# High Voltage Measurement

I went for an interview for scientist position in BARC India. They asked this problem

How will you measure a high voltage like 765kV AC ?

I replied that we can step down using potential transformer and then measure it using a voltmeter.

Then they said tell me some other method?

How will you measure HVDC line of 500kV?

• I suppose the answer "very carefully" is not what they're looking for? Jun 10, 2018 at 1:55
• I am not able to understand Jun 10, 2018 at 1:56
• IIRC you can use polarisation rotation effects in optical fibre to measure voltage gradient in a fibre that just goes up to the line and down again. Was I dreaming, or is this a thing? Jun 10, 2018 at 6:25

HVAC is often measured with the cable interface insulators called bushings ( usually porcelain , glass or polycarbonate) that have built-in electrodes with an specified low capacitance value , supplier to large diameter coaxial jacks and have the same high BIL ratings Basic Insulation Level, to a low voltage higher capacitance load that acts as a capacitance voltage transformer of 10k:1 or cascaded to the desired end voltage range.

HVDC can be measured with E field sensors using a known impedance current loop as a voltage divider. These must be clean dry air gapped E Field sensors designed to withstand the HVDC breakdown in a very clean environment . Although I am not up to speed on laser E field tracking and other modern methods.

• and in case of HVDC? Jun 10, 2018 at 2:11
• Can you please share a link about bushing which I can refer Jun 10, 2018 at 2:12
• Is this what you are refering to en.wikipedia.org/wiki/Electrostatic_voltmeter Jun 10, 2018 at 2:47
• The correct answer is the bushing capacitor for HVAC and a wireless E field sensor for HVDC. en.m.wikipedia.org/wiki/Bushing_(electrical) Using any small meter on 0.5 MV would be fatal. Using resistors that can be exposed to contamination would fail. For these reasons , the other answer is incorrect. Jun 10, 2018 at 3:54
• please take a look at this problem electronics.stackexchange.com/questions/379060/… Jun 10, 2018 at 7:31

Caddock makes a line of ultra-stable resistors with a 20 KV rating and .025% accuracy. They can be soldered in series. The 40 KV Fluke test probe is just 2 of these at 500 M each for a 1 G ohm probe.

50 of them in series would give you 1 million volt isolation, if they are covered in heat shrink and have ample clearance to any conductor. The impedance would be so high you would need a CMOS op-amp to buffer the resistor divider before amplifying it to a useful level.

If it was AC you could use simple capacitive dividers.

• And in case of HVDC ? Jun 10, 2018 at 2:12
• I have never seen anyone in the transformer or power industry use this method , although it may work, it is not standard practice.being tested for your experience in this question. Due to creapage failures , this method is not suitable. Jun 10, 2018 at 3:48