PDA

View Full Version : disadvantages of losing a heatspreader on p4's?



Amoeba Assassin
07-13-2003, 03:05 PM
I was just wondering, I hear you get slightly higher ocs. Thanks

bcn
07-13-2003, 03:13 PM
yes, we do get them , but I also get like 3-5C lower on the load, though my system temps raised the same level up after removing IHS.

I am using aircooling.

well, and you get some difficulties for mounting the heatsink, but that is not a pain in the ass... :)

CCW
07-13-2003, 07:08 PM
more chance of damaging core, also wouldnt a heat spreader be more benficail because it gives you a larger surface area to cool?

Craig

Creative
07-13-2003, 07:53 PM
my IHS came off lastweek and I then put it back into the machine without after glueing the IHS back on....I dont think it was getting good contact cos it would OC ~200mhz less than previous.

I then took the IHS off and put it into the machine and couldnt get the machine any stable over default 2.8. 10 mins later I take out the cpu and there was no contact being made with the vapo and cpu - Stupid IC7 :mad:

Tried it in the IS7 and it made beatiful contact, I was very impressed....but alas the pc wouldnt even boot up.....the cpu is dead somehow....and its quite sad cos it was a 2.8 that could bench at 3.9 @ 1.65vcore and Im sure would have been 4ghz benchable had it worked without the IHS....:(

Im tempted to try it again, but maybe with a heatsink of a friends....I cbf putting it into the vapo and then having it not work and take it all out again :rolleyes: :D
heheheh

saaya
07-13-2003, 07:56 PM
well depends on the flowrate of the coolant you use. if you use good air/watercooling or even phase change cooling you want to get as close to the die as posible to remove the heat directly from the core.

ctgilles
07-14-2003, 12:50 AM
With a prometeia, I know I had heat spots (not every place is cooled as good) and sometimes the necessary freeze :) using AS Céramique and custom silver/copper coldplate goes best, but a §§§§§ to make good contact

Amoeba Assassin
07-14-2003, 03:36 AM
lol I bet its hard to put back

rashio
07-14-2003, 06:41 AM
just wondering but is there any thermal paste between the die and the ihs?

Creative
07-14-2003, 07:31 AM
Yep there is a rubber like tim that I have seen on the bottom of other heatsinks before....I found the hole in the IHS is actually more a suction type thing that I think Intel put some sort of vacuum into to suck the IHS down or something.

Then there is also some glue type stuff in the fubber bit, kinda like a rubber gasket kind of thing in waterblocks scenario if you know what I mean :)

rashio
07-14-2003, 07:38 AM
so could you not replace the rubber bits with as3 or some sort of stuff?
also between the die and the ihs could you not put as3 in there or some thing else?

i am not saying this will work, but i was just wondering if any 1 has tried this or something like this?

Major
07-14-2003, 09:20 AM
It seems to me that logicaly you should see better performance with out the IHS, IMO that getting good contact pressure would be the issue.

bcn
07-14-2003, 09:25 AM
Yes, the performance is better WITHOUT the IHS, cause you get lower temps first and then if you think, physics you cannot change. the more surface contacts you have for the heat to pass away, the hotter the core will run. yes, I put AS3 to the core and it runs fine.
the hole in IHS is ment for the gases to go away during the manufacturing proces of P4 - this you can see in Intel documents.

And for the pressure. I use even more pressure on the core than before - I note that on the levers of the heatsink, when I mount it - it takes more force to mofve them and the core seems to be just fine :)

macci
07-14-2003, 09:25 AM
Originally posted by saaya
if you use good air/watercooling or even phase change cooling you want to get as close to the die as posible to remove the heat directly from the core.
And if you use poor aircooling..? :D


also wouldnt a heat spreader be more benficail because it gives you a larger surface area to cool?
It only adds thermal resistance between the core and cooler base.

AndrOvr
07-14-2003, 09:41 AM
Hiya guys,
How do u remove the IHS in a easy way?
W out razors...possible?

CCW
07-14-2003, 09:44 AM
JC always said he had problem gewtting good contact with no IHS.

saaya
07-14-2003, 05:54 PM
the hole in the ihs is to let air get in or out. the temp diferences make the air inside the ihs expand and pull together, if it wouldnt have a hole it would probably pop off because the air inside would expand and creat some pressure.
thats probably why some people lost the ihs, they tapped or closed the whole with termal grease or had the hsf right on it.

PureBooYah
07-15-2003, 08:18 AM
Originally posted by CCW
JC always said he had problem gewtting good contact with no IHS.

i took mine off and had that same problem, i had to reseat the HSF 3 times before i finally made good contact.

once i got good contact i was able to oc 5fsb higher without the ihs, which isn't much, but every little bit counts :D

PureBooYah
07-15-2003, 08:22 AM
Originally posted by AndrOvr
Hiya guys,
How do u remove the IHS in a easy way?
W out razors...possible?

i used razors and after working it in and around each corner it came off pretty easily (i did slice my finger once :( ). In each of the corners the IHS isn't connected to the silicon so there's a little gap that a razor can easily fit into. i don't really know any other way to get it off w/o using razors.

karmakid
07-16-2003, 04:36 AM
also wouldnt a heat spreader be more benficail because it gives you a larger surface area to cool?

LMFAO :)

thermal paste is bad mmkay. CPU<->IHS<->cooler = 2xthermal conducting material

CPU<->cooler=1xthermal conducting material (paste and so on)

and the transfer area on the die dont get bigger cause you put a metalplate on it. and the metalplate is weird cause I found out that it was magnetic. !! that means it caontains IRON. IRON !?!??!