Emissive Surfaces Outdoors

[drawoh]

I have a requirement to specify a material and finish on a housing that will be installed on top of a vehicle. We want to dump heat from the electronics inside, even on a hot day. This will not happen if the sun's radiation heats it up.

The vehicle will be left outdoors and will experience rain, snow, hail, sleet, dirt, and all the other crap that reaches the top of a car or SUV.

I can see very low values for emissivity for highly polished metals like aluminiun, and low values for oxidized and anodized materials, again, like aluminium.

Is there a protective finish I can put on a polished aluminium housing that will prevent oxidation, and maintain a low emissive surface?

JHG

 

[unclesyd]

Normally you would brite-dip and anodize and apply one of many clear coatings available for the automotive market, like Por-15

Personally I would give the Pernalac people a call about their clear lacquer. I've had very good success with this product.

http://www.permalac.com/product.asp

 

[IRstuff]

Note that low emissivity also works against you, since the radiated heat will also be low.

The clear coat adds a layer of insulation, which will decrease the convection efficiency.

TTFN

FAQ731-376

 

[btrueblood]

Drawoh,

What you are really looking for is "low solar absorptivity paint" or coatings. Google the quoted term, top hit was:

http://www.pcimag.com/Articles/Cover_Story/645bb044256a7010VgnVCM100000f932a8c0____

Note that the performance of any coating is subject to degradation over time due to weathering.

 

[unclesyd]

You could go for the gold standard of IR reflective surfaces, that is Gold.

I would look at the thinner variety of IR reflective roof type paints or coatings.

Also they make films for glass that are quite effective. I have some that are seven years old and still looking new.

Whatever type surface or coating you choose make sure it block only from the outside.

 

[drawoh]

You could go for the gold standard of IR reflective surfaces, that is Gold.

I am considering gold. Non-corrosive is good. I have a price for nickel plating, and it is fairly reasonable. I need to verify that it will survive in its shiny, non-emissive state. Oxidized nickel absorbs heat.

JHG

 

[btrueblood]

Was going to suggest gold plating also, it's what we used for hot surfaces on spacecraft thrusters. There are some types of low-alpha paints that still beat gold (because the emissivity of the paints is very high at "room temperature", and the ratio of alpha (solar absorptivity) to epsilon (thermal emissivity) is lower than for gold. But, the paint I'm familar with was a silicone base, good to 400 F, very expensive (more expensive per pound than gold) and difficult to apply. Gold can be plated over a nickel strike very, very thinly. At least on 304 stainless, the gold is very stable (our process was copper strike, nickel plate, then gold plate). Make sure your surfaces are as smooth as possible, the shinier the better, before sending off to the platers.

Wow. Sorry about the rambling post, but it's late and my mind is fried.

 

[KENAT]

Think again about what IRstuff said.

How/when are you planning to lose heat from your housing?

Is the box only dumping heat when the vehicle is moving so you can rely on convection or do you need to dump heat even when stationary for extended periods?

If dumping when stationary then I'd think you want to radiate a lot of the heat, not rely on beneficial breezes etc.

So, making it gold or similar will work against you.

Maybe you make it black but with a sun shield that allows air flow but limits direct heat from the sun or something.

To give an extreme example of the issue IRstuff is talking about, consider the SR71 Blackbird that was painted a very dark purple, almost black. Obviously this is going to absorb heat from sunlight. However, it also helps radiate heat. In fact when flying at high speed it not only radiates what it absorbs from the sun it also radiates some of the heat generated from surface friction. The blackbird was the dark color to help keep it's temperature down.

(Sorry if you didn't want a more tangential post)

Posting guidelines faq731-376

http://eng-tips.com/market.cfm?

(probably not aimed specifically at you)
What is Engineering anyway: faq1088-1484​

 

[drawoh]

KENAT,

This box will have a heater inside it for sub-freezing temperatures. My boss wants passive cooling at high temperature. My worst case is the vehicle sitting still in the bright sunshine at maximum ambient temperature. My device is rated to 45°C. My maximum ambient is 40°C. I am figuring that the sun's radiation will kill my [Δ]T. My heat conduction to outside the box must be worked out accurately for any of this work.

I am trying to get them to select a device than can run at 55°C, and/or install a fan, and/or use a thermoelectric cooler to control heat.

JHG

 

[unclesyd]

Here are two approaches that may have to potential to help alleviate some of your heat load. It is Micro Colling from Wolverine Tube and the Heat SpreaderShields and Graf/Hx from GraphTech.

There is local firm is using a flat plate heat pipe to handle a very high heat load. I've have no info on which one.

http://www.microcooling.com/

http://www.graftechaet.com/eGRAF/eGRAF-roducts/SPREADERSHIELD-2-D-Heat-Spreaders.aspx

http://www.graftechaet.com/GRAFIHX/GRAFIHX-Home.aspx

 

[IRstuff]

With only a 5°C delta, I don't think you can do much of anything. How much power is being dissipated within the box? Even if you heatpipe everything to the surface, you'll still wind up with less than a few degrees for convection or radiation.

Even with 1% emissivity, anything more than 0 W of dissipation will drive the surface temperatures up by about 4.5 degrees per exposed surface.

One option is mirrored sunshades around the box, with about 2 inches of gap between the box and the shades. The top and bottom should allow air passage. This would cut the load on the surfaces down, and pure passive convective might be able to get some internal heat out. Even still, your best case only allows about 30W of dissipation for each square meter of surface area. You might need to put pins or pin-fins on the outside surfaces of the box.

TTFN

FAQ731-376

 

[unclesyd]

You need to get the hold the prototype. It has a turn down ratio of 1.000,000 to 1.

Seriously there might be some more information about the sunshield lurking around the NASA site.

http://www.nasa.gov/centers/goddard/news/features/2010/webb-sunshield-testing.html