User-agent: Mozilla/5.0 (Windows NT 6.1; WOW64; rv:38.0) Gecko/20100101 Thunderbird/38.2.0
Matt,
I still don't think you need any resistors at all in the circuit.
The NST is capable of putting out 30 mA onto a short circuit, but it
would be putting out only a few hundred volts at that current draw.
NSTs have high leakage inductance built in, because they are purpose
built for driving neon tubes. They can put out 30 mA at zero voltage,
or 15 kV at zero current, but never both at the same time. The
current/voltage curve is a strong inverse hyperbolic relationship.
A neon tube is an extremely non-linear load. It takes high voltage to
break down the gas inside, at which time the tube has nearly infinite
impedance. Once the gas breaks down, the tube impedance drops to a few
hundred ohms. The NST then provides a few hundred volts at 30 mA to
keep the tube lit.
Even direct contact with the output of a NST is gonna sting bad but
would be unlikely to be lethal except in the most unusual
circumstances. Not that I am advocating taking hits from a NST -- it's
just that NSTs are reasonably safe unless you do something deliberately
stupid.
I think it would be extremely unlikely to get lethal currents out of
your VDG even in the event of a runaway corona system event, unless you
have a "Boston Museum of Science" sized top electrode. The max stored
charge of your VDG is limited by the dimensions of the top terminal, the
quality of the insulating column, and the dryness of the air.
The corona spray system allows the collector to charge faster to its
maximum value than it would otherwise, but it can't accumulate any
greater amount of stored charge or a higher voltage that it would
without the corona spray system.
As I envision it, your system already has a high resistance built into
it -- the air gap between the ends of the points of your corona
electrode and the surface of the belt. Unless the applied voltage is
too high, or the electrodes are too close to the (presumably grounded
metal) lower roller, the air gap will provide all the resistance you
need. Just adjust the spacing of the corona points, and the voltage of
the NST with a Variac for optimum operation.
Ordinary paraffin from a home canning supply place is a good potting
compound for your diodes, and it has the advantage that it can be melted
out if you need to repair a part. Some RTV silicone rubbers release
corrosive acetic acid during curing which can damage components, and
epoxy is overkill.
BTW, if you haven't already built your HV rectifiers stack, Goldmine
Electronics was offering ready made integrated 30 KV rectifiers at an
attractive price. They should have more than enough capacity to handle
even a 30 mA short circuit current from your NST.
No that is exactly correct. Another user mentioned that I should be careful
using NST for VDG seemingly because there could be short down in the inside
of the VDG to the HV brush, allowing potentially lethal currents up to the
collector. I've not been able to confirm this exact reason - maybe someone
else can chime in on this.
I was mostly only concerned about the maximum wattage I'd need for
resistors, which I realized I can actually easily calculate by assuming a
short to ground. This gives me the maximum possible current flow through
the resistor network and thus I can calculate maximum wattage.
With this in mind I've discovered my 500mw resistors are not quite beefy
enough, and I'll need to setup some parallel/series chains to get higher
wattage. The NST puts out 30ma and I want to get that down to 10s or
hundreds of microamps for safety.
I did consider using the DMM in the ground path and that seems the safest
option to me. Of course I'll need to float the case away from ground during
this process but it should only be temporary.
Thanks for the hints I'm working on potting my Diodes and resistors now,
and am using generic silicone - I'm wondering if that's going to be
workable inside 1inch electrical PVC conduit or should I switch to HV epoxy
(can't find anywhere).