Is Arduino able to handle a Microphone? What Microphone Sensor to choose?

Using Arduino Sensors
1

I'm about to start a project, in which I detect insect sounds in trees. Although, as I'm not exactly sure if Arduino is the right hardware to handle such thing. But it's not like I'm aware of "why". And I'm not even sure if the Microphone Sensors provided for Arduino can support me in such project or if I should buy other ones.

The sounds, as they're coming from tree scratching, they're probably very low frequencies. All Microphones as much as I know act similarly in low frequencies.
But because I want it to act nicely even when the distance between the Microphone and the source is big, I know that I want a Condenser Microphone, because of its sensitive sensor, which even if the source has high frequencies, it will be able to handle them. Also, that's comfortable since my source is definitely not loud, so Overload Avoidance it is.

But the source is not one, so the sound is multidirectional. It can come from all the different directions. So that means that I want a Random-Incidence Microphone.

Okay till here.
The thing is, that Condenser Microphones require a much bigger amount of Voltage (40V as I saw?). Which means an Arduino can't handle it, not with an 5 Voltage Output. I need like ... 8 Times this!!!

Now, regarding Arduino handling Microphone information, obviously as much as I know it can only handle <=5Vols, 0 to 1023. So the answer is no, an Arduino can't handle Microphone Information. So I'm like stuck, not knowing what to do. Oh man, I feel terrible. Why am I here again? I feel shame.

Basically, I need some guidance because I'm not sure where to start from :confused: An Arduino obviously can't do the job since I need such sensitive microphones.

2

ksapsy:
The sounds, as they're coming from tree scratching, they're probably very low frequencies.

You might want to verify the frequency components that make up the sound. Can you "characterize" the sound using an audio FFT module?

This will show you the frequencies you will need to monitor as well as the sound's bandwidth. From there, you'll have to do a Nyquist sanity check: Take the highest frequency component found in the sound's FFT characterization and multiply it by 2. This will be the minimum sampling frequency the Arduino will NEED to achieve without distorting the signal.

Assuming the sampling frequency isn't too fast for the Arduino, you can filter the input signal from the mic. This can be done either in analog electronics or in software (DSP - digital signal processing). Your filter(s) should pass only the frequencies present in the FFT characterization and attenuate all other frequencies. The Arduino can then determine if the output of all of the filters are high enough to guess that, in fact, the bugs are making the noise you're trying to detect.

ksapsy:
Okay till here.
The thing is, that Condenser Microphones require a much bigger amount of Voltage (40V as I saw?). Which means an Arduino can't handle it, not with an 5 Voltage Output. I need like ... 8 Times this!!!

You can use an external power supply to power the mic. You can also use the external power supply to power the Arduino (provided you use a voltage regulator to first step the 40V down to about 7V).

ksapsy:
Now, regarding Arduino handling Microphone information, obviously as much as I know it can only handle <=5Vols, 0 to 1023. So the answer is no, an Arduino can't handle Microphone Information. So I'm like stuck, not knowing what to do. Oh man, I feel terrible. Why am I here again? I feel shame.

There are all sorts of cool tricks to measure high voltage with Arduinos (or any other MCU/ADCs). The simplest is a voltage divider. This literally divides the voltage based on two resistor values. You can also use an opamp circuit to do the same thing (google non-inverting opamp amplifier).

Another thing you must realize is that the signal coming from your mic is like a barn door - it swings both ways (positive and negative voltages). Because of this, you need to first offset the voltage by its maximum peak value and only then can you do amplifying and filtering as mentioned above. This circuit will be more complicated, but very doable.

What is the maximum peak output of the mic you want to use?

3

The big issue is here: what do you really want your project to do?

OK, listen to insect sounds. And then. Record the actual sounds? Record there mere presence? This determines the kind of microphone you need, and the kind of other hardware around it.

As you saw yourself the first problem is in the sensor: what does it have to pick up. Arduinos are meant to read sensors and act on it. The ADC is also not suitable to record sounds, 10-bit resolution and at most 9615 samples per second. For digital sound recording you need something different.

4

ksapsy:
I'm about to start a project, in which I detect insect sounds in trees. Although, as I'm not exactly sure if Arduino is the right hardware to handle such thing. But it's not like I'm aware of "why". And I'm not even sure if the Microphone Sensors provided for Arduino can support me in such project or if I should buy other ones.

The sounds, as they're coming from tree scratching, they're probably very low frequencies. All Microphones as much as I know act similarly in low frequencies.

"propably very low" needs turning into numbers... No, microphones can differ a lot in their
low frequency handling, some roll off at >100Hz, some at 10Hz...

But because I want it to act nicely even when the distance between the Microphone and the source is big, I know that I want a Condenser Microphone, because of its sensitive sensor, which even if the source has high frequencies, it will be able to handle them. Also, that's comfortable since my source is definitely not loud, so Overload Avoidance it is.

But the source is not one, so the sound is multidirectional. It can come from all the different directions. So that means that I want a Random-Incidence Microphone.

long range and omni-directional response are at odds with each other - for longer range you need more directional response.

Okay till here.
The thing is, that Condenser Microphones require a much bigger amount of Voltage (40V as I saw?). Which means an Arduino can't handle it, not with an 5 Voltage Output. I need like ... 8 Times this!!!

Now, regarding Arduino handling Microphone information, obviously as much as I know it can only handle <=5Vols, 0 to 1023. So the answer is no, an Arduino can't handle Microphone Information. So I'm like stuck, not knowing what to do. Oh man, I feel terrible. Why am I here again? I feel shame.

Basically, I need some guidance because I'm not sure where to start from :confused: An Arduino obviously can't do the job since I need such sensitive microphones.

You need a condensor microphone preamp with phantom power, that's a given. You probably want
one of the modern ones that's a USB device.

How loud are these noises? The rustling of leaves is going to be a big source of noise to you I fear.
Picking out a scratching sound from a background of rustling without directional information sounds
hopeless to me.

BTW quality condensor mikes may be unsuitable for long term use in an outdoor environment,
they are designed for indoor humidity regimes. There may be specialist manufacturers for outdoor
microphones.

5

@Power_Broker I've got to give you some credit, your information and suggestions were very useful!

@wvmarle No recording included. At least not on the Finished Product. The finished product is just to warn you if there's an insect in the tree or not.

@MarkT

MarkT:
"propably very low" needs turning into numbers... No, microphones can differ a lot in their
low frequency handling, some roll off at >100Hz, some at 10Hz...long range and omni-directional response are at odds with each other - for longer range you need more directional response.

So ... it can't be LongRange and Omnidirectional you say? Goood (God with o's) ... that's annoying. Because trees are long you know, it's not like I want to check up only 10cm of the tree, I want all of it.

MarkT:
How loud are these noises? The rustling of leaves is going to be a big source of noise to you I fear.
Picking out a scratching sound from a background of rustling without directional information sounds
hopeless to me.

Well, I was hoping that after I place it IN the tree, like making a small hole for the microphone (which hopefully will gonna be regenerated by the poor tree :confused: ), the microphone wouldn't be disturbed by the environment noises, but only by the Inner Tree's soundwaves. But you know, it's better said than done. I still haven't tried anything because I'm at the beginning.

And the beginning is where you encounter most problems.

6

When placing it IN the tree, the whole picture changes from the microphone pov. Issues like directionality and distance are pretty much gone. You now listen to the tree directly; almost like a doctor's stethoscope.

My approach to this problem would be to first find trees that have the insect infestations.

Find a microphone that can be mounted in/against the tree, and see whether you can hear the munching of the insects using a normal amplifier. Then at least you know the microphone works, and whether you have to deal with ambient noise, and what this could be. A tree may act as soundboard, allowing you to hear insects from a distance, but maybe also branches rubbing against one another. It will also give you an indication of the range sounds travel inside the tree.

Next, find a way to detect those sounds. If there's little or no ambient noise, this should be pretty straightforward (take a recording of the first experiment to play back to your Arduino, just to make life a little easier). If there's a lot of sound filtering to do, it's going to be a much harder problem and one that an Arduino likely can't handle. You may have to look at a Raspberry Pi or even more powerful system.

After that it's field testing. Find trees that are infested, and check that your alarms go off. Find trees that are not infested, and see that your alarms stay silent.