Intelligent sound activated switch

[wesli-1]

Hi,

I have designed this sound activated switch, but I want to know if you think it will work before I go ahead and build it. There's probably a few little things I have overlooked as usual...

Ok, the idea is that it responds to hand claps, and will perform a different action depending on the number of claps recieved. At the moment it controls two devices, which can be switched on or off depending on the number of claps detected.

The sections are as follows:

Audio preamp stage: pretty straightforward, just an inverting amplifer with variable gain, decoupled input and output because it is not working with dual supplies.

Pulse shaping stage: uses a 555 timer (U2) to generate a 0.25s wide positive pulse at it's output (pin 3) whenever a clap is detected.

Missing pulse detector: uses a 555 timer (U3) to generate a positive edge at it's output (pin 3) when the first clap is detected, and a negative edge when a clap is not detected for more than 0.75 seconds.

Decade Counter: enabled whenever U3 is active. Advances its count on the falling edge of the pulses generated by U2. (it's done on the falling edge so that it won't be clocked until after U3 pulls the reset pin low).

Flip flops: clocked on the falling edge when U3 detects a missing pulse. C7 and R11 ensure that U4 cannot reset until the flip flops have read their inputs and changed state.


As far as I know, The 4017 and 4027 are clocked on the rising edge of the pulse, hence the use of the transistors as inverters to make the IC's clock on the fallilng edges as I require. I know I could use an inverter/buffer IC like the 4009 or 4049 instead of the transistors, but I will be building this on a piece of veroboard so using transistors makes the layout much less complicated.

What are the applications of this circuit?
1) controlling several lights in a room (duh)
2) controlling something which has several power settings, such as a dimmable light, a fan, motor, etc.
3) to make yourself look smart, and transform yourself into an uber-chick-magnet. (well probably not)

Thanks in advice for looking at this, and of course feel free to use it for your own diabolical plans to take over the world.

 

[audioguru]

Hi Wes,
You don't need an inverter for the clock of your 4017. Just tie the clock pin high and the counter will advance on the low-going edge of the clock enable-not pin. The clock enable-not input doesn't have the Schmitt trigger of the clock input but it doesn't matter with your fast logic edges.

 

[Roff]

You need to add a resistor in series with the base of each NPN transistor. The value is not critical - 10k would work.

 

[wesli-1]

Hi Wes,
You don't need an inverter for the clock of your 4017. Just tie the clock pin high and the counter will advance on the low-going edge of the clock enable-not pin. The clock enable-not input doesn't have the Schmitt trigger of the clock input but it doesn't matter with your fast logic edges.

Thanks, that's a really handy thing to know - I guess a few other IC's would have a similar input stage too.

You need to add a resistor in series with the base of each NPN transistor. The value is not critical - 10k would work.

Doh. Thanks for pointing that out and saving me from some smoke.


I was also wondering if anyone has seen a circuit like this before or if it appears to be a completely original idea.

Cheers,
Wes.

 

[nos_slived]

I was also wondering if anyone has seen a circuit like this before or if it appears to be a completely original idea.

Um.... the clapper.