Hi there,
For the past few years I've been dreaming, researching and tinkering with the possibility of a high frequency microphone + digital audio sampler project.
I want to make something that is portable, 3.7V battery powered and smallish, that could record onto an SD card or something low power like that. I envision going out backpacking and coming back with hours of bird and mouse choruses, then slowing them down to a quarter or an eighth of their ordinary speed and playing them back. I'd also like to explore what kinds of ultrasonic noises are to be found in the city.
More and more I realize how big of a project this is, but I keep coming back to it! I've started to conceptualize it in two components, the sensor and the sampler.
The sensor, for now, is just gonna be a microphone, but it could be anything down the line. To me, any kind of signal could be interesting to sample and slow down. And in the future, who knows what kinds of crazy sensors will be available? Anyhow, for now I've found an inexpensive MEMS microphone that is reasonably flat up to 80 KHz (Knowles SPU0410LR5H-QB). Its sensitivity in that range is +/- 10 dB compared with its response at 1 KHz. For me, that seems good enough to make a frog/bat/bird/bug/alien recorder.
The sampler component would basically be an ADC with some amplifier circuitry, a microcontroller and some good memory (likely SD with a little RAM). Some of the newer Arduinos and similar devices are more than capable of dumping 200 Ksps onto an sd card at 16 bits/sample. I have successfully written at ~2 MB/s onto an SD card using a Maple microcontroller, but have read that with the Arduino platform and some newer libraries, you can do way better than that.
My goal is to have the sampler be more broadly usable. I've got artistic goals, but I don't see why a portable, high quality digital sampler wouldn't be perfectly useful for research purposes. What if some super wizard ultrasonic microphones come out in 10 years that are sensitive up to 500 KHz (does air even do that?! water maybe?) Well then I'd like to use it! What about a hydrophone? I believe dolphins can interact up to 150 KHz or so.
So with that in mind, I've been looking into some newer SAR ADCs. The LTC2378-20 looks really good to me. It's low power and has a good SNR even up into the high frequencies.
But what to do next?! How to design, let alone build a high frequency audio circuit that isn't noisy! I can program microcontrollers and built simple circuits, but dealing with crazy high frequency noise from misplaced ground loops, prototyping with surface mount components, oh my... yikes, where to begin?
Am I crazy? And even if I am, does anyone have any suggestions of good resources for embarking on such a project? A book, previous project, some common sense, tips, whatever? Maybe I need to team up with someone around my area that is an audio engineer.
For the past few years I've been dreaming, researching and tinkering with the possibility of a high frequency microphone + digital audio sampler project.
I want to make something that is portable, 3.7V battery powered and smallish, that could record onto an SD card or something low power like that. I envision going out backpacking and coming back with hours of bird and mouse choruses, then slowing them down to a quarter or an eighth of their ordinary speed and playing them back. I'd also like to explore what kinds of ultrasonic noises are to be found in the city.
More and more I realize how big of a project this is, but I keep coming back to it! I've started to conceptualize it in two components, the sensor and the sampler.
The sensor, for now, is just gonna be a microphone, but it could be anything down the line. To me, any kind of signal could be interesting to sample and slow down. And in the future, who knows what kinds of crazy sensors will be available? Anyhow, for now I've found an inexpensive MEMS microphone that is reasonably flat up to 80 KHz (Knowles SPU0410LR5H-QB). Its sensitivity in that range is +/- 10 dB compared with its response at 1 KHz. For me, that seems good enough to make a frog/bat/bird/bug/alien recorder.
The sampler component would basically be an ADC with some amplifier circuitry, a microcontroller and some good memory (likely SD with a little RAM). Some of the newer Arduinos and similar devices are more than capable of dumping 200 Ksps onto an sd card at 16 bits/sample. I have successfully written at ~2 MB/s onto an SD card using a Maple microcontroller, but have read that with the Arduino platform and some newer libraries, you can do way better than that.
My goal is to have the sampler be more broadly usable. I've got artistic goals, but I don't see why a portable, high quality digital sampler wouldn't be perfectly useful for research purposes. What if some super wizard ultrasonic microphones come out in 10 years that are sensitive up to 500 KHz (does air even do that?! water maybe?) Well then I'd like to use it! What about a hydrophone? I believe dolphins can interact up to 150 KHz or so.
So with that in mind, I've been looking into some newer SAR ADCs. The LTC2378-20 looks really good to me. It's low power and has a good SNR even up into the high frequencies.
But what to do next?! How to design, let alone build a high frequency audio circuit that isn't noisy! I can program microcontrollers and built simple circuits, but dealing with crazy high frequency noise from misplaced ground loops, prototyping with surface mount components, oh my... yikes, where to begin?
Am I crazy? And even if I am, does anyone have any suggestions of good resources for embarking on such a project? A book, previous project, some common sense, tips, whatever? Maybe I need to team up with someone around my area that is an audio engineer.