I've made a wireless microphone. So let's make a wired one. As one of the wiser viewers has pointed out:
"Not everything needs to be wireless". There will be a small test at the end of this
video to see if you've been paying attention, so stick around to the end. There are three microphones in my collection
of components that I'd like to try out. I've got two analogue microphones, an Electret
microphone and a dynamic microphone. And I've got an ICS-43434. This is a digital I2S MEMS microphone. If you haven't got one of these then the INMP441
is equally suitable and very affordable. How do these three microphones work? An electret microphone is essentially a parallel
plate capacitor with the distance between the capacitor plates varying depending on
the sound wave hitting the diaphragm. They have a built-in Field Effect Transistor
that amplifies the signal. In order to work the Electret microphone needs
to have a bias voltage of 1 to 2 volts and this is typically supplied by the microphone
pre-amp. A MEMS microphone is a Micro-Electro-Mechanical
System. It has a pressure-sensitive diaphragm etched
into a silicon wafer. The principle of operation is actually quite
similar to Electret microphones where the capacitance varies based on the sound waves
hitting the diaphragm. One of the nice things and MEMS microphones
is that all the electronics are integrated into the package so you end up with a nice
digital interface. They are also very compact so you'll find
them in a lot of portable electronics. Dynamic microphones are much simpler in operation,
they just consist of a coil of wire moving in a strong magnetic field. The movement of the coil is caused by sound
waves hitting a membrane and this movement causes a voltage to be induced in the coil
due to the magnetic field. This very small voltage needs to be amplified
by a microphone preamplifier to make it audible. How are we going to do the USB interface? You'll find out after this quick word about
the channel sponsor. PCBWay have been supporting the channel for
a while, they offer PCB Production, CNC and 3D Printing, PCB Assembly and much much more. I've made quite a few boards with them and
they are great to work with. You can find a link to their details in the
description. So, how are we going to do the USB interface? This is actually surprisingly easy - we're
going to use a Teensy. A Teensy can pretend to be many USB devices,
including a USB audio device. What's really cool with the Teensy is that
you can design your Audio project in a graphical environment so it's pretty easy to get some
simple code up and running. Here I'm connecting the I2S input through
an amplifier and out to the USB interface. This generates some very simple code that
we can copy and paste it into our sketch. We just need to set the amplifier gain in
our setup function. If you wanted to get fancy then you could
hook the value of this gain up to an analogue input and have a volume control on your microphone. There are also a whole bunch of entertaining
effects that you can add - all of these run in real-time on the device so you could easily
build yourself a real-time voice manipulator. If we plug a Teensy running the simple sketch
into the computer it shows up as an audio device and we can record audio from it. That's pretty amazing for just a few lines
of code. How do we get all our microphones hooked up? The Teensy can only act as a single USB device
and only supports two channels, so I've got two Teensys connected to support my three
microphones. Hooking up the I2S MEMS microphone is trivial. The really nice thing about this kind of microphone
is that you can just connect it directly to the I2S interface of your microcontroller. To make sure it's got a nice clean power supply
I'm using a battery and a low drop out regulator to feed it 3.3volts. This is probably completely unnecessary but
we've seen when using the ESP32 how easy it is to get power supply noise interfering with
the microphone. For the two analogue microphones there's a
bit more work to do. The electret microphone requires a small bias
voltage to work and I've got a nice little MAX9814 breakout board that can supply this. This board also has a handy built in 5v linear
regulator so once again we can power it from a nice clean battery - I'll also tap off from
this 5v supply for some of the other boards we need. The MAX9814 is a very nice little chip and
include an automatic gain control so it will be interesting to see how it performs. The dynamic microphone requires a very high
gain pre-amplifier as the signal coming from the moving coil is very low. I've got a small breakout board based around
the AD828 op-amp that has been configured as a microphone pre-amp. This requires a 5v power supply so I'm taking
a line off the MAX9814 board to power it. Both of these microphone pre-amplifiers will
give us an analogue signal so we need to pass them both through an analogue to digital
converter. Fortunately, I have another breakout board
that contains a PCM1808 which is stereo Analogue to Digital Converter that outputs an I2S signal. This requires both a 5v supply and a 3.3v
volt supply for the 5v supply we'll use the clean power coming from the linear regulator
and battery. For the 3.3v supply, we can just use the power
from the Teensy as this is the supply for the digital part of the IC. Now that our signal is I2S we can just connect
it straight to the Teensy and use it as a stereo microphone. With everything wired up and plugged in I've
mapped all these inputs along with my shop bought USB mic through to one single audio
device so I can record them simulataneously. I've normalised the four different samples
to remove any difference in volume. This is my shop-bought USB microphone. And this is the ICS43434. And here's the dynamic microphone. And finally here's the electret microphone. To my hearing, both the analogue microphones
seem slightly flatter than the shop-bought USB microphone. The MEMS microphone doesn't seem to have as
much base as the shop-bought microphone, but it sounds pretty good. If we look at the spectrograms of each signal
we can see some differences. Both the analogue microphones seem to be missing
some of the higher frequencies. This makes me suspect that the A2D converter
may not be giving us the full range. The MEMS microphone seems to go to the full
20KHz range, but also seems to have picked up some high frequency noise. All this has made me wonder what's actually
in my shop-bought microphone. So I've taken it apart and see what makes
it tick. Looking at the microphone it's an electret
microphone, we can confirm this by checking to see if it's magnetic and also by checking
the bias voltage which is pretty low. The electret microphone capsule is bigger
than my electret microphone - it's 1.5cm vs 1cm - this means that my shop-bought microphone
is going to be more sensitive than my home made version. In terms of electronics, everything is being
done by this one chip, a RealTek ALC4042. It's quite hard to find very much information
on this chip, but it does seem to be used in a lot of products. Having looked inside the microphone, it's
quite tempting to attempt some kind of upgrade to it, but that's probably for another video. I promised you a test - so here it is. The audio for this video was actually recorded
using my new microphones. The introduction and first section was recorded
using the MEMS microphone, the third and forth sections were recorded using the electret
microphone and the fifth and sixth sections were recorded using the dynamic microphone. This final section is being recorded using
my shop-bought USB microphone. Could you tell any difference in the audio? Let me know in the comments. I'll see you in the next video!
