Current transformer range & sensitivity

[Jonesy]

No, the ADC does not take bipolar signals. And the problem is I have NOT ENOUGH resolution on the ADC (1024)

where I need at least 2000 resolution for 10A, 5mA resolution bipolar.

 

[sgugan]

then go for 12bit ADC

 

[FvM]

I don't see, that the measurement has been yet clearly specified. Do you intend to measure an averaged rectified or instantaneous current value? Only the latter would result in a bipolar signal, but must provide more overhead for the sine peak voltage or a heaven higher crest factor. Processing an instantaneous value may be reasonable e.g for power factor calculation or harmonic current measurement, but would also require your ADC and software to follow the waveform.

Basic AC current measurement would suggest an analog averaging rectifier or possibly true RMS detector circuit in front of the ADC.

 

[mtwieg]

12 bit and even 10 bit accuracy is likely infeasible due to the effect of magnetizing currents, unless you either have a huge core with lots of turns (for very large magnetizing inductance), or you apply some correction factor to null out the effect of magnetizing current on the measurement. The latter may be easy if you are working with a narrow bandwidth (like only measuring a 60/50Hz).

 

[Jonesy]

I must be able to measure instantaneous currents as I must plot the graphs out. yes, the line is 0-10A 60 Hz. Is this possible?

 

[mtwieg]

Has anyone ever tried loading a CT with a transimpedance amplifier? Seems like it would get rid of the issues with magnetizing current since its output voltage would be fixed at 0V...

 

[FvM]

Has anyone ever tried loading a CT with a transimpedance amplifier?

It's possible, I already suggested it in post #9. I've often used the method with wideband CTs. But unfortunately, the secondary winding of usual small CTs (e.g. from Talema or Coilcraft) has a considerable resistance and won't allow to cancel the magnetizing current completely. Ideally, a CT with two secondary windings gives the option to a apply an active compensation circuit.

---------- Post added at 21:03 ---------- Previous post was at 20:45 ----------

I must be able to measure instantaneous currents as I must plot the graphs out. yes, the line is 0-10A 60 Hz. Is this possible?

Yes, surely. Then the full peak-to-peak current has to be mapped to the 10 Bit ADC range, and the measurement is of course bipolar. Most digital oscilloscopes have maximum 10 bit resolution, in so far it should be suitable. Timing resolution is another point.

 

[Jonesy]

How do I bias the bipolar voltages, as my adc only accepts 0-5V?

 

[mtwieg]

The CT should be isolated, so it should just be a matter of using a resistor divider to make a bias voltage and use that as a virtual ground for the CT.

 

[Jonesy]

Can u show how the resistor divider must be connected to the C.T to enable the bias to work? Cause I have tried it and it doesn't couple/add to the AC signal.

Would normal voltmeter be able to measure this change? Should I set my voltmeter to AC or DC voltage measuring after the bias?