Vacuum Rectifiers X-rays report / Projects / Forums

Forums

4hv.org :: Forums :: Projects

« Previous topic | Next topic »

Vacuum Rectifiers X-rays report

first previous 11 12 13 14 next last

Move Thread

LAN_403

radhoo

Fri Feb 18 2011, 06:49PM

Registered Member #1938 Joined: Sun Jan 25 2009, 12:44PM
Location: Romania
Posts: 701

Pictures speak for themselves:

3 American 2X2 variants from different periods and the russian 2X2 I've mostly used in this thread.

I'm going to check the emission in regards to:
- the different Anode bell metal
- the different Anode Bell size
- the different oxide-tube size
- differences in internal pressure, I expect the 1954 to have a softer vacuum and so a poor performance

Regarding the points of emission. The images I took previously show the very uniform emission. Why do I keep insisting on that?
Well its because there are only a few regular geometric shapes/bodies inside the 2X2 tube and those are:
- the anode bell as a closed end cylinder
- the cathode oxide tube
In my opinion these two are responsible for the X-ray emission and the field distribution.

If the x-ray emission resulted from another point inside the tube, than I wouldn't have obtained images like these:

If the source was a single point inside the tube, then we would have seen one (or a few ) brighter spots instead this uniform illumination.

Even further, in "09.Vertical emission angle"

I demonstrated the emission is given by a source at the same vertical level as the the oxide tube covering the filament.

These two observations indicate the emission results from a bombardment of the oxide tube by electrons emitted from the WHOLE inner surface of the anode bell. The shape of the two is reflected in the uniform illumination of the fluorescent screen.

Why the oxide tube and not the filament inside?
- Because the tube is a perfect cylinder. If the target was the filament, having the shape of two wires in close proximity, the resulting 360degrees X-ray images (see blue images above) wouldn't have been identical . Instead as the shape of the filament is not geometrically regulated, we would have seen more shadows vs. intense spots and not a uniform distribution of the X-ray emission field all around the tube.

Why all the bell's inside surface is an emitter?
- because the field is stronger near the center of the bell and decreases in a gradient towards the top, where : the distance to the center electrode is bigger, -or- there is not oxide on the top of the cathode tube.

Another possibility regarding the x-ray emission ,but not related to the anode/cathode shape, is that of electrons impacting the glass. As shown here:

some tubes took advantage of this property to totally exclude the use of a metallic cathode target, and use glass for x-ray emission.
See the pictures I posted of the operating tube: it has a very strong blue glow.
However , there are some issues with this theory: first because the anode bell works as a shield that would block any inside emission, and than its the polarity applied to the tube. The Anode bell is negative while the internal oxide tube is positive. The electrons would take the opposed path, going towards the center and not the glass walls.

Proud Mary

Fri Feb 18 2011, 07:36PM

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

It's interesting to compare the Russian and American interpretations of the A-suffix ruggedized 2X2 'For applications critical as to severe shock and vibration.'

The Russians have applied a dark body radiator coating to their 2X2A anode, to reduce its temperature, and hence the heat conducted into the top cap glass seal.

They have shortened the cathode and filament assembly support rods to reduce mechanical advantage on the glass seals, but the price of this has been
the need for a heat shield to deflect radiant heat away from the glass-metal interfaces coming up from the base.

Finally, they have replaced the '4 Small Pin' base with the much more transplantable International Octal base configuration, to carry the design forward into the next generation of thermionic technology.

radhoo

Sat Feb 19 2011, 10:16PM

Registered Member #1938 Joined: Sun Jan 25 2009, 12:44PM
Location: Romania
Posts: 701

12. 2X2 CCCP vs. USA ??!
Note: 4 variants of the 2X2 Tube has been used. The tubes stand in vertical position. The camera has been set on a tripod, with unchanged parameters (same focus/exposure time/etc). All the tubes has been powered up in inverse polarization, using the same potential difference.
Objective: compare the x-ray emission output by using the fluorescent screen
The tubes used:

From left to the right:
The russian 2X2 (that I've used several times in this thread).
Three american 2X2 variants, the forth appears to be the oldest (manufactured in 1954)

In the same order, here are the results:

The Russian 2X2 emits the most radiation. Actually the X-ray output a few times the output of the other tubes. An impressive difference.
The setup has not been changed, extra care was taken when replacing the tubes not to change the distances / fluorescent screen position from the camera, so differences in the emission field can also be observed.

