Check out my latest contraption, a B+ regulator for tube amps, adjustable from 200 to 300 V. The error amplifier/shunt MOSFET bias servo op-amps are fed 5 V from a 7805-based external board placed right next to the regulator board.

I'd like to know if my protection measures are adequate. I'm relatively new to using op-amps in high-voltage regulators.

B+ regulator circuit

I have applied the protections I think are necessary: at the top of the schematic there's D2/R5 for discharge of the last filter capacitor in case of a short circuit, and D3 (UF4007 or the like) at the far right for reverse spikes. Prototypes seem to survive fine without them but of course you want protection in a final design; I admit I didn't try shunting the outputs on the prototypes :-)

The op-amp is nothing fancy, LM324 or similar single supply chip. The LM324 has EMI protection diodes on all pins; is this sufficient or should I fit reverse Schottky diodes from the error amp non-inverting input (TP6) to GND and/or VCC? Schottky diodes would be required because their forward drop is lower than the op-amp input's maximum reverse voltage allowed.

That being said, I'm not sure I like the idea to have reverse voltages shunt to the external regulator, and I can't use Schottky diodes to shunt to the HV rail because the voltage is too high.

In normal operation all other inputs should never see any reverse voltage since they are connected to DC regulators so no reverse voltage protection needed there, right? Last thing, should I fit a Zener diode as a crowbar on the error amp's non-inverting input (TP6) to GND?

  • \$\begingroup\$ What did your simulator tell about there effectiveness? \$\endgroup\$
    – Andy aka
    Aug 17, 2022 at 19:10
  • \$\begingroup\$ @Andy aka Efficiency is appaling like all shunt regulators :-) Efficiency is definitively not the goal here - I like shunt regulators' sound best. Also I live in Canada where it's winter half of the year, this is just heat my main home heating won't need to supply lol \$\endgroup\$
    – Joe
    Aug 17, 2022 at 19:26
  • \$\begingroup\$ I said effectiveness not efficiency. \$\endgroup\$
    – Andy aka
    Aug 17, 2022 at 19:37
  • \$\begingroup\$ @Andy aka Sorry I misread you. Sim gives an output Z below 2 mOhm from 20 Hz upwards, ripple 5mV pk-pk at 270V. Of course this is too good to be true because it's all theorical, I have not input parasitics of capacitors... \$\endgroup\$
    – Joe
    Aug 17, 2022 at 20:06
  • \$\begingroup\$ You "like their sound best"? What does that mean? \$\endgroup\$
    – Hearth
    Aug 18, 2022 at 1:12


Your Answer

By clicking “Post Your Answer”, you agree to our terms of service and acknowledge that you have read and understand our privacy policy and code of conduct.