Heatsink Colouring/painting

[DaveW]

I was looking at some heatsinks on a website, and i'm considering buying them, but they're blue, and it completely ruins my case's colour scheme. Anyone know a way to paint these so that they still work with good efficiency? Does your standard spray-clearcoat ruin it's heat dissipation rate? I mean, they must have got them blue somehow, so there must be a way of doing this. Ideas, please?

[DaveW]

I just placed the order, so i guess it's too late-but i'd still like to know if it's possible, painting the heatsink would be worth knowing. If no-one knows, i guess i could try and find out myself

If this is something you want to see done, then tell me, i'll get an old heatsink, something hot, a wee box, and a temp probe and find out once and for all...

[crazybillybob]

I know most heat sinks are anodized.....Which means there's allot of work in there to get paint to stick. Anodizing hardens the sureface of the aluminum, which make the paint not stick to it...with much heat it will bubble. To keep the paint on it you have to sand off the pigment (color) then down into the softer alum so the etching primer has something it can bit into....Then the paint will stick.....As for thermal conductivity.....Never though much about it. I'd love to see you try the experiment.


Good luck,
Crazybillybob

[dgrmkrp]

in theory, commercial coloured heatsinks are anodized for a reason... I did try painting a heatsink with normal spray, in light coats.. i avoided painting the sole and it went allright on the mosfets, although i didn't use it much or with a thermal probe. But it did get hot, so... ?
I painted another heatsink with a thicker coat of a special paint, which contains more metal but needs "cooking" and the results where bad.. that thing kept absorbing heat.. and it didn't get cooler with a fan blowing on it.. weird..
i would say that anodizing is the securest way of "painting" a HS.

[DaveW]

I might try the experiment with a couple of different paints, i can take an old HS and cut it into segments of equal size, then test their heat dissipation rates somehow. If i do it, i'll post it up here.

-Dave