WLD System
WLD System
WLD System
By:
Ryan Hanson
Senior Project
Electrical Engineering Department
California Polytechnic State University
San Luis Obispo
2017
Table of Contents
List of Tables…………………………………………………………………..…………………………. iii
List of Figures………………………………………………………………….......................................... iv
Abstract………………………………………………………………………............................................. v
Chapter 1: Introduction…………………………………………………………………………………… 1
Chapter 2: Design Specifications………..………………………………………………………………... 4
Chapter 3: System Design and Functional Decomposition……………….…………………………….... 7
Chapter 4: Project Planning…………………………………………………………………………….... 12
Chapter 5: Design and Testing……………………………………………………………………………17
Chapter 6: Conclusions and Recommendations..…………………………………………………………48
References………………………………………………………………………………………………... 49
Appendices
Appendix A: Senior Project Analysis………………………………………………................................. 51
ii
List of Tables
Table 2-1: WLDS Requirements and Specifications ................................................................................5
Table 3-1: WLDS Functional Requirements ............................................................................................7
Table 3-2: WLDS- 1ASLOBLO Fuse ......................................................................................................8
Table 3-3: WLDS - LED ..........................................................................................................................9
Table 3-4: WLDS - Power Module.......................................................................................................... 9
Table 3-5: WLDS - Water Sensor Module .............................................................................................. 9
Table 3-6: WLDS - Microprocessor Module...........................................................................................10
Table 3-7: WLDS - Actuator Module......................................................................................................10
Table 4-1: Deliverables Table…………………………………………………………………………..14
Table 4-2: Estimated Project Costs .........................................................................................................14
Table 4-3: Winter 2017 Project Costs…………………………………………………………………..15
Table 4-4: Total Project Costs…………………………………………………………………………..16
iii
List of Figures
Figure 3-1: Level 0 Black Box Diagram.................................................................................................... 7
Figure 3-2: Level 1 Block Diagram........................................................................................................... 8
Figure 4-1: Project Gantt Chart………………………………………………………………………….12
Figure 4-2: Winter 2017 Project Gantt Chart……………………………………………………………13
Figure 5-1: ALSROBOT Water Sensor…………………………………………………………………17
Figure 5-2: Arduino Microcontroller with Water Sensor Schematic……………………………………18
Figure 5-3: Serial Monitor Output of Water Sensor…………………………………………………….20
Figure 5-4: Pinout of XBeeS2C…………………………………………………………………………21
Figure 5-5: Image of XBee USB Adapter…………………………………………………………….....22
Figure 5-6: XCTU Software Parameters………………………………………………………………...23
Figure 5-7: Block Diagram of XBees in Transparent Mode…………………………………………….24
Figure 5-8: Successful Communication Between Two XBees in AT Mode……………………………24
Figure 5-9: Block Diagram of Arduinos Driving XBees………………………………………………...25
Figure 5-10: Successful Communication Between Two XBees Using Arduino MCU…………………26
Figure 5-11: Serial Console of Router XBee……………………………………………………………27
Figure 5-12: Serial Console of Coordinator XBee……………………………………………………...28
Figure 5-13: 12V DC Eco-Worthy Linear Actuator…………………………………………………….29
Figure 5-14: Control Relay Options…………………………………………………………………….30
Figure 5-15: 2-Channel Relay Terminals……………………………………………………………….31
Figure 5-16: Complete Setup of Actuator Control Module…………………………………………….32
Figure 5-17: Arduino GSM Shield 2…………………………………………………………………...33
Figure 5-18: Adafruit FONA 3G……………………………………………………………………….36
Figure 5-19: Arduino and GSM Shield Schematic……………………………………………………..39
Figure 5-20: SMS, Buzzer, XBee and Water Sensor Network…………………………………………42
Figure 5-21: Actuator Control…………………………………………………………………………..43
iv
Abstract
The water supply shortage has increased in recent years due to overpopulation, climate
change and obsolete water facilities, where deteriorated pipes cause most of the water leaks. The
problem is not the size of the leak, but the time it takes to detect it. This paper presents the
implementation of a system installed in the hydraulic facilities of a residence, to detect water
leaks. The system consists of a water sensor installed by a water reservoir of interest, a
microprocessor to interpret the data and evaluate whether it is a water leak or not, an SMS alert
message, and an electrical actuator to shut off the main water supply to avoid leakage.
v
Chapter 1: Introduction and Background
With the growth of the world population, the demand of fresh water has increased
causing serious problems in the field of water supply. Therefore, control of water has become a
considerable issue today. Scientists, technicians, politicians, and generally, many other
inhabitants of the planet become increasingly educated on the subject. The threat of pollution
hovers over and limits water supplies. The shortage of this vital liquid requires great attention.
The proportion of fresh water found in rivers, lakes, and underground sources comprise only 3%
of the total amount of water on earth [2]. In addition, the water found needs treatment for human
consumption, to eliminate particles and organism harmful to health, and ultimately must
distribute through pipes to homes safely [3].
This work focuses on the issue of distribution, more specifically, on the issue of “water
leaks” in residential areas. Anyone who’s had a water heater, dishwasher or burst pipe disaster in
their home knows how important early detection can be. Even those slow leaks that only cause
mold damage require expenses to repair. The more water spilled (or splashed) the more money
the repairs cost to residents. For this reason, it’s crucial to have some system installed in
residences to detect water leaks.
The water detector electronic device is designed to detect the presence of water and
provide an alert in time to allow the prevention of water damage. A common design includes a
small cable or device that lies flat on a floor and relies on the electrical conductivity of water to
decrease the resistance across two contacts. The device then sounds an audible alarm together
with providing onward signaling in the presence of enough water to bridge the contacts. These
provide usefulness in a normally occupied area near any infrastructure that has the potential to
leak water, such as HVAC, water pipes, drain pipes, vending machines, dehumidifiers, or water
tanks.
Water leak detection is an expression more commonly used for larger, integrated systems
installed in modern buildings or those containing valuable artifacts, materials or other critical
assets where early notification of a potentially damaging leak proves beneficial. Specifically,
water leak detection has become a necessity in data centers, trading floors, banks, archives, and
homes. The water leak detection industry, small yet specialized, has only a few manufacturers
1
operating world-wide [8]. The original application was in the void created by "computer room"
floors in the days of large mainframe computer systems. These use a modular, raised floor based
on structural "floor tile" usually 600 mm square and supported at the corners by pedestals [18].
The void created gave easy access and routing for the mass of power, networking and other
interconnecting cables associated with larger computer systems - processors, drives, routers etc.
mainframe computers also generated large amounts of heat so a void under the floor could also
serve as a plenum to distribute and diffuse chilled air around the computer room. Chilled water
pipes generally run through the void along with the drains for condensates associated with
refrigeration plant. In addition, designers found the floor void a very convenient place to route
other wet services feeding bathrooms, radiators and other facilities [15]. A leak occurring within
a floor void would therefore go unnoticed until the hydrostatic head of pressure meant that the
water found its way through to floors below where its dripping through the ceiling penetrates the
joints and connectors of the power or network cabling and cause system failure from short
circuits.
Current digital water leak detection systems can locate multiple water leaks to within 1
meter resolution over a complex network of cables running several kilometers [20]. This
functionality reduces the downtime and potential damage caused by inaccurate reporting
common with older analogue based systems [20].
The Z-Wave equipped Fibaro Flood Sensor ($59) includes a tilt sensor (so the user knows if
someone moved it), temperature sensor, a siren, a light, a built-in Z-wave network range tester,
and it can be wired or wireless. The company says it works with any professionally-installed
alarm system, but the user can also set it up. With a Fibaro Relay installed, the sensor can also
automatically shut off a solenoid valve. The Flood Sensor requires Fibaro’s Home Center 2 or
another Z-Wave hub.
