Terrible temps with new loop.

matty114

matty114

matty114

Associate
Joined
25 Nov 2007
Posts
1,560
Location
Crewe
Got my loop set up today and I am getting bad temps. Within seconds of starting AIDA64 FPU stress test my temps go above 90c at 1.2v

http://i.imgur.com/cPix6yV.png I have tried turning the case in all directions to try to get air out, I have reseated the block 3 times. and it is still the same.

I7-4790k

XSPC RayStorm Block
XSPC D5 Photon 170 Reservoir/Pump Combo
XSPC EX360 120mm Radiator
Scythe Gentle Typhoon 1850RPM fan.
 
did you take the plastic off the bottom of the CPU block?

If the fans are spinning and the pump is flowing the shouldn't be an issue with that setup.

Make sure you don't have the pump set to 1 on the speed dial or something.
 
ubersonic said:
did you take the plastic off the bottom of the CPU block?

If the fans are spinning and the pump is flowing the shouldn't be an issue with that setup.

Make sure you don't have the pump set to 1 on the speed dial or something.
Click to expand...

Plastic is off, i have the flow direction right, the pump is on max speed.
 
Can we get a picture of your loop?

What paste did you use and how did you apply it?

Also is your block tightend down properly making sure to use even pressure ?
 
Cant see any issues that would explain it not shifting the heat did you make sure the block had the proper jet plate for your cpu in it?

although this alone would not justify shuch high temp maybe 4.6 is the threshold for you chip.

What is t he temps under real world usage/gaming?
 
Last edited:
This is my first Water cooling attempt, I do not know what the jet plate is. I will try removing the springs when I get in from work later.
 
when you look at your cpu block you will see 4 screws, you can undo these and change a small plate inside that dictates the flow for different cpus.

I think the raystrom I had came with the 1150 socket plate already in it just check your documents that came with block should tell you.
 
I posted this in another thread where somebody was havin trouble with raystorm temps. When I initially built my pc I was getting idle temps of 50c @ stock clock. I took the springs off the block, reseated it with new paste and screwed the knuts down equally in each corner (3 on one, then 3 on the opposite) do this all the way round.
After I did this my idle went down to 26c.

 
Well my idle temps are 25-30 it jumps up to 90 seconds after starting the stress test. That doesn't sound like the same problem?
 
I am looking at the picture but can't work out the flow direction with certainty.

Are you making sure your flow complies with the inlet/outlet direction of your block?
 
when I first put my raystorm on it was backwards in flow. After a week I figured I had better swap it araound and remounted the block.

Net result in temps was zero. some blocks it matters but on the raystorm it doesnt seem to. well not for me anyway and not to get these sillt high temps.
 
mjallan123 said:
I am looking at the picture but can't work out the flow direction with certainty.

Are you making sure your flow complies with the inlet/outlet direction of your block?
Click to expand...

On the pump the one in the middle and the cap are input s and the one at the back is the output, that is going to the "in" on the cpu block then to the rad.
 
Tried mounting the block without the springs, I took out my res/pump so It can stay upright while I turn the case in all directions, no air came out. My pump is on the max speed and is flowing into the "in" on the block.

It is at 78/82/80/75 after 5 minutes of AIDA64 FPU stress test at stock speeds volts. Could this be a problem with the paste under the IHS? I have no idea what to try now :/
 
Last edited:
Yep, likely the paste. Despite what Intel say the paste on these chips is still rubbish. Mine gets as hot as that at stock.
 
The problem has never really been with the paste, it's not as good as MX2 or AS5 but it's still good stuff, the issue is the is way too much of it because the IHS glue holds the IHS further away from the die than it would be with solder (this distance varies chip to chip), changing the paste was never going to fix that.
 
You must log in or register to reply here.
Back
Top Bottom