Mic Issues In Music Frequency Project

Discussion in 'The Projects Forum' started by cowhock35, Jul 27, 2014.

  1. cowhock35

    Thread Starter New Member

    Dec 25, 2012
    26
    1
    Apologies for the hand drawn diagram.

    I am trying to make a project that takes input from either an electret mic or line level input from an aux cable. The input is user selectable using a switch. Its basically a color organ, with 6 frequency bands.

    The input goes to a bandpass filter IC that takes an AC input and outputs DC voltage. The output voltage is used to trigger a MOSFET (BS170). The MOSFET is used as a switch to turn on LED strip lights based on when the frequency occurs in the music.

    It all works perfectly using the line level input. However, the mic circuit doesn't seem to work quite right. Some of the lights don't seem to shut off immediately when the music stops. Especially the low frequency (63Hz) channel, if I clap my hands it will blink at the sound, then stay on for half a second then faintly blink once more. When music is playing, the lights don't quite follow it, as if there is some kind of delay in them. If there is a fast baseline in the music, the 63Hz lights stay on constantly. The 200k pot is to adjust the gain and sensitivity of the circuit to accommodate for different volumes of music.

    I imagine it has to do with a capacitor in the mic circuit discharging and recharging and smoothing the output, but I can't figure out how to fix it.
    Perhaps its the poor quality of the ebay electret mics? The signal on the scope is much noisier than the signal from the line input, but that is expected.

    Any suggestions that might help me out would be much appreciated. The BA3830 is somewhat complex, and I don't think it is related to the problem, as the the line input works great with it.

    The mic circuit is here: http://www.electro-tech-online.com/attachments/electret-20mic-20-1-png.39643/
    The only variance is that mine runs on 12V and the Pot is 200k instead of 470k.

    The full circuit is attached.
     
    Last edited: Jul 27, 2014
  2. bertus

    Administrator

    Apr 5, 2008
    15,645
    2,344
  3. cowhock35

    Thread Starter New Member

    Dec 25, 2012
    26
    1
    Its a two terminal type with leads already connected. I simply put the negative to GND and the positive to the circuit
     
  4. bertus

    Administrator

    Apr 5, 2008
    15,645
    2,344
    Hello,

    How long are the leads of the microphone?
    It looks like there is a pickup of the mains.

    Bertus
     
  5. cowhock35

    Thread Starter New Member

    Dec 25, 2012
    26
    1
    about 15mm? Its all in a non solderable breadboard currently. What is a pickup of the mains?
     
  6. bertus

    Administrator

    Apr 5, 2008
    15,645
    2,344
    Hello,

    Pickup of mains, is the mains voltage in the naberhood of the circuit that is picked up by the high impedance input of the circuit.
    The frequency can be 60 Hz when you are in the USA or 50 Hz when you are in europe or japan.

    Bertus
     
  7. cowhock35

    Thread Starter New Member

    Dec 25, 2012
    26
    1
    It isn't just the 63Hz channel though, the slow and non ideal response is present on all channels. I was just using the 63Hz as an example because I think it is the most prevalent.
     
  8. bertus

    Administrator

    Apr 5, 2008
    15,645
    2,344
    Hello,

    I do not see a decoupling capacitor on pin 8 in your schematic.
    (see the datasheet)
    You also have noted the wrong pin numbering for the pins 10 to 18.

    Bertus
     
  9. cowhock35

    Thread Starter New Member

    Dec 25, 2012
    26
    1
    Yeah I realized I labeled the pins wrong I just forgot to change it. It is hooked up correctly I just got the numbers wrong when I copied the diagram to paper.

    I had a cap there at one point but I am trying to remove excess components and I removed it without any apparent change in functionality. I tried putting it back in and it didnt seem to help. What is its function in the circuit?

    Its hard to explain exactly what the problem with the mic activated lights are other than its not quite right. I just tried snapping my fingers and all the lights go on and then each frequency progressively turns off starting with low and with the high frequency staying on the longest, probably like half a second.
     
  10. bertus

    Administrator

    Apr 5, 2008
    15,645
    2,344
    Hello,

    You have the reset, pin 18 floating.
    How does the circuit react when you ground this pin?

    What are the input levels in Line and Mic selection?

