HOW TO REMOVE i7 IHS

Hi guys. Today, my 4th i7 920 arrived. So here’s the guide how to get the pants off this baby.

First this is what you need besides the CPU:

1.razorblades
2.adhesive tape
3.permanent marker
4.Lighter / gas soldering iron or anything like this
5.something to clean the HS after burning (acetone + hankies)
http://www.abload.de/thumb/dsc00246tr31.jpg
If you have everythig collected, the show can begin.

Step 1 – prepare 6 razorblades for use
-->use some of the tape to protect your fingers
http://www.abload.de/thumb/dsc00247v0sx.jpg

-->mark one of the prepared blades with the waterproof pen as shown in the picture. The cutting depth is max. 4mm for the i7! If you cut deeper, you may damage some of the small ceramic caps.
http://www.abload.de/thumb/dsc00248qrmd.jpg


Step 2 – Cutting
-->Take your marked razorblade and cut from the corners to the middle (each side two cuts from the right corner to the middle, then from the left corner to the middle = 8 cuts around the CPU)
http://www.abload.de/thumb/dsc00249bjg5.jpg

-->now cut the rest like this (each side one cut = 4 cuts, all in all 12 cuts)
http://www.abload.de/thumb/dsc00250mvew.jpg


Step 3 – get ready to burn
-->We’ll make a little fire under the IHS, but nobody wants to roast the PCB. So do the following:
http://www.abload.de/thumb/dsc00251r7qi.jpg

-->the melting pot (if you heat up the IHS, the solder melts and gravity will do the rest)
http://www.abload.de/thumb/dsc00252k6co.jpg

Step 4 – Melting the solder
I’ve done this with a simple gas lighter. Sometimes you need to shake a little bit as you can see in the video.
Videos say more than thousand pictures:
http://www.youtube.com/watch?v=YWTgS2ZuiJQ

After this, clean the Heatspreader using acetone

So heres the result, tested & works:
http://www.abload.de/thumb/dsc00257efdj.jpg
This is a photo of another i7, wich is currently running in my rig:
http://www.abload.de/thumb/dsc001719pek.jpg

Carefully remove the tin with your razorblade.
Have fun with your naked i7 !;)

How the story continues: The custom-waterblock.

Copper 40x15x1000 vs. HS
http://www.abload.de/thumb/dsc001748ugc.jpg


What was your i7 Batch?
http://www.abload.de/thumb/dsc00177dqis.jpg
http://www.abload.de/thumb/dsc00178atwt.jpg
http://www.abload.de/thumb/dsc00180a8q6.jpg

The copper met a milling machine, result: 29x38x12 + two holes
http://www.abload.de/thumb/dsc00181z5mm.jpg
threaded (G1/4 inch thread)
http://www.abload.de/thumb/dsc00245zwy8.jpg

I think you can imagine the leftover.
Performance:
- The water runs directly across the core now. CPUTemps are max. 30°C @ 4GHz full load, wich allows a nice 24/7 OC @ 4,4GHz; 1,39VCore (selected CPU, limited by the max. stable BCLK of my DFI LP UT mobo with HT&TurboMode on) :cool:

Regards from Germany,
Frank

30°C under full load with water? and that's with HT on? That's insane :shocked:

Quote:

---

Originally Posted by ghost_recon88

30°C under full load with water? and that's with HT on? That's insane :shocked:

---

Thats XS;)

I'll post some screenys of the rig when ready. Next stop: testing 8SSDs@Areca 1261ML on it.

Clearly not for the faint of heart!!.

Really nice :up:. How did you manage to seal the waterblock watertight to the core?

What are your idle temps? And that's with chilled water I'm assuming.

I soldered the IHS onto my waterblock. After this glued it on the CPU (two-component epoxy resin)

Yep, I'm using a "chiller". It cools the water down to 20°C. Idle Temp is 21°C.

Ok, thanks. I like how on the top of this forum there is a big warning not to pop the IHS and what do I see some threads lower……:yepp:

Quote:

---

Originally Posted by corross

I soldered the IHS onto my waterblock. After this glued it on the CPU (two-component epoxy resin)

Yep, I'm using a "chiller". It cools the water down to 20°C. Idle Temp is 21°C.

