Surge Current and Power in FETs

Care to explain what is wrong about it?

No because I have done so before and there wouldn't be any point repeating it.

It's hard to train operators to flip switch slowly, but they can't do it too fast with opening switch box. The power caps are pretty big. I think it's 5000uF 170V.

Ok, so it was something that was based on a misunderstanding of the system then.

So you'll have to decide whether it's NTC, resistor, FET or some combination of those. Or one of the other suggestions such as a series inductor and TVS. Assuming you want to add an inrush current limiter at all.

Yes but not my misunderstanding.

Obviously it was. In fact, that is likely why you won't even talk about it other than to play your games.

I expect that from many here.

"Quick as possible" does that really mean100 ms or just 1 second ? Anyway, the pressure of a gas does not immediately reach final value when the pump is started. For faster pressure variation, let the pup run continuously into a tank and use a fast valve to let some gas through, as in normal pneumatic systems.

Regarding the PID startup issue:

1.) Modify the PID coefficient during startup, e.g. reduce gain and when up and running. restore original coefficient, 2.) With a fixed predetermined ramp, accelerate the motor to at least the minimum normal operation speed and only then, give control to the PID. You may have to reset the PID sometimes during the ramp to get rid of long integration effects or set the I-term to zero during acculturation during the startup ramp.

Accelerate the motor with a ramp and start the ramp slightly before full power is needed.

Do not let the motor stop completely to avoid startup friction during next startup.

A gas tank and a pneumatic valve will let the pump run constantly and is capable of much faster cycles.