Strange problem with temps shooting up on load

Syr has you well in hand here, not much I can add.

One question, that is if your temp spikes aren't going hand in hand with CPU activity. Have you run this mobo with just a HSF on it, and if so were your temp's stable?

The reason I ask is because I've had a couple of Asus mobo's that would do the same thing yours is doing, they had faulty v-reg's on them.

I generally take off and straight-edge any stock sinks I leave on my mobo to make sure they are making good contact, a lot of the time the sinks wont be making contact on the components very well.

Quote:
no, the spikes in temps happen in a cycle every 2 minutes or so when my CPU activity is at 100% (small TFFs prime95 on 4 cores)

Quote:
Pump seems to work fine but it hard to see if it stops for few seconds once in a while. i don't have a flow indicator.

Quote:
I can't say i did... that may indeed be the problem. I guess am gonna have to purge my system to have a look

Quote:
Will do tomorrow. It goes reservoir > pump > rad > CPU > reservoir

Quote:
I haven't tried the mobo with just an HSF on it. I haven't actually got an HSF to test w/. I've lost the original one that came with the CPU.

Quote:
What are v-regs?

Quote:
not sure what you mean by straight-edge. I have tried reseating the CPU waterblock.

Thx you both for your help

Also, I've just realised i didn't check if the cpu block was in the correct orientation. doh. that's most likely the problem. will check tomorrow & report back.

Quote:
Vregs- voltage regulators for the cpu...some are voltera(spellcheck) when you change the vcore..you're basically controlling your v-regulators..

One thing i observed with the vcore in the video is that the vcore remain stable..i would doubt the vregs creating this problem where the temperature would increase steadily.

I'm more likely seeing the culprit to be the dts sensors on the cores. these are ptc probes..normally they would change resistance with varying temperatures. the current in that circuit would be monitored and converted in terms of temperature. It could be the ptc sensors(inside the cpu) faulty or the monitoring circuit(mobo) is faulty.

Another temp sensor is the tcase sensor(inside the cpu). This is the sensor for the cpu case. Some programs monitor this, eg. Everest. normally intel specifies max Tcase. You can run prime 95 with coretemp/realtemp and Everest and see if the Tcase temp also rise with the coretemps. if this happens then it might be the voltage regulators.

Quote:
if you watched the video...it is very unlikely because the rise in temperature is very slow...normally waterloops even with bubbles will have almost constant temperatures when loaded...his case on the other hand has some problems...i've tried more than 5 cpus and primed them..and i've never seen his problem before..which is sorta interesting...