1

Hey guys I am looking for a circuit where I can convert 100mV AC to 1V DC and then feed it to my microcontroller's ADC pin. Can any one suggest any circuit to achieve this?

I want to interface a split core current transformer (CT sensor) to my microcontroller and its rating is for 0-30A so it will give me an output of 0-1V AC. So usually I will get 100mV AC as my CT sensor output. I don't want a constant 1V DC. I just want to convert AC to DC so that I can give that as an input to the microcontroller and convert that DC voltage to some numerical values. For a microcontroller giving AC voltage to an ADC pin is not advisable right?

PeterJ
  • 17,131
  • 37
  • 56
  • 91
Vijay Anand Chandrasekaran
  • 80
  • 3
  • 13
  • Going to need more information, what is creating this 100mV signal, and what about it are you wanting to measure? Do you want a constant 1V DC or is your ADC supposed to be reading the peak of your AC signal, multiplied by 10? – Al Longley Mar 18 '15 at 11:13
  • Hello Longley I want to interface Split core current transformer(CT Sensor) to my microcontroller and its rating is for 0-30A it will give me an output of 0-1V AC. So usually i will get 100mV AC as my CT sensor output. I dont want a constant 1V DC. I just want to convert AC to DC so that i can give that as an input to microcontroller and convert that DC voltage to some numerical values. For Microcontroller giving AC voltage to ADC pin is not advisable right?? – Vijay Anand Chandrasekaran Mar 18 '15 at 11:19
  • Why you want to make an AC signal be DC if you will already use the ADC of your uC? It does not make sense... Explain better what you need and what's your project for better answers. Also, probably your uC have an AREF port which you may use in despite of amplify the signal. – Pedro Quadros Mar 18 '15 at 12:20
  • @VijayAnandChandrasekaran since you have only positive values on your signal it's ok to connect it to your ADC pin. Just remember to create a current loop through a resistor, since you not want current on your port, but voltage. – Pedro Quadros Mar 18 '15 at 12:24
  • @Pedro Quadros i do not have Vref in my uC. The CT sensor output is AC voltage. My concern is can i give direct AC voltage to uC since uC works on DC voltage. Anyhow i achieved the output by interfacing directly and its working fine. If any one wants to know how i did this please post your que in this tag only. I will elaborate the ans Thank you for your advice. – Vijay Anand Chandrasekaran May 23 '15 at 06:31

2 Answers2

5

We might need more information from you about what kind of signal this sensor is actually outputting. By 100mV AC, do you simply mean that the sensor is giving an analog signal output that has an amplitude of 100mV? Or, do you mean that your sensor is outputting an AC signal (like a sine wave with a fixed frequency) and the amplitude of this wave changes over time, depending on what is sensed?

If the case is the latter, as I suspect given the information you've provided, here's a solution that I'd go with:

First, AC couple your signal so that your 100mV swing will be centered around 0V (GND). This can be done easily by placing a capacitor in series with your output signal.

Now, your signal is centered around 0V and has a 50mV swing on each side for a total of a 100mV swing. You're going to want to amplify this signal to have a total 1V swing, so amplify the signal to get a 500mV swing on each side using an op-amp, bringing your signal to a range of -500mV to 500mV. Now, just add in a DC bias using a resistive divider or other means of 500mV, and your signal will now be your same AC signal from 0-1V.

Note: if you don't have a negative supply for your op-amp, you'll want to do the math and add a bias to your signal before amplifying it. This thread might help: Single supply op-amp audio amplifier

To get the DC level from this, you can low-pass filter your signal to smooth out the AC waveform and get a level that you can feed into your ADC. Rather than do the math, I prefer to mess around with the capacitor values (goes between the output and GND) and scope the output till I get a response that I like.

Good luck with your project!

Community
  • 1
justinrjy
  • 764
  • 4
  • 7
  • You did not mention the requirement of a diode at the output of the op-amp feeding the output capacitor (peak detector). It's also worth noting that this approach may result in undesirable output ripple depending on the input impedance of the sampling device and capacitor value chosen. If the input signal is will always be 100mVpp then this solution will work perfectly. If not, a parallel resistance with the output capacitor will be required in order to provide a discharge path. – sherrellbc Mar 18 '15 at 13:42
0

If you are interested in the RMS value I suggest looking for a AC to RMS converter IC. If you want to sample the signal you could offset one leg of the CT to 1/2 the rail and use the ADC to measure the other side. The CT is floating so by referencing it to half the rail you can m easure both plus and minus voltages.

AndrewF
  • 51
  • 2