International Journal of Electronics, Communication & Soft Computing Science and EngineeringISSN: 2277-9477, Volume 3, Issue 6

1

Wireless Sensor Network Based System Design forChemical Parameter Monitoring In Water

N. Neha Beri

Abstract- Determination of natural freshwater qualities in theabsence of significant direct human impact, the critical waterquality parameters are essential to be monitored for the healthof the creatures. An autonomous real time device to measurethe physical and chemical parameters such as (PH,Temperature, Turbidity) in water using Arduino Atmega 2560Microcontroller is developed which is capable of generatingalarms & sending information in the form of alerts of theabnormality of a particular quality parameter or of an over-headed tank to the authenticated user using Zigbee, as thesensors detects the threshold level values. The data ismanipulated through Atmega microcontroller which transmitsthe data through Zigbee and also displayed onto the LCD. Thedevice is designed for domestic purpose with much importancegiven to cost & power consumption. This paper is an attempt toexplore basic physical & chemical sensors as per the WHOguidelines.

Keywords - Real time water quality monitoring, Wireless SensorNetwork, Physical and biological sensors, Machine intelligence,Remote water metering, Arduino

I . INTRODUCTION

There has been lots of research on development ofsensors for measuring the different water quality parameters.These sensors have been associated with a LCD displaydevice to display the measured value. This can only be usedas a stand-alone device for measuring the parameter on-site.There has been a great progress in the wirelesscommunication research to develop wireless sensor zoneusing the sensor node comprising of sensor board, processorboard and transceiver. To the best of our knowledge, there isno such indigenous technology which integrates the sensorsfor water quality monitoring and the WSN. There is still alack of indigenous real-time integrated drinking waterquality monitoring system available indigenously which canmonitor and control the quality of drinking watercontinuously in different places, transmit data to the centralserver via the wireless communication and generate alarms.This proposed research paper refers to technologies in themove towards next-generation water quality monitoring. Thesystem shall provide a simple and efficient means to detectand analyse the water quality regularly and automatically. Itis based on low cost wireless network for remote sensormonitoring the quality of water from diverse features, withless manual intervention, and heavy expenses. The proposedsystem mainly deals with wireless communication, sensortechnology, database management system and artificialintelligence techniques. All these branches are well studiedand lots of literature and modules are available for reference.The proposed system can be used in different institutions,housing societies, apartments in a small scale The system

can also be scaled up to monitor the drinking water quality ina public water distribution system in a large scale.

A. Proposed Work

As per stated in the paper [1], Limitation of humanresource to do real-time monitoring and data collectionregarding this environmental issue will lead a major problemto fresh water supplies at coming decades. This paperdescribes an approach of Wireless Sensor Network (WSN)application to do real-time data collection at the fresh waterresources such as rivers, lakes or wetlands areas to obtainproper water quality parameters measuring. The WSNsystem is used as a platform to monitor the fresh waterquality readings, deployed at distributed location which eachnodes will able to interface with various water qualitysensors. This proposed system is powered by PIC16F886(4/35 I/O pins, 20 MHz oscillator/clock input, operatingvoltage range 2.0V-5.5V) nano-watt MCU, with RF XBEE802.15.4, ISM 2.4 GHz module for each node while theCoordinator device integrated with GSM/GPRS modem andmonitoring LCD.[1]

B. Problem Definition

As the sensor nodes along with all the sensors/probes,different electronic circuits are supposed to be in remotelocations and thus are battery powered and operated, a smartpower system and networking activities are mostly taken intoconsideration for WSN platform design to solve this criticalissue as every battery has a limited life and are also noteasily replaceable due to remote an inaccessible locations,energy optimization is an big issue. Energy optimization canbe dealt by optimizing the electronic circuit design, sleepscheduling of the nodes, time synchronization forcommunication etc. Different sensors has different samplingrate. A standard data packet has to be formed before the datacan be sent to the base station. One way of addressing thisissue is to consider Lowest common multiple (LCM) of allsampling rates to collect the data, form the packet and sendto base station. But then packet size is also an issue whichneeds to be considered. The problem with the basic physicaland chemical sensors is that it cannot detect the pathogensand microbiological components in the water. This paper isto explore the possibilities of using soft sensor (softwarewhere several measurements are processed together) todetect pathogens (infectious agents). The contribution to thiswork is to study and develop a suitable, intelligent andadaptive decision support system for monitoring both thequantity and quality of water in different water bodies andwater flow. This paper aims to identify source of water

International Journal of Electronics, Communication & Soft Computing Science and EngineeringISSN: 2277-9477, Volume 3, Issue 6

2

pollution, to detect water leakage, to meter the water supplyremotely with the help of a wireless sensor network zone.

