Aluminum Heat Sink

The best approach is to use similitude. This means that you take series of profiles which are for electronic cooling (usually aluminium extruded) and analyse their geometry and design your system according to it. In different catalogues such profiles are given with all details and with their transfer characteristics for free convection conditions.

If you need more input after you made your sketch then send the drawing with dimensions (as a JPG picture) and expected power dissipation or expected thermal resistance in °C/Watt and we shall try to give you more indications.

It would be good to look at a book about heat transfer as well in the direction of conduction as in the direction of convection. You have also to decide if you want a free convection or if you will use a fan for forced convection. This will have an impact on the dimensions.

Thanks.

I have already designed a shape.

The design must consider free convection. Final products are solar-powered LED street lamps. I intend to build a few ones with different power rates.

Will try to send you the drawing with dimensions and the expected thermal performance.

This seems pretty much like an academic exercise. If you don't have anything like ABAQUS, ANSYS or ALGOR to do this for you, then you need to calculate the heat transfer through the aluminum by using Fourier analysis (good luck with that, lots of paper and pencils) and then calculate the heat radiating into the free space around the sink (usually upwards) across the metal to air thermal discontinuity.

You could always build one and test in a controlled environment to measure operating temperature versus environmental temperature to get yourself a decent curve of performance.

It is not necessary to go to such a complexity as you think. There are already for different geometries available equations. The body can be parted in a series of simple bodies. The only real problem is the convection coefficient which depends on air speed at the wall and this is difficult to define at every point so that again the surface are cut in slices. In general the high conductivity of aluminium leads to an almost constant temperature of the base and only the fins have to be analysed as temperature drop and only in forced convection. The errors are within acceptable limits.