Teensy Microphone Module

Davidelvig said:
1) Would it be better to just ship it back so you can poke around?
2) Send another beta my way (if you trust me).
3) I'd also like to try my hand at the process of ordering and assembling.
Click to expand...

I would like to get it back sometime to see if I can tell what failed. You should not have been able to kill it too easily and if so, I might need to add some defenses! My first suspicion is my home-brew soldering operation and a bad solder joint.

I definitely trust you and I'll definitely send you a couple Beta mics so you can continue testing. Your application would probably benefit from lower gain, so let me know if you want me to add the solder bridges to cut the gain 20dB. Otherwise, you can test how easy it is to change the gain with those small solder jumpers.

You are most certainly welcome to build some yourself! That IC is a real pain to solder without a stencil (I did three by hand!) but certainly it's possible with a pin-point soldering tip.

Thanks!
 
Yep - bad soldering job! I confirmed a wonky output and discovered a whole pile of solder balls between some pins. Heat it with the heat gun for a few seconds and it's back working again. FullSizeRender.jpg

They are kind of tough to see in the picture but between the last few pins on the left side, you can see some gray lumpy stuff. Those aren't suppose to be there!!
 
Please let me know if you start selling these mic boards. At the very least, I'd like to add links in the design tool documentation and hardware compatibility list on the website.
 
PaulStoffregen said:
Please let me know if you start selling these mic boards. At the very least, I'd like to add links in the design tool documentation and hardware compatibility list on the website.
Click to expand...

Paul,
Thanks! I received a couple of data points that it would be great to be able to disable the auto-gain-control system and a simple jumper will do just that. I'll be spinning the board "one-more-time" (famous last words) for that update.

I haven't decided if I'll sell the boards, primarily because I have no idea what that would involve and there is no way I can compete with the pricing from Adafruit or Sparkfun. I suppose I could make a smallish batch, price them to cover costs and see what happens. If you have any pointers please let me know.

Thanks!

Leon
 
Neutronned said:
I haven't decided if I'll sell the boards, primarily because I have no idea what that would involve and there is no way I can compete with the pricing from Adafruit or Sparkfun. I suppose I could make a smallish batch, price them to cover costs and see what happens. If you have any pointers please let me know.
Click to expand...

The big advantage your board has is it's already biased to 0.6V instead of 1/2 VCC like the other boards, so there's no need to do shenanigans like this to connect use it:

https://forum.pjrc.com/threads/4046...io-Lib-results?p=126317&viewfull=1#post126317

I know there's a Teensy section on Tinde

https://www.tindie.com/browse/teensy/

and some of the forum members here make some of these products. They might be able to tell you what potential sales numbers could be.
 
Mic AGC test.jpg
A quick test of the Auto Gain Control circuit. Top waveform has it enabled and the bottom has it defeated by shorting out a level set resistor. Based on popular feedback, I need to add a convenient way to shut that AGC off.

Code: 
// Teensy 3.2 to stream audio from analog channels to USB
//  Test microphones are connected to A2 and A3
//  Using Audacity to record data

//*********************************************************************************** INIT
#include <Audio.h>
#include <Wire.h>
#include <SPI.h>
#include <SD.h>
#include <SerialFlash.h>

AudioInputAnalogStereo   adcs1;          //xy=264,317
AudioOutputUSB           usb1;           //xy=496,314
AudioConnection          patchCord1(adcs1, 0, usb1, 0);
AudioConnection          patchCord2(adcs1, 1, usb1, 1);

//*********************************************************************************** SETUP
void setup() {
  AudioMemory(12);
}

//*********************************************************************************** LOOP
void loop() {
}
 
Last edited:
What is the component of FB1(ferrite)??and the purpose??
Also pcb layout has many drill holes instead of corner 4 holes. Why is that?
 
Wicky21 said:
What is the component of FB1(ferrite)??and the purpose??
Also pcb layout has many drill holes instead of corner 4 holes. Why is that?
Click to expand...