The last 2X2, from 1954 seems to have 0 x-ray output (vacuum inside changed because of age? ).

radhoo

Sat Feb 19 2011, 10:28PM

Registered Member #1938 Joined: Sun Jan 25 2009, 12:44PM
Location: Romania
Posts: 701

13. Targets
Note: the camera has been brought dangerously close to the setup, to take photographs of the electron targets:
1) In normal polarization the source is the normal central cathode , and the target the inside of the electron bell
2) In inverse polarization, the source is the inside of the electron bell, and the target the oxide cylinder covering the filament.

The 1954 American 2X2, used in inverse polarization:

Remember, this tube has very little (if any) x-ray output. Since I used the fluorescent screen for detecting radiation, I used no dosimeter (Radex 1706) in this test scenario. The third picture has the exposure reduced, so you can see the oxide tube getting hot-red because of the electron bombardment. No other component inside the tube seems to be targeted (not even the wires supporting the oxide tube).

Another American 2X2, in inverse polarization, with the oxide tube glowing because of the electrons:

An now, the last tube again, in direct polarization. In only a few seconds (aprox. 6 seconds), the anode bell is incandescent, because now it becomes target of the electrons emitted by the central cathode:

On the other hand, my tests showed that the Russian 2X2 doesn't get incandescent not even after 15 seconds!

Luckily my camera survived the dangerous proximity of the setup, and no arc-overs occurred to it.

radhoo

Sat Feb 19 2011, 10:38PM

Registered Member #1938 Joined: Sun Jan 25 2009, 12:44PM
Location: Romania
Posts: 701

14. Pinhole camera - a Paradox?
Note: Using the Russian 2X2, a sheet of lead has been placed in front of the tube, in vertical position. In front of the hottest x-ray emission zone, a pin hole has been made in the lead sheet. The distance between the tube and the lead sheet is 6 cm. On the other side, at 4 cm distance, the fluorescent screen has been placed. So the distance between the fluorescent screen and the tube is a total of 10cm.
The camera has been brought close to the fluorescent screen and focus attempted using the super-macro setting.
The setup:

Based on my previous experiments, I would have expected the camera to record a filled spot (since I assumed we have an uniform SINGLE x-ray emitter source - the central oxide bar). Instead... :

The images have been enhanced in software. Now a lot of damage on the CCD (random pixels) also becomes visible.
Comments:
If I'm not wrong, it appears there are 2 points of emission in horizontal plane? If yes I can only think of the glass walls (left-right).

Proud Mary

Sun Feb 20 2011, 12:55AM

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

This is an excellent start, Radu, but there is a risk of certain assumptions and methodological errors clouding our understanding of what is happening here.

1) How does the perceived relative brightness of a fluorescent screen relate to the energy and fluence of the X-rays exciting it? What does it mean to say that one optical event is "twice as bright" as another to the human eye? Does the light output of your screen have a linear relationship to X-ray energy, and to X-ray fluence, or is it non-linear? And so on.

2) As you do not record the potential difference across the tubes, nor the current in each case, there is a chance of being seriously mislead. It is perfectly possible that the American valves had a lower impedance in field emission mode, and drew more current than your supply could deliver without a significant drop in voltage. Once this voltage falls below ~20kV, the X-radiation will be increasely attenuated by the glass envelope, but this will not stop the glass from fluorescing blue from photons striking its inner surface, as we see in your excellent pictures. This would explain why the poor X-ray emitters have the effective anode (i.e. the maker's cathode) white hot, while the good Russian tube does not have enough current flowing through it to heat the efffective anode to incandescence. (You will recall that the old TV EHT thermionic diode rectifiers only became an X-ray hazard when their heaters failed, and their impedance rose to a high value, so the full EHT voltage appeared across them)

I shall finish working on my dosimetry experiment tomorrow night, using the two Russian 2X2As I have been able to find, and so will be able to tell you the
actual dose rate in Gy/hr across 15kV - 40kV in 5kV increments. I'll post up the graphs on Monday morning.

radhoo

Sun Feb 20 2011, 10:31AM

Registered Member #1938 Joined: Sun Jan 25 2009, 12:44PM
Location: Romania
Posts: 701

Proud Mary wrote ...

