KateGleasonCollege of Engineering
Rochester Institute of Technology
Rochester, New York14623
Project Number: 08427
LED LIGHTING SYSTEM FOR DEVELOPING NATIONS: LIGHTING A THIRD OF THE WORLD
Ian J. Frank/Project Manager / Matthew Walter/Chief EngineerJesse Steiner/Electrical Engineer / Nicholas Balducci/Mechanical Engineer
Mike Celentano/Electrical Engineer / Lucas Spencer/Industrial & Systems Engineer
Copyright © 2009 Rochester Institute of Technology
Proceedings of the Multi-Disciplinary Senior Design ConferencePage 1
Abstract
In addition to providing clean, reliable, high-quality lighting at an affordable price with a design that can be built in the target nations, the Light Emitting Diode(LED) lighting systems project seeks to establish working relationships with developmental organizations with focus in developing nations. The project team selected a two module concept. The first module is a low-cost lighting unit that can be charged by the second module – a communal power station. The power module is designed to be used as a micro-business where the owner would charge customers a fee to recharge their lighting modules. The end result of the project is a working relationship with two non-profit organizations with a focus in Haiti. Additionally, the team was successful in designing the base technology for the proposed lighting solution. The current revision of the design meets the majority of the engineering specifications set to guide the team throughout the design process. Potential improvements have been identified for rectifying deficiencies in both modules and are suggested for future iterations of the project.
introduction
Although citizens of developed nations have come to take electricity and clean lighting for granted, according to the International Energy Agency (IEA), nearly two billion people around the world lack access to electricity. As of the year 2000, the IEA estimates that 14% of urban households and 49% of rural households in developing nations are without electricity [1]. In addition, many of the clinics, schools, workplaces, and markets in these parts of the world are completely without light. In most developing regions of the world, the population is growing significantly faster than the ability to provide electricity to them. In those areas without electricity, lighting is currently provided by fuel-based lamps which are inefficient, costly to operate, and produce large amounts of potentially harmful byproducts and green house gas emissions [2]. Fuel-based lighting accounts for 33% of all residential lighting and 12% of all lighting [3]. These fuels include kerosene, diesel, propane, biomass, candles, and yak butter[2, 4]. In addition to the aforementioned portions of the population, there are 35 million people in the world living in refugee camps, most of whom are without any lighting at all [1].
Lighting is one of the principal sources of power consumption. There is a great deal of disparity in the efficiency of these systems. While some may be able to obtain efficiencies of about 100 lumens per watt, there are many who are forced to use systems that operate well below one lumen per watt [5]. Additionally, it is important to note that these less efficient light sources provide less light and produce a less uniform dispersion than the more efficient systems. For example, a candle provides about 10 lux (lumens/m2) on a workspace if at a distance of 1 foot from the source, while typical on-workspace levels found in industrialized countries range from 400 to 500 lux, where this is accomplished with either desk lamps or ceiling lighting. It is estimated that fuel-burning lighting systems are the primary, if not only, source of light for 25% of the world’s population. These applications consume some 17% of the world lighting energy budget, but produce only about one tenth of one percent of end-use lighting. Thus, the costs of lighting for a family in a developing nation are comparable to a family in a fully-powered, industrialized nation [4]. The bottom line is that end-users pay a great deal less for a better end product when using electricity rather than burning fuels for their lighting needs. Additionally, the inefficient and low-quality nature of fuel-based lighting makes it very difficult to use for such applications as working, reading, or cooking, while also posing a health and fire risk.
Although the efficiency of these lighting systems has been examined quite thoroughly, the impact to the world as a whole has only recently come under consideration. As stated by Evan Mills, it is estimated that fuel-based lighting accounts for about 20 billion gallons of fuel annually at a cost of some 38 billion U.S. Dollars [3, 6]. In addition to the economic hurdles presented by fuel-based lighting, there is a heavy environmental burden as well. The byproducts of the combustion of these fuels are numerous and vast in nature. A single kerosene lamp will produce a sooty residue which will adhere to all surfaces of the area in which is it used, and also produce more than 200 pounds of carbon dioxide each year. This fact, in conjunction with the number of fuel-burning lighting applications, accounts for more than 200 million tons of greenhouse gas emissions into the atmosphere every year. The use of biomass as a lighting, cooking, and heating fuel also leads to a