Corsair H50 Running Hot

Are you sure the locking screws are sunk in the back plate properly? I drilled 4 holes in the back panel during installation because I found that the nuts had a tendancy to drift out, I would guess that this would cause the tensioning to be uneven and cause air gaps. The holes allowed me to check everything was seated properly.
 
I am going to resit the cooler today. It might seem like there is no need but heh, its a Sunday and I have nothing else to do. It'll be interesting to see if there is a problem with the contact though, to be honest, I'm starting to think there isn't.

After playing Metro 2033 and Crysis on full for >2hours the temps have been sub 78c which is very good. It only seems to be this stress test that is causing the temps to go high. One thing I have noticed, the CPU fan does not increase in speed with temp increase. It just sits at ~1560rpm. The cpu fan in this case, is the fan attached to the chassis side of the radiator; pushing cool air from outside, through the rad and inside the case.
 
One thing I have noticed, the CPU fan does not increase in speed with temp increase. It just sits at ~1560rpm. The cpu fan in this case, is the fan attached to the chassis side of the radiator; pushing cool air from outside, through the rad and inside the case.

Perhaps you need to enable that in the BIOS somewhere
 
For reference:

21740074.jpg


Ambient is 26oC in here, small room door closed all day with the PC on which it never was in the heat, it was 29oC at its highest this year.

Sounds like your sample is dodgy

Hawker
 
97236939.png


Test ran for about 4 seconds on the settings shown. Take a look at the Maximum Temps as I was a bit late with the printscreen.
Due to the fact on two of the cores are getting hot, that could indicate that the heat plate is uneven... Would anyone agree?
 
Test ran for about 4 seconds on the settings shown. Take a look at the Maximum Temps as I was a bit late with the printscreen.
Due to the fact on two of the cores are getting hot, that could indicate that the heat plate is uneven... Would anyone agree?

You need more than 4seconds to put CPU and cooler on the right temp.

In the first 4 it doesnt even load the test fully ...

You want to check at least 30seconds before you check any temps, anything below that is irrelevant, for 4 seconds I could run this thing passively and get same temp.
 
For what it's worth I picked up a second hand h50 and fitted it to my (also second hand..go ebay!) i7 920 C0 and am disappointed with the temps at stock clock speed.

To the point where I'm not going to bother overclocking it until I get a better cooler. I've reseated the h50 twice with no joy - but I did note that the distribution of thermal paste when taking it off each time was very very poor on the chip leading me to agree with you that it's not a particularly snug fit.

Mine is second hand so could be ropey, or there could be a bad batch but my conclusion is that despite the excellent press the h50 gets around here it's really not that great.

I also found installation to be very fiddly - which is one of the things I was looking to avoid.
 
I should add I'm running it from the cpu fan header (at full tilt in the bios) and not a molex. I'be intrigued to know if it would increase the pump RPM any...something to try I guess.
 
Just an update for anyone interested. I resat the H50 using some MX3 and the temps are look much much better. I blame the original high temps on two things. 1) The factory fitting thermal paste corsair apply to their coolers & 2) The fiddle-ality of fitting the H50, causing you to catch the applied paste while performing the insert and twist lock thingy.

Temps are now;

Idle Max (running metro for 2 hours, crysis for around 1 (both on full))
40 57
39 56
40 57
38 55

Just incase your interested, his hd5870 max 64c (idle 35)
 
Back
Top Bottom