# How do residential analog and smart meters measure power?

I'm building my own digital meter in order to upload the data to a SQL database; up until now I know that there are several parameters being measured into a digital watt-meter:

volts, current, apparent power, instantaneous power, actual power, power factor

However, I still have to understand which one of these is the real value used to increase the counter on both devices. To be more precise with my question, old analog meters couldn't do all these calculations, and still they worked out as intended. However with, no linear loads being more and more common, I'm guessing the power factor in a common house would be hanging around .7 or .8 so how does this change in the type of load would affect the measurement in a digital vs analog watt-meter?

My first guess would be that analog meters measure REAL POWER, but I can't be so sure about digital smart meters. How do they work?

The analog meter is built around a motor. The magnetic fields that produce the torque to drive the motor are proportional to the voltage and current at any instant. So, in that sense, it is measuring instantaneous power. Because it spins and turns a counter, it also measures the total energy consumed.

The digital meter measures voltage and current directly, and over the course of many samples can measure and accumulate voltage and current readings, as well as calculate the apparent power used. Real power is calculated by multiplying the voltage and current at any instant (one pair of samples). Apparent power can be calculated by measuring the amplitudes of the voltage and current over the course of one or more cycles and multiplying the two results together.

An additional consideration is seeing what you are charged for. Residential customers (at least in the U.S.) traditionally paid only for real power consumed. This was probably due to the fact that originally most loads were predominately resistive, and also the fact that while analog meters that can measure power factor do exist, it wasn't cost effective to use them in residential neighborhoods.

Nowadays, our uses are quite different, and switching power supplies can really mess with the power being supplied to the grid. This has resulted in regulations forcing power supply builders to get the power factor back closer to 1. I'm still not aware on anyone other than industry paying for power factor, but if the smart meters can measure it, that situation could change.

Analogue meters measure the average value of the instantaneous values of current and voltage multiplied together: -

• power = Average(v x i)
• energy = integral(v x i) Power factor is irrelevant because there is only one way to measure power and that is how I have stated it.

Trying to "use" power factor, RMS voltage and RMS current in order to calculate power is a fruitless task given the number of harmonics present in many current waveforms.

A moving coil power measurement device (aka the dynamometer): - These days microprocessors are used and they sample voltage and current waveforms and multiply the numbers. Surprisingly high sample rates are used to avoid the harmonics in the current waveform aliasing the ADC.

If power factor is needed to be displayed it is done by taking real power and dividing numerically by the product of RMS voltage and RMS current. The same samples that are used to calculate power can be used for calculating RMS voltage and RMS current.

A company that I used to work for (Landis and Gyr) pioneered the hall effect power meter - this was basically a solid state version of the two-coil dynamometer shown above.

Just in case there are any doubts that the power measurment method can cope with harmonics of current take a look at this: - Here we have voltage at 50Hz and a common harmonic (3rd) of current at 150Hz - see that the instantaneous waveform of power has zero average, just as it should do.

I too have wondered about the precise functioning of my recently installed smart electricity meter. However, I would expect to be billed for the real (kWh) energy I consume and this has to be accurately sensed and computed by the smart meter, which can presumably accommodate both low power-factor (eg of fridge motors) and/or harmonic-laden current waveforms (eg of electronic equipment). I would expect this also to apply to the measured energy (kWh) consumption and and power (kW) levels viewable on the in-house display unit. I can only conclude that true power and consumption figures are always determined and presented. In this respect smart meters are far superior to earlier monitors which used a clamp-on current transformer but were unable to sense mains voltage and so could not determine true power or consumption.