How to sample a domestic AC voltage supply for analyzing?

[charitha jayanath]

My project is about analyzing the voltage variations of a domestic AC voltage supply.

The standard rms value is 230 +/- 6%.

So I need to analyze the AC voltage signal and find out when the supply exceeds the given standards.

This device will be fixed to the live and neutral wires at the power meter.

The way I thought of doing this was to filter the signal to remove the noise, inverse the negative part of the wave, identify the peak of it by taking a fixed count of maximum values (eg: max 100 values-assuming 50 Hz wave is sampled at 2MHz sample rate : this will remove any residue error due to noise) and average it to get the peak, get the rms and check whether it is beyond the upper or lower boundaries and if so, record it with the date and the time. So at the end of the monitoring period, you'll get a data log of the variations of the supply voltage beyond the upper/lower limits.

The first thing I need to do now is to sample a 230+\-6% V(rms), 50 Hz wave.

So if any one has any idea of how to sample a 230 V(rms), 50 Hz AC signal, please share it with me. I'm thinking of using PIC18f4550 as the micro controller to process the sampled data.

Dropping the peak voltage using a step down transformer is not effective for it will drop down the variation (+\- 6%) of the signal.

I would very much appreciate any information you'd share with me regarding the sampling of a 230+\-6% V(rms), 50 Hz wave. (important: RMS Voltage may go a lot beyond these limits some times such as 190V - 270V. Beyond that, it can be taken as surge and ground it.)

Your Message

 

[Pretest]

If your circuit doesn't need to be isolated from the main supply, you could use a resistor divider to lower the signal and then apply a dc voltage to your output signal using an op-amp. This way you don't need to rectify the signal.

If the peak you want to sample is at 90° after the zero crossing, you can fire an interrupt at the zero-crossing and then take 100 measurement when you are around 90° and average it.

As for the zero crossing detector, inversing the negative part of the wave will be enough. You still need a resistor divider.

 

[milind.a.kulkarni]

Yes Potential devider network followed with the RMS detection using zero-corssing detection is one alternative....but I will do in this way +/- 6% accuracy can be modeled using step down transformer may be 230/6-0-6 and convert the voltage to lower range and apply the threshold switching for indication will also be possible....may be need to do some experiments there but will be more safe .....bcz if you try to add the resistane of the potenial devider .....one need to be vary careful as volatage is very high .....so normal electronic resistors will not be helpful....

Good Luck

 

[sreepss]

Potential divider will work if you are not need to isolate your device.