---

Let's see some more pics of the whole set-up. :yepp:

Quote:

---

Originally Posted by Conquistador SW

Ok, thanks. I like how on the top of this forum there is a big warning not to pop the IHS and what do I see some threads lower……:yepp:

---

http://www.xtremesystems.org/forums/...d.php?t=209279
Yes, this is a great sample of craftsmenship. Looks like boble used a wrecking bar...

Quote:

---

Originally Posted by [XC] Hicks121

Let's see some more pics of the whole set-up. :yepp:

---

No problem, but I have to complete the rig first, like I mentioned before. The SSDs are on the way. Moreover, I'm building water blocks for the DFI, because the Mosfets get quite hot.

Quote:

---

Originally Posted by Conquistador SW

Ok, thanks. I like how on the top of this forum there is a big warning not to pop the IHS and what do I see some threads lower……:yepp:

---

Made me lol too


Sticky: Intel Core i7 - DO NOT REMOVE IHS! ( 1 2 3 4 5)
.
.
.
HOW TO REMOVE i7 IHS

Damn, direct-die watercooling. You have balls lol!

Quote:

---

Originally Posted by ownage

Damn, direct-die watercooling. You have balls lol!

---

The core aint that fragile that people tend to believe. I even lapped the core of my C2D E6600, and not a little either.

Wow--did you happen to have pre-removal tests for comparison? Because 30C... that's SWEET!!!

Before this mod, the CPU temp was around 45°C with the chiller @ 20°C. So it dropped 15°C. :)

FINALLLY i have been wanting to see someone do this with an i7.
Awesome job!!!!

I'm going to try this over the weekend :D

That's awesome dude. Popping a soldered IHS is quite a feat in itself but that direct-die water cooling is even better. Can't wait to see them pics of the whole setup!

Awesome.

Very nice :up::yepp:

Thats not eXtreme.. thats totaly insane. :D And that video .. wouldnt like to see what could happen if flame met CPU directly.. btw. how can CPU survive that much heat? (Yes I know, since it was soldered, it must been done with some heat too..).

Im tempted to try it too.. but its kinda expensive fun.

you have to remember intel puts these chips in an oven to start with. Cant remember the temp but its pretty hot. In that vid right as the IHS popped off it looked like the CPU exploded!
Awesome work. Might have to give this a shot. got any hi res shots of the Die BTW?

Thats awesome :up:

Can't wait to see the cooling setup ;)

Can you run the core naked under a regular Heatsink on air or you need some shims?

Quote:

---

Originally Posted by Fafeifa

Can you run the core naked under a regular Heatsink on air or you need some shims?

---

Should be able to...

By the way, how did you clean the core from the solder?

Quote:

---

Originally Posted by dejanh

Should be able to...

By the way, how did you clean the core from the solder?

---

I haven't done it yet, now evaluating the risks before getting my hands dirty. :shrug:

Quote:

---

Originally Posted by dejanh

Should be able to...

By the way, how did you clean the core from the solder?

---

Would appear he used acetone to remove the solder right after he heated it up to remove the IHS to begin with.

Idle of 21*C and 100% load of 30*C...we 'air-cooled' people can only dream...

So if I successfully remove the IHS on my i7, I would need to remove the latch from the cpu socket I guess?
The one that keeps the cpu in place I mean...

Thanks for your feedback!:bday2:

Quote:

---

Originally Posted by liftedcj7on44s

... got any hi res shots of the Die BTW?

---

I'll add some to post #1.

Quote:

---

Originally Posted by ChaosMinionX

Would appear he used acetone to remove the solder right after he heated it up to remove the IHS to begin with.

---

The acetone is only used to clean the Heatspreader. Take a razorblade to plane the solder tin from the core. Intels tin is malleable @ ambient temperature.

Quote:

---

Originally Posted by Fafeifa

Can you run the core naked under a regular Heatsink on air or you need some shims?

