Better to have a smooth or rough inside of block?
 

So - I was just thinking - is it better to have a really smooth or a really rough or somewhere in between surface on the inside of a block? I've heard that you want turbulence - so it seems that rough would be good. But most of the blocks I've seen look rather smooth on the inside.

A rough surface will increase the turbulence, which is good. Most blocks are machined, and so they're smooth. (At least that's what I've heard)

The question is: How much difference does it make? :)
Hard to measure.

Well then - why not give it a jagged surface? I mean - give it perhaps a sawtooth sort of surface - and then go over it with a coarse grit sandpaper. Would that be an improvement?

my suggestion is that a rough surface produced a hardly noticible turbulance. if you need a good anoumt of turbulance inside you block, you will need to implement something else like pins or dimplets or some kind of fins.

The golfball aproach? That one is quite easy anyway. just drill shallow holes all over the block...

This is a tricky question.

Because of the relatively low flow rates we use, there's not much added flow restriction from the blasted surface.

On the cooling side though, you're looking at trapped molecules of water, but also an increased surface area (can increase by 50%, from perfectly flat, but that's theoretical, not practical).

I suspect it doesn't make much difference, if any. It sure looks nice though!

The better blocks that have been out don't have room for any roughening of the surface. Not like you can really sand the inside of a White Water to effectivly. BUT I do have a couple sand blasters at work that would add a lot of tiny dimples to anything. it might eat those thin fins though. I was thinking about sand blasting my REV. 4.0 but I doubt the difference could be measured.

Ok. There is a heat layer over the metal that is the radiated heat. The barrier is say... 1-2 mm. If you have water moving over this barrier, you are carrying the heat away. If you have a bumpy surface, say a sand blasting where the surface is not 1mm varience, then your heat barrier is not effected and you could call the surface change cosmetic. If the surface deformity extends in a significant way to bring this heat barrier out into a faster flow of the water, or increase surface to water heat transfer area, then you have an advantage. I look at hara's dimple block, jaydee's dimples in his avatar, or the microchannel. On those examples you have a large enough deformity to add to heat->water exchange. The sandpaper approach or sand blasting, doesn't do jack from my understanding.

Well not really dimples, but engraved pyrmid pins.
http://www.customcooledpc.com/micro3/004.jpg
http://www.customcooledpc.com/micro3/002.jpg

My theory on this was to add as much surface area as possible over the core in which would be used and not to add any excess surface area that wouldn't be used either by length, width, or higth. It seemed to have worked well.

which block??