I'm a dumbass

[pIII_Man]

I'm feeling like a dumbass right now and just thought i would let the world in on my stupidity. Today i was bored and was in a mischivous mood so somehow i put it into my head that i was gonna remove the IHS (intigrated heat spreader) from my cpu. The fact that i tried to remove it from a tualatin core celeron is beyond me as it ran super cool, but still i was a bit "out of it." Anyhow i went about doing it in a simple way. I put the cpu in a slocket (slot 1 to socket 370 adapter) Then i carfully put the IHS part of the cpu in vice, then i kinda rocked the slocket side to side until with some force the IHS finally twisted off...sucess!...Kinda. At first i was happy that i poped the thing off within 1 minute of pulling it out of my computer but then as i was cleaning the thermal paste off of the core (it was a nice core too, dark black hue without a tint of purple, definitly from the middle of the wafer) i relised that the paste (between the cpu die and IHS) was kinda stiff, it was a phase change type grease that only gets soft when warm. I finally got the grease off with a warm paper towel, and noticed i chiped a corner off of my nice 1.6ghz tully celleron :frown: . The corner of the cpu's core was still embeded in lump of thermal paste on the bottom of the IHS. I suppose the moral of the story is, if it ain't broke don't fix it and if you do want to do something about the IHS just lap it, it is a nice thick slab of copper and really won't effect temps much.

Anyhow i figured i would share my pain with you guys and hopefully someone can learn from my mistake.


If it isn't a P6 then it isn't a procesor
110% BX fanboy

 

[HardWareBoss]

Did you learn your lesson? IF it ain't broke don't mess with it.

 

[pIII_Man]

ya i did, the hard way hehe


If it isn't a P6 then it isn't a procesor
110% BX fanboy

 

[HardWareBoss]

LOL

 

[Crashman]

You should have asked me first! I've SUCCESSFULLY removed an IHS before and found the results of running without it were negligable.

<font color=blue>Only a place as big as the internet could be home to a hero as big as Crashman!</font color=blue>
<font color=red>Only a place as big as the internet could be home to an ego as large as Crashman's!</font color=red>

 

[gaviota]

[-peep-] happens...I once dropped a hard disk from a PC that belonged to my boss because I was working in a small table that was full of stuff. Felt like a dumbass too.

__________________________________________________
It's not important to know all the answers, as long as you know how to contact someone who does.

 

[Kelledin]

For future reference (in addition to Crashman's comment)...

Most thermal epoxy will go brittle if you leave it in the freezer for a few hours--just seal it in an anti-static bag first. Then it will break easier without taking any of the CPU core (or GPU oxide) with it. I've successfully used that trick to change the HSF on my vidcard. Only thing the old epoxy took with it was a trace of the nVidia logo print.

(This doesn't work on Arctic Silver epoxy though, unless you mixed it with regular AS3 paste before applying.)

<i>I can love my fellow man...but I'm damned if I'll love yours.</i>

 

[ksoth]

How big of a chunk was knocked off? A Duron 700 that I had got all grinded and chipped along the edges from removing and replacing heat sink several times while switching motherboards, and the thing still works. Have you tried the chip to see if it works?

-------------------------------------------
<font color=blue> "Trying is the first step towards failure." </font color=blue>

 

[lhgpoobaa]

LOL Twit

<b>I am not a AMD fanboy.
I am not a Via fanboy.
I am not a ATI fanboy.
I AM a performance fanboy.
And a low price fanboy.
Regards,
Mr no integrity coward.</b>

 

[pIII_Man]

Naw this thing doesn't even post. Well i am gonna get a celley 1.2 now, and see what happens.


If it isn't a P6 then it isn't a procesor
110% BX fanboy

 

[pIII_Man]

What method did you use to remiove it?

Were you using a copper cored heatsink?


If it isn't a P6 then it isn't a procesor
110% BX fanboy

 

[HardWareBoss]

No new 1.6 Telley?

 

[HardWareBoss]

Correction Tully.