The Wally system doesn’t use Z-Wave or Bluetooth or Zigbee. It uses a proprietary
wireless system that takes advantage of the copper wires in your walls and treats them like
antennas for the system. The moisture sensor (which also detects temperature and humidity)
helps with large water emergencies and also alerts the user about slowly developing moisture—
the kind that can lead to mold damage. The user also needs the Wally hub and the sensor for this
system. A customer can purchase the hub with six sensors for $299. Individually the sensors cost
$35.
2
This paper presents the instrumentation of a Water Leak Detection System (WLDS), a
simple but cost effective water detection system, implemented in residential areas and offers a
detailed description of the system throughout the following chapters. The next chapter dives into
the WLDS system specifications and customer needs.
3
Chapter 2: Design Specifications
Chapter 2 describes the thought process behind creating the WLDS, the customers likely to buy
this product, and the design specifications of the product.
4
that no further leakage occurs. The WLDS detects and eliminates a water leakage in a short time,
because in most cases, the problem originates not from the leak itself, but the time to detect it.
The microcontroller deactivates the water pump, if the water sensor detects 20mL or
more water leakage. ¼ mL of water equals approximately a drop of water from a faucet, so
20mL of water ensures no false alarm [9]. The actuator and water sensor used to deactivate the
water pump is small and light enough for any user to carry and install. The actuator can obtain
power from a wall socket which negates the need for additional power [4]. The water sensor
features battery powered hardware, allowing maximum mobility so that it may run for at least 30
hours. Ideally, installation only requires mounting the water sensor, installing the actuator into
the water pump, and plugging in the motor to the nearest residential power outlet. Table 2-1
summarizes detailed marketing requirements as well as the engineering specifications of the
proposed WLDS.
TABLE 2-1
WATER LEAK DETECTION SYSTEM REQUIREMENTS AND SPECIFICATION
Marketing Engineering Justification
Requirements Specifications
2,5,7 Meets standards of International The unit sits in an outdoor setting exposed to
Protection code IP51 various amounts of dust and condensation. This
requires a level of weather resistance.
2,5,7 Meets NFPA 70 National Product uses electrical equipment. Meeting this
Electric Code Safety Standards standard keeps the user safe from electrical
hazards.
2,5,7 This product meets Functional Product meets UL Standards to keep user safe.
Safety Listed to applicable UL
Standards and requirements by
UL
1,5 All materials cost less than The price ceiling keeps the system affordable.
$100
2,5,8 Requires no additional user Residents don’t have to worry about their water
5
input beyond initial installation pump system.
5,6,8 Battery powered actuator Mobile actuator incase users need to move to
different location/pipe.
5,6 Actuator must run for at least 30 Actuator must be able to run continuously for at
hours. least 30 hours to ensure cost effectiveness for
the user.
5,6 Battery powered water sensor A mobile sensor so the user doesn’t have to
worry about plugging into a wall socket.
5,6 Water sensor must run for at The sensors must run for an extended period
least 30 hours. before batteries need replacement.
1,3,5 Actuator consumes less than During its inactive state, it uses minimal power
500 mW during standby to maintain efficient function.
operation
5,6,8 Installation takes less than 1 The device comes ready out of the box,
hour for an untrained user requiring only that the user install the sensors,
attach their actuator, and plug the device in.
5,6,8 Actuator and water sensor must This size restraint ensures the device places
have dimensions smaller than easily and does not require excessive space.
8”x5”x3”
5,6,8 Actuator weighs less than 5 lbs. A user can more easily carry a lighter device.
5,6,8 Water sensor weighs less than 2 Makes sensors light enough to place at user
lbs. friendly locations
5,8 Speaker outputs warning beep Typical alarm clocks are built between 50-100
no louder than 80 dB, and no dB. Warning sound must have dB in this range
quieter than 50 dB. to hear from anywhere in house.
Marketing Requirements
1. Affordable
2. Autonomous
3. Low Power Consumption
4. Accurate Pump Control
5. Unobtrusive
6. Easy to Install
7. Weather Resistant
8. User Friendly
6
Chapter 3: System Design and Functional Decomposition
Chapter 3 will focus on the main design of the WLDS system and a functional breakdown of
each component within the system. Figure 3-1 and Table 3-1 show the level 0 block diagram
and the input signals of the WLDS system consecutively. The microcontroller within the WLDS
relies on the data from the water sensor to accurately control the water pump. For simplicity and
ease of use, the AC power source provides power to the fuse and actuator. The microcontroller
then makes a decision based on the status of the sensor, and it outputs a signal to the actuator,
sends a text alert to the user, and sounds an 80 dB alarm so the user can hear when home.
TABLE 3-1
LEVEL 0 WATER LEAK DETECTION SYSTEM FUNCTIONAL REQUIREMENTS
Module Water Leak Detection System
7
Figure 3-2: Level 1 Block Diagram
Figure 3-2 shows the level 1 block diagram of the system which features the major modules and
their connections. The line voltage feeds into theAC/DC power module [4]. The resulting voltage
powers the microprocessor and LED. The line voltage splits off to feed the actuator which relies
on a control signal from the microprocessor to determine whether the line voltage passes on to
stop the water pump.
TABLE 3-2
LEVEL 1 WLDS - 1A SLOBLO FUSE
Module -1A SLOBLO FUSE
Table 3-2 shows information about the 1A SLOBLO fuse. This fuse provides overcurrent
protection for the AC/DC converter [2]. SLOBLO allows short surges of current through
without blowing the fuse. This ensures the components can source the starting current they need.
8
TABLE 3-3
LEVEL 1 WLDS - LED
Module -Status LED
Table 3-3 shows information regarding the LED. The LED uses the power output of the AC/DC
converter to produce light, signifying power to the system.
TABLE 3-4
LEVEL 1 WLDS - POWER MODULE
Module -AC/DC Converter
Table 3-4 shows information about the AC/DC Converter module. The AWPC accepts universal
AC input such that the AWPC could be used anywhere regardless the available local residential
AC input. In this module the system rectify the AC signal provided by an electricity company,
then step down the DC voltage to 3.3V [6].
TABLE 3-5
LEVEL 1 WLDS - WATER SENSOR MODULE
Module -Water sensor
9
Table 3-5 shows the module for the Water Sensor. The module monitors a given area for a water
leak. As an example, the sensor has an output signal of 3.3V which outputs a control signal to the
microcontroller when the system detects a leak.
TABLE 3-6
LEVEL 1 WLDS - MICROPROCESSOR MODULE
Module -Microprocessor
Table 3-6 shows the input and output signals of the Microcontroller module. The microcontroller
processes data acquired from the water sensor and sends an output signal to the actuator. The
microprocessor also produces a sound alert and text alert, when it receives the input.
TABLE 3-7
LEVEL 1 WLDS - ACTUATOR MODULE
Module -ACTUATOR
Table 3-7 shows the input and output signals information of the actuator module. This module
waits for the controls signal 2 to activate to stop (or deactivated to start) pumping water to into
the user's home.
10
The WLDS modules building process is recorded below in Chapter 4, as well as the estimated
and actual cost of the project.
11
Chapter 4: Project Planning
Chapter 4 includes the WLDS Gantt charts for Fall, Winter, and Spring Quarter. The major project
milestones are within these Gantt charts. Along with the planning, this chapter includes cost estimates for
each quarter and the actual cost at the end of the project.
12
Figure 4-2: Winter 2017 Project Gantt Chart
Figure 4-2 above shows certain project milestones completed by the end of winter quarter.
TABLE 4-1
13
Project Deliverables
Delivery Deliverable
Date Description
February 16 Design Review
March 17 EE 461 Review
w/Advisor
March 24 EE 461 report
April 28 EE 462 Design Review
May 19 ABET Sr. Project
Analysis
June 1 Sr. Project Expo Poster
June 15 EE 462 Report
Table 4-1 above shows the milestones completed for the WLDS throughout Fall, Winter, and
Spring quarter.