C. Objectives

The paper aims to build an autonomous real timedevice to measure the physical and chemical parameters(PH, Temperature, Dissolved Oxygen, Conductivity,.) ofwater using Arduino Atmega 2560 Microcontroller (54digital input/output pins (of which 15 can be used asPWM outputs), 16 analogue inputs, 4 UARTs (hardwareserial ports), a 16 MHz crystal oscillator, a USBconnection, a power jack, an ICSP (In circuit SerialProgramming) header, and a reset button).[7] The sensorsused are capable of capturing the analogue data and sendsthem to the ADC, then the data is manipulated throughAtmega microcontroller which transmits the data throughserial communication to the Zigbee modem then, thesensor output data is displayed onto the LCD. This paperis an attempt to explore the sensor probes thatcontinuously measure the basic physical, chemical andbiological sensors such as pH, Dissolved Oxygen (DO),Conductivity, that generate alerts in terms of mails/messages/ alarms following violation of safety norms asper WHO standard or according to the local sensitivity.In addition, Soft sensor concept (software where severalmeasurements are processed together) is related tomonitor water quality. The expected outcome shall be anew prototype model for on-line, real-time water qualitymonitoring system in a predefined wireless sensor zone.This paper aims to provide simple, efficient, costeffective and socially acceptable means to detect andanalyze water bodies and distribution regularly andautomatically. This paper aims to design wireless sensornetwork zone architecture for drinking water qualitymonitoring. This paper aims to design and developwireless sensing hardware for collecting water qualityparameters like pH, DO, Temperature and pathogens likeAlgal toxins (Cyanobacteria) etc. This paper aims todesign wireless sensor network zone architecture fordrinking water flow and quality monitoring. This paperaims to develop the interface modules (both hardwareand software) for the wireless sensor nodes and probes.This paper aims to develop an expert system for adaptivesetting up of new bench mark for water quality ofdrinking water and other usage.

II. SYSTEM OVERVIEW

In the proposed architecture, each water reservoir will beattached with a sensor node equipped with a set of sensorprobes capable of measuring the parameters like pH,Turbidity etc. According to the specifications of the sensorprobes and the processor board of the sensor the signalconditioning circuit will be designed to generate the sensoroutput to the processor board through Analog to Digital

Converter. The processor board processes the data accordingto the quality specifications and transmits to the centralserver through the transceiver. The measured data in each ofthe reservoir shall be sent to the central server through therespective transceivers either directly or indirectly throughother sensor or repeater nodes.

Figure 1: Transmitter Section

Figure 2: Receiver Section

Table 1: Standard specifications for drinking water[10]

SERIALNO.

PARA-METER

S

DESIRABLE RANGE

FORDRINKING

WATER

MAXIMUM RANGE

FORDRINKING WATER

UNITS

1. Turbidity 5 10 NTU

2. pH Value 6.5 - 8.5 - -

3. Totalhardness(CaCO3)

300 600 mg/l

International Journal of Electronics, Communication & Soft Computing Science and EngineeringISSN: 2277-9477, Volume 3, Issue 6

3

4. Iron 0.3 1.0 mg/l

5. Chlorides 250 1000 mg/l

6. Dissolved Solids

500 2000 mg/l

7. Calcium 75 200 mg/l

8. Copper 0.05 1.5 mg/l

9. Manganese

0.1 0.3 mg/l

10. Sulphate 200 400 mg/l

11. Nitrate 50 - mg/l

12. Fluoride 1.0 1.5 mg/l

13. Mercury 0.001 - mg/l

14. Cadmium 0.01 - mg/l

15. Selenium 0.01 - mg/l

16. Arsenic 0.05 - mg/l

III. PROCESS FLOW

Figure 3: Process flow

IV. PLATFORM DEVELOPMENT

Since this project is at preliminary stage, a completewater quality experiments are not carried out yet. Thetemperature of hot and cold water using LM 35 is tested(Figure 4) and the Zigbee communication has beenperformed. (Figure 5) Interfacing circuits for signalconditioning of Sensors are generated in Multisim 13.0

