mike said:
I stared at this for a long time. Still can't figger out what the diode
does. Depending on the source impedance, the base is gonna melt or the
bond wire will blast off long before the diode turns on??? Yes? No?
Stated another way, I ain't mollified.
Won't a couple of base resistors protect the base?
mike
It is assumed that the circuit is in series with a current limited load
- like 200mA. The diode clamps the forward bias of both b-c junctions
simultaneously to approximately 0.25V at this current, preserving
significant hFE in the transistors- as much as 20 @500mA Ice, so that
the base current is now limited to Ice/hFE or about 5mA in this
application. Now the majority of current flows through the
EC,pnp->CE,npn path, so that the total voltage drop is ~ 2x Vce +Vdiode
, and Vce~Vbe,sat-Vdiode for a total of 2xVbe,sat-Vdiode where Vbe,sat
is the BE voltage at the edge of saturation for Ic=200mA- approximately
the same for both transistors. The circuit can be used safely to switch
as much as 500mA. Without the diode, the current splits into two
parallel paths (EC,pnp+ BE,npn)||(EB,pnp+CE,npn) and the split will be
essentially equal since the transistors are complementary- so that
contrary to WHill's slander, it is quite safe and reliable at 200mA.
There is no need for series base resistors and they would only serve to
greatly increase the on-voltage across the circuit. Also, the Schottky
does tend to leak profusely so that the base pulldown resistors should
be reduced to something like 470 ohms - reducing triggering sensitivity
which is a non-issue in this case. The thread was all about another
failed attempt at slander by Hill: first knock the circuit for using one
extra part, then knock the circuit on the technical merits implying the
basic topology is unsalvageable, then get butt handed back to himself,
then return to knock circuit on using one extra part, etc...
