I am currently working with a system that produces high voltage transients for system testing, and have some cables that need to be made/modified to work with some new equipment.

The operating voltage through these cables is (relatively) quite low, with a supply limit of 300 VAC RMS, at 16.4 A absolute max from the supply. The system is not typically required to operate near those limits, but should be capable, if required.

All of the connectors are Class III (1000V) or Class IV rated, which includes a surge rating that is more than sufficient. The wire is 10 AWG with heavy silicone insulation, sourced from extra cables that were provided by the system manufacturer.

What I would like to be able to do, is manufacture 'new' cables, by connecting new terminals/connectors (solder) and cover them with shrink wrap to prevent any external contact with the conductive elements while they are in use.

The problem I am running into, is how to assess the insulation requirements of the shrink wrap, for times when surge testing is being performed. We regularly produce 1.2/50 microsecond and 10/700 microsecond pulses, up to 6kV, that will run through the new cables. In a short circuit current condition, this can produce 1kA for the same duration. At maximum repetition, these pulses are once every 10 seconds.

The most information that I can find about many products is the dielectric breakdown (in Volts per unit thickness) and by that metric, many of the heavier duty products would be more than adequate.

My question is, can that number (dielectric breakdown) be used to reliably assess safe insulation levels for such high energy pulses, assuming that the pulse voltage is lower than the breakdown?

Or: Is there a way to assess Basic Insulation Level from information that is readily available?

I have tried contacting several manufacturers directly, but have either been ignored, or have been directed to products that are rated for 6kV+ of constant voltage on bus bars (and were MUCH too big for wires that are only 10 AWG to be protected).


1 Answer 1


I don't want to pretend that I have experience with very high voltages, but I can give you a couple pointers.

"My question is, can that number (dielectric breakdown) be used to reliably assess safe insulation levels for such high energy pulses, assuming that the pulse voltage is lower than the breakdown?"
- I would say, use that number only as reference point, but aim for at least double the calculated insulation. Personally, if I were expecting a 600V peak in my cables, I would make sure their insulation is capable of withstanding AT LEAST 1000V.
Safety is a tricky "business", and whatever you calculate, the real world is full of unexpected surprises. You could have a great insulation, but some sharp object may puncture or thin it and allow for those short pulses to eventually cause its breakdown, meltdown and finally a powerful arc or a short circuit, or simply an exposure of people to the high voltage(s). This is a very realistic scenario. A high temperature source nearby may weaken/soften the insulation, or cause it become brittle, creating cracks through which the electric field can puncture, or conductive liquid may seep into. Lightning or similar unexpected transients could superimpose and create much higher voltage peaks.
Anything of such nature can happen, just use your imagination. A safety "padding" of 50% can never be considered "over the top". Not even 66-75% (meaning the insulation capable of withstanding 3-4 times the maximum voltage.
The higher the voltage, the more unexpected paths that can be formed for a current flow. It all depends on how safe and confident in your insulation do you want to feel.
I hope that I have somewhat answered your question or helped in your decision-making process.

  • \$\begingroup\$ Currently, the product that is most appealing, would actually be almost 10x overrated for dielectric breakdown (55kV). It also lists a generic voltage rating of 600V though. \$\endgroup\$ Commented Apr 16, 2020 at 21:52
  • \$\begingroup\$ The system itself has to be run in isolation from the rest of the building, as it is actually intended to simulate lightning transients. Nothing is handled while the system is active, so the major purpose of protecting them is to avoid anything accidentally making contact in an area that is not visible to the operator before, or during, testing, as well as preventing any possible flash-over (which is further minimized by proper creepage/clearance distances) \$\endgroup\$ Commented Apr 16, 2020 at 21:59

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service and acknowledge you have read our privacy policy.

Not the answer you're looking for? Browse other questions tagged or ask your own question.