Adding Bias Voltage to HV Generator - HV Circuit Solutions

[ArchieDave]

Hello all. I have a HV generator that puts out ~200 nsec, 10 kV pulses. I'd like to add a bias voltage to this but I'm not sure how this is normally done. If I add a parallel DC source I could use a diode to protect the DC source from the kV pulse, but I can't seem to find kV rated diodes that can pass several amps (the current I need to pull through the bias leg). So should I add an RL leg such that the time constant is much greater than the pulse width? I would like to eventually gate the bias circuit as well.

 

[Baluncore]

No diode is necessary.
To bias the circuit you can use a DC supply with a parallel ceramic capacitor sufficient to attenuate the spike energy.
Connect the DC supply, through a high voltage RF choke or a resistor chain, to the load and spike output.

 

[berkeman]

Hello all. I have a HV generator that puts out ~200 nsec, 10 kV pulses. I'd like to add a bias voltage to this but I'm not sure how this is normally done. If I add a parallel DC source I could use a diode to protect the DC source from the kV pulse, but I can't seem to find kV rated diodes that can pass several amps (the current I need to pull through the bias leg). So should I add an RL leg such that the time constant is much greater than the pulse width? I would like to eventually gate the bias circuit as well.

What the heck is that generator used for?

 

[ArchieDave]

Baluncore - Thanks. I don't want to suppress the spike from the load so I can't just add a snubber, so you're saying I need to add a choke between the load and the added parallel capacitor. To my knowledge, a choke is a type of inductor. If so, the RL time constant needs to be greater than the duration of the spike. Is this correct?

berkeman - I can control the rise time of the HV output and I'm wanting to study breakdown in a gap as a function of the rise time. So I won't to keep the spike and add a bias across the load, which is a low pressure air gap that will break down and behavior as a fairly stable resistance.

 

[sophiecentaur]

Could you make up a DC bias circuit, with its own (special) supply transformer, wound with loads of isolation. If you literally need a Bias with very few mA, this could be not too hard to make with a ferrite ring. Same idea as a TV LOPT.

How many volts do you need and for how long? Did you consider doing it with batteries?

Some people really want to play with lethal stuff, don't they?

 

[Baluncore]

Here is attachment.

 

[sophiecentaur]

Here is attachment.

The series system would be good (I actually sketched exactly that in my mind as I read the OP) as long as you can stand the HV generator up like that. Which is why I wanted to know what actual bias volts were needed. It would not be unsafe to operate the HV source at a few volts above ground but 200V could be a problem.

Just a thought - would it be possible to DC bias the Other End of the equipment? That might simplify matters.

 

[Baluncore]

Just a thought - would it be possible to DC bias the Other End of the equipment? That might simplify matters.

Yes, that is the series arrangement but with the order of the load and pulse generator reversed.
The DC bias generator still needs a ceramic capacitor capable of handling the pulse.
Maybe a fast “gas discharge surge arrestor” could take some of the pulse off the ceramic cap.
http://en.wikipedia.org/wiki/Surge_protector#Response_time
http://en.wikipedia.org/wiki/Surge_protector#Gas_discharge_tube_.28GDT.29

 

[sophiecentaur]

I meant that the other end of the final load could be biased, rather than be grounded. ( but it would depend in the actual details whether tat approach could be practicable.

 

[Baluncore]

Sophiecentaur. If you draw your schematic arrangement you will see it to be the same series arrangement, but with a different order of connection. The DC supply has the same requirements either way. It comes down to generator and load isolation requirements.

 

[ArchieDave]

Baluncore. Thanks for the clarification. I'll see if I can work out the specs for the components I need and check availability. The series option isn't something I would have thought of.

Thanks to all for discussion.

 

[sophiecentaur]

Sophiecentaur. If you draw your schematic arrangement you will see it to be the same series arrangement, but with a different order of connection. The DC supply has the same requirements either way. It comes down to generator and load isolation requirements.

The DC supply could be earthed at either end of the chain, true. But it may be more convenient to have the 'target' end at an elevated voltage than the pulse generator. There could be a significant difference so the option is worth submitting as a possibility.