Hi, Appoligies to the moderators if this is in the wrong place, but 'High Power Controls' or 'Home Automation' appear to be the best bet for this subject.
Has anybody tried to make a power measuring system using a four quadrant analogue multiplier like the AD 633? I believe the DC output of one of these could be feed to the Arduino's analogue inputs. Using such a chip means that you are only digitizing 1 input instead of 2. By oversampling and decimation, rather than averaging, it would also be possible to improve on the 10bit accuracy of the native ADC.
I'm looking to make a system that diverts the output from my Solar Panels that I don't use during the day to batteries so that it can be used a night. (at the moment 60+% of production is sold to the grid and I buy in at least the same again at night)
I've seen several implementations of the 'Open Energy Monitor' that use the Arduino's analogue inputs, but the sampling rate of multiple volt and current waveforms takes tens of mains cycles and several seconds to come out with the power being transmitted on the mains side. It also strikes me that there is precious little processing time left to get the Arduino to then control what one might do with the numbers obtained.
The system is also prone to accumulated errors including:-
The problem of multiplying the instantaneous current value with the voltage value that by definition was worked out earlier.
This is a built in Power Factor error.
The DC biasing in the circuit means only a 5bit accuracy of current or voltage. this severely limits the range of the output given that the RMS value of the Voltage will not alter very much.
Theoretical error figures of over 15% are quoted in some places.
I don't think that the input side of the 'Open Energy Monitor' is up to simultaneously measuring the three 'power' figures that I'll be looking for. The AD633 has a probable error of 2% and some other analougue multipliers are eve n better. Has anybody seen and used anything better?