When you build a hv device, you will probably want to know what voltage it gives. You can measure the maximum distance between two metal balls the voltage arcs over; there are tables relating distances to voltage, see Measuring HV by sparklength . This method is simple but inaccurate. A better one is a suitable resistor expanding the measuring range of an ordinary voltmeter.
The simplest way to build a hv resistor is to fill a plastic tube with water, see super-cascade. But for an accurate hv resistor, there's no way around soldering. For my version, I soldered 100 standard metal resistors (10MOhm +/- 1% each) together, wound the whole chain around some rod and pulled it off, so that the (still wound) chain fitted into an empty bike air-pump. I glued in terminals, sealed the whole thing and filled it with vegetable oil.
Each single resistor is rated for 500V max., so I can measure up to 50kV (I've even used it at 80kV). With a total resistance of 1GOhm, I have 50uA at full voltage, so the resistor plus a simple moving-coil meter makes a perfect HV-meter. For more accuracy, shunt a digital voltmeter module (-/+200mV) with 1kOhm and connect the HV-resistor in series.Hints: Sealing the tube oil-tight is easily done with epoxy glue. Glueing in a cable (as I did with one end) is not a good idea, however, as the oil creeps through the cable (like through a wick). I didn't know this and searched for the leak for days! Another mistake (probably) is using a black plastic tube, as this often is slightly conductive. I've never really tested for this, but the readings from my meter always seemed plausible, so it can't be too bad.
[Back to Index] [Other links]