PPD42 sensor

[Dave Hylands]

I started a wiki page for the PPD42 here: https://github.com/mozilla-sensorweb/sensorweb-wiki/wiki/Shinyei-PPD42-Particle-Sensor

The sensor has 2 outputs. P1 is used to measure 1um particles and by default P2 is setup to measure 2.5um particles.

The T2 (threshold input) can be used to adjust the voltage (and hence the sensitivity) used for the P2 comparator. Connecting T2 to ground through a 10K resistor will make the voltage used for P2 be about 0.491v or allow it to measure particles about 0.5 um.

By default P1 has a threshold voltage of 1v and P2 has a threshold voltage of 2.5v.

The sensor is supposed to mounted in a vertical position (with the connector along the bottom edge and the metal shroud on the top side.

So, looking at the sensor output (from the datasheet), it seems that it outputs a pulse which has a width of 10 msec to 90msec. The time that the output is low is what is significant.

This is not something that linux is good at measuring, unless there is specific hardware support for it (for some reason I had heard that the sensor had a serial interface).

So I started to look at the various processors we're considering to see what hardware support they have.

The i.MX6 has a GPT (General Purpose Timer) which has an Input Capture capability. The GPT is essentially a 32 bit timer/counter and has 2 channels of input capture. Rising or falling edges can trigger a capture event and the value of the counter will be stashed into a capture register.

The ATSAMA5D3 appears to have six channels of Timer/Counter. It is a bit more sophisticated than the i.MX6, in that you can setup one channel to capture rising edges and a second channel to capture falling edges (and thus measure the difference).

The AM4378 (from TI) has 3 eCAP (Enhanced Capture) modules which can capture 4 event time stamps.

The devil will, of course, be in the details. This means making sure that we can MUX all of the various functions onto usable pins.

Do we know what other sensors we'll be connecting?