Use Sound-Activated Switches to Indicate Sound Direction

Thread Starter

matt787

Joined Oct 3, 2012
1
I'm looking for a simple and cheap way to indicate the basic direction of origin for a sound. The application would be something I can use in a forest or outdoor environment to determine the general direction of a loud sound (above background noise). I know advanced system compare the time the sound was picked up by each microphone to give a direction, but this is beyond my electronics capability and I'm going for cheap.

Instead I want to use 4 (or potentially 8) sound activated switch circuits placed in a cross. When the first switch to detect the sound is activated it will also trigger a transistor to shut off the indicator for the other circuits. So in essence the first MIC that hears the sound will light the indicator light and lock out the other circuits. The circuit I plan on using is this one: http://risecircuits.blogspot.com/2012/08/simple-sound-activated-switch.html

My question is do you think it will work? Am I overlooking anything? Is there an easier way to do it?

In my head (with my very minimal electronics knowledge) the speed electricity travels is much faster than the speed of sound, so the first mic to detect the sound should be able to create an open circuit on the indicator lights of the other circuits (preventing them from activating) before they detect the sound. And if I want more accuracy I can use more circuits.
 

wayneh

Joined Sep 9, 2010
17,496
... the speed electricity travels is much faster than the speed of sound...
That's not just in your head, and is quite true, by several orders of magnitude.
I can't comment on whether another approach might be better, but the logic of your approach is valid. The first sensor to hear a sound is the one closer to the source of the sound. There might be some odd cases where you could cause that to be untrue, but those are exceptions and not the rule.
Fast detectors should have no problem reacting quickly enough to beat out the sound waves racing to the next sensor.
 

THE_RB

Joined Feb 11, 2008
5,438
Sound travels at approx 1 foot per mS. So if you can put your 4 sensors about 10 feet apart in a square you will get anywhere between 10 and 14mS between the first sensor triggering and the last.

The big problem you will have is gain, and eliminating local noise/wind noise. A gushot in the distance will likely be quieter than a branch creaking 20 feet away. And if you want to detect something quiet like an animal call in the distance well good luck with that.
 

wayneh

Joined Sep 9, 2010
17,496
Yup, signal to noise is a bee-itch.

It's so easy for our brains but so very, very difficult for a machine.
 
Top