Amplifier for 0-100V?


for experiments with a gas discharge tube I need a kind of amplifier.

From a DC source I get about 100V, 50mA.
With a DAC, which I will connect to an Arduino, I want to "modulate" this DC.
The result will be fed into the tube then.
The modulation frequency is between 0Hz - 1000Hz (1kHz) max (I have not checked
yet, whether the software is able to reach 1kHz with teh DAC...but at least, this is the plan :slight_smile:
The resulting signal should follow the waveform of the input as close as possible,
but it does not need to be "true HiFi" or such. The frequency is in the audio range,
the usage is not.... :slight_smile:
The current through the tube should be current limited to avoid sputtering the electrodes in there.

The problem is: This is analogous technique...YYYYIIEEEEEEKKKKKS!

I read about beasts called N-channel MOSFETS and such, but this kind of items seem to be
used as simple on-off-switches only.

What I need is a simple as possible circuittry, which does this kind of "amplification"...but I have
littarally no idea how to do this.

Any help is very appreciated...thanks a lot in advance!

You can buy high voltage DACs, up to a couple of kV, or use a low voltage DAC and a high voltage op amp.

You will have to think about how many voltage steps and DAC bits are required.

Hi, jremington,

THAT sounds good! One Op-Amp will make the circuit rediculous simpler! GREAT!
Remains the problem with the current limitation...

Any idea?


Series resistor providing small signal voltage as a feedback into your arduino driver

Hi jackrea,

thanks a lot for your help!

Ich I understand that right:
The current through that resistor creates a controlling voltage.
That voltage goeas back to the arduino, which in turn in-/decreases
the absolute values of the values sent to the DAC which in turn
in-/decreases the amplitude of its output.

This is a long chain has some delay. Together with the non linear
behaviour of the conductance of a gas around the point of ionization
the whole feedback loop will tend to oscillate on its own.

Is there a method to do this without using a feedback loop…more in the way
a fuse would do this (unfortunately only a single time…it is an
exampe only)?

Thanks a lot for any help in advance!

There are standard ways to stabilize a feedback loop, the simplest is to provide an alternative
high frequency feedback through a low-value capacitor, which provides a delay-free path that
short-cuts the actual physical system.