Close

Does this project spark your interest?

Become a member to follow this project and don't miss any updates

Remote Water Consumption Display

Design and build a simple, inexpensive, and manufacturable remote indicator for the most common type of mechanical water meter in use today.

Similar projects worth following
This project is submitted for

This project was created on 06/13/2014 and last updated 7 months ago.

Description
My project is to design and build a remote display for the most common type of mechanical water meters in use today. This system needs to be inexpensive, simple to install, and simple to use. In short, my Dad has to be able to use it. In its basic form, there will be a "sender" that is installed outside at the water meter, and a receiver placed inside the home. The receiver will display immediate water usage, along with historical consumption data. Further, the receiver will be able to connect with the consumer through the Internet to provide regular reports, as well as alert them to possible problems (pipe breaks, leaks, etc.).
Details

PROJECT INTRODUCTORY VIDEO

Water is an issue now, and it will continue to be an issue in the future. Global warming, watershed reduction, waste, whatever the reasons, the availability of clean water is already a crisis in many parts of the country, and there is nothing to indicate the situation is going to improve any time soon. A major part of dealing with the water crisis is conservation, but that is not exactly easy.

Sure, "just use less water" is pretty easy, but we're humans. Unless you have your local water cop tapping you on the shoulder a few times a day you're going to forget. And if you're like the vast majority of Americans, you have absolutely no idea how much water you've used until your monthly utility bill comes. And that's a problem. You simply can not address a problem that you can not see, and you can not address the problem effectively if you do not have rapid feedback from changes and visibility of immediate measurements. That's what I want to solve.

There are a few similar projects out on the Internet, and while they are good, they have not been designed for simplicity, or for manufacturing.

While this project will likely result in a manufacturable product I have no intention of doing that myself. With all of my designs for this project being totally open and free, it is my hope that someone with the resources to bring it to market will use my work as either an inspiration or an actual foundation for a real product. If nothing else, maybe a lot of Makers can duplicate my project, and together we can save a little water. :-)

As it stands right now, I have nothing more than an idea and a rough block diagram. I will be chronicling the entire project here, from initial design, prototyping, design revisions, right on through to the final form. I hope you click on "follow" and come along for the ride.

SYSTEM DESIGN:

I really think this is going to be fun. 8-)

John

Oh, by the way ... I'll probably be talking about this on Twitter and G+

       https://twitter.com/JohnS_AZ

       https://plus.google.com/u/0/102867471209152798204/posts

Components
  • 1 × PIC12LF1840 Transmitter: Microcontroller
  • 1 × MAG-3110 Transmitter: 3-Axis Magnetometer Sensor
  • 1 × TWS-BS Transmitter: RF transmitter module, 4800 baud serial
  • 1 × MCP1702-3302 Transmitter: 3.3V LDO Voltage Regulator
  • 1 × 6V 900mAH rechargeable battery pack Transmitter: Power supply
  • 1 × Photovoltaic Solar Panel Transmitter: Battery charger, weatherproof, at least 7VDC
  • 1 × 1N4001 Transmitter: Blocking diode for solar battery charger
  • 1 × TLC59282 Display, Water Wheel : 16-Channel, Constant-Current LED Driver with 4-Channel Grouped Delay
  • 16 × LED Display, Water Wheel : Indicator
  • 1 × LCD Display Display, Text : 4X20 Character generic LCD display module

See all components

Project logs
  • Log #14 Hostile Environments

    7 months ago • 4 comments

    So, I've been working on my receiver code off and on as time permits. During this time my prototype sensor/transmitter was out at the water meter. I had placed it in zip-lock bag and tightly closed it with a rubber band. That was in another bag, and that whole assembly was held onto the water meter with a much larger rubber band (I think it came from a bunch of broccoli). All of the bag openings were down, and the large metal cover was placed over the meter box. Situated like that I had no worries about water getting to the sensor. Besides, I'd have PCBs soon and the final sensor would be fully potted.

    I can only guess, but here's how I imagine it happened.

    The city meter reader must have opened the meter box (who else would?) and noticed plastic bags wrapped around the meter.

    I'll avoid ascribing mental attributes to this unknown person, but for whatever reason, he (or she I guess) pulled everything off of the meter. That by itself would have been fine. But for whatever reason, this person removed the outer bag, then pulled off the rubber band and removed the inner bag. I presume he lost interest because he then tossed the bags, rubber bands, and the exposed sensor assembly (still attached to its power supply) several feet from the meter box. He replaced the metal cover, and left.

    Again, that would have all been okay but ...... that afternoon and night, it rained.

