0
\$\begingroup\$

I want to make a mains voltage sensor for a power meter project. Since space is an issue, I looked for an alternative to a full size mains transformer. Is possible to use a voltage dropping capacitor to lower the mains voltage to about 12v and then feed it in a 1:1 small transformer for separation ?

\$\endgroup\$
5
  • 1
    \$\begingroup\$ You revealed no details of how your power meter project is powered. There is frequently a way to tap off a representative value of the mains voltage. But that depends on the design of your power supply. If you are using a switchmode power supply, it may be an advantage to use a vintage technology linear power supply with a transformer. Your question is difficult to answer as a generic question absent the context. \$\endgroup\$ May 5, 2016 at 22:41
  • \$\begingroup\$ I am only asking about the measuring sensor circuit. The measuring is done with an Arduino chip that has a dedicated power supply. Normally I would have to use a mains transformer, a voltage divider and a shifter. I want to reduce the size of the mains transformer by using a voltage dropping capacitor in series with the primary coil of a 1:1 small separation transformer (2 coils of 1200turns, 24v in, 24v out). I never saw this done and I want to know if is possible or I just have to squeeze in a mains transformer. \$\endgroup\$
    – Mihai
    May 7, 2016 at 5:56
  • \$\begingroup\$ A capacitor is NOT recommended if your goal is any kind of accuracy. I would use a small power transformer from a discarded "wall wart" power supply. Using anything not already rated for direct mains connection is too dangerous unless you had significantly more experience. Blowing off the question about the Arduino power only limits your possible solutions. \$\endgroup\$ May 7, 2016 at 6:03
  • \$\begingroup\$ I see. I will use then a mains transformer. Saw some very nice PCB mount ones that are a good compromise in size. One more question, if you can help me. Can I put 2 mains transformers close each other (let's say at about 5mm apart) ? I have 2 sensors inside the enclosure measuring 2 lines of 230v AC. \$\endgroup\$
    – Mihai
    May 7, 2016 at 6:12
  • \$\begingroup\$ It is common to rotate one of the transformers 90 degrees to minimize magnetic coupling between them. \$\endgroup\$ May 7, 2016 at 6:19

2 Answers 2

1
\$\begingroup\$

Not a legal or safe way to do that.

Reason why is at 60HZ a .22uF capacitor will allow 10mA to flow into a short circuit, plus the capacitor needs to have a voltage rating 3 times the line voltage, and be a high quality Mylar film or polypropylene to be safe, but the voltage tracking will be very non-linear.

So how do you know what load your smaller transformer is as a reactive circuit? (the capacitor in series with the transformer primary).

The best and safest and legal way to do this is to buy a transformer rated for the line voltage with an output 1/10th to 1/20th the line voltage (if using 1/20th line value you can use a tiny thread of software to multiply the results x 2.).

The transformer output is converted to DC peak value with a rectifier and capacitor. 70.7% of that value is your RMS value. Add a 10K 1 watt resistor across this capacitor to act as a minimal load, and to bleed off the charge if not in use.

Use a 1 or 2 watt potentiometer to adjust the output so your AC volt-meter is tracking the line voltage fairly close. It will work ok over a 50% change in voltage, but at low line voltages both designs would start to become non-linear, with a reading lower then is actually on the AC line.

NOTE: Seriously consider adding a properly rated fuse to the transformer primary, regardless of the design. If anything goes wrong, as in a short circuit or severe over current the fuse will pop and stop any chance of a fire.

\$\endgroup\$
3
  • \$\begingroup\$ I have proper capacitors and this circuit is intended only for measuring the mains (reduce the voltage to a lower value, shifted to have all values positive and fed in an Arduino analog pin). Thing is I never saw this before being done at small level. Only in HV transformers, using capacitive dividers. So, I better squeeze in a mains transformer ? \$\endgroup\$
    – Mihai
    May 7, 2016 at 6:06
  • \$\begingroup\$ I re-wrote the first 4 paragraphs of my answer. Please read them. \$\endgroup\$
    – user105652
    May 7, 2016 at 17:23
  • \$\begingroup\$ @Mihai: Did you notice that the capacitive divider uses two capacitors so there is no phase shift. Capacitor - resistor or capacitor - transformer will cause phase shift. If you are trying to measure instantaneous voltage and current to calculate instantaneous power you need to avoid phase shift. In any case, pay attention to Sparky's comments about capacitor failure modes. \$\endgroup\$
    – Transistor
    May 7, 2016 at 17:43
0
\$\begingroup\$

I have done this to convert a small 110 volt transformers to 240 volt. I selected the capasitor by trial and error. 0.1uf to 0.2 uf depending on the manufacturer of the transformer, there were two, same physical size. I checked the voltages, mains = 240 transformer = 110 capasitor = 300. Are we getting close to a resonant circuit? I measured the transformer inductance and plugged values into an online resonance calculator. 200Hz. This is a cheap and cheerful fix. I have never seen it used for a transformer.

\$\endgroup\$
1
  • \$\begingroup\$ Welcome to EE.SE. Since the question is about voltage monitoring for a power .meter the phase shift introduced by the capacitor would be unacceptable. \$\endgroup\$
    – Transistor
    Jun 17, 2018 at 7:58

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.