Figure 4: Arduino interface with LCD, Keypad andZigbee

Figure 5: Arduino interface with LM 35

A. pH Sensor

Signal from a pH probe has a typical resistance between

International Journal of Electronics, Communication & Soft Computing Science and EngineeringISSN: 2277-9477, Volume 3, Issue 6

4

10MΩ and 1000MΩ. Because of this high value, it is veryimportance that the amplifier input currents be as small aspossible. So here we are using LMC6001 having less than25fA input current is an ideal choice for this application.The theoretical output of the standard Ag/AgCl pH probe is59.16 mV/pH at 25ºC with 0V with pH of 7.00. This outputis proportional to absolute temperature. To compensate forthis, a temperature compensating resistor, R1, is placed inthe feedback loop. This cancels the temperaturedependence of the probe.The LMC6001 amplifies the probe output providing ascaled voltage of +/- 100mV/pH from a pH of 7. Thesecond op amp, TL072 provides phase inversion and offsetso that the output is directly proportional to pH, over the fullrange of the probe. Total current consumption will be about1mA for the whole system.

@pH 7 output of probe is 0 volts, @pH<7 voltageis positive, @pH>7 voltage is negative

Total pH range is 0(strong acid) to 14(strong base) If probe generates -59mV/pH unit then our

effective range would be +/- 7*59mV or +/-.414volts

Depending on temperature the voltage generatedper pH unit varies from -54mV@0c to -74mV@100c

Table 2: pH Properties

pH Properties Range

pH value fordrinking water

6.5-8.5pH

Operatingtemperature

20-100(Celcius)

OperatingVoltage

2.5-5.5Volts

Pressure range 0-100Pounds/squareinch gauge

In order to build an adequate amplifier there are afew considerations other than those pointed out by the idealprobe section. One consideration is the very high impedancethat a pH probe has. Not only are the probes very highimpedance they also are susceptible to noise, and the inputstage is very vulnerable to drift/offset characteristics of theamplifiers used to interface the probe. There are a lot of Op-amps that can be chosen for the job, don’t just look at theInput Impedance of the Op-Amp.A typical probe has an impedance of anywhere between50MΩ and 500MΩ, and since 100MΩ*1nA=.1v evenhaving a single stray nano amp can throw our measurementoff by almost 2 entire ph units. The goal then is to choose anop amp that is adequate enough that will not load down theprobe but that also has characteristics which will keep both

the cost down and the accuracy up.In order to normalize (Zero for 7pH) we will need to add anadjustable offset control to our simple circuit. Since the pHprobe should produce 0v at pH7 the gain portion of thecircuit will not affect this reading, which is why we adjust itin the offset portion. When combined with gain adjustmentswe can make a simple circuit that is able to be calibratedmuch like most commercial pH units.In the most basic sense the gain controls our slope (think theabove graphs) and the offset zeros it around 7. First find therange of the difference between the 4 and 7 calibrationreadings then by dividing the voltage difference by our gain;we get our voltage swing between the two solutions. Fromthe ideal pH probe scenario we know that a pH probe shouldproduce about 59mV per pH unit, by dividing our observedrange by this step number we can see if our amp is workingcorrectly. Once we can check what condition our probe isin, even with a cheaper Op-Amp on breadboard, we canachieve very good results for little investment.

Figure 6: pH sensor Signal Conditioning Circuit

Figure 7: pH sensor practical lab setup

International Journal of Electronics, Communication & Soft Computing Science and EngineeringISSN: 2277-9477, Volume 3, Issue 6

5

Figure 8: pH Buffer solutions

B. Turbidimeter

Figure 9: Turbitimeter Practical Setup withArduino

An open-source affordable turbidimeter based on a light-to-frequency sensor is built using off-the-shelf electroniccomponents. The principal housing components are two-part case and a cylindrical cuvette holder. The cuvetteholder houses a near-infrared (860 nm) light emitting diode(LED) and a light-to-frequency sensor, placed 90 degreesapart in a ―single-beam design. The light-to-frequencysensor outputs an electrical pulse train with frequencycorresponding to the intensity of detected light. A single-beam turbidimeter contains:

a light source that is directed through a liquidsample

a chamber to hold the liquid sample and one photodetector (a light-to-frequency sensor

TSL235)

Figure 10: Single Beam Turbidimeter[10]

TSL 235 Technical details;

• It consists of a photodiode and a current tofrequency convertor.

• Supplied voltage is 6.5volts.• Output frequency is 200-300kHz.• The device is temperature compensated.• Ultraviolet to visible light range is 300 to 700nm.• It can communicate with the microcontroller

directly because of its TTL logic.• The generated output from light to frequency

sensor is a square wave.

C. Fluid Level SensorUltrasonic sensors (also known as transceivers when theyboth send and receive, but more generally calledtransducers) work on a principle similar to radar or sonar,which evaluate attributes of a target by interpreting theechoes from radio or sound waves respectively.[13] Thistechnology can be used for measuring wind speed anddirection (anemometer), tank or channel level, and speedthrough air or water. To measure tank or channel level, thesensor measures the distance to the surface of the fluid.Technical details;

• Working Voltage is 5 Volts DC.• Working Current is15mA.• Working Frequency is 40kHz.• Max Range is 4m.• Min Range is 2cm.

•

Figure 11: Practical Setup of interfacing UltrasonicSensor to check the fluid level

Photodetector

Light Source

International Journal of Electronics, Communication & Soft Computing Science and EngineeringISSN: 2277-9477, Volume 3, Issue 6

6

V. RESULTS5.1 pH Results:

Figure 12: pH Versus Standard Deviation

Figure 13: pH Versus Variance

Figure 14: pH versus Voltage

Figure 15: pH Buffer versus pH measured

Figure 16: Voltage versus Temperature

Figure 17: pH measured versus Temperature

5.2 Turbidity Results:

Figure 18: Turbidity versus pH

International Journal of Electronics, Communication & Soft Computing Science and EngineeringISSN: 2277-9477, Volume 3, Issue 6

7

Figure 21: Turbidity versus Frequency

Figure 22: Frequency versus pH

Figure 23: Turbidity versus Standard Deviation

Figure 24: Turbidity versus Variance

VI. CONCLUSIONS

The sensor node observations are not error free.Sensors like pH are having lot of errors in observation.Parameters like pH are dependent on the temperature of thewater. The same water gives different pH values at differenttemperatures. A temperature compensation circuit tocompensate the pH value with regard to temperature of thewater is developed. With GSM module it is observed that itis having difficulty in its antennas, the SIM card can bestolen. Multiple SIM cards with the same number can’t beused to monitor from various stations.

VII. FUTURE WORK

In order to smoothen the observation (avoid errors),moving window technique may be followed like taking theaverage of 10 observations in each sample. The first

observation can be taken as average of observation 1 to 10;second observation can be taken as the average ofobservation number 2 to 11 and so on. Resolving the energyefficiency and power control by Energy efficient protocols,conditioning circuits, compensation circuits, node clusteringalgorithms, data aggregation schemes has to made. DASH7Module can be used to cover large areas for communication.To increase the number of sensors MUX can be used. Thesignal conditioning circuits or RTD, DO, CONDCTIVITYcan be implemented by the Multisim schematics developed.

Figure 25: Resistance Temperature Detector SignalConditioning Circuit

Figure 26: Conductivity Sensor Signal ConditioningCircuit

International Journal of Electronics, Communication & Soft Computing Science and EngineeringISSN: 2277-9477, Volume 3, Issue 6

8

Figure 27: Dissolved Oxygen Sensor Signal ConditioningCircuit

VIII. ACKNOWLEDGEMENT

The author acknowledges deep sense of gratitude toDr. Samrat L. Sabat, Associate Professor, School ofPhysics, University of Hyderabad, for his guidance,valuable suggestions, constant encouragement and support.The author takes this opportunity to thank Dr. Siba K.Udgata, Director, Centre for Modeling, Simulation &Design, University of Hyderabad for his valuablesuggestions during the work.The author would like to express special thanks to Mrs. K.Bramaramba, Assistant Professor, Electronics andCommunication Engineering, Stanley college ofEngineering and Technology for Women, Hyderabad, Indiafor her guidance and constant encouragement.Finally, the author would like to thank the Institution andthe faculty members without whom this work would be adistant reality. The author also extends her heartfelt thanksto the research scholars and all the helpful people atUniversity of Hyderabad. I would like to thank Arduino forproviding IDE platform to work upon.I wish to express my sincere gratitude to my parents fortheir unconditional support and encouragement throughoutmy life.

