Feedback elliminator

Discussion in 'The Projects Forum' started by Earl Jefthas, Oct 9, 2009.

  1. Earl Jefthas

    Thread Starter New Member

    Aug 23, 2009
    6
    0
    Hallo guys

    I posted a thread about my project a few weeks back. I need to build a feedback elliminator for a PA system. I settled on a final design and I just want to know if my idea is feasable or not, and if there is any suggestions to do some part of the design in a other way.

    1. The circuit will consist of a main microphone, low-pass filter, frequency to voltage converter(LM2907) and a voltage controlled amplifier(VCA).

    2. The output signal of the microphone will be split in parallel to the input of a pre-amp and frequency to voltage converter.

    3. The output signal of the microphone will go into a pre-amp, which output will than be inputted into a low-pass filter with a cut-off frequency of 15kHz. This output will than be amplified by the voltage control amplifier.

    4. The frequency to voltage converter will convert the frequency of the output signal of the microphone to a voltage that will control the gain of the VCA. When a high voltage is outputted by the frequency to voltage converter, the gain of the VCA will be low.

    - The way the feedback will be elliminated is to reduce the gain of the VCA when feedback occurs.

    For the VCA I wanted to use a LMH6505 ic from national semi-conductors but this component is not available in Cape Town. Can you please suggest a electronic device to use in the place of that?

    Thank you for helping me
     
  2. Earl Jefthas

    Thread Starter New Member

    Aug 23, 2009
    6
    0
    This should be a prototype for a subject that I have called Design Projects
     
  3. JDT

    Well-Known Member

    Feb 12, 2009
    658
    85
    Don't think this will work! The normal signal from the microphone will be a wide range of frequencies. The reason that the feedback is normally a whistle at a single frequency is that the frequency response of the microphone-amplifier-speaker-room is very "peaky" (even with very good quality equipment - mainly due to the room). So, at one particular frequency, the loop gain will be more than 1 and oscillation will start.

    Commercial "feedback eliminators" use a digital processor to shift the frequency of the signal a few Hz. Not enough to sound out of tune. But that when the speaker output is picked up by the microphone it is now at a slightly different frequency. Even they don't work that well!
     
  4. kubeek

    AAC Fanatic!

    Sep 20, 2005
    4,670
    804
    I can only agree with JDT, commercial digatal feedback eliminators will work better than anything you could do in analog. In the small box you have a DSP processor that can do a lot more complex way of filtering, so the analog circuit to do it the same will have to be really complex, expensive and big. And I don´t really think your circuit could reduce the feedback at all, or it would need a very carfeful, and probably frequency dependent setting of the trigger level.

    Maybe if this is a stable PA, you can make it work, but with a lot of fine-tuning to the conditions on the stage. Maybe it would be easier to make some changes in the mic - speaker placement, or place some baffles into the room or between the mic and the speaker.
     
  5. Audioguru

    New Member

    Dec 20, 2007
    9,411
    896
    Record the message then turn off the microphone and play back the message for no feedback.

    Talk into a microphone that is in a different building from the speakers. No feedback.

    Turn down the gain then "eat" the microphone so it can't hear the output from the speakers. No feedback.

    Turn off the speakers that are nearest the microphone. No feedback.

    Face the directional microphone away from the direction of directional speakers. No feedback.
     
Loading...