Overclock.net › Forums › Cooling › Water Cooling › I think my cooling solution has kicked the bucket
New Posts  All Forums:Forum Nav:

I think my cooling solution has kicked the bucket

post #1 of 3
Thread Starter 
So I've recently noticed that my system has been turning itself off randomly under load for about a month now, and tonight it happened again, and I finally decided something was genuinely quite wrong. I changed out the GPU about 4 months ago, but haven't had any problems until just recently, so I tentatively eliminated it as the source of my problem.

However, upon stress testing the CPU, I was able to get the system to power off in about 5 minutes! Once I started monitoring the temps, it came to my attention that my AMD FX-8350 was reaching 93 degrees Celsius; I don't need anyone to tell me that that is NOT normal, because I KNOW it isn't. (and obviously nowhere near safe for the CPU) The T.j. max of the CPU is 90 degrees.

I cleaned out the radiator using a vacuum (I could see a lot of light through at the end and it didn't look visibly dusty afterwards) which helped keep the system running for longer, but Prime95 eventually crashed it a couple minutes later. After repeated testing and another round of cleaning, I booted up my system to find that my idling temperatures in the BIOS were sitting at a steady 72 degrees, and I realized that the most likely candidate for blame is the pump of my Corsair H75 Liquid cooling AIO. This was supported by the fact that one of the rubber tubes to the radiator was blazing hot, the other one was stone cool, and the air coming from the back of the radiator was about room temperature.

After this round of testing, the system now shuts itself off just running Google Chrome/booting up, never mind playing a game. I guess my intense stress testing might have finally pushed the pump over the edge.

I also confirmed that it wasn't the PWM header; I swapped it around onto another PWM header and plugged a different fan into its original PWM header. The fan ran fine, and system temps remained unbearably high.

My intuition tells me that my pump has gone bad (by slow process of elimination) but I just wanted to pass it by the much more experienced community and see if they had any counter-ideas, or alternatively, that thou might confirm my previous supposition.

Oh yes, and fan speeds look entirely normal; forgot to mention. I've scrolled all through the different fan settings from Silent to Turbo, HWMonitor returns reasonable RPM's, and everything appears fine on that front. It does not however, affect temperature in the slightest. And I've also confirmed that all my CPU OC settings (CPU/NB load line, etc.) are set to default.

AMD FX-8350
Nvidia GTX 1060
16 GB G.Skill Ripjaws @ 1866 Mhz
Corsair H75 Liquid AIO Cooler

Thanks very much for your input!
post #2 of 3
I think you have already guessed right.
Just another check: A working pump must emit some noise, yours ?
post #3 of 3
Thread Starter 
Not any decipherable noise. I fortunately had the stock cooler that I never threw away (thank goodness) so hooked that up and temperatures are MUCH more stable! It's clear the cooler was at fault. Also fortunately, my AIO is still under warranty so I should hopefully be able to get it replaced. Annoying, and I'm hopeful I didn't massively decrease the lifespan of the processor, but it would appear there is no harm done thumb.gif

I was concerned that perhaps over time the pump had started to lose contact with the processor, but I've discounted this on the grounds of one of the pipes being as hot as it is, and with the other pipe not heating up at all, even as temperatures climbed. Clearly heat was being transferred, but was not getting through the radiator as well as it should be.
New Posts  All Forums:Forum Nav:
  Return Home
  Back to Forum: Water Cooling
Overclock.net › Forums › Cooling › Water Cooling › I think my cooling solution has kicked the bucket