Grenadier's big thread of RÃ¶ntgen related shenanigans / Electromagnetic Radiation / Forums

Forums

4hv.org :: Forums :: Electromagnetic Radiation

« Previous topic | Next topic »

Grenadier's big thread of RÃ¶ntgen related shenanigans

first previous 17 18 19 20 next last

Move Thread

LAN_403

Adam Munich

Thu Apr 07 2011, 03:43AM

Registered Member #2893 Joined: Tue Jun 01 2010, 09:25PM
Location: Cali-forn. i. a.
Posts: 2242

I got my meters in the mail the other day a 50uA meter for measuring anode voltage and a 3mA meter for measuring anode current. I have no pics yet, though I will tomorrow.

In series with the 50uA meter is a 1.8G resistor, and that turns it into a 90kV meter. Below 40kV there is no problems with measurement, but anything above and all hell breaks loose inside the meter. The needle starts swinging back and forth at a million miles an hour presumably due to electrostatics. When the CW is allowed to spark this problem is greatly reduced, yet not entirely eliminated. Nonetheless the meter seems to check out ok when it comes to measurement, so that's a good sign.

Although the meter works when the CW is allowed to spark, I am wondering about what will happen when there is an x-ray tube in the way. An x-ray tube is not really a short circuit like a spark is, so I am concerned that the electrostatics will once again reap havoc on the meter. The question is, will the tube appear as a high resistance load on the CW? If so then electrostatics will be a huge issue. If ohm's law fits the situation then it should appear as a 25 meg resistor if 3mA gets through at 75kV...

Wat do.

Proud Mary

Thu Apr 07 2011, 08:20AM

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

Do you have the negative lug of the 50Î¼A meter tied to Earth, and the positive to your 1.8G resistor?

What is the voltage rating of this resistor?

It's never a good idea to run a C&W without a load.

It's perfectly valid to think of an X-ray tube as a somewhat non-linear resistor in series with a diode. It will pull down your voltage quite a lot, so measure the voltage directly across the tube so you can be sure you know what is going on.

Wire a gas arrestor or small neon directly across the meter terminals to stop the whole thing going live if the meter should fail open circuit.

uzzors2k

Thu Apr 07 2011, 03:21PM

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

Is your meter plastic like this one?

I have the exact same problem with my meter, at roughly the same voltage. My setup consists of a 1G resistor made up of five 200M ones in series. These have been placed in a PVC pipe and encased in wax. I thought the meter jumping was caused by arc-overs in my 1G resistor, but perhaps it's something else. I'm sure it's not electrostatics, as that only caused slow deflections on a different meter (metal) during my first x-ray experiments. I should also add, during my first x-ray experiments I used the same 1G resistor but it was sealed with oil, and I used a voltmeter in series with a smaller 1.1M resistor. No problems then, so I'll try reverting back to this setup.

Try making a resistor divider with your 1.8G resistor and measure voltage instead. If that doesn't work any better check if it's arcing-over somewhere. With 50ÂµA sensitivity it won't take much to peg the meter.

James

Thu Apr 07 2011, 06:01PM

Registered Member #3610 Joined: Thu Jan 13 2011, 03:29AM
Location: Seattle, WA
Posts: 506

I used a switching regulator I build with a LM2576 set to about 4.2V. If you are worried about damaging the filament, make a crowbar circuit out of an SCR and 5V zener diode so that the SCR latches and shorts across the output if the voltage exceeds the zener breakdown. The problem with batteries is that they wear down quickly and your filament current dips.

You can get nice clear mineral oil easily, it's sold as a laxative for both humans and animals. I originally was getting small bottles at the pharmacy but then I found agricultural supply shops sell it in gallon and larger jugs meant for livestock. If you have problems with flashover you may need to find someone with a vacuum pump and chamber to degass the oil as it absorbs moisture and other gases.

Proud Mary

Thu Apr 07 2011, 08:14PM

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

James wrote ...

The problem with batteries is that they wear down quickly and your filament current dips.

I haven't experienced this problem myself, as yet. I use combinations of Enersys Cyclon 2V 8AH lead-acid cells and NiMH 1.2V 9AH cells, and find they stand up very well for a couple of hours supplying heater current of an amp or two - perhaps the simplest solution to the need for a floating heater supply when using a grounded anode X-ray tube.

The XRD and XRF tubes I use are designed to be operated for hours at a time without overheating, if need be. But if you are using diagnostic radiography tubes - dental types for example - the tube will be destroyed by overheating long before the heater batteries run out.

James

Thu Apr 07 2011, 09:11PM

Registered Member #3610 Joined: Thu Jan 13 2011, 03:29AM
Location: Seattle, WA
Posts: 506

The issue I saw was not so much the batteries wearing down completely, but that the output would always sag a bit. Do a dozen consecutive exposures at say 0.75sec each and the first couple come out overexposed, the next few are perfect, and the last ones underexposed. It may vary by tube though, perhaps some give a much larger change in anode current for the change in filament voltage. Regardless of the power source, a regulator is a big improvement.