    *sigh*

    The pictures show the result of less than 24 hours exposed to the elements (under power). The strange thing is that pin number 5 that almost totally eaten through. That connector is the ICSP (programming) connection, and there was nothing plugged into it. All I can imagine is that it ended up with a VERY good connection to ground.

    At any rate, the PIC is cooked, and I have to replace the connector on the mag sensor, but I don't have high confidence that it works either. I haven't decided if I will recreate the prototype board, or just order and wait for custom board to come in (that I can actually waterproof).

    Oh well .... onward and upward.

    John

  • #13 Maybe you can buy one! (Just not from me)

    9 months ago • 2 comments

    Interesting!

    I'm still working on firmware, and am waiting for another (faster) PIC for my sensor and will report all that when I have some accomplishments to talk about.

    Tonight, a good friend on IRC pointed me to a KickStarter project, the Water Hero, that is pretty much matches my project.

    Well, kind of.



    Let me say right up front, they have designed a very cool device!

    There are a number of differences between their design and mine and it really doesn't meet my requirements, but it may yours.

    - The Water Hero (WH) depends on WiFi from the sensor and talks directly to the internet. Mine talks to a base station.

    - The WH has no local dedicated display, a feature that I believe is crucial for awareness and conservation.

    - The WH is only suitable for indoors installation and virtually all water meters here are outside and underground. The KickStarter does say that an outdoors version is under development.

    - The WH requires an AC outlet to plug in the power supply. (My sensor is solar powered)

    - The WH requires you install an app to access the system, and they offer both a free and a paid 'premium' service to monitor your system. Again, no local dedicated display.

    - The WH has an optional shut-off valve actuator. (very nice!)

    - As far as I can tell, the WH is totally closed. Nothing "open" to it at all.

    It looks like they've done some fine work there, and I hope their project succeeds. If their product meets your needs, I would encourage you to support them.

    Several folks have asked if I am going to support their KickStarter.

    Well ..... no.

    I've already built one. :-)

    John

  • Log #12 Can you hear me now?

    9 months ago • 0 comments

    Not a whole lot to report yet. I've been working integrating a real time clock (RTCC) to my project, and optimizing my code that drives the LCD display. But one interesting thing did happen ... I got interviewed.

    Elecia White, one of the Hackaday Prize judges, and her husband Christopher White host and produce a podcast called "Embedded". I am a bit of a podcast junkie, and theirs is one of my favorites. I was a bit startled when Elecia contacted me and asked if I'd like to be a guest on their show. I mean, they have some pretty notable guests on there and I had a bit of an "I'm not worthy" moment. But one of the up-sides of being my age is that you can approach a lot of things saying "What the hell, sounds like fun." So last Saturday we recorded the show and now it has been published.

    In it, we discuss the hackaday prize contest and my entry, or course. We also talked about contests in general and a raft of other topics. All in all it was a great deal of fun talking with them and the 70 minutes flew by amazingly fast (and I don't think I came across as too big a doofus).

    So THANKS Elecia and Chris for having me! And I hope you folks enjoy listening to it.

    Here's the link: EMBEDDED #74 : ALL OF US CAME IN SIXTH

View all 14 project logs

Build instructions
  • 1

    Build Instructions.

    Huh.

    I'm honestly not sure what to put in this section, but the rules say I need build instructions.

    I mean, I am posting all of the schematics in the Build Log entries, All of the source code is up there. When I get to the point I have PCBs the artwork will be posted, and everything about this project is open source.

    So, I guess the Build Instructions are as follows:

    1) Buy all the parts as I've listed them above.

    2) Wire them together as per the schematics I've posted above.

    3) program the microcontrollers from the source code I've linked up there.

    That's about it. I'm really not trying to be flip or insulting. As I go along if some aspect of this project needs a more in depth explanation, I'll certainly add it.

    And in the mean time, if YOU think of anything I should include in this section PLEASE do not hesitate to leave a comment for me!

See all instructions

Discussions

hack wrote 10 months ago point
Are you having any problems measuring peak and fall on the sensor at higher RPMs? I imagine that the mains meter will have the potential to spin rather fast.

Are you sure? [yes] / [no]

John Schuch wrote 10 months ago point
I wondered about that but haven't seen any problems. I turned on a garden hose on full and turned on a sprinkler circuit and the sensor seemed to keep up. If I do see a problem I have the option of dialing back the oversampling in the sensor which will boost the sample rate from 20Hz to 40 or 80.

Are you sure? [yes] / [no]

Eric Tsai wrote 11 months ago point
Hi,
I'm doing a home automation project myself. Your idea of using the magnetometer is great. I didn't know the wheel thing has a rotating magnetic field. I'm excited to give this a try.