TABLE 4-2
ESTIMATED PROJECT COSTS
Cost Estimates
Item
Cost
System Parts
Controller $15
Sensors $25
Subtotal $100
Testing Parts
Tubing $20
Containers $20
Subtotal $140
14
Table 4-2 shows the anticipated cost estimates for the Water Leak Detection System. Simulating
a water pump requires tubing and containers. Cost estimates include wires and circuit boards
because the system requires wires to make connections between components and the circuit
board to integrate the components. The section labeled other includes tools such as hot glue, and
electric tape. And finally, the enclosure material serves as the material cost estimate for the box
used to protect all the circuitry from the weather, dust, and water that may endanger the inner
circuitry.
Assuming the average electrical engineer receives a pay between $25/hr to $35/hr and an
estimated time of 200 hours to make the product, the labor costs range from $5000 to $7000.
Using a most likely cost of $6000, the estimated labor cost equals the most likely cost.
TABLE 4-3
WINTER 2017 PROJECT COSTS
Table 4-3 above depicts the cost of the WLDS system parts for Winter 2017.
15
TABLE 4-4
TOTAL PROJECT COSTS
Item Cost
PVC Tubing $8
Piezo Buzzers $5
Comparing tables 4-4 and 4-2 one will notice that the project only went over budget by 15%.
The subsequent chapters include the bulk of the report which include design decisions, testing,
results, and conclusions.
16
Chapter 5: System Design Decisions and Testing
Chapter 5 includes the bulk of the report and consists of the different modules for the WLDS.
This includes design decisions and testing for the water sensor, XBee Wi-Fi modules, buzzer
alarm, SMS alert system, and actuator control. This chapter also includes detailed explanations
of each module and system schematics.
Design
The liquid level sensor, built by ALSROBOT, displays the part number RB-02S048A. Amazon
or EBay sell this part for under $2. It's very inexpensive. The sensor outputs an analog voltage
proportional to the amount of liquid exposed to it. The sensor has a series of parallel wires across
the board. These wires sense the liquid level that the board is exposed to. I simply connect the
analog pin, represented by an S, to an analog pin on the Arduino board to read the analog value.
The simplicity, cost, low power consumption and compatibility with the Arduino makes this
water sensor the best decision for the project. The sensor operates on 5V and needs less than
20mA for operating power current, which means the Arduino can easily provide this (so no
external power needed to power it). The liquid level sensor has 3 pins. It's very basic. Figure 5-1
below shows the pinout of the sensor [25].
Two of the pins power the device, 1 connecting to the +5V of the Arduino and the other
connecting to the ground terminal of the Arduino. The other pin, with an S, stands for the signal
pin. This pin outputs the analog voltage signal proportionate to the amount of water on the sensor
covered with liquid. This pin connects to an analog pin on the Arduino board and monitors the
value.
For this circuit, the LED lights up when the liquid level of the sensor grows greater than 600.
17
This value meets the sensitivity specification because an analog value of 600 equates to
approximately 6 droplets of water (1.5mL). After the LED worked with the Arduino for testing
purposes, a buzzer alarm took its place. [25].
The liquid level sensor board needs 5V of power from the 5V and GND pins on the Arduino.
The signal pin (S) gets connected to an analog pin on the Arduino board. Here we connect it to
analog pin A0. This allows the Arduino board to read the analog voltage value. The LED light
has its anode terminal connected to digital pin D13 and its cathode.
Code
/*Code for Liquid Level Indicator Circuit Built with an Arduino*/
void setup() {
Serial.begin(9600); //sets the baud rate for data transfer in bits/second
pinMode(sensorPin, INPUT); //the liquid level sensor input to the arduino
pinMode(ledPin, OUTPUT); //the LED output
}
void loop() {
liquid_level= analogRead(sensorPin); //arduino reads the value from the liquid level sensor
18
Serial.println(liquid_level); //prints out liquid level sensor reading for debugging
delay(2000); //delays 2 s
Serial.println(“Buzzer Buzzing!”);
}
else {
digitalWrite(ledPin, LOW);
}
}
The commented code above explains itself for the most part. The sensorPin initialized to 0 and
LED to 13. The liquid level variable holds the value of the analog value output from the sensor,
serving as the sensor reading representing how much water leaks on the sensor. The setup
function sets the baud rate.
The loop function repeats over and over. It reads the value from the sensorPin and stores it in the
variable liquid_level. The serial monitor prints this value for debugging purposes.
The analogRead function, the most important function for the module, reads the sensor pin and
outputs a value anywhere from 0 to 1023, proportional to the voltage from the sensor. If the
sensor is submerged in no liquid, the Arduino registers a 0 reading. If the sensor is fully
submerged in liquid, it registers a full reading of 1023. A reading of approximately 600 equates
to 1.5mL of water. Thus, when the reading grows above 600, the LED/speaker turns on,
indicating a leak. Otherwise, the LED and speaker turn off.
19
Testing Module 1:
20
Module 2: XBee Radio Communication
Design
Wi-Fi could work for my project but the power consumption is too high and the high data rate
proves unnecessary. Bluetooth is useful for control but not applicable for sensor networks such
as the WLDS. The XBee has a low data rate (250kbit/s), low power consumption, and is used to
create mesh-type sensor networks making it the perfect fit for the WLDS. Each XBee device can
communicate with one another, and through each other via a mesh network to devices out of
range.
The schematic above depicts the pinout of the XBee Radio Module. The main pins used for this
project include the Data Out/in, Gnd, 5V, and one of the i/o pins which connects to the water
sensor. The XBee, programmed through software called XTCU, allows users to directly assign
the XBee parameters such as IP Address, sample rate, router/coordinator functionality.
ZigBee defines three different device types: coordinator, router, and end device.
Coordinator
ZigBee networks always have a single coordinator device. The coordinator starts the network,
selecting the channel and PAN ID. It also buffers wireless data packets for sleeping and end
device children and distributes addresses, allowing routers and end devices to join the network.
The coordinator manages the other functions that define the network, secure it, and keep it
healthy. This device cannot sleep and must constantly run.
Router
A router equates to a full-featured ZigBee node. This device can join existing networks and send,
receive, and route information. Routing involves acting as a messenger for communications
between other devices too far apart to convey information individually.
21
Can buffer wireless data packets for sleeping end device children. Can allow other routers and
end devices to join the network. A router cannot sleep and may have multiple router devices in a
network [27].
End device
An end device describes essentially a reduced version of a router. This device can join existing
networks and send and receive information, but cannot act as messenger between any other
devices. Cannot allow other devices to join the network.
XBee features Transparent Mode (AT) or Application Program Interface Mode (API). The first
mode makes sure the XBees function properly and work with XCTU software and a USB cable
only. In transparent mode, if I sent a packet of data saying “Hello World” from the router XBee,
the coordinator XBee should receive the information and display “Hello World” on its serial
monitor. API mode utilizes the XBees’ data pins, allowing interfacing with a microcontroller.
The WLDS utilizes API mode to transmit the analog data from the water sensor to a remote
location via wireless communication [27].
The XBee transceivers have a 2mm pin spacing which does not interface with the standard 0.1
inch breadboard. There are several different breakout boards that bypass this spacing [27]. The
various adapter boards also allow for connection through USB or serial to your computer. The
breakout board used here has a UART to USB conversion circuit to connect the XBee to the
computer and X-CTU software easily.
One can see from the image above that adapter has just the right number of pins to mount the
XBee, solder points that make it easy to hook up the adapter to a breadboard, and most
importantly a mini USB port used to connect the XBee directly to the computer to change
parameters.
