HV current mirror

Hello, I need to create a high voltage, extremely accurate current mirror. I tried to make this circuit with analog circuitry and opamps but I get oscillations that I can not manage to quell and I don’t feel too comfortable floating a 16v opamp up to near 200v to act as a high side current mirror either, at least not as a long term solution.

This is what I had in mind.

Basically the M2 circuit is doing its own thing and I need M1 to replicate the DC current conditions of M2 as accurately as possible.
How accurate? Well the opamp I was using is accurate down to 1uv. This is even more an issue now that I’m using the feedback device on the low-side instead of high-side like I was before so now I have the accuracy of R6 7 8 and 9 to deal with.

Actually on that topic, I was wondering if some sort of auto calibration could be done to eliminate the error of those resistors. Even a small misreading will generate significant DC error at M1, even without those voltage dropping resistors in the way.

How feasible is this and what would be required?

Thanks in advance, I really need this circuit to work.

Anyone? I'm at least looking for a basic outline of what I would need to get it to work.

No one?

You might be better off asking in an electronics engineering forum rather than an Arduino forum. Sorry.

I don’t know how much current is involved, but the enclosed would do it for small values.

The accuracy depends on the matching of R1 and R2 - 1k would be ok for a few mA

The second circuit is a conventional Wilson current mirror - to be accurate it needs matched transistors and close thermal coupling. It’s accuracy can be improved with resistors in each of the emitters connected to HT.


Isrc.pdf (20.6 KB)

Isrc2.pdf (14.7 KB)