THE MAGAZINE FOR COMPUTER APPLICATIONS
Circuit Cellar Online offers articles illustrating creative solutions
and unique applications through complete projects, practical
tutorials, and useful design techniques.

THERMAL CONSIDERATIONS IN ELECTRONICS DESIGN

by George Novacek

Start ý Why Worry? ý Reliability ý Design Steps ý Practical Example ý Heatsink Selection ý Sources and PDF

HEATSINK SELECTION

The heatsink, or more accurately, the heat exchanger selection, is a subject in its own right. A wealth of design information can be found at heatsink manufacturersý web sites. Another excellent source of heatsink design information is the International Rectifierýs Power Modules Designer Manual. The main problem is that the heatsink-to-ambient thermal resistance (T_ha) depends on many factors, such as the orientation of the heatsink within the cabinet, exposure to air movement, and even color and the surface finish. For example, while polished aluminum heatsink radiation cooling efficiency is about 0.1, black anodized finish can achieve 0.9.

You will find heatsinks predominantly made out of aluminum. It has a volume heat storage capacity of 40.5 J/in.³ýC. Although there are materials with higher storage capacity (i.e., brass, 50.5; copper, 57.5; or nickel, 67.0), aluminum wins hands down in heat storage capacity per weight, with 413 J/lb.ýC. Brass, copper, and nickel are 165, 178, and 208, respectively. Although its thermal conductivity is not particularly high (gold, copper, and silver are higher), it is reasonably priced, easy to machine, and corrosion-resistant. Consequently, exchangers made of material other than aluminum are the result of specific design requirements.

VERIFY BY TEST

I canýt overemphasize the need to verify the calculations by test. Although the semiconductor manufacturerýs catalog characteristics are worst-case, these conservative numbers should be taken at face value. There are two quantities you need to confirm by test because they can have a significant effect on the final result. The first one is the heat exchanger thermal resistance. As I stated, too many conditions that cannot be accurately forecast will affect the heat transfer efficiency. The only way out of the dilemma is to measure a prototype in the intended environment.

The second quantity that affects the result is the dissipated power (P_D). As with the example, its worst-case scenario can be determined accurately for DC operation, but an AC or pulsed operation is a different matter. Although the mathematics and the necessary data to plug into equations certainly exist, experience with thyristor and PWM drives taught me to accept the numerical results as guidelines at best. There is nothing more convincing than hard data obtained by measurement.

PREVIOUS • NEXT