Blob vs spread (applying thermal paste)

Page 3 - Seeking answers? Join the Tom's Hardware community: where nearly two million members share solutions and discuss the latest tech.

Conumdrum

Distinguished
Nov 20, 2007
2,397
0
19,960
86
RS sells MX-2? Wow! Nice to know. Now when my 30 gram tube will ever run out.........

MX-2 is a bit better in tests than the others. But any 3 will do. MX-2 is easy to apply.
 



No, RS sells Artic Silver 5 and Artic Silver Ceramique 3.5g tubes under the Thermaltake label.
 

adamh9

Distinguished
Jan 16, 2010
164
0
18,680
0



yea, I don't mind the noise, it's actually quite soothing and helps with my sleep :D


they sure move some air, it's like a wind tunnel in my case, but my temps dropped 10C after adding 3 of these(2 front 1 side)
 

OvrClkr

Splendid
Moderator


Correct, but that is only for the first fan you throw in your cart. When you buy off the Egg you need to make sure you buy more than one product just to justify the shipping charges. I purchased a DIM fan, a 140mm (exhaust) fan and some CPU paste and I only paid 7.23$ dollars for shipping ;)
 



Yeah well its definitely the better option to buy the fans as an addition to a larger purchase for sure.
 

adamh9

Distinguished
Jan 16, 2010
164
0
18,680
0
BTW. if you go to amazon.com right now they have a one month trial for amazon prime where you get free 2 day shipping for a month. Just make sure you delete your credit card after you sign up!!
 

TheRealMaxor

Honorable
Jul 15, 2012
1
0
10,510
0
I've been a technician for a while, and before that I built computers in my home. I've done test upon test, and every time no matter the thermal paste.... I've come to find a few things out.

1) Thermal paste is only there to provide a connection, thermally conductive connection, to the heat sink. We all know that thermal paste will NEVER be as thermally conductive as the heat sink itself, so with that in mind, I've always found less is actually more, but always maintaining enough for even FULL coverage of the contact area of the heat sink. (Sometimes the entire CPU is not in contact with the heat sink, in that case, try and limit the paste to just the area that will make contact) BTW the ones that don't use the full contact area of the CPU, are junk!

2) With keeping that less is more, I've found that the thinnest layer of paste across the entire surface to be the best. I use a plastic sticker applicator I sized down to be easily held. With some practice I can do a one swipe pass and get a thin, almost even, layer of thermal paste.

I'm not just using the time I've spent building computer after computer as a technician, but I work in a small shop that takes pride in our builds, and we benchmark every single machine that leaves here. The ones with the lowest temps are the ones that we use the above method on.

We've had other technicians using various other methods, some of which where mentioned here, and less is ALWAYS more, and thin even layer turns out to run the coolest. So take that how you want, I'm just laying out the facts as we've seen them.



Also lapping/buffing/honing the heat sink and CPU surfaces is a VERY inexpensive way to permanently drop 5-10 degrees Celsius, if you can do a precision job, you actually do NOT need thermal paste.

We've created three machines this way, and we even TRIED thermal paste! Every time it added 1-5 degree's depending on the paste. Obviously the precision allows for a perfectly mated surface that doesn't actually have room for thermal paste or it will lose contact with the other side. This method will ALWAYS run the coolest, and the pressure of the heat sink clamps allow it to never move, and will resist the minor warping caused by heat fluctuations in metal.
 

Mupples

Honorable
May 10, 2013
15
0
10,510
0
A skittle size drop is the best method, never prespread. You just have a toothpick and qtip handy if it starts showing around the edges. I've been doing it for years and I've got the accurate amount down precisely, I usually see a little on the sides but never something coming out.
 

Calculatron

Honorable
Apr 3, 2013
2,240
0
12,460
384
This site recently did an excellent two-part article on this subject:
http://www.tomshardware.com/reviews/thermal-paste-heat-sink-heat-spreader,3600.html
http://www.tomshardware.com/reviews/thermal-paste-performance-benchmark,3616.html

Hardwaresecrets did a less indepth two-part article, devoted mostly to just the *how* of the application:
http://www.hardwaresecrets.com/article/What-is-the-Best-Way-to-Apply-Thermal-Grease-Part-1/1303/10 (dealing with flat-based heatsinks)
http://www.hardwaresecrets.com/article/What-is-the-Best-Way-to-Apply-Thermal-Grease-Part-2/1392/8 (dealing with HDT heatsinks)

I now personally own three aftermarket cooling solutions: two HDT (Enermax ETS-T40, Arctic Cooling A30 Freezer) and one flatbase (Be Quiet! Shadow Rock 2). For the ETS-T40, which has the first generation direct-touch heatpipes (that has the aluminum dividers between them) the transversal stripe worked best. The A30, which has the second generation direct-touch heatpipes, but still has little gaps where they meet, it wasn't too picky, but the small dot and transversal stripe seems to operate the best. With the Shadow Rock 2, which has the traditional flat base, I only have tested the dot and the spread method - it is far to difficult to install to facilitate easy testing - and the dot method netted me just under 5C gain in temperature. So my finding more-or-less followed that of HardwareSecrets.
 

ASK THE COMMUNITY

TRENDING THREADS