Painted Heat Sinks

[Montie]

Morning,

Have you had any experience with painted heat sinks? I have a customer
who would like a painted finish and I'm having a hard time finding the
thermal conductivity of paint so I can calculate what the thermal
effect would be.

Any suggestions?

Thanks,
Montie
[email protected]
Montie Design
www.montie.com

 

[John Larkin]

Morning,

Have you had any experience with painted heat sinks? I have a customer
who would like a painted finish and I'm having a hard time finding the
thermal conductivity of paint so I can calculate what the thermal
effect would be.

It's huge compared to the thermal conductivity of still air, so it
won't change things much. The only interesting case would be with
high-velocity forced air flow, where it may be worth doing the math.

Why not anodize? That's probably easier than painting.

John

 

[w_tom]

With each change of medium is an increase in thermal
resistance. CPU to thermal compound to heatsink has higher
thermal resistance (conducts less heat) than CPU direct to
heatsink. Thermal resistance increased due to medium changes.

Yes, you can paint the heatsink. Probably matters little
the type of paint used. Medium is heatsink, through paint,
then to air (3 mediums) instead of heat direct to air (2
mediums). The third medium causes thermal resistance
increase. Then there is this thing called black body
radiators which also would be lost in painting.

Up front number is 'degrees C per watt' which includes
thermal resistance of those medium changes. That number is
what heatsink manufacturers (the responsible ones) and you
both worry about. Painting will only adversely increase that
number.

Would not worry about paint's thermal conductivity number.
A medium change will increase thermal resistance anyway.

 

[Ignoramus4324]

I suspect that using black paint can actually improve heat
dissipation, due to increased radiation of heat as opposed to a shiny
reflective surface.

i

With each change of medium is an increase in thermal
resistance. CPU to thermal compound to heatsink has higher
thermal resistance (conducts less heat) than CPU direct to
heatsink. Thermal resistance increased due to medium changes.

Yes, you can paint the heatsink. Probably matters little
the type of paint used. Medium is heatsink, through paint,
then to air (3 mediums) instead of heat direct to air (2
mediums). The third medium causes thermal resistance
increase. Then there is this thing called black body
radiators which also would be lost in painting.

Up front number is 'degrees C per watt' which includes
thermal resistance of those medium changes. That number is
what heatsink manufacturers (the responsible ones) and you
both worry about. Painting will only adversely increase that
number.

Would not worry about paint's thermal conductivity number.
A medium change will increase thermal resistance anyway.

--

 

[GregS]

It's huge compared to the thermal conductivity of still air, so it
won't change things much. The only interesting case would be with
high-velocity forced air flow, where it may be worth doing the math.

Why not anodize? That's probably easier than painting.

John

Powder coat is probably thicker than standard paint, which may hinder things.

greg

 

[John Larkin]

I suspect that using black paint can actually improve heat
dissipation, due to increased radiation of heat as opposed to a shiny
reflective surface.

Radiation effect is tiny at the temperatures seen on ordinary heat
sinks. The paint's thermal resistance will likely make things worse
than any increase in radiation.

John

 

[Dave]

Radiation effect is tiny at the temperatures seen on ordinary heat

sinks. The paint's thermal resistance will likely make things worse
than any increase in radiation.

John

Radiocity equation at the end of this paper may be insightful,

http://power.ece.uiuc.edu/Balog/images/How to keep your cool when working with power electronics.pdf

 

[Spehro Pefhany]

Radiation effect is tiny at the temperatures seen on ordinary heat
sinks. The paint's thermal resistance will likely make things worse
than any increase in radiation.

John

Radiation works both ways, though, and if you've got hot stuff nearby
(such as those thermionic devices or some kind of space-based
thermonuclear reactor) you could get the HS 'seeing' the hot stuff
rather than reflecting the IR off. So maybe they should be black on
the cold side and shiny on side that sees the sun or whatever.


Best regards,
Spehro Pefhany

 

[Keith Williams]

Radiation works both ways, though, and if you've got hot stuff nearby
(such as those thermionic devices or some kind of space-based
thermonuclear reactor) you could get the HS 'seeing' the hot stuff
rather than reflecting the IR off. So maybe they should be black on
the cold side and shiny on side that sees the sun or whatever.

The "color" in visible light is irrelevant, unless the HS is
radiating in the visible spectrum. I suggest this is usually
badness in this bidness. ;-)

 

[Spehro Pefhany]

The "color" in visible light is irrelevant, unless the HS is
radiating in the visible spectrum. I suggest this is usually
badness in this bidness. ;-)

Black and shiny are not colors.


Best regards,
Spehro Pefhany

 

[w_tom]

John's post adds details to this previously posted sentence:

Then there is this thing called black body radiators
which also would be lost in painting.

Heat sinks are colored dull black so that the minor
additional cooling from radiation will decrease the important
number - 'degrees C per watt'. But as John notes, the
additional cooling by radiation is minor.

