Electrostatic discharge from a choke?

[Ed/KB1HYS]

I have a  7 Henry choke with about 3500V on it.  I can hear a popping noise in a nearby reciever when the supply is powered up.  I could not find anything arcing after a long inspection of the supply.  I did find that the choke frame will arc slightly to gound. When grounded the poppilng noise goes away.  The whole supply is built on a sheet of plywood and therefor the frames of the chokes and trannys are isolated and float.

 I would have thought that internal arcing would result in a rapid failure of a direct short to ground.   This just seems to be a static buildup that discharges periodically.  I don't have a megger to check leakage or anything but I've left this supply powered up for over an hour and nothing really changes.

[WD5JKO]

Ed,

    If the intention is to float the choke frame, then do so with something better than plywood. At 3500v, I'm sure plywood will be somewhat leaky. Is the choke insulation rated for 3500v? If so, just ground it. An alternative of course is to ground the choke case through a 1 meg-ohm resistor and see what happens.

    If the choke is part of a choke input filter with solid state rectifiers, then you might be seeing the result of the periodic back emf spikes; could be from the choke or power tranny. SS diodes have a reverse recovery time that for a brief instant (~ 1-100us??) can short out the power tranny at every zero crossing.

Jim
WD5JKO

[WA1GFZ]

C between windings and core charge core to dc output voltage due to ripple pulses until it flashes over. Better float, ground it or put in in the return leg.

[Ed/KB1HYS]

MV rectifiers in a bridge config.

I guess I will put it in the return leg. That will be the easiest and safest fix. Thanks guys.

[The Slab Bacon]

Ed,
     I have lost a few plate transfomas by putting the choke in the negative lead. The AC ripple voltage will now appear on the center tap of the winding, causing a possible insulation breakdown. (including a 150lb man-killer)

I would HI-POT test the choke and see if it stands up to the voltage you're using. If it does, ground the case and put it in the positive lead. If not mount it on porcelain stand-offs and be very wary of it or enclose it in a protective shield to protect yourself from it.


                                        The Slab Bacon

[KM1H]

If its a simple corona discharge you may be able to eliminate with a transformer varnish bath. Preferably in a vacuum pot but if you bake it in the oven it will do almost as good. Id still put it on a 1/2" to 3/4"  insulator of Delrin or other hard plastic using nylon hardware and topped with a DANGER HV sign.

Carl
KM1H

[WA1GFZ]

Carl,
I think the ripple on the supply is pumping the core voltage by the C between the windings and the core. It is a very high reactance but in time I think a charge is set up on the core.

[Ed/KB1HYS]

I've brought the choke up to 4500V. That's as High as I can reasonably go DC. There was no current over the bleeder current, the poping noise did not increase with voltage, it wasn't audible in the receiver when the voltage was <1500 or so (not sure exactly).  I grounded the case, and kept the voltage up. No change in current on the supply. The popping stopped in the receiver. 

If the choke has an internal arc, wouldn't the high voltage on the coil and grounded frame have caused a catastrophic arc failure??

[The Slab Bacon]

YEPPIR ! ! ! ! ! !