X-ray power supply

[Jon Rosenstiel]

I've been using a GE 100kVp tube type, (Kenotron), x-ray power supply to power my fusor.

I'd like to be able to have a choice of using either positive, negative, or both outputs. (Other projects). To do this would it be ok to cut the filament power to the respective Kenotrons? (I could bring the filament xfmr primary leads out through the top cover Cannon plug).

If this is doable I’d only need one bleed resistor, and then the power supply would fit into a nice cabinet I have!

Jon Rosenstiel

[DaveC]

Don't forget the Xray shielding for your Kenotrons!! They usually have a fairly thick anode cup, which will help absorb low energy xRays, but there are always backscattered electrons which will generate xrays when they hit something solid. Even TV Hv rectiifers and dampers produce xrays.

Dave Cooper

[Richard Hull]

Watch out for where your ground-ground is!!

X-ray supplies are notorious for locking down a center tap or one end of the HV transformer to the case! If this is the situation you may not be able to just flip the supply around. Here is where drawing out a schematic would help.

Richard Hull

[Jon Rosenstiel]

I'm leaving the power supply in its original container, (under oil), so x-rays from the Kenotrons shouldn't be a problem.

Let me see if I can explain what I want to do a little more clearly:

My plan is to mount the complete x-ray power supply inside a 19" rack cabinet I acquired recently. The problem is that the cabinet isn't quite tall enough for two bleed resistors with a possible 100kV between them.

Jon Rosenstiel

[Jon Rosenstiel]

Richard,
I've brought each hv winding ground end out of the case through one of the Cannon plugs on top of the power supply. Depending on what polarity I need I connect up the milliammeter and ground accordingly.

Only thing is that now I have both positive and negative hv present at all times. (Need lotsa space for a possible 100kV)! I thought that if I could cut the filament power to the respective Kenotrons I could simplify the power supply by just having one bleed resistor inside my cabinet. Then, if I require a positive hv supply I can re-wire the milliammeter circuit, switch on and off the proper Kenotron filaments, and change the x-ray cable from the cathode socket to the anode socket.

So, my original question is still unanswered… Will I harm either the Kenotrons or the HV xfmrs by switching off the Kenotron filament power?

Jon Rosenstiel

[Richard Hull]

You will not harm the kenotrons. You will still rectify though through field emission though. At really high HV levels if the thing is in circuit it will try and suceed at tugg electrons out of dead fils.

A classic example is E-Beam services in MD. A technician on the linear accelerator killed the fils to work on the accelerator with the voltage at reduced level, (only 55kv). He later lost his arm after it putrified through morbidity due to extreme X-ray burns.

Richard Hull

[Richard Hull]

What I understand is that the rectifiers will still be in circuit with dead fils. They will still rectify. They will still contribute their full voltage rectification as if the filaments were on. They will just not supply much current at all as the electrons are not boiling off of a hot filament. They are still being supplied by the filament though through field emission. Smart money would remove them from the circuit with broad gapped replugable jumpers. This is the manner of polarity reversal used in the 5KV Ortec NIM bin supply and a lot of other HV supplies where reversal of polarity is to involve only one set of filters or bleeders.

Richard Hull

[Jon Rosenstiel]

Richard,
Thanks for your insights. It appears that if I stay with my original plan I'll have to put a bleed resistor from the "dead" side to ground also. I'll have to do some tests to see just how much field emission current the "dead filament " Kenotrons supply. Just off the top of my head I'd guess in the uA range.

Jon Rosenstiel

[Richard Hull]

Microamps sounds about right. I have seen 500ua supplies of about 50kvdc use a special field emission rectifier tube with tungsten fuzz for a cathode and the upper hemisphere of the tube coated with acquadag as a plate. When powered up, a careful rubbernecking for a good look up the tube's "skirt" showed the tips of the cathode fuzz incandescing due to field emission. Field emission is a viable, usable methodolgy. Some modern miniature low voltage x-ray tubes are now made with such special cathodes and, of course, save the weight, bulk and expense of a filament transformer.

Richard Hull

[Jon Rosenstiel]

I did a quick field emission test on my power supply today. (After realizing that all I had to do was to turn on the breaker and the hv switch while leaving the filament switch off)! Duh!

Anyway, with the variac full open, (235Vac), I was getting 3.4mA into about 200k. Quite a bit more than the 500uA I was guessing! With a just 90 meg bleeder in the circuit the supply would develop about 40kV.

(At the time of the test the temperature of the power supply was 35C)

Actually, I can produce neutrons with the field emission current! (20kV at 1mA).

So, I can see that my original idea of killing the filament power to disable one side of the power supply won't really fly after all. I'll do as Richard Hull suggests; remove the unused rectifiers.

Jon Rosenstiel

[DaveC]

This is where a reliable Xray monitor is invaluable. From the work i do with Ebeam tubes, a uA or two at 50 Kv will produce a pretty strong xray output. So if you are getting mA, thats a serious xray hazard...

Dave Cooper

[Jon Rosenstiel]

Dave,
Thanks for the heads up on the x-ray hazard. I checked my power supply earlier this evening.

With the Kenotron filaments off, input voltage at maximum, (235Vac), and 3mA field emission current I recorded the following: (Measured with a Victoreen 440RF survey meter)

8 inches from the power supply tank the Victoreen indicated, (are you ready for this?), 200mrem/hr! TALK ABOUT A SERIOUS HAZARD!!

8 feet away, (where I normally sit while operating my fusor), I measured 5mrem/hr!

As expected, with the Kenotron filaments on there were no x-rays emitted.

So, you are quite right, at mA levels the x-ray hazard is quite serious.

Again, thank you for bringing this matter to my attention.

Jon Rosenstiel

[DaveC]

Jon - good you checked.. that is a pretty serious Xray level. I use a "Monitor 4" geiger type xray meter. These have the quenched LND geiger tubes running at something around 900 to 1100 volts.

Since we make an ebeam tube that emits through a very thin window.. I can check geiger sensitivity to low voltage photons. In tests done in the last two months, I detect Xrays down to about 1500- 1600 keV electron output. (One of the things I particularly like about the the Monitor 4 unit, is its very low energy Xray sensitivity)...

With the end windowed geiger tube directly opposite the ebeam tube output...a few microamps of beam current will peg the meter (>1000 cpm) Xrays at low voltages will not penetrate the skin... making them "only" a skin cancer hazard!!

Several years ago, the company did some outdoor tests with an ebeam tube emitting at 90 keV. The xrays were detectable above the background rate (~16 cpm where we are in So. Calif) about 100 feet away. with the unit emitting vertically into the air.
Detectable levels were also measured on the far side of a concrete block wall, although these were perhaps only a factor of two or three above background.

So please... be careful with voltages above about 40 kV, as xray penetration depth goes up approximately as kV ^2. We want to leave no fusioneers behind!!

Remember, with the high currents being used here for the fusors, normal shielding thicknesses may not be adequate. Personally, I like to see little or no increase above background when I am running tests.

Dave Cooper

[DaveC]

Whoops!! The photon energies are NOT 1600 keV but 1600 eV. Sorry for the typo.

Dave Cooper