Scenario:
3-phase into a 3-phase rectifier.

The change:
Somebody substitutes the 3-phase input with a single phase input. As long as the ratings are not exceeded, no problem right? The ripple increases, but from the point of view of the rectifier, it is now a single phase full wave bridge that has a pair of unused diodes.

The problem:
What happens if somebody connects a neutral or center-tap to the last diode pair on the rectifier?

I can't see any problem with the open circuit output voltage of the rectifier under ideal conditions, but I'm not sure how things end up with a real load. Would the potential worst case scenario be double current on the neutral leg, or something completely different?

 

You should be fine, as long as the maximum reverse voltage of the rectifiers aren't exceeded.

See the attached. Between the neutral and the hot will be two opposed rectifiers on either side of the bridge.

Connecting the neutral up to the bridge won't hurt anything, but it won't help anything either.

 

You should be fine, as long as the maximum reverse voltage of the rectifiers aren't exceeded.

See the attached. Between the neutral and the hot will be two opposed rectifiers on either side of the bridge.

Connecting the neutral up to the bridge won't hurt anything, but it won't help anything either.

Any chance that heavily inductive or capacitive loads or high speed switching (PWM or PFC) would alter that, assuming that everything remains within spec for the recitifiers?

 

Not really.

A heavily capacitive load would just smooth things out.

Power rectifiers are relatively slow to switch on or off, so high speed switching loads would have no impact.

If a heavily inductive load was placed on the output, and current were interrupted, spikes would be shunted to the neutral. I can't see that being any problem, either - more of a benefit.