HOw to detect the flow of water through a pipe

I worked on this some today... If you apply 5V/ground to the red/black the other two wires (blue and yellow IIRC) will go high (5V) when the meter increments another full unit then falls low (0V) before the halfway point (need to test more to pinpoint this...manually blowing air into the fitting on my bench was getting old fast) to the next full unit and back to high at one full unit. It appears each instance of going high marks 1 unit.

Now to decide how to use this info :stuck_out_tongue: I am thinking of polling the unit every 30 seconds, recording the state (high or low) then incrementing a number if the state has changed. This number will then be sent to Cosm...

If anyone has a better/different solution I am certainly open to ideas!

For the record I wish this sensor (elster amco meter v-100 3/4) had better resolution =( It looks like it will be great for "I used this much water today/this week" but not so great for what I wanted which is more of a "when is water being used and at what rate". For example I just flushed a toilet and it went from a little under 14.2 to a little over 14.3 which is going to result in exactly nothing on whatever graphing solution I end up with. I am thinking an analog sensor that changes based on the rate water flows through it might be more my speed since I care more about flow/duration

Also for the record the male threads on the meter seem to be not quite right/not standard plumbing threads. This is according to two folks in different plumbing departments and me since one connection worked on the first try with plumbers dope and the other leaked with plumbers dope and still leaks (ever so slight drip once every few minutes) on the second try with another round of fittings and teflon tape. Disclaimer: I have never used plumbers dope before so I might not have used enough or let it cure long enough...