Watthour meter

[windy]

 I have a GE 100 amp watthour meter that was used for metering electricity into our house. I'm thinking of using it to moniter the watts being generated by my 3hp 3 phase conversion. I hooked it up to one phase of the generator and it would slowly rotate with the voltage at 40 volts and drawing 1 amp per phase. It is a 240 volt,single phase,60hz meter. I will be using the generator for hot water heating using three elements,so if it works I would take the reading times three because I'm only hooking the meter between one phase.

 My question is, would the meter be accurate at any given voltage and frequency as these constantly change with wind speed. Or,for these meters to be accurate ,do they need a constant 240 volt at 60 cycles.

 windy

[Ungrounded Lightning Rod]

They're accurate over a wide range of voltages and currents.

 - Magnetic coils wound on cores produce a moving field that produces a moving drag on the disk that's proportional to the product of the current and voltage.

 - A permanent magnet produces a drag that is proportional to speed.

If these were the only forces on it the disk would spin at a speed proportional to the power used - and a direction indicating which way the power is going.  But the bearing produces a friction that makes the meter read low at very low power and stop if the power use is too low to overcome the friction.

The resistance of the disk changes with temperature, but this affects drag from the moving and static fields equally, so it reads accurately over a wide range of temperatures.

I'm not sure if they lose accuracy with frequency changes.  The moving field would run at a different speed (which shouldn't matter, since it's moving vastly faster than the disk) and its penetration of the disk and strength of its drag might be affected (which WOULD matter).

[Flux]

You will loose accuracy with the low voltage and low frequency.

It may still give you useful results, but working at less than 70% volts is not good and I suspect the frequency issue may be much more serious.

There are various tricks used to compensate pivot friction and other factors and these may be fairly frequency dependent. I have never had reason to check one off frequency.

Flux

[Ungrounded Lightning Rod]

The only point where either voltage or current affects the accuracy (of an electromechanical type) is where they get so high they fry the meter or saturate the cores, the disk is moving fast enough that it's at an appreciable fraction of the speed of the moving field, or the total power is so low that friction dominates.  The meter doesn't care whether a low current*voltage product comes from low current, low voltage, or both.

But I expect frequency to be a major issue.

Assume, for the moment, that it's low enough that the fields penetrate the disk nicely.  Also assume that the moving field from the coils is the same strength as the damping field from the magnet.  The disk should stabilize at the speed where they produce equal but opposite drag - which means at half the speed of the moving field.  But the higher the frequency, the faster the field moves.  So this speed is directly proportional to frequency.

So the meter's reading should be the product of the frequency and the power.  OOPS!

This means at 30 hz it will read half the power that flows through it, at 15 hz a quarter of it, and so on.