Uzzors' Coolidge X-ray Machine / Projects / Forums

Forums

4hv.org :: Forums :: Projects

« Previous topic | Next topic »

Uzzors' Coolidge X-ray Machine

previous 1 2 3 4 next last

Move Thread

LAN_403

uzzors2k

Tue Feb 03 2009, 09:53PM

Registered Member #95 Joined: Thu Feb 09 2006, 04:57PM
Location: Norway
Posts: 1308

I received a response from the Ritter guy, and Harry your assumptions are almost dead-on. Apparently this tube type has been around since 1957, I wonder how old mine is. It certainly doesn't look very worn.

Proud Mary

Wed Feb 04 2009, 02:14AM

Registered Member #543 Joined: Tue Feb 20 2007, 04:26PM
Location: UK
Posts: 4992

Uzzors, I think you can be sure that the recommendation not to use filters to cut out soft X-rays would not be permitted nowadays. Trying putting the filters in and out on RadPro with 70KVP/10mA at a distance of 1 cm (inside your mouth!) and see how much difference it makes to the absorbed dose. Radiological safety criteria have changed very much in the last 50 years!

You should consider the possibility that your tube was removed from service because its emission began to fall below the level needed for dental X-ray service - which is not to suggest that it is dead - far from it.

I would suggest you try out this tube in a clear container of oil, to prevent flash-over along the glass surface, and so that you can see any glass fluorescence, and the presence of any glowing plasma which might suggest that the tube had gone "soft" (i.e. lost some vacuum). Start with microamperes only and only increase current very slowly. Watch the work in a mirror so you can see what you are doing, without the hazard of direct line-of-sight exposure. (though some of the rays are, of course, scattered, reflected and refracted from the background)

A GM tube is not suitable (ikke!) for measurments with this kind of apparatus, except for telling you that radiation is either present or is not present. You must make an ionization chamber - (the parallel plate type being much the easiest to make in the home workshop) to know how many Grays you have.

If you need help with the ionization chamber, just ask.

usaman65

Tue Feb 10 2009, 07:10PM

Registered Member #1364 Joined: Thu Feb 28 2008, 09:09PM
Location:
Posts: 55

Are you intending on using that? better have some good sheilding ans a geiger counter or an ion chamber survey meter because of such high REntogens.

kev

uzzors2k

Thu Feb 19 2009, 11:38AM

Registered Member #95 Joined: Thu Feb 09 2006, 04:57PM
Location: Norway
Posts: 1308

Hehe, picking up some Norwegian, Harry? I completed improved tin can ionization chamber as shown here. It works fairly well with background radiation reading ca. 20mV, and the radium watch-hands Leslie gave me read ca. 170mV when they're close to the chamber window. With the watch hands a few centimeters distance the reading drops to 70mV. This chamber is much more useful than the Geiger counter I constructed, and only cost about 1Â£. Do you think it would be worth the trouble to roughly calibrate it? I was thinking of acquiring a few samples of mineral with known intensity, and plot the voltage/radiation in a graph. Or is it even reasonable to expect the chamber to be linear enough?

Proud Mary

Thu Feb 19 2009, 01:15PM

Registered Member #543 Joined: Tue Feb 20 2007, 04:26PM
Location: UK
Posts: 4992

Erik, you can make some very basic linearity tests by placing your source on a measuring ruler, and taking a reading off your meter for every 5mm, or 10mm, that you move the source away. Then you can plot this on a graph, and see what sort of graph you have. Calculation will be easier if you have a non-directional source - an isotropic radiator.

You can use the same method for drawing graphs of directional sensitivity.

This basic apparatus with be affected by temperature, pressure, supply voltage, and perhaps capacitance to your hand, and other local objects.

But it is still much better than a GM tube.

You could calibrate it with a gamma source of known intensity, but you can also calibrate it roughly against your tube anode voltage, if you can measure and vary this.

