Saturday, December 25, 2010
power monitor
With the price of electricity rising quickly in Ontario, Santa thought it would be sensible to bring us on of these Black and Decker Power Monitors. After hooking it up and seeing that it worked with my meter, an Elster model, I began to be curious how it worked, and whether there would be a way to collect the same usage information as the power monitor and be able to feed it into more useful forms.
So to start with I began exploring exactly how this Black and Decker power monitor works. What this one does is quite simple really. The Elster meter, like may other models, has an infrared port that generates a pulse for every watt-hour of consumption. So all the power meter really does is detect and count the pulses.
I started with just a bog-standard IR photo-transistor biased with a 9V battery and looked for the pulses with my oscilloscope. It only took a few minutes to confirm that the 5 ms pulses were there. So the next step was to put together a proper circuit for detecting clean pulses.
I noticed that my phototransistor is not particularly selective to the IR range, and that bright sunlight (which also contains a fair slog of IR) was swamping the signal of the LED pulse. I made an infrared-pass optical filter using a piece of exposed and developed E-6 film. I understand unexposed C-41 (colour negative) film would work as well. I then have the whole thing attached to the front of the meter using HVAC putty — perma-gum is what they call it.
Next I built a circuit using an operational amplifier and a 7414 Schmidt trigger to amplify and clean up the signal from the phototransistor and provide a clean signal as input to an Arduino microcontroller. This is approximately the circuit I wound up with.
