>  Since he is using a PNP transistor, he has it configured in the
>  upside down form which is correct for a PNP transistor.   This
>  is so that you can relate it to an NPN transistor so that the bias
>  polarities are correct.  So, in connecting a resistor from Base to
>  Emitter, this is similar to connecting a resistor from Base to
>  Collector on an NPN transistor.  Except that on a PNP, this resistor
>  will force the transistor to turn off until driven low, which turns
>  it on.   In an NPN transistor, this resistor would keep it turned on
>  until drive off by a low.

Yes, I understand all that (and have for about 25 years, some of which were
spent earning a masters degree in electrical engineering from RPI).

>  And as far as the PIC pin is concerned, it's general knowledge that
>  a TTL/CMOS output is better at sinking current than at sourcing it.
>  So, therefore, pull the pin up, and drive it low.

I agree with your facts but not your conclusion.  PICs use totem pole CMOS
outputs.  These actively pull both down and up.  It may be true that the
pullups are a bit weaker than the pulldowns, but shouldn't matter much
because there should be virtually no current when the base is driven to very
near the emitter.  In other words, we are looking at the PIC's open circuit
high output voltage, which is very close to the power rail.  The base won't
start drawing meaningful current until it is forward biased at least a few
100mV.

I think what was happening is that the transistor wasn't driven hard enough.
This kept it from going into saturation, which caused the voltage drop
accross it to increase, which caused it to heat up.  Its temperature
probably exceeded specs, which caused it to leak accross the base-emitter
junction, which caused it to stay on.  The correct solution is therefore to
provide more base current in the on state by lowering the base resistor to
1K ohms.


********************************************************************
Olin Lathrop, embedded systems consultant in Littleton Massachusetts
(978) 742-9014, olin@embedinc.com, http://www.embedinc.com

--
http://www.piclist.com hint: The PICList is archived three different
ways.  See http://www.piclist.com/#archives for details.