Editor’s note: This question was asked in response to Electronics Cooling’s recent webinar by Roger Stout. To view the webinar, click here.
Question: I didn’t understand what you were getting at with those “normalized transient thermal response” curves? Can you be more clear?
Answer: Yes, I worried about that. There really wasn’t time to get into it to the depth it deserved. The problem is, short-time transient thermal response does, in fact, depend very strongly on the die size or active area – and, for a given device, on nothing else. When you “normalize” a heating curve, you’re saying that the entire curve, from short-time to steady-state, goes in proportion to the final value, which is patently ridiculous. In other words, just because you mount a device on a super heatsink such that the steady-state theta-JC drops, say, 80%, the short-time response at 1 millisecond won’t be any different at all. Normalizing the curve makes it look like the short time response also drops 80%, as if the die size magically grew by 5x! That’s one of my pet peeves. Normalized transient response curves should be outlawed!
– Roger Stout