---

It depends on your "regular" heatsink. Intel-boxed-cooler=no way because of the push pins. A DIE is fragile like porcelain. If you concentrate high pressure on a small area, it'll break. You should prefer a cooling solution with 4 bolts or anyting like this to press the cooler firmly down to the cpu surface. For example just mod the boxed cooler (4 skrews M3x40, 12 plastic washers,4nuts, 4 knurled nuts;))

BTW: I don't think, that an IHS removal makes sense for an air cooling setup. Get a nice cooler and sand the IHS down.
http://www.caseking.de/shop/catalog/...er::11849.html

Quote:

---

Originally Posted by Garrett

So if I successfully remove the IHS on my i7, I would need to remove the latch from the cpu socket I guess?
The one that keeps the cpu in place I mean...

---

:yepp:Removing the latch isn't necessary in most cases. Look at the bottom of your Heatsink if it fits.

Just insane!

Quote:

---

Originally Posted by Garrett

So if I successfully remove the IHS on my i7, I would need to remove the latch from the cpu socket I guess?
The one that keeps the cpu in place I mean...

---

Lol the increased possibility of bending those pins makes me more queazy than removing the IHS.

A very good guide. Thanks. :)

damn :D

very nice how to :D but i won't be doing this i think lolz :D

would this be good for phase? excellent guide

Oh.... This is just too AWESOME, you are awesome.
And it would be even more awesome if you could put up a alittle more detailed guide of the "How the story continues: The custom-waterblock."-part of the guide.
I'm really tempted doing this myself :)
So if you dont have more pics, you can maybe atleast try explaining more like "step-by-step".

I hope that its not all too much expected :)

I have already posted this step-by-step guide here. Believe it or not: one day later, XS was down and my posts were lost. :clap:
Thanks to your awesomeness, zerophax, I am working on a new guide now. Will post the link in a few days.

To the XS-server owners: STOP OVERCLOCKING YOUR RAID CONTROLLERS, WE NEED A STABLE FORUM!:D

This is so insane, I would be too scared to try myself, you got ballz dude :up:

Quote:

---

Originally Posted by Bobbylite

would this be good for phase? excellent guide

---

lol i dont think so...condesation on a ihs i 1 thing,but on a direct die...?....water+direct die = no good i think.chilled waterblock should be just fine.


i have cracked naked northbridge chips before,same things as cpu naked i would imagine.my mcw-30 cracked the corner right off of my old dfi 680i.must VERY careful tighting anything down on something like this,slow and even.1/8 in tire lug patern turns only til snug,then done..no more.


i am impressed with your thread dude.tho imust admit my new C0 stepping e8400 with my new multi core(mc)dd tdk block at 30c 1.28 volts 4ghz keeps me content.tho....20 c would be nice =) i am certian i could pop mine off and put my new dd block on it being its 4 screw turndown setup and i know the miamal pressure and method to tighten without damging now.


now a phase change head on direct die i would imagine from a critical pressure point would crack it instantly because the second u gently tightend down on the corners the actual phase "neck" with all the plumbing,wiring,thick foam insulation etc would pull 1/8 inch one way as soon as u let go and .....crack!

So... How much do I have to shell out to get my hands on a direct-die WC block of my own? :D

Seriosly man, WOW. I wish I could get my Xeon running 30c load @4

:shocked:

might try this. it doesnt look like its dificult

wow, props! You're much braver than I could ever be. Wow... that's all I got to say. :D

Quote:

---

Originally Posted by railmeat

lol i dont think so...condesation on a ihs i 1 thing,but on a direct die...?....water+direct die = no good i think.chilled waterblock should be just fine.


i have cracked naked northbridge chips before,same things as cpu naked i would imagine.my mcw-30 cracked the corner right off of my old dfi 680i.must VERY careful tighting anything down on something like this,slow and even.1/8 in tire lug patern turns only til snug,then done..no more.


i am impressed with your thread dude.tho imust admit my new C0 stepping e8400 with my new multi core(mc)dd tdk block at 30c 1.28 volts 4ghz keeps me content.tho....20 c would be nice =) i am certian i could pop mine off and put my new dd block on it being its 4 screw turndown setup and i know the miamal pressure and method to tighten without damging now.


now a phase change head on direct die i would imagine from a critical pressure point would crack it instantly because the second u gently tightend down on the corners the actual phase "neck" with all the plumbing,wiring,thick foam insulation etc would pull 1/8 inch one way as soon as u let go and .....crack!

