Intel IHS Removal

[largon]

I think I just killed my E6400...

[HaxR3]

Ouch :S

[Entsafter]

looool...
You should train with low-cost cpus -> celerons!!! It's not that easy when you do it the first time, but after a few ihs removals it's not that hard...
It would be better for your cpus if you sent it to me...

[giorgioprimo]

Intel Celeron D 356 w/IHS removed and completely cleaned:

did you use CrazyXP1700 way to do it? or did you find some alternative way?

[GordonFreemanLP]

Entsafter, so you're using the E6600 without HIS under wather and you didn't crap the core by pressing on it the WB? It's fantastic!

Ciao giorgio, cos'hai intenzione di scappellare te?? lol lol

[Entsafter]

Yes Gordon! The core can take more pressure than you think.. it's like a core of the old Amds... no problems. The only important thing is a cpu cooler with height adjustment -> screws and springs...

[largon]

The chip is dead as a brick.
Oh well, it was useless with the IHS on anyways for the chip was priming at 75ºC (CoreTemp) with water cooling.

Absolutely horrible IHS-die contact it seems.

btw,
Don't even bother using a butane lighter for the heating.
I emptied the whole lighter but the damn plate didn't move at all, these things just don't seem to generate enough heat to melt the TIM.
I had 2 package knife blades stuffed under the opposite corners but they merely warped the PCB.

edit: Hmmm... How about an E6600 next...

[soki]

The chip is dead as a brick.
Oh well, it was useless with the IHS on anyways for the chip was priming at 75ºC (CoreTemp) with water cooling.

Absolutely horrible IHS-die contact it seems.

btw,
Don't even bother using a butane lighter for the heating.
I emptied the whole lighter but the damn plate didn't move at all, these things just don't seem to generate enough heat to melt the TIM.
I had 2 package knife blades stuffed under the opposite corners but they merely warped the PCB.

edit: Hmmm... How about an E6600 next...

Mine 6400 also reach 75oC+(actually 80) with 1.47Vcore in coretemp during prime with lapped hsf and thermalright ultra 120. But it's stable @ 3400 mhz so no need to kill it.

[largon]

I finally popped (literally) the IHS of the presumed dead chip...

I bought a new lighter, stuffed the blades under two of the corners and heated the IHS while holding the cpu upside down. Suddenly the (I)HS just took off with a bang.

The core is flawless. I thought I had actually cracked some corners or partially ripped the die off of the package.
There's a slight possibility that it's still alive but just refuses to post due to the extremely warped package causing improper contact with the socket. The cpu is now clamped in a screw press in order to straighten the package.

Update expected within 24 hours.

p.s.
I'd like to elaborate that no prying whatsoever is needed during the removal. Just cut the black silicone and heat the middle of the IHS with a lighter while holding the cpu upside down.

Removing LGA775 IHS is actually very easy if done right.

Just remember:
DO NOT TRY TO PRY IT OFF

[CrazyXP1700]

Hence the 4 razorblades suspending the cpu... so it just falls off...

prying screwdrivers is too risky IMO... after killing 2 cpus that way...
i mean if your using percision screwdrivers... their still twice as thick as a razorblade, it will flex the package too much, possible seporate the DIE fom the PCB, and will almost definetly warp the package every time...

screwdrivers are not necessary... and too risky...

[menlatin]

it was pretty funny when i literally poppe dmine off as well. We were slowly tourhcing the top with the cpu suspended on the blades, and then all of a sudden a "POP" and it was done. It seems to me, because of the pop, that dudring installatin of the IHS, they must put ALOT of pressure in the center trying to get good contact, and the the flex of the IHS is still present when removing.

[largon]

Update expected within 24 hours.

It's still warped.
Think I'll tighten the screws...

Funny thing is that there's no visible damage but the damn thing just doesn't post... Shutdown after 3 secs.
The cpu does power up for it gets quite warm during bootup.

[freecableguy]

you'll kill the chip if you flex the substrate.....

[Stigma]

shamino pried upwards, wich lifted the core off, if he had twisted, he would have had no prob

This is true, but why risk it at all? Let gravity do the work! When the soldr is actually running, the amolunt of force needed to make it pop off is miniscule. If you use crazy's method, the force of gravity allong with the slight tension of the razor blades will pop it off, AND are garantueed to not get overzealous and apply too much force, as Sham did wrong =)

-Stigma

[michaelz]

I have a couple of IHS-less northwoods, and I have made shims using this. It seems to do the job.

