high voltage regulator problems

[metebalci]

Hi,

I have a sulzer like voltage regulator. Since I am using it in high voltages, I need to elevate opamp supply rails and the ground potential of active voltage reference. I am sampling the voltage at output directly after series-pass device. I have a voltage divider connected from output to ground. I have connected the middle point of voltage divider (that is 5 volts below the output) to the opamp negative supply rail and also to the ground of active reference device. Opamp positive supply rail and the active reference supply is elevated about 10V on top of the negative supply rail. Is there any side-effects of such a connection ?

Having measured the circuit output, it is continously changing between [Vexpected-3, Vexpected+3]. Is it normal ? If not, what could be the reason ?

Is there any way to measure the output resistance of this circuit for different frequencies by a DMM ?

Is there any reason to support more current than total quiscent current required ?

Thanks in advance.

MB

[jackinnj]

Originally posted by metebalci
Hi,

I have a sulzer like voltage regulator. Since I am using it in high voltages, I need to elevate opamp supply rails and the ground potential of active voltage reference. I am sampling the voltage at output directly after series-pass device. I have a voltage divider connected from output to ground. I have connected the middle point of voltage divider (that is 5 volts below the output) to the opamp negative supply rail and also to the ground of active reference device. Opamp positive supply rail and the active reference supply is elevated about 10V on top of the negative supply rail. Is there any side-effects of such a connection ?

Having measured the circuit output, it is continously changing between [Vexpected-3, Vexpected+3]. Is it normal ? If not, what could be the reason ?

Is there any way to measure the output resistance of this circuit for different frequencies by a DMM ?

Is there any reason to support more current than total quiscent current required ?

Thanks in advance.

MB

may help to describe the voltage you are trying to derive from the rails and the permissible excursion.

[metebalci]

Output is 380V.
Negative rail is 375V, positive rail is 385V.

[metebalci]

Circuit is like....

[jackinnj]

i have used the following as a HV regulator -- its from Horowitz and Hill "The Art of Electronics" -- I use a DAC to derive the voltage reference -- shown here is a fixed reference. you will have to choose your own compensation RC values for best results, as well the voltage divider for the error amplifier.

[metebalci]

Hi jackinnj,

How do you get supply rails of error amplifier (opamp) and fixed reference here ?

Regards,

MB

[jackinnj]

In the design in which I modified an Heath HV Power (Microprocessor Controlled Adjustable Power Supply) supply there was a spare center-tapped filament winding. In another supply I used a wall wart, doubled, rectified, filtered and regulated -- in the full wave doubler you pull the ground from the junction of the two capacitors. if you are careful in the layout it will work just fine. You can do the same thing with a non-center tapped filament winding.