Removal of a Northwood B's IHS

Vapor

Distinguished
Jun 11, 2001
2,206
0
19,780
That's right!

It's off my old 2.66, but not my 3.06 yet (don't have any AS right now).

It's surprisingly easy to do...it also promises cooler temps, or better yet, a higher OC. Anyway, a guide is below...I forgot to take pics (I will when I do it to my 3.06), so for now you get diagrams.

With the text on the IHS oriented the way it should be (i.e., readable), here is the bottom of the processor:

<font color=green>|</font color=green> | o<font color=white> F I L L E R T E X T </font color=white> | <font color=green>|</font color=green>
<font color=green>|</font color=green> ---------<font color=red>----</font color=red>---------- <font color=green>|</font color=green>
<font color=green>_______________________</font color=green>

Another way to know the right corner, it's the one with the whole on the IHS

The red part is the only place (at least on the Northwood B) where there is no adhesive between the PCB and the IHS. Willamettes have the same scenario on all four edges (meaning only the corners have adhesive). I do not know about NWa or NWc, but I imagine they are the same as my NWb. DO NOT ATTEMPT THIS ON PRESCOTT--there are reports of the IHS and core being adhesed together...you will kill your CPU if you procede to do this (via the removal and/or breaking of your core).

Alright, back to the red space (in my diagram). Get a good blade (like an exacto knife) and some sort of small wedge (I found one of those newfangled staple removers...a small flat head screwdriver will do, however). Alright, go to town with the blade on those two corners, utilizing that space to insert the blade. You will know when a corner is free, the PCB will bend slightly; work on the other corner. TIP: when you get some lifting space, insert your wedge--it'll help A LOT. When you get both corners unglued, insert the wedge (or the blade first to get it to lift enough to fit your wedge), and lever the IHS off.

Now that the IHS is off, get some 99% isopropyl alcohol (no lower--the other stuff has conditioners to counter skin dryness that leaves residue) and clean it up. You're now ready to achieve cooler temps and possibly a greater OC! Enjoy.

NOTE: I AM IN NO WAY RESPONSIBLE FOR ANY OF YOUR MESS UPS, or any other processor problems after doing this procedure. Also note that I have only tested to see if my 2.66 can still run or not. Because I have no thermal compound, I am unable to say if my processors acheived lower temps. I am also unable to validate any increase in OCing for the same reasons in addition to the fact that my system is incapable (RDRAM and no PCI/AGP lock with a very sensitive RAID controller). Also, the IHS has a function...to prevent core cracking!! Removal of it increases the danger of breaking your core because of poor heatsink installation.

Taking those warnings in mind, everyone else has noticed reduced temps (so you and I will, too), some have reported increased OCing.

Enjoy!!

Maxtor disgraces the six letters that make Matrox.
 

Cybercraig

Distinguished
Dec 31, 2007
2,058
0
19,780
An oxy-acetylene torch is much faster! I just don't have that kind of patience! :lol:


Abit IS7 - 2.8C @ 3.220 - Mushkin PC4000 (2 X 512) - Sapphire 9800Pro - TT 420 watt Pure Power
Samsung 120gb ATA-100 - Maxtor 40gb ATA - 100
Sony DRU-510A - EAT ME MSI!
 

Tavelkyosoba

Distinguished
Oct 14, 2003
188
0
18,680
uhm...why would the heat spreader be there if it impeeds cooling??? that just doesn't make sense.

intel gets those big ass heatsinks because their heatspreaders hurt performance?

i like my AXP, the die look neat haha

I've got a book of matches,
I've got a can of kerosene,
I've got some bright ideas involving you and me.
 
I've got to agree. Why wack on a heat spreader when it impeads cooling performace???

<font color=blue>My dick is so big, that my dick has a dick. And my dicks' dick is bigger than yours.</font color=blue>
 

Vapor

Distinguished
Jun 11, 2001
2,206
0
19,780
It impedes cooling because of the increase in thermal resistance...it saves Intel money because it can handle a lot more abuse and improper HSF mounting that just a raw core can. Anyway, the IHS is bloody heavy (more than half the weight of the processor) and the thermal compound between it and the core looked to be crap--it was all hard too. Anyway, it does make sense for Intel to use them to increase the durability while sacrificing only a bit of a thermal advantage.

Maxtor disgraces the six letters that make Matrox.
 

jammydodger

Distinguished
Sep 12, 2001
2,416
0
19,780
As the name suggests the heat spreader increases the surface area allowing for heat to be dissipated more quickly. This is usefull when using an aluminium heatsink because they are notgreat conductors. However when you use a copper heatsink the heatspreader actually impeads cooling because it has a higher thermal resistance than copper and therefore acts as a barrier between the core and the heatsink.
 

Tavelkyosoba

Distinguished
Oct 14, 2003
188
0
18,680
well i know what its supposed to do, and it makes sense. the copper core in axp heat sinks acts like an external heat spreader-rapidly spreading the heat from the die to a large surface of aluminum.

but i see no need for an IHS if it lowers heat disspipation and increases durability...i've never broken any of my athlons or axp's. the only thing i could see damaging a chip would be dropping it(wich i've done) and the only thing that gets hurt are pins...and you just pop out your pocket knife and bend em back.

now i suppose it makes for less expensive heatsinks because they can be made of aluminum instead of copper cored...

i just feel like its big business trying to save a buck by switching to single ply in the company bathrooms. haha

I've got a book of matches,
I've got a can of kerosene,
I've got some bright ideas involving you and me.
 

Cybercraig

Distinguished
Dec 31, 2007
2,058
0
19,780
Yea, but look at the socket 478 H/S mounting arrangement. When you lock down those cammed arms you are putting a lot of force on that chip! I just changed MB's and my lapped copper heatsink had a GOUGE in it! I had to re-lap before re-installing! Now that's some pressure, matey! :smile:


Abit IS7 - 2.8C @ 3.30 - Mushkin PC4000 (2 X 512) - Sapphire 9800Pro - TT 420 watt Pure Power
Samsung 120gb ATA-100 - Maxtor 40gb ATA - 100
Sony DRU-510A - EAT ME MSI!
 

Tavelkyosoba

Distinguished
Oct 14, 2003
188
0
18,680
oh my, i didn't know they applied that much pressure.

I've got a book of matches,
I've got a can of kerosene,
I've got some bright ideas involving you and me.