Supply Question

[matthecht]

Hello everyone,

Can someone give me an idea on how to "vacuum seal" an xray transformer enclosure ?

I purchased an xray transformer of unknown make and model off ebay.
It looks virtually identical to the pictures in Tyler Christensen's 5/19/2009 post labeled "XRAY TRANSFORMER DESTROYED".

In any event, I have the thing in a 12 gallon tote with some 80kv (PIV) .2A diodes to give a dual rail configuration. I tested it to 1kv DC no problem.

I also have 5 gallons of Shell AXA oil that I purchased from chemical strategies.

From what I can garner on these forums, I need to vacuum seal the device, and operate the transformer at no more than 1/20th of the rated output power for continuous operation.

Any thoughts would be greatly appreciated. I don't want to kill my baby (lol!).

Thank you,
Matt

[Tyler Christensen]

The simple answer is you need a container large enough to safely mount it in that will also hold up to the vacuum. I never found an enclosure that is good for this (don't use a 5 gallon bucket or it'll implode!)

My transformer ended up being useless because the core couldn't handle a no-load situation which is basically what the fusor is. It was good to about 25kv but above that it just ate too much primary current. I imagine yours will act the same as all big x-ray cores do, but hopefully it will be different. You might need to run it at higher than 60hz, however even that didn't fully solve my problem and I ended up burning out the secondary.

[DaveC]

A couple of points to make.

1. You DO NOT want to try to operate a HV transformer like yours, under vacuum in oil. It will have have a low breakdown voltage, and a tendency to develop bubbles deep in the windings. These will ionize and steadily destroy the insulation.

2. You DO want to have a good sealed tank for its enclosure as good as you can get or make..... with one very important proviso. It needs to have either a vent to air through a desiccant cartridge, to allow for thermal expansion of the oil, or... it needs a gas buffer space...(dry nitrogen is ideal for this), with some sort of bellows to again allow for thermal expansion. The gas volume must be at least as large as the probable oil expansion at normal operating temps.

If you build a truly hermetically sealed tank, without expansion provisons, you may destroy (by cracking) the high voltage feedthroughs. With the very stout Federal Standard bushings, these MAY be strong enough to withstand the pressure, but it depends on how warm the transformer gets, whether there was any gas space left inside, and how stiff the tank walls are, and etc.

If the unit has Alden type HV receptacle(s) these will almost certainly be cracked as the unit comes up in temperature, if the tank has no provisions for thermal expansion..

3. Vacuum Treating the Oil to dry it...etc..:.
If you have NEW, never used Shell AXA, it may well be usable "as is". Vacuum drying to about 100 microns at 80C will make sure, however, if you can put together some sort of tank for the oil

4. Vacuum drying the transformer Tank:
In order to be sure that the transformer itself is in the driest possible condition, it could be placed in a heated vacuum oven (in its opened tank) and pumped on until the pressure gets down to 100 microns, or thereabout. Holding at that pressure for a few hours while heated to 60 - 80 C, will remove as much of the water as it is practical to do. Venting with dry Nitrogen will prevent the transformer and tank from, getting wet from atmospheric air. Promptly filling with dried oil, and closing up the tank (per the discussion above) and you will be good to go.

Hope that's some help.

Dave Cooper

[lutzhoffman]

To Dave: Very nice useful instruction set, thanks. Tyler's note on the bucket made me smile, as I remembered mine collapsing, as my vacuum pump went to work.

Reference the secondary, these are designed to be loaded at a set ma load, for a given input voltage, and current. Sometimes it can help to load the secondary with a HV resistive load to keep it more stable. Yes this does waste some power, but it can work quite well along with slowly ramping up the input voltage while monitoring your output KV. For very long exposure fluoroscopy work, the transformer load is about 5 ma average for most work. A couple of foot long ohmite resitors in series under oil can do the trick.

[Dustinit]

I thought the reason for placing it under vacuum was to cause entrapped bubbles within windings to expand and bubble out. I didn't think it would be required to maintain it under vacuum.
Dustin.

[Linda Haile]

I think Dustin is correct here. Once you pump out the air and bring the system up to atmosphere there is no longer any air trapped in the windings as long as the coils stay under oil. If, for any reason, you remove the transformer from under oil, it will need the air removing again.