Proud Mary

Thu Apr 07 2011, 09:36PM

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

James wrote ...

The issue I saw was not so much the batteries wearing down completely, but that the output would always sag a bit. Do a dozen consecutive exposures at say 0.75sec each and the first couple come out overexposed, the next few are perfect, and the last ones underexposed. It may vary by tube though, perhaps some give a much larger change in anode current for the change in filament voltage. Regardless of the power source, a regulator is a big improvement.

Have you thought this through? Your 12*0.75 sec exposures at say 2A using the 2V 8 AH Cyclon lead-acid cells I recommended will require 18 C out of the 28800 C ideally stored in the cell - a loss of 0.0625% of the charge available.

In any event, I don't see any reason why Grenadier would want to work in grounded anode mode with a dental tube, so the problem of providing a floating filament supply with a high stand-off voltage will not arise.

James

Fri Apr 08 2011, 12:20AM

Registered Member #3610 Joined: Thu Jan 13 2011, 03:29AM
Location: Seattle, WA
Posts: 506

This was a dental tube, grounded cathode, so that wasn't an issue. I was using NiMH cells, not SLA as it was what was on hand, and they did sag over time. It's been a while since I've messed with SLA since they always seem to be dead by the time I need them. LM2576 works just great though, it can run from the same 15V supply that powers the mosfet driver, runs cool and efficient, and the voltage stays put.

Adam Munich

Fri Apr 08 2011, 05:12AM

Registered Member #2893 Joined: Tue Jun 01 2010, 09:25PM
Location: Cali-forn. i. a.
Posts: 2242

After an hour of WTF (things weren't obeying ohm's law) I realized the millameter wasn't mesuring right (too sensitive). Turns out the 100 year old shunt resistor was 109 ohms; 'twas supposed to be 3.4. So... I took couple feet of 39ga and made a new shunt. 23.5v through a 10k resistor ought to give 2.3mA, the meter now says 2.3mA so that's one problem solved.

Now here's another issue; with a 20 meg resistor as a load the CW supplies 2.5mA at 5kV (supposed to be 35-40kV). The resistor burned up in a hurry (it's only 1/4W) but that's still not right; it's a huge drop. I think the problem might be the 15 meg current limiting resistor I put in there. If I do remove it (I'll probably have to) I'll have to make certain I don't let the CW spark where it's not supposed to or else it's curtains for the 20mA diodes.

Seeing as I don't have a 20 meg 300 something watt resistor to use as a dummy x-ray tube it looks like I'll have to just test the circuit using the coolidge tube. By putting it in my backyard and having the tube inside of that big 1cm thick lead box I made before, and by powering up the circuit remotely I should be safe. I could use my video camera to monitor the meters and see if things work all right. I really don't need more than 1mA out of the CW, but theoretically I should be getting only 1.6kV drop per mA:

(0.001 / (60,000 * 5e-10) * (((2 / 3) * (4^3)) + ((4^2) / 2) - (4 / 6)) = 1 666.666666666666666

Another problem, there is a little bit of tiny arcing inside the mA meter. The coil is arcing to the magnet inside, though it can't be more than a few uA in that arc. There might be something I can do there to solve that.

Pic of the meters

they are bakelite while the faces are metal and the windows are glass. For some odd reason the swinging problem on the kilovoltmeter stopped, so that's good.

Proud Mary

Fri Apr 08 2011, 08:44AM

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

Grenadier wrote ...

Now here's another issue; with a 20 meg resistor as a load the CW supplies 2.5mA at 5kV (supposed to be 35-40kV). The resistor burned up in a hurry (it's only 1/4W) but that's still not right; it's a huge drop. I think the problem might be the 15 meg current limiting resistor I put in there. If I do remove it (I'll probably have to) I'll have to make certain I don't let the CW spark where it's not supposed to or else it's curtains for the 20mA diodes.

OK, so the voltage drop across your 15MÎ© current limiting resistor when 2.5mA is flowing is:

V = IR

V = 0.0025A * 15000000Î©

= 37,500V

As W = VI

= 37500V*0.0025A

= 93.75W - almost 100W gone up in smoke, with nothing to show for it!

Now, to do away with the 15MÎ© resistance and not risk flashover, you should supply filament current to the X-ray tube a good 3 min before you supply the HV. This will ensure that the tube conducts at once when high voltage is supplied, which will pull the supply down to a safe value before it can get up to any mischief.

If you would post up a schematic of the whole setup showing where you have put the meters, and any shunts or droppers that you may have wired up to them, we can see what else can be done to sort all this out.

first previous 17 18 19 20 next last

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