The properties of metal filters (i.e. aluminium) can also be used as an aid to calibration (see RadPro)

And you can, of course, go back to the primary definition of One Roentgen, and start with the volume of your chamber, the voltage applied, and the current that flows, if you have the means to measure accurately very tiny currents.

uzzors2k

Fri Feb 27 2009, 11:59PM

Registered Member #95 Joined: Thu Feb 09 2006, 04:57PM
Location: Norway
Posts: 1308

I've made a basic x-ray machine, with a tube holder, filament current controller and shield. I'll add more details later, but I need to sleep now. My question is; how hot is the filament supposed to be? I've got an intensifier screen, some CRT dust and a glow-in-the-dark toy under the tube and none show signs of excitation. Currently I've run the filament from barely dull red to glowing yellow. Does it have to get bright white before it starts working? My multimeter goes nuts when I attempt to measure tube current which is unfortunate. I'll try again tomorrow.

Proud Mary

Sat Feb 28 2009, 12:37AM

Registered Member #543 Joined: Tue Feb 20 2007, 04:26PM
Location: UK
Posts: 4992

Erik, none of those methods will show much to the naked eye.

What about your ionization chamber? If you know for sure that chamber works from using it with other sources, then if it does not show any indication I would suggest no X-rays are being emitted.

Is your anode voltage holding up when your apply heater voltage? If you have no way of measuring it, I will send you 5 x 200M 2W HV resistors so you can make a HV potential divider, if you haven't got any yourself.

You must NEVER apply anode voltage to an X-ray tube - or any other power thermionic tube - before the heater is properly hot - 3 minutes to be very safe, or the tube may be seriously dsmaged.

The right colour for the heater would be bright yellow to yellow-white, but you must only turn it up if you can watch the anode current, or at least the voltage across the tube.

Are you sure your PSU can give out 7mA @ 70kV without falling flat as the heater comes on and the tube starts to draw current?

Set up some good tests with the ionization chamber - including a test with, and without, a shield - and make sure the HV ions around the machine can't directly interfere with the chamber. ( f.eks. put it under a kitchen sieve connected to Earth)

Then we'll see what happens next, Erik.

uzzors2k

Sat Feb 28 2009, 04:09PM

Registered Member #95 Joined: Thu Feb 09 2006, 04:57PM
Location: Norway
Posts: 1308

My ionization chamber became desensitized after the very first test. I had it directly under the tube thinking it was shielded from HV ions with all the tin-foil, which it wasn't. The reading is now -30mV with just background radiation and just 12mV with the radium watch-hands VS 30mV background and 300mV watch-hands before. I assume the pA-sensing Darlington has suffered, I'll replace it later. Either way, before the first tests I was unsure of it's practical use since it takes so long for the reading to stabilize. I made a small 130V inverter which boosts the chamber voltage and increases speed and sensitivity, but it still takes several seconds for the reading to stabilize. And if exposures are in the 1-5 second range that's a bit impractical.

The plan was to run about 500ÂµA to 2mA of current through the tube for reasonable x-ray intensity and low tube power. It's easier to manage long exposures than ones in the ms range, especially when anode voltage takes some time to climb. I have no idea how filament temperature and anode current correspond together, so I thought it would be safe to start with the filament barely warm. I've heard tubes can be damaged by low filament temperatures before, but never understood how or why. Could you explain? I have run the tube without the filament heated to check if my machine could withstand the HV, but otherwise I always let the filament warm up first. If it's actually warm enough is another matter.

I have no way of measuring the anode voltage unfortunately. I have tested the multiplier and transformer before, and they should be capable of 3kW. I doubt supplying 100W or even 500W is a problem. See the video I made of the multiplier arcing (through a water resistor).

I do have 1000x 1M mini-resistors which I bought for this purpose, but not the will power to solder them together. I would greatly appreciate it if you could send the resistors. Soldering 5 beats 1000! How much do you want for them?

