-

[prontospyder]

<<prontospyder ~ i bet that was scary wasnt it? :Q>>

Definitely....I couldn't believe the cooled off thermal pad was that strong. I know I'll be using Arctic Silver from now on.

 

[THUGSROOK]

DAPUNISHER ~ thx for test
would be nice to know if maybe they somehow fixed that bug. maybe they change the temp ratio or minimum speed.

R08 ~ if your cpu fan was a "sucker" id say yes that would be a good idea.
but if its a blower like the P4 retail ...that would be a bad idea. the fans would fight each other.
try just removing the fan like DAPUNISHER is and run some tests. the grille will still look kewl there

Evan Lieb ~ the guy that ripped his cpu out of the socket seemed pretty accurate :Q
we arent saying this will happen to everyone every time - just some possible warnings, tips and tricks here

AkumaX ~ friends dont let friends run stock

terminalmind ~ dont be teriffied - just make sure the HS is free from the cpu. a few carefull wiggles to be sure it aint stuck. heating it up 1st is a great idea

 

[lung]

For removing the thermal pad I would strongly suggest not using a razor blade as there is too much of a possibility of scoring the surface of the HSF. I used a piece of thin, hard plastic to remove the vast majority of the pad, then a white artist's eraser to remove the rest, then rubbing alcohol to get off the last bit of adhesive. Worked great and by doing all that, it even helped smooth out the finish of the HSF.

 

[SupermanCK]

Hee hee...funny thing is everytime i remove my heatsink...thermaltake volcano 7+ with ACIII...i always forget to do a couple of twists...but the retension clips are definitely not a good design at all...i always have a b!tch time trying to unclip it...do other volcano 7+ owner have the same problem?
anyway...after the long struggle...i always forget to do a couple of twist and the cpu is glued to the volcano...but nothing was damaged for the few times i had taken it off...lucky me...the lock machanism is still lock too...wheewww...

 

[paralazarguer]

I've reinstalled the same retail P4 HSF at least 10 times with the thermal pad. Nothing was damaged, and CPU temps on the 2.4A P4 I was using never got higher than 45C, and usually were around 39-40C.

I agree with this statement. I've personally removed at least a dozen retail P4 heatsinks with pad and none of them have been sticky in the least. the thermal pad is not dangerous. Using rail pollish remover and wd40 is a bad idea. it prevents thermal paste from properly filling in the gaps in the heatsink after cleaning is done. xylene based cleaner, (Goof Off and some carburetor cleaners) acetone, mineral spirits, or 99% pure isopropyl alcohol are the only things that should be used. The retail pad is not eh least big sticky and will not pull the processor out of its socket. Something's wrong if that's happening.

 

[SupermanCK]

i don't know about you 7757524, but when i take out my cpu with AC3 on...it gets pretty glued on to the heatsink...
i don't think the thermal pad is that sticky as like the AC3...and i dont' remember ever taking the cpu with the stock heatsink either...
just if anyone who uses ACx should pay attention when he/she takes out the heatsink...

 

[DAPUNISHER]

Well I've had it running since last night with the side case fan disconnected and the CPU fan RPMs are between 3082-3150 now and the CPU temp is 41-42c and the case temp is 31-32c almost identical readings except for the CPU fan RPMs being a little higher and the case and CPU temps fluctuating a degree higher or so. I think I'll leave it this way for now since the side fan seems to be superfluous with my case cooling as I have a top blowhole fan and front intake fan still running.

 

[fatbaby]

hmmm

i removed the thermal pad..applied aa and now mbm says i'm getting 260 f temps and bios says 40 but both of em play that warning msg!

Err and when i press down on the hsf, 2 mm of aa comes out on each side...am i putting too much?

 

[dbwillis]

Yeh, 2mm coming out when pressed down is too much...maybe way too much

 

[fatbaby]

so how much aa do i apply on the hsf? the arcticsilver website only has instructions for the chips that have the cores exposed

 

[gururu]

What a nightmare. I made the mistake of using the thermal pad for the first few days. When I got in some arctic silver 3, I tried to remove the thermal pad from a 'warm' cpu. Yes, it happened to me too. The cpu peeled out of the socket with the HSF unit. I also had several bent pins (fixed with a paper clip). Needless to say I almost cried... Anyway after cleaning up with IPA and a lint free rag, I put a thin coat of arctic silver on the die with a disposable shaving razor. I also colored the heat sink surface with the AS3. I broke in in for a few days, but only achieved a 2 C drop after AS3 fixing. I was a little disappointed.

As far as having a fan blowing on the heat sink. I don't know. Do whatever gives you a lower temp. Forget about the speed of the stock fan. I was able to achieve a 3 C degree drop by having a strong fan blow right onto the HSF. Needless to say my stock fan is still up in the 3500s. Could be because my idle is set 1t 49 C...

careful with the theraml pad it is a monster!

gururu

 

[paralazarguer]

What a nightmare. I made the mistake of using the thermal pad for the first few days. When I got in some arctic silver 3, I tried to remove the thermal pad from a 'warm' cpu

You tried to remove your HSF while it was still warm? You got what you deserved. The box for the P4 does say that it's meant to be professionally installed and presumably removed. Where is a source which states that the P4 fan is thermally controlled!? Mine's not. With a 10degree difference in case temp it spins the same speed. I can make it thermally aware using Q-fan in the bios but I don't believe the fan is actually speed controlled otherwise. If you have a source for this please point it out. I can't make it change speeds without Q-fan.

