to be cooled down, the pump cease working (I made every possible check to be sure that this is the issue) !!!

Can someone help me or give me an idea on how is this possible ? I tried to google my issue, but I found nothing like it...

Anyway, thank you for your time !

 

[Vanzin]

HI,

this is really weird and I never had something like this in my 5 years since I am working here. Did you check your BIOS configuration that the pump is always running with 12V? It can also be that the screws are to tight. Try to lose them a little and make sure that both screws are tighten with the same pressure. If that also doesn´t help it can be that the solder point from the cables are defective.

 

[Geonaute]

Thank you for your anwser !

Actually, I was wrong, but this gets... weirder than every issues I have ever had...

Apparently, it is not the heat that deactivates the pump but... a short-circuit caused by the contact betwin the processor and the bloc's copper plate... I am as flabbergasted as anyone can be !
This time I'm 99% to be right : I started my PC holding a metal wire betwin the CPU and the bloc, and as soon as I just touched the bloc with the wire, the pump stopped...

Can anyone explain such a bizarre behaviour ?

I repeat that this issue just happened one day for no reason ; one day the pump was working ok, and the next day this happened...

Thank you for your time and help, if you think of anything

 

[Eddy]

This is a very strange problem. As far as I know, we've never had anything like this before, and I've been with the company for almost 8 years.

The heatspreader of the CPU must not be under power at all. This means that it has no contact to any current-carrying components. If a short circuit occurs here, the heatspreader must be in contact with current somewhere. Remove the cooler and the CPU from the socket. You may see something here that is in contact between the CPU heatspreader and the motherboard. This can also be e.g. thermal paste with some dust.