I'm working on a Class A Follower power buffer that will use Tokin SITs so I will have a couple of hundred Watts to dissipate.
I have two large heatsinks that are natural anodised finish. To maximise their performance I would ideally like to finish them in black so the choices seem to be black anodising or powder coating. Any thoughts on which finish will be best?
Oh, one other thing, if you run a finger nail along the heatsink fins they 'sing' and I'm wondering if powder coating would give a modicum of damping?
Thanks
I have two large heatsinks that are natural anodised finish. To maximise their performance I would ideally like to finish them in black so the choices seem to be black anodising or powder coating. Any thoughts on which finish will be best?
Oh, one other thing, if you run a finger nail along the heatsink fins they 'sing' and I'm wondering if powder coating would give a modicum of damping?
Thanks
Powder coating has a thickness that would reduce heat loss.
Anodise or very thin black aerosol paint for a DIY finish.
You won't hear them sing above the noise of the cooling fan - I assume this doubles as a fan heater for winter use?
Anodise or very thin black aerosol paint for a DIY finish.
You won't hear them sing above the noise of the cooling fan - I assume this doubles as a fan heater for winter use?
Don't powder coat it. Powder coating will greatly reduce it's efficiency as a heat sink.
Anodizing improves a heat sink's efficiency, however this is most evident on small heat sinks. I'm guess that your heat sink will be large, so anodizing might not get you as big an improvement as you are hoping for.
BTW - if you are actively cooling (i.e. have a fan blowing over the heat sink) the anodization will offer almost no improvement in cooling performance.
Anodizing improves a heat sink's efficiency, however this is most evident on small heat sinks. I'm guess that your heat sink will be large, so anodizing might not get you as big an improvement as you are hoping for.
BTW - if you are actively cooling (i.e. have a fan blowing over the heat sink) the anodization will offer almost no improvement in cooling performance.
A very thin coat of very high temp spray paint like the stuff they use on exhaust headers. Just don’t cake it on. It’s made for heat transfer and won’t flake off a clean surface. There is stuff called VHT spray paint that comes in lots of colors if you want but black is best.
Many years ago I saw some black engine paint with graphite mixed in for heat conduction, possibly an idea.
Anodized surfaces have far-IR emissivity close to 1.0, you don't need to paint them black for top performance. Its bare metal that doesn't emit well (emissivity about 0.0).
Look up the derating for natural convection and dusty conditions.
Anodising will prevent corrosion and dust sticking.
Use too big a heat sink.
And if possible, keep it as an external part of the housing, better convective heat transfer.
Anodising will prevent corrosion and dust sticking.
Use too big a heat sink.
And if possible, keep it as an external part of the housing, better convective heat transfer.
Hello,
Just a side comment. I prefer to make sure I have more heat sink than the minimum required as electronics that runs cool tends to last for decades. (given quality parts with parts used well within their ratings)
In my opinion, just fairly cheap insurance to avoid problems.
FWIW
Just a side comment. I prefer to make sure I have more heat sink than the minimum required as electronics that runs cool tends to last for decades. (given quality parts with parts used well within their ratings)
In my opinion, just fairly cheap insurance to avoid problems.
FWIW
That's really interesting and helpful Mark, thank you. I have always just assumed that things needed to be black to maximise heat dissipation.
That said, maybe the information you've shared has been staring at me all the time - as I said the heatsinks are recycled from inverters and no commercial business is going to use over-sized heatsinks just so they can be silver - if having black heatsinks meant they could be half the size/lower cost they would do it.
Perhaps I'll just leave the heatsinks natural anodised.
Before starting the project I used the heatsink calculator website to check and these heatsinks are more than up to the task.
Just a side note: actual measuring beats guessing any day of the week
Build a test jig: 2 transistors in parallel (so same voltage), passing same current ( a reasonable emitter resistor should equalize that) so same dissipation, say 12V @ 1A each so 12W each, one mounted in an anodized HS, one in a black one (a fine mist from a spray can should do), measure temperature after 30 minutes.
Wind less room of course.
Build a test jig: 2 transistors in parallel (so same voltage), passing same current ( a reasonable emitter resistor should equalize that) so same dissipation, say 12V @ 1A each so 12W each, one mounted in an anodized HS, one in a black one (a fine mist from a spray can should do), measure temperature after 30 minutes.
Wind less room of course.
For the amount of power you will be dissipating this is already an overkill.
Imho not worth the effort to paint them black for some unknown and probably unneeded gain.
No need to experiment. Power supply guys have already done this for us in great detail. If your heat sink is already clear anodized, don't coat it with anything. All you'll do is degrade it's performance.
At 2000 degrees C, yes. At room temperature they only have to be "black" in the far infra-red. Most materials tend to have high emissivity at far IR, except for bare metal surfaces and some selected materials that lack IR absorption lines such as germanium, silicon, calcium fluoride, polythene etc.
If you've ever experimented with a heat camera these things are easily demonstrated.
Most heat sinks rely mainly on convection, but the radiation is a useful extra mechanism of heat loss.
Another slightly off-topic side-note:
I always thought a dark colour was improving the ir-absorption, thus it would heat up more/faster when there‘s radiation onto the surface? I can‘t relate this to a better temperature emission of a dark surface vs. a bright one…
I don‘t have any material to prove or verify this, this just goes again my „physic-gut-feeling“…
I always thought a dark colour was improving the ir-absorption, thus it would heat up more/faster when there‘s radiation onto the surface? I can‘t relate this to a better temperature emission of a dark surface vs. a bright one…
I don‘t have any material to prove or verify this, this just goes again my „physic-gut-feeling“…