[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]

Re: surface breakdown was Re: 20 joules at 100 bps vs 4 joules at 500 bps



Original poster: Jim Lux <jimlux@xxxxxxxxxxxxx>

At 12:18 PM 8/4/2005, you wrote:
Original poster: Greg Leyh <lod@xxxxxxxxxxx>



Original poster: Jim Lux <jimlux@xxxxxxxxxxxxx>

At 07:25 AM 8/3/2005, you wrote:

Original poster: Greg Leyh <lod@xxxxxxxxxxx>

You're correct -- a toroidal electrode would have provided far better field control than the spherical electrode used on Electrum.
The spherical electrode geometry was part of the overall aesthetic requirement however, and had to be accommodated. As a result, the operating instructions for the machine specify to immediately remove power for 5 seconds if a tower strike is observed.


Based on experience, 5kV/inch seems to be a reasonable surface gradient to maintain on the outside of a secondary form.


Is that for creeping discharges on the surface, or a "free air" distance from toroid to base? Seems a bit low for the latter.


That's for creeping discharges along insulating surfaces. In my experience free-air field gradients between smooth, blunt features can be considerably higher; up to 20kV/in for dry, well ventilated areas.

The next secondary coil that I'm working on will have convolutes on the outer casing, like this: http://www.lod.org/Status/90L10KPhotos.htm The intention here is to support higher surface fields than a standard external winding, without requiring significant amounts of transport care or maintenence.


What sort of manufacturing process would you use for this? Cast the corrugations after winding? I assume you're not actually winding on the inner surface of the corrugations.

What about making a regular tubular secondary, then bonding disks with a hole in the middle along the tube. You'd want a real good bond with no pinholes. The advantage might be that you could use different materials for the two components.

Or, on a very larger scale, what about building the secondary as a stack of modules that get bolted together(like they do for HV string insulators or for high power EHV converters).


Your idea with the glued baffles would most likely work well. However for this design the goal was to completely protect the windings from the environment, and to increase the outer surface flashover length to the point where dust, moisture and other normal contaminants will not cause breakdown or tracking. The ribbed structure also protects the windings from physical trauma, and rough handling.
One version of the ALF coil towers use a modular design; each tower consists of five 'bobbins', wound in place at ground level. The windings do not have baffles, but depend solely on the air gap between turns [about 1" clear.] The end of each bobbin holds a support arm for the equipotential/strike rings.


The two identical 1:12 prototype coils (one shown in the photo) were made by first winding the secondary wire onto cardboard tubes as we did for Electrum [ in fact we used the same winding jig, after cleaning off a bit of rust.] The cardboard tubes w/windings were then inserted into the corrugated pipe, and potted into place. Once the potting set we removed the cardboard tubes, leaving the windings exposed to the inside of the tube. Lsec and Fo of the windings did not measurably change after potting.


So the corrugated tube is the "outer wall" of the potting mold.