Paul,
QMX has no spike voltage issue at end of transmission because of the built-in wave shaping system.
Regarding Zeners for QDX, they may provide some protection against the hi-Z SWR case, but they are not a good solution to the inductive spike problem. At the moment the BS170 transistors turn off, there is approximately an amp +/- of current flowing through L14. That current will now go to the Zeners, where it most likely will be shared unequally as they are not identical. The datasheet for 1N4756A shows a maximum current of 95 mA.
The voltage at the BS170 drains will be well above the Zener voltage of 47 volts, as that high current and the Zener diode resistance conspire to elevate it well above the BS170 breakdown spec. The diode instantaneous power will be very high, perhaps as much as 75 watts each for a very brief time.?
These are not healthy conditions, either for the Zener diodes or the transistors. A commutating diode across L14 which is capable of handling the coil's current avoids all these problems.
JZ KJ4A