My home automation project has many of the elements yours does - wireless transmitter, a server platform. I'd like to do some alarming for long periods of water usage, which might indicate that someone forgot to shut off something.

Just wanted to express my appreciation for the idea.

Are you sure? [yes] / [no]

John Schuch wrote 11 months ago point
Thanks, and good luck with your project!
I'd be interested in hearing how it goes.
John

Are you sure? [yes] / [no]

Jac Goudsmit wrote a year ago point
Congratulations on making it to the last 50 of the Hackaday prize! http://hackaday.com/2014/08/25/50-semifinalists-selected-for-next-stage-of-the-hackaday-prize/

Are you sure? [yes] / [no]

joram wrote a year ago point
Also consider metric units, for those non US residents. You might also add a second sensor on the hot water flow for example, to measure the consumption of heated water, with temp sensor. As a matter of fact, a tempel sensor on the cold input alsof makes sekse. We alsof Reuss rain water, zo mocht as welk measure that amount. Thuis gives an overview off all the water consumption of a household. As input we could measure the inflow on the rainwater tank. Just if you run out of ideas...

Are you sure? [yes] / [no]

Jelli wrote a year ago point
You will find the hardware you are looking for at Sensus.com, BadgerMeter.com and tehamawireless.com. There are several major companies releasing consumer products very soon.

Are you sure? [yes] / [no]

Rogan Dawes wrote a year ago point
I'm fascinated with the idea of measuring water flow, and doing real-time reporting, not so much for individuals (although that's certainly important!) but also at a utility level.

For example, i f you have a main pipe that is recording 100l/s flow, and 3 subsidiary pipes that are reporting 10l/s flow, you either have exceptional demand from the consumers between the main pipe and the subsidiary pipes, or you have a burst pipe somewhere. Being able to detect that and react in real time without waiting for a resident to report it would be incredibly useful.

One problem I have is powering the sensor and transmitter, of course, but I was wondering how difficult it would be to "siphon" off a little energy from the flow of water itself. Obviously I don't want to put anything into the pipe that would obstruct the flow, or be at risk of damage when the inevitable grit and small stones get into the pipes during maintenance. I have just discovered that water is diamagnetic, and wonder if there is a way of using that property in conjunction with a powerful magnet to generate enough energy from the flowing water to power the sensor and the transmitter.

Are you sure? [yes] / [no]

CasualCasual wrote a year ago point
Have you done power consumption tests for the magnetometer and wireless transmitter? I imagine the wireless transmitter will be the big power hog. I've seen power specs for that Mag at about 24uA, along with 10uA for the PIC, with .55ah per 9 volt is about 2 years of run time, but the transmitter needs about 8mA, which if running constantly would only give you about 3 days of run time.

This is a really interesting project, and I'd be interested to see exactly how much water I'm using during my overnight lawn watering.

Are you sure? [yes] / [no]

DeepSOIC wrote a year ago point
And how about optical readout? I've seen people make covers (or shields) over standard water meters which have a led-phototransistor pair sensing the rotation of a black cogwheel. Does your meter have such a cogwheel (a black asterisk-like thing that spins)? Looks like these cogwheels are added deliberately for the purpose.

Are you sure? [yes] / [no]

DigiGram wrote a year ago point
I really like this idea. Our water measurements are not taken monthly for the bill, they just estimate (recently bought a house and the meter was not read in over 2 years!) So we have to read the meter monthly and email the reading in. Something like this can be scripted to automatically send the readings monthly (and my wife has asked me for something like this just last month). I'll be following your progress.

P.S. I like your way of going about it without changing the plumbing etc!

Are you sure? [yes] / [no]

John Schuch wrote a year ago point
Thanks for the comment! My motivation .. a few years ago we had an underground irrigation pipe break and I had no clue until we received a multi-thousand dollar water bill. (I settled the bill for much less) But still, I want to be able to keep an eye on it.

Are you sure? [yes] / [no]

Adam Fabio wrote a year ago point
Great Idea! Many people take it for granted, but water is a precious resource. I can't wait to see how you implement your meter side sender. Don't forget to document your design as you go along, and thanks for submitting to The Hackaday Prize!

Are you sure? [yes] / [no]

John Schuch wrote a year ago point
Hi Jac. (Don't know why this thing won't let me reply directly to your comment) I think there are a few issues with using a RPi and camera in this application. First is obviously cost. Then there's complexity. I envision a system where you place a sensor near the meter and you're done. With a RPi and camera you have to mount and aim the camera, and teach the vision system what to look for on the water meter: Fails the "my Dad can do it" test. :-) And power. A PV panel and battery pack for a sensor and simple transmitter will be fairly modest, while powering a RPi and camera would take a LOT more juice. And finally there's the nincompoop city meter reader. If he bumps that camera, you'd get to start all over again. I don't think it would be "cheating", and it would be a cool project. It just doesn't mesh with what I'm after.

