Sign in with
Sign up | Sign in
Your question

Lap CPU and Heatsink but no temp. change

Last response: in Overclocking
Share
December 17, 2010 6:08:40 AM

Hey guys, can anyone tell me if a 2-3 Celsius difference on idle or a 5-6 Celsius difference during load is bad?
I wanted to prepare my E8500 for OC and decided to reapply the thermal paste (Artic Silver 5). I let settle for a couple days and now core 0 stays at 35 C and core 1 is about 37-38 C when idling. After playing a game (Fallout New Vegas) for a little bit I alt-tabbed out to check temps and core 0 was at 38 C and core 1 was at 43-44 C.

Have never checked load temps like this that I can remember so this could have been happening before I reapplied the paste, but before, the idle temps were always even.

Don't think this is good. The heatsink is a Sunbeam core contact and uses Heat-pipe Direct Touch tech.

After this, I have now actually went through the lapping process both on my cpu and cooler. But the temps are exactly the same. 35 C on core 0 and 37 C on core 1 when idle. After being on load for a min or two, the temps are: core 0 40 C and core 1 45-46 C. Why would the temps not lower and why do I still have a difference like this? Do I just need to let the paste settle more?

Also, I am using HWmonitor and Realtemp to watch temps and both seem to never want to go below 35 C. Am I right in thinking they both naturally don't record that low of temps?
December 17, 2010 7:33:14 AM

First, it is normal that not every core have the same temperature.
The method you used to lap your CPU is comparable to this one?:

To reduce the thermal difference, you can try to tighten the screws and put a small drop of TIM (less better). The gravity could influence also, as one part of the cooler makes more pressure on the CPU.

When you check load temps, be sure to use programs like Prime or LinX. Watch out because the temperature climb pretty fast so if you overclocked randomly without one of these program, you must bring your CPU back to stock and overclock from that point.
m
0
l
a b à CPUs
a b K Overclocking
December 17, 2010 2:37:24 PM

I use Arctic silver and it seemed like with more time my temps got better.
m
0
l
Related resources
December 17, 2010 6:33:12 PM

Thanks for the replys guys. I do know that a temp difference between cores could be alright, I just never had one before and didn't know if a 5-6 C difference during load was bad.

My lapping process was about the same, except I did the back and forth then rotate 90 degrees thing. And yes, I know exactly what your saying Just a nickname about the cooler not making direct contact due to gravity. It seems that does play a part in it.

Ultimately, after doing more reading, it seems that artic silver 5 has an extended curing process (200 hours!) and most things are worked out after the tim cures. Guess I will just wait.
m
0
l
December 17, 2010 9:45:16 PM

Or you could get mx-2, no curing time, spreads better, non-conductive, etc.

Superior and cheaper [:tigsounds]
m
0
l
a c 172 à CPUs
a c 197 K Overclocking
December 17, 2010 11:35:17 PM

I have been using AS5 for years and I have never noticed an extended "curing" time. Because I overclock and do 24 hour stress tests to check for stability, I use most of that semi-mythical 200 hour time period testing. And I keep a maintenance log on my systems, I am not forgetting what my original temperatures were.

As far as the "no temperature difference after lapping" is concerned, all that means is that the CPU ISF and the heatsink base were flat enough before you started.

m
0
l
!