 

[John Larkin]

Radiocity equation at the end of this paper may be insightful,

http://power.ece.uiuc.edu/Balog/images/How to keep your cool when working with power electronics.pdf

His example 2 is just silly. If you glop on tons of silicone grease
and apply even modest mounting pressure, it oozes down to under 100
microinches.

John

 

[Gandalf]

Morning,

Have you had any experience with painted heat sinks? I have a customer
who would like a painted finish and I'm having a hard time finding the
thermal conductivity of paint so I can calculate what the thermal
effect would be.

Any suggestions?

Thanks,
Montie
[email protected]
Montie Design
www.montie.com

If you're using convective cooling and have cool surroundings to radiate to,
glossy black oil based enamel will considerably boost the cooling over bare
aluminum. Anodizing doesn't seem to do anything for the long wavelength infrared
radiation needed for the lower heatsink surface temps.

On one project, I needed .35C/W cooling but the best aluminum extrusion I could
buy would only give .5C/watt. By painting with black oil enamel, I got my
..35C/W.


If you're using forced air on a relatively cool (less than 100C) aluminum
heatsink, anodizing would be the simplest way to get color without affecting
performance.

 

[John Larkin]

If you're using convective cooling and have cool surroundings to radiate to,
glossy black oil based enamel will considerably boost the cooling over bare
aluminum. Anodizing doesn't seem to do anything for the long wavelength infrared
radiation needed for the lower heatsink surface temps.

On one project, I needed .35C/W cooling but the best aluminum extrusion I could
buy would only give .5C/watt. By painting with black oil enamel, I got my
.35C/W.


If you're using forced air on a relatively cool (less than 100C) aluminum
heatsink, anodizing would be the simplest way to get color without affecting
performance.

According to the little prog I have, a heatsink of 100 sq cm area, 50
C above ambient, emissivity = 1.0, radiates a tad under 4 watts.
That's about 12.5 c/watt, and a 50C rise is pretty radical in most
apps. For a 30C rise, more reasonable maybe, it's more like 2 watts.

If it's inside a non-black box, some of its radiation will be
reflected back. Similarly, parallel fins mostly just radiate at each
other, which can defeat most of the radiative cooling effect.

John

 

[Adrian Tuddenham]

John's post adds details to this previously posted sentence:

Heat sinks are colored dull black so that the minor
additional cooling from radiation will decrease the important
number - 'degrees C per watt'. But as John notes, the
additional cooling by radiation is minor.

That may be good for the written specification, but it is not what you
want for outdoor P.A. gear which may be used in strong sunshine; it will
absorb more radiant energy because of it.

I'm not sure that all polished metals reflect infra-red wavelengths
well, either. I have recollections of 'reflective' car tools getting
very hot when left lying around in the sunshine.

 

[Roger Hamlett]

That may be good for the written specification, but it is not what you
want for outdoor P.A. gear which may be used in strong sunshine; it will
absorb more radiant energy because of it.

I'm not sure that all polished metals reflect infra-red wavelengths
well, either. I have recollections of 'reflective' car tools getting
very hot when left lying around in the sunshine.

Yes. And some colours that can be quite 'unexpected', can be very good IR
radiators. Certain white paints, are actually good IR absorbers and
radiators in the NIR, and IR sections of the spectrum, while some quite
dark appearing coatings are poor (hint, normal 'anodised' aluminium is not
good in this section of the spectrum...).

Best Wishes

 

[Keith Williams]

Black and shiny are not colors.

Ok, black or shiny in the visible range may not be so black or
shiny in the far infra-red.

 

[John Larkin]

That may be good for the written specification, but it is not what you
want for outdoor P.A. gear which may be used in strong sunshine; it will
absorb more radiant energy because of it.

I'm not sure that all polished metals reflect infra-red wavelengths
well, either. I have recollections of 'reflective' car tools getting
very hot when left lying around in the sunshine.

While playing with my new IR thermometer, I tried measuring the light
fixture above my drafting table. The polished brass part measured well
over 100C with a thermocouple but barely above room temp with the IR.

John

 

[John - KD5YI]

According to the little prog I have, a heatsink of 100 sq cm area, 50
C above ambient, emissivity = 1.0, radiates a tad under 4 watts.
That's about 12.5 c/watt, and a 50C rise is pretty radical in most
apps. For a 30C rise, more reasonable maybe, it's more like 2 watts.

If it's inside a non-black box, some of its radiation will be
reflected back. Similarly, parallel fins mostly just radiate at each
other, which can defeat most of the radiative cooling effect.

John

So, how much power does it take to raise the 100 sq cm heatsink in your
first paragraph to 50C? How much of the total power is radiated?


John

 

[Daniel Lang]

Yes. And some colours that can be quite 'unexpected', can be very good IR
radiators. Certain white paints, are actually good IR absorbers and
radiators in the NIR, and IR sections of the spectrum, while some quite
dark appearing coatings are poor (hint, normal 'anodised' aluminium is not
good in this section of the spectrum...).

That is why telescope domes are painted white using a paint that is dark
in the thermal IR region. It mimimizes heat absorption during the day
and helps the domes cool off faster at night.

Daniel Lang