Ceramique: Alright, what gives?

[fdisk2003]

I recently built a P4 system:

Asus p4p800 deluxe
Geil platinum 512x2 PC-3500
P4-p2.4c
no OC (yet)

I'm using the stock intel HSF currently to see how well it works.. I lapped it with 400, 600, then 800 grit. Decent smooth surface, can't feel any ridges.

I had a week to go on my Ceramique to arrive so I did with the generic silicon stuff you get from PC stores.

Idle: 31-33c
Load: 41-44c

(Window open at night, nice and cool out, case open)

Not bad..

Okay, so yesterday Ceramique arrives. I put it on and wtf, look at my temps:

Idle: 33-35c
Load: 47-49c

!!!!

What is going on?

I email Arctic Silver, tell them how I installed it (correctly, btw). They said to clean it off and try again.

So THIS time, I go with some Brasso on the HSF, then I clean it again with xylene based carb cleaner, and finish off with a coffee filter so there's no lint. Okay, so here I have a HSF with a mirror like reflective surface. I apply the Ceramique and rub it in to the HSF, I put a dab on the IHS on the P4 CPU. I lock it down and run it under load (this 3d game I have)

Idle: (not tested)
Load: 45c - 47c

I should also add, it hasn't been on long, the temp appears to be steadily increasing. It might reach the other temps in a bit, I'll have to see.

(Window open at night, nice and cool out, case open)

Okay, what in the world gives here?

Why is Ceramique performing so crappily?

I'm applying it right. I rub it into the HSF with a bag over my finger then I wipe it with a coffee filter so the HSF is "hazed", just like the AS site says to. Then I put a rice grain sized glob on the IHS. I put the HSF on, turning 1-2 degrees to get rid of bubbles. Lock it down..and presto.

And after all this, temps are still higher then with the generic silicon goo that I can obtain from any PC store.

Thanks in advance for any suggestions. This is driving me nuts after waiting a week for Ceramique to arrive only to have it not perform as advertised.

PS: No, I didn't grind too much off the CPU - it locks down nice and tight.
PSS: I cleaned off the silicon goo with xylene based carb cleaner, then wiped it with coffee filter so no silicon goo would interfere with heat exchange.
PSSS: Using Speedfan and Asus PC Probe to find temps.

 

[Mikewarrior2]

has your room temp gone up at all? Any increase in room/case temp will effect overall cpu temps.


Mike

 

[fdisk2003]

the first and last temp are both at night with the windows open. middle temp is during the day when it was hot. so really, the middle and last temp are probably the same.

which means 2 tries using ceramique its outperformed by generic goo using the same exact hardware setup.

 

[Elcs]

If Ceramique is like AS£, then let it cure.

Although Ive only reseated my sink about 4 times, Ive noticed my temps start at a level, increase slightly then lower over time. Worth a try imo.

 

[RalfHutter]

Pull it apart, clean all the Ceramique off, put it back together with the generic goop and see what your temps are, then you can see if it's the Ceramique that's responsible for your higher temps or if it's something else.

 

[citsacras]

 

[boyRacer]

I think Ceramique sucks when you have a wider area to spread because its much harder to apply it evenly... there was a review of it online where it didn't do too well with P4s but it did great with Athlons... along with the Nanotherm.

 

[Cerb]

Maybe you're doing too much work.
I got a difference, but a) with an Athlon, 2) only cleaning w/ 100% acetone, and 3) no lapping of perfectly good surfaces.

 

[Mikewarrior2]

How did you lap it? ANd i agree with the above post suggesting that you clean it all off and try your generic grease. It could be a problem from your lapping (way too common).


Mike

 

[fdisk2003]

it was lapped with 800 grit for both the generic goop and the first ceramique application.

the method i used was:

start with 400 grit wet/dry silicon sandpaper i got from an auto store. taped it to a flat surface (my kitchen counter). i put a few drops of water on it, and a drop of liquid soap to lube it good. i then made several passes with the HSF oriented in the same direction, then I turned the HSF 90 degrees and repeated. I did this several times until I got a uniform bottom. Then I switched it to 600 grit, then 800.