REFERENCES

[1] Muhammad Azwan Nasirudin, Ummi Nurulhaiza Za’bah & O.Sidek,”Fresh Water Real-Time Monitoring System Based on Wireless SensorNetwork and GSM” 2011 IEEE Conference on Open Systems-ICOS,Malaysia, 2011.[2] R. Balaji, R. Ganesan, “Remote Water Pollution Monitoring SystemUsing GSM”, Proc. of the Intl. Conf. on Advances in Computer,Electronics and Electrical Engineering Editor In Chief Dr. R. K. Singh,Copyright © 2012 Universal Association of Computer and ElectronicsEngineers. All rights reserved. ISBN: 978-981-07-1847-3 doi:10.3850/978-981-07-1847-3 P0517[3] Xiping Yang, Keat G. Ong, William R. Dreschel, KefengZeng, CaseyS. Mungle and Craig A. Grimes, “Design of a Wireless Sensor Network forLong-term, In-Situ Monitoring of an Aqueous Environment” USA, 2002[4] Dong He, Li-Xin Zhang, “The Water Quality Monitoring System Basedon WSN”, 978-1-4577-1415-3/12/$26.00 ©2012 IEEE, China[5] Zulhani Rasin and Mohd Rizal Abdullah, “Water Quality MonitoringSystem Using Zigbee Based Wireless Sensor Network” Malaysia,International Journal of Engineering & Technology IJET-IJENS Vol: 09,No: 10 91410-7575 © December 2009[6] Christopher D. Kelley, Alexander Krolick, Logan Brunner, Alison

Burklund, Daniel Kahn, William P. Ball and Monroe Weber-Shirk, “AnAffordable Open-Source Turbidimeter”, Sensors 2014, Open AccessSensors, ISSN 1424-8220, 22 April 2014.[7] Arduino Cookbook, 2nd Edition by Michael Margolis[8] Programming Arduino Getting Started with Sketches 1st Edition(Paperback) by Simon Monk[9] Building Wireless Sensor Networks by Robert Faludi[10] WHO guidelines for drinking water[11] “An affordable open source Turbidimeter” Sensors 2014, ISSN 1424-8220.[12] "Human Agency, Environmental Drivers, and Western JuniperEstablishment During the Late Holocene", locatedon:http://libres.uncg.edu/ir/asu/f/Soule_Pete_2004_Human_agency.pdf[13] “Ultrasonic Sensor”, located on:

http://en.wikipedia.org/wiki/Ultrasonic_sensor[14] "Operational Guide", located on:http://www.gemswater.org/common/pdfs/op_guide_for_data_2005.pdf[15]"Coordinated Activation and Reporting for Energy-Efficient TargetIntrusion Detection, Tracking, and Reporting in Wireless SensorNetworks", located on:http://www.cis.umassd.edu/~vvokkarane/publications/car.pdf[16] "SiGe BiCMOS PAM-4 Clock and Data Recovery Circuit for High-Speed Serial Communications", locatedon:http://mountains.ece.umn.edu/~sobelman/papers/mthsieh_soc03.pdf[17] "Miniature Fiber Optic Pressure Sensors for Intervertebral DiscPressure Measurements in Rodents", located on:http://drum.lib.umd.edu/bitstream/1903/7394/1/umi-umd-4810.pdf[18] "Integrating Heterogeneous Wireless Technologies: A Cellular-Assisted Mobile Ad hoc Networks", locatedon:https://www.cerias.purdue.edu/assets/pdf/bibtex_archive/2004-106.pdf[19] www.arduino.cc

AUTHOR’S PROFILEN. NEHA BERIObtained Master of Engineering, E.C.E,Embedded Systems, Stanley College ofEngineering and Technology for Women,Hyderabad, Telangana, India, in the year 2014.Current Research focuses on EmbeddedSystems, Wireless Sensor Networks, Waterquality parameters, Arduino interfacing andRemote monitoring.Research work for this paper was conducted atUniversity of Hyderabad, School of Physics.Email id: nehaberi@outlook.com