1-Wire Power Usage Monitor

This is a future project. I have not started yet but these are my initial thoughts ...

Mains Power Monitor

Initially, this started out as a project to enable my with the ability to sense whether mains power was present but my work to manually enter gas and meter readings into my Home Control System also led me to look at automating the monitoring of electricty usage and to get a more real-time view of power consumption over the day and to see what devices and loads are making the biggest impact on my electricity bill. There are devices available to monitor this but they are not easily interfaced into my Home Control System. The Electrisave  is one such device. A version due soon will also have a USB interface though. What I'm really after is a 1-wire networked device to do this though.

DS2423 1-Wire counter

Some of the later electricity meters have flashing lights with a pulse rate proportional to measured usage. Jon00  has built a 1-wire monitoring circuit using the DS2423. This is a 16-bit counter (0-65535) which cycles. The meter flashes once per 1/1000th of a KW. This means you need to sample quite often to check for counter loop around. I'm planning to use this technique and similar hardware but, I will store and increment another counter as part of my HCS, each time the counter loops to allow me to store absolute meter readings (for the duration that the HCS is running).

Clamp Meter

The device used to measure AC current without breaking into the wiring circuit is called a clamp meter . This basically wraps a two-part ferrite core around one conductor coming into the house (if you used both on an a.c. supply you would get a zero reading as the currents would cancel out). Around the other side of this ferrite core is a secondary winding, effectively creating a mini transformer. By measuring the current induced in this secondary winding, and taking account of the characteristics of the transformer and the frequency and waveform of the current being measured, the size of the current flowing through the conductor can be determined.

But it isn't that simple! If any of the loads in your house are inductive (and that is usually most of them), them the mathematical relationship becomes much more complex and a simple rectifier type circuit to provide an RMS  value will be inaccurate.

Do I care though? Surely any indication of power usage would be better than none.

Circuit Design

The first part of the circuit to measure the supply voltage uses a mains transformer to step down from 240Vac to a more usage voltage. You can't get a transformer that will give the required voltage level to feed into the DS2450, so the transformer output is rectified and then fed via a smoothing capacitor, into a high impedance resistor bridge. The smoothing capacitor, removes the rectification ripples but does slow down the response to voltage variations.

Maplin - Ferrite cores: http://www.maplin.co.uk/module.aspx?ITAG=SPEC&ModuleNo=32792&doy=11m9#spec

Gas Usage

I don't think tracking gas usage is going to be possible any time soon. I did think of putting a camera in the gas meter box and using image/character recognition software to read the meter. Technically it is possible and because the position and size of the numbers would be fixed, it should be quite simply to do accurate number recognition. I also have easy access to the meters for cabling but, I'm not sure what the gas board would say.

References