My Q1 pulls the load up, and Q2 pulls it down. That's a little nonlinear, but the transfer function is continuous and it can be made to work fine.
-- John Larkin Highland Technology, Inc picosecond timing precision measurement jlarkin att highlandtechnology dott com http://www.highlandtechnology.com
I've had more luck with follower topology in that kind of situation, as in these fragments:
The diode (or zener in the mosfet cases) helps down-programming by linking the common base level shifter to the output.
But if your app is just to bias photo-detectors the current demand is likely to be very low and you might think about a shunt regulator which has the great advantage of being inherently short-to-ground proof?
piglet
The n-depletion pass fet doesn't need a big/slow pullup, and it can add a current limit:
Using another one on the low side could have startup advantages.
-- John Larkin Highland Technology, Inc lunatic fringe electronics
Thanks piglet, Well after trying the fix the 'not really broken' part of my circuit. I finally discovered I needed to by pass my the HV voltage divider resistor with a small cap. So it's working now.. though might be better. My step response time is ~100us at the moment that seems to be good enough but if not I've got half a dozen (thanks now to you) different configs to try. I ordered some HV pnp's for Phil H's floating push-pull (totem pole).
Speed-wise it seems all the single pass element circuits are going to have one slow side.. either turning on or off through the gate/ base resistor and whatever capacitance is there.
Re shunt. Does this always have some resistor it works against? In which case I want the thing to dial down to zero volts too, and when I did the numbers on the resistor.. well it was dissipating a lot of power at zero volts.
thanks again, George H.
Nope, doesn't have to be resistor. Current source would work well, could be as simple as a single depletion mode fet. Since you are contemplating using a current limiter in the series pass version I think the big picture dissipation should be pretty similar?
piglet
Running transistors hot causes them distress, resistors not so much :)
piglet
OK, If I get to that point I'll hit up SED for design ideas. (I must owe everyone here at least a case of beer by now.)
Hey I hooked up the spendy Spad today. It didn't blow up, I'm counting photons, the dark count is high, but that's expected.
I'm declaring victory for the day and going home before I do break something. :^)
George H.
ElectronDepot website is not affiliated with any of the manufacturers or service providers discussed here. All logos and trade names are the property of their respective owners.