MOSFET Heating Question

Hmmm seems odd...I'd written one reply but deleted it on re-reading your post.
(I rarely read anything correctly first time )
Maybe a small current sensing resistor with a 'scope across it would tell you WTF is going on.

When you say in-circuit, do you mean it's a surface mount device soldered to the pcb or is it standing up on it's legs.

Break out the 'scope and the ouija board .

Del

Although you are all nature English speaker, we still guess a lot. is this procedure?

How many voltage did you apply on the fet when you measured them? The power dissipate is 2 (your current) x tube voltage drop

Maybe your work station is different in the two ways. ie. different power dissipates.

Do you know heat ohm law? if there would no be a fun in the box, the condition of on baord situation is of cause worse than in free air. if it would be a surface mount soldered on the pcb rather than it stand up on it's legs, that would be more worst.

so the two condition you hav to describe to get clear answers.

best bet is take a photo to show us.

295C = 563F

That seems a bit warm to me.

I agree

Thet is a bit warm

If that is case temp then the junction temp is even higher

might be some measurement error.

incircuit is it enclosed? if there is no air circulation then there can be a heat build up.

Also are all the conditions? the same?

measure all voltages, currents and signals and compare

Maybe it`s a problem of temperature measurement instead of the MOSFET`s physics laws.

I have found rather high differences in temperature readings when measuring contactless with IR cameras or IR sensors because the reflection factor determining the reading has to be adjusted if you take the samples on different spots or under different angles - which may be the fact when the semiconductor is mounted (compared with free-air measurement).

Regards Uwe

Do you mean: In Circuit Test? i.e. with a vacuum fixture.

Overheating was a common problem with vacuum fixtures: very little air to cool the component.

Change to a "Top Hat" fixture.

What is the part number of the MOSFET?

Wow, thanks to all of you for your quick response.

Let me answer some of the questions first. The FET is an FDD3706 and is installed standing up with no heatsink. What I meant by in-circuit was that the FET was soldered to the board as intended. The free-air test was done with the FET removed from the pcb, set on the test bench and alligator clipped to a dc power supply set to 4 volts and current limited to 2 amps.

Our next procedure which we'll get to today is, as Del noted, 'scoping the FET to obtain some traces. We're also going to repeat the tests with several more samples - perhaps we just met an anomaloy.

Thanks again folks

Grae

and current limited to 2 amps.

Maybe that's the answer....
When it's in circuit is it driven from exactly the same supply?

Del

Hi Del, nice to have you so available.

The circuit is normally battery operated, but for these tests, we need to use a controllable power supply.

Batteries are difficult to adjust. Every time we twist the little knob on top the damn thing breaks off or the battery explodes. Go figure.

Grae