---

What is the issue with phase? This thread shows direct water cooling on die to begin with... I'm not sure what you're warning against ;)

VERY nice results man!

I mean it :)

Sorry for bumping this thread, but where did you get these razor blades?

Ltd. edition, no longer available:lastweek:
The Croma razor blades provide a soft and smoothing shave and fit all Intel processors.:yepp:

Quote:

---

Originally Posted by corross

Ltd. edition, no longer available:lastweek:
The Croma razor blades provide a soft and smoothing shave and fit all Intel processors.:yepp:

---

So this guide is identical to what is necessary to remove the IHS for my QX9650?

Sweeeet!

I think you can remove the IHS from pretty much all CPUs. Not sure tho..
Then again there is always this
http://www.xtremesystems.org/forums/...d.php?t=209279

Quote:

---

Originally Posted by Fafeifa

I haven't done it yet, now evaluating the risks before getting my hands dirty. :shrug:

---

I used steel wool to clean the solder from the die

Insane, simply insane.

I must try this with my new TEC+TRUE cooling.

@OP nice work m8.....

thanks for the info....

Would that significantly reduce temperatures using air cooling?

Quote:

---

Originally Posted by zalbard

Would that significantly reduce temperatures using air cooling?

---

see page 2 up top he answers that question I believe

Quote:

---

Originally Posted by Mescalamba

Thats not eXtreme.. thats totaly insane. :D And that video .. wouldnt like to see what could happen if flame met CPU directly.. btw. how can CPU survive that much heat? (Yes I know, since it was soldered, it must been done with some heat too..).

Im tempted to try it too.. but its kinda expensive fun.

---

Heat isn't what kills a CPU, it's electromigration and for that to happen you need to have it up and running ;)

Quote:

---

Originally Posted by Smartidiot89

Heat isn't what kills a CPU, it's electromigration and for that to happen you need to have it up and running ;)

---

Well if you melt it, im sure it wouldn't be working as well ;)

You are a very brave and skillful guy! Well done! I don't have the nuts for this!

Truly awesome work I'm impressed you had the guts to do it! :up:

Made a thread at a Swedish forum about this thread just so you know.
http://www.sweclockers.com/forum/sho...hreadid=866224

Here is fallwind's link that tried direct die watercooling, may wish to read that as well. It increased load temps by a huge amount in addition to seeing a 16C difference between cores, and he offered a screen shot at stock settings and load. And btw, that is what one would expect.
http://img36.imageshack.us/img36/4885/loaded.jpg
http://www.xtremesystems.org/forums/...d.php?t=225354

And another thread on direct die.
http://www.xtremesystems.org/forums/...d.php?t=227964

Quote:

---

Originally Posted by rge

+1...which is why passive direct die cooling wont work on i7, I drew a pic back when arguing with someone else ...there is no getting around the very small surface area of die, and .58 W/M*k heat conductance of water is not going to dissipate 180W. I could see IHS/waterblock combo without thermal paste, ie soldered as doing better, but not direct die.

http://img188.imageshack.us/img188/1...rectdiepst.jpg

---

i think you can't argument with the heat conductance of water / copper...
You have to mind that the water comes with high speed and only stays for a very short time directly on that cpu. So take a stick of copper and heat one side of it... Don't you think the waterflow will be faster than the heat conducting through the metal?

The fakt is, that direct die watercooling offers continuously new cold material (water) to dispate the heat. Copper heats on and on, there is no new material coming... heat have to to through the material.

I think direct die cooling only works if the water comes with a quite fast an continuous stream over the die. :)

Example:

Try to brun a hole in a plastic bottle filled with water. Think you can't...
Try to burn a hole in a bottle filled with copper :D - or a brick of copper packed in plastic... i think you will know what i'm talking about...

Don't forget, while Copper provides "666x" faster heat transfer then just the water, the performance will be judged by the poorest component not the best, in this case by either the core ->TIM-> IHS application, or the IHS->TIM-> waterblock.

