I finished building several
tweet-a-watts power meters this weekend. I'm still working on the software to record the data, but here's a sample. I've turned off the y-axis units because I'm not 100% sure of my math. The relative differences should be correct. The x-axis is in UTC, so the numbers are 7 hours ahead.
Currently this is what each sensor is recording:
Sensor 1: The air conditioner.
Sensor 2: 2 linux computers.
Sensor 3: 1 windows computer
Sensor 4: 1 linux computer.
Sensor 5: Refrigerator.
I have one more sensor, but something is wrong with it and I haven't figured it out yet.