Removing the ihs socket clearance issues??

toddm27

Member

​

Joined

Aug 31, 2003

Location

Smyrna, TN

• Apr 4, 2006

• #1

I was reading that removing the ihs you may have clearance issues with the socket, I was wondering if it does hit on a storm cpu block and a dfi mobo, I plan on unlidding my 165 tomorrow and was looking to see if any sanding of the zif socket or anything was gonna be required, I know it will need shorter screws.

 

Quailane

Member

​

Joined

Oct 18, 2003

• Apr 4, 2006

• #2

On the original plastic AMD heatsink holder, ther are four little nubs that the heatsink rests on. The IHS is EXACTLY the thickness of one of those, so you can figure out from there.

 

OP

toddm27

Member

​

Joined

Aug 31, 2003

Location

Smyrna, TN

• Apr 4, 2006

• Thread Starter
• #3

I understand that, but I was wondering mainly if the waterblock is going to hit against the zif socket with the ihs removed, I've read some stories of it not making contact with the core perfectly flat after removing the ihs from resting on the edge of the socket lifting it up?

 

jonspd

Super Speedy Senior

​

Joined

Jan 18, 2003

• Apr 4, 2006

• #4

I dont see why if wouldn't make 100% contact with the core using a waterblock. Unless mounted wrong. As stated do a few trial run's and setting it and see what kinda imprint you get.

 

adamwinn

Member

​

Joined

Apr 25, 2005

Location

Berkeley, California

• Apr 4, 2006

• #5

id get out some of that cheap white lithium paste that we all have 80000 packets of lieing around, and use that to test fit and adjust/sand/file as necessary-

the tolerances are so close that i wouldn't take anyone's experience as a guide considering how things change from chip to chip, and heatsink to heatsink

 

OP

toddm27

Member

​

Joined

Aug 31, 2003

Location

Smyrna, TN

• Apr 4, 2006

• Thread Starter
• #6

well sounds good, I'm gonna pop it in the morning, wish me luck as my chip will boot at 3ghz with stock voltage and it is somewhat stable but any increase in voltage even running at 2.65 and it will fail prime instantly, passes at stock vcore at 2.65 for 12+ hours. I'm really hoping for a horrible ihs mount job and 3ghz after that.

 

adamwinn

Member

​

Joined

Apr 25, 2005

Location

Berkeley, California

• Apr 4, 2006

• #7

Weird about the voltage - that sounds more like a motherboard problem than a cpu problem to me. have you tried any other chips in the same mobo?

 

OP

toddm27

Member

​

Joined

Aug 31, 2003

Location

Smyrna, TN

• Apr 4, 2006

• Thread Starter
• #8

no, but it is giving me some other issues as well, not booting after shutdown, I will have another board to test tomorow afternoon as I bought a couple ultra d's from monarch, so I'll give it a shot, I also have a 146 sitting around I can test in it, either way I'll give it some goes, one thing I noticed is the temp skyrockets with any extra voltage and it runs hot period compared to my last 165, idles at 40-42, loads at 48-50 stock vcore at 2.65

 

adamwinn

Member

​

Joined

Apr 25, 2005

Location

Berkeley, California

• Apr 5, 2006

• #9

Ah gotcha. Makes more sense now w/ the temps jumping like that. G'luck w/ the ihs

 

OP

toddm27

Member

​

Joined

Aug 31, 2003

Location

Smyrna, TN

• Apr 5, 2006

• Thread Starter
• #10

well I don't have any clearance issues, it went smooth as silk and temps dropped like you wouldn't believe, idle at 20c and load at 2810 25c stock vcore I habn't tried higher yet but I think this thing has a lot of life in it yet, I was wondering how off my temps were on my expert as they seemed to report a little high and to turn out my chip had some rock hard themal compound on the ihs, I can't recommend this enough to anyone and again be very carefull and take your time.


I'll keep you guys posted how high it goes as this is just the tip of the iceburg it seems, before removal anything over 2650 resulted in core 1 failing instantly and now its been priming 30+ minutes at 2.8 btw my chip is a ccb1e 0606vpmw

 

adamwinn

Member

​

Joined

Apr 25, 2005

Location

Berkeley, California

• Apr 5, 2006

• #11

Nice results and thanks for the large picture. It really helps to see exactly how much (or how little) space there is between the ihs seal and the surface mount chips

 

OP

toddm27

Member

​

Joined

Aug 31, 2003

Location

Smyrna, TN

• Apr 5, 2006

• Thread Starter
• #12

well I upped it up to 2900 and it ran prime for 3 1/2 hours before an error at stock voltage, so I backed off to 2850 and well see how that goes, this thing absolutely rocks without the ihs

 

Jacobman

Member

​

Joined

Jun 28, 2004

Location

redding, California

• Apr 5, 2006

• #13

Very nice work, i hope to do something like that when i get my opteron.

 

OP

toddm27

Member

​

Joined

Aug 31, 2003

Location

Smyrna, TN

• Apr 5, 2006

• Thread Starter
• #14

idle was 35-38 prior and now its around 20-22, load was 44-45 prior and now 26-28, It made all the difference for me