This simplifies the process from "core->TIM-> IHS-> TIM->copper block -> water" down to just "core->water". Though all our theorizing doesn't take away from the fact that the OP is getting crazy good temperatures that none us water or air cooled guys could dream of.

Pull your IHS off a core i7 and prove us wrong (not that hard, I have removed soldered IHS's before) , only shoot a youtube video of it working, if you want to seemingly defy physics and expect belief, you have to show proof. I have youtube videos of a thermocouple in IHS and one directly in the die via drilled holes in IHS and drilled into die, showing actual gradient through IHS and dies, posted in realtemp thread. So you can easily make a video after you pull the IHS off showing both the setup and temps/screen all running and prove it works...screenshots are easy to fake, not that I have even seen a screenshot of any successful direct die cooling.

I am not pulling IHS off on my i7, as I know what will happen when trying to merge 666 lanes of traffic into 1 lane. Yes the cars could all drive incredibly fast...but speeds would be unrealistic. If you think you can compensate by theoretically running water fast enough, I would suggest you start with this pump . As to how to find a waterblock that wont rupture or keep die from cracking under that pressure, I will leave to you.

trueg50, as for simplifying the process down to core>water, you are ignoring surface area.

Direct die = 10,000 cars (heat) in parking lot A drive in single lane to parking lot B (water) at 1 per minute (represents slow .5w/m*k transfer water). That is simplest transfer via shortest route, and the rate limiting step occurs immediately (but low surface area for transfer).

IHS/waterblock = 10,000 cars in parking lot A drive in 160 lanes or 160x fast (160x represents rate tim1/solder vs water) then those speed up to 300 per minute (copper IHS) as lanes widen/go faster, then across tim2 at now still 50X faster than water (surface area increase x tim2 (3-6W/m*K), then through copper water block again 300x fast, then finally through 1 per minute now slow water, but x 1000 lanes (massive surface area of waterblock copper pins) = 1000 per minute to lot B, ie here is where surface area disagrees with your rate limiting arguement being same regardless of where it occurs.

IHS/Waterblock heat (cars) has longer to go to get to parking lot, but guess which parking lot empties faster?

Nice to see that the core is still intact, good job!

i'll bet the heat transport in massive copper is much slower than in a fast water stream... So this direkt di cooling will beat a 5x5x5m cube of copper sitting on your cpu ;)

In your opinion, a double size copper water block shoud be much more effective than a smaler one. I think pure copper limits the heat with could be taken away.
Think of heatpipes, with fluid material inside... This dosn't make sence with your arguments.

So we have a increase in the ability of taking heat away:

pure copper -> copper combined with fluid or gas (moving molecule, atoms, whatever) inside (heatpipe) -> direkt fluid contact at high speeds.

good-> better-> best

And you have to think that the gas / fluid inside a heatpipe is much slower than in water cooling cirquit...

You forget about surface area, that is needed to transfer heat to the water. Ergo: the most effective way of cooling i7 is with a top performing block (good flow rate, large surface area with lots of pins etc.) directly on the CPU.

Also, you cant compare heatpipes to pure copper or water alone. Heatpipes work on the principe of condensation.

I suggest you read the rather good thread in the watercooling section about this too. But it seems that you simply ignore the test quoted above anyway.

Quote:

---

ut it seems that you simply ignore the test quoted above anyway.

---

and so you ignore the test in post#1?

Why did he get very low temperatures with direct touch? Think Frank should offer a few screenshots and pics from the hole setting...
This is no easy topic. Somewhere, there have to be a maximum of cooling performance with water. The more surface the better temps? Perhaps the minimum of copper compared with the most posible surface is the solution, don't know.

Quote:

---

Originally Posted by Krautmaster

Perhaps the minimum of copper compared with the most posible surface is the solution, don't know.

---

That is pretty much what everyone has been trying to tell you, and that is pretty close to the current solution of IHS/waterblock. If you want to try to better the current solution, solder/attach a copper waterblock directly to die, so bottom of copper waterblock becomes the heat spreader and eliminates tim2. Only post a video of it working this time as might have some positive results.