I might have to make myself a shim or two. Running IHS-less Mobile Northwoods passive under a Ninja, ended up chipping the core after a few re-mounts. Lucky replacements are readily available on ebay.

Also tried IHS removal on a celeron D before this thread came along. Core came clean off attached to the back of the IHS.

[guess2098]

just kill one x6800 and one 630

yeaaaahhahahahaa

i use lighter and heating the ihs about 20 to 30 sec
and it just pop out!!

then i clean the top, put into mobo!!! .... hahaah no boot!
i feel so good now!!!!!!!!

core is perfect look tho hahahaha

[MentholMoose]

I removed the IHS on a 2.8C northy with a soldered IHS. I put the edge of a single thin razorblade under the IHS, held the blade with pliers so the IHS was face up, and heated the IHS with a heat gun for about 15 seconds. It popped off and fell a few inches onto my carpet.

It posts fine and there is 5c idle temp drop as measured by BIOS. I haven't tested further because I still need to remove the remaining TIM.

just kill one x6800 and one 630

I am so sorry for your loss. Please accept my condolences.

[Dani]

just kill one x6800

..sad to hear

[menlatin]

peeps, you NEED to use a torch. You can buy them at most hardware stores. you can use the ones that take the disposable tanks. Sometihng like this
http://cgi.ebay.com/Ace-Professional...QQcmdZViewItem

[MentholMoose]

peeps, you NEED to use a torch. You can buy them at most hardware stores. you can use the ones that take the disposable tanks. Sometihng like this
http://cgi.ebay.com/Ace-Professional...QQcmdZViewItem

Honestly, I'm too nervous to use a torch. I'm afraid that I might heat the CPU too much. I don't see why such extreme heat is necessary considering that the TIM melts at less than 100c.

A heat gun set to 550c worked fine on my 478 northwood (2.8C SL6WT). I finished cleaning it and it works fine at default settings (didn't test any OC yet).

[celemine1Gig]

just kill one x6800 and one 630

yeaaaahhahahahaa

i use lighter and heating the ihs about 20 to 30 sec
and it just pop out!!

then i clean the top, put into mobo!!! .... hahaah no boot!
i feel so good now!!!!!!!!

core is perfect look tho hahahaha

I also thought that I killed my CPU after removing the IHS. BUT, it is alright. You have to watch very carefully if the CPU and the heatsink do have proper contact. If either the CPU contacts don't get forced down on the socket pins with enough pressure, or the CPU-Die and heatsink don't have enough contact, then the CPU just won't boot. It's sometimes kinda tricky to find out. Sometimes it seems as if there would be enough pressure, but in reality it just isn't enough enough and you get no boot. I found out that my CPU was OK, after I had remounted the IHS and tested again.
Although you could indeed have destoyed the CPUs by heating them too much for too long. A normal lighter isn't really appropriate as it takes much too long to heat the surface.

[Kasparz]

Honestly, I'm too nervous to use a torch. I'm afraid that I might heat the CPU too much. I don't see why such extreme heat is necessary considering that the TIM melts at less than 100c.

A heat gun set to 550c worked fine on my 478 northwood (2.8C SL6WT). I finished cleaning it and it works fine at default settings (didn't test any OC yet).

550c are too hot. Just remember THG videos, when they burned A-xp's without cooler. They died after 300c temp.

[celemine1Gig]

550c are too hot. Just remember THG videos, when they burned A-xp's without cooler. They died after 300c temp.

The flame temperature of ~550°C is absolutely no problem, as you don't heat up the Die material directly, but the IHS instead. You just need to really carefully watch the time you spend heating it and of coure the IHS temp. You have to remember that although the flame may have a temp of 550°C, the IHS and Die temps won't instantly shoot up to that level, too. They will rise kinda slowly to the level that you desire. I used a little torch and that thing reaches flame temps of over 1000°C. Guess what, it worked well. Just don't point the flame anywhere else than directly on the IHS's metal surface.

[Methylphenidate]

Sorry to revive an old thread, but does anyone know what kinds of complications may arise if I were to attempt this on a Kentsfield ES?

[camouflage]

My advice is to wait until QUAD COREs overflow the marked - then C2D will be maybe ~$100 and the experiment will be less pricy........

Alternative I did this: