One of the application im working on needs three phase power measurement to be done through micro-controller. The power measurement is in the range of 90kVA. Is there any way to noninvasively/invasively measure the current/voltage to get actual power consumed by load(compressor in this case).

I want to understand a way to measure a Power consumption with a sensor and update data to server with IoT to remote server. micro-controller aspect here is peripheral and any WiFi module can be used to update data to server.

While referring to some material I did came across some methods as Abraham electrostatic voltmeter or Transformer ratio method and some other for voltage measurement.

High Voltage Engineering - J R Lucas 2001

But need to understand which method is used in commercial measuring instruments. Also is it practically possible to keep any WiFi or RF node for data collection at its vicinity?

Note:I clearly understand the risk at these Power levels. Im here to understand the technology so that i can concentrate on my part of design.

  • 7
    \$\begingroup\$ Welcome to EE.SE! KVA is not a voltage measurement unit but rahter apparent power. Also, if you need to ask about something that's 90 KVA, you need to seek professional help by someone certified. \$\endgroup\$
    – winny
    Feb 8, 2019 at 13:00
  • 3
    \$\begingroup\$ As mention VA isn't VOLTAGE, it is POWER. Also it is lower k not upper K (K=Kelvin, k=kilo). now if you are dealing with 90kVA,, high chance this is 400Vac but this isn't "VERY HIGH" this is legally speaking LOW-VOLTAGE. The uC is the least concern here. You need to know what you are measuring, the safety surrounding grounding. Do you have access to neutral or is a virtual star suitable. Then... simple case of high-resistance, buffer, ADC and isolators. I won't post more details because anything over 50Vac you should know and if you don't, you shouldn't be working on it \$\endgroup\$
    – user16222
    Feb 8, 2019 at 13:27
  • 3
    \$\begingroup\$ At 90kVA you need a professional electrician... \$\endgroup\$
    – Solar Mike
    Feb 8, 2019 at 13:39
  • 2
    \$\begingroup\$ so you are after VOLTAGE and CURRENT measurement. What voltage ? this will drive the current and voltage measurement scheme. \$\endgroup\$
    – user16222
    Feb 8, 2019 at 13:56
  • 5
    \$\begingroup\$ "Question is very simple" At 90 kVA, nothing is simple. \$\endgroup\$
    – winny
    Feb 8, 2019 at 14:41

2 Answers 2


Generally speaking, non-contact power measurement is not possible. That is, you cannot make a device which will make no electrical connections to an isolated set of wires and determines the power being provided to the load.

Power (kVA) is the product of voltage and current. It is possible to make clamp-on current meters, which take advantage of the fact that AC current produces an AC magnetic field which can be sensed by, for instance, Hall effect sensors or a specially-constructed transformer. These are pretty standard devices, and come in a large range of scales.

Voltage, on the other hand, does not lend itself to non-contact measurement.

With that said, one might (in theory) build an AC power system with a built-in load (such as a resistor of known resistance), and it would be possible to do a non-contact measurement of the current in the load and thereby back out the voltage. But I doubt that this is what you're thinking of.

However, all of the above is pretty much beside the point. As others have commented, 90 kVA is nothing for someone with no experience to even think about messing with. Your question suggests that this is not a homework exercise, but rather a business situation. Based on what you have asked, I suggest very strongly that you tell your boss that he needs to hire either an experienced engineer or a consultant with a proven track record. I'm sure you'd rather not plead inexperience to your boss, but in this case it is very much the best option. A commercial high-power installation is not a good laboratory for self-education.

  • \$\begingroup\$ Dear @WhatRoughBeast, This is what i was expecting. As there is very less information I found through google, I came here expecting to get some knowledge. Its Obvious that I was not expecting a solution here. I understand there are other ways to get information. But look at this thread and see how its helpful for others. At least its helping me. \$\endgroup\$
    – Rajendra
    Feb 10, 2019 at 9:11
  • \$\begingroup\$ I understand from the beginning the seriousness and risk involved in this domain & at this voltage, but there is no way to show it in my question. Frankly speaking Im into IoT domain and Im no expert in Electrical and High Voltage area. My world revolves around 5v. Ive no boss... im the owner of small startup. Eventually I will improve my question as im slowly getting the details of this topic. \$\endgroup\$
    – Rajendra
    Feb 10, 2019 at 9:12

To take a three-phase power measurement you need to measure the instantaneous voltages and currents so that you can calculate instantaneous power and integrate that value to measure energy consumption.

Current measurement can be done relatively safely by using split-core Hall-effect sensors of which there are many types.

enter image description here

Figure 1. Split-core Hall-effect sensors of which there are many types.

These can be clamped on while the equipment is powered down and the signal wiring routed safely away from the mains and out to the energy meter.

Voltage measurement presents bigger challenges as it generally requires connection to the supply being monitored, requires proper fusing and isolation from the energy meter analog and logic circuitry. This would best be done by three small mains transformers but you may have trouble getting these at your required phase to phase voltage if there is no neutral available.

Since the compressor is most likely a balanced load you may be able to monitor one-phase only and multiply the energy consumption by three.

Go on over to the Open Energy Monitor project where most of this has been thrashed out already.

  • \$\begingroup\$ Thank you for info@Transistor, but we are talking about voltage up to 90kVA, does all above sensors will work the same way at this range? \$\endgroup\$
    – Rajendra
    Feb 14, 2019 at 5:29
  • 1
    \$\begingroup\$ 90 kVA isn't a voltage - it's a power measurement. This error suggests that you are missing some very fundamental understanding of your project. For current measurement you need a current sensor that will work in the range of current (A, not kVA) that your compressor uses. \$\endgroup\$
    – Transistor
    Feb 14, 2019 at 7:12
  • \$\begingroup\$ word voltage is coming because I will also need to measure it to get Energy/Power consumed(correct me If Im wrong), while current is must and thats what we are discussing in all above post. I think Ive tried to clear my question above by making changes in it. If some mistake are made unknowingly should be ignored. You have also already mentioned about voltage in your answer. \$\endgroup\$
    – Rajendra
    Feb 16, 2019 at 19:33

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.