Power Supply from an "interesting" XRT

[Rich Feldman]

100 degrees C forever won't damage any idle diode, resistor, or transformer winding. They get hotter than that inside when working at rated current, and are exposed to much higher temperatures during manufacture.

You can learn a lot from "How It's Made" TV episode about "Electric Pole Transformers", about 5 minutes long.
I wasn't going to post a URL, because the first 10 hits in YouTube search were obviously pirated. Then saw one which appears to be "official", a detail I think matters for creative works. Baking and oil potting appear 3 minutes into video.
https://youtu.be/8F9m_HN6ueE?t=185

The order shown is: bake in air, put into tank hot, pull vacuum and add oil. Order of the last two steps is unclear. If and when I pot my own dental XRT, am planning to heat it within evacuated tank using DC current in the windings (to remove moisture from 10 years of storage).

There are many reports on fusor.net about amateurs skipping the vacuum, or applying vacuum after the transformer is submerged in oil, with varying success.
I think the pros always evacuate chamber before admitting oil. The tanks (even big rectangular ones mounted on concrete pads outdoors) are reinforced to withstand full vacuum.
High frequency transformers, with fewer turns of thicker wire, might get an exception.

Suppose your finely wound XRT secondary carries tiny bubbles of air deep into the oil. When you pull a perfect vacuum above the oil, the pressure at deepest bubbles can't get below 10 torr (unless you turn off Earth gravity).
Bubbles will expand by factor of 4 or 5 in diameter, which I think isn't going to make them all escape. If the oil bubbles under vacuum even without a transformer, then it hasn't been degassed properly.

Please tell us what your expert friend recommends. Experience and practice count more than my theory and talk.

[Pablo Llaguno]

Back with updates from the past few weeks. I learned that it does take time to make your own power supply. Rich was right in that baking at 100ºC won't damage components, however if anyone does this with their own supply, make sure your wires can take the temperature. I left it all in a bucket and my HV line (made from a RG213 cable) couldn't take it, the dielectric melted. Otherwise baking it actually helped a lot in getting rid of humidity from the coils. Here is a picture before it was baked:

IMG_0055.jpg (93.91 KiB) Viewed 4389 times

After baking it for 2 days at 120ºC we put the power supply in a vacuum chamber for about a day. Then we tested the dielectric strength with a dielectric strength tester. At 5kV from the tester, between ground and the HV line we had a good impedance (in the order of Gigaohms). Not many have access to this devices, but they sure make your work a lot easier when building the tank for your supply, it also helps in making sure your HV wire can take it. Once all that was done, we put the thing in oil and tested it.

So how did it perform? Well, Mark was right in that these small size transformers are quite limited. At no load the thing does output -32kV, however with a load it is another story. My chamber is rather large as it is a Kurt J Lesker PVD system (a large tank with tons of cf flanges) so this might be a reason why my output was limited to just -12kV. I did create a plasma though:

IMG_0071.jpg (53.62 KiB) Viewed 4389 times

I put a fuse between the line from my variac to the power supply, and it kept blowing 1A, 3A, 5A and even 10A fuses. Ultimately I decided to leave only the fuse from the variac (20A). I think I pushed it too far, because I also blew that one. That is 2.4kW of input power, on the output I had -12kV and some 30mA at around 20mTorr. I assume a lot of power was being lost in the windings/core of the transformer.

Sadly, silly me made a huge mistake. When realizing the supply was limited under the plasma load, I decided to test the power supply to see if it still achieved 32kV under no load. I placed the line in an oil tank (as a insulator) and started raising the input voltage with my variac all the way to 32kV. Then, realizing I could still raise the voltage a bit more I pushed it to about 38/40 kV. Then something blew up and burned the oil and the thing is behaving really weird. Btw burnt transformer oil smells horrible and it sticks in your nose!!! Here is a picture of the mess:

IMG_0078.jpg (90.58 KiB) Viewed 4389 times

Notice the dark oil and also how some of the plastic deformed during the baking of the tank.

Due to a certain virus my university is cancelling classes, so I am not sure when I am going to be able to inspect the supply as it is in the lab. I am hoping that I blew the diodes and not a secondary coil.

Here is the whole thing

IMG_0076.jpg (126.52 KiB) Viewed 4389 times

[Pablo Llaguno]

Back with some updates, the XRT is basically done.

I opened up the supply and pulled the XRT, removing the diodes, ballast and measuring resistors. Initially there was hope, as with 1V input I was measuring 400V between both secondary windings. However, as I increased the input voltage I noticed input current was way too high under no load. At about 20V input, there was some smoke and I could see the traces of contaminants from the failure point under the oil. Took the XRT out of the oil, removed the plastic stripping that act as insulators and identified the failure, a burnt secondary winding.

I am still not sure how I burnt a secondary winding. All operation was under oil so I am guessing it was either extended runs with the fusor or when I pushed it to 40kV (probably the latter). While I could wire it up as to use only a secondary winding and a half wave rectifier, or rewind the damaged secondary, I don't think it will be worth it. What I learned from this is that as Richard and others have said, you need a heavy XRT to achieve fusion on these larger sized fusors.

Also, one can conclude that the stress on these smaller sized transformers when operating with a HV load is quite significant. Rich Feldman`s viewtopic.php?f=11&t=4805&p=27643 really shows how much power is heating these kind of devices. As I learned during Fusor operation. Also, using any radiator or pump to cool it down will not leed to success, as much of the heat is trapped inside of the windings.

As to the future of my fusor, this was certainly a huge blow, but I've learnt a lot. I hope I can find a heavy weight XRT as much of the process of building the tank and wiring it up would be similar. I am also considering making a HF supply.

[Mark Rowley]

As a group we should not offer hope that these smaller XRTs work for Fusor use. I also traveled down this rabbit hole of smoked secondaries and was met with the same results. The entire forum is littered with similar failures. Yes, one does learn some things but it's hardly worth the wasted time and money spent in the process.

Pablo, I'm saddened to hear about the failure but there is no doubt you will forge through with a working power supply. Obviously, keep your eyes open for the unobtanium heavy weight XRT, but dont count on getting one. Pursuing a high frequency arrangement is your best bet.

Looking forward to your soon-to be Neutron Club membership!

Mark Rowley

[Richard Hull]

If it is a portable, swing out over the patient x-ray head with no external supply, it will not work well or at all for a real fusing fusor. Period!

The x-ray supply you want should not be able to be lifted or moved without at least a hand truck! (the reason: Transformer iron and gallons of oil)

Only special high frequency, modern HV systems can get away with near zero mass in the transformer and absolutely no oil anywhere in the system. $$$$

Richard Hull