Kilo-Lumen bike headlight

Step 5: Machining the Housing

This is yet another step where I assume some knowledge of milling machine techniques.

I used a 1/8 inch end mill to machine 70 mil deep channels across the length of the housing. This provides heat sinking and more than doubles the thermally emissive surface area of the housing. The channels were cut front to back with respect to direction of travel of the bike. This allows the airflow from riding to dissipate the heat better. Cross cutting in the other direction resulting in a checker board pattern would increase turbulence and potentially increase cooling effectiveness. I have found though that after my 15 minute commute on cool evenings that the housing is just barely warm to the touch.

There are some excellent comments below regarding the coating of the housing. It turn out that for an application where convective airflow is the dominant cooling mechanism (as is the case here) that coating the surface of the aluminum will not measureably improve its heatsinking performance. In any case I decided to powder coat mine for durability and aesthetics. As stated above, the cooling performance of the housing is sufficient, so overall it seems to be working fine.

Comments

[aburton]

I'm sorry, but I'm not quite sure what you mean when you say "Bare aluminum is not as good a conductor as one might think." Do you think you could explain this a little further? As it is currently worded, this is a deeply misguided statement.

[kc6qhp]

If you aren't quite sure, how sure are you that it's deeply misguided? :)

Anyway, a polished, or smooth aluminum heat sink will dissipate heat poorly compared to a heat sink where the surface is anodized, painted, or otherwise roughed up. Of course if you paint it with an inch thick of rubber that no longer holds true. The point is that aluminum when machined as I did, comes out pretty shiny and giving it certain coatings will help the aluminum dissipate heat better. So maybe the word "conduct" was inappropriate.

effect of anodizing

Here's a table I found for thermal emissivity:
Polished aluminium 0.05
Polished copper 0.07
Rolled sheet steel 0.66
Oxidised copper 0.70
Black anodised Al 0.70
Black enamel 0.85
Dark varnish 0.89
Black oil paint 0.92

I suppose I would hae been better off with a thin flat black coating.

[roosta]

this is because dark surfaces radiate heat far more effectively than a shiny surface. its stated in one of the laws of thermodynamics, but i hate physics, so im not going loking for it :P i cannot understand why people dont paint their radiators matt black. far more efficiant than white.

[mortso]

So the inverse of Solar radiation is happening, where Black absorbs more solar heat than shiny or white surfaces. I never considered the reverse to be true, very cool thing to know. thanks!

[aburton]

Right, but in this case it is not a strong mode of heat transfer. You're better off leaving the aluminum uncoated.

[aburton]

Thermal emissivity only applies to radiant heat. In this situation, radiant heat is going to be negligible compared to forced convection. The properties you want to be looking at are surface area and thermal *conductivity*, not thermal emissivity. You're right that it is preferable to increase the turbulence across the surface of the housing, but it is HIGHLY unlikely that you'll be able to coat the surface of the housing with something that can increase radiant heat and keep the convective heat constant. Anodizing won't help either. When you anodize, you're coating the surface of the aluminum with a (relatively) thick layer or aluminum oxide. This also doesn't conduct heat as well as pure aluminum (or 6061 aluminum, which is probably what you're using.) The result will be that you're actually insulating the aluminum from convective heat transfer. The important thing to keep in mind is that you're not radiating much heat. You're going to get several orders of magnitude more heat transfer from forced convection. If you want to increase the heat transfer, try to maximize your surface area.

[kc6qhp]

Yes, you are correct. Now here's a question (because I honestly don't know)... At a microscopic level what is the mechanism of heat exchange during convection? Molecules of the cool air (while in their convection cycle) get in close proximity to the heatsink, or even touch, but what imparts the thermal energy from the hot heat sink to the air? Is it direct contact (conduction)? Is it microscopic radiant heat exchange? In any case you do need excellent conduction of heat to that interface surface, but what goes on at that boundary in convective heat exchange? Also, as you said, increasing the surface area is important. From the perspective of a surface, a polished surface has the smallest possible surface area. Anodizing chemically etches the surface and results in a larger surface area. By how much I don't know. Finally there's the issue of material longevity. Bare aluminum will form an oxide on its own, and eventually that oxide can (depending on the alloy) get thick, thicker than an anodized surface which I believe will stop the oxide growth to some extent.

[aburton]

It's actually a very complicated issue, and we're actually getting into some basic fluid mechanics. On a very, very small scale (think atoms), the interface between the air and aluminum is what's called a "no slip condition" (http://en.wikipedia.org/wiki/No-slip_condition). In this small region, the air acts as a conductor, not as a convector. As you move away from the surface of the aluminum, the atoms begin to move more and more, and the collision of molecules transfers the kinetic energy of heat through convection. You need good conduction to have good convection.

There are three forms of heat transfer: convection, conduction, and radiation. Materials which are transparent to infrared, like air, do not absorb or emit radiation. Thus, you would not have microscopic radiant heat between molecules.

Aluminum is EXTREMELY reactive with oxygen. It's virtually impossible to expose a piece of aluminum to the atmosphere without it immediately forming a thin film of aluminum oxide. If you scratch the surface of that film, the surface of the scratch will immediately form an aluminum oxide film. Once the film has formed, it doesn't spread down any farther through the aluminum. The entire purpose of aluminum anodization (other than the obvious aesthetic one) is to increase the thickness of this layer to an unnatural thickness. I think what you're thinking of is a 3000 series aluminum, which has pretty poor oxidation properties, and tends to get some white scale on it when it's left outside. 95 times out of 100, you'll be a 6061 or 6063 aluminum to make your housing, and this won't be an issue.

You will find that chemically etching the metal, as in anodizing, will not give you a measurable increase in heat transfer. Try to put thermal fins (like a CPU heat sink) on the body, or better yet, try to improve the thermal interface between the heat source and heat sink. Thermal greases usually outperform epoxies, so you might want to use a mechanical fastener to physically attach the LEDs, and a thermal grease to actually transfer the heat.

Coating or anodizing an aluminum housing *will* decrease the amount of heat transfer from this application. Perhaps not a significant amount, but it should be made clear that coating the housing is not that important. Especially from a functional standpoint.

[kc6qhp]

Wow, very interesting! The no slip condition really make a lot of sense and now I can visualize what is going on at the interface and how the heat exchange takes place. I machined the 'fins' on the ehat sink as a way to increase the surface area and it seems to work well. I probaly could have gotten away without them, so further improvements aren't necessary at this point. I'm using thermal grease between the heat spreader and the housing/heatsink. I'm surprised really that thermal grease does better than silver epoxy. The manufacturer states that the % by weight of thermaly conductive material (ceramic or silver) is higher in the grease by 10-20%! I will change the wording in this section to reflect what I've learned. Thanks!