audio filter, does phase matter?

Discussion in 'General Electronics Chat' started by JasonL, Jul 11, 2014.

  1. JasonL

    Thread Starter Active Member

    Jul 1, 2011
    I have a circuit that filters audio for bass, mid, and treble.

    The filters apply different phase shifts. Does the phase shift matter? I plan to have an adjustable amplifier connected to each filter.

    I want to try averaging the three signals in the end, and output it to one speaker. My guess is, phase will matter in this case; the phase differences will cause bass and mid to subtract because they're 180 degrees apart. If that's a problem, is there a way I can get the signals to be in phase?

    Will phase be a problem if I chose to output the signals into three separate speakers?
  2. Ramussons

    Well-Known Member

    May 3, 2013
    1) Your schematic shows the 2 inputs of the Opamp interchanged.
    2) All filters (Low Pass, Band Pass, High Pass and Band Reject) introduce Phase Shifts across the cutoff frequencies. This is an inherent property of filters. And audio amplifiers continue to use these for Tone Control.

    JasonL likes this.
  3. crutschow


    Mar 14, 2008
    The phase shifts from the various filters occurs at different frequencies so they will general have little effect on each other when summed together. The one caveat is that at the frequencies outside their filter region the phases must be the same. Thus you cannot have a 180° phase-shift, such as from an inverting amplifier, in one of the filters if the others don't. In that case you will get some signal cancellation.
    JasonL likes this.
  4. JasonL

    Thread Starter Active Member

    Jul 1, 2011
    Thanks crutschow, I think I understand what I needed to do.
    Because there is little overlap between the magnitude response for the LPF and HPF, I don't think the phase difference matters much. The phase between LPF vs. BPF and HPF vs. BPF matters though because there is more overlap in the magnitude response.

    The phase between the LPF and BPF were 180 degree apart for low frequencies. And, the phase between HPF and BPF were about 180 degree apart for high frequencies. So, I inverted the output for the BPF to try to match the phases.

    I added the three signals together to see if there were any problems.

    simulation2 is without inverting the BPF.

    simulation3 is with inverting for the BPF.
    schematic3 shows this change.

    I think this should work for filtering the signals? I plan to add a controlled amplifier for each filter.