22
Testing Module 2: Connecting to and Using the XCTU Software
The Figure above shows the configuration of two different XBees. Both XBees connect to the
computer using the adapter shown in figure 5-5 and a USB to mini USB cable. The XBee in the
top right corner with a C by it stands for coordinator XBee and the other with no letter stands for
router XBee. Once one XBee initializes to the coordinator, all subsequent XBees connected
automatically initialize as Routers unless changed within XTCU. Most of the XBEE parameters
remain unchanged except for a few important ones.
The PAN ID allows the XBees to find one another remotely, so this number is the same for each
XBee connection. The Channel Verification (JV) equals zero for the coordinator and one for the
router. This parameter makes it easier for the router to find the coordinator, maintain connection
while the coordinator remains powered on, and disables connection otherwise. Baud Rate equals
9600 and Sampling Rate equals 50ms [27].
23
Testing Module 2: AT Communication Between Two XBees Using Serial Communication
The block diagram above depicts a simple connection between two XBees and a computer. The
XBees use adapters to function properly with the computer via USB.
Figure 5-8 above shows the router XBee communicating with the coordinator XBee in AT mode.
The green connection button in the image shows that the connection is open between the two
XBees. In the router’s serial console (BLUE) I typed HELLO WORLD and as seen above, the
Coordinator processed the information wirelessly and displayed the message received from the
router in red.
24
Testing Module 2: AT Communication Between Two XBees With Two Arduino MCU’s
Figure 5-9 shows the block diagram of two XBees in AT mode each connected to a MCU and
this time the MCU connects to the computer via USB instead of the XBee. The difference
between this setup and the previous setup (Figure 5-7) is that instead of typing the message in via
serial monitor and manually sending the message, this setup utilizes the Arduino MCU, crucial
for the WLDS to work properly. The goal of this setup includes programming one Arduino to
transmit a message and the other to receive and display the message wirelessly all through Code
Composer Studio instead of XCTU software.
void setup () {
Serial.begin(9600);
}
void loop() {
Serial.println("Hello World");
delay(5000);
}
This code simply starts the serial monitor and prints “Hello World” to the screen every 5
seconds.
25
Code for Arduino w/XBee Coordinator
void setup() {
Serial.begin(9600);
}
void loop() {
if (Serial.available()) {
Serial.write(Serial.read());
}
}
The code checks if the serial monitor’s availability. If available, the Coordinator XBee receives
the message from the router XBee, sends it to the serial read port of the Arduino, and prints out
“Hello World” wirelessly.
Figure 5-10: Successful Communication Between Two XBees Using Arduino MCU
The figure above shows correct communication between XBees. First, the usbmodem numbers at
the top of the two serial monitors differed from one another. This means that I have two
Arduinos plugged into my laptop, one running the Coordinator Code, and the other running the
Router code. Usbmodem1421, the coordinator XBee, clearly receives “Hello World” from the
router XBee every 5 seconds. Wireless communication is established between the two XBees
using only the Arduino MCU, and now the router XBee can wirelessly transmit water sensor data
to the coordinator XBee.
26
Testing Module 2: XBee Communication Using API Mode
API (Application Programming Interface) mode is a frame-based method for sending and
receiving data to and from a radio's serial UART, an alternative to the default transparent mode
[27].
The WLDS needs API mode to function properly, because it allows transmitting of data frames
between the water sensor, Router XBee, and Coordinator XBee. The data below shows an
example of the router sending a message to the coordinator in frame form, how to read the
frames explained in the following passages [27].
The data below shows the frames received by the coordinator when the router sent the message
“Hello” in packet form.
Figure 5-11 above shows a screenshot of the router XBee’s serial console. Seen above, packet
zero, which translates in hex to “Hello”, transmits via mesh networking to the Coordinator XBee.
27
Figure 5-12: Serial Console of Coordinator XBee
As seen in the figure above, the coordinator XBee received the packet of data from the router
XBee.
7E 00 11 90 00 13 A2 00 41 66 59 D2 81 3B 01 48 65 6C 6C 6F 37
Start delimiter: 7E
Length: 00 11 (17)
Frame type: 90 (Receive Packet)
64-bit source address: 00 13 A2 00 41 66 59 D2
16-bit source address: 81 3B
Receive options: 01
RF data: 48 65 6C 6C 6F
Checksum: 37
The data above represents different information received by the Coordinator from the router.
Each set of Bytes represent different parameters about the XBee. For this line of Bytes, the light
blue set holds the highest precedence. This set represents the ASCII characters “H,e,l,l,o” in Hex
starting with byte 48 and ending with byte 6F.
28
Module 3: Actuator Control
The figure above shows the actuator used to stop water flow when the WLDS detects a leak. The
specs include an input voltage of 12V DC, a starting current of 0.8A, and max current of 3A. The
actuator has a max load of 1500N/Push, a speed of 5.7mm/s, stoke length of 4 inches and duty
cycle of 25%.
A 2-channel relay utilizes communication between the Arduino Uno Microcontroller and the
Eco-Worthy Actuator. The relays control the direction the actuator moves. They work by using
current from the input source to activate an electromagnet, which pulls a switch that allows
higher currents on the opposite side of the relay to flow.
29
Figure 5-14: Control Relay Options [28]
On the control side of the relays a GND pin, IN pins numbered from 1 to 8 depending on the
relay model, and a VCC pin exist. The relays require power to stay activated, so a stable 5V
power supply works. Otherwise the Arduino has trouble powering the higher channel relay
modules. On the control side of the relay, 5V power supply connects to the VCC and GND pins.
Next, the IN pin connects to the corresponding Arduino pin, then the relays activate once the IN
pins connect to the GND pins. On the relay side, there exist three main parts of each relay, and
three screw terminals [26].
30
Figure 5-15: 2-Channel Relay Terminals [28]
These terminals refer to the Normally Closed (NC) connection, the top one, the Common (COM)
connection, the middle one, and the Normally Open (NO) connection on the bottom. If no
connections to the IN pin exist, then the relay connects between the NC and COM terminals. If
the 5V power source connects to the IN pin, then the relay connects between the NC and COM
terminals as well. Finally, the user connects the IN pin to the GND pin, the relay connects
between the NO and COM terminals [26].
31
Figure 5-16: Complete Setup of Actuator Control Module [27]
Figure 5-16 above shows the complete setup of the actuator module. This tests the actuator and
makes sure it functions properly. The code below controls the actuator using Arduino.
void setup() {
void loop() {
digitalWrite(forwards, LOW);
digitalWrite(backwards, HIGH); //Activate the relay one direction
delay(2000); // wait 2 seconds
digitalWrite(forwards, HIGH);
digitalWrite(backwards, HIGH); //Deactivate both relays to brake the motor
delay(2000);// wait 2 seconds
digitalWrite(forwards, HIGH);
digitalWrite(backwards, LOW); //Activate the relay the other direction
delay(2000); // wait 2 seconds
32
digitalWrite(forwards, HIGH);
digitalWrite(backwards, HIGH); //Deactivate both relays to brake the motor
delay(2000);// wait 2 seconds
GSM
GSM, an international standard for mobile telephones, stands for Global System for Mobile
Communications, or also 2G.
To use GPRS for internet access, and for the Arduino to request or serve webpages, the user needs to
obtain the Access Point Name (APN) and a username/password from the network operator. Among
other things, GSM supports outgoing and incoming voice calls, Simple Message System (SMS or text
messaging), and data communication (via GPRS) [28].
The Arduino GSM shield 2 acts like a GSM modem. From the mobile operator perspective, the Arduino
GSM shield looks just like a mobile phone. From the Arduino perspective, the Arduino GSM shield 2
looks just like a modem.
GPRS
GPRS is a packet switching technology that stands for General Packet Radio Service. It can provide
idealized data rates between 56-114 kbit per second. The user can also access the internet using the GSM
shield. Like the Ethernet and WiFi libraries, the GSM library allows the Arduino to act as a client or
server, using http calls to send and receive web pages [28].
