Cap lead broken, Zobel network resistor overheated. Are they linked?

[Chris Wilson]

Strange scope patterns from the output of my 136kHz Class D 1kW amp, both pre and post LPF. Took it apart and found one of the 2kV 4n7 caps on the primary of output transformer T1 had a broken lead. The 22 Ohm, 25 Watt panel mount metal cased resistor in the Zobel network (shown as "Z" in the schematic, a 22R wire wound and a 4n7 poly cap in series), had also got mad hot and exuded an end cap, explaining the hot smell! Resistor was still intact and still sat at 22 Ohms. Hadn't got a spare wire wound so refitted it, with a new 2kV DC poly cap. Trace now good. 50W wire wound on order. Would the effectively missing cap circled in red in the schematic attached cause the Zobel network wire wound to overheat, or do I have other issues? Thanks.

[Phoenix]

The snubber circuit (zobel as you and the circuit designer call it) may not be well designed, a poorly designed snubber will dissipate a lot of heat - it's a balance between loss and snubbing really. I also don't really like the placement of those 4n7 caps (like the one you circled), they might do more harm than good.

The purpose of that RC is to damp out very high frequency ringing caused by rapid dv/dt or di/dt exciting a resonance. If the RC break point is too low the circuit will dissipate more energy than it needs to damp the ring. If the resistor is too small it will also dissipate more energy. You really need to stick a scope on the drain-source and look for ring (without the snubber and 4n7) and compare to with them in place.

https://www.maximintegrated.com/en/app-notes/index.mvp/id/3835 might have a better idea on snubbing this topology.

[T3sl4co1l]

BTW, what's the transformer look like?