Stefan,

If arc-induced breakdown paths between turns is an issue, have you considered adding a homebrew corona inception detector with feedback to the modulator or switch driver? I haven’t made one for a VLF coil, but it might be worth a try if repairs are an issue; there is good precedent at near-VLF frequencies.

The idea would be to limit voltage or skip cycles before breakdown occurs; the voltage-limiting method could often be nearly transparent to real-time operation.

If repairs are rare then the above might not be worth it, just thought I’d mention the possibility.

73, Jim AA5BW  

Hi Bob,

Of course, if the coil is protected against rain, then it is no problem. But the coil is quite large and due to the high voltage, the housing should ideally have a distance of 30cm or more to the coil. So the housing must be quite large, even more. And that makes it dangerous because of wind blasts on the to of the building in 35m AGL...
Anyway it would be nice and i also thought about using a simple foil across the coil, without additional distance.
However on LF, there are partial discharges on the top of the coil which burns holes into the protecting foil. And the VLF winding is 2000m long. It is a lot of effort to repair it after a short cut between some turns due to arcing. So it is a high risk.
I think i will make a test in the near future and see if 9 kHz partial discharges are as agressive as 137 kHz partial discharges :-)
73, Stefan/DK7FC

Am 26.12.2013 16:32, schrieb Bob Raide: