>
>
> > hFE doesn't matter much. My DC calculated load is around 350ma @ 12v DC=
;
> in
> > fact, I would be using PWM with variable duty cycle so the current will
> be
> > lower than calculated.
>

Whether current gain matters depends on the current drive capability of the
source and the frequency.
At beta=3D30 (as specified in one  suggested device) Ib =3D 350/30 ~=3D 12 =
mA.
That's beyond the dive capabilities of some microcontrollers.
A microcontroller's max pin drive spec may be 10 or 5 or even 2 mA max in
some cases.

If you are using on/off PWM you do not need a "5 Watt" transistor.
IF you can get Vsat =3D say 0.3V you dissipate about 0.1 Watt worst case.
Lower Vsat =3D less dissipation.

An eg BC337/327 or BC817/807 (NPN/PNP   TH/SMD   TO92/SOT23 ) would work
with good base drive.
Get the -40 version with beta >=3D 250 (250-600 with mean =3D 400) and base
drive can be well above the notional 350 mA / 250 ~+ 1.5 mA needed giving
good saturation voltage and low dissipation.
The SOT23 parts would need thermal calculations and Vsat checked but should
be able to be designed to wpork well.
The through hole parts would handle this easily.

Lower than higher base drive will make switching faster, all else being
equal.

What is the drive source?
What drive current capability.
What is PWM frame rate and min/max on % required (apart from 0 and 100)
eg a 10 kHz PWM frame rate with a 1% step size gives 100 uS frame length
with 1 uS pulse width at 1% on.
IF you want a 1% pulse width at a 10 kHz rate you need a switch that will
operate at some MHz.
"Easy enough", but not trivial.



         Russell McMahon
--=20
http://www.piclist.com/techref/piclist PIC/SX FAQ & list archive
View/change your membership options at
http://mailman.mit.edu/mailman/listinfo/piclist
.