3930k temps seem too hot

Crion

New member
Hey guys,

When I was testing my 3930k on an 100i, everything seemed fine with my BIOS reading temps in the low 30s. Now that I've finished and attached my Apogee HD I'm getting temps around the mid 40s. Granted the GPU is in the loop but I've never seen idle temps this high on water without something being wrong. I've already reapplied thermal paste and reseated the waterblock. Can anyone think of something that I'm doing wrong? I've got a 480mm Black Ice SR-1 radiator cooling everything and I don't think I don't have enough rad space.

EDIT: Oh, btw, ambient room temp is around 24C
 
Last edited:
No, the CPU is running completely at stock right now. I usually don't apply an overclock until I've finished setting up a system. IE windows installed, programs installed, etc.
 
At that ambient,40's sounds a touch high...

Okay,go thru the check list.

1. Did you remove the sticker on the block face? You wouldnt be the first to do that.

2. What pump are you using and what speed?

3. What speed are you fans running and what are they?

Pic of the loop will help.

Stock means auto. Auto means the volts will be running a lot higher than needed.

n0000b lesson #1

If they rig hasnt changed and it was running stock during the h100 testing then this isnt the cause.
 
Last edited:
At that ambient,40's sounds a touch high...

Okay,go thru the check list.

1. Did you remove the sticker on the block face? You wouldnt be the first to do that.

2. What pump are you using and what speed?

3. What speed are you fans running and what are they?

Pic of the loop will help.



If they rig hasnt changed and it was running stock during the h100 testing then this isnt the cause.
1. This is a block I've been using for about a year now so yes.
2. MCP655 and currently speed is set at max. Plan to lower it back down to about 3 for normal use.
3. Silverstone AP-121 and at 1450rpm right now. They're configured in pull and as an intake.

Only difference on the rig between now and the test was the card was H100 vs custom loop so components otherwise are the same.

EDIT: Whoops. Forgot the photo. Rad's in the basement of the case.

EDIT2: Oh, and btw, my video card is idling at 30C so the rest of the loop seems to be doing it's job. Hence why I thought I'm missing something related to the CPU block.
 

Attachments

  • IMG_20130112_103010.jpg
    IMG_20130112_103010.jpg
    101.3 KB · Views: 79
Last edited:
I have my 3930k with ram and rive waterblocks at 1.35 for 4.5ghz and my idle temps are between 37-42 coldest core to hottest core. Now I really dint have a great deal of knowledge of thermodynamics but if the pump is pushing the liquid through at full speed is there enough time for the heat to be extracted. Is this a possibility im sure a member of the forums has to be gifted at science/engineering could clear that possibility.
 
I believe you need to use the raised ports on this block,i may be wrong but im sure someone else had a problem like this with his apogee....


EDIT: Having looked at the block top,you have the fittings on the wrong ports,You must use the raised ports in the IN/OUT config displayed on the top
 
Last edited:
Then you may not have anything wrong at all.
Load temps are what you should be concerned with,run a IBT 10 run session and report your load temps?
 
Then you may not have anything wrong at all.
Load temps are what you should be concerned with,run a IBT 10 run session and report your load temps?
Max temp was 72 on hottest core. Most of the time running in mid to high 60s. Seems normal to me. Does the 3930k just have higher idle temps? This is my first 6 core CPU that I personally have owned and done water cooling on.
 
like tom said before, auto functions tend to error on the high-side.
voltages, especially. and higher voltage make more heat. think you might have an
air pocket stuck in the CPU block? all the water pressure is at the lowest point
and air does not travel down to well.
 
like tom said before, auto functions tend to error on the high-side.
voltages, especially. and higher voltage make more heat. think you might have an
air pocket stuck in the CPU block? all the water pressure is at the lowest point
and air does not travel down to well.
I don't think so as I did give the system a few good tilts/shakes while bleeding so that should have dislodged any air pockets although it's not impossible. Any advice or place to point me to on the voltage adjustments. Everty time I've played with voltage it's always been increasing it for an overclock so I've never really lowered it before.
 
like tom said before, auto functions tend to error on the high-side.
voltages, especially. and higher voltage make more heat. think you might have an
air pocket stuck in the CPU block? all the water pressure is at the lowest point
and air does not travel down to well.
That is irrelevant as the rig hasnt changed,he has taken one unit off and placed another on,there is a 10c discrepancy between the H cooler and his custom loop. Voltages dont matter as the baseline hasnt changed.
OP-Did you wind the pressure on with that block? It has a bowed surface and needs some pressure wound on the mount to establish a good contact.
 
That is irrelevant as the rig hasnt changed,he has taken one unit off and placed another on,there is a 10c discrepancy between the H cooler and his custom loop. Voltages dont matter as the baseline hasnt changed.
OP-Did you wind the pressure on with that block? It has a bowed surface and needs some pressure wound on the mount to establish a good contact.
Yes, the block is screwed in as far as it will go. The temp difference is odd but in your experience do these temps still feel in the normal range to you for this CPU?
 
No.
They seem high to me for stock settings,these kind of temps are the kind of thing you expect when OC'd
I can't think of anything else with this block as far as problems go. I'm about to the point where I'm going to switch blocks. I know the Raystorm is still considered an excellent block but are there others that get equal or close performance? I've had trouble with XSPC's mounting hardware in the past.
 
Back
Top