Test accuracy of algorithm with a single microphone
Ultrasonic Wind Sensor
Updates:
Documented the test procedure, readjusted the voltage thresholds to 70% of the maximum value
Implemented wave counting using voltage thresholding and hardware interrupts, for one and then two microphones.
Ran experiments involving counting wave periods and the wave-counting algorithm
Was able to play a 1kHz signal and have both microphones measure the time it took to count 1000 wave periods
Derived the original frequency based off of the number of periods (1000) and the elapsed time (f = periods/time)
Results were fairly accurate (mic 1 was generally within 1 Hz and mic2 was within 5 Hz)
Because mic2 is farther away, we think that some waves might have been “lost”, or didn't have enough energy to successfully trigger mic2 in order to be registered.
To Do:
Think of ways to improve the accuracy of mic2 (maybe use a louder source?)
Implement the algorithm to derive the propagation time between mic1 and mic2 (and thus the wind speed).
Create a Schmitt trigger using an op-amp and replace the comparator circuit