Need a BIG 10kv power supply

[macona]

A bit off topic but I am hoping someone might have some ideas.

I have an electron beam gun I am going to install in one of my vacuum chambers, it will handle 12kw max, thats 1.2 amps at 10kvdc. I really dont need that much, 300 to 600ma (3 to 6kw) would be fine.

Anyone have any ideas? I am toying with the idea if building a SMPS with some of the giant IGBTs I have and rewind a x-ray transformer to step up. Voltage is kind of critical, the position of the beam in the hearth is directly related to the accelerating voltage of the electron gun. So fixed at 10kv and settable current. Ideally both would be adjustable.

[NobleGas]

That's the sort of power supply that's used in big microwave transmitters, either klystron based or TWT (although TWT have more complex supply arrangements). If you can locate a surplus unit, you'll be looking for something upwards of 3kW output, which will have a power supply capable of 10 -12kW.

[absolom7691]

6 microwave oven transformers, wired in series (center tapped, three on each side of ground) will get you about 12.5KV at around 1A. You could control the voltage with a variac stack and probably build a cheap inductor to limit the current. I use an MOT stack for my tesla coil, minus the variac. My inductor is two 500' spools of 12awg electrical wire, right off the shelf from Home Depot. It limits it to about 500ma.

[macona]

I have been keeping an eye out for a large plate x-former but no luck yet.

There are issues putting MOTs in series, namely the insulation between the core and secondary was never intended for that high of voltage. Quite a bit of risk.

-Jerry

[JasonB]

There are issues putting MOTs in series, namely the insulation between the core and secondary was never intended for that high of voltage. Quite a bit of risk.
-Jerry

....If absolom7691 meant to wire the high side of each transformer in series to develop ~12.5KV how would the voltage potential each individual high side winding sees exceed it's rated ~2KV? In other words, the sum of the voltages in series makes a large voltage potential of 12K, but each individual winding only 'sees' the voltage that it outputs.

Unless I am misunderstanding ? In theory what absolom7691 stated makes sense.

If we are lucky professor Planters may chime in

-Jason

[macona]

Its because the primaries are all in parallel which means there is a large potential difference between one end of the stack to the other and it can jump from the secondary of one end, through the core and into the primary. It's just condsidered a bad idea. Possibly I could take the transformer apart and add some kapton insulation to beef it up a bit.

[planters]

There are issues putting MOTs in series, namely the insulation between the core and secondary was never intended for that high of voltage. Quite a bit of risk.

Actually, this is what I have used very successfully in my dye laser power supply. A 10 KVA variac (crude but very simple) supplies a 2:1 step down/up transformer (so 240 to 480V) this is sent to 5 MOTs wired in series and isolated from each other to well over the 2,400 V per stage. These feed a resonant stack of microwave capacitors. 10KV AC is then used to charge each 1/2 of the main capacitor bank in a voltage doubler arrangement for a total of 20KV at discharge. 6KVA is easy, the forced air cooled MOTS can run all day and this big linear set up is very stable.

[absolom7691]

I have never had an arc-over with my MOT stack. With 3 xformers on each side of ground, the potential isn't too high. I also have mine submerged in oil, as well. It looks like a poor man's pig.

Speaking of pigs, could try to locate a pole pig on ebay, they are heavy but they'll run forever.

[macona]

Well, maybe it is worth a try. Planter's method seems the safest.

Now I need to find three or four MOTs. I have one. the 240-480 xformer is easy.

Thought about a pole pig but would really like to keep this thing contained in a 19" rack.

[planters]

A very nice thing about MOTS is that they are CHEAP, really cheap. I ordered 12 at a time for about $150. One was DOA, but who cares. You could probably put 6 nicely in a 19" rack mount. The more MOTs the better for heat dissipation and smaller incremental voltage steps.

I have a couple of power supplies based on this same set up; variac to step up, to series MOTs, to resonant caps through a rectifier to storage caps. Today, I was experimenting with some flash lamps and set the charging voltage to 4,120V metered by a Fluke VOM through a 1:1000 Caddock voltage divider. The charge could not be set any finer than (read) 4.120 +/- 0.002V, but once set it remained nailed to less than 0.001V on the meter (1V on the cap).

I like thick PVC plates as insulators. I built the entire enclosure for my power supplies from heavy plates of PVC. Nylon 1/4-20 cap screws hold everything in place. I use comp. case fans to exhaust air from the enclosure and place holes wherever I want a stream of incoming air to cool a component. This keeps high voltage away from metal (like a fan guard) and allows for a compact enclosure.