[Richard Hull]

A vacuum is traditionally pulled in high tension oil filled capacitors and transformers when they are dry. This pulls air out of little traps and micropores leaving a vacuum or greatly reduced pressure. The item is valved off and oil is then introduced, flowing into every nook and cranny, even mircopores, inorder to abhor the vacuum. This warrants complete oil saturation. The device is then free to balance itself to atmospheric pressure and is more of less hermetically sealed at atmosphere against moisture and oil leakage.

As Dave noted, large units will incorporate a hollow, vented bellows for expansion and contraction.

Richard Hull

[Linda Haile]

Richard, I used to work in a place where they would immerse the coil, then pump down. Air would bubble out and be displaced when the system was returned to atmosphere.

I can see that the method you suggest might eliminate the possibility of air being absorbed by the oil, but will this really be an issue?

If it does prove to be an issue, would not periodic pumping down be a viable alternative for an amateur who doesn't have large enough pockets for metal bellows, etc?

[DaveC]

To Lyn and Dustin -

Correct all around. The issue with putting a "wet' coil in oil of also unknown condition and then pumping the entire thing until bubbling stops is actually simple.

Transformer oil and most other dielectric fluids, have a rather low saturation level for water. In transformer oil, it is around 25 ppm. Above this level, the water will "preciptate" out and deposit on surrounding structures... insulating paper, metal surfaces, or wherever. Unless you also raise the temperature to reasonably high value, it will take forever to fully dry the oil and the transformer to below the saturation level. In fact, even getting to the saturation level in real equipment is difficult.

When the transformer and oil are both wet, it is ususally necessary to circulate the oil so that deeper areas are brought to the surface, where they can desorb the moisture more readily.

Wide and shallow tanks are easier to dry than narrow and deep ones, because of the hydrostatic pressure of the oil height. For rapid vacuum drying of oil, a spray system is used... the oil to be dried is sprayed through a fine nozzle - at a fraction of a gph (10 - 20 l/hr).

This type of operation requires a pretty large capacity pumping system but gets the job done quickly. Also you normally don't want to use either a turbo or diffusion pump. Too low a pressure and the transformer oil will evaporate and contaminate your vacuum pump oil.

To achieve very low moisture levels, one usually needs a desiccant in the oil, or in the gas blanket above the oil. (The latter method is very simple to implement, but slow in extracting moisture. Usually it is used to stabilize or "polish up" the lowest level achievable.

The 80C temperature I cited, is a safe of thumb, since most paper insulations, and varnishes are good for moderate exposures to temps about 105C...some newer insulations being capable of 130 to 150C if the oxygen level is low. You can always push the 80C number at least 10 -20 degrees, but remember to check for the hottest temperature in your heater system. Vacuum ovens are notoriously non-uniform in temperature.

Thus, Richard's observation that one usually degasses a previously dried transformer with (previously) dried oil under vacuum. Much, much quicker this way.

Dave Cooper

[derekm]

Dave Cooper wrote:
>....
> To achieve very low moisture levels, one usually needs a desiccant in the oil, or in the gas blanket above the oil. (The latter method is very simple to implement, but slow in extracting moisture. Usually it is used to stabilize or "polish up" the lowest level achievable.
>
...
> Dave Cooper
is activated alumina or molecular sieve a suitable dessicant?
I have to mod an oil filled HT transformer and am looking for a correct procedure.
e.g.
drain
mod the tank
vacuum pump (rotary only) to 10^-1 torr
refill under vacuum with rotary pump oil with an insert of activated alumina.
luckily the tank will fit inside my bell jar

[Rich Feldman]

We see agreement that commercial practice generally uses oil tanks designed to withstand full vacuum -- from dental tubeheads up to big pad-mounted boxes. Not a big cost or weight penalty when contents are a mains-frequency power transformer.

There's an important point that seems to have been missed in this thread,
in advice against the shortcut of introducing oil -before- evacuating a tank.
The absolute pressure and "bubble expansion ratio" at each level of the tank are limited by the hydrostatic pressure of the oil.

I will add my voice to the chorus of concern about what happens to moisture in the windings, Kraft paper layers, etc. if the transformer is not baked out under vacuum while NOT under oil.