Enclosures, paint and heat dissipation.

Not that I know of. But if it looks flat black, chances are it's got good emissivity. Or, to rephrase that: if it's got good even emissivity in visible light, chances are it has good even emissivity across the spectrum.

--
www.wescottdesign.com

Is radiative cooling from a case even significant compared to conducted/convenctive cooling?

Isn't a layer of epoxy paint on an aluminum case going to significantly increase the thermal resistance?

--
Grant Edwards               grant.b.edwards        Yow! I like the way ONLY
                                  at               their mouths move ...  They
                              gmail.com            look like DYING OYSTERS

Good question, found the following:

Ditto. Increase, yes. By how much? Relevant or not?

-- Roberto Waltman

[ Please reply to the group. Return address is invalid ]

ts.dir/JEP1991.pdf

one of the first hits on google was this:

hed-vs.-Coated

-Lasse

Thanks - What search pattern did you use?

-- Roberto Waltman

[ Please reply to the group. Return address is invalid ]

Engine black.

You can get a similar product for motorcycle exhaust header pipes, but I think it has to be cured at high temperature to make it set properly (blowtorch).

Well, for aluminum, you anodize it black. Small scale kits are available

There's probably a way to get nickel to turn black (chemically.) I don't know it off the top of my head, but ISTR "black nickel" plating, so it's likely not too hard to find.

With steel, heat and oil will get you there (black oxide finish). Or you can try stopping at one of the pretty colors before black.

The advantage of these treatments over paint is that they don't add an insulating layer (paint) to the metal to change its color. Still, for starting with a shiny mirror surface, you can easily get better results from turning it black with paint than from leaving it be - you just might be able to do better by turning it black without paint.

Rustoleum Barbeque Black is a classic. Flat is better than glossy, but for some things glossy may look better and be good enough. Baking it helps it cure better.

--
Cats, coffee, chocolate...vices to live by
Please don't feed the trolls. Killfile and ignore them so they will go away.

Both of these appear to address cooling of components on a circuit board, not the cooling of the case. I wouldn't think the two cases would be analogous. The heat transfer from compoents on a board is typically generated in small, localized spots with pretty large temperature differences between the heat source and environment.

Heat trasferred from a case is usually over a very large, uniform area that's only a few degrees warmer than the environment.

Good question.

--
Grant Edwards               grant.b.edwards        Yow! Gee, I feel kind of
                                  at               LIGHT in the head now,
                              gmail.com            knowing I can't make my
                                                   satellite dish PAYMENTS!

Depends of the surface temperature. At moderate temps, like won't burn your hand, radiation is a minor factor.

Not much. The first tenth of an inch of air on each side will have a much higher thermal resistance than the thin layer of paint.

John

If that's a concern then for metal cases one of the commercial blacking products is probably the way to go since the resulting layer is only a few microns thick. I have vague memories of "Carr's Metal Back": good model shops (particularly ones with a railway emphasis) should have it.

--
Andrew Smallshaw
andrews@sdf.lonestar.org

My recollection is that as a rule of thumb at normal ambient temperatures passive radiative cooling becomes important at about 50C and above and grows as T^4. I was surprised how painting a classic old style linear regulator PSU box (of shiny folded aluminium) dirty black immediately resolved a thermal problem once long ago.

What matters is whether the paint appears "black" in the long wave thermal infrared. It does seem a shame to spoil the nice case - does it really run hot enough to justify molesting its appearance?

Regards, Martin Brown

While searching on this subject found a page claiming that any paint based on organic pigments has a high emissivity ( > 0.8 ) in the infrared, regardless of the visible color.

I was asking in general - If you are referring to the TrimSlice, some owners posts suggest it runs too hot. The manufacturer's response is (a) it is within limits and (b) they are still working the bugs out of the Linux power management.

And no, I would not "molest its appearance", only because it will probably void the warranty. Now, after the it expires...

-- Roberto Waltman

[ Please reply to the group. Return address is invalid ]

On a sunny day (Tue, 23 Aug 2011 15:02:53 -0400) it happened Roberto Waltman wrote in :

Sometimes this sort of problems can be solved by pointing a small fan on the case. I use a small 12 V PC processor fan to cool the external heatsink of my CB transmitter.

On occasion I've had good results with thick black marker pen and it sticks to aluminium better than most paints.

Yes, but... From the materials emissivity table mentioned in the first post, there are better options, even if not so pretty:

Aluminum Anodized 0.77 (doesn't mention color)

Black Enamel Paint 0.80 Black Epoxy Paint 0.89 Black Silicone Paint 0.93 Black Parson Optical 0.95

And from

Chemglaze Black Paint Z3O6 0.91

3M Black Velvet Paint 0.91 Martin Black Paint N-15O-1 0.94 Martin Black Velvet Paint 0.94

-- Roberto Waltman

[ Please reply to the group. Return address is invalid ]

Probably not, especially if thin. According to an engineer at a radiator company I called with concerns about some painted aluminum radiators, there is no detectable increase in thermal resistance if the paint is less than about .002 inches thick, at least in forced air flow radiators.

If you go through the calculations I think you will find that the paint thermal resistance is small compared to the air "film resistance", the thermal resistance of the thin layer of air which is semi-stuck to a solid surface, not taking much part in convection flow. This should be covered in any heat transfer book.

Black whiteboard pens have an emissivity close to 1. That's cool, because you can dab a part, thermally image it, and then wipe it clean.

The OP could paint the box black with one of those and see if the temperature changes much, thermocouple somewhere inside maybe.

John

[ Putting on my OP hat ] I ordered a TrimSlice but have not received it yet, so I don't know first-hand if heating is really a problem.

My question (triggered by some TrimSlice forum posts) was of a more general nature. Wanted to know, for now and for future reference, to which extent heating problems could be addressed using paint or coatings. (Answer: more than what I expected.)

I was surprised to find paint/coatings for roofs, for example, specifically mentioning heat issues, but nothing for electronic systems.

Thanks for all the replies,

-- Roberto Waltman

[ Please reply to the group. Return address is invalid ]

Years ago I had a contract to convert a motley assortment of VGA mono monitors for Atari use, one particular make was especially prone to overheating, either the line transistor, B+ rectifier or the PSU in general would blow.

In the short term I kept things under control by blacking the heatsinks with black marker pen - that bought me time to think of replacing the bipolar line driver with a MOSFET.

No other modifications were needed once that fix was established.