enter image description here


I got a simple problem. My objective is to measure voltage and power factor of a 16V AC through a differential ADC (ADS1115) that measures voltage difference between 2 channels. I only managed to measure current accurately through a CT.

In the positive cycle of the the circuit shown, the voltage across the 1k resistor is 1.45Vp which is measurable. In the negative cycle, the voltage across the resistor will be -14.54V which will burn the ADC.

Hence, how do I modify this circuit to achieve a safe voltage measurement? I will need to detect the waveform at point 0V in order to compare it to the current then measure power factor. I thought of connecting A2 to the ground of the DC supply instead of the nuetral, but I was not sure if that's the right approach.

Sorry i'm a beginner. Any input/ advice will be highly appreciated.

  • 1
    \$\begingroup\$ Do you need to digitize the AC waveform or are you just interested in measuring the peak-to-peak voltage? \$\endgroup\$ – kjgregory Nov 9 '17 at 15:57
  • \$\begingroup\$ Is your ADC ground the same as the "DC Ground" in your schematic? \$\endgroup\$ – The Photon Nov 9 '17 at 17:32
  • \$\begingroup\$ @kjgregory, Just interested in measuring the peak to peak voltage as well as the 0 crossing really for power factor. \$\endgroup\$ – Moha Sawyer Nov 10 '17 at 2:34
  • \$\begingroup\$ @ThePhoton, Yes indeed it is. The bridge rectifier is also supplying the ADC. \$\endgroup\$ – Moha Sawyer Nov 10 '17 at 2:36

You can capacitively couple and re-bias the AC voltages into the range of your ADC:


simulate this circuit – Schematic created using CircuitLab

VCC is the power supply voltage of your ADC. I've chosen the resistors assuming this is 3.3 V, and that the AC signals are 16 V peak-peak (based on your saying the A1 and A2 nodes in your circuit go to -14.5 V). If I've misinterpreted, you can adjust the resistor values to change the scaling from the AC input to the ADC input.


Use an AD4950, AD8138 or similar differential amplifier. These parts will accept a common mode value from your ADC or a vref circuit and will output a differential waveform centered around that voltage reference. The gain can be adjusted in order to keep the maximum Vpp output (you probably need about 1/4 to 1/8 gain or so). You may need to produce a negative DC supply rail in order to bias the AD8138 into its linear region.

Alternatively, use a second transformer to step down the AC voltage to within the ADC's range. You'll want one with three windings so you can set the common mode of the output with a vref circuit.


Your Answer

By clicking “Post Your Answer”, you agree to our terms of service, privacy policy and cookie policy

Not the answer you're looking for? Browse other questions tagged or ask your own question.