Transient voltage spike snubber question

[boylesg]

In the circuit below I have found that the 4 x 75V/5W series zener snubber is not sufficient to deal with the voltage spike created by a flyback transformer from a tv when it is running at about 20kHz - the zener diodes fry to the point of smoking.

So I have calculated and implemented an RCD clamping snubber as per here: https://www.daycounter.com/Calculators/Snubbers/Snubber-Design-Calculator.phtml The snubber can handle 74W assuming a peak inductor current of 8A from a car battery (will introduce an approrpriate resistor in order to limit the current to about that).

What if I was to leave the zener diode snubber in place and have the additional RCD snubber? Would the power of the transient spike be dissipated across both snubber or would there be some unanticipated (by me) effect that would cause one or both to fail? The reason I am asking is because I was formerly considering having more than one strings of zener diodes in parallel to share the excessive power until I read some where that one string would conduct until it burns out followed by the next..........

 

[FvM]

High voltage zeners have a positive TC, so there's a good chance that paralled diodes share the power dissipation fairly equally. My suggestion in your previous thread was sligthly different:

Then there's no other means than further increasing the total diode power rating, either using diodes with higher individual rating, or more diodes of lower zener voltage in series connection.

Additional RCD snubber should work too, although introduce unwanted statical power dissipation if connected parallel to the Z-diodes. The more interesting question is, what causes most of the input power to be transferred to the snubber diodes rather than the output load?

 

[boylesg]

High voltage zeners have a positive TC, so there's a good chance that paralled diodes share the power dissipation fairly equally. My suggestion in your previous thread was sligthly different:



Additional RCD snubber should work too, although introduce unwanted statical power dissipation if connected parallel to the Z-diodes. The more interesting question is, what causes most of the input power to be transferred to the snubber diodes rather than the output load?

As I understand it, the only way to transfer this kick back current to the load, in this case a flyback transformer, would be to use a non-dissipative inductor based snubber where you store the kick back voltage and then connect the inductor in series with the battery when the mosfet turns on again.

Is there any reason why you could not just use a free wheel diode across the inputs of the flyback transformer in this situation so as you don't have to worry about power rating of resistors etc?
For example a FRD 31DF6 ultra fast rectifier diode with a reverse recovery time of 35ns - which is comparable with zener diodes.

Can anyone explain why you would choose a RCD clamp snubber over a free wheel diode or a free wheel diode over a RCD clamp snubber?