1) How does the perceived relative brightness of a fluorescent screen relate to the energy and fluence of the X-rays exciting it? What does it mean to say that one optical event is "twice as bright" as another to the human eye? Does the light output of your screen have a linear relationship to X-ray energy, and to X-ray fluence, or is it non-linear? And so on.

LE: Let's not discuss brightness in the biological context, eg. as perceived by the human eye, but use instead another sensor such as a CCD camera. By doing so we would need to quantify any image focused by the camera lens on the ccd into pixels. For simplicity we won;t use colors, but gray scale colors instead, where complete blackness would be attributed a value of 0, and white would be 255. A photon striking the LCD will create an electrical signal related to the photon initial energy, and so we will get a darker or brighter pixel (0...255). Of course there are limits to the CCDs like too much energy (>255) that the CCD won't handle (CCD blindness), but it's too early to discuss error cases. There won't be too much light to get anyway.
The "fluence"/intensity of x-rays relates not to the brightness perceived, but to the exposure time needed to capture the image. It's similar to water flowing through a pipe, filling a glass, or to electric current filling a capacitor, since. Intensity is a function of time.
An event would be brighter than another, if the image as recorded by the CCD is brighter, meaning it contains at least one pixel with a higher value than all other pixels in the image we relate to.
The light brightness output of the screen can either have a linear relationship to energy or a non-linear, it's not really important, since a scale can be determined empirically by using a tube and a variable High voltage supply. For each increasing step, we would record the difference from previous frames, since X-ray emission contains the max energy in a small fraction, and all the other lower energy levels.
So do you still thing this is not doable? Your CdS+fluorescent screen dosimeter comes with a similar approach (also requiring calibration for any useful purpose).

But probably you're right, it's not good to jump to conclusions with so many uncertain premises. At least we have something new to think of and this story goes on. Speaking of something new, here is a crop of the original pinhole camera image (I selected the best image I got, from 3 attempts):

See the big image. I've selected in blue something barely visible, that attracted my attention while looking over the photos.
Too bad I didn't mark the position of the pinhole in regard to the fluorescent screen! I'll have to do this again.

LE: first at didn't want to puncture the only lead sheet I have and tried to use a 0.5mm Aluminum sheet, that made no difference, as the x-rays were passing right through it. The lead sheet was a different story: it is 1.6mm tick.

Proud Mary

Sun Feb 20 2011, 12:00PM

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

I've got to get the Sunday roast into the oven, so can only say a few words for now.

Firstly, I shall be cross if you spoil your camera, because you will be needing it.

If you are finding it difficult to get lead glass, ordinary plate glass as used in big shop windows will give significant protection to your camera at the photon energies being used in these experiments.

As to your pin-hole radiographs, do you now think that the gap between the two nebulae is an image of the gap between the two filament supply rods? If it is, then the gap should disappear if you turn the valve through 90Â° on its axis.

If you think the X-rays are being produced by electrons leaving the edge of the heat shield impacting on the glass - or the flashed getter, or both - then taping a narrow strip of aluminium foil round the outside of the glass on the plane of the heat shield and then connecting this to Earth should increase the x-ray fluence (if you can stop flashover, and you don't get dielectric puncture of the glass! )

Ah well, the kitchen calls!

plazmatron

Sun Feb 20 2011, 08:13PM

Registered Member #1134 Joined: Tue Nov 20 2007, 04:39PM
Location: Bonnie Scotland
Posts: 351

@ Radhoo:
your pinhole cameras performance is a lot better than I would have expected! If you manage to get hold of some thick plate glass to protect you camera as Proud Mary suggests, it may be worth making a much smaller pinhole, and trying a much longer exposure to see if you can obtain a higher resolution image.

Also kudos on the much needed comparison of the American and CCCP 2X2's I am really looking forward to trying this out now!

Proud Mary wrote ...

I shall finish working on my dosimetry experiment tomorrow night, using the two Russian 2X2As I have been able to find, and so will be able to tell you the
actual dose rate in Gy/hr across 15kV - 40kV in 5kV increments. I'll post up the graphs on Monday morning.

I will be interested to see that too!

Proud Mary

Mon Feb 21 2011, 09:25AM

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

plazmatron wrote ...

I will be interested to see that too!

Later - Monday morning events in real life have intervened!

first previous 11 12 13 14 next last

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