List of Parts
(x) = have

12V Lead-Acid Battery
12V Mechanical Commutator (x)
5000V Power Supply

Plexi Case
2 Copper Rods (x)
2 Copper Mesh Tubes
Carbon Resistor
Silver Tips

5KV 12uF Capacitor
5KV Diode

--------------------------------

This is going to be my first big build. The largest challenge
I'm facing with this so far is finding a 5000V supply. There
are plans for a variable-output high voltage supply at Instructables,
but I am uncertain whether it would work for this purpose.

Some of the parts listed above I may forgo obtaining until later,
such as the silver tips and a carbon resistor, the latter of which
I would be unsure of where to find or what the specifications ought
to be.

I'll be spending my time on this, and wish to end up with a display
quality unit - an operational demonstration model that will be kept
running. Once I have it working, my attention will likely drift
onwards to using it for production of HHO.

I've begun a couple of other discussion threads entitled
'DIY Tesla Switch Guide' and 'PC Fan Generator' for those interested.

The PC Fan Generator has evolved into something a little less Half
Baked than it started out as - I am now using it as a driver for a
commutator switch, one that I will implement using in this Gray's
Tube project in place of the 'Rotor' as shown below.

As of yet I haven't obtained even half of the parts on my list,
but as I do so and continue to progress its construction, all will
be documented and shared.

Items I Need Help With so Far
5000V Power Supply
Shielding to Prevent Damage to Sensitive Electronics

---

Do you know what kind of current you need at 5Kv? If you don't require much current you could possibly use a flyback transformer. There are some very simple circuits out there for them.

Good luck. I will definitely be watching this. I've always been interested in Gray's tube.

- Quick Update -
I have now found a source for 5kv diodes, and the copper mesh as
well...

innovation_station,
You state that 3 'mots' will provide close to 5kv, but unfortunately
I'm not familiar with that term.

I've got access to an old microwave, as well as a small television
that would likely not be missed... so I'm thinking there ought to
be some nice high voltage items rattling around in them like flyback
transformers that I can put to practical use.

Still, I'm uncertain whether going over the 5kv mark would make
much difference - like if I've got to be right on the dot with it
or die from exposure to EMF or something.

The diode for the HV supply was one of the things I've been
in search of for a while now, so no problem if its already
built in.

Below is a winding I grabbed from a 15w fluorescent ballast,
and the other components - I'm guessing one of them is likely
a diode of some sort - the one with shiny wrapping?

I've got a CFL bulb somewhere around here that I could salvage
for its inner goodness and hook onto it... although what the
resulting volts would be has me guessing -

This is a microwave oven transformer I salvaged today, along
with the matching 2100V capacitor and diode.

I'm thinking of using the ballast from a low-wattage compact
fluorescent bulb as the input, which would likely create a
spectacular amount of voltage at a lethal current.

Most importantly what I'll have to figure out, is the model
of diode that I'll need for the 12V positive supply on the
short copper rod with such a large voltage.

In the diagram it is matched with the voltage being fed into
the other rod - 5KV... so if I end up producing 20,000V, I'm
guessing that I'll need at least a 20KV diode.

This of course would far exceed the amount of voltage as
shown in the drawing, so I've got no idea what the result
would be.

My alternative to the CFL Ballast - Transformer idea is to use
a Cockroft-Walton Voltage Multiplier, and while I can easily
see how they are built, the underlying method - matching
the capacitors, diodes, and input source has me confused.

What I would ideally like to accomplish is to use a 9-12V
lead acid battery hooked to a CW-Multiplier - preferrably the
same one used to charge the low-voltage rod.

Any ideas on what parts I would need?

The cheapest place to buy High Voltage power supplies is at Thrift Stores like the Salvation Army, Purple Heart, etc.  I have seen used Room HV Air Cleaners for around $10.

Tear out the HV Unit, Plug it in the Wall, & you will have a 12 KV Pulsating DC supply READY TO GO, almost free. 

.

These are the inner workings of a 30w 550mA CFL bulb.

On each side are two leads wound with a bit of copper
wire... I'm a bit uncertain which two I ought to use,
and what the resulting polarity would be - is there any
way to be certain?

While your power supply idea has plenty of merit, it would
require that I possibly break an operational unit.

If I can find out how many volts this CFL board puts out,
it will serve quite nicely as a starting point for the
CW Voltage Multiplier, although it would require that I
use 120V AC.

Ideally I'd like to charge the high voltage rod directly
from the 12V lead acid battery through the chain of diodes
and capacitors, so as to remove the necessity of having to
use alternate power sources... as shown below - in regards
to which I must ask the following -

When the charge passes through the capacitor(s) following
the electromagnetic coil, then back to the battery past
the motor driving the +12V, what is to prevent this from
overloading it as well as the 5KV supply - also hooked onto
the 12V battery?

It seems that this power supply is in fact 25KV, with
a high-frequency driver feeding a CW multiplier.  It is
unknown to me how this could have any effect on the
operation of the tube discharge in comparison to
using a regular low-frequency high voltage supply.

I'll be recieving a shipment of copper mesh soon, so
all that is needed besides would be a high-voltage diode
to attach on the 12V end. I've already got a 2100V
microwave capacitor, so perhaps I won't need to get
a different one until later...  I'm planning on connecting
the output of the Tube onto the capacitor, which is then
connected back to the +12V battery terminal - Oh right,
and I'll be needing to find a 12V battery as well.

I'm still a bit confused about something
as it turns out; as can be seen in the following
diagram, the return from the coil goes to charge
a 5Kv 12uF capacitor which discharges back against
the +12V from the battery to recharge it, but in the
way is the timing mechanism which is going to be in
the direct path of this...  ??  It is my best guess
that Mr. Bedini knew what he was doing when he drew
it this way, but it's still got me wondering.