Should i lap my cpu & cooler

[K404]

I disagree, but using the same arguement as you! LOL

You`re right, the closer you look at the finish the less smooth it appears, and an atomically flat surface is physically impossible to achieve.

Assuming you lap by hand, no matter how fine-a grit you use the micro scratches you add to the surface will be bigger than whats present and leftover from the machine, assuming the machine is in good condition.

We use thermal paste because of the micro troughs no matter how good the surfaces are.

K

 

[Wellcomef3-a]

Assuming you lap by hand, no matter how fine-a grit you use the micro scratches you add to the surface will be bigger than whats present and leftover from the machine, assuming the machine is in good condition. K

K404, I take it you are talking about attempting to lap a surface that has been previously lapped professionally, such as on the range of SwiftTech products? For me that is a clear no-no, I would never attempt to improve on a machine lapped surface for obvious reasons.

However there are lapping methods utilized in the production of slip gauges which go far beyond the methods used on heat sinks and the like.

As a finish, there is no better than lapping and then polishing. Not even surface grinding comes close.

 

[Master_G]

It seems to me that we are talking using different terms here (myself not using the machining terms).

To me, when i say flatness, it encompasses both roughness and flatness. I am imagining this on a micro level, so i would only call a heatsink and IHS perfectly flat if they had no air gaps in between them. This appears to be both flat and a zero level of roughness (having looked into this more).

It was the word "finish" that got me, if you had said roughness i would have agreed, but when most people talk about "finish", they are talking about how it looks to the naked eye, not how flat (and rough) it is.

I imagine that regardless of what it does to flatness, lapping should always reduce the roughness of the surface, and so therefore it is only flatness that counts, and reduced roughness will come through the lapping process.

(Starting from a rough and non-smooth IHS/heatsink) Increasing flatness through lapping necessarily lowers the roughness of the surface, whereas lowering the roughness of the surface through lapping does not necessarily increase the degree of flatness.

Is that correct? That is the reasoning behind me only referring to flatness in reference to lapping.

G

EDIT - just seen the post above you mention polishing? Using polish is a big no-no from what i have read, since the polish makes the surface look good by filling in the micro holes, and so reduces heat transfer as you want that to be done by TIM and not polish.