ilifeguard Post Mortem April 19, 2011 Dr. Andrew Rawicz School of Engineering Science Simon Fraser University Burnaby, British Columbia V5A 1S6

Transcription

1 April 19, 2011 Dr. Andrew Rawicz School of Engineering Science Simon Fraser University Burnaby, British Columbia V5A 1S6 Re: ENSC 440 Post Mortem report for ilifeguard Dear Dr. Rawicz, Please find enclosed a document entitled Post Mortem for ilifeguard. ilifeguard is an underwater communication system built to prevent the loss of lives due to drowning by alerting the lifeguards when a swimmer is in need of help. Children below the age of twelve and adults with medical conditions are more likely to face the risk of drowning. As such, they form the backbone of our target audience. Within this document we have discussed the current state of our system design, the major problems that we encountered during the implementation of the design, comparison of the actual design versus the functional specification, the future extension and finally, a personal description of the project experience by the members. GESS is comprised of five passionate and industrious engineers namely: Suleiman Mohamed, Mehdi Elahi, Elis Micka, Gurman Thind and myself. The product has an immense social benefit which we intend to service by providing a reliable, affordable and appealing life saving device. For any questions or concerns please contact us at ssk15@sfu.ca. Sincerely, Shivam Kishore Shivam Kishore Executive Manager GESS Inc.

2 22 nd January 2011 Proposal for Guardian Electronics System Solutions Executives: Shivam Kishore Chief Executive Officer Elis Micka Chief Design Engineer Suleiman Mohamed Chief Software Engineer Gurman Thind Chief Hardware Engineer Mehdi Elahi Chief Research Engineer Contact Person: Shivam Kishore 1 P age

3 Executive Summary According to lifesaving Society, a Canadian charity organization, drowning is the third leading cause of accidental death for people 60 years of age or older. Each year almost 500 Canadians die in water-related incidents. Red Cross reports that approximately 6% of drowning population in Canada are toddlers and infants. Though young children below the age 12 and adults 60 years or older face the greatest risk of drowning, people of all ages who have medical problems are also prone to the risk of drowning. Current swimming pools have lifeguards around the clock. However, due to the large number of swimmers, lifeguards have to divide their attention and may at times misunderstand a gesture for help. Making a device that alerts lifeguards the instant a swimmer needs help would tremendously reduce the risk of drowning and lessen the emotional burden experienced by lifeguards who would live through the guilty of not saving a swimmer under their watch. This document discusses the current state of an assistive device, which would alert the lifeguard when a person is drowning, and in need of immediate help called ilifeguard. This would be a three-component device, which through a relay of acoustic signal and radio frequency (RF) signal transmission would alert the lifeguard via an alarming mechanism. To determine if the person is drowning, the pressure at a particular depth and the amount of time the person spends at that depth are the parameters that are considered. The document further compares the working prototype of ilifeguard with the solutions proposed in the earlier documents and proposes various advancements that could make the product more marketable in the future. GESS inc. comprises of five fourth year Engineering Science students from Simon Fraser University. All the members are studying Electronics option but have diverse work and coop experiences, which provide an excellent background and expertise for the company. 2 P age

4 EXECUTIVE SUMMARY INTRODUCTION CURRENT STATE OF THE PRODUCT Bracelet Transceiver Receiver PROBLEMS ENCOUNTERED Bracelet Underwater Communication PCB Layout and Design Real Time Programming Waterproofing Radio Frequency Transceiver Design of the RF board: Microphone: Radio Frequency Receiver Design of the RF board and Frequency Alarming System BUDGET AND TIMELINE Timeline FUTURE IMPROVEMENTS PERSONAL REFLECTIONS 15 Shivam Kishore 15 Gurman Thind 15 Suleiman Mohamed 16 Mehdi Elah 16 Elis Micka CONCLUSION 17 3 P age

