DIY 30kV supply from a beginner, dangerous

[oz2cpu]

I "scored" this one a few weeks ago, the seller told me hos cool and fantastic it was
so i got it home, when i opened it, it was not exactly what i had in mind :-)
so i released a video about it all

https://youtu.be/SrBwraFqDv8

the comments are super funny, yes it is a bit dangerous, and must be redone quite alot to work
and be safe.

[Phil1977]

Do you really think it´s worth to rework this? It seems it has quite a sophisticated regulation, but beside that I think you can get everything else much easier and in a safer way by building it up from scratch.

30kV with 100uA easily are in the range of an off the shelf transformer with a voltage multiplier. The most complicated part of a DIY setup may be the voltage divider and regulator.

[coromonadalix]

i would seriously stay away from this pile of ....  just seeing the internals ...

[HighVoltage]

  I have seen many high voltage power supplies over the years but never one this bad.
As others have said, stay away and start from scratch or get a good quality supply.

[Ice-Tea]

Throw it (far) away.

If you want to review a well designed 30kV isolation tester, let me know. I can send you one on loan ;-)

[Haenk]

"Hey mor, I built a death trap!"

[Gyro]

I think you lot are being partially blinded by the look of the internals.

I certainly wouldn't call it a death trap. Much of the circuitry is low voltage and CRT LOPTs especially with separate Triplers (as opposed to diode spilt secondaries) are getting much harder to find. In addition, you have a reasonably enclosure, ready fitted meters (ok the volts one's calibration leaves something to be desired) and matching high voltage output connectors.

Yes, the construction is crap, but it is basically functional, withstands arcing of the output, and apparently works reasonably well as an EHT leakage detector so the basic circuit must be fairly sound. I've seen much worse and far more dangerous starting points for an EHT supply! I think it's worth at least putting a higher value HT rated resistor in series with the output*, partly to reduce the 'shock value' (remember the tripler has capacitance), but mainly to protect the tripler itself from overload, and characterizing it a bit more before considering whether it is worth a proper rework.

Don't forget to drill down to the mains primary side and make sure everything is well sleeved - an EHT flashover to mains can cause a nasty (hopefully constrained) high current flash events - this was occasionally a failure mode in old CRT colour TVs.


Edit: *On second look, I can't see from the diagram whether the output series resistors are 100k or 100M (in which case they should be ok).


P.S. Did you eliminate corona discharge at the end of the output cable as a source of the sound at the end of the video? Another culprit may be the un-potted resistor in the top of the tripler. My 10kV insulation tester produces plenty of noise from the end of the probe when the ground connection is that close and will trip on the bottom two current ranges without visible arcing. Get it to that same state, stand away from it and darken the room, the source should be obvious.

[inse]

The front panel says Ri=200M, so max 150uA output current.

[Gyro]

I missed that. It sounds pretty safe then as the resistors are down-stream of any capacitors in the tripler (coax cable capacitance could still sting a bit though).

It's worth remembering that the 100M resistors will dissipate 4.5W each under 30kV short circuit conditions and, according to the schematic, are only rated at 10kV each.

[inse]

Most probably the voltage will already drop after the tripler.
If Thomas has the right equipment, he could do some measurements.
I bought me a Philips HV probe for cheap, capable of 30kV, it is huge.

[Gyro]

Yes, I have a big old 30kV 3G Marconi one. There have been times when it has been worth its weight in Gold, for DC at least.