I'm designing a 24V to 350V DC-DC Converter based on a H-Bridge. The power requirement is 500W and the circuit operates at 20KHz. The design works fairly well and I've achieved about 90% efficiency at 200W load. The main issue with the circuit is ringing. The waveforms distort/ring when the transformer is connected to the H-Bridge. Without the transformer the waveforms are extremely clean, even under load. Picture below shows waveforms with the transformer connected but without any load.
I found that adding a capacitor across all my FETs helped minimize the distortion severely. Here's a picture from my o-scope demonstrating this (left is without load, right is with 200W resistive load). Note that the output from the transformer is rectified with a full-bridge rectifier and smoothed by a capacitor:
So my question is: why are capacitors across my FETs minimizing the distortion? What is taking place in the circuit? I initially added a RC snubber across the FETs but the circuit works much better without the resistors and just the capacitor!
Here's a picture of the schematic and the layout:
I'm not entirely sure which picture is for which configuration but adding the capacitors is not a good idea because they will reduce the efficiency of your switching circuit. The transformer produces the ringing because it is inductive and it has self capacitance - together, when the FETs switch you will get a resonant tuned circuit and ringing will occur.
I'd do nothing other than try to improve your transformer. Maybe you should give details of this. There are also other effects that can appear to cause ringing namely your o-scope earth point not being close to your measurement point and getting an induced voltage in that loop from the switching currents. This commonly happens. Make sure your measurement technique is sound.