33
Digital pins 2, 3 and 7 are reserved for communication between the Arduino and modem and cannot be
used by one’s sketches. Software Serial library handles communication between the modem and Arduino
on pins 2 and 3. Pin 7 is used for the modem reset. When the yellow status LED turns on, it means the
modem is powered, and the user can try connecting to the network.
This sketch checks the modem's IMEI number. Modems have unique IME numbers, and they use the
numbers to identify valid devices that can connect to a GSM network. Once the modem reads the number,
the Arduino prints out the connected network carrier and the signal strength of the network over the serial
port.
// PIN Number
#define PINNUMBER ""
void setup()
{
// initialize serial communications
Serial.begin(9600);
Serial.println("GSM networks scanner");
scannerNetworks.begin();
// connection state
boolean notConnected = true;
34
}
}
void loop()
{
// scan for existing networks, displays a list of networks
Serial.println("Scanning available networks. May take some seconds.");
Serial.println(scannerNetworks.readNetworks());
}
After double checking the circuitry and the placement of the sim card, the program would still
not run correctly using the Arduino GSM. I spent countless hours troubleshooting , and finally
solved the problem. I figured out the sim card and code functions properly. Instead, the cell
phone providers around the area (San Luis Obispo) do not support 2g anymore and the Arduino
gsm shield that I had purchased only works on a 2g network. I bought a new 3G shield and
everything worked perfectly.
35
Module 4a: New 3G Text Message Alert Module:
The figure above shows the important pinouts of the Adafruit FONA, used to send an automated text
message alert when a leak is present.
Device Specs
● Quad-band 850MHz GSM, 900MHz EGSM, 1800MHz DCS, 1900MHz PCs - connect onto any
global GSM network with any 2G or 3G SIM.
● American Version dual-band UMTS/HSDPA 850/1900MHz WCDMA + HSDPA
● Fully-integrated GPS (Qualcomm PM8015 GPS) controlled and query over the same serial port
● Make and receive voice calls using a headset or an external 8Ω speaker + electret microphone
● Send and receive SMS messages
● Send and receive GPRS data (TCP/IP, HTTP, etc.)
● AT command interface used with 300, 600, 1200, 4800, 9600, 19200, 38400, 57600, 115200,
230K, 461K, 961K, 3.2M, 3.7M and 4.0Mbps
● Native USB support - plug it into a computer an
On its own, this module can't do anything. It requires a microcontroller to drive it! I chose an Arduino, but
any 3-5V microcontroller with a UART can send and receive commands over the RX/TX pins.
36
Required accessories to make FONA 3G work
● Required Lipoly Battery - 500mAh or larger. I used a 3.7V 1200mAh battery.
● 3G mini sim card. At&t GO sim cards work great.
● MicroUSB cable for charging the battery and communicating with the module over USB
● External uFL GSM Antenna - this slim one works great (or, if the designer wants to us an SMA
antenna - a uFL to SMA adapter cable.)
● External Active GPS Antenna (needs a uFL to SMA adapter too)
Below are the pinouts and a description on what each one does.
● JST 2-pin - the battery input connector. It works with any Lipoly batteries, but, since the charge
rate equals 500mA (and the cellular module can spike high current draw), a
500mAh or 1200mAh batteries works best.
● MicroUSB connector - the LiPoly/LiIon battery charging port.
● Headset jack - the 'standard' TRRS 3.5mm phone headset jack with stereo earphone and mono
microphone
● The GSM antenna plugs in up top. To use the module for any voice or data communications as
well as some SIM commands, the user needs an antenna.
● The designer can either use a uFL GSM attena like this, or use a uFL to SMA adapter and then an
SMA antenna
The most important pins are broken out at the bottom of the board. All of the following fall under 3-5V
input safe.
● Vio - THE MOST IMPORTANT PIN! The designer MUST drive with an external voltage from
3V-5V to set the logic level converter. The converter also buffers the indicator LEDs so
NOTHING appears to work unless power reaches this pin! The user should set the voltage to
whatever voltage the microcontroller uses for logic. A 5V micro (like Arduino) should have it set
to 5V, a 3V logic micro should set it to 3V.
● Key – The power on/off indicator tied to the button in the top left. Tie this pin to ground for 3 to
5 seconds to turn the module on or off. It's not a level signal ("low means off, high means on") -
instead the designer must pulse it for ~5 seconds to turn off/on. The module comes by default off.
Tie this permanently to ground if the designer never wants the micro to turn off the FONA for
power saving.
● 5V - The USB 5V from the microUSB connector when it’s in and powered. Good if the designer
needs to know when the microUSB plugged in and/or want to recharge the battery from an
external plug.
● PS - The Power Status pin. Low when the module has no power and high when the module has
power. If using the Key button or pin, the designer monitors this pad to see when the modules
booted up.
● NS – The Network Status pin. It lights up/blinks to signal the status of the module and tied to
the Net LED so for more detail see the LEDs section below.
● Reset - Module hard reset pin. By default, it has a high pull-up (module not in reset). Toggle this
pin low for 100ms to perform a hard reset.
● RX (in) & TX (out) - The module uses UART to send and receive commands and data. Auto-
baud so whatever baud rate sent "AT" after reset or boot equals the baud rate used.
● RTSin - The hardware flow control pin. The pin stops and starts data transfer from the FONA 3G
37
to the microcontroller
● CTSout - The hardware flow control pin. This pin determines when the FONA 3G's serial buffer
becomes full and when to stop and start data transfer to the FONA 3G from the microcontroller
● DTR - The hardware flow control pin, used with the FONA 3G to hang up calls, control
data/command mode for TCP/IP, etc.
● RI – The Ring Indicator, basically the 'interrupt' out pin from the module. By default, high and
designer can configure RI to go low when it receives a call. The designer can also configure RI to
go low when an SMS received.
● SPK+ and - : Used toconnect an external 8 ohm speaker.
● MIC + and -: Used to connect an external electret microphone and to bias the mic with 2V.
LEDS
38
Figure 5-19: Arduino and GSM Shield Schematic
The figure above is the schematic of the Arduino, FONA GSM, and computer tied together.
Wiring with Arduino Microcontroller
● Vio connects to 5V
● GND connects to GND
● Key connects to GND (always on)
● RX connects to digital 2
● TX connects to digital 3
● RST connects to digital 4
#include "Adafruit_FONA.h"
#define FONA_RX 2
#define FONA_TX 3
39
#define FONA_RST 4
// We default to using software serial. If the coder wants to use hardware serial
// (because softserial isnt supported) comment out the following three lines
// and uncomment the HardwareSerial line
#include <SoftwareSerial.h>
SoftwareSerial fonaSS = SoftwareSerial(FONA_TX, FONA_RX);
SoftwareSerial *fonaSerial = &fonaSS;
//
// HardwareSerial *fonaSerial = &Serial1;
void setup() {
while (!Serial);
Serial.begin(115200);
Serial.println(F("FONA SMS caller ID test"));
Serial.println(F("Initializing....(May take 3 seconds)"));
Serial.println("FONA Ready");
}
void loop() {
40
if (fona.available()) //any data available from the FONA?
{
int slot = 0; //The slot number of the SMS
int charCount = 0;
//Read the notification into fonaInBuffer
do {
*bufPtr = fona.read();
Serial.write(*bufPtr);
delay(1);
} while ((*bufPtr++ != '\n') && (fona.available()) && (++charCount < (sizeof(fonaInBuffer)-1)));
The code above checks to see if the Fona can receive and send a text. After getting the Arduino
and Fona to work together correctly, the next step was to put all the modules together and get a
working system, shown in detail below.
41
Complete System:
The WLDS includes two main modules: Module 1 and Module 2. Module , the bulk of the
system, consists of the water sensor, sms alert, buzzer alarm, and xbee router. Module 2 consists
of the actuator, relay module, and xbee coordinator.
