I'm designing high-voltage enclosure circuitry that engages a lock when either of two HV supplies reach 50V or greater. Both are capable of 6kV output. Below is what I have so far, where R7 represents the lock:



The two similar blocks form comparators against a 5V reference, and Q3 and Q6 form an AND gate. Q9 drives the output.

I've had issues with op-amps and logic ICs burning up with previous designs, so all components are discrete.

This design works well in simulation so far, but I have only yet cranked up the sources to 60V. In reality, a 100:1 resistive divider (10M:100k) will be used to yield a 0-60V range, or thereabouts (200:1 for 0-30V, perhaps).

The issue is that, at 50V, the voltage at the comparator input is too low to overcome Q1's or Q4's \$V_{be}\$. I thought about adding a common-emitter amp with a gain of 10 at the inputs that just saturate a bit above this 50V mark, but that would also require another inverting amp following it.

Is there a simple way to go about this that I'm not thinking of? Or is a change in configuration needed?

  • 1
    \$\begingroup\$ I'd probably try to figure out why the op amps / logic ICs burned out before and then correct the issue, the circuit would be a lot simpler without having to homebrew it with transistors. Failing that, could you decrease the divider so the input voltage goes up, and then shunt the excess voltage via a zener+resistor to ground? Not posting as an answer because I'm not sure I understand the issue enough \$\endgroup\$ – Flyingfirepig Mar 29 '18 at 19:17
  • \$\begingroup\$ @Flyingfirepig The burning-out issue is due to the DUT's gnarly EMI/EMC tossing around the housekeeping supplies like packing peanuts. The discrete BJTs have proven to be much more robust, so that's the game plan. I thought about the Zener, too, but if I want a 5V signal from 50V on the supply I'd be dealing with up to 600V from 6kV. I figured I'm overlooking something, but that may be the way to go I suppose. \$\endgroup\$ – calcium3000 Mar 29 '18 at 19:25

You can make it work by providing a negative supply for R1/R2, something like -5V will do.

But there are probably easier ways, especially if you know why exactly you had part failures. If your failures are due to the divider flashing over, then discrete parts are not going to be all that much more robust.

I would suggest something more along these lines (concept only at this point), where U1 is an open-collector or drain output dual comparator with low input bias current:


simulate this circuit – Schematic created using CircuitLab

|improve this answer|||||

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service, privacy policy and cookie policy

Not the answer you're looking for? Browse other questions tagged or ask your own question.