Since originally building my 10kV power supply, I’ve made a few changes and
fixed a few problems along the way. Most obviously I’ve given it a coat of paint, added a ground connector and replaced the lock switch.
The paint is just standard black gloss, I’ve added high voltage warning stickers next to both terminals to make absolutely clear that both can be at lethal voltages depending on configuration. I’ve replaced the lock switch because the old one started arcing internally when switched off; I found it in my bits box without any ratings stamped on it, it seemingly was only intended for 12/24Vdc rather than 240Vac. The replacement is smaller so I had to patch it into the hole with a piece of aluminium can; I’ll sort this properly at a later date. I’ve also added a ground terminal to the front so that it is easy to raise things to ±10kV; I needed to do this for some ion motor experiments, but this will be useful for a lots of other situations.
As you can see, I’ve made a variety of changes inside as well
Notice first that there are 2 separate ground wires leading to the front panel; the ground terminal is not connected to the grill/switch until the point at which they are both connected to the mains supply ground. This is so that, should the ground wires become damaged or detached, it is much less like that the controls will end up connected to lethal voltages.
The lock switch on the front panel is no longer directly switching the power to the transformer. 240V 6A switched are physically large, difficult to find and expensive; I have worked around this by using a 1A switch to control a 17A relay. I didn’t have a 240V/240V relay to hand so used a 12V one and the 12V transformer from a microwave. I have also added a 6A circuit breaker to the MOT supply line to cut the power if the output is shorted. Note that this is to protect the circuitry only; it will not make you any less dead if you accidentally include yourself in the circuit.
I’ve slightly altered the high voltage side of the circuitry too. I first realised that there was a problem when the supply was only outputting around 6kV; this was after repeated sparkovers on the output(I was experimenting with etching Lichtenberg figures into wood). I opened the case up and found that the capacitors on the negative side were bulging and leaking electrolyte.
It turns out that with the output of the original circuit shorted, these capacitors have voltage applied across them the wrong way; this blows electrolytic capacitors. I replaced the whole capacitor bank even though the ones in the middle were not showing any visible damage, they may still have been damaged internally. To stop this happening again, I added a resistor to the positive side of the output stage; this will prevent a reverse bias voltage in the case of short circuit.
Here is an up to date diagram of the whole circuit. Please don’t build one yourself unless you fully understand the risks; making a small mistake with high voltages is a very quick, roast pork scented, route to your grave.
I have intentionally omitted the fan and indicator lamp to improve legibility of the diagram; they are non-essential, although the fan is certainly a good idea.