thermal shutdown

Nice part, thanks.

I'll probably stick with what I have. It's close on parts count and actually a bit cheaper.

It's a nuisance to use a new part. I have to fill out forms.

John Larkin Highland Tech Glen Canyon Design Center Lunatic Fringe Electronics

aren't you the boss? ;)

Sure, but I have to fill out a form.

To tell production what our stock number will be, and what mfr/mpn's to buy, and how many to stock.

And I have to tell my PCB guy how to create the part in the PADS library, including what I want the schematic symbol to look like.

The form also allows me to make notes about this part.

Nobody else can decide all that. It might take days, or a week, for the PADS part to be created, checked, and released to the library. Meanwhile, my schematic is stalled.

We have

In stock 8,428 4,808,803 $2,235,140.75

so I don't want to create any more if it's not necessary.

John Larkin Highland Tech Glen Canyon Design Center Lunatic Fringe Electronics

If you have a reference voltage available, it would be better to use a bridge for the comparison. When you rotate that V vs Temp graph, you have a pretty steep slope in Temp vs V out of the divider. This would make comparing to stable reference resistor ratios better than relying on a voltage source, especially the 12V. You can use a comparator with internal hysteresis built-in or add a smidge externally. Are you planning on latching the driver off, or allowing the DPAK to cool down and re-trying enabling the line driver?

The hitemp shutdown is to prevent frying the parts, which is a judgement call, so it needn't be super precise.

I figure that I can limit my pulse generator's voltage swing while the dpak resistor is too hot, and remove the limit after it cools down a bit. A user could ask for a +42 volt pulse and short our output.

I plan to add some hysteresis.

Something like this:

This shuts down at 100c and recovers around 75. That can be tuned later.

John Larkin Highland Tech Glen Canyon Design Center Lunatic Fringe Electronics