PS: now I recall that R 2K was not chosen just "empirically" of course,
but (although I'm a hobbyst) I made some engineering efforts on it.

Even in the worst case (400V) with 2K we will have 200mA flowing through
the base of the PNP transistor. This is well within its capabilities, as
it is within the capabilities of the NPN transistor (300mA max collector
current).

The Icbo of the NPN transistor was sheeted as being just 0.1uA max (with
Ices 0.5uA), so I thought that a base resistor in the PNP one wasn't
necessary (however, I left pads of it on the PCB). But as Vasile suggests,
I have probably underestimated EM coupling or miller effects. Let away
that I'm now afraid that those values of Icbo and Ices that datasheet
gives will be very different in case of higher ambient temperature.

DC-gain of the PNP transistor is ~25 and of the NPN is ~100.

Now I think that those pads must be really populated.. but how do I
calculate the value, given the "dynamic" nature of use of the circuit?

Also, base-emitter resistor on the NPN transistor wouldn't harm, but it's
probably really unnecessary as the PIC output forces it to logic "0" or "1"
(sure, there's the 200K pull-up, but if I put a base resistor into the
base-emitter of the NPN transistor, then I can say goodbye to the nice
"bootstrapping" feature that powers my MPU even when starting from total
inactivity).

With kind regards,
Mario

--=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
.