All the components in Module 1 can run off the 9V battery, but the Fona 3G requires extra
amperage to run properly via the external LiPoly Battery. The water sensor continuously
monitors for water at a certain sensitivity between its conducting wires. When the sensor reaches
that certain sensitivity, the Arduino sends a signal over to the Fona through serial interface,
42
which then sends a text using it’s 3G capabilities to the user, thus notifiying the user that he/she
has a leak and should return home shortly. The XBee Router also takes in the analog data from
the sensor and sends it via mesh networking to the XBee Coordinator, part of Module 2
described below.
The figure above describes Module 2 of the WLDS. After Module 1 detects a leak, the XBee
coordinator receives this data and processes the data via the Arduino MCU. When the sensor
detects water, the Arduino sends a signal to the relay activating the actuator and shutting off the
main power. Once the sensor detects no water, the actuator retracts inward simulating turning the
water back on.
43
CODE FOR MODULE 1:
#include "Adafruit_FONA.h"
#define FONA_RX 9
#define FONA_TX 8
#define FONA_RST 4
#define FONA_RI 7
#include <SoftwareSerial.h>
SoftwareSerial fonaSS = SoftwareSerial(FONA_TX, FONA_RX);
SoftwareSerial *fonaSerial = &fonaSS;
void setup() {
pinMode(FONA_RI, INPUT);
pinMode(buzzer, OUTPUT);
pinMode(FONA_RST, OUTPUT);
digitalWrite(FONA_RST, LOW);
delay(100);
digitalWrite(FONA_RST, HIGH);
while (!Serial);
Serial.begin(115200);
Serial.println(F("FONA SMS Alarm Notification v1"));
Serial.println(F("Initializing....(May take 3 seconds)"));
fonaSerial->begin(4800);
if (! fona.begin(*fonaSerial)) {
Serial.println(F("Couldn't find FONA"));
while (1);
44
}
Serial.println(F("FONA OK"));
while (1) {
uint8_t n = fona.getNetworkStatus();
Serial.print(F("Network status "));
Serial.print(n);
Serial.print(F(": "));
if (n == 0) Serial.println(F("Not registered"));
if (n == 1) Serial.println(F("Registered (home)"));
if (n == 2) Serial.println(F("Not registered (searching)"));
if (n == 3) Serial.println(F("Denied"));
if (n == 4) Serial.println(F("Unknown"));
if (n == 5) Serial.println(F("Registered roaming"));
if (n == 1) break;
delay(500);
}
void loop() {
// read the sensor and store it in the variable sensorValue:
int sensorValue = analogRead(A4);
if (sensorValue > 36) { // if the sensor reading equals less than threshold
var = 1;
} else {
var = 2;
}
switch (var) {
case 1:
Serial.println(sensorValue); // displays sensor reading during alarming status
Serial.println("ALARM DETECTED!"); // displays that alarm detected
Serial.println("Sending first notification...");
Serial.println(notifyPhoneA);
digitalWrite(buzzer, HIGH);
fona.sendSMS(notifyPhoneA, "Dear Ryan, hurry home your house has a leak! Shutting off
45
the main water pump");
delay(5000); // delay to check for alarm reset
break;
case 2:
Serial.println("System armed - not alarming"); // displays armed status
Serial.println(sensorValue);
digitalWrite(buzzer, LOW);// displays sensor reading during armed status
delay(3000); // delay to rerun program if alarm not sounding
break;
}
Serial.begin(9600);
pinMode(forwards, OUTPUT);//set relay as an output
pinMode(backwards, OUTPUT);//set relay as an output
void loop()
{
if (Serial.available() >= 21) { // Make sure the frame all there
if (Serial.read() == 0x7E) { // 7E =the start byte
for (int i = 1; i<19; i++) { // Skip ahead to the analog data
byte discardByte = Serial.read();
}
int analogMSB = Serial.read(); // Read the first analog byte data
int analogLSB = Serial.read(); // Read the second byte
int analogReading = (analogLSB + (analogMSB * 256));
Serial.print(analogReading);
}
46
else{
digitalWrite(forwards, LOW);
digitalWrite(backwards, HIGH);//Activate the relay the other direction, they must differ to move the
motor
delay(2000);// wait 2 seconds
}
47
Chapter 6: Conclusions and Recommendations
The main goal of this project is to provide an autonomous system that prevents excessive
water damage within a household. After working on this project, I better understand the
difficulties people face when trying to create a product from scratch. The WLDS detects a water
leak anywhere the user chooses to put it and alerts the user with a text message, in case of house
vacancy. The WLDS differentiates itself from many products on the market today, because when
leakage occurs, not only does it alert the user with an alarm and text message, but it also powers
an actuator installed at the main water pump which shuts off the main water preventing further
leakage. This project is more proof of theory than application, because of time and money
constraints. This means that I only created one sensor network and created communication
between the sensor and actuator, but if I had more time, the creation of multiple around the house
would increase the product efficiency. In addition, incorporating the actuator into the piping
creates a better water shutoff network because it doesn’t shut off all the water in the house,
instead, just where leakage occurs. All the specifications listed in Table 2-1 are met besides the
price constraints. The prices of the components used in the project totaled $229, coming in 15%
over budget. The final product met all performance requirements after testing, using water to
simulate a leak. However, future improvements on the WLDS can still be made. Color coding
the sensors would make it even more clear which one goes where. This implementation would
make the installation process much easier for the customer by reducing the chances of incorrect
installation. Water resistant packaging seems crucial to the product working correctly and
meeting NFPA safety standards. Also, the XBee ADC pin can only reach 1.2V total, and the
water sensor I used went up to 5V, so I spent too long troubleshooting this problem, before I
realized the pin of the XBee malfunctioned instead of something wrong with my connections. To
prevent this, incorporating a different Wi-Fi module into the system, or utilizing a simple voltage
divider both fix the problem. Finally, the Arduino includes a low-power mode not utilized in this
project. By changing the inputs of the sensors to interrupts, one could leave the microcontroller
in low-power mode until it needs to perform an action. This could dramatically reduce its power
usage.
48
References:
[1] TEXAS INSTRUMENTS INCORPORATED, 'DC-DC CONVERTER', 9 160 229, October 13,
2015.
[2] Hadipuro, 'Indonesia’s Water Supply Regulatory Framework: Between Commercialisation and
Public Service?', Water-alternatives.org, 2010. [Online]. Available: http://www.water-
alternatives.org/index.php/allabs/111-a3-3-1/file. [Accessed: 17- Oct- 2016].
[3] Iec.ch, 'IEC - World Plugs: List view by location', 2016. [Online]. Available:
http://www.iec.ch/worldplugs/list_bylocation.htm. [Accessed: 17- Oct- 2016].
[5] N. Sokal, K. Sum and D. Hamill, 'A capacitor-fed, voltage-step-down, single-phase, nonisolated
rectifier', APEC '98 Thirteenth Annual Applied Power Electronics Conference and Exposition, vol. 1,
1998.
[6] R. Radzuan, M. Raop, M. Salleh, M. Hamzah and R. Zawawi, 'The designs of low power AC-DC
converter for power electronics system applications', 2012 IEEE International Symposium on
Computer Applications and Industrial Electronics (ISCAIE), 2012. 42
[8] Weifeng Huang, Tao Zeng, Liping Ye and Zhen Li, 'A self-acting water pump control system for
residential buildings based on resonance water level sensor', 2011 International Conference on
Electric Information and Control Engineering, pp. 265-357, 2011.
[9] B. Popkin, K. D'Anci and I. Rosenberg, "Water, hydration, and health", Nutrition Reviews, vol.