5 1.0 Introduction According to lifesaving Society, a Canadian charity organization, drowning is the third leading cause of accidental death for people 60 years of age or older. Each year almost 500 Canadians die in water-related incidents. Red Cross reports that approximately 6% of drowned populations in Canada are toddlers and infants. Although young children below the age 12 and adults 60 years or older face the greatest risk of drowning, people of all ages who have medical problems are also prone to the risk of drowning. Current swimming pools have lifeguards around the clock. However, due to the large number of swimmers, lifeguards have to divide their attention and may at times misunderstand a gesture for help. Making a device that alerts lifeguards the instant a swimmer needs help would tremendously reduce the risk of drowning and lessen the emotional burden experienced by lifeguards who would live through the guilt of not saving a swimmer under their watch. Over the past semester, our team has devoted the time and effort to make the concept of ilifeguard a reality. We faced many boulders on our way to the final stage, but managed to successfully develop the product according to the requirements and the standards that we proposed in the earlier documents. 4 P age

6 2.0 Current State of the Product As proposed in the project proposal, the ilifeguard saves the lives of drowning people in swimming pools. The swimmers need to wear the bracelet at all times when they are in the swimming pool. The bracelet itself is composed of a printed circuit board that has a microcontroller and it is connected to a pressure sensor and a buzzer. When the swimmers start drowning and they sink towards the bottom of the pool our system is designed such that when the swimmers go beyond the threshold depth the microcontroller will start a timer. The reason for the timer is because people like to dive deep in the water and hence we do not want to have a false alarm if someone is going at the bottom of the pool and they are fine. However, when the person is at a certain depth for an excessive amount of time then the timer expires and the microcontroller sets the buzzer on. The buzzer outputs a 5 KHz signal. This signal is picked up by our microphone which is connected to one of our radio frequency printed circuit boards that is lying by the side of the pool. When the microphone picks up the signal from the buzzer it alerts the RF printed circuit board which sets the siren and the large light emitting diode in the on mode for the lifeguards to be alerted. Furthermore, the RF printed circuit board also sends another message through means of radio frequency to our second RF printed circuit board that is by the side of the lifeguards. This way the lifeguards are guaranteed to be alerted and they can rush to the rescue of the drowning swimmer. Figure 2.1 below shows the complete overview of our system. Figure 2.1: System Overview 5 P age

7 2.1 Bracelet Currently the bracelet uses a microcontroller ATMEGA644V which is the brain and the powerhouse of the printed circuit board. This bracelet is powered up by a 3.6 lithium batter. Figure 2 below shows all the components of the bracelet. On the left side of the figure as mentioned is the printed circuit board with all the corresponding connections and the microcontroller. In the very middle of the figure we can observe the analog pressure sensor which provides the input to the printed circuit board and the microcontroller. When the input of this pressure sensor is greater than our certain threshold value the microcontroller starts the timer and eventually sets the buzzer pin in the on mode. The buzzer that we are using and which in this case is functioning as the output of the bracelet is the one that is shown in the right side of Figure 2.2 below. This buzzer functions best with a 12 volts battery and it outputs a 5 KHz acoustic signal. Figure 2.2: Bracelet PCB, Pressure Sensor, Buzzer 2.2 Transceiver Switching our attention to the receiving part of our system we analyze the first part which is the transceiver. Currently the transceiver uses a microcontroller ATMEGA644PV as the brain and powerhouse of the printed circuit board. The transceiver is composed of an acoustic receiver, a microphone and an RF printed circuit board capable of transmission and reception. Here we are using the transmission feature to communicate with the RF printed circuit board that the lifeguards hold. This part of our system is to be placed at the corner brim of the swimming pool. It takes as inputs the acoustic signal received by the microphone shown in Figure 2.3 below. 6 P age

8 Figure 2.3: Microphone When the input has been detected by the microcontroller of the RF printed circuit board then the board has two main functions of equal importance to execute. The first step is to set the pins of the siren and the big light emitting diode in the on modes. As it can be observed in Figure 2.4 below the executing times need to be in real time. The moment the RF board receives the input signal from the buzzer of the swimmer s bracelet through the microphone it needs to alarm the lifeguards right away of the situation so they can respond as fast as they can. Figure 2.4: Overall Transceiver System 7 P age

