I have the following circuit connected to a transformer via J2. The schotkys are not installed
The transformer goes to a glass chamber to produce Ozone by coronal discharge in the 5-20KHz and 5-20KV range



Below are a series of O-scope screen caps of the mosfet gate and drain terminals

1.) MOSFET gate with NO rectified AC applied
Notice after the pulse, the signal is completely flat straight


2.) MOSFET gate and drain WITH 10V 60Hz rectified AC applied (4.48% duty cycle)
It seems the signal on the drain is affecting the gate signal when the pulse goes low


3.) MOSFET gate and drain WITH 10V rectified AC applied (14.06% duty cycle)


4.) MOSFET gate and drain WITH 10V rectified AC applied (22.3% duty cycle)


5.) MOSFET gate and drain WITH 10V rectified AC applied (48.32% duty cycle)


Can anyone provide some insight as to why the gate is being affected when it goes low? Is this parasitic capacitance/inductance?
Here's by board layout

 

For sure no clamps, no schotkys on MOSFET, will give you lots
of ringing on waveforms, which you have. In fact you might consider
using infinite persistence on scope, or change trigger point, to see if
that is violating HV specs on MOSFETs.

Gate R calculations -

https://www.diodes.com/assets/Calculators/BipolarGateDriver.xls

https://toshiba.semicon-storage.com/info/docget.jsp?did=59460

http://www.ti.com/lit/an/slla385/slla385.pdf

Capacitor ESR vs Technology -

Regards, Dana.