68, no. 8, pp. 439-458, 2010
[10] Jong-Chull Shon, “Water level sensor,” US6810732 B2, 2 Nov., 2004.
[11] R. Ford and C. Coulston, Design for electrical and computer engineers. Boston: McGrawHill,
2008.
[14] "Electronic & electronics components depot United States," 2016. [Online]. Available: https://
www.emartee.com. Accessed: Oct. 21, 2016.
[15] J. Millman and A. Grabel, Microelectronics: Digital and analog circuits and systems, 2nd ed.
49
New York: McGraw-Hill Companies, The, 1987.
[16] G. Martinsanz, "Sensors for fluid leak detection," Sensors, vol. 15, no. 2, pp. 3830–3833, Feb.
2015.
[17] Ghazarian, John, Ruggieri, Tom and Balaster, Ammon. (2009). “Secure Wireless Leak
Detection System. World Intellectual Property Organization (WIPO)”, WO/2009/017512.
[18] Gray Douglas (GB).(2008). Water leak alarm for premises using acoustic signal sent from water
meter. Intellectual Property Office. GB2447903 (A).
[19] Yi-Bong Lin, Imrich Chlamatac. (2001). “Wireless and Mobile Network Architectures”.
Editorial Wiley. ISBN 0-471-39492-0.
[20] S. Ferdous Hoque, “Water Conservation in Urban Households : Roles of Prices, Policies and
Technologies”, IWA Publishing, April 2014.
[22] D. Tan, “8 Things You Should Know About Water & Semiconductors”, China Water
Risk. http://chinawaterrisk.org/resources/analysis-reviews/8-things-you-should-know-about-water-
and-semiconductors/ July 11, 2013.
[23] W. DeOreo, “California Single Family Water Use Efficiency Study” The California Department
of Water Resources. http://www.irwd.com/images/pdf/save-
water/CaSingleFamilyWaterUseEfficiencyStudyJune2011.pdf June 1, 2011.
[24] D. Meadows, “Leverage Points: Places To Intervene In A System”, Whole Earth Winter, 1997.
[26] Belloc, Hilaire. On. Freeport, N.Y.: Books for Libraries Press, 1967. Print.
[28] Industries, Adafruit. "Adafruit Industries, Unique & Fun DIY Electronics And
Kits". https://www.adafruit.com/product/3147. N.p., 2017. Web. 29 May 2017.
[29] Mao, Guoqiang, and Bariș Fidan. Localization Algorithms And Strategies For Wireless Sensor
Networks. Hershey, PA: Information Science Reference, 2009. Print.
50
Appendix A: Senior Project Analysis
Project Title: Water Leak Detection System
The Water Leak Detection System WLDS has a couple of main components. The water sensors,
the microcontroller, and the DC Motor. The Motor runs off line voltages between ~80-264
VAC. The sensors are battery powered as well as the microcontroller. A user installs the motor
directly into the main water pump and places the microcontroller/water sensor system around the
house where they feel a water reservoir needs monitoring. With no further actions necessary, the
WLDS monitors the water reservoir for any leaks present and alerts the user when it finds one.
Responding to the found leak, the motor shuts off the main water pump preventing any further
leakage.
● Primary Constraints
There exist both monetary and location-based constraints associated with this project. The
WLDS must function in various countries such as the U.S. which uses 110V or Indonesia which
uses 220V. One must also keep in mind the various wall outlet types [3]. Making the system easy
to install and independent of the users’ water system offers yet another challenge. This may
require designing protection circuitry in case a user makes a mistake during installation. The
water sensors must also meet NEMA standards.
Building a non-intrusive actuator system is also challenge. Non-intrusive systems exist, but they
sell for a much greater price than intrusive systems, which indicates greater implementation
difficulty. If this fails, I plan to relax my spec to an intrusive system. The power consumption of
the WLDS also presents difficulty. The actuator operates off the electrical mains, but has a
backup battery as well. The backup battery must last for at least 30 days, which sets a limit on
the power consumption of the water sensor. I attempted to solve this problem firstly by
optimizing the system to use a minimal amount of power. Because that didn’t meet the
specification, I moved onto a more expensive battery. The microcontroller deactivates the water
pump if the water sensor detects 20mL or more water leakage. ¼ mL of water equals
approximately a drop of water from a faucet, so 20mL of water ensures no false alarm [9]. The
actuator and water sensor used to deactivate the water pump is small and light enough for any
user to carry and install. The actuator can obtain power from a wall socket which negates the
need for additional power [4]. The water sensor features battery powered hardware, allowing
maximum mobility so that it may run for at least 30 hours. Ideally, installation only requires
mounting the water sensor, installing the actuator into the water pump, and plugging in the motor
to the nearest residential power outlet. Table 2-1 summarizes detailed marketing requirements as
well as the engineering specifications of the proposed WLDS.
51
● Economic Impact
As with all products, there exist many economic impacts likely to result following the
implementation of this project. Many companies benefit from the need for goods and services
associated with the project such as shipping, raw materials, parts, manufacturing, and labor.
Almost every aspect of the creation of the WLDS requires labor in some shape or form making it
an important aspect [20]. Local banks also benefit from the additional purchase made by the
businesses and the product user. It saves water districts money that they can then allocate to
other conservation efforts. The most significant economic impact, however, is the reduced water
waste, stepping closer towards sustainable usage of the Earth’s natural resources [20].
Users benefit from the WLDS through using an effective and efficient water leak detector and
low power consumption. With an effective control system in place, a water leak becomes
detected and terminated resulting in less money used for repairs and replacements. Beside the
initial investment of buying the product, there exist no upkeep costs beyond its use of power
except for repair costs if broken down. This becomes increasingly likely throughout the
product’s lifetime until a replacement becomes necessary. Users have the option of replacing the
water sensors, the only replaceable part, in case they alone fail. The WLDS requires the user to
already have a compatible water pump to use with the control box. The user of the WLDS,
someone who seeks autonomous control of their home water system, pays low overall costs for
the affordable product. Not including design or testing cost estimates, the estimated sum of parts
totals $100[4]-[7]. The project takes an estimated 150 hours of development time ending around
April of 2017.
The estimated fixed costs total $9150 to design, and development a prototype followed by an
estimated $100 per unit plus labor likely meaning $200 per unit. If each unit sells for $250, a
positive profit margin would emerge after 185 units sold assuming inefficient manufacturing.
Assembling this product through lean manufacturing practices on a production line would result
in much lower production costs. However, this also results in a higher initial investment to create
a manufacturing line. If it costs an estimated $50,000 to create a manufacturing line that reduced
production costs to roughly $100 and the selling price remained $250, it would require 334 units
sold to make profit. However, factories can produce at much faster rates when they use
manufacturing lines [21]. If 500 units sell per year, the device makes $75,000 yearly profit.
● Environmental Impacts
The materials that make up the WLDS include electronics composed of metals mined out of the
earth. Sometimes, these mining operations impact ecosystems quite seriously due to the overall
destruction of the surrounding environment. The manufacturing process uses harsh chemicals,
which if not disposed of properly, seep into water supplies contaminating drinking water for
humans and animals alike [2]. Like all other electronics, the WLDS consumes power which puts
pressure on power generation plants to produce more energy. This requires burning more fossil
fuels resulting in more greenhouse gasses emitted into the atmosphere. Greenhouse gasses also
52
accumulate from the increased vehicle presence caused by the need to transport the WLDS. The
atmosphere warms as the gasses accumulate leading to the degradation of various ecosystems
reliant on very specific temperatures such as aquatic ecosystems [2]. Once the lifetime of the
product has elapsed, the users must properly recycle the waste so it does not decompose quickly
and release toxic chemicals. Often, this waste ends up in developing countries negatively
affecting both the people and animals living there [9].