The ferrite along with C1 and C3 act as a fairly heavy filter for noise coming from the Teensy. Most of the holes are vias to connect the ground plane on the bottom to grounds on the top side to help reduce noise. I removed the mounting holes from the board because I was running out of room :)
 
Any specific component value for ferrite bead??
If i able to make the MICOUT circuit which includes (R4, R5, C4, C3, C1, ferrite) on a breadboard with adafruit max9814 module, is that possible to get the same output results ??? (Just for testing purpose)
 
Wicky21 said:
Any specific component value for ferrite bead??
If i able to make the MICOUT circuit which includes (R4, R5, C4, C3, C1, ferrite) on a breadboard with adafruit max9814 module, is that possible to get the same output results ??? (Just for testing purpose)
Click to expand...

Qty Parts Value Description DigiKey Manufacturer
3 C1, C2, C9 0.1uF CAPACITOR, American symbol 399-1170-1-ND Kemet
1 C3 100uF POLARIZED CAPACITOR, American symbol 478-8156-1-ND* AVX
1 C4 0.022uF CAPACITOR, American symbol 399-1163-1-ND Kemet
1 C5 2.2uF CAPACITOR, American symbol 399-3522-1-ND* Kemet
1 C8 0.47uF CAPACITOR, American symbol 399-8100-1-ND Kemet
1 FB1 FERRITE Ferrite 2.5K 1276-6377-1-ND* Samsung
1 GAIN JUMPER Normally open (Gain selection)
1 J1 3-Pin PIN HEADER
1 MIC MIC1-PTH Microphone 102-1721-ND* CUI
1 R1 150k RESISTOR, American symbol 311-150KCRCT-ND* Yageo
1 R2 100k RESISTOR, American symbol 311-100KCRCT-ND* Yageo
1 R3 2.2K RESISTOR, American symbol 311-2.21KCRCT-ND* Yageo
2 R4, R5 1.0K RESISTOR, American symbol 311-1.00KCRCT-ND* Yageo
1 U1 MAX9814TDFN14 MAX9814TDFN14 MAX9814ETD+TCT-ND* Maxim
1 PCB NeutronNed MAX9814 Rev 2.0 PCB Way

Sorry about the formatting! This is a cut&paste from the Excel BOM. As you can see above, the ferrite is available from Digikey P/N 1276-6377-1-ND.

Yes - with similar power supply filtering and that output divider, you should get pretty close to similar performance. If you just need one for testing, send me a private message with your address and I'll mail you one.
 
This mic has proved useful to me. I'll need more, and am happy to help get these into some kind of production (by buying some)

It's a mic designed just for the Teensy's 0.6 V adc inputs.
Seems to be very low noise, and with some useful amplification options.
As for improvements, Leon, I think I'd like an easier option to disable the max9814's auto-gain control.
Also, I think there's extra pcb space on the sides to make the mic narrower.

This would be a great Teensy add-on, and would have saved me time last year as I learned how to use the adc object.
Any others on the forum interested in jumpstarting a new mic?
 
Guys i'm just asking
I hope to use this micro phone board on my project which can capture frequency range from 300Hz - 2000Hz . If so sampling from teensy around 44.1KHz is useless right? According to nyquist theorem 4.5KHz is much more enough right ?? Just to avoid aliasing should put an anti-aliasing aliasing filter right. Can you guys please let me know what are the additional components needed to include on this Leon's board?
 
Wicky21 said:
Guys i'm just asking
I hope to use this micro phone board on my project which can capture frequency range from 300Hz - 2000Hz . If so sampling from teensy around 44.1KHz is useless right? According to nyquist theorem 4.5KHz is much more enough right ?? Just to avoid aliasing should put an anti-aliasing aliasing filter right. Can you guys please let me know what are the additional components needed to include on this Leon's board?
Click to expand...

Sampling at 44.1 kHz for signals between 300 Hz and 2 kHz is fine.
If you put a weak antialiasing filter (e.g. RC, or 20 dB/decade) in you signal path then noise frequencies above 20 kHz are suppressed by 20 dB.
If you plan to decimate 44.1 kHz audio by a factor of 10 then you need a good digital filter around 2 kHz.

