Note the current from the Opto-Coupler, displacing the "myth" mentioned above in #16 that's it's going to "blow up" due to the excessive 1A base current required by the output pass NPN at Turn On ![Face Palm :palm:](https://www.eevblog.com/forum/Smileys/default/xfacepalm.gif.pagespeed.ic.EBDwh1hCfo.png)
Yes, NPN Opto couplers will naturally current limit, as they are current-gain devices.
That said, they still need some care as they have a puny 150mW NPN rating, and scant SOAR information.
LTSpice gives optimistic CTR values, the real high voltage Optos have poor specs, they mostly just give 20% or 50% min and 600% max.
That means getting enough drive for a lazy BJT/opto pair, has power envelope issues with a high gain opto, especially at current limit active.
That means they will need 10mA or more LED drive, and finding high voltage NPN parts with good HFE is also difficult.
Strangely, at 400V and above, the PNP FZT958 looks better than any of DiodesInc's NPN parts.
An opto coupler with MOSFET output aka telecom relay types with a surge resistor+BJT may be a better drive choice.
eg a GAQY216S is 600V and shows 80mA average 1A peak On current, 35 Ohms typ at 5mA IF, and 30c/500 at lcsc
Addit: another candidate part is something like BCW214A, which is a Dual SSR, rated to 140mA continuous and 600mA peak/100ms per channel.
Parallel the outputs and series the LEDs, and it's done in one part.
A part like CPC1983B has a higher price, and higher ratings.
It has a curve showing 3A @ 10ms, 2A @ 100ms, 1A @1s, and 500mA continual. For 20mS it is rated > 2.5A