ADS1115 measuring AC voltage ?

I would like to measuring AC voltage from a small inductive coil.
The voltage is in range 0.5 mV to 25 mV.

What is the best way to achieve this ?

  1. amplify it to a more manageable voltage, applying an offset as needed to make sure the whole signal is positive.
  2. read it with a suitable (resolution, sampling rate and range) ADC.

Hello wvmarle.

So - ADS1115 will not fit the requirements for direct measuring ?
then please - what components I shall use in order to measure this range ?

Based on the very limited info I'd say you need some kind of instrument amplifier, bring that signal to a maximum of 3-4V.

After that it's a question of whether the fairly low sampling rate of the ADS1115 is good enough for you.

The ADS1115 can do 0.256volt full scale with a PGA setting of 16.
Could be just ok for that ~70mV peak/peak.

But as wvmarle pointed out, the ADS1115 is relatively slow.
So what are you trying to measure.

OP mentioned AC 0.5-25 mV. Don't know if that's the peaks or RMS or what... indeed 70mV may work but you still would need to offset it by some 30-50 mV.

We need more details on the signal!

I am just having a bigger coil which create induction from 50 Hz AC ( it's one machine for boots )
There is a smaller one which is like a tester for the work of the big one.
( the circuit shall remains with this parameters unfortunately for me )
So the field is more or less constant, no peaks I believe or so much above these 25 .. maybe 30 mV..
I just need to measure it more or less precisely..

Please define "more or less precisely". What resolution? What sampling rate? What's the allowed error? Defined in absolute numbers, not vague wordings.

I still have no idea what you're actually trying to measure here. If you're doing something like measuring the current of a 50Hz AC supply, the field is not constant, as it's a 50 Hz AC field. So the voltage is also not constant but following the same pattern.

This is what I am trying to measure - the AC Voltage which inducts into the second coil.
Under "more or less precise" I mean about 5-10% measurement error is acceptable.

Measurement error of what, exactly? Waveform, peak value, RMS value?

So your voltage will go negative as well? If so: apply a ~0.1V offset (2.5V reference diode + voltage divider), 3x gain OpAmp stage, and read it with the ADS at 8x gain (0-512 mV). Should give you a very decent signal. As you need just 5-10% accuracy that'd do just fine.

  • peak value is the best.

  • negative is great if possible.

  • I am not an expert into the field, and that's why I am looking for a specific circuit with OpAmp which is tested and proven that will work.

Do a quick Google search and you can find the required circuits. The testing and proving it works for your specific application you will have to do yourself. Mind you, dealing with such low voltages may prove a serious challenge.

Or if you don't care about the negative part, just connect it directly to the ADS1115's input, and set it to 16x gain. You may even be able to use the differential input in this case. Start sampling and see what you get.

Before I post here I did quite a lot of google searches.
My only point were to find an actually proven and tested circuit which fits my needs at the best.

Without EXACT specifications of your setup you won't get a very specific suggestion for that. You will have to show the actual waveform you get, for starters. The frequency and amplitude both have an effect on the circuit design. The impedance of the coil may also be of importance.

Then if you absolutely want a tested and proven circuit for YOUR exact setup, you have to find someone that happens to have the exact same setup and uses it in the same way you want to. That failing, hire an engineer and have him or her build and test the circuit for you using the exact same components you plan to use.

Sinusoidal, 50 Hz, amplitude - max 30 mV. impedance of the coil - 4 Ohm.

Some sort of microphone preamp could be needed.
Replace the electret capsule for the coil, and leave R3 out.
Parts values might need some tweaking, so a scope would be handy.

quite a good idea tbh. Love it, and going to try it. Thanks.