w/c cooling temps question

ArchAnGeL

ArchAnGeL

ArchAnGeL

Associate
Joined
4 Apr 2003
Posts
1,821
Location
Manchester
i recently upgraded my w/c loop so I could give my phase unit a bit of a rest (noise, power consumption etc.)

The new loop is

cpu - storm
chipset - mcw30
gpu - mcw60 with ramsinks
pump - mcp600
rad 1 - mcr320 + 2x loons (3 won't fit until I change my mobo)
rad 2 - unknown 120.1 + loon
tubing - 1/2" tygon with jubiliee clips
1/2" ek barbs

The fans in the top pull air in through the 320 and I am pushing air out of the rear through the 120


Which was cobbled together from my exitsing loop and some second hand bits.

pics
http://picasaweb.google.com/george.gornall/NewMonitorAndWatercooling?feat=directlink#


My ageing rig is

e6400
asus p5n-e sli
8800gt 512

My cpu temps at 1.5v/3.4ghz are around 65 on die but the cores are reporting up to 80ish under load with everest or realtemp which seems high to me.

The GPU and northbridge are reporting 40ish and the gpu increases by around 8 degrees under load. Room temp is 20ish and internal case temp 23ish with an external digital themometer.

The cpu block is hot around the edges.

So why the high temps?

Is the pump not pushing enough water? Visually when filling it seemed to be pushing pretty quickly and there is turbulance in the res.

or is the storm block not really good enough for dual core cpu's?

I was worried about the contact between the cpu and block so I lapped the cpu but it made little difference

There are no kinks or restriction that I can see


Thanks for the help
 
Last edited:
Is the rad warm? The Storm is no longer ideal for chips with IHS but I'm surprised at the temps you're seeing. With the rad you have and the other temps you're seeing, the problem would appear to be limited to the heat transfer from cpu to block which could be simply down to a bad mount. Have you removed the block and checked the TIM pattern?
 
MikeTimbers said:
Is the rad warm? The Storm is no longer ideal for chips with IHS but I'm surprised at the temps you're seeing. With the rad you have and the other temps you're seeing, the problem would appear to be limited to the heat transfer from cpu to block which could be simply down to a bad mount. Have you removed the block and checked the TIM pattern?
Click to expand...

I did that initially as i was worried about the curve in my mobo from over tightening the block and sure enough the tim had only transferred around the edges.

I then lapped the cpu which revealed it was raised around the edges and reseated but I haven't checked it after doing this. So that is likely the next step. I am also going to have a look at the water temps tonight to see what they are like.

I wouldn't say either rad was warm
 
If the rads aren't warm then the water temps aren't the problem which means it must be heat transfer (or bad temperature monitoring).
 
Hi

I know this will be a silly question, but as Mike will agree I am very good at asking silly questions! :)

Is the loop totally free from air? Looking at the thread you know what you are doing, but if the rads are only half full they will not be able to release the heat and be cold.

As I said, probably a VERY silly question but its always worth checking.

Other way I solved a similar problem with contact until I lapped my block and CPU was to use loads of paste between them! I know its not the best way and ultimately you want to lap them if you have a problem, but it sorts it short term!

Hey Mike. Hope you are well mate?
 
Stuck at work wishing I was at home!!

Oh Mike
Check out the Water cooling advise thread by Faceoff and see what you think would ya? ;)
 
Last edited:
i have not managed to strip the block off but I have had a look at water temps tonight.

Running at stock there is probably only 0.4 deg increase in water temp when running orthos for 1.5 hours

I get about a 3 degree increase from idle when stressing the gpu using crysis benchmarking util

so the contact with the cpu must be iffy I guess?
 
no idea what was going on with the storm but after a few days messing with it and trying to get good contact I have swapped to a xspc delta v3 and the temps are much better. The water is also moving a lot faster.

cpu 48 / 58 (core) under load at 1.5v 3.3ghz using everest
35 idle
 
How long has the loop been running? Could be the jets in the storm getting blocked.

Is this a recent build of the loop or has it slowly been getting worse?

If it's slowly been getting worse:

1 Clean the storm out (if its dirty you might wanna reflush the rads and complete loop again to minimise it re-happening)
2 Lap the bejesus out of the cpu
3 Get a backplate for the mobo cpu mount if you haven't already
4 Remount it

If that don't fix it, or it is a recent build, then it might be due to not enough flow; the storm is impingement and loves high pressure. With all the components in your loop there will not be a huge amount of flow i'd guess. Maybe look at twin DDC pumps, or 2 loops, or replacing the storm with something less restrictive and better suited to IHS chips.


edit/ i should learn to read all the thread before replying LOL

ArchAnGeL said:
no idea what was going on with the storm but after a few days messing with it and trying to get good contact I have swapped to a xspc delta v3 and the temps are much better. The water is also moving a lot faster.

cpu 48 / 58 (core) under load at 1.5v 3.3ghz using everest
35 idle
Click to expand...
 
Last edited:
You must log in or register to reply here.
Back
Top Bottom