Problem of a sound controlled alarm

Discussion in 'The Projects Forum' started by radetonator, Mar 14, 2012.

  1. radetonator

    Thread Starter Member

    Aug 1, 2011
    50
    0
    I am designing a sound controlled alarm circuit.

    The requirement: when you loudly clap your hands within 5 meters of the device, the device is to emit a 1kHz tone for 30s.

    the circuit:
    [​IMG]

    The problem is
    When I clapped my hands, the speaker could not work. The 555 monostable circuit cannot be triggered. The signal at pin 2 of U5 is shown below:
    [​IMG]


    but if I blown air to the mic, the device worked. The signal at pin 2 of U5 is shown below:
    [​IMG]


    What should be modified to make it work? Thank you.
     
    Last edited: Mar 14, 2012
  2. DickCappels

    Moderator

    Aug 21, 2008
    2,651
    632
    If its is a problem with sensitivity, then its probably because your comparator threshold is set to one volt, and it will take a 1 volt pulse out of the first LM741 in order to get the output of the comparator to change. The LM339 can handle inputs much closer to ground, so consider lowering the voltage on U1A pin 4.

    Also since the LM339's output has pretty good current sinking capabilities (and its the negative edge that triggers the '555, consider using U3A as an additional analog gain stage ahead of the comparator, since its probably not really adding anything as a buffer for the input of the '555.
     
  3. Audioguru

    New Member

    Dec 20, 2007
    9,411
    896
    The first opamp has a fairly low voltage gain of about 37. It is inverting with a low input resistance which reduces its gain. If it is non-inverting then its high input resistance will not load down the mic and its gain can be anything you want. Try a gain of 200.

    Many lousy old 741 opamps do not work from a supply as low as 9V. If the 9V is from a battery then it can drop to 6V. The 741 opamp was designed to use ONLY a 30V supply.

    Your schematic wrongly shows the output of the LM324 opamp connected to +9V.
    You have it always at ground since the input of the comparator is always at ground and the comparator and LM324 opamp are both non-inverting. The comparator or the LM324 should be an inverter so that the output of the LM324 is high then a sound causes it to go low and trigger the 555.
     
  4. radetonator

    Thread Starter Member

    Aug 1, 2011
    50
    0
    How should the first stage be changed? could you help me draw a circuit?
     
  5. Audioguru

    New Member

    Dec 20, 2007
    9,411
    896
    Here is a non-inverting opamp used as an electret mic preamp:
     
  6. radetonator

    Thread Starter Member

    Aug 1, 2011
    50
    0
    Thanks.

    How about this circuit? I want the frequencies below 2kHz, how should value of C13 be calculated?


    [​IMG]
     
    • 3.JPG
      3.JPG
      File size:
      108.8 KB
      Views:
      28
  7. Audioguru

    New Member

    Dec 20, 2007
    9,411
    896
    The alarm will be triggered by any kinds of sounds (dogs barking, etc) which have frequencies above the calculated cutoff frequency.
    The formula for the cutoff frequency is simple: 1 divided by (2 x pi x R x C).
    You have 4.7uF and 1k ohms so the frequency where the gain begins reducing (-3dB) is 34Hz. Frequencies above 34Hz will have the maximum gain of the circuit. The alarm will be triggered by nearly any sound.

    Your comparator has an input that is floating which is bad. It also does not have a reference voltage which is also bad.
     
  8. Audioguru

    New Member

    Dec 20, 2007
    9,411
    896
    Your mic preamp circuit is missing the 1k resistor and 47uf capacitor to ground (in my circuit) that filter out the battery voltage jumping up and down with the signal. So your circuit might become a low frequency oscillator.

    A hand clap is mostly high frequencies but many sounds are also high frequencies.
    Then your circuit will be triggered by many high frequency sounds, but maybe not the low frequencies of a dog barking.

    Your circuit does not need the LM324 opamp, the 33uF capacitor C4 and the R10 and R14 voltage divider. Because the output pin of the comparator and its pullup resistor can trigger pin 2 of the 555.
     
  9. Audioguru

    New Member

    Dec 20, 2007
    9,411
    896
    Your 9V battery does not have the 100uF bypass capacitor on my schematic so the +9V will jump up and down with the signal.

    Your circuit has transistor Q1 driving the speaker that is in series with a 1k resistor. Then Q1 is not needed because the speaker current is only 4mA. The 555 can drive a load with up to 200mA.
     
  10. radetonator

    Thread Starter Member

    Aug 1, 2011
    50
    0
    I just tested the circuit. If U4 pin8 is not connected to power supply, voltage of U5 pin3 can vary between low and high. But if U4 pin8 is connected to power supply, U5 pin3 always has low voltage. What is wrong?
     
  11. Audioguru

    New Member

    Dec 20, 2007
    9,411
    896
    This website does not show your schematic properly so I cannot see it.
     
  12. radetonator

    Thread Starter Member

    Aug 1, 2011
    50
    0
    U5 could work properly only when U4 was not connected to DC power supply. What is wrong?
     
  13. Audioguru

    New Member

    Dec 20, 2007
    9,411
    896
    The opamp in the bandpass filter is not biased at half the supply voltage so it is a rectifier.

    I simulated it (with its (+) input biased properly):
    1) It is extremely simple so it passes frequencies from 75Hz to 270Hz at its peak. Lower and higher frequencies are cut.
    2) It has a voltage loss at its peak of more than 200 times.
    3) Its resistor values are so low that most opamps cannot work with them.

    Why did you use this horrible bandpass filter anyway?
    The capacitors in the mic preamp already cut low frequencies and the lousy old LM324 opamp already cuts high frequencies (very low slew rate).
     
    Last edited: Mar 15, 2012
Loading...