Peak Detector For measuring AC Voltage

[Gursahib Singh]

Hey guys is the circuit correct for measuring AC voltage using a PIC Microcontroller?

What would be its reliability?

A 5.1V zener can also be connect so that it does not give more than 5V.

Need your advice...

 

[KlausST]

Hi,

It doesn't work without proper GND connection.

I don't like measuring AC voltage with the help of peak voltage. I find peak voltage is not that reliable.
But I know many people do it this way. It only works good when it is undistorted sinewave.

But main voltage is distorted. Often it has a flat top, so it shows less RMS voltage than expected.
On the other hand every spike near the top of the sine will result in showing higher voltage than expected.

Maybe, maybe, maybe both effects compensate each other and it shows the correct value.

Klaus

 

[Gursahib Singh]

Then how should i measure the AC voltage?

 

[betwixt]

First tell us where the AC voltage comes from. Klaus correctly warns you about the risks of measuring the peaks of the waveform which could be distorted but it depends on the particular circumstances. AC covers a multitude of possibilities, are you measuring 440V AC line voltage or maybe the square wave output of a signal generator, is it 50Hz, 60Hz or a 30GHz satellite downlink signal? Please be more specific about the source of the voltage and give an idea of how accurately it has to be measured.

Brian.

 

[crutschow]

Then how should i measure the AC voltage?

For more accurate measurement you can use the rectified average value of the AC waveform.
That value is less affected by distortion and spikes in the waveform.
The average value is obtained by low-pass (RC) filtering the rectified signal. In your circuit just place a resistor of the appropriate value between the rectifier output and C1.

But note that, if the voltage is from the mains, it is a serious electrocution hazard since you show no isolation. That's a no-no. You need a transformer or other form of isolation.

 

[KlausST]

Hi,

Rectified average is a good solution.

But when you speak of an AC voltage it is meant as the RMS value.
So the best solution is to take a lot of samples an mathematically calculate the true RMS voltage.

I usually do it this way. But it needs to use interrupts to get a constant sample rate...

Klaus

 

[Gursahib Singh]

Hi,

Actually i have to design a Panel meter which continuously shows Average/RMS voltage (Whichever is easier to design) on Seven Segment Displays. I am using a PIC Microcontroller for the same.

The AC voltage can be from any source. But mostly from Mains. Isolation is not to be used. My boss has restricted me from using transformer. Max Voltage to be measured is 500V(~700 Peak).

 

[KlausST]

Hi,

shows Average/RMS voltage (Whichever is easier to design)

Is it a known waveform? Or mains voltage?
Then average is definitely the easiest: Just show "0". Finished ;-)

I hope the shown value is always "RMS", but you mean it is regardless how to measure this value..
If so, then the measurement with "rectified average" is the easier method from software side.
From hardware side the true RMS may be easier, because you need no rectifier and no average filter.

Klaus