ideas on wire feedthrough sealing of oil filled HV device?

Oil is always a mess. Why not use a plastic potting material?

--

John Larkin         Highland Technology, Inc 

lunatic fringe electronics

snipped-for-privacy@highlandsniptechnology.com wrote in news: snipped-for-privacy@4ax.com:

Plastic? Do you mean an RTV type?

Oil filled containment is typically a vertical canister, by design, and wire ingress and egress would be in and out the top 'lid', or at least well above the oil fill line, and on a unit never meant to be tipped or aggitated. Usually found in permanent, hard mounted installations.

Yet another reason RTV potting media is a far better choice (especially if it is a small device/package volume, and even more serviceable inasmuch as it should not have been potted unless it worked, so it should not fail after potting unless it is a bad design. However, potted is easier to work on as small excavations can be easily back filled in all instances *except* HV sections.

What I just described I actually had for over 8 years without leaking. However, the cause for the small amounts of oil that finally leaked into a larger container just for the purpose remained a mystery, so I decided to first try using paraffin to repot everything after first removing the oil. While paraffin worked, it shrank a lot upon cooling and I didn't like that plus the fact that any heating of the transformers might compromise any insulating qualities he paraffin provided leading to possible arcovers and such. Then, I had read about additives for paraffin that increased its strength and melting point, such as EVA glue sticks. However, I couldn't find any mixing ratios on EVA/ paraffin mixes, so abandoned the idea. Budget is limited and this is a hobby device, so trying to keep costs minimal. So it was back to mineral oil like used originally, except this time in a smaller PVC junction box housing and better feedthroughs into and out of the mineral oil filled box so that I wouldn't have leaks in the future. Unless I read anything more, I will probably use epoxied solid short wire lengths as feedthroughs and then either use alligator clips to attach the transformer wires to them, or just solder everything before applying epoxy.

BTW, another possible DIY potting material I came across was "machinable wax". There actually is a recipe well known that uses a mix of HDPE and paraffin. This ends up significantly raising the wax melting point (to

240 F) and making it machinable. However, the big drawback is that 240 F is needed to melt it all together and I'm sure not going to try potting my transformer at this temp, let alone ever trying to melt it if there were problems. So... back to mineral oil.

Since a picture says a thousands words, here is what I had and now what I want to do. I have included captions describing what you are seeing with the last indicating my uncertainty about proper HV feedthrough set up:

Comments and suggestions welcome. Thank you.

Hysol white epoxy, its ceramic loaded. Ive used it in lab applications in excess of 45 Kv.

Steve

Hysol 1c off white is ceramic loaded. I've used it in lab applications at 45 KV.

CENTER your feed wire in a pvc tube, fuse or chemically weld it to your box and fill the space around the wire with some sequential layers of 1c.

It doesnt outgas either, once drt.

Steve

No, polyurethane or some such. Tap Plastics and other hobby shops have nice clear flexy casting material.

--

John Larkin         Highland Technology, Inc 

lunatic fringe electronics

p:

I still say you can mitigate the shrinkage of the paraffin by coating the d evices in layers. You can still fill in the entire container if you wish, but by coating the devices in layers each one only shrinks a bit, then more is added after the shrinkage.

Or you can fill the enclosure with paraffin only an inch or so at a time an d deal with the shrinkage that way. Since the molten paraffin will melt th e previous hardened paraffin somewhat, the layers should not cause any issu es with gaps between layers.

Oil does seem like a terrible way to do this unless your container is seale d. After seeing the RTV added as a layered insulator I think you should no t try that again. Bring the leads out the top of the container above the p araffin line.

--

  Rick C. 

  - Get 2,000 miles of free Supercharging 
  - Tesla referral code - https://ts.la/richard11209

snipped-for-privacy@gmail.com wrote in news: snipped-for-privacy@googlegroups.com:

Good advice. Make a hole and place a short PVC tube in it to keep distances. Fashion and center the wire in the tube and fill with RTV caulk or epoxy or other FIRM type potting with good adhesion. Once that cures, the tube/wire assy can be placed in the breech in the tank and fixtured in place by whatever means (more firm epoxy).

My initial try with paraffin did work. There was some shrinkage in the center, but it didn't seem to be enough to expose anything. I had two concerns about using paraffin over the long term: 1) the device is going to be in an area with poor climate control, so temps in the winter might dip into the 40's, but be in the high 80's during summer so, would this cause cracking over time, and (2) the transformers might, over a period of time, generate enough heat to start melting the paraffin. This was a very remote possibility, but I still addressed it.

In the end, I went with mineral oil once again, but I had a lot more trouble fitting the transformers into the smaller junction box. However, the project is now completed with the small junction box housed within the larger older one. This time, for leaks, the second box will catch the oil plus I custom trimmed an absorbent disposable bed pad underneath it to help absorb any oil. Believe it or not, I have already had leakage, but it is from the junction box top as that doesn't make a good seal and I don't think I want it to either since someone else pointed out how mineral oil can expand and contract quite a lot. Come to think of it, I did all of this rehousing and reoiling, but I think the problem with the original's leakage was simply from oil expanding out of the top cover during hot weather. That's the only way it would have ended up on the bottom of the tub in which the device was housed. I didn't realize this at the time and thought the leaks came from within the junction box somehow. Don't think so. I could have simply placed an absorbent pad underneath in that case too and probably would have gone on for many more years. Live and learn.

This revamping started to become costly because I had to figure out how to keep the transformers stationary within the new smaller junction box, and yet they had to be easily removable too if the time ever came. I figured it out but it wasn't cheap.

On Oct 24, 2019, Jim Horton wrote (in article ):

[snip]

The traditional approach is simply to provide some expansion space above the top of the normal oil level.

If it must be hermetic, a metal bellows can be used. Or a long bit of vertical pipe with a porous vent plug at the top so air pressure equalizes with weather and thermal expansion. The maximum observed range of air pressure variation is 26% peak to valley.

One can make hermetic seals using epoxy as the insulator, but one must match the thermal coefficients of the materials used in the assembly, or it will disassemble itself under thermal cycling.

Joe Gwinn