And using the same hardware, the generic goop was lower.. so then I try using braso to get a super fine mirror finish. I was really suprised how little work that took, I felt very confident. then I repeated the ceramique installation, this time trying to be much more careful. and all I got was a 2c difference in temps. the first temp (higher one) was during the day, the 2nd was at night. so really, I think there was no change - the change in temps was probably just due to the difference between day and night time air.

I will try to put the generic goop on and see how it works.

Should I try maybe applying the ceramique a little thicker on the CPU? What I mean, is Ceramique is supposed to be applied in a rice grain sized blotch on the center of the die - not spread evenly over it. It's supposed to spread itself with heat or whatever. Should I just do that part for it maybe?

The Ceramique site says nothing about temps being higher during the "setup period".. incidentally, how long is the setup period?

 

[randomboy]

Well..........heres an idea.

Ceramique = hot
Grease = cool

cool > hot

Grease > Ceramique

Put grease on, and enjoy. Yay.

 

[BZeto]

Originally posted by: randomboy
Well..........heres an idea.

Ceramique = hot
Grease = cool

cool > hot

Grease > Ceramique

Put grease on, and enjoy. Yay.

Exactley.

Just go back to the grease if ceramique isn't cuttin it...

 

[Mikewarrior2]

Let us know how hte regular grease works?

And what is your kitchen counter made of? is it an absolutely flat surface?

And i think you need more ceramique than a rice size drop? When you took off the Heatsink, did the ceramique cover the entire IHS? I believe the rice-sized drop is for socket-a /small contact surface CPU's.



Mike

 

[fdisk2003]

Well, I'm on day two of testing the 2nd Ceramique application. Getting temps as high as 45C with my Window and cases open. I don't think it's working out so I'm going to try re-applying the grease tonight.

When I removed the HSF this is what I saw:
<img src=http://www.arcticsilver.com/images/p4_cmq_dab2.jpg>
Basically a 1" diameter of where Ceramique "melted" (or whatever) and spread itself around evenly.

According to the guide, I'm applying this properly.. I just don't understand.

The kitchen counter top is a flat laminate. The surface I have now on my HSF is a very nice mirror like finish. I can put the brasso can next to it and clearly read it like it was a mirror. It is a very flat/uniform surface.

 

[MrThompson]

Brasso isn't helping the situation. It fills the microscopic surface imperfections where the thermal compound should be.

 

[l3ored]

Originally posted by: Elcs
If Ceramique is like AS£, then let it cure.

try some as 5 link

 

[RalfHutter]

Wait a minute! :Q

Your pic shows that you have removed the IHS and are (apparently) running without any thermal compound between the CPU die and the IHS? You can't do that!!! :Q

You need to have thermal coumpound at that interface too. Don't you remember the compound that was there when you removed the IHS? It still needs to be there now. You also need to clean up all the black glue from the CPU PCB and the IHS so the IHS will sit all the way down on the die. Scrape the glue off the IHS with a razor blade and try alcohol and some thing not-too-sharp on the PCB.

After that, apply some Ceramique to the top of the CPU die, reapply it to the top of the IHS too (but cover the entire thing evenly), reassemble the whole mess and recheck your temps. You should be a bunch cooler now.

 

[fdisk2003]

Nono hehe, that is a stock picture from the Arctic silver site. I didn't remove the IHS - I was just showing the 1" diameter white dispersal pattern of Ceramique across the top of the IHS.

Also after I applied Brasso to the HSF, I did a thorough cleaning with a Xylene based carb cleaner. So there shouldn't have been much remaining if any. Day 4-5, temps are 44-45c. Haven't had the opportunity to run it during the daytime heat, and I won't until this coming Sunday.