9 2.3 Receiver Lastly to complete the overall system we have the very last part of this communication which is the RF receiver board. This includes the radio frequency communication of the RF transceiver printed circuit board mentioned above and the RF receiver printed circuit board that is positioned by the side of the lifeguards. The receiver part of this overall system is explained in more detail regarding its components by Figure 2.5 below. As it can be observed this board is sending signals at 2.4 GHz centre frequency and its antenna is on the printed circuit board. It also used a microcontroller ATMEGA644PV which is the brains and powerhouse of this board. Figure 2.5: Receiver Board 8 P age

10 3.0 Problems Encountered 3.1 Bracelet The bracelet, which is responsible for transmitting acoustic signals comprises of a printed circuit board (PCB) mounted microcontroller along with the acoustic buzzer and the pressure sensor. There were several issues that we came across and are listed below: Underwater Communication Correct form of underwater communication between the bracelet and the transceiver was a very important component of our system. Since, none of the team members had any experience with any sort of underwater communication it proved to be very challenging. A lot of research, time and effort dictated the use of acoustic signals of high frequency for communication. Acoustic signals attenuate minimally underwater and have high signal to noise ratio underwater at high ratio and thus were the perfect and best fit for our system. Furthermore, choice of components like the pressure sensor and the buzzer was very crucial since they had to be waterproof. Finding the appropriate buzzer proved to be cumbersome and it took us quite a few tries before hitting on the right buzzer PCB Layout and Design The first major step in the design of the bracelet was to design the PCB, which would basically act as the skeleton for the other components. Since, the PCB had to be the size of an average watch dial; it was difficult to design the layout of the PCB with all the components mounted on it. Several schematics were designed before the final schematic could be approved. Furthermore, etching the PCB also involved great effort, since it was the first time experience for all the group members. Ultra violet exposure method was used to etch the design on the PCB. The process involves a serious of steps, requiring very precise exposure time and solution compositions, thus it took several tries and a lot of patience to get the perfectly etched PCB Real Time Programming The analog values from the pressure sensor had to be converted to digital values via the micro-controller before they could be used for depth measurements. These values had to be read and used real time without any propagation or delay. This posed to be a big challenge as there was always a delay to the processing of the values and one delay lead to another thus making the real time feature faulty. To deal with this issue, timers were incorporated in the code at appropriate places and multi-threading was used to process several tasks 9 P age

11 simultaneously. This approach greatly improved the real time reading and processing ability of the micro-controller Waterproofing Waterproofing the PCB was by-far the most challenging issue that we faced. The PCB had to be worn by the swimmer as a bracelet and hence it was of utmost importance to waterproof it completely. Our initial approaching of conformal coating with epoxy resulted in failure multiple times, reason being the nature of epoxy compound. Epoxy expands gradually on and after cooling and thus it entered the thin layer between the surface mounted micro-controller and the PCB causing the micro-controller pins to become nonconducting and thus resulting in the micro-controller becoming obsolete. We also tried the plastic gun, which basically formed a layer of thick plastic over the PCB. Again, it resulted in failure as there were small air holes through which water gradually made its way. In the end we ended with the PCB being placed outside the water with wires connecting the buzzer and the pressure sensor, which were placed inside the water. 3.2 Radio Frequency Transceiver The transceiver was responsible for receiving the acoustic signal transmitted by the bracelet and transmitting the RF signal to the receiver held by the lifeguard. During the design Several issues were encountered and are listed below: Design of the RF board: The RF board is a PCB which is capable of being a RF transmitter and a receiver. Again, to minimize the board size, the layout of the board was very carefully designed and it required a lot of trials to successfully design the perfect layout. The board has two parts, one having micro-controller as the center and another having the RF chip as the center. Both the parts have their own ground which separates them and enable them to function correctly as two separate entities but in cohesion. This fact was unknown to us and it took a tremendous amount of research to figure this property. Furthermore, the antenna used to transmit the signal is also etched on the PCB and in a folded dipolar configuration which makes it very compact and powerful. In order to use this feature, the RF chip had to carefully chosen and the corresponding circuit had to accurately designed on the board, which required a lot of research and trials Microphone: The choice of microphone was very important since it was responsible of receiving the acoustic signals transmitted by the bracelet. The microphone that we chose was a magnetic microphone and this did not add any impedance to the circuit. Also, magnetic microphones 10 P age