So did anyone try removing IHS for air-cooling setup and running it that way? I believe the chip should be able to handle even the normal case's vertical mounting if you attach the heatsink properly.
I am more concerned if one gets better temperatures that way...
http://www.xtremesystems.org/forums/...d.php?t=223542 Someone get a hold of this guy! :up:
I am getting i7 next month, and wondering if it's worth trying.
I don't believe noone tried... Perhaps, if not air, then just a normal (not direct die) water setup? Results should be similar... Anyone? :)

Edit: one more local linky, direct die cooling and way worse results... http://www.xtremesystems.org/forums/...84&postcount=2 And I believe it would be even worse for normal air or water...

I understand that direct die most likely wouldn't be the best but surely removing the IHS and mounting a waterblock directly on the core would help alot :)

rge i think that you forgot that there is no material that will take 100% of heat and spread also 100% of it. if that thing would exists we didn't have to pay energy bills :D

more things on the way of heat is worse compared to direct die watercoling. think of it. heat have to go thru IHS, thermal paste, bottom of water block, and what are the results? you are colling a 55* buttom of waterblock (when cores are 80*) with ambient water. what will be better? if your theory about 666x better colling with standard waterblock would be true nobody would care about phase change lizquid nitrogen or whatever, everybody would just sit on AC bc of more copper... nonsense

you guys hold up to your name. nice job

easy work...thanks man

http://i107.photobucket.com/albums/m...6/DSC05077.jpg

Let us know your temps!

lol, I just noticed.

The "Intel Core i7 - DO NOT REMOVE IHS!" thread is no longer sticky. :rofl:

results please! :)

Quote:

---

Originally Posted by zalbard

So did anyone try removing IHS for air-cooling setup and running it that way? I believe the chip should be able to handle even the normal case's vertical mounting if you attach the heatsink properly.
I am more concerned if one gets better temperatures that way...
http://www.xtremesystems.org/forums/...d.php?t=223542 Someone get a hold of this guy! :up:
I am getting i7 next month, and wondering if it's worth trying.
I don't believe noone tried... Perhaps, if not air, then just a normal (not direct die) water setup? Results should be similar... Anyone? :)

Edit: one more local linky, direct die cooling and way worse results... http://www.xtremesystems.org/forums/...84&postcount=2 And I believe it would be even worse for normal air or water...

---

I'm not sure if it's much of a comparison, but I have put naked socket 939 chips on stock coolers and aftermarket coolers without too much trouble.

It doesn't hurt to be TOO careful. I just wouldn't trust any coolers that use the push-pins. Those mount with a ton of uneven pressure since it's a full mount every time you push a pin in. With screws, you can give a small turn to each of the 4 one-by-one... so much safer.

I wouldn't be afraid to mount an aftermarket air cooler... that's if I had the $$ to afford an i7 and then taking the risk to pop off the IHS. :) I haven't had an abundance of money since the 939 days.

By the way, GREAT TUT!!

Quote:

---

Originally Posted by prznar1

more things on the way of heat is worse compared to direct die watercoling.

---

Which is why you should take of your IHS, and do direct die water cooling and prove your theory. No point in arguing what you can prove correct? So look forward to seeing a video of your direct die cooling system with better temps. Because if you honestly believe that, is it very easy to do. Takes about 15 minutes at best to remove IHS on soldered die.

Or are you going to be yet another poster who claims they can better temps with very little surface area of water, but yet fails to prove it.

Funny how all the people that posted any kind of proof after removing the IHS, got worse temps. Also corross says he got better temps, not using water but nanofluids with very high concentration of cnt, so seems to me he thinks water will not work either.

So can you point to even one individual who did get better temps with direct die on 100W plus heat load cpu using water... that even bothered to fake a screen shot, yet alone post a video?

Also regarding your theory of more things in way of heat make it worse, guess I need to stop using tim, it is just more things in the way, especially since tim is only 4-8 W/m*K heat conductance. Also cut fins off aircooling, fins are just more copper in the way.

Or if you want to try something that might work, remove IHS, solder and figure out how to reflow indigo xtreme solder like paste to connect waterblock to die, so bottom of the heat sink serves as IHS, that might actually lower your temps, assuming it does not damage the die.

it survived the ihs removal but didnt the mounting...should of took more time...but im just to lazy

it was a week 39 and had high temps on water couldnt even clock it to 3.5 the soldering had some holes in it but cant say if it was from the removal or not

Huge bump on this thread. But while reading it I had an idea which people never tried for removing the IHS. Why not using some chemical, that reacts and dissolve solder, but not silicon? So all that it would be needed would be drilling a little hole on the IHS, or inject it with a needle below the IHS(removing the glue first).
I think there are plenty of acids able to do that. Silicon is unaffected by most acids, including sulfuric acid.
That would be definitely less dangerous to the CPU than desoldering the IHS if the PCB is also acid resistant. If it's not I think it would still be feasible be slowly removing the IHS area over the die and adding acid until all the solder from the die is removed, without dropping any acid on the PCB.

Forget about removing it, it's not worth it unless you are going to solder the waterblock directly to the die.

just got my IHS removed whats the best way off cleaning off all that solder?

Well finished polishing down to the actual core too bad though my waterblock doesn't reach it even with the IHS back in place its roughly 55C in bios. Looks like I need to go out tomorrow and find some kind of copper shim.

Quote:

---

Originally Posted by =[PULSAR]=

Well finished polishing down to the actual core too bad though my waterblock doesn't reach it even with the IHS back in place its roughly 55C in bios. Looks like I need to go out tomorrow and find some kind of copper shim.

---

Remove the bracket on the mobo and use a block that has threaded bolts all the way down so you can control the pressure very safely and evenly.

I'll have to look in the morning but isn't the bracket riveted onto the board usually?

Quote:

---

Originally Posted by =[PULSAR]=

I'll have to look in the morning but isn't the bracket riveted onto the board usually?

---

The top part comes off, I can't post from there at work (blocked site) but if you go to www.xcpus.com and look for Sirrobin4ever's thread, he removed the IHS and got some very good results with water and the bracket removed.

Very impressive but seriously, this just looks like its to much of a pain and a high risk factor. just Lap it a bit and call it a day.

Quote:

---

Originally Posted by RAMdHERsoftware

Very impressive but seriously, this just looks like its to much of a pain and a high risk factor.

---

Welcome to XtremeSystems :p:

is this possible with an i7 3930K ?

Try it and let us know :)

nothing is impossible :D

I am planing to remove the IHS from my Sandy Bridge 2600K. I will remove the bracket from the Mobo and use other Srews to use my Waterblock direct on the DIE - I am doing it right now, with the IHS. I am planing not to use a lighter after cutting the edges, I want to use a flat iron. I think I get enough heat to melt the Iron Solder and flip the IHS down. But what is the best thing to remove the rest solder from the DIE without destroying it. Direct after removing the IHS, is the rest solder on the DIE warm enough to take it up - with what?

Sandpaper (Maybe 800/1000 grit).

I don't think you will be successful, you will most likely kill the chip. It isn't as simple as just heating with a lighter or flat iron, you MUST watch your temperatures, I would not go further than 160c...

Anyway, why would you do such a thing? You won't gain any overclock headroom, and temperature difference will be very minimal, maybe even unnoticeable.

Quote:

---

Originally Posted by BeepBeep2

Sandpaper (Maybe 800/1000 grit).

I don't think you will be successful, you will most likely kill the chip. It isn't as simple as just heating with a lighter or flat iron, you MUST watch your temperatures, I would not go further than 160c...

Anyway, why would you do such a thing? You won't gain any overclock headroom, and temperature difference will be very minimal, maybe even unnoticeable.

---

Unless you got an IVY.

I hoped for less heat. Less Heat is always Good and I hoped for more OC Headroom, because I can give more VCore

Quote:

---

Originally Posted by theix

Unless you got an IVY.

---

I would love to see someone take it off on one of those and then replace the TIM with better stuff and then put it right back on to see just from doing that how much it lowers temps.

Sorry guys.

But threadsurrection for awesome direct-die block.