Ok, I retract my suggestion of any current-limiting resistor on the base. No purpose to it. A nice feature of the emitter follower configuration.

 

Ok, I retract my suggestion of any current-limiting resistor on the base. No purpose to it. A nice feature of the emitter follower configuration.

Not this application - but someday you might find use for a stopper resistor.

Probably not so likely with an emitter follower, but its a trick up you sleeve for that; you never know moment.

For anything to do with current control - calculating the resistor isn't a precise science, so better avoided.

 

I have hit a problem buying a bigger fan, transistor doesn't get hot on full speed but low speed it does. I think I have done the maths right and worked out 1.8w at low speed settings, 0.4w at high speed. Does that sound right? I measure 10v from C to E and 0.2A at low settings and 1v C>E 0.4A.

Maybe I can link up three of the little guys. On second thoughts I might go and buy a one rated for better power dissipation.

 

I have hit a problem buying a bigger fan, transistor doesn't get hot on full speed but low speed it does. I think I have done the maths right and worked out 1.8w at low speed settings, 0.4w at high speed. Does that sound right? I measure 10v from C to E and 0.2A at low settings and 1v C>E 0.4A.


Maybe I can link up three of the little guys. On second thoughts I might go and buy a one rated for better power dissipation.

10V @0.2A is 2W,

This is why pwm is better, no power wasting due to high voltage drop across the transistor.

 

I have hit a problem buying a bigger fan, transistor doesn't get hot on full speed but low speed it does. I think I have done the maths right and worked out 1.8w at low speed settings, 0.4w at high speed. Does that sound right? I measure 10v from C to E and 0.2A at low settings and 1v C>E 0.4A.

Maybe I can link up three of the little guys. On second thoughts I might go and buy a one rated for better power dissipation.

Its all about VxA=W. At half way; the transistor has current and voltage about half maximum which multiply together and make Watts. Flat out; the transistor has all the current but no voltage, at minimum; all the voltage at more or less a tiny leakage current. PWM takes advantage of this fact by making sure the transistor is only ever saturated or cut off. The amount of time its on determines the average power.

Transistors are harder to parallel than MOSFETs, but MOSFETs can be tricky to bias in linear applications - you'll probably need to add current sensing nfb.

If you're not going PWM' a bigger transistor is the easy answer - a bigger Darlington transistor would be easier still.