Are you sure? [yes] / [no]

John Schuch wrote a year ago point
Yes, that would be ideal, but adding a new flow sensor would vastly increase the cost. I've decided that a design requirement is that the system can not require modifications to the existing plumbing. Essentially, I want to read the meter that is already there. The mechanical meters I'm talking about have an internal magnetic coupling, and I plan to detect that externally.

Are you sure? [yes] / [no]

zakqwy wrote a year ago point
Makes sense. Great start!

Are you sure? [yes] / [no]

Jac Goudsmit wrote a year ago point
Have you thought of just putting a Raspi with a camera on the meter, and running some software to either transfer pictures over Wifi, or (a bit more difficult) read the numbers straight off the camera picture and process the data in cool ways e.g. build statistics files, show the difference between now and yesterday etc.)? That way you don't need sensors to sniff out the magnetic coupling. Of course, on the other hand, you could argue that that method would be cheating, and probably also less fun ;-)

Are you sure? [yes] / [no]

zakqwy wrote a year ago point
Have you thought about an electromagnetic flow measurement system? Better reliability and accuracy vs mechanical meters, but a bit tougher to design and build.

Are you sure? [yes] / [no]

CasualCasual wrote a year ago point
You'd have to put that in-line with the main supply valve, and water companies get skittish when they see that you've mucked about with their system. (to the tune of replacing the meter and sending you the bill for it)

Are you sure? [yes] / [no]

zakqwy wrote a year ago point
What about a homemade clamp-on meter? Think that's in the realm of possibility for a DIYer? Seems like the transducers would be tough to build..

Are you sure? [yes] / [no]

CasualCasual wrote a year ago point
Hmm, I hadn't ever heard of indirect flow meters. There's an ultrasonic version that requires bubbles or particulate matter in the water to sense flow. http://www.omega.com/prodinfo/ultrasonicflowmeters.html This one in particular says that it won't work with clean drinking water.

Maybe this paper could prove useful? http://www.sersc.org/journals/IJCA/vol5_no4/18.pdf I don't have time to read it right now though.

Are you sure? [yes] / [no]

zakqwy wrote a year ago point
Ultrasonic meters generally work on one of two principles; the unit you describe uses the Doppler effect, bouncing a signal off of entrained particles or bubbles. Other units, called 'transit time' meters, use a pair of spaced transducers to compare the relative sonic velocity of the fluid with and against the flow direction, and use the time differential to calculate fluid velocity (since sound is just a pressure wave moving through a medium).

Acoustic coupling and ignoring incorrect pulses are two of the many challenges to building one of these setups yourself, but it would be a fun project!

Are you sure? [yes] / [no]

CasualCasual wrote a year ago point
Somehow your description made it sound much more complicated than this video I found. https://www.youtube.com/watch?v=rqJ9snaqYiY
Here's another informative video, https://www.youtube.com/watch?v=b1V4fzpDvAM

Seems like it'd be worth trying though. It might be tough to mount in this particular application since you'd have to have a good section of pipe, but it'd be a neat test that others might find useful. Most of the indirect flow meters I've seen in my tiny bit of research have been very expensive, so an open source version might be pretty nifty.

Are you sure? [yes] / [no]

CasualCasual wrote a year ago point
So, with some angled wedges and some ultrasonic emitter/sensors an indirect flow meter could be made. Repurposing a range finder might work, especially if the price can be found low enough, http://www.epictinker.com/Ultrasonic-Distance-Sensor-p/hc-sr04.htm . Maybe unsolder either the transmit or receive transducer, add a cable extension that way you can get the distance between the transducers correct. 3d print a housing with some hose clamps to attach it to a pipe.

I'm not sure what the wedges would need to be made of. The specs I've seen say "sound conductive material" which makes me think a hard silicone or a large block of pencil eraser, but I'm not too sure. Maybe even hot glue pumped into place within the housing attached to the pipe so there wouldn't be any air gap to cross.

With more free time it'd be a fun little project to try. Not sure how great the battery life would be for a remote sensing application. This type of sensor would need to be running continuously to keep up with flow.

Are you sure? [yes] / [no]

CasualCasual wrote a year ago point
Something like this maybe, http://i.imgur.com/ghYIrvU.png

Are you sure? [yes] / [no]

Similar projects