optical disdrometer project feasibility

[bees]

Was hoping to get some opinions on a project I'd like to do:

I'm working on building an optical disdrometer, a device that gives information about rain. The hardware amounts to some LEDs, two photo transistors, and opamps. My plan was to use the two ADC channels to sample the opamp voltages at 10KHz, do signal processing (some filtering, linear regression, cross correlation) on the teensy, and finally relay the information back to a computer via serial for more processing and display.

At a high level, does this sound reasonable?

Not sure if this is the right place to ask this, but does this ADC library work with teensy 3.2? Or would I be better off just implementing the features I need myself?

 

[GremlinWrangler]

The 10khz sample rate will be near the limit for a Teensy, but should be do-able. Unsure how many samples you will be able to take for filtering at that rate.

The ADC library should work normally with the 3.2 since the changes were to board layout and bootloader not the IC.

 

[bees]

Awesome! The sampling rate was kind of arbitrary (saw it in a paper somewhere), should be fine to lower it if the Teensy is struggling. Will share my progress with you all.

 

[PaulStoffregen]

Fast sample rates usually require interrupts or DMA. It's doable, but more work at the hardware level.

The audio library uses the ADC at 44100 Hz, and much faster speeds are possible. But to sustain low-jitter fast sampling, you've got to use the hardware to trigger and move the data into memory.

 

[bees]

Would it be sufficient to have the PDB trigger the ADC then have the PDB isr read the ADC result into a buffer? I'm unfamiliar with using DMA, so the isr/update functions here are beyond my grasp right now. Or would I be better off reading the reference manual and using DMA? It sounds like I should do the latter.