Why is this Zener across the collector and emitter of this transistor driving a solenoid coil?

Question

I am trying to understand a vintage (1970s) impact printer circuit. Each of the pins in the printer are driven by a small individual coil driver circuit, that was originally specified to look like this:

(Detail from a schematic here. )

I understand how this works since I asked a question about it earlier and got some great answers, but the company that sold the printer issued an addendum to the circuit with an update, where a 62V Zener diode (1N4759) is added from the collector to the emitter:

(SWTPC PR-40 addendum detail.)

What is the purpose of this additional diode?

I'm assuming it's to protect something (the coil?), since it was an after-the-fact addition to the circuit design, but I'm having a hard time intuiting what's happening here.

I don't think the Zener will do anything in normal operation-- the 40VDC rail is below the 62V breakdown voltage so it won't bypass the transistor. Is this a protection for if the 40V rail somehow manages to jump up to > 62V? Is something else going on?

Answer 1

It's there because the coil has inductance. When the transistor tries to turn off the inductance fights against the dying of the current by changing voltage in the direction that tends to maintain the current as much as possible. That means it goes from about +40V to 0V and continues to go negative until something stops it. That something could be the transistor breaking down (which is very hard on the transistor and could certainly affect reliability) or it could be something across the coil or the transistor conducting. Or it could just resonate with the coil distributed capacitance and ring for a while if the transistor has a high enough breakdown voltage (it might be hundreds of volts though).

If we didn't give a toss about speed we would just slap a (normally reverse-biased) diode across the coil and the Vce of that transistor would be limited to maybe -41V, since the coil voltage would only go to one diode drop below ground. Easy and cheap. But that's a bad solution if we want the maximum speed since the current will continue for longer that if we allow the voltage across the transistor to get higher, perhaps to -62V by using Zener diode.

That's harder on the transistor, but not as hard as letting the transistor absorb the energy that now goes into the Zener diode. The total energy in the coil is \$ L I^2/2\$, and the initial current will be the coil current when on. Some of the energy goes into the coil resistance, some into the Zener diode, and some goes into the transistor as it is trying to switch off.

The transistor will have a specification called SOA (Safe Operating Area) that indicates the allowable current and voltage that transistor can reliably handle.

Answer 2

My question is: what is the purpose of this additional diode?

It's to protect the transistor from excessive back-emfs when it tries to deactivate the solenoid, Energy gained in the magnetic field of the solenoid has to be released and, to avoid an excessively high voltage being generated, the Zener diode provides a path to safely discharge that energy.

This sort of thing is normally achieved using a diode directly across the solenoid but, if you want to deactivate the solenoid quickly, you need to burn off that stored energy at a higher rate than a simple diode achieves so, the Zener diode allows the collector to fall 62 volts below its emitter. The transistor has to be rated to withstand probably 80 volts or more for this to be reliable.