Sign in with
Sign up | Sign in
Your question

Pump not working properly...?

Last response: in Overclocking
October 9, 2011 4:38:50 AM

So I hooked everything up. I have it set up Res>pump>rad>cpu>res

I just filled it the for the first time, I can't hear the pump but when i touch it it is making sort of a bumping feeling inside it. At first I thought that it was just pumping the air pockets through, I haven't had it on for very long (not even 5 minutes) but it doesnt seem to be getting any better.

Personally I suspect two things; either I can't have the pump lying on its side (EK-DCP 4.0), I have the intake on top and out going out to the side, I read that the pump is fairly flexible about the orientation it can be placed as long as there is water going into it.
So it's either that or I hooked up the res backwards (put inlet on outlet and outlet on in). I have the XSPC single 5.25" bay without a pump in it. If you look at it from front on, I have the right hole as intake and the left as out (so in other words the hole that is in the middle of the res I have as outflow).

Could anyone tell me if its one of those problems or anything else?

I can post up pictures if that helps...

More about : pump working properly

October 9, 2011 4:40:26 AM

So basically the pump makes a small vibration every half a second
October 9, 2011 8:05:09 AM

Righteo so it seems the pump isn't pumping at all (even though I can feel it vibrating). I checked the CPU temps and they are between 90c and 92.5c so something is not right here
Related resources
a b K Overclocking
October 9, 2011 9:35:49 AM

pumps work with a semi sphere sitting inside itself. the impeller is attached to a rotating peice of steel or aluminum with the center cut out. the semi sphere it drops into has a shaft with a plastic bearing on it that this impeller attachment rests on. inside the half sphere wich the impeller is attached to is graphite that is the base for this pillar to rest on.

the base of the pump contains a powerful magnet witch forces this half sphere/impeller combo to rotate and pump the water.

if you feel it moving and bumping around it does not have water in it. OR the graphite is broken and the pump is not properly balanced anymore. take the pump out of the comp and put it in your bathtub, see if it spurts water smoothly. if it does then you just need to prime your system. gravity can be used if you have a light case or are strong just rotate the case in the air. I have my world strong man champion house mate rotate my 130Lbs comp when i need to do this. you can also use a hand pump in the rez to prime the system.

with 90c temps you are not getting any water in your water block. even stagnant water should have lower temps than that.
October 9, 2011 11:30:22 AM

I got the pump working, idle temps are 36c. I'm not convinced these are good temps but I think its because of the way (and amount) of thermal paste I put on the CPU. I'm going to re-apply it either tonight or tomorrow and see if temps improve.
a c 324 K Overclocking
October 10, 2011 2:06:43 PM

Idle temps of 36C are pretty good, no matter what your ambient is (unless you live in the arctic and your box is outside).

Load temps are what you should really be concerned about.
a b K Overclocking
October 10, 2011 10:08:53 PM

36 is fine, check your load temps tho with prime and make sure they dont get to high. you can run a temp program like core temp or speed fan. look at all your cores. if all cores are with in 2C of each other then id say your goop spreading is fine, if one core is hotter than the rest tho then your goop is not properly covering your CPU.

idle temps dont mean jack, check your load temps with prime for at least 10 min then post em up
October 11, 2011 3:56:23 AM

45 min of test time:
Core #0: 44c
Core #1: 48c
Core #2: 50c
Core #3: 49c

I'm very happy with those temps, I just thought my idle temps would have been lower but I wasn't concerned they were too high

Best solution

a b K Overclocking
October 11, 2011 6:59:17 AM

temps are low enough, now its time to start with the OC.

the 44c - 50c temperature difference stands out to me although the 44 is the anomaly so perhaps the power saving features were giving core0 less power and thus less heat. either that or only 1 core had proper goop over it or the water block you have is less than efficient when it comes to water flow and the outsides of the block stay hotter than the center (i would expect core0 to be roughly in the center of the water block)
October 12, 2011 6:12:05 AM

Yeh I was going to say that my cores often exceed the 2c difference between them. Core #0 is often the coolest. For example right now (I only have browser open and downloading something off steam) Core 0: 25, core 1 28, core 2: 28, core 3: 30.

I suspect it might actually be the way I applied the paste again, because when I reapplied it from the first time it was hard for me to align it properly and have even pressure because I already had all the tubing hooked up so it restricted my movement a bit with the block.

I'm getting a gpu block and another rad soon (either friday or monday (hopefully)) so I might do the paste again when I remove all the water so I can put in the gpu block and rad.
October 12, 2011 6:14:44 AM

I just read up a bit about core 0 being cooler and it seems a few other people with the i5 and i7s that are getting the same results, even underload
a c 324 K Overclocking
October 12, 2011 1:44:48 PM

Most people with a multi-core CPU have 1-2 cores that run cooler at idle and's just how it is. Some of it has to do with how the IHS actually spreads heat and the sensors read it. The rest is likely to do with the work load the cores are doing, even at idle on various threads of workload.
October 24, 2011 12:53:45 AM

Best answer selected by aleckazee.
a c 205 K Overclocking
October 24, 2011 1:51:55 AM

This topic has been closed by 4ryan6