12 do not require any dc supply to function. This fact was also confirmed by the datasheet of the microphone. However, without a dc supply the microphone gave very inconsistent and inaccurate results. After performing many tests it was figured that the microphone needed a very minute offset voltage of about 100mV. This realization proved to be very valuable and the microphone then behaved consistently. 3.3 Radio Frequency Receiver The receiver was responsible for receiving the RF signal from the transceiver and enabling an alarming system comprised of sirens and LEDs to alert the lifeguard Design of the RF board and Frequency The RF board designed for the transceiver was also used for the receiver. This time around, its receiver functionality was used. It was very important to ensure that the RF transmission and reception did not interfere with the normal frequencies used for radio and cell phone communication. Thus a 2.4 GHz signal was used which conforms to the IISM standards and thus does not interfere with any in use frequencies Alarming System The choice of alarming system proved to be really difficult. It was important that it is effective and efficient along with being light and compact since the receiver was to be held by the lifeguard. It needed intensive research to figure out the system and in the end LED indicators along with high amplitude sirens were used as the alarming system. 11 P age

13 4.0 Budget and Timeline Figure 4.1 compares the budgeting compares the actual cost of prototyping vs. the estimated cost. Components Projected Prototype Cost Actual Cost Bracelet $195.9 $ Transceiver $ $ Receiver $ $ 354 Total $ 890 $ 995 Figure 4.1: Cost Comparison Looking at the comparison table it is evident that we were quite close to our estimated projected cost in making the prototype. The slight increase in cost was due to the unexpected breakage of the pressure sensor due improper handling and also due to the small extra components like clock crystals, capacitors and LEDs that we had to buy in excess. Furthermore, we had to buy additional batteries as backup which added to the additional cost increment. 4.1 Timeline Figures 4.1 and 4.2 shows the proposed timeline and actual timeline for our project respectively. Comparing the two charts we realized two major patterns. Firstly, our group spent less time for documentation than anticipated. But this time was compensated for by the design of each individual component which required much more time than we anticipated. Documentation was done throughout our project as shown in the actual timeline and part ordering was also continuous throughout the development of the project. 12 P age

14 Figure 4.1: Original Schedule Gantt chart Figure 4.2: Revised Schedule Gantt chart 13 Page

15 5.0 Future Improvements One of the major thoughts that we considered while developing the prototype model was of improving and making the system more efficient in the future. Since this product has very immediate and a massive market, one of the most important considerations is the cost. The cost of the entire system can be greatly reduced by mass production and choosing the components more carefully. Table 5.1 displays the approximate cost of the actual system when estimates of ten thousand units are produced in comparison to the actual prototype cost. Component Actual Cost Projected Unit cost Bracelet $ $ 50 Transceiver $ $ 70 Receiver $ 354 $ 74 Total $ $ 206 Figure 5.1: Estimated Actual Production Cost Another important step in making the product successful in the market is to make the components as small and appealing as possible. The size of the bracelet can be reduced threefold by using miniature surface mount components. Similarly size of the RF boards can also be reduced. Furthermore, better ways to waterproof the bracelet have to be used. Making waterproof PCBs is one option but further research needs to be done regarding the methods that can be used to make the waterproofing more efficient and effective. 14 P age

