Leak detection...

[Externet]

Good day.
The title considers ideas and projects design.
Not proposing yet, but am after your opinions if it is worth proposing. Water meters were read by reading them for many decades. On knees, under hot sun, or snow. Now it is a lazy guy sitting in air conditioned truck cruising the neighborhood and its laptop or whatever polling the meters by the sidewalks and reading/logging the monthly consumption while having a sandwich and soda on hand. The modern 'hard' work.

I suppose there is a meter identifier and a consumption reading being transmitted by demand. How complex would it be to make a monitor to have at home and tell if there is a water leak in the dwelling ? Say every 8 hours or at 03:00 AM to check for a second if there is consumption, lit a red LED at the gadget as warning.

I do not know the details of its operation, but may need a RFID sort of circuit. What do you know ?

 

[Pommie]

I thought these systems relied on a cell network to comunicate the reading.

If you can interrogate the meter then a better system might be to log a reading every so often (15min intervals) and if any two readings are the same then no leak exists.

Mike.

 

[For The Popcorn]

Some utilities monitor water meters for continuous flow and are somehow informed if there is continuous flow over a 24 hour period. At my partner's grandparents, we came home to a notice on the door. "You have a leak. Fix it within 24 hours or your water will be shut off." After I few panicked thoughts, I remembered I had left a garden hose faucet on the day before that had a leak around the stem. The nozzle was shut off, but the leak around the packing was enough to trip the alarm.

 

[Nigel Goodwin]

This is our core business - water meters (at least in the UK) generally don't have any kind of transmitting or external capability. What they do have is an option to add a sensor that detects the 'rotation' of the meter and provide a pulse output - depending on meter type, each pulse represents a specific number of litres - the handheld test unit we manufacture can be set from 0.5L per pulse to 1000L per pulse (the last was added on request, and is obviously a VERY large meter on a water main).

From the sensor various items can be connected, including loggers (which transmit the meter reading to a central location, most use SMS, mine use GPRS), BMS's (Business Management Systems), and local readers - a device mounted up on the wall, which the water guy reads using a handheld reader, presumably RFID of BT etc, very short range anyway. As the sensors are only single units, to connect multiple device requires splitters, which we manufacture and is a large part of our business.

As far as I'm aware there no 'people driving round in vans, sitting there and reading meters wirelessly' in the UK, it's either done manually, via a device that you have to essentially touch, or remotely with no human intervention at all (I have a number of loggers due to come in shortly this morning, from deepest darkest Scotland )

So it really depends on exactly what the OP has, he would need to examine his meter and anything attached to it.

 

[Ramussons]

Good day.
The title considers ideas and projects design.
Not proposing yet, but am after your opinions if it is worth proposing. Water meters were read by reading them for many decades. On knees, under hot sun, or snow. Now it is a lazy guy sitting in air conditioned truck cruising the neighborhood and its laptop or whatever polling the meters by the sidewalks and reading/logging the monthly consumption while having a sandwich and soda on hand. The modern 'hard' work.

I suppose there is a meter identifier and a consumption reading being transmitted by demand. How complex would it be to make a monitor to have at home and tell if there is a water leak in the dwelling ? Say every 8 hours or at 03:00 AM to check for a second if there is consumption, lit a red LED at the gadget as warning.

I do not know the details of its operation, but may need a RFID sort of circuit. What do you know ?

If you can define what is "Leak" and "Use", you have the answer.
This means that you have to monitor each utility outlet, sum it all up and compare it with the reading on the main meter periodically, and any difference means leak.

Or you use the principle of statistics and declare that water utility cannot be "on" all 24 hours. Any "use" indicated by the main meter on a continuous basis can be considered as a leakage.

 

[Externet]

There has to be dozens of methods. ----> https://www.ecosia.org/images?q=wireless water meter reading device

The one am more familiar to guess about is the driver with a helmet as if it was a racing course, doing nothing but cruising the neighborhood

with a laptop on board probably with a

And later dumping the readings to the main computer at the office... another machine prints the invoices, another machine puts them in envelopes and prints postage... minimal human intervention, more water service raises for management to scratch their 'tonsills', never to lower water service cost with the savings from reduction of personnel.

Any system they use, is probably cheap enough to make a reader to watch for leaks at home.

 

[Nigel Goodwin]

There are numerous meters, here's one of the sensors we wire to splitters for the water companies:

Cyble Sensor

www.itron.com

I still can't see the point in sending a man round in a van?, if you're transmitting wirelessly from the meter/logger why not just do it directly via SMS or GPRS?.

Quick thought - as I understand it the USA has ceased 2G, that will have broken pretty well every meter logger out there, as they normally use 2G.

My loggers can be set (remotely) to transmit at various periods you want - minutes, hours, days, weeks or months. The biggest issue is battery life - normally they transmit once a week - but I've got two on test (one at work, and one at home) that have been transmitting daily for about 2 years now, so that's about 14 years at once a week. The reason for the test is that these two are using one 18650 Li-Ion battery, normally we use a much larger non-rechargeable Lithium battery.

 

[Nigel Goodwin]

If you can define what is "Leak" and "Use", you have the answer.
This means that you have to monitor each utility outlet, sum it all up and compare it with the reading on the main meter periodically, and any difference means leak.

Or you use the principle of statistics and declare that water utility cannot be "on" all 24 hours. Any "use" indicated by the main meter on a continuous basis can be considered as a leakage.

For locating mains leakage they use a technique called 'step detection' - using a branching technique.

So in the middle of the night, when water usage should be minimal, and users are less likely to be disrupted you fit a transmitter to a main water meter in the street - this transmits the cubic meters per hour going through the meter.

From that meter/water main the pipes will branch off in different directions - like the branches on a tree.

So you go to one of the main branches, carrying a receiver displaying the meter flow, and turn off that branch, then wait a minute or so, and see if the water flow drops - if it does then the leak (or at least the high consumption) if down that branch. You then turn it back on, and either work further down that branch if the leak is that way), or go and try one of the other main branches.

It's called 'step detection' because if you graph the results you get a 'step' down on the graph when you turn the leak off.

We've currently got a number of such units we're developing out on test with a couple of water companies, except ours use GPRS again, so work any distance as long as you have phone coverage, where the normal radio ones are short range.