the Next Generation AC Ammeter

It is a commonly encountered problem that the technician engineer wants to monitor directly the AC current drawn by a mains powered appliance. There is a small assortment of traditional tools.  The classical AC ammeter is a moving iron instrument with the instrument coil either carrying the full load current or is driven from a shunt. It is rare to see this sort of instrument available second hand because the coil is invariably blown by excessive current.  The moving iron instrument is the gold standard, however there is not a simple relationship between the meter deflection and current magnitude. The meter scale is distinctly non linear and you are forced to have faith in the meter and its original calibration.  The moving iron instrument is not self calibrating in the sense that if you have one good data point, you can interpolate the rest.  This design  meets this condition.

The alternative is the now ubquitous hand held digital multimeter with a 10A shunt. I have little faith in the calibration of the cheap units, most of them could not sustain 10A continuously anyway.

I wanted a self powered, inherently linear and self calibrating AC Ammeter panel meter that was not a moving iron instrument because mainly I did not want to spend the considerable sum of money for a real instrument that would still be subject to destruction by a transient overload.

The alternative AC ammeter solutions involve the use of resistive shunts and measuring the sub volt ac signal with an AC millivolt meter.  This is an elegant by not straight forward.  The moving coil instrument is a DC instrument, better suited to measuring voltages of the order of ten volts.  It would need a rectifier that adds a high level of uncertainty into the reading when taking into accound the forward voltage drop of any real world diode bridge.  

I wanted my AC ammeter to have a full scale deflection of 10A.  If I used a 0.1ohm shunt, this would only drop one volt, and dissipate only 10W.  Thats a reasoneable amount and I certainly would not want to use more resistance than that.  A one ohm shunt would dissipate 100Watts at 10A and that is just silly.  

The 1 volt developed across a 0.1ohm shunt is not enough to drive a diode and moving coil voltmeter without a grossly non linear  meter scale.  I did not want to go to the trouble of supplying a separate DC supply and ac amplifier, thats already getting excessively complex.  How to step up the voltage to something more measurable ? Use a step up transformer across the shunt and use a bog standard diode bridge to drive a simple DC voltmeter that now has, at full load, maybe 30 volts to measure, which is trivial to do.  The step up transformer is nothing more than a plug pack transformer with the low voltage secondary used as the "primary".  This steps up the voltage nicely. Even better is the fact that the "voltage source" driving the transformer has a 0.1ohm driving impedance, so the load presented by the transformer and meter load is not going to influence the reading.  Even better is the fact that the meter circuit can be grounded, it is not floating at mains potential, and this opens up the possibility of safe remote measurement of mains current.


In practice ,  this circuit when tested with a radiator as dummy load and a large variac  showed totally  linear and expected readings from 100mA to full scale deflection at 10A.

The circuit is calibrated by monitoring the resistor shunt voltage with an AC millivolt meter.  The pot is tweaked to give the correct meter reading.  It is sensibly linear and does not exhibit the pathology that a current transformer and moving iron instruments show.  A BNC outlet permits direct monitoring of mains potential current with a scope. (usefully for debugging switch mode PSUs or other impulsive loads) A modern digital voltmeter with an AC range is sufficiently sensitive to calibrate this instrument.   The only other requirement for calibration is a known high wattage resistor or another moving iron ammeter that you have faith in.