16 Due to the nature of this system, it has applications in a wide variety of places. They can range from controlled environments like swimming, private bath tubs and theme parks to uncontrolled environments like small ponds, lakes and even seas and oceans. Personalising the system depending upon the need and the environment is one area that has great potential and certainly can be a huge benefit in marketing and increasing the scope of the ilifeguard system. 6.0 Personal Reflections Shivam Kishore From this project I gained enormous experience with various developmental stages of the product design. Reflecting upon the project, I realize we should have spent more time on the research before actually going to the development phase. Also, we should have been more careful with ordering of parts and should have thoroughly researched our needs and requirements before ordering components. Working as a part of all the teams within the company I became familiar with various new hardware techniques like etching PCBs and software s like SoildWorks for 3-d design. Acting as the CEO of the company I learnt about effective team and resource management. Overall the combined effort and expertise of all the group members made the creation of ilifeguard possible and I enjoyed been and working as a part of this team. Gurman Thind Through undertaking ilifeguard system project, I had opportunity to work with group of enthusiastic and responsible student of engineering. While working on the project I have applied the knowledge I have attained from SFU engineering courses. My work involved debugging and running analog plus digital circuitry of the system. Most challenging part of 440 project was setting up acoustic communication under water. Leaning the basic principle of sonar communication and actually applying them in practice was most amazing. It all started from ordering components that work underwater and studying their behavior once applied under water for testing. Transmitting signal through the piezo electric transducer and detecting the type of acknowledgment on the receiving mode was a part of communicating process for our project with the aid of micro controller. I have also learnt the PCB eagle schematic designs and got the full knowledge of process of making an actual PCB board. Also I got the chance to enhance my software skills in C P age

17 Being a member of enthusiastic team we learnt that it was important not to take criticism personally and avoid giving any personal criticism unless it really affecting the project. Organizing weekly meeting and communicating effecting on project s success and failure was all part of team-work. Suleiman Mohamed While the skills I learned from my capstone project span from hardware to software to better ways of collaborating with people, none parallels the technical skills. These skills include designing parts and components using solid works, and programming in real time for various functions. In addition, I gained a great deal of knowledge in building electric circuits, making the circuits into schematics in Eagle and printing the circuits on printed circuit boards (PCBs). Though paying for PCBs to be printed is easier and time efficient, the etching option is more rewarding in terms of skills gained and cheaper. Above all, except a few skirmishes the product schedule went on time, each member did their part and as a result the project was a success. Mehdi Elahi As a research engineer my main task was to ensure that the team had a firm background before they started on development of any component. This lead to extensive research about various components and techniques which greatly increased my knowledge level. Furthermore, having some prior experience with PCBs, I acted like a guide for the team to etch and design various PCBs for the project. I also worked with programming the microcontrollers, which helped me improve my software skills and gave me great insight into real time programming. I also improved my people skills, due to constant interaction involved within and outside the team during the course of the project. On the whole it was a great idea and project and I thoroughly enjoyed working with the team. 16 P age

18 Elis Micka Working as the design engineer for GESS inc, I worked on the mechanical and design aspects of the components. It involved a magnitude and variety of tasks like etching of PCBs, drilling holes into the PCBs, cutting and packaging the RF boards and the bracelet. I also worked on making the various switching circuits that were needed to ensure that accurate voltage is supplied to the components. I was rigorously involved in the integration of the system and also in the testing phases. I believe that we lacked some amount of initial research and we hurried at times in to the development stages. This did prove to a problem in some stages of the development, but was sorted out with the effor and patience of the team. I greatly enjoyed working with this team and I am glad we were able to finish what we aimed for in the beginning. 7.0 Conclusion This document outlines the final progress towards the making of ilifeguard. It also compares the claims made in the functional specification to the actual working prototype. The problems that were encountered are also described along with the solutions that were used to resolve the problems. This document concludes with the analysis of the project from each members point of view. The past semester has been a very challenging and yet a rewarding experience for the GESS team. We applied most of the knowledge and skills acquired in our young engineering careers to date, as well as patience and determination, to research, design, implement, test, and document hardware and software modules to achieve a working model of ilifeguard. 17 P age