I just now did some more testing today. I increased the filament voltage further, up to 1.6V and it glows white now. To see if it's working at all I placed some photographic paper in the intensifier screen, and x-rayed it for 10-20s. The tube doesn't get warm

and the paper was unexposed after development. I also tried measuring the anode current again, but the multiplier ruins my readings. Same goes for the ionization chamber, even at some meters range. I did a crude voltage test, with a salt-water resistor placed at the anode. While the set-up was running as normal I drew an arc from the resistor. The arc initiated at 4 cm, suggesting only 40kV. I don't find this surprising as the transformer is actually wound for 10kV, but ringing and other effects seem to give twice the output voltage (at least when the arc first initiates). I'm guessing the multiplier is loading down the transformer enough for it to behave ideally again, which only gives me 40kV. Still, I would expect the anode to become warm and some fluorescence on the screens at the very least.

Proud Mary

Sat Feb 28 2009, 05:33PM

Registered Member #543 Joined: Tue Feb 20 2007, 04:26PM
Location: UK
Posts: 4992

The 5 x 200M 2W resistors are a gift. I have many happy memories of Norway. If you'd like to send me your address in a private message, I'll put the resistors in the post on Monday morning, together with anything else that might be useful.

Why is low heater voltage damaging? It can sometimes over a period of time cause a condition called "cathode stripping." - stripping of the barium oxide cathode coating. But low heater voltage is not at all likely to result in immediate catastrophe, as can excessive heater voltage. The biggest stress on heaters is the surge that flows at the moment of switching on, as the heating element expands, bends and flexes as it expands.

So you should always increase heater voltage only gradually, Erik.

An X-ray tube is no more than a special case of a diode valve, and there are two basic failure mechanisms:

1. Heater failure. You know this has not happened because your heater lights up.

2. Vacuum loss. If your tube is unable to maintain high-voltage stand-off, but starts arcing inside when the anode is at 20 - 30kV then you have lost some vacuum. (Arcing at full voltage (70kV in your case) is not necessarily a sign of failure, but may be the operating conditions are not set properly.)

If there is any sign of barium oxide - a white powdery material - around the heater pins or the anode structure, then your tube is definitely gone. (The BaO is produced by the conversion of all the getter into its oxide)

You need to put a moving coil meter on the cold, Earth side of the tube so you can see what is going on, Erik. I have a box of 50-0-50uA somewhere and will send you one, but you'll have to make your own shunt.

My guess is that your tube is not gone, but that you don't have enough HT, but we shall see!

Don't wreck it before you get the measuring kit from me!

uzzors2k

Fri Mar 13 2009, 07:41PM

Registered Member #95 Joined: Thu Feb 09 2006, 04:57PM
Location: Norway
Posts: 1308

With the marvelous contribution from Harry I now have some more insight into what my x-ray machine is up to. I was given a moving coil ammeter, a bunch of high resistance and precision resistors for the voltage divider, and some extra goodies! I must say the meter looks cool on the x-ray machine! Thanks a bunch, Harry!

1236973054 95 FT62720 Highimpedance Divider

Today I took some measurements.

1) There is no current flowing through the tube, even with the filament heated. If there is any it's below 1~10ÂµA.
2) The output voltage from the CW multiplier is 48 - 50kV. That should be enough for testing, as the D model is rated for 40-70kV. I'll see if I can't add another stage to the CW tower, which should boost it up to 60kV.

Currently I run the filament at 1.5V, at the most. Given a total lack of current flowing through the tube, Harry expecting the filament to require 5V and various x-ray datasheets quoting filament voltages in the same region I'm going to buff up the filament driver. I should be ready to test with increased filament current tomorrow.

previous 1 2 3 4 next last

Moderator(s): Chris Russell, Noelle, Alex, Tesladownunder, Dave Marshall, Dave Billington, Bjørn, Steve Conner, Wolfram, Kizmo, Mads Barnkob