I will use a function generator module which outputs 0-1V sine-wave for frequencies of interest from 0.5Hz to 30kHz importance. So my input will be 0-1V sine wave but I want to obtain an AC sine wave with 1V amplitude and 0V offset i.e symmetric swing. I also want to achieve this not by using too complicated circuitry.

Theoretically below circuit works to eliminate such offset:

enter image description here

And here is the output 1V with symmetric swing:

enter image description here

I don't have experience with these but I have couple of questions:

1-) What component in practice should replace Voff for a better accuracy? Voff can be a voltage divider or a reference voltage; but what in practice is better for a reliable offset removal?

2-) The circuit above assumes that the incoming sine is always a 1Vpp with 0.5V offset. What if the Vin's offset varies, is there a solution for that? I mean is there a zero offset Vout solution for a possible varying offset of Vin?

I don't know what is the main issue in such circuits so these are the questions I need to figure out. Maybe some other things like power supply stability, opamp type ect. more important. It would be great to hear if some could share experience.

  • \$\begingroup\$ The simplest "DC-offset removal" is a capacitor. Try 3.3µF capacitor and 1Meg resistor (high-pass filter) at the input of a voltage follower. Or use a voltage divider instead Voff. Connect the voltage divider between Vcc and Vee R1 = 2.2k and R2 = 3.3k \$\endgroup\$ – G36 Nov 13 '18 at 17:19
  • \$\begingroup\$ Are you sure the cap will work down to 0.1Hz input? \$\endgroup\$ – panic attack Nov 13 '18 at 17:46
  • \$\begingroup\$ You say 0.5Hz in your question. For 0.1Hz you need a bigger cap \$\endgroup\$ – G36 Nov 13 '18 at 17:48
  • \$\begingroup\$ Im not sure the cap will work well for such low freq. \$\endgroup\$ – panic attack Nov 13 '18 at 17:50
  • \$\begingroup\$ AC coupling doesnt pass DC and I might need down to 0.5Hz or even lower freq like 0.1Hz accurate way \$\endgroup\$ – panic attack Nov 13 '18 at 17:51

Within the limits of the frequency band that you are interested in the "easiest" way to remove DC offset from an AC waveform would be to use AC coupling at the input of your amplifier using a suitable capacitor.

Another possible way is to sample the input AC waveform with a long time constant low pass filter filters enough to pass just the average value of the AC signal. You could then try to use that filtered voltage to subtract the offset from the original AC waveform.

Both of these approaches will self adjust for variation in input signal amplitude and offset. They are not a perfect solutions however because if the changes in amplitude or offset happen too fast they will not be tracked by the long time constants in the respective circuits.

  • \$\begingroup\$ I just want to point out that the two approaches are mathematically equivalent. If the pole is the same, both implement the exact same high-pass filter. \$\endgroup\$ – Edgar Brown Nov 13 '18 at 17:12
  • \$\begingroup\$ Agreed with the mathematical equivalency. \$\endgroup\$ – Michael Karas Nov 13 '18 at 17:21
  • \$\begingroup\$ Are you sure the cap will work down to 0.1Hz input? \$\endgroup\$ – panic attack Nov 13 '18 at 17:46
  • \$\begingroup\$ I need a symmetric sine Im not sure at 0.1Hz it would work. \$\endgroup\$ – panic attack Nov 13 '18 at 17:48
  • \$\begingroup\$ AC couupling no good for very low freq close to zero \$\endgroup\$ – panic attack Nov 13 '18 at 17:53

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.