Applying Arctic Silver Thermal Paste 1

[biosol]

I'm trying some Arctic Silver 5 on my 3.2 prescott with the retail hsf. After cleaning both the hsf and cpu according to a couple guides I put one dot of the AS 5 on the cpu and then 2 dots on the stock hsf. It didn't smear around easily but I got most of the area covered. Then I wiped off a fair amount, mounted it, and fired the system back up. I started the Windows based HW monitor by the board manufacturer and watched for a couple minutes as the temp went from the low 40sC to 47C. I checked a couple emails and opened one web page that had some flash on it. Then I decided to reboot and see what the BIOS had for a temp. At first it wouldn't reboot. After a couple minutes it did and the BIOS cpu temp was 83C. It seems as though I need to redue something. Do you think I should only put 1 dot of Arctic Silver on the heatsink???

Suggestions?

Thanks in advance.

 

[Kiernan]

I put one drop on the HS, rub it in so it fills any scratches, then wipe it off. You should be left with a little residual silver paste. Then put one drop on the CPU and smear it around for full coverage about the thickness of a piece of paper. Then mount the HS.

Just remember to clean the CPU and HS with rubbing alcohol to remove what you've already put on.

 

[micker377]

By "full coverage", that means coverage of the CPU chip. The silver content can be conductive, make sure it doesn't get on the other components on the die.

 

[biosol]

Ok, I'm going to try it again. Will clean off and re-apply just one drop on the HS and will wipe it off better.

 

[franklin97355]

Read this first!

http://www.arcticsilver.com/arctic_silver_instructions.htm

 

[biosol]

Thanks franklin, that's one of the guides I have printed out. I wiped off all the AS 5 paste this time after applying it to the HS. After firing the system up again I'm going straight to the BIOS. It started out at 37C and is just idle there. After a couple minutes it seems to idling ~47C.

Which one is usually more accurate, the BIOS or the MB software utility in Windows (temps)?

TIA

 

[biosol]

Does this have a breakin period? I mean I'll probably run it for 30 minutes now then shut it down for the night before running it for an hour or so tomorrow morning before shutting it down again. Is it better to run for a couple hours first?

 

[stduc]

biosol
Does this have a breakin period?

Not at all. Heat pads need to be hot long enough to to do their re-crystalation thing, but arctic silver doesn't change its state.
As all electronic equipment suffers its biggest threat when powered up (electrical & thermal stress etc) I have a personal rule to prolong the life of my PC. Basically, if I plan to use it tomorrow, I leave it on. You may want to ponder over that!

 

[paparazi]

biosol
I am a bit confused about what you are doing!
Quote: "I wiped off all the AS 5 paste this time after applying it to the HS"
There should be a thin almost translucent application of paste to both CPU core and heatsink base, any excess paste will, under pressure, ouuse out so to speak.
Not applying enough may leave non heat conductive, pockets of air (not good for heat transfer)
A DOT of paste? well that depends of the size of your DOT? P4 and new Athlon64 cores need rather more paste applying than say an older Athlon, the contact area is substantially larger, a guestimate amount for a P4 core would be 3 or 4 rice grains in volume, enough to completely cover the core and the same for the heatsink base.
The paste as we know, is not supposed to act as a layer between the two contact surfaces but there needs to be enough paste aplied so that microscopic surface inperfections and voids are replaced by the heatsink paste material.
Martin


We like members to GIVE and not just TAKE.
Participate and help others.

 

[biosol]

Sorry for the confusing wording paparazi. The stock heatsink has a copper circular contact area. I put one small drop (about the size of the one in the picture from the Arctic Silver how to guide) on it and spread it around. The contact area has circular grooves. I wiped off 'most' of the compound on that area leaving it in the grooves and a thin layer as best I could tell. I didn't use any cleaning fluid, just a non lint paper. Then I added one drop to the center of the cpu contact area as the guide instructed and after mounting the HS gave it a slight twist in both directions.

This morning is it has been running for about 15 minutes and is still in the high 30sC. I asked about a break in last night because as I used the system to do some basic stuff, the temp started to drop over time toward the low 40sC.

 

[stduc]

Sounds okay to me.

 

[biosol]

Yeah, it looks to be operating ok. One of the reason I asked about a break-in period is because the issue I'm trying to address is the heat under 100% load when ripping, encoding, music, movies etc. Otherwise it worked fine before the AS 5. Right now I've had to pause the ripping/encoding every couple minutes to let the process cool down.

Should I wait at all to try that or go for it now?

 

[stduc]

With arctic silver - you've either got good thermal conduction - or not. Actually the same is true of a heat pad. So you may as well stress test the CPU straight away - as wait.

But at what temp are you pausing? Up to 70C - or even more for short periods is fine. What's you case temp? Perhaps if thats higher than 10C or 15C above ambient you may want to consider improving airflow through the case.

 

[biosol]

stduc, thanks for the info. I am actually testing it right now, by analyzing and encoding a DVD. The first part went better than before, I only paused it once (although the cpu usage seemed less than before as it wasn't pegged at 100% like I remember it being). During the encoding I've paused it twice at 63c and 65c.

For the case, the temp is 32c. I've had the case side panel off for a couple weeks now. Like I say for everything else it runs fine, it's just the heavy/max cpu stuff. A friend of mine is sending me an aftermarket, TheramlTake Copper cpu HS, so I think/hope that may be necessary/do the trick.

This 3.2 cpu seems kind of like a high maintance euro sports car that always needs tuning.?

 

[biosol]

Oh, I also forgot during this test that I had better results before by setting the program to use only 1 of the 2 HT cpus. This cut the cpu usage in half. I'll try that again and see if I can run the whole program without pausing.

 

[garebo]

From the guide:

Important Reminder:
Due to the unique shape and sizes of the particles in Arctic Silver's conductive matrix, it will take a up to 200 hours and several thermal cycles to achieve maximum particle to particle thermal conduction and for the heatsink to CPU interface to reach maximum conductivity. (This period will be longer in a system without a fan on the heatsink or with a low speed fan on the heatsink.) On systems measuring actual internal core temperatures via the CPU's internal diode, the measured temperature will often drop 2C to 5C over this "break-in" period. This break-in will occur during the normal use of the computer as long as the computer is turned off from time to time and the interface is allowed to cool to room temperature. Once the break-in is complete, the computer can be left on if desired.



Good advice + great people = tek-tips

 

[stduc]

[blue]biosol[/blue] A better heat sink will make a noticeable difference. It could well reduce your temps at 100% CPU by 10C or more.

[blue]garebo[/blue]I've read that claim of the Artic Silver people. But I haven't seen any justification for it. I'm not sure that 2-5C would be very noticeable in a real world situation either. Why should the break in period be longer without a fan? Are they trying to say that the vibrations caused by a fan aligns the silver particles?

Call me an old cynic if you like!

 

[garebo]

I wasnt alluding to the fan business, only the fact that in time the 2-5C temp difference would apply.
Plus, the question was asked and i only showed the answer to the question.


Good advice + great people = tek-tips

 

[stduc]

lol - I wasn't critising you! Just questioning the verasity of the arctic silver people - not that it isn't good stuff. But have you ever noticed a drop over the first 200 hours of use?

 

[biosol]

Thanks stduc,
I'm looking forward to receiving the after market HS. I think there has been a reduction of a couple degrees C between the AS 5 and the regular Intel goo after idle and working temps. And maybe it takes longer to really heat up at 100%. But I'll try put the new HS as well. I read one post on another forum where someone said that even at 100% their cpu temp ran in the 40s.

First timer with thermal material, thanks for the info and suggestions.