 

[WangButter]

Originally posted by: gururu
I broke in in for a few days, but only achieved a 2 C drop after AS3 fixing. I was a little disappointed.

It's probably your ambient / case temps. You may need to get some more ventilation in that case. Do you get lower temperatures with your case open? Also, what are your room temperatures like? The heatsink won't radiate heat very well if the air surrounding it is warm too.

 

[Antoneo]

After Gururu's post (and comments from other threads about the supplied thermal pad), I am coming to the conclusion that the difference in temperature between the AS3 and thermal pad is only a couple of degrees. I currently have a P4 1.6A still in its box and forgot to order AS3; now I am trying to convince myself that AS3 is not a requirement for overclocking (I plan to only go 133 FSB... :.

Has anyone recorded the temperatures with the thermal pad and with the AS3?

 

[THUGSROOK]

plans change

SupermanCK ~ an alpha can also rip the cpu from it socket - you gotta be very careful - and not use too much ASII - that makes it even tougher to remove from the cpu.

DAPUNISHER ~ good work
you just saved yourself a fan, some dust, and some noise

 

[paralazarguer]

I currently have a P4 1.6A still in its box and forgot to order AS3; now I am trying to convince myself that AS3 is not a requirement for overclocking (I plan to only go 133 FSB... :.

Has anyone recorded the temperatures with the thermal pad and with the AS3?

Good choice. You can run at the 133fsb with the retail cooler and retail pad no problem. I'm doing it as I type. According to sharkyextreme's northwood overlclocking guide here any 1.6a should be able to do this. Also, there are no fans in this case and it's fine.

 

[GotoDengo]

Damn, I must be doing something wrong:
1.6a 150 fsb @1.6vcore
retail hsf, as3
Case temp: 29c
CPU idle: 44c CPU full load: 58c

I've tried 3 re-applications of the AS3, which hasn't made for any improvement. Besides possibly being too thick a layer, what else might I be doing wrong? Is it possible the surfaces aren't contacting cleanly? How much would lapping the HSF help? Could my Epox 4g4a be over-reporting the temp?

I'm also wondering if I can remove, for cleaning purposes, the heat spreader that covers the P4 core. With all these re-applications of AS3, I'm afraid I've got some leaking into that tiny hole on top of the heat spreader.

thanks,

Goto

 

[RalfHutter]

Originally posted by: 7757524
Where is a source which states that the P4 fan is thermally controlled!? Mine's not. With a 10degree difference in case temp it spins the same speed. I can make it thermally aware using Q-fan in the bios but I don't believe the fan is actually speed controlled otherwise. If you have a source for this please point it out. I can't make it change speeds without Q-fan.

I've built about 5 or 6 P4 Northwood rigs and all of the retail HSFs have been thermally controlled. Before I deliver a rig to it's new owner I always run Prime95 and Memtest86 to make sure everything is stable. All of the retail HSFs (according to both MBM5 or Intels' "Active Monitor") run around 1450-1550RPM at idle. After heating up the CPU by running Prime95 ALL the HSF fans end up running between 2800-3300RPM. After I close Prime95 the HSF RPMs slowly drop back down to normal. I know this isn't conclusive proof that they are thermally controlled, but I believe they are.

 

[THUGSROOK]

RalfHutter ~ thank you - excellent info

 

[RalfHutter]

I recalled the old adage that "patience is a virtue" and kept searching Intels' site and I finally found "Conclusive Proof" that the Intel retail HSF fan is thermally controlled. Here's an excerpt of the relevant section:

"In addition to the Thermal Monitor feature, the boxed Intel Pentium 4 processor fan heatsink uses a high quality variable speed fan which allows the processor to remain within its operating thermal specifications by running at different speeds over a short range of internal chassis temperatures. The processor fan operates at a low speed while internal chassis temperatures are low. If internal chassis temperatures increase beyond a lower set point, the fan speed will rise linearly with the internal chassis temperature until the higher set point is reached. As fan speed increases, so does the noise that the fan generates"

Here's a direct link to the page. It's even got some neat charts and graphs showing the various set points.

 

[nemo160]

i was working on my friends p4 2.2/stock hsf/as2 or 3 and had it pull out of the socket on me (luckily no bent pins)
i think the reason the cpus pull out of sockets has less to do with the interface material and more with the sheer surface area of the p4 heat spreader
more area = more force required to break the surface tension
maybe socket 478 should be beefed up a little bit i.e. higher grip strength on the pins (not sure on the feasability of that)

 

[THUGSROOK]

even a small retention clip on the CPU PCB (what little there is) would be nice.

 

[SupermanCK]

i have to disagree on the stronger grip...that is because the pins on the CPU is really easily bent or break...i think it is good as it is now...
if the clip retenting was to be stronger...i think i would've broken both of my 1.6a's already...

 

[MrPeepers]

I replaced my 1.8A retail HSF tonight. Nothing fancy, some thermal grease (not artic) and a PC power and cooling HSF. The damn retail HSF ripped the CPU out and it was stuck on. I had to use a hairblower to warm it up and slide it off. Then I used finger nail polish remover to get the goop off. Once everything was working my load cpu temp went from 49-50C to 45-47C, not bad.

Currently at 2.38Ghz (132x18) with 3:4, so the DDR is pumping. I think my system is finally done, back to gaming instead of tweaking.

 

[THUGSROOK]

Originally posted by: MrPeepers
....The damn retail HSF ripped the CPU out and it was stuck on!

this is turning out to be an excellent thread guys - thx for all of your replies so far