All-semiconductor regulated HT PSU

[Nick]

...This "voltage stress" has nothing to do with V-squared-over-R power dissipation - this is a separate mechanism. It can take many years to develop. And most film resistors are rated at 250 volts (within their power envelope)...

Agree with everything you've said here.

The point about the working voltage range for resistors is more often than not completely missed by most - if you buy a bag of mixed resistors on eBay (from wherever), how do you really know the SOA of that component?

The parameter you need to look for is RCWV (Rated Continuous Working Voltage) - you can sometimes work this out by taking the square root of the rated power * resistance - i.e. Vrcwv = SQRT(V^2/R * R). e.g. a 1W, 100K resistor should have a RCWV of 316V, a 2W, 100K would be 447V...

...However, sometimes the component has a lower LEV (Limiting Element Voltage) - the LEV is the safe voltage generally for the component, regardless of value - you need the manufacturer's data sheet for that.

So, the ACTUAL safe continuous working voltage of a resistor is the smaller of LEV and RCVW, i.e. Vwv = MIN(LEV,RCVW).

This point is SO important for safety, but so often missed. For high voltage work (anything over a 100 or so volts), you really need to make sure that all your components are rated accordingly, remembering to de-rate them if air-flow is constrained, the ambient temperature is high or the environment is humid etc. A general engineering principle for this arena is to allow between 50 and 100% headroom, e.g. if you calculate an 0.5W resistor would probably do, use a 1W; if you are working with 300V, use a 450V component etc.

Safety First.