Reversible polarity 50kV supply suggestions please. / High Voltage / Forums

Forums

4hv.org :: Forums :: High Voltage

« Previous topic | Next topic »

Reversible polarity 50kV supply suggestions please.

1 2 next

Move Thread

LAN_403

Proud Mary

Sat Jul 04 2009, 09:18AM

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

A compact, cased, 50kV/1mA supply is required to have a reversible output for a transmission target device as follows:

Output state A: Earth/0 - +50kV
Output state B: Earth/0 - -50kV.

No 'hot switching' is required, nor are changes of polarity likely to be needed in a single session. The change could be effected, by, for example, a screw-driver port at the rear.

It is vital that the polarity could not be changed 'by accident' (expensive parts!) - or by power failure as could happen with a vacuum relay implementation.

The question seems to revolve around turning the (potted) C&W 'upside down' - easy to think about, but not so easy to implement in a compact, sealed unit.

Any suggestions, gentlemen?

HV Enthusiast

Sat Jul 04 2009, 02:22PM

Registered Member #15 Joined: Thu Feb 02 2006, 01:11PM
Location:
Posts: 3068

Typically, these type of HV power supplies have isolated output windings which can be referenced according to whatever polarity you need.

Also, compact and 50kV usually don't go in the same sentence.

Proud Mary

Sat Jul 04 2009, 02:39PM

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

Dr. GigaVolt wrote ...

Typically, these type of HV power supplies have isolated output windings which can be referenced according to whatever polarity you need.

In fact, most bench-top transmission target PSU's with a switchable earthed anode option (ie. Va = 0V, Vk = -50kV)
have two completely separate flyback and C&W chains)

Dr. GigaVolt wrote ...

Also, compact and 50kV usually don't go in the same sentence.

Some of the commercially available modular 50kV PSU's used in conjunction with micro-focus target transmission devices are about the size of a packet of cigarettes, but you have to be able to pay for them!

We're only talking a few watts here, X-ray tubes as thick as a pencil and 10mm long.

Steve Conner

Sat Jul 04 2009, 06:11PM

Registered Member #30 Joined: Fri Feb 03 2006, 10:52AM
Location: Glasgow, Scotland
Posts: 6706

Hi Harry

I once spent quite a while pondering this problem. My solution was an ignition coil-based flyback supply with a HV diode assembly that could be swivelled mechanically through 180 degrees by a front-panel polarity switch. The switch would have other poles that took care of reversing the polarity of the primary drive and metering.

You can also devise a C-W module with contacts on both ends that you can pull out of a hole in the instrument and insert the other way round. The end inside the instrument gets driven and the HV output comes from the end that sticks out of the panel. Metering and bleed resistors can be included in the assembly too.

plazmatron

Sat Jul 04 2009, 07:20PM

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

No reason why this cant be fairly compact. I would be inclined to make a small LOPT driver. Rather than turning the CW upside down, why not have two potted CW`s side by side, one positive out, and one negative out, with the outputs tied to each other. (there will be some leakage down the unused CW during usage) then simply switch the output of the LOPT, between the input of either the + CW or the - CW.

That way you will only be switching 10kV or so, without the need for any funky vacuum relays etc.

Proud Mary

Sat Jul 04 2009, 09:06PM

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

Steve McConner wrote ...

Hi Harry

I once spent quite a while pondering this problem. My solution was an ignition coil-based flyback supply with a HV diode assembly that could be swivelled mechanically through 180 degrees by a front-panel polarity switch. The switch would have other poles that took care of reversing the polarity of the primary drive and metering.

Good man! That's exactly what I was thinking! With the power entering the C&W chain at mid point, rather than end-fed.

I have a few heavy duty 10kV ceramic rotary switches - the big ATU type of thing - and these might be a good starting point if they will work as well submerged in dielectric oil.

Just as well I don't subscribe to the same miniature ethos as the rest of the micro-focus fraternity!

Thanks Steve, that was just the fillip I needed.

I think it's nearly time to start a project thread on my X-ray microscope, so many months in the planning.

Proud Mary

Sat Jul 04 2009, 10:11PM

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

plazmatron wrote ...

No reason why this cant be fairly compact. I would be inclined to make a small LOPT driver. Rather than turning the CW upside down, why not have two potted CW`s side by side, one positive out, and one negative out, with the outputs tied to each other. (there will be some leakage down the unused CW during usage) then simply switch the output of the LOPT, between the input of either the + CW or the - CW.

That way you will only be switching 10kV or so, without the need for any funky vacuum relays etc.

You know how much I like my vacuum relays, Leslie!

I have some glass and steel 30kV relays so beautifully engineered that connoisseurs have called them "the Talisker amongst relays."

Having one xformer and two C&W chains is a popular solution, as you point out, at the risk of having the unused chain floating and leaking.
At first sight, it appears to be an uncomplicated engineering problem, but the fact that commercial solutions to the problem continue to be patented tells a different story.

