Hi,

I've designed a simple two stage op amp, and everything works up to ~100 deg celcius. Once I reach high temperatures like around 100 degree celcius, my pmos active load of the diff amp and the pmos of the 2nd stage amp both go into triode.

Can someone please advise why this may be and anything I can do to fix it?

Thank you!

 

at high temp my loop gain drops from ~80db to ~40db
phase doesn't change

I also noticed that decreasing my pmos W/L f by increasing L fixes the problem.. but I don't understand why?

 

All semiconductor properties change with temperature because silicon has a significant coefficient of expansion. Transistor gain, diode forward voltage drop, Rds, leakage current - everything changes. Check the transistor data sheet for various properties tempcos.

ak

 

Look at the thermal characteristics of the device in its datasheet. They often have graphs and tables saying how things change with temperature.

When you design a circuit you really need to read the datasheets of the components you are using, and make sure your circuit can handle any fluctuations in how each device operates. That's a critical part of circuit design, and without reading up on how your specific components work, you will likely run into trouble, just as you have this time.

Regards,
Matt

 

I am no expert when it comes to cmos and pmos devices, but I agree completely with AnalogKid. You have to consider all the variables that are taking place when you are near the 100 degree point. If you are driving inductive loads: you can expect inductive kickback to cause unusual results.

Changing the W/L may be causing the circuit to become tuned to a more efficient frequency, and not causing any of the devices internal protection to engage. Try adding a larger heat sink, or even simpler: blow an external fan on the device to see if it lets you drive the load harder without experiencing the problem. We would have to see the full circuit for more thorough troubleshooting.

Hope that helps!