I'm trying to find a sound sensor that quantifies the sound level on some scale. I would like the sensor to be able to detect sounds as quiet as a whisper up to the sound of a loud motorcycle. Also, I would like the sensor to be directional so that I can determine the direction of the sound.
1) What products would you recommend for this application?
2) Also, I've seen "sound sensors" on the net that were microphones and others that were microphones incorporated into a board. Given that, I'm assuming that this application is more complex than just hooking up a microphone to an Arduino - I would guess that there's signal processing being performed. Are boards required for this type of application or would a simple microphone element work?
Thanks
Recommendations for Directional Sound Sensor
Moderators: adafruit_support_bill, adafruit
Please be positive and constructive with your questions and comments.
-
easternstargeek
- Posts: 347
- Joined: Mon Dec 13, 2010 1:39 pm
Re: Recommendations for Directional Sound Sensor
This is a very interesting problem- one that you will have to attack on a variety of fronts.
Start with a sound-to-voltage transducer- in other words, a microphone. If you want to experiment on the cheap, you can build a various types of microphonic transducers from electret-condenser BANNED. Good BANNED can be had for a couple of dollars.
To measure sound intensity, you will have to build a precision rectifier and peak-detector circuit if you want to use an analog input into your microprocesser. You can also go all-digital by sending the mic's signal to a DSP chip, but you may still need a pre-amp.
To determine direction, you must first define the space you need to resolve. The easiest would be using two mics to detect direction in a single quadrant of a single plane (An arc sweeping 90 degrees). If you want to resolve a full 360 degrees in a single plane, you will need a minimum of 3 mics. If you want to detect direction in all planes and all directions, you will need to build a tetrahedral array, where a mic capsule is placed at the center of each face of a regular tetrahedron.
Unfortunately, that was a gross simplification of the easy part of the task. There is still a great deal of work to do!
Here is a partial list of some things to consider, in no particular order:
1. The most important question is why do you want to do this? Are you just trying to see where "noise" is coming from, are you looking to demonstrate the principle of echo-location, etc? If you give us an idea of the end-goal, that will help narrow down and prescribe the right approach to what is a very complicated problem.
2. What frequencies are you interested in? Most acoustic environments are a rich soup of different frequencies and amplitudes thereof. If, for instance, you wanted to build an electronic "Bat" you could use a simple single-frequency sound source (the higher the better) something like a 20kHz+ ultrasonic transmitter to provide the sound you are trying to locate. That is a much simpler problem to solve than building a broad-spectrum "directional noise detector."
3. The gross acoustic environment is very important. Objects in the space will reflect, absorb, or refract sound waves to differing degrees depending on their frequency. Phase cancellation could become a problem if a primary sound wave and it's reflection from a nearby surface hit the microphone at the same time. A "Pressure-zone microphone" device will help mitigate that problem. All such phenomena greatly complicate the signal-processing.
4. Naked microphone elements by themselves will not likely do the job. What gives a finished microphone it's directional characteristics is determined by the housing, mounting methods, and sometimes more than one elemet employed with special circuitry. Read up on how Directional, Omnidirectional, Shotgun and Pressure-zone mics are built.
There is much more to learn before you can do this successfully, but you'll have a much better chance of a successful outcome (and much less work to do) if you narrow the scope of your goals.
There is a lot of very interesting work being done in spatial acoustics and acoustic environments. Since LadyAda and other cohorts are connected to MIT, which is a real hotbed for this kind of research, let's hope that a few of them show up to give you better information than I have.
Good luck to you- I'm looking forward to hearing about the results of your experiments!
Start with a sound-to-voltage transducer- in other words, a microphone. If you want to experiment on the cheap, you can build a various types of microphonic transducers from electret-condenser BANNED. Good BANNED can be had for a couple of dollars.
To measure sound intensity, you will have to build a precision rectifier and peak-detector circuit if you want to use an analog input into your microprocesser. You can also go all-digital by sending the mic's signal to a DSP chip, but you may still need a pre-amp.
To determine direction, you must first define the space you need to resolve. The easiest would be using two mics to detect direction in a single quadrant of a single plane (An arc sweeping 90 degrees). If you want to resolve a full 360 degrees in a single plane, you will need a minimum of 3 mics. If you want to detect direction in all planes and all directions, you will need to build a tetrahedral array, where a mic capsule is placed at the center of each face of a regular tetrahedron.
Unfortunately, that was a gross simplification of the easy part of the task. There is still a great deal of work to do!
Here is a partial list of some things to consider, in no particular order:
1. The most important question is why do you want to do this? Are you just trying to see where "noise" is coming from, are you looking to demonstrate the principle of echo-location, etc? If you give us an idea of the end-goal, that will help narrow down and prescribe the right approach to what is a very complicated problem.
2. What frequencies are you interested in? Most acoustic environments are a rich soup of different frequencies and amplitudes thereof. If, for instance, you wanted to build an electronic "Bat" you could use a simple single-frequency sound source (the higher the better) something like a 20kHz+ ultrasonic transmitter to provide the sound you are trying to locate. That is a much simpler problem to solve than building a broad-spectrum "directional noise detector."
3. The gross acoustic environment is very important. Objects in the space will reflect, absorb, or refract sound waves to differing degrees depending on their frequency. Phase cancellation could become a problem if a primary sound wave and it's reflection from a nearby surface hit the microphone at the same time. A "Pressure-zone microphone" device will help mitigate that problem. All such phenomena greatly complicate the signal-processing.
4. Naked microphone elements by themselves will not likely do the job. What gives a finished microphone it's directional characteristics is determined by the housing, mounting methods, and sometimes more than one elemet employed with special circuitry. Read up on how Directional, Omnidirectional, Shotgun and Pressure-zone mics are built.
There is much more to learn before you can do this successfully, but you'll have a much better chance of a successful outcome (and much less work to do) if you narrow the scope of your goals.
There is a lot of very interesting work being done in spatial acoustics and acoustic environments. Since LadyAda and other cohorts are connected to MIT, which is a real hotbed for this kind of research, let's hope that a few of them show up to give you better information than I have.
Good luck to you- I'm looking forward to hearing about the results of your experiments!
Please be positive and constructive with your questions and comments.