Re: [TCML] Alternative RSG Motor and shielding

[Jim Lux <jimlux@xxxxxxxxxxxxx>]

On 8/29/13 12:37 AM, Yurtle Turtle wrote:
> I have done so, and managed to kill 2 VFDs. I finally found evidence that my electrodes were arcing to the motor base, so that's easy to fix.
>
> Like you, I have a question on the best way to not kill my last remaining VFD. I've read about protecting the three phases, but don't have the specs handy.
>
> I'm currently only running three wires to the motor, and letting the motor housing float.

Interesting questions..

So, what you want to do is "not allow spark currents to flow through the 
motor windings and VFD".

There's two possible sources of such current:
1) Primary, either directly conducted (e.g. spark gap) or induced (the 
current flowing in the primary loop around C, L, and sparkgap)
2) Secondary, either a direct strike or induced (magnetic field from 
loop consisting of topload, spark, ground return path)


Look at it systematically:
1) Direct contact: make sure you have an adequate insulating path or 
bushing between your rotary contacts and the motor. If your rotor disk 
is made of insulating composite (Most are.. G10, FR4, etc.) that might 
be good enough. Might is the operative word.  You could get creeping 
discharges along the surface of the disk at a surprisingly low voltage. 
 Ballpark, I'd assume you need at least 3" of surface insulating distance.
2) Induced from magnetic fields from primary.  The key here is to make 
sure that the wires between primary, tank cap, and spark gap are short, 
and things are close together.  And, that the motor and all electronics 
are not near that loop.  Shielding is hard to do effectively for 
magnetic fields.  The other thing is to keep all the motor 
wires(especially including the case ground) together, so they don't form 
a loop. That bundle of 4 wires should go to the box holding the VFD, and 
the 3 or 4 power wires going to the VFD should be bundled too.  You 
don't want anything where there is a loop of conductors that has 
appreciable area.

3) Secondary direct strikes:  Put some mesh between the motor and 
electronics that is connected to the bottom of the secondary (RF 
ground).  That means the spark will hit the mesh, not the motor or 
electronics.  Basically you're building a little faraday cage for your 
motor and electronics.  It does not have to completely cover it (it's 
not "shielding the motor"; it's more like providing a safe place for the 
strike to hit.

4) Induced transients from secondary.  These are really tricky. Fast HV 
pulses can destroy things fairly easily by causing internal sparks (ask 
anyone who has built a Marx bank and fried their power supply charging 
it).  The key here is to make sure that the loop area is minimized.

This is the kind of thing where "more grounding" is not necessarily 
better. Adding random ground wires can create a loop that can intercept 
the magnetic fields and create voltage differences between points on 
that loop.

Here's a way to get bitten by this.

Wall socket ground and power goes via one power cord to VFD, which then 
goes to motor case.
Wall socket ground and power goes to primary side power transformer, 
with a different cable and a different wall socket.
Primary side low voltage ground (e.g. NST case) is connected to RF 
ground (for safety, so that if you touch the RF ground, you won't get a 
shock).

So far, no problems..
Now, you add a wire from primary low voltage side ground (NST case, Pig 
tank) to the motor.

You've just created a potentially large loop: Wall ground-motor-new 
ground jumper-primary ground-wall ground-wire inside wall to ground in 
other socket.

And that large loop will have induced currents from the transients in 
the primary and secondary.

_______________________________________________
Tesla mailing list
Tesla@xxxxxxxxxx
http://www.pupman.com/mailman/listinfo/tesla