I am making a circuit for measuring AC mains voltage and current and I want to avoid using a transformer because it introduces distortions. But everyone says that the circuit will be considered unsafe (without explaining why).

I know one of the reasons for this is that using a transformer separates the neutral and it is harder to complete a short circuit. But this isn't a big downside for safety in a case enclosed circuit (or at least that's what I think).

What are the other reasons? How is using a transformer different than using a voltage divider (apart from the neutral separation)? Is it because of possible malfunction of a resistor causing it to short?


marked as duplicate by PeterJ, uint128_t, Asmyldof, Bence Kaulics, dim Jun 26 '16 at 21:13

This question has been asked before and already has an answer. If those answers do not fully address your question, please ask a new question.

  • \$\begingroup\$ They explain only about the neutral separation in this question. \$\endgroup\$ – Vasil Kalchev Jun 25 '16 at 0:22
  • 2
    \$\begingroup\$ Galvanic isolation (what you're calling "neutral separation", I guess) is often considered a big enough benefit in itself to justify using a transformer. If you don't provide galvanic isolation, you will find you need to take extraordinary measures in other areas in order to get the safety marks you need to have to sell your product. \$\endgroup\$ – The Photon Jun 25 '16 at 1:14
  • \$\begingroup\$ Solution to your problem: have your measurement devices in the non-isolated power line domain, then use isolated power/data connections back to your user interface, etc. \$\endgroup\$ – user2943160 Jun 25 '16 at 2:28
  • \$\begingroup\$ What distortion?? \$\endgroup\$ – soosai steven Jun 25 '16 at 8:54
  • 1
    \$\begingroup\$ Beyond galvanic isolation, one of the benefit of a transformer is that it intrinsically limit the power level at the secondary. If you short out the secondary inadvertently the fault current will be much smaller than when shorting the mains wires. The energy released in this fault event could be handled more easily by further protection devices (e.g. a thermal cutout in the transformer itself) and could cause less damage before the fault is removed (e.g. a fuse breaks the circuit). \$\endgroup\$ – Lorenzo Donati supports Monica Jun 25 '16 at 8:56

Transformers are safer because they are isolating your voltage from the mains (even if your transformer is 1:1 and there is no voltage difference).

The reason is the ground connection (please see this article for difference of ground and neutral).

A really good answer with pictures you can find here

To sum it up. If your ciruit is not isolated and you touch a single wire of it the current can flow through your body to ground. If your circuit is isolated by a transformer, the floor you are standing on has no connection to the secondary circuit of your transformer. The voltage potential will be only between the secondary wires of your transformer. So touching a single wire will not let any current flow through your body. Only when you touch 2 positions of your circuit with different potential, current will flow through your body. Its considered safer as the chance to accidentally touch 2 positions in your circuit is less then only one!

A voltage divider does not have this effect, as it will not isolate your circuit from ground.

  • \$\begingroup\$ Saying "the floor is not ground any more" is misleading at best. The floor IS ground, but the secondary circuit of the transformer does not have a connection to ground. Therefore, even if you make an accidental connection to ground through your body, there is no second connection to complete a circuit and allow current to flow. \$\endgroup\$ – Dave Tweed Jun 25 '16 at 1:13
  • \$\begingroup\$ you can still get electrocuted through a transformer. The electrical distribution system uses many transformers between generator and load; even the wall 120V (in USA) has passed through many transformers. I think the transformer steps down the voltage so that makes it safer; also the max current is LIMITED by the secondary, not the wall source. But you can still have current discharge from a secondary through your body to ground, afaik. Ground acts as an infinite charge sinks; so even though the secondary voltages are floating rel. to ground, charge can still enter ground.mb i am misinformed \$\endgroup\$ – jbord39 Jun 25 '16 at 2:49
  • \$\begingroup\$ Thank you for the answer, but I was looking for other upsides of using a transformer beyond the galvanic isolation. \$\endgroup\$ – Vasil Kalchev Jun 25 '16 at 11:38

I am making a circuit for measuring AC mains voltage and current

1) If your looking to accurately track the AC mains voltage (L-N) over a wide range an isolation transformer will give you poor non-linear results. As the AC voltage drops lower the transformer core transfers energy less efficiently, so your low voltage readings are actually higher than your meter indicates.

2) To get accurate voltage measurements use probes with differential inputs, rated for 500vac or twice the voltage your testing. They have a common ground which can be tied to earth ground if not done by design. I used that type of setup to measure the voltage and current on a 0-30KVDC wire used to charge a capacitor bank for surge testing.

3) For voltages below 1,500vac you can use analog-isolators such as those that Analog Devices make, if you do not mind the $30 USD cost. If your tracking an AC voltage over a narrow range then a transformer is good enough.

How is using a transformer different than using a voltage divider?

4) Avoid voltage dividers because you do not have galvanic isolation from the AC mains, where as the methods I mentioned do, including use of a isolating transformer.

5) Think about what your needs are, and ignore the 'he said/she said' statements, or you would be to scarred to build anything. Be careful no matter what method you use. If you use a transformer use a fuse on the primary in case of things going wrong downstream.

  • \$\begingroup\$ Thank you for the suggested measuring methods. About the transformer, I was looking for other upsides of using it, apart the galvanic isolation. \$\endgroup\$ – Vasil Kalchev Jun 25 '16 at 11:46

As others have said, galvanic isolation is recommended when interconnecting a low voltage circuit with a high power circuit. There are various commercial off-the-shelf components that provide galvanic isolation—e.g., isolation amplifiers and opto-couplers. Also research how to properly route air gaps on/in circuit boards to prevent arcing from the high voltage section to the low voltage section. (There are various questions/answers on the subject of routing air gaps on PCBs this forum.)


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