• WARNING: Tube/Valve amplifiers use potentially LETHAL HIGH VOLTAGES.
    Building, troubleshooting and testing of these amplifiers should only be
    performed by someone who is thoroughly familiar with
    the safety precautions around high voltages.

SS rectifier diodes for 275-0-275Vrms


Long story short: I need to extract a negative rail from a 275-0-275Vrms transformer winding. As the transformer is a Hammond, we can assume the voltage will be a bit higher in reality. The negative rail will be used for some Mosfet followers direct coupled to a quad of 6S4S (soviet 6B4G) so a sudden loss of voltage due to diode failure would be distastrous.
Would one 1N4007 or UF4007 per "phase" be enough or should I put two in series?
My choices are pretty much limited to: 1N4007, UF4007, 1N5408 (all 1000V PIV) or C3D02060F (600V PIV SIC Schottky).
275 volts AC RMS will have a peak value of SQRT(2) x 275 volts or 389 volts. On the negative half cycle it will be -389 Volts. So there can be 778 volts across a diode connected from the voltage source to a capacitor.

If you use a 600 PIV diode, fireworks should be expected.
  • Like
Reactions: 1 user
Yes, ordering higher PIV diodes might be the best option. I just saw that George uses single UF4007 for the exact same job in his TSE II boards, where many builders use transformers with quite a bit more voltage. OTOH, I just saw a thread where on of those diodes had split in half, resulting in a burned out mains transformer...
Joined 2011
You don't wanna run semis close to their spec limits. I routinely double the required diode voltages, at least.
If your job (or your house burning down) depended on it, why would you not properly derate parts?
Too difficult to multiply by two?
I agree, ordering more suitable parts is the best solution here. I could live with pushing the ratings a bit if it was the B+ supply, but the negative rail has a bigger potential for distaster.
The best I can find locally is BY255 (1300V 3A) or BY133 (1300V 1A), should be enough?
Joined 2011
Certainly those are better than 1kV parts.

Perhaps a half century ago, using series diodes (with the required balancing resistors) was reasonable,
but not today with better processes. Why have lower reliability due to more parts and connections?
This is like saying "I drive my Ford Pinto every day, and have never had an explosion".


  • impala-vs-pinto.jpg
    18 KB · Views: 22
Solid State Rectifiers . . .

A word of wisdom:
Use a good spring clamp that heat-sinks the wire At-the-diode body, when you solder them into the circuit.
Soldering heat sometimes destroys semiconductors, but much more often it Fractures the Plastic case, exposing it to the environment.
You may not even see the Fracture, but the environment "will".

This is my experience with a test and measurement manufacturer that built military ruggedized test equipment, and had a world class reliability department.


Using 1 K PIV diodes would work in a bridge. But with a center tapped transformer and just two diodes caution should come into play. But you already know that.

I once worked on a project where the lead didn’t quite get it. He had 10,000 PIV diodes hooked up to a high voltage transformer. He was disappointed that when he increased the voltage to just above about 3,500 VAC the diodes blew up with a bit of a show. I did explain to him that at that AC voltage he did surpass the rated voltage.

The other issue that comes into play is tolerance. An 800 volt diode will of course be good for 800 volts and even if it didn’t break down until 999 volts, it would still be rated 800 volts. As a 1N4000 series diode tops out at 1,000 volts, some rated at 1,000 volts probably work a bit above that, good practice of course limits a decent designer to actual specifications.
Last edited:
  • Like
Reactions: 1 user