However, the WLDS makes its users more aware of their water usage and decreases overall
usage. If the WLDS does indeed reduce its customers’ water usage, its benefits may outweigh its
costs. Assuming it reduces each user’s water consumption by only 2%, 1 million installations,
and an average household water usage of 360 gallons per day [23], we save 7.2 million gallons
of water per day. These savings exceed the water costs to operate a large fabrication house of
4.8 million gallons per day [22]. Increased water savings can also save energy, and outweigh the
energy cost of manufacturing the Smart Water Meter since much of our energy goes toward
transporting and treating water.
● Manufacturability
The WLDS serves as an easy-to-use control box requiring no user input beyond initial
installation of the water sensors and motor. As a result, users likely struggle when attempting to
fix malfunctions, and when they occur, instead buy a new unit. While this benefits profit, it also
leads to less sustainability as more units require more resources. The WLDS must last a long
period necessitating a quality build. With no regular maintenance required, the unit requires very
little upkeep associated with the product aside from power costs. The product could potentially
include easily replaceable internal parts as an upgrade, but this would drive up manufacturing
costs. Also, the implementation of the electronics within the WLDS requires the use of rare earth
metals. Users can help sustain this resource if they disable the components and recycle them
correctly. However, this often does not happen resulting in aggressive unsustainable
consumption of resources. The modular design of the WLDS allows the individual
manufacturing of the modules. The circuitry likely requires the longest manufacturing time due
to its small size and precision design. The other components such as the microprocessor and
water sensors come pre-made and ready to integrate into the assembly. Installation of the WLDS
consists of placing the water sensors, connecting the motor to the water pump, and supplying
power from a nearby wall outlet.
● Sustainability
The WLDS strives to find leaks before any serious damage occurs to the user’s house. The
system needs minimal maintenance for the sensors used to monitor leakage [7]. The system only
requires connection to residential wall socket voltages to perform its functions. Potential
improvements, the incorporation of a solar panel to reduce energy usage would make the product
greener for the environment. However, the implementation of a solar panel would require a
battery to store the resulting charge. The batteries contain chemicals such as sulfuric acid which
can harm the environment. The WLDS could also potentially benefit from a wind generated
power source depending on average weather conditions of the region.
53
● Ethical Considerations
Regarding the Institute of Electrical and Electronics Engineers (IEEE) code of ethics, the WLDS
upholds ethical principles as an ethical device. The designers of said product accept no bribery
and discriminate against no persons based on race, religion, gender, disability, age, national
origin, sexual orientation, gender identity, or gender expression. Designers perform research
when deciding all price estimates and completion time estimates to ensure they honestly reflect
likely end costs. Designers make the completed product with the welfare of the public in mind by
including warning labels to help prevent incorrect usage. If unforeseen or unintended conflicts
arise resulting from the production or use of the device, responsible parties must take steps to
remedy said conflicts. While the designers of the project have intention to uphold the IEEE code
of ethics, the power to do so may lie with different people. If a governing body purchased the
WLDS in bulk and distributed them as they see fit, discrimination may become an issue. This
applies to the acceptance of bribes and setting fair prices as well.
From a utilitarian perspective, the WLDS could be either beneficial or harmful, depending on the
consumers who utilize it. An individual user utilizing this device gains happiness by not having
to constantly monitor his/her water system. This allows for optimal water monitoring, so the user
can enjoy daily activities without worrying about water leaks. If someone only introduces a few
devices into a community, those without the WLDS have greater chances of severe water
damage. This could lead to unhappiness. However, if all members of a community used the
WLDS then all members would benefit equally and the average happiness of the community
would rise. With initial successes, the governing body of the area may decide to include the
device as part of the water utility spreading the technology for others to benefit from resulting in
even more happiness. According to the utilitarian ethical framework, the WLDS has overall
utility, which makes it an ethically justified project. Several arguments exist for the good it
provides. Saving water (and thereby energy) has utility, because California is currently in a
drought and water usage needs reducing [23]. This also contributes to sustainability, proving
very useful in the long term. In addition, the revenue generated by the project creates jobs and
the money saved by using the product reduces financial strain on households [20]. Both provide
good for many people.
Several ethical dilemmas exist as well. Inevitably, the manufacturing process introduces new
social and political inequities or sustains old ones. The labor involved in manufacturing
typically is not paid well and goes to the person who accepts the least amount of money for
it. Inequity created by unfair treatment of workers, however, diminishes when the product
switches to manufacturing by a responsible company in the US. This costs more, but minimizes
the social and political inequities associated with my project. Overall, the good the product
provides outweighs the negative possibilities, assuming it is manufactured responsibly by a
company that treats its workers fairly.
54
● Health and Safety Impacts
The use of the WLDS helps users protect their property. Having a water leak detection system
helps maintain public health and safety. In addition, the autonomous nature of the WLDS allows
people living with physical disabilities or ailments to forgo the arduous work associated with
periodically fixing their water reservoirs or losing possessions due to water damage. Not all
aspects of the WLDS benefit people’s health. The electronics inside the WLDS require a
manufacturing process reliant on harsh chemicals. This is dangerous to the labor force if work
sites do not take proper precautions. As stated in section 5, these chemicals can sometimes leak
into water supplies, making anyone who uses the water for drinking ill. In addition, the
manufacturing setting generally resents a danger for workers due to the large machinery often
present. The motor runs on high voltages which can deliver painful and dangerous shock if
irresponsibly tampered with. Because the setting of the system necessitates placement outdoors
and near bodies of water, the device has an increased chance of becoming wet and, therefore,
higher chance of electric shock to the user. This same high voltage concerns during testing
because wrong wire placement can result in electric shock nearby people and damage to
surrounding components. Also, any metals the water sensors contain may rust and compromise
the safety of the water [7].
The WLDS relies on a water system where a reservoir supplies water and needs a pump to move
it to a residential storage tank. In countries like Indonesia, the government supplies the supplied
water [2]. If they feel like this product gives users an unfair advantage over others with regarding
water access, lawmakers may introduce new legislation. Those with expendable money can
afford this product giving them an advantage over those who cannot afford the product. This
further solidifies the relationship between money and power. The user benefits directly from a
well working water system making him/her the main stakeholder. Both water districts and
households are most likely to utilize the product, making them the main stakeholders. The
WLDS has several indirect stakeholders as well though, primarily the manufacturing workers as
well as plumbers. The WLDS has a net positive effect on plumbers, for whom it would create
jobs. It has a net negative effect on manufacturing workers since it creates inequities, but those
are minimized by selecting a responsible manufacturer. Other stakeholders may include any
investors who expect a return of profit. This could potentially cause issues if investors only
contribute if they can influence specific changes made to the product. Despite negative concerns,
the implementation of the WLDS would introduce technology into a community simplifying
otherwise difficult aspects of everyday activities, especially for the physically handicapped. If
the governing body in an area decided to purchase the WLDS for all residential buildings in an
area, it would help solve the issues of inequality and promote a sense progress in the community.
Other stakeholders include companies working to manufacture the WLDS. With sufficient
demand, the WLDS supplies opportunity for manufacturers to make money and create jobs.
55
● Developmental Comments
The design of the project system has necessitated the research of power electronics, specifically
converters between AC and DC voltage and DC to DC step-down [4]. The circuit designs
required sensitivity analysis Monte Carlo simulations. The software chosen, LTspice, warranted
research and practice in how to perform the desired analysis. This project required knowledge of
new tools such as Gantt charts which helped establish a clear schedule of expected milestones
related to the project. Other research conducted includes information on conditions in countries
like Indonesia, existing patents related to the WLDS, datasheets for potential components, and
worldwide wall-socket voltages. I also learned about adding new information flows to a system
and how they can change its behavior from the Donella Meadows article [24]. This motivated
my project to introduce real-time data in units that make sense to humans in the household water
consumption system. Adding this information hopefully affects everyday choices that add up to
a huge amount of wasted water and energy.
56