    Bertus
     
    Last edited: Jul 27, 2014
  11. to3metalcan

    Member

    Jul 20, 2014
    228
    23
    Try lowering the gain on the mic channel...it sounds like you might just simply be overloading it! I suspect this would be quite obvious if you could HEAR the output. The behavior only seems strange because you're seeing a visual representation of it...sustain with a weird "tail" response is a common response from a heavily clipped audio stage.
     
    Last edited: Jul 27, 2014
  12. ronv

    AAC Fanatic!

    Nov 12, 2008
    3,282
    1,233
    A couple of things you can try.

    Add a 1K across the 47Ufd. Most electrets don't like over 10 volts driving them.
    the .35 Ufd can be removed the 10Ufd. at the input to the chip will take care of any DC.
    Try about 100pf from the output of the op amp to the negative input. If it is oscillating because of the breadboard this may stop it.
     
  13. cowhock35

    Thread Starter New Member

    Dec 25, 2012
    26
    1
    Grounding reset seems to have no effect.

    The line level is P to P 1-2 volts at max, and its usually run a little lower than than for better sensitivity.

    Coming from the mic, there is a high bias Voltage of like 5 V with about a 1V PP ac signal in it. The capacitors seem to drop that bias voltage down to about 2 volts and make the ac signal entirely negative.

    When I play around with removing and changing the values on the .33uf, 47k, and 10uf in series leading to the input, it seems to have significant changes on the response of the lights. Do you know what the purpose of those elements are? If I knew what they did perhaps I would understand more about why they affected the circuit.

    Playing with those elements for the line input results in a similar poor light response to the mic output.
     
  14. cowhock35

    Thread Starter New Member

    Dec 25, 2012
    26
    1
    If the gain is lowered, the lights won't flash, as the input to the filters will be too low. Looking at the signal on the scope, it doesn't have the appearance of being clipped, and audiogurus mic pre amp circuit which I use lists a 470k gain resistor and I only use a 200k, so I wouldn't think clipping to be an issue.
     
    to3metalcan likes this.
  15. bertus

    Administrator

    Apr 5, 2008
    15,645
    2,344
    Hello,

    I have taken the comments from the others and put them in your drawing:

    [​IMG]

    Bertus
     
  16. cowhock35

    Thread Starter New Member

    Dec 25, 2012
    26
    1
    I replaced the .33uf with a .14uf, it seemed to have a quicker response on the line level input. Without that cap, the 63Hz lights stay on too long, and don't flash to the base.

    The 100pf didn't seem to have an impact.

    The 1k across the 47uF seemed to be quite positive. It didn't totally fix the problem but it seems to have solved some of the randomness of the mic input lights.

    Thank you so much for all the help everyone!

    Now it leaves me with the accuracy of the mic. Here are two videos I just took, one of line level: https://www.youtube.com/watch?v=4IaqGljHa2o&list=UUA8xCgRBsq00a0M5esBVsHw

    And the other of Mic Level:https://www.youtube.com/watch?v=2tK-tRNEl40&list=UUA8xCgRBsq00a0M5esBVsHw

    As you can see the Mic level just isn't accurate. 3 or 4 channels all light up between 50 and 100hz. How would I go about fixing this problem?
     
  17. bertus

    Administrator

    Apr 5, 2008
    15,645
    2,344
    Hello,

    How is the circuit powered?
    Where does the 12 Volts come from?

    Do you use decoupling capacitors on the 7805?

    Bertus
     
  18. ronv

    AAC Fanatic!

    Nov 12, 2008
    3,282
    1,233
    Now it sounds like you are just picking up Huummm. The mic output is very small so it is easy to pick up hum from things like wiring in your house. So keep the blue wires short.
     
  19. cowhock35

    Thread Starter New Member

    Dec 25, 2012
    26
    1
    I'm using an older 5/12/15 V supply. University surplus. I am not using decoupling caps on the 7805.
     
  20. cowhock35

    Thread Starter New Member

    Dec 25, 2012
    26
    1
    The wires were all breadboard channels and resistors and caps, but I cut them all as short as possible, nothing more than a few mm sticking out of the breadboard. It may have helped a little bit but it doesn't look like much.
     
Loading...