Flow rate sensor is slow and not accurate at low flows

[hallboyone]

It could be replaced with something like this. However, this will require retuning everything, but this is probably okay since the flow control never worked great anyway.

[hallboyone]

I'm testing a new sensor (linked above) and have collected the following data.

The slope of the trend line is the conversion between the pulses and volume. The exact values to 6 decimal places are

slope     =  0.009472
intercept = -0.054317

At low flows, the measurements get pretty noisy, but it should be good enough. As an added benefit, the pulses are much more frequent (about 60Hz at 0.5ml/s) so the discrete derivative will be able to update much faster.