Test and calibrate decibel readings of fully assembled sensors

[tcsullivan]

During development we tested the accuracy of decibel measurements from the bare sensor PCB, which proved the sensor to be accurate within a couple of decibels (dBA). However, the decibel readings appear to shift slightly once the PCB is assembled into the enclosure (which includes a layer of foam to protect the microphone). I'm not sure if this is just a linear shift in amplitude, or if the frequency response of the microphone is changed by the case/foam.

Ideally, we can do some studio testing again to fully validate this. Anyone with a couple of the sensors (one with enclosure and one without) could also do basic testing for this: use a computer with a serial monitor like PuTTY or minicom installed to view the serial output of the sensor over USB; this will show the dBA of every measurement taken. The serial port should be configured for 115200 baud.

If we can identify the discrepancy and make a software fix, that would be ideal. Otherwise, we'll need to look into foam selection and the size and placement of the microphone port to make the sensor as accurate as possible.

Let me know if you need to borrow PCBs to do this testing. Rev 2 or rev 3 PCBs with the 1551SFL enclosure should be used for this.