I built a simple full bridge rectifier with AC Mains input, but I noticed that:

  • When the input is 220 V rms the DC output is only 293 V, but I expected the DC output voltage to be 220 x sqrt(2) = 311 V.

  • The DC out drifts over the time. eg for Vin = 220 VAC I may get 298 VDC output, but if I disconnect the circuit for a while and then reconnect it I may see 293 VDC out for 220 VC in.

The circuit consists of four RL207 diodes, a 10 uF 450 V capacitor and a 1 MegOhm bleeder resistor across the capacitor.

I use voltmeter UNI-T UT50A on its 1000 VDC range .


For 220 VAC input

Why is the DC output voltage not 311 VDC as I would have expected?

Why does the DC output voltage vary as described above?

  • 3
    \$\begingroup\$ Please edit your question to explain what type of voltmeter you are using, make and model and, if possible, link to the user manual. What range is selected on the meter for each measurement? \$\endgroup\$ – Transistor Mar 29 '19 at 17:58
  • 3
    \$\begingroup\$ Also, please please please put a small resistor load in parallel with the capacitor. It's called a bleed resistor. It will aid in getting more repeatable measurements and it will help discharge the capacitor quickly when mains power is removed. Even a modest \$390\:\text{k}\Omega\$ resistor would be nice. \$\endgroup\$ – jonk Mar 29 '19 at 18:06
  • 1
    \$\begingroup\$ @Ibrahim: You can't expect us to be familiar with a UT50A. Please read the comments carefully and try to supply all the missing information. Use <Enter> x 2 for a new paragraph. \$\endgroup\$ – Transistor Mar 29 '19 at 18:24
  • 2
    \$\begingroup\$ Test your multimeter link. "Page not found." You have added a 1m resistor. That means 1 milliohm. You mean 1M? Capitals matter. \$\endgroup\$ – Transistor Mar 29 '19 at 18:58
  • 2
    \$\begingroup\$ @Vinny is not clear. He is being cool by using jargon and using the equivalent of the often rude "dictated and signed unread" as he zips off to the next rep scoring opportunity. || VTC = Voted To Close - ie he has indicated that the question is not clear, to him, even though it's meaning is entirely clear (to many) so he has voted to be impolite and unhelpful to a new contributor. Helpfully others will be more helpful. \$\endgroup\$ – Russell McMahon Mar 31 '19 at 8:17

The \$\sqrt{2}\$ relationship between RMS and peak voltages only holds for perfect sine waves. Your meter is a typical average-of-absolute reading meter with an 11.1% span shift to make it pretend to read RMS with sine waves.

The readings you are getting are more an indication that your mains power is somewhat distorted. It's not unusual to have the tops of sine waves flattened by circuits like those you are building (but drawing substantial power) unless they incorporate power factor correction, because they only draw power near the peak of of the waveform.

You should be seeing about 310VDC if the sine waves were perfect and you are seeing 293, that's not a huge difference, a bit over 5%. It's also on the low side, corresponding to the flattening I mentioned.

Similarly, the difference between measurements indicates that your mains voltage waveform is probably changing a bit.

See, for example, this answer which has actual mains waveforms from a US lab.

Both this and the household mains show a peak voltage that's lower than what we'd expect for a pure sine wave with the measured RMS value.

Your meter (when measuring AC mains) is measuring the average |voltage| and scaling it so it approximates RMS for a sine wave. When measuring the DC across the capacitor, after the bridge, it's measuring the peak voltage (minus diode drops), approximately.

For a perfect sine wave and no diode drops the ratio of Vpeak/Vrms is \$\sqrt{2}\$.

If the tops of the sine waves are flattened off, the peak voltage is decreased more than the average or RMS voltage of each half cycle. Think of it approaching a square wave where a true RMS meter would read 1.00 for Vpeak/Vrms rather than about 1.414. It will also slightly affect the reading on your meter on the AC range, which will tend to read higher than the actual RMS value. A square wave would read 11.1% higher value than true RMS on a meter like yours. Those two factors add, so the measured peak voltage is less than expected from the apparent RMS value.

If you make the mains measurement with a true-RMS reading meter, you should see a bit less of a difference. Hopefully that's intuitive if you think about what RMS really means.

  • 1
    \$\begingroup\$ Thanks @spehro , I thought about harmonics before but didn't expect thay have a such influence , and what makes that clear is that 220 AC/ reaches about 300 DC is that happen ant later night when heavy loads switched off , that happened here \$\endgroup\$ – Ibrahim Alshrif Mar 29 '19 at 20:43
  • \$\begingroup\$ can you please clarify this point you have mentioned " It's also on the low side, corresponding to the flattening I mentioned." \$\endgroup\$ – Ibrahim Alshrif Mar 29 '19 at 21:32
  • \$\begingroup\$ Thanks @spehro every thing is very clear now \$\endgroup\$ – Ibrahim Alshrif Mar 30 '19 at 2:53

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.