 

[pIII_Man]

i overclocked the 1.1 to 1.6...i was actually not limited by the cpu but rather the agp bus speed.


If it isn't a P6 then it isn't a procesor
110% BX fanboy

 

[HardWareBoss]

How far will the 1.2 go?

 

[pIII_Man]

146 x 12 at least i hope


If it isn't a P6 then it isn't a procesor
110% BX fanboy

 

[Crashman]

I cut the rubbery glue (latex cement?) all the way around the edge with a razor, then gently pried the cap loose. I was using a Thermaltake Volcano II, no copper core.

<font color=blue>Only a place as big as the internet could be home to a hero as big as Crashman!</font color=blue>
<font color=red>Only a place as big as the internet could be home to an ego as large as Crashman's!</font color=red>

 

[pIII_Man]

well if you are not using a copper core heatsink i can see why the temp diffrence was marginal, but i think with a copper core heatsink that on its own can spread the heat to the rest of the heatsink, the cooling performance should be a good bit better as the copper core will act like a heat spreader.

Looks like you used the most common way of doing it, i would have done the same except i could not get my screw driver under any of the gaps.


If it isn't a P6 then it isn't a procesor
110% BX fanboy

 

[Crashman]

Actually I was one of the first to do it, back when I was experimenting with a Celeron 1.2 that wouldn't do 133MHz bus. I used the edge of the razorblade to pry it up, not a screwdriver.

<font color=blue>Only a place as big as the internet could be home to a hero as big as Crashman!</font color=blue>
<font color=red>Only a place as big as the internet could be home to an ego as large as Crashman's!</font color=red>

 

[pIII_Man]

lets hope mine fares better...

oo BTW a freind of mine game me a mobo, its an asus cucl2-c I'll try that out to see if i can run a higher bus speed than what the bx limited me to. I have heard the cusl2 fares well with tualatins after doing the little pin isolation trick and connecting ak4 (i beleive) to VTT.


If it isn't a P6 then it isn't a procesor
110% BX fanboy<P ID="edit"><FONT SIZE=-1><EM>Edited by piii_Man on 10/22/03 10:11 PM.</EM></FONT></P>

 

[Crashman]

What about your experiment to feed an AGP bus it's own clock signal?

<font color=blue>Only a place as big as the internet could be home to a hero as big as Crashman!</font color=blue>
<font color=red>Only a place as big as the internet could be home to an ego as large as Crashman's!</font color=red>

 

[Guest]

Captain Obvious gives piii_man a big reward!

What the reward is should be obvious!

<b><font color=red>Captain Obvious To The Rescue!!!</font color=red></b>

 

[pIII_Man]

Well i found the 440bx has trouble running an asynchronous agp bus. I actually did get it to post but even in the bios screen there were some "anomalies" do to what i beleive is data corruption because the bx cant run async. I went about locking the bus by using a hex inverter and a 66.6 mhz 3.3v oscillator and a .1uF capacitor. The 66.6mhz oscillator was fed into 3 inputs of the hex inverter. On the output side of the Hex i fed the inverted signal to the GCLKO pin(graphics clock out) on the chipset, GCLKI (graphics clock in) and the final input went to the CLK pin of the agp card. In order to effectivly "inject" the clock signals i had to remove 2 resistors from the the pcb so i wouldnt get interferance comming from the signals the BX was putting out to the gpu.

I was thinking of using a clock divider chip and running the agp at 1/2 of the fsb thus keeping some sense of synchronism (if that's a word) between the 2 of them, however now that i got this new board its not worth the trouble. No matter how bad the 815 performs a cpu at 146mhz fsb will always run better than one at 110mhz fsb. The 815 should alow (with the 1/2 divider) 150mhz fsb before my agp card starts freaking out.


If it isn't a P6 then it isn't a procesor
110% BX fanboy<P ID="edit"><FONT SIZE=-1><EM>Edited by piii_Man on 10/22/03 10:34 PM.</EM></FONT></P>