TIM: Manual Spread vs Blob in centre

fastdude

fastdude

Active Member
Some sources are telling me it's better to persevere manually for a more even coat and to avoid leaving corners bare, while others are telling me that simply leaving a pea-sized blob in the centre and letting the pressure from the CPU cooler do the work is a better method.
 
Pea size is slightly large for most TIM(some of the most viscous ones are an exception), its more in line with ~2 .177 copper BB's. It is the superior method to use over spreading, as spreading has a much higher chance of foreign contamination, in addition to the fact it prevents air bubbles, as spreading TIM manually will cause a large number of air pockets to form.
 
I find that i got better temps on both my Q6600 system and on my I7 system if i spread out the Thermal paste myself using a cut up credit card or something similar rather than just letting it do itself when you put the Heatsink on.
 
bomberboysk said:
Pea size is slightly large for most TIM(some of the most viscous ones are an exception), its more in line with ~2 .177 copper BB's. It is the superior method to use over spreading, as spreading has a much higher chance of foreign contamination, in addition to the fact it prevents air bubbles, as spreading TIM manually will cause a large number of air pockets to form.
Click to expand...

Nevakonaza said:
I find that i got better temps on both my Q6600 system and on my I7 system if i spread out the Thermal paste myself using a cut up credit card or something similar rather than just letting it do itself when you put the Heatsink on.
Click to expand...

This is exactly the kind of thing I'm talking about. Two reputable members contradicting each other. While I appreciate that different people get different results, i kind of need a definitive answer.

I imagine it depends on the heatsink used. I'll be using a Gelid Tranquillo

I think I'll stick with Bomber's method seeing as from what I've gathered this area is one of his specialties. Thanks for the input nev anyways
 
fastdude said:
This is exactly the kind of thing I'm talking about. Two reputable members contradicting each other. While I appreciate that different people get different results, i kind of need a definitive answer.

I imagine it depends on the heatsink used. I'll be using a Gelid Tranquillo

I think I'll stick with Bomber's method seeing as from what I've gathered this area is one of his specialties. Thanks for the input nev anyways
Click to expand...

I dont think you will really get a "definitive answer",because not everyones setups are the same,all the processors are diffrent,Even if you compared 10 people with i7 920s on the same cooler ,some run cooler,some run hotter,Then you got Ambient temperatures to take into consideration....diffrent case setups etc.

The only way to find out which is the best method for yourself is to try them both!
 
Nevakonaza said:
I dont think you will really get a "definitive answer",because not everyones setups are the same,all the processors are diffrent,Even if you compared 10 people with i7 920s on the same cooler ,some run cooler,some run hotter,Then you got Ambient temperatures to take into consideration....diffrent case setups etc.

The only way to find out which is the best method for yourself is to try them both!
Click to expand...

Fair enough, see your point. Every CPU is different, blah blah
So many annoying variables eh

I think I'll do that, though I'll try the blob method first and if I'm happy with the temps, not bother trying manual. Maybe if I'm bored one day I'll check out the temp increase/decrease between the two methods

thanks guys
 
fastdude said:
Fair enough, see your point. Every CPU is different, blah blah
So many annoying variables eh

I think I'll do that, though I'll try the blob method first and if I'm happy with the temps, not bother trying manual. Maybe if I'm bored one day I'll check out the temp increase/decrease between the two methods

thanks guys
Click to expand...

You mean if you are bored for a week. No point in taking it off an hour later before it has had a couple days setting in, you will always get a couple of degrees lower after a few days and a few heat/cool cycles, that temperature difference differs between materials.

I always have and always will use the "blob" technique, less air bubbles and my temperatures are awesome without the spreading
 
The clearest way to demonstrate why spreading is an ineffective method is through this video:
http://www.youtube.com/watch?v=EyXLu1Ms-q4

Air is a very poor thermal conductor, especially when compared to TIM (which in reality isn't exactly an ideal conductor anyhow, as it is nowhere near as conductive as solid metals). The majority of times when the spread method produces better results is when not enough thermal interface material is used, not enough pressure is applied, or the material is very viscous. In the end, remember that AFAIK, nobody on this forum is a thermal engineer, and to find what works best for your heatsink/thermal interface material combination. In some rare cases i believe the spread method may work better, for example the HDT style heatsinks.
 
Blob in the centre has always yielded the best results for me and it's the only method I use now.

Most sources can also be wrong as TIM such as AS5 requires 200 hours cure time.

IMO all TIM needs 1 week normal use (turning on, loading, idling (repeat these two) and shut off) before you can accurately compare. One of my goals is to create an accurate TIM comparison.
 
I use the "blob method" because I'm too lazy to spread it, much less find something to spread it with that I don't need to clean up afterwards.
 
hackapelite said:
I use the "blob method" because I'm too lazy to spread it, much less find something to spread it with that I don't need to clean up afterwards.
Click to expand...
Nice logic

linkin said:
Blob in the centre has always yielded the best results for me and it's the only method I use now.

Most sources can also be wrong as TIM such as AS5 requires 200 hours cure time.

IMO all TIM needs 1 week normal use (turning on, loading, idling (repeat these two) and shut off) before you can accurately compare. One of my goals is to create an accurate TIM comparison.
Click to expand...
:good:
bomberboysk said:
The clearest way to demonstrate why spreading is an ineffective method is through this video:
http://www.youtube.com/watch?v=EyXLu1Ms-q4

Air is a very poor thermal conductor, especially when compared to TIM (which in reality isn't exactly an ideal conductor anyhow, as it is nowhere near as conductive as solid metals). The majority of times when the spread method produces better results is when not enough thermal interface material is used, not enough pressure is applied, or the material is very viscous. In the end, remember that AFAIK, nobody on this forum is a thermal engineer, and to find what works best for your heatsink/thermal interface material combination. In some rare cases i believe the spread method may work better, for example the HDT style heatsinks.
Click to expand...
OK, the video was helpful.
AIR BUBBLES!:eek:

Aastii said:
You mean if you are bored for a week. No point in taking it off an hour later before it has had a couple days setting in, you will always get a couple of degrees lower after a few days and a few heat/cool cycles, that temperature difference differs between materials.

I always have and always will use the "blob" technique, less air bubbles and my temperatures are awesome without the spreading
Click to expand...

Yeah, I know about curing. Understood, I'll put a blob in the centre soon as it comes.

Should I buy some AS5 or is will the TIM that comes with the cooler be fine?
 
Unsure as to UK pricing, but i'd go with MX-2/MX-3/MX-4/TX-2/TX-3/TX-4 or OCZ Freeze over AS5 if its similarly priced.
 
You must log in or register to reply here.
Back
Top