It's a bit like making the perfect bacon and eggs, isn't it, Leslie? Any sleazy restaurateur can smother poor ingredients in exotic sounding sauces, but once everything must be rendered clear and distinct - as with the perfect bacon and eggs - the weak spots will shew.

I'm not much influenced by mass-manufacture engineering economics, and so see no need to struggle for either the miniature, or the mechanically reproducible, but to hold ever before me the image of the hand-built Bugatti, with each bolt cut to order....

"Compact" chez-moi still has toggle switches with rubber boots and safety bars when I'm doing it.

I expect the case will be about the size of a shoe box, which is my idea of 'bench top.' I work from a right-angle section aluminium frame, and then fill it in with aluminium sheet, with sub-assemblies in Eddystone die-cast metal boxes, which adds enormously to the robustness of the chassis.

And whilst I have your attention, sir... if you are able can you send me a quick technical note on your opinion of the best developer and other details of push processing Ilford HP5 sheet film ISO 400, to 1600 and 3200 and perhaps more, in language suited to the beginner.

If the film be pushed, at what magnification will the grain size become large enough to significantly degrade the information content? What is the actual diameter of the halide grains at the pushed speeds? Can I use post-processing proprietary software to mitigate or de-granulate the grains?

I used to enjoy the chat we once had about X-rays, so here are a few refreshing points of view for you.

A micro-focus device 50kVp/250uA/silver anode has a collimated beam diameter 00,00mm1. Converted to Gy/m2, how does this compare with your giant Marxist apparatus?

Steve Conner

Sun Jul 05 2009, 05:35PM

Registered Member #30 Joined: Fri Feb 03 2006, 10:52AM
Location: Glasgow, Scotland
Posts: 6706

Harry wrote ...

I work from a right-angle section aluminium frame, and then fill it in with aluminium sheet, with sub-assemblies in Eddystone die-cast metal boxes, which adds enormously to the robustness of the chassis.

While we're wandering off topic, do you have any pictures of this construction method? I'm hoping to build another hi-fi amp soon, and I can't figure out how to make a sturdy case with integral heatsinks.

LutzH

Wed Jul 08 2009, 08:56PM

Registered Member #1721 Joined: Sat Sept 27 2008, 08:44PM
Location:
Posts: 136

Hello:

How about using a small HF flyback, followed by 2 half, or 2 full wave VM's each configured for the opposite polarity.

The switching would be then be at the input to the VM stacks. This way you could use a rather conventional switch, to switch 2-5KV HF to either the + or the - VM input, a HAM radio antenna switch would work, but it may be to big. If the ends of the VM's were connected, the diode orientation in the idle one, should block the output of the other, or maybe not.

To make it very compact you could use ceramic capacitor stacks, like the ones from Voltage Multipliers Inc, or Morgan Electroceramics. The cheapest would be to buy HV ceramic disk caps, and dissolve away the epoxy using methylene chloride solvent. (You can distil this from paint remover, or maybe use the paint remover as is, if you cannot find MeCl at your local chemical company. Then you could stack the capacitors, add the diodes, and re-pot for a very compact assembly.

I have not tested this idea in real life, but it would be easy to do with a few diodes and capacitors.

Aloha....Lutz

plazmatron

Sun Jul 19 2009, 12:25AM

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

OK.
Push processing film, is otherwise known as overdeveloping.this can be accomplished by leaving the film in the developer longer, or developing at a temperature, higher than the manufacturers specification (usually ~20Â° C).
`Pushing` is generally used for `artistic effect` where large grain size is unimportant, or even required.
If you must push process the film, I recommend developing for a longer period, rather than increasing the temp, since warm developer will soften the emulsion more, which will lead to the inevitable scratch, plus developing film slower, gives much finer control over the amount of development you wish to do.

`Pushing` 35mm HP5 to iso 800 produces visible grain in enlargements greater than 4x5 inch prints, but I have no actual data on grain size.

Personally I try to avoid having to `push` film or paper. I used to do this in the beginning, but it really does degrade the image somewhat. the best course is to get the exposure time right in the first place, if you want really detailed images.

Contrast is more easily improved when you come to make a print, than it is to remove grain, (by despeckling etc).
I recommend the program `Gimp` (Gnu Image Manipulation Program) for post processing on computer, It is available for Linux, Widows, and Mac, is totally free, and will handle almost anything you care to throw at it.

With regard to exposure times, its going to be a LONG exposure with your tube, as a ballpark figure I once did an exposure on iso 400 film with a 30kVp source, with a beam current of just 50ÂµA! The exposure time was about an hour, with a tube to film distance of 8 inches to get a visible image, so you want a stable vibration free setup.
The emulsion side needs to be facing the x-ray source. When you are working in pitch black, this seems a little difficult to get right but....
... once simple `trick` you can do when handling the film, in the dark, is wet a fingertip, and touch the surface of the film, in an area you dont care about damaging (a corner, or the row of holes if its 35mm), the emulsion side is somewhat stickier than the plastic side, since it contains gelatin.

Given the low wattage of the tube I wouldn`t worry about running the tube for such a long time, a little fan wouldnt hurt though.

Les

1 2 next

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