# How to measure voltage of each bank in a Marx Generator?

If I have a 10 stage Marx generator with each stage at 2.5kV, when I erect all the stages the last stage's negative terminal will be at 22.5kV with respect to ground.

If I want a voltage measurement of these capacitors what is the best way to do this? I am worried about violating the standoff internal to my differential probes since the probes would all share a common ground on the supply/measurement side.

• What value Line input and C value for each stage? – Sunnyskyguy EE75 Mar 5 at 22:29

VR37000003005JR500 30MΩ 3.5kV working voltage

IEC 60950-1-07 PCB Pollution Level 1 creepage 4mm/kV > 1kV

Use ~ 1cm air gap around the body to nearest part or above cellulose or dry wood substrate.

• Thank you I will give this a shot – MaxwellEE Mar 6 at 18:25

Make e.g. a voltage divider with 30 resistors that rated for 1kV each. When needed, use a capacitor parallel to each resistor to compensate for parasitic capacitance of the resistors. Benefit of using multiple resistors is that a high resistance can easier (lower cost/availability) be achieved and power dissipation is distributed over more resistors.

For measuring these kinds of high voltages, you might consider using a shielded copper sulphate ($$\\text{CuSO}_4\$$) solution in a cylinder of perspex as one of the resistors in a resistive voltage divider. You can decide on the divider ratio to use used.

Typically, a hollow perspex cylinder is used with a dilute solution of $$\\text{CuSO}_4\$$ using de-ionized water. For example, in this case you might use a cylinder that is $$\30\:\text{cm}\$$ long and $$\1\frac14\:\text{cm}\$$ in diameter. (Increasing the cross-sectional area improves the non-inductive behavior.) The cylinder design also suppresses the effect of coupling capacitance with the surroundings. Advantages for using a solution of $$\\text{CuSO}_4\$$ is that it is easy to adjust to the resistance you want for the divider (given some other value you've also chosen for the non-inductive carbon film resistors used in parallel, perhaps, to make up the other resistor of the divider), it can withstand many megavolts across it, and you can make it in a variety of convenient shapes.

Enclose the non-inductive carbon film resistors (use a number of them in parallel for the value you want -- do not use just one by itself) in an aluminum cylinder that extends partway up the outside of the perspex cylinder to provide shielding from EMI for the non-inductive, parallel carbon film resistor group.

Look these up on the web. They are built and used very often for Marx generator voltage measurements in laboratories.