Removed IHS Results thread

**misteroadster** · 09-28-2006, 12:14 AM

Failed and killed one E6600 and one E6700 , pcb of 6600 was probably too much bended while warming, and core cracked with 6700. Hard loss.

Avoid removing IHS.

Maybe i will try another method anyday (Cabriolet IHS by lapping).

**vudoodoodoo** · 09-28-2006, 02:28 AM

Blagh. I only have a heatgun. I don't wanna have to go and buy a torch.

Gonna just lap the IHS then. ;x

**pH(x)** · 09-28-2006, 08:41 AM

Originally Posted by vudoodoodoo

Hmm I wonder if a heatgun would get it hot enough to remove the IHS...

That's a big Negative vudoo.

Your standard torch (propane gas powered) blows a flame in excess of 5,500F. And the heat at the tip is very concentrated, most heat-guns you will find, only blow at 400F by heating atmospheric gas with coils of wire.

**Kjaks** · 09-28-2006, 09:16 AM

Originally Posted by misteroadster

Failed and killed one E6600 and one E6700 , pcb of 6600 was probably too much bended while warming, and core cracked with 6700. Hard loss.

Avoid removing IHS.

Maybe i will try another method anyday (Cabriolet IHS by lapping).

I'm sad to hear, but honestly, the top of the lighter flame is about 600 degrees C and the core will be much warmer than the rubber by the time it melts. I think it would be better to just cut the rubber all around the edge and then hold the CPU around the bottom pcb and warm the middle of the IHS with a lighter for a shorter period of time and then twist the IHS of if it doesnt fall off by itself.

If I saw your post earlier I would've warned you..

**menlatin** · 09-28-2006, 11:22 PM

Originally Posted by misteroadster

Failed and killed one E6600 and one E6700 , pcb of 6600 was probably too much bended while warming, and core cracked with 6700. Hard loss.

Avoid removing IHS.

Maybe i will try another method anyday (Cabriolet IHS by lapping).

Wow, that sucks. I think maybe that glass of water might have had something to do with that. Or did you crack the core during mounting? In that case, did you use and shim/spacer/feet to prevent your cooling method of rocking back and forth?

**corsair8575** · 09-29-2006, 10:25 AM

If you have learned anything in High School class is that if you do a high delta in temperature on a substance it will crack and bend.

Please dont drop hot things in water.

**celemine1Gig** · 09-29-2006, 01:59 PM

Originally Posted by corsair8575

If you have learned anything in High School class is that if you do a high delta in temperature on a substance it will crack and bend.

Please dont drop hot things in water.

Exactly what I wanted to add.

I can only agree with corsair8575. Letting the CPU cool slowly would be much better for the material. I even heated my CPU to ~150°C or higher accidently (hadn't fully cut through the black rubber), but let it cool slowly afterwards. The CPU is still working fine.

**misteroadster** · 09-30-2006, 08:59 AM

Yes, i'm thinking that was the mistake on the 6700. But now i'm not really confident to try a third time :/

**menlatin** · 10-09-2006, 01:36 AM

What up folkity folks. Well, i finally got my e6400 working again... except only 1 core is functioning. Kinda sucks, but i can limp for the time being. Turns out i wont be able to get a new cpu this month, so is going to have to last me.

Anywayses. Here's some results. I could only get the cpu to work with the Ds3 and not my P5b-deluxe. Sucks, yes, but atleast it works and overclocks. It gets to 3.6ghz on 1.425 bios volts, 1.41v windows, 1.39v load.

Ambient 25C
Core Temp 54C idle / 64C load
Speed Fan 33C idle / 45C load

The cpu overclocks and works, but it does some weird things too. cpu-z is missing stats, cpuid is missing stats, and PC wizard is alittle off. I'll get screen shot later.

**Silvermirage** · 10-14-2006, 09:03 AM

There is far too much risk involved for me to try it. Nice job!

**icon57** · 10-16-2006, 07:37 AM

i have never had the ballz to remove the ihs, but thank you guys for sharing such detailed info, and pics.

**menlatin** · 10-28-2006, 05:11 AM

got a new e6400 from The Nemesis today. Its nice. after booting into windows @ 4ghz, with high temps, i decided to lap. temps still high, but i got 3997mhz 1m pi stable. Open the windows to the cold outside, and got 4005mhz 1m psi stable. Pulled the IHS off, and now i'm @ 4000mhz dual 32m pi stable with normal (68*F)
Pics

**chicconet** · 11-08-2006, 09:06 AM

From my experience I always found a benefit removing the IHS from the cpu, a better cooling for sure. Anyway the naked core is very fragile and I'm often worried to break it during cooling device mounting procedure. Now I use this rubber frame to prevent craking and I sleep better tham before

**celemine1Gig** · 11-08-2006, 10:23 AM

Wohooo, the price of these foam pads is hilarious.

I'd bet that their profit should be around 100-200% with these things, if not more. Still, it's one of the best ways to protect your CPU-DIE from broken edges etc. .

**menlatin** · 11-08-2006, 05:39 PM

Or you could just get a hold of 1" of electric tape and some scissors. 5 layers of tape is pretty darn close to the height of the core. 4 layers on A64's and 5 on Core2. I've done 5 bare cpu's like that so far.

EDIT: I guess while im at it i should add the latest results

**~~Zucker2k~~** · 11-08-2006, 06:45 PM

I went Ihs-less last weekend with my E6700. Temperature gain is about 6-7C. Used a lighter and a box cutter. You'll probably need a glove too.

DISCLAIMER: You do this at your own risk!

Quick guide:
1. Cut sides with the box cutter or razor blade. Conroe cores are located in the center of the pcb so unless you're determined to cut into the core, you probably won't unless you're stinking rich and won't mind. Don't do this while you're high, or drunk, or both

.
2. After cutting sides, fire the lighter (I used a stove lighter, bigger flame more heat). Invert CPU and hold lighter so that the flame is about an inch below the surface of ihs. You will need to wear gloves.
3. Tricky part. In order to ensure that you don't burn the core, while holding the flame to the CPU, insert the box cutter between pcb and ihs and pry gently. When the TIM hits the right temperature, the ihs will come flying off. That's it.

PS: PCB (CPU) will most likely bend a little bit (this may increase with the application of too much heat). But you can remedy this by straightening it as much as you can while still warm. Don't apply too much pressure, it'll straighten out completely when you install it (heatsink pressure+heat from core would do the job).

PPS: Another tricky part is the mounting. My waterblock is sitting right on the core with AS5 in-between them, and a lot of good pressure for my own good. How you approach yours is up to you. There are a few guides around. (Check here: www.thetechrepository.com).

Have fun, and goodluck.

Oh, before I forget, I lapped the core with a very fine grit foam and cleaned the whole topside of the pcb with isopropyl alcohol.