choosing component for audio signal source selector circuit

Discussion in 'General Electronics Chat' started by T3STY, Nov 13, 2016.

  1. T3STY

    Thread Starter New Member

    Oct 4, 2016
    25
    0
    Hello!

    I'm trying to design an audio source signal selector, for 3 stereo inputs, and one stereo output. My idea looks like this:
    audio_switcher.png
    A micro controller will choose which signal source to switch on (never more than 1 at a time). But I need a device (in red in the pseudo-schematic above) to actually do the switching on command.
    It totally looks like a relay would do the trick, but they're really bulky for the purpose; I'm trying to keep this as small as possible, discrete SMD components are very much preferred.
    I thought about using transistors, but I'm not sure how to place them here - and I only know a bit of BJTs, I don't know if JFET or MOSFET would be more suitable for this.

    Can anyone help me with this?
    I'm looking forward to learning, so if you bring a brief explanation as well I would really appreciate it.
     
  2. AlbertHall

    AAC Fanatic!

    Jun 4, 2014
    8,134
    2,016
    The ADG509F will switch 4 stereo inputs.
     
  3. AnalogKid

    AAC Fanatic!

    Aug 1, 2013
    8,079
    2,285
    Analog Devices, Linear Technology, and Maxim make analog multiplexer chips specifically designed for audio, with better harmonic distortion performance and "pop" reducing circuits.

    ak
     
  4. T3STY

    Thread Starter New Member

    Oct 4, 2016
    25
    0
    Thank you for your suggestions, guys!

    I have found the MAX4932 to be pretty interesting and quite easy to use.
    I also wanted to ask if there should be any filters on the audio input/output pins? I often see a capacitor in series with the input, sometimes a high-pass filter, but I never understood what was their purpose and which benefits they bring.
     
  5. AnalogKid

    AAC Fanatic!

    Aug 1, 2013
    8,079
    2,285
    A capacitor in series with an audio input usually is a coupling capacitor to remove any DC component from the signal. You are correct that it could form a high-pass filter, which is why it is calculated, with the input impedance of the following circuit, to place that corner frequency far below the audio passband.

    ak
     
  6. T3STY

    Thread Starter New Member

    Oct 4, 2016
    25
    0
    So, since the audio band is from about 20Hz to 20kHz, and DC current is most likely a low frequency, I should calculate the capacitor as a high-pass filter with a 20Hz cut-off frequency. Am I right?

    The MAX4932 IC has a on-resistance of < 0.5 Ohm when powered at 3V to 5V.
    If my calculations are correct (assuming the 0.5Ohm max resistance), I could use a 15uF filter capacitor?
     
  7. crutschow

    Expert

    Mar 14, 2008
    23,091
    6,850
    That's not the impedance you use.
    For the series capacitor of a HP filter you use the resistance to ground of the load you are driving, not the series resistance of the switch.
     
  8. T3STY

    Thread Starter New Member

    Oct 4, 2016
    25
    0
    I know I'm going off-topic, but please, bare with me.

    I am not sure what the load resistance is going to be. The output of the MAX4932 will go to a 45k potentiometer for volume control. Can I consider this the load? after the potentiometer, it will eventually go to an audio amplifier, does this matter as well?
     
  9. crutschow

    Expert

    Mar 14, 2008
    23,091
    6,850
    Yes, that is the load along with whatever load it is driving (45K is an odd value :confused:).
    Not knowing the amps input impedance, but assuming it's a least 10k, use 5k as the impedance to calculate the capacitance value.
     
  10. T3STY

    Thread Starter New Member

    Oct 4, 2016
    25
    0
    The odd 45KOhm potentiometer is because I'm going for a digital one. I found many available, but an appealing one was the DS1882.
    The power amplifier will be a class D one, most likely, a TDA7492 (60k input impedance) or TPA3116 (9k to 60k input impedance, depending on the gain settings) -based amplifier.

    So, to the 45k potentiometer, I should add the 60k input impedance? if my calculations are correct, I get a 79pF capacitor (whereas 75pF is the closest commercially available value). Is this correct? Isn't this just too low to make any difference with or without the filter?
     
  11. crutschow

    Expert

    Mar 14, 2008
    23,091
    6,850
    No.
    If you look at the circuit you will see they are in parallel, with the equivalent resistance value depending on the pot wiper setting.
    The worst-case would be with the wiper at the top, where the resistance would be the two values in parallel.
     
  12. AlbertHall

    AAC Fanatic!

    Jun 4, 2014
    8,134
    2,016
    This is not correct. For 45k the capacitor is about 180nF.
    C=1/(2*PI*R*Fc)
     
  13. T3STY

    Thread Starter New Member

    Oct 4, 2016
    25
    0
    My calculations assumed 45k+60k. For ~100k I think it was correct.

    However, crutschow is right, the potentiometer and the impedance are in parallel. For the worst case scenario (pot wiper at top, 60k impedance only) I get 132pF.

    I think I can handle this from here. Thank you all for your help :)
     
  14. AlbertHall

    AAC Fanatic!

    Jun 4, 2014
    8,134
    2,016
    You definitely need to get a new calculator :(
     
  15. T3STY

    Thread Starter New Member

    Oct 4, 2016
    25
    0
    This is the calculation that I have performed:
    C = 1/(2 * 3.14 * 60000Ohm * 20Hz) = 1,32 * 10^-7
    10^-3=uF, 10^-6=nF, so 10^-7 must definitely be pF. What am I doing wrong?
     
  16. AlbertHall

    AAC Fanatic!

    Jun 4, 2014
    8,134
    2,016
    10^-3 is mF, 10^-6 is uF, 10^-9 is nF, 10^-12 is pF, so 10^-7 is 0.1uF or 100nF.
     
  17. crutschow

    Expert

    Mar 14, 2008
    23,091
    6,850
    Try again. ;)
    45k in parallel with 60k is 25.7k (the pot resistance is still there).
    The capacitor to give a 20Hz corner is then 309nF.
     
  18. T3STY

    Thread Starter New Member

    Oct 4, 2016
    25
    0
    I was never good at math. Thank you for the heads up!
     
Loading...