Designing New 10kV to 30kV Analog and Digital Signal Isolators

Yes, apart from electrostatic voltages, isolation becomes an issue when measuring from the higher voltages found in traction and other MV power supplies. It is also applicable to the power supplies for some sensors or control circuits, depending on the application. I appreciate that you are also covering a wide range of frequencies - in the traction world that allows us to look at all the parameters in the power supply as well as locomotive side, from 3kV/5kA DC to 25kV AC single phase and all in between. I'm not a designer, just a user, but am looking forward to any discussion here.

Hi gideon:

In first step to start with I am testing insulation quality of various insulated conductor wires used in isolation transformer, and general wiring in HV Circuits.

I have tested a set of PTFE insulated wires and coaxial cables. I found RG213 Solid PE insulated coaxial cable excellent, however PTFE cables failed after 10kV and were giving hissing sound from Corona discharges through micro holes they may have.

I am measuring Corona current up to 15kV right now but will extend this measurement up to 30kV very soon. I am limiting the maximum allowed current of 10uA to trip the HV Supply and normally look for 1uA level to record the usable limit.

Perhaps I can charge a HV Capacitor and measure total charge to see the actual charge leakage. I will also like to know if there are short duration very high current peaks that may be of serious concern.

The main design criteria is do you have some control power on the HV side or not?

With a suitable power source, you can float any type of electronic sensor that will send you the results on an IR or RF or optical fiber link in analog or digital format.

If you don't have such a power source available, things become much more complex and you are in for a difficult task.

You may want to consider using a solar cell to provide an isolated power source even if you have to shine a light to it using a bundle of optical fiber.

Good luck. I am looking forward to see some results.

The initial design has 30kV isolation in its DC power and no battery to be used. This design often suffers Corona charge leakage and that is to be minimized to as low as possible.

We can measure Corona charge in pico Coulombs or femto Coulombs.

30kV Isolation requires minimum 3 Inches creeping distance between Floating output and mains Ground/Earth and all other points at Earth potential. Right now one direction it is 4" or 100mm but in real mounting, all sides are to be maintained. One way to achieve is to seal all connectors and not to have any exposed points at floating end.