DId I put to much thermal Paste

Pro_Tactics

Reputable
Nov 11, 2015
32
0
4,530
CPU: I5 4690K
CPU COOLER: Hyper 212 evo

About 2 months ago when i built my pc i think i put a bit much thermal paste. While putting on the cooler (hyper 212 evo), I was moving it around a bit placing it and just a tiny bit got on the motherboard, which i cleaned off quickly (didnt seem harmful). Anyways after 2 months the CPU idles at around 28-37 degrees at home screen

The amount of thermal paste i placed was around just a little larger then pea sized and was placed in the middle. Is anything wrong?

Thanks
 
Solution

In principle, this should be one of the worst possible techniques since unless your thin application is perfectly flat, is perfectly level and the surfaces are mated perfectly straight and smoothly, it creates its own air bubbles.

Try joining two glass plates with optical adhesive by applying a thin layer on either or both sides. Working air bubbles out of the joint is practically impossible. The pea/bead method on the other hand can yield practically perfect and repeatable results with minimal effort, albeit at the expense of a slightly thicker adhesive layer than absolutely necessary.

I only use the...
It probably was a bit too much. I personally think a pea is too large. I eat peas all the time, and I definitely go smaller than a pea. A lentil is a better size dot than a pea. I think you did put too much, but the better question is what your temperatures are when under a stress test. Idle is one thing, but under a stress test is the most important temperatures.
 

Rookie_MIB

Distinguished
On my last cooler install I did some checking around on different application methods and I went with the 'very thin x'. Under normal conditions, idle was low 30s, load was 55s. OC'd, idle is now low 40s, load is low 70s.

Build is an Asus P7F7-E WS motherboard, Xeon X3470 (stock 2.9ghz), OC'd to 4.0ghz on a low profile cooler. I have a Hyper 212 Evo waiting, just need my IO plate to come in so I only have to take it apart once.

Still, a site did some testing with some plexi and different install methods to check for air bubbles when the clear plate was screwed down. Small pea sized and very thin x actually had the lowest number of air pockets. Large pea (to the point of overdoing it) had little to no effect in spite of it coming out the sides and making a bit of a mess. Thermal difference between the different application methods was almost within margin of error, only 1-2 degrees difference max.

While that might mean something on a LN2 build, here in the real world it's not going to make much of a difference. Overall, just a little dab will do ya.
 

InvalidError

Titan
Moderator

In principle, this should be one of the worst possible techniques since unless your thin application is perfectly flat, is perfectly level and the surfaces are mated perfectly straight and smoothly, it creates its own air bubbles.

Try joining two glass plates with optical adhesive by applying a thin layer on either or both sides. Working air bubbles out of the joint is practically impossible. The pea/bead method on the other hand can yield practically perfect and repeatable results with minimal effort, albeit at the expense of a slightly thicker adhesive layer than absolutely necessary.

I only use the "thin coat" method as a preparation step when joining a rough surface to fill scratches or the gaps between pipes on something like a 212+, then the pea/bead method to put the two surfaces together.
 
Solution


I've never had an issue with air bubbles on the CPUs or GPUs I've had to clean off and apply thermal grease. Temps have always been good the only issue I remember was with a P4 which ran so hot the stock cooler it came with was useless and I had to upgrade to a Zalman model that made it usable. I had to redo the thermal grease until I got the right coating (light film) to keep the temps at their lowest. Still using it today a mater fact from time to time and it is still running good.