Thoughts on Water flow sensors

[Wade]

I know you said you don't want a project. but with what I think was simple programing, I setup an Arduino (NodeMCU) to read pulses from a simple flowmeter. It counts them and then using the great arduino plugin from Enigmatheater I send that to HS. I do the toalizing at the arduino and send a command to reset the total every monday morning. I have virtual devices in HS that store the most recent 6 week total and rely on the Device History plugin for overall trending. I have the NodeMCU programed to set a "flow" flag as long as there are pulses (i.e. flow)

Searider I'm planning something similar and it will be my first foray into Arduino territory. Would you be willing to post a how-to on your metering setup?

[AllHailJ]

I have some engineering questions on their claims. I looked at all the devices listed and only the Phyn provided the accuracy specs. Their flow meter can detect .1 gpm. It is basically a doppler meter. This accuracy converts to 7571 drops per minute. I'm dubious that they can detect a small drip in the wall. Even if the others have twice the accuracy (I doubt it) then you are left with the smallest leak that can be detected is about 3500 drops per minute. That's a big leak.

I have leak detectors in all wet areas and the fibaro and homeseer hockey pucks have caught two leaks. One in the kitchen when the garbage disposal backed off and a spill in the laundry room.
There have been multiple nuisance trips but the turning the water back on is just the press of a button. I love the shutoff actuators as it is very convenient to turn off the water if you will be gone for a while. I think the insurance company is in corporate mode trying to improve it's bottom line by reducing risk to them.

[mminehan]

I have some engineering questions on their claims. I looked at all the devices listed and only the Phyn provided the accuracy specs. Their flow meter can detect .1 gpm. It is basically a doppler meter. This accuracy converts to 7571 drops per minute. I'm dubious that they can detect a small drip in the wall. Even if the others have twice the accuracy (I doubt it) then you are left with the smallest leak that can be detected is about 3500 drops per minute. That's a big leak.

I too am a little dubious. I can't seem to find many user reviews for these products. A couple for the Phyn Plus seem to suggest that once the 'learning' phase is done they seem pretty good at working out where the water is being used. Two reviews on BestBuy USA say the smaller Phyn SWA throws a lot of false alarms. But no end user reports of accurate leak deteaction. Most of the formal reviews are just rehashes of the marketing information.

They are not cheap. So I think it is case of 'watch this space'.

[Searider]

Searider I'm planning something similar and it will be my first foray into Arduino territory. Would you be willing to post a how-to on your metering setup?

I have no problem sharing how I set things up. Might make more sense to add it as a thread in the Arduino forum. let me see what I can put together.

[Wade]

I have no problem sharing how I set things up. Might make more sense to add it as a thread in the Arduino forum. let me see what I can put together.

Great, thanks.

[jwbryan]

Hi Searider, were you able to post the meter setup? I need to get two flow rates in my house (one for the well head, one for the water being treated) and hoped to build one instead of buy one.

[Michael McSharry]

Tasmota firmware in any of the ESP8266 devices is able to count pulses. I modified a version, described in the mcsMQTT manual, where I had the pulse-to-ounces conversion done in the ESP8266 and configured via the web page. mcsMQTT also has this ability so any pulse input can be viewed in HS as a volume or flow rate measurement.