I'm working on a project to build a rig for testing various sized Variable Frequency Drives (VSD's / inverters), the kind that you'd use to run a squirrel cage induction motor. The range would be from 10 to 140A (75kW max), 0 - 65Hz @ 415VAC. There will also be a variable mechanical load on the motor shaft.
The test I need to perform is to measure the current in each of the motor output phases, U, V & W through the VSD's speed range of 0 to 60Hz. At first this sounds easy, just measure the RMS current of each phase with a CT or hall effect current sensor. But the devil's in the detail.
The output of a VSD is not a pure 3 phase sinusoidal waveform but is a PWM square-wave that is smoothed out into a rough sinewave by the inductance of the motor windings. This means there's a lot of harmonics that spoil the readings of most RMS digital current meters. There are some meters that will now compensate for this using a built in low-pass filter. Link
Also the VSD output is variable voltage (Hence current) and variable frequency. AC current transformers become useless as the frequency approaches zero. A hall effect based sensor would probably be better here.
So, I'm looking for some kind of current sensor, probably something that has a built in low-pass filter for reasonably accurate (5%) readings and gives out a proportional signal, e.g. 0-10V / 0-20ma / 4-20ma. The lowest frequency I would expect it to work at - from my experience - would be about 10Hz.
Ideally, something like an industrial sensor version of the Fluke 87V DMM referenced in the link above.
If all else fails, then just a current shunt and analogue meter would suffice but ultimately I'm looking to capture the data into a PC for graphing.
Any advice would be appreciated... ;-)
Edit: Although this sounds like a shopping question - it probably is - it may end up as an electronics design question because I think there's nothing on the market that exists to solve this problem.