Transistor selection, PWM fan control

[skimask]

Ok Thanks, I'll try some of those.
I like the to-220 package better. Do you think it will get hot mounted in a small project box with the rest of the circuit?

Only one way to find out. You can do the calculations based on the fan's current, PWM rate, duty cycle, thermal resistance of the MOSFET, dissipation of the device based on voltage/Rds(on)/current flow, etc.etc.etc...or you can just try it. One thing for sure, if you use it without a heat sink to save a buck, it's a dumb way to find out if the pieces/parts will heat up or not.
What is the end result of this project anyways? Are you driving an automotive type radiator fan for the purpose of being an automotive type radiator fan or something else?

[124C41]

I knew there was a formula, I kept getting stuff about air conditioners and science projects when I searched.

Yeah, It reads a coolant temp. sensor and controls the radiator fan speed. This is my first time using PWM, got started yesterday on a breadboard with a 80mm PC fan. 4 kHz was noisy, 20kHz is working real well. I’m more excited about PWM than I should be, I’ve been connecting other things to the circuit to see how they respond.

Ordered some MOSFETs , looks like they will be useful for all sorts of mischief.
Thanks for you help.

[skimask]

Yeah, It reads a coolant temp. sensor and controls the radiator fan speed. This is my first time using PWM, got started yesterday on a breadboard with a 80mm PC fan. 4 kHz was noisy, 20kHz is working real well. I’m more excited about PWM than I should be, I’ve been connecting other things to the circuit to see how they respond.

I can't think of any vehicle application that actually uses full PWM to control a radiator cooling fan. 2, maybe 3 or even 4 settings, maybe...but not a full range of 0% to 100% duty cycle controller of a fan motor.
Sounds like some serious overkill...