Prelapped Heatsinks

[koslov]

Saw this review of a prelapped heatsink and had to check out how they were lapped. Elitecool.com describes their method here: http://www.elitecool.com/privacy.php .

Here's a snip:

We lap out heatsinks by using a flat piece of glass and then using 600, 1000, 1200 grit sandpaper to sand it down until it is perfectly smooth. This usually takes 15-20 minutes. We then use metal polish and buff it to clean it up. The finished product gives excellent performance.

While I think its a good idea for a service retailers can offer, I think their method is terrible. 1200grit isn't all that great, considering 2000grit is easily available. Then they use *polish* and buff it?? Then they have the audacity to call it "extremely flat" and charge an $8 premium??

While I do all my own lapping, I would like to see an "enthusiast" retailer offering lapping on their products. Obviously they would have to do a little better than elitecool, maybe using a Lapmaster like N8 has referred to: http://forums.procooling.com/vbb/sho...&threadid=5728

[Brians256]

Sad, but until the consumer is educated, this kind of pseudo-service (right idea, but wrong method) will abound. HOCP will probably tout this service (unless they irritate Kyle and/or Steve in some way) because they typically under-research products before they recommend or warn against them.

[pippin88]

You do not really want to go above 600-800 grit if you plan on using thermal compound. This removes any irregularity or lack of flatness, but still allows the thermal paste to interface well.

I think lapping has been taken a bit far, with people thinking that mirror finish is necessary etc. when the main aim of lapping is to get rid of any bow or irregularities in the heatsink.

Mirror finish is not best.

They should also not be using anything on the surface after lapping really.

[g.l.amour]

who was it that said that polishing up copper kills it thermal properties?

[Brians256]

Uhhh, a bonehead? Polishing a material doesn't change it's thermodynamic properties as far as I know. That's a property based upon the constituent atoms and their structure (i.e. crystaline structure? if so, what lattice arrangement, etc...). Unless polishing changes the atoms present in the surface (such as removing an oxide layer), then it won't do diddly to change the material's properties.

If you are talking about the thermal properties of the interface, then that's a whole different ball game. Then, you are talking about the size and frequency of the gaps between surface A and surface B. But, we all know that.

Hopefully, I'm not misinterpreting g.l.amour's terse post.