I've been looking at various ways to measure current for a while now, and am stumped on how to read high-side AC Current Measurements ohmically (no Hall Effect sensors). My initial thought was to use PNP and NPN BJTs in parallel (one of each), connecting them across a 0.1 Ohm current sense resistor (basically making a differential amplifier). I can't seem to make a circuit that works in this configuration, however, and differential amplifier ICs can't handle the high voltages I'm looking to measure (up to 600VAC). I've seen a op-amp from Analog Devices that can handle common mode voltages up to 600V, but am trying to make a circuit just from discrete circuit elements to start.

My question is mainly this: is it possible to measure high-side AC current with just a few BJTs and/or MOSFETs? My ideal circuit would do the following:

Vin=600VAC, 0.1Ohm current shunt resistor on the high side, Current=1 Amp AC. --so-- V1=600VAC V2=600VAC-(0.1 Ohm) * (1 A) = 599.9VAC

BJT circuit takes V1 and V2 as inputs, and gives 0.1VAC (V1-V2).

Thanks for all your help! The replies don't have to be an answer to the question, more trying to understand how something like this could be possible.

  • 3
    \$\begingroup\$ The usual way is with a current transformer, then your problem reduces to measuring an isolated 5A (usually) AC current. It could be done a shunt, precision amplifier and some means of signal isolation- either analog or digital (transformer, optical, RF, capacitive etc.) but you also need to create a power supply that lives on the 600V rail. Rogowski coils are also used sometimes. \$\endgroup\$ – Spehro Pefhany Feb 13 '20 at 3:04
  • \$\begingroup\$ Why not measure neutral currents? \$\endgroup\$ – Tony Stewart EE75 Feb 13 '20 at 3:54

For safety reasons, AC current measurement typically performed using either a current transformer (CT), or a Rogowski Coil, or a Hall Effect sensor. For various reasons—not the least of which are equipment safety and personnel safety, current measurement in an AC power distribution line is typically not performed by inserting a series resistance into a current carrying path and measuring the voltage drop across the resistance.

And in addition other's comments to your question, be certain to provide sufficient galvanic isolation between the high voltage system and any low voltage system that uses the sampled output from the high voltage system. Galvanic isolation is typically implemented via photonics (e.g., opto-coupler), or electromagnetic induction (transformer), or electrostatic field (capacitance), or an isolation amplifier component (example, example), or some combination thereof. (NB: A CT, a Rogowski Coil, and a Hall Effect sensor provide at least some level of galvanic isolation. Additional isolation might still be necessary, however.)

The measurement system must also be designed to operate safely for the specific measurement category wherein it will be used.


Failure to provide (a) adequate galvanic isolation and (b) adequate "measurement category" design can result in destroyed equipment and can kill people when a circuit fault occurs. So this is very important.

See also:


Your Answer

By clicking “Post Your Answer”, you agree to our terms of service, privacy policy and cookie policy

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