@NickAlexeev Looks like Dave already google translated it, but it was simplified Chinese, for the record.

@ScientistSmithYT Did you start the arc at a shorter distance? Or with a higher voltage (for example due to cutting off current to an inductor)? It doesn't take as high a voltage to sustain an arc as it does to start it.

@ThePhoton I started it out at a shorter distance. I didn't measure the starting arc distance so much as the end arcing distance. But I approximate it to have started arcing at around 0.5-1 mm. (The Breakdown Voltage of the surrounding air)

@ThePhoton I started the arc at the same 2,300 VAC. I also used 3 1.05 uF HV AC capacitors in parallel.

@ScientistSmithYT 2300 * sqrt(2) = 3250 V peak. At 1 mm that's within a factor of 2 of the usual 6000 V/mm breakdown voltage for air. If your electrodes are a little bit pointy, it seems reasonable you could start an arc with that.

Folks, what's the best practice for layout of a switcher (buck or boost, non-isolated) on a 4-layer board?

What should I do with the ground plane?

Should I make an island for the switcher's PGND? Should I create the PGND pours on the top and bottom layers and keep the ground plane intact? (ping @AdamLawrence )