Insulating a 10 kV solder connection

[trubetz]

Greetings everybody

I have a task of permanently connecting two wires (20 AWG in rubber insulation 4 mm in diameter) that operate at a voltage of up to 10 kilovolt. The question is: how do I insulate the solder joint? The heat shrink tubes that I have at hand have a voltage rating of 600 volt, so these are clearly not enough. I think there must be an industrially correct way of doing this, but I do not know how to find this either. Please advice.

[mikeselectricstuff]

You need to exclude all air to prevent corona - probably need to use a sleeve filled with something like silicone

[Gyro]

I hope this isn't a high energy situation!

Assuming low current cabability, I've found that multiple layers of 3:1 ratio adhesive lined heatshrink work well - the smallest sized to be a close fit on the rubber insulation before shrinking, to exclude air as Mike says.

[trubetz]

I hope this isn't a high energy situation!

Absolutely not, we are talking about a power supply that has a short circuit protection at about 100 uJ energywise.

Speaking of the layers, how "multiple" are we talking about? Obviously I cannot stack 5+

[Psi]

If it was me, I would use some electronic rated silicone RTV to bridge the gap all around the solder join and increase the diameter to match the insulation.
Once that cures I would heatshrink over the join using the kind of heatshrink that comes with built-in glue inside.


But it wont be able to handle much bending stress, if any.

[Someone]

Starting with some heat shrink that specifies a minimum wall thickness and significant dielectric strength (such as XPLO) is only half the battle, that gets the voltage rating in theory but controlling the fields (avoiding voids!) is an art in its self. The cable supplier should have some suggestions.

[Ian.M]

It may be preferable to do a rigid potted joint, as getting all the bubbles out is easier if you can de-gas pourable resin in a vacuum chamber vs trying to apply silicone sealant without trapping any air.

[nfmax]

Possibly paraffin wax? If you can fully immerse the joint into melted wax and allow time for it to fully soak in. You may need a protective mechanical layer as well, though.

[robert.rozee]

I have a task of permanently connecting two wires [...]

a. is it practical to just replace the two wires with a single (longer) wire?

b. if a single wire is not possible, can you use a plug and socket rated at 10kv?


cheers,
rob   :-)

[TimFox]

If you must solder the two wires together, be sure there are no sharp edges or points sticking out, where the local electric field will be far above the average value.
Can you solder them into a copper sleeve before wrapping them in appropriate insulating tape or heat-shrink?
The 600 V rating on your heat shrink assumes that two conductors are separated by the tubing wall, with 600 V difference between them.
It's the same way insulation ratings for normal wire are specified.

[Doctorandus_P]

A short search:
https://html.duckduckgo.com/html?q=high+voltage+heat+shrink

quickly finds:
https://www.mcmaster.com/products/high-voltage-heat-shrinks/

which claim to have heat shrink up to 5kVac, so 2 layers should suffice.

Another tip I saw in a 3M datasheet: With the glue filled types, start crimping them from the center, and then slowly moves outwards, to reduce the chance of air bubbles.

Another option is to have a look at how high voltage cables are coupled, and use the same materials. It looks like they use the self-vulcanizing rubber tape instead of heat shrink. Topping that off with a layer of heat shrink may be a good idea too. The high voltage power cable splices are built up from multiple layers.

I also don't understand where the "600V" rating for "regular" heat shrink comes from. Nearly all plastics have dielectric strengths of 10kV to 40kV per mm. the 600V rating is very likely just a waiver so they are not responsible, can get away with less stringent quality control and can therefore put the stuff on the market for a lower price.

[TimFox]

For really high voltage cables (100 kV), we used a special high-dielectric constant (and somewhat conductive) tape around the conductor, with an EPR self-fusing rubber tape over that, and finished with a self-fusing silicone rubber tape.
With self-fusing tapes, one applies them "half-lapped" and the layers adhere to each other.
When we started using the high-dielectric constant tape (in grad school), the technician helping me and I both wondered about the other guy, since the tape smelled like dirty tee shirts and we each thought the other bathed normally.

[Psi]

... and finished with a self-fusing silicone rubber tape...

Self-fusing tape is a great idea for high voltage.
It's not an idea i had considered, but it makes a lot of sense.

[Zoli]

... and finished with a self-fusing silicone rubber tape...

Self-fusing tape is a great idea for high voltage.
It's not an idea i had considered, but it makes a lot of sense.

3M 70 is a tape I worked(for high(~150^oC) temperature ); it has a 34kV/mm specified dielectric strength.

[CosteC]

10 kV is relatively low
one layer of tape use for transformers usually withstands 5 kV, so 3-4 layers makes sensible insulation.
Main challenge is to make it in even, overlapping layers, without trapped air.
Heatshrink tubes with glue shall work well too. Standard heatshrink tubes have withstand of 10-20 kV/mm.
Silicon tubes can work in higher temperatures but are not so conveniently adjustable in size/diameter.

IPC-A-610/620 recommends smooth, round soldering joints. Spikes are not acceptable as cause high field gradient as well as can damage insulation mechanically.