How do I fix a Wedge lapped cpu

[Trainwreck]

How do I fix a wedge lapped Q9450 cpu any good ideas

I'm willing to take the time to do it right but how do I keep the low side up consistently and away from the emery paper I have the easypckits - Premium Lapping Kit and
Coarse Lapping Kit

 

[Xaero_toast]

What is a wedge lapped cpu?

 

[Trainwreck]

When you lap a cpu excessive pressure on one side will cause that side to wear down faster causing the IHS to have a wedge shape its flat on top but on a angle /

 

[jediman]

i would guess additional pressure on the high side while lapping would help, but my guess is you will then end up with a bowed proc (which might not be bad). alternatively, depending on your hsf you may be able to lower the hsf on the low side

 

[Archer0915]

A fixture is a good start.

 

[Trainwreck]

A fixture is a good start.

What do you mean a fixture

IE:something to hold the cpu level with the glass?

Like what, a copper shim might work
Piece of lexan might work as well


















111

 

[Archer0915]

What do you mean a fixture

IE:something to hold the cpu level with the glass?

Like what, a copper shim might work
Piece of lexan might work as well
111

Yes essentially yo can use wood or AL if you have access to a shop. Make a hole in said fixture that allows the IHS to penetrate while catching the shoulders of the PCB media. It will be slow going an I suggest using calipers.

 

[Trainwreck]

Yes essentially yo can use wood or AL if you have access to a shop. Make a hole in said fixture that allows the IHS to penetrate while catching the shoulders of the PCB media. It will be slow going an I suggest using calipers.

Yes !!! calipers are a must so I know where to start High side low side or a micrometer.
AL might not be a good idea it would wear at a faster rate and gum up the lapping sandpaper so would Lexan (plastic) coopper or steel would be a way to go

I must Mic the CPU for a accurate thickness on the shim first thing

Stainless steel !!
Any one know a good place to buy cheap shim stock.

 

[Hipcrostino]

you could try setting up to "rails" of wood on the sandpaper that sit static to the cpu when your lapping. You would run the cpu up and down the rails to sand it and only it. You'll have to periodically lower the rails some how. Say taking off a slice with a plain or somthing.

 

[CharlieCS]

You could de-lid the sucker, that would inherit more risk but just trowing that out

 

[Bobnova]

You could de-lid the sucker, that would inherit more risk but just trowing that out

That's what i was thinking, assuming it isn't a soldered chip.

Failing that, get a TRUE or something else that is spring loaded, it'll sit flat regardless of the wedge. Assuming it's not so bad the the edge of the HSF hits something.

 

[LabRat23]

The question I would ask is: Is it a problem? I would say most likely not. All heatsink mounting systems are spring loaded to nullify variances in dimensions. I don't think it is possible for a wedge shaped cpu IHS to provide enough variance to be a problem. It just not thick enough for that.

If the heatsink still mounts on firm as it did before, don't worry about it.

 

[Trainwreck]

you could try setting up to "rails" of wood on the sandpaper that sit static to the cpu when your lapping. You would run the cpu up and down the rails to sand it and only it. You'll have to periodically lower the rails some how. Say taking off a slice with a plain or somthing.

Now there is a idea any thing flat and the proper thickness would work even two strips of (lexan)plastic or metal

You could de-lid the sucker, that would inherit more risk but just trowing that out

Definitively I thought about it but the risks far outweight the end results.

The question I would ask is: Is it a problem? I would say most likely not. All heatsink mounting systems are spring loaded to nullify variances in dimensions. I don't think it is possible for a wedge shaped cpu IHS to provide enough variance to be a problem. It just not thick enough for that.

If the heatsink still mounts on firm as it did before, don't worry about it.

It doesn't seem to be a issue except for cores 0 and 2 run hotter by 6-8 degrees than cores 1 and 3
I lapped it for flatness even though it was wedge shaped and now just looking to see it there is a way to get it level purely out of curiosity if that is the reason for the temp differences and aesthetics.

Yep the HSF is a water block spring loaded and sits on it good

Thanks everyone I have some really good Ideas to go forth with from here. I even thought a machine shop might be able to mill it for a small fee!!

 

[Xaero_toast]

The question I would ask is: Is it a problem? I would say most likely not. All heatsink mounting systems are spring loaded to nullify variances in dimensions. I don't think it is possible for a wedge shaped cpu IHS to provide enough variance to be a problem. It just not thick enough for that.

If the heatsink still mounts on firm as it did before, don't worry about it.

^hits nail on head, IMO^

 

[Shadowmonk]

It doesn't seem to be a issue except for cores 0 and 2 run hotter by 6-8 degrees than cores 1 and 3
I lapped it for flatness even though it was wedge shaped and now just looking to see it there is a way to get it level purely out of curiosity if that is the reason for the temp differences and aesthetics.


I wouldnt worry about 6-8c between the different pairs of cores. I have a constant 6c difference. I've seen this with rigs ive built, spent a week reseating and trying different configurations to relieve this but to no avail. I wouldnt advise de-capping it though, unless your confident that you can do it without killing the chip.