Are you planning to use audioboard or bare Teensy ADC?
 
I'm going to make real time application. Due to that i'm hoping to use bare Teensy ADC. Somehow i'm going to make dsp biquad filter or FIR filter. Still im not sure which one to use. Also i have doubt about this anti-aliasing filter. If using dsp filtering no need to use Anti aliasing filters??? I read an article. They have mentioned if aliasing occurs by using dsp filtering it cannot recover.
 
With an appropriate mic, the Teensy 3.2 is
In my application I don't use any software filters (yet).
The result is a signal that is fine for pitch detection and it sounds pretty lifelike
What is your planned use of the signal?
 
Last edited:
I'm also going to use Teensy 3.2
In my case i'm doing a research level project to detect mosquito wing beat sounds. So simply i need to capture frequencies between 300Hz-3000Hz. i thought to use anti-aliasing filter before sampling and also dsp filtering. At the end
from those resultant samples going calculate FFT(need to identify fundamentals and harmonics).
This microphone board seems pretty well on my project. Instead of 102-1721-ND microphone on leons board i thought to use some noise cancelling microphone which has frequency range from 100Hz-12KHz.

Please somebody help me to design anti-aliasing filter(Low pass filter, which has cutoff frequency around 3.5kHz) on leons Microphone board. Simply some extra components needed add on MICIN path
 
Guys do you'll have any idea of much better microphone board? Which can capture mosquito wing beat tones?? Something can be use with teensy ? Any idea about what sort of microphones used inside mobile phones(Just curious)?
 
Wicky21 said:
Guys do you'll have any idea of much better microphone board? Which can capture mosquito wing beat tones?? Something can be use with teensy ? Any idea about what sort of microphones used inside mobile phones(Just curious)?
Click to expand...
what means better?
Have you thought how far you can hear a mosquito? I guess less then a meter. so the soundlevel is very very low.

How far do you want to detect the wing beat tones?
Once you figured out what the expected received level is you get some requirement on sensitivity AND allowed noise level of pre-amp filter and ADC.

concerning senitivity:
the mic you mentioned (CMA-4544PF) has a sensitivity of -44dBV (1 1kHz tone of 1 Pa (94 dB SPL) generates a voltage of 10^(-44/20) = 6.3 mVrms)
checking with Digikey I found one (POM-3535L) that says -35dBV, resulting to 17.8 mVrms for a 94dB SPL tone.
So the second one is more sensitive.

This is for analog mics. for digital mics the values are different but that is a different story
 
Last edited:
Thanks for replying WMXZ
Better means my requirement is to capture the low sound level signla which comes out from mosquito.

Yes less than 1-meter is enough. So POM-3535L is perfect on my case right?
I prefer Analog mic

And also after the microphone i thought to use Anti-aliasing filter (Low pass Butterworth filter which has order of 5) and found a IC which is much suitable for my case = MAX7414 . Cutoff frequency should be around 3000Hz. Im not quite sure which sort of filter should use Pasive,Active or IC based filter.

If i change the microphone on the Leons board is that enough? Do i need to change other biasing resistors ?
 
Leon's mic will be biased to 0.6v.
That's perfect for the Teensy Audio Library's adc input object
My guess is that it will get a reasonable signal from mosquito wings within a meter in a very quiet room.
I'd test it, but I can't capture the mosquitos without making them permanently quiet.
 
I would suggest not filtering the mosquito buzzing sound until you've capture sufficient audio samples to confirm that you have all the interesting frequencies. Don't trust your ears - sample fast and then analyze the frequency content. I'll bet the Teensy 3.2 is fast enough that you can run the ACD input into a software filter to either filter out or in the interesting signals.

Digikey's website is down right now so I can't look up pin-compatible replacements for the CUI electret mic element. From the MAX9814 datasheet: "An internal microphone bias voltage generator provides a 2V bias that is suitable for most electret condenser microphones." so there should be plenty of options.
 
You must log in or register to reply here.
Back
Top