power transistor lifespan

[Nick Nack]

Forgive me if this is the wrong forum.

I know there's a rule of thumb that a semiconductor's life expectancy halves with every 10 (or 15) degrees above 25, but what is the expectancy at 25? The power transistors I'm using are rated to 200 Deg and I expect the junctions to be as high as 80 or 90, though it's hard to know what the resistance between sink and die.

 

[JohnRoberts]

Forgive me if this is the wrong forum.

I know there's a rule of thumb that a semiconductor's life expectancy halves with every 10 (or 15) degrees above 25,

MTBF drops in half for every 10 decrees C temperature rise.

but what is the expectancy at 25?

25C is probably very good.

The power transistors I'm using are rated to 200 Deg

200'C? Device technology has improved over the years. My memory is a little fuzzy but as I recall plastic transistors were limited to something like 150'C to insure package stability, but I suspect that has improved. TO3 (metal) power transistors were good up to something like 200'C . Silicon can tolerate 300'C but even metal packages will not be reliable much above 200'C.

I have been out of the trenches for years and power transistors were just starting to transition to plastic packages. Amp designers pushed them harder and they survived, so answer is "I don't know" what practical upper temp limits are for modern plastic power devices. Also in the middle of all this, class D power amps became cost effective, so audio power amps with huge power dissipation have gone the way of the dodo bird.

and I expect the junctions to be as high as 80 or 90, though it's hard to know what the resistance between sink and die.

Most device data sheets specify theta (thermal resistance) Junction to case. I recall trying to keep case temperatures at or below 100'C

JR

PS: One of my latter patents at Peavey was as co-inventor for an audio power amp heatsink design, but like I said already I have been out of those trenches for years. (6,515,859 Roberts , et al. February 4, 2003).