Ideas on how to test a pump?

So here's my dilemma... The temps on my CPU cooled by an EK Predator have been a a good ten degrees higher than I'm used to seeing while gaming. When I open the side panel I hear no noise from the pump and I can't seem to feel any vibration (other than what's coming from the fans). I've turned the system on and off a couple times with my hand on the pump to see if I can feel it start up and I don't think it is. The fans (which are connected to the same hub as the pump and running off the same CPU header) are ramping up and down based on temps like normal.

The Predator doesn't have a res, so no help there... The fan headers on the AIO are in a very inconvenient place that would require me to disassemble damn near my entire system to get to them (so I could hook the pump up directly to the MB to check that way). I REALLY don't want to spend half a day tearing my system apart during midterms. Any ideas on how I can test the pump without getting at the headers?

I am running the unit on the CPU header. My MB is an MSI G45 Z87 so the CPU header is set to PWM and there is no way for it to be accidentally set to DC. In any case, all the setting are exactly the same as when I was running my previous Predator which I had to RMA because of a leak.

Amigo, if the pump ceased functioning....your system under load or idle would crit-temp and emergency shutdown to prevent damage after an hour of even idle. The fact your CPU temps are "only" 10C higher than previously, and under gaming load at that, is indicator enough that the pump is working to some extent.

If it weren't (and you overrode the thermal safeties most motherboards should have), you would see idle temps in the 100C range after a few hours. You would also feel heat radiating off the block itself as well as the hose connects (they'd be hot enough to hurt your hand)...the hose itself would be malleable to the point of being just shy of jello. I say this as someone who had to clean up a Fukushima-style pump failure that was not noticed for 14 DAYS of 24/7 uptime of idle and gaming.

Skripka said:

Amigo, if the pump ceased functioning....your system under load or idle would crit-temp and emergency shutdown to prevent damage after an hour of even idle. The fact your CPU temps are "only" 10C higher than previously, and under gaming load at that, is indicator enough that the pump is working to some extent.

If it weren't (and you overrode the thermal safeties most motherboards should have), you would see idle temps in the 100C range after a few hours. You would also feel heat radiating off the block itself as well as the hose connects (they'd be hot enough to hurt your hand)...the hose itself would be malleable to the point of being just shy of jello. I say this as someone who had to clean up a Fukushima-style pump failure that was not noticed for 14 DAYS of 24/7 uptime of idle and gaming.

Click to expand...

That's exactly the kind of answer I was looking for. I thought it was kind of strange that my temps were only 10 degrees higher... But I don't have any experience with faulty pumps, so I wasn't sure. I didn't notice the temps while I was gaming... I instead noticed the max temps on RealTemp have been about 10 degrees higher than the highest I previously recorded on the CPU (81 as opposed to 72). I'm waiting for a new 1440p monitor to arrive so I've been messing around with DSR on a 900p screen. Maybe goofing off in SWTOR with DSR on is just a lot more taxing on my CPU than I previously thought (although usage is still pretty low).

I asked this same question over at Overclock in the Predator thread and got about 10 replies containing 0 real help. One response here and everything is taken care of. May I never take [H] for granted, lol.

pump failure would be immediately noticeable, as Skripka mentioned. What you are seeing is more than likely a flow restriction (kink or too-sharp of a bend in the tubing), corrosion/buildup internally (not common in an AIO), the pump being air bound (not common in an AIO and usually pretty noisy), or something like the heating compound breaking down or rad getting clogged with dust.

Try giving the system a bit of a rock (like, rocking chair) to see if a bubble got stuck in the rad or block.

Gave the system a shake and also released a little stress on the tubing... Not sure which one did it, but within 30 seconds or so there was the swooshing sound of water and then silence. Temps dropped almost immediately by a couple degrees and have gone back to what I was used to.

I think I overestimated the bends I could achieve with the EK ZMT.

Thanks!

Fixall said:

Gave the system a shake and also released a little stress on the tubing... Not sure which one did it, but within 30 seconds or so there was the swooshing sound of water and then silence. Temps dropped almost immediately by a couple degrees and have gone back to what I was used to.

I think I overestimated the bends I could achieve with the EK ZMT.

Thanks!

Click to expand...

Swooshing of water means the system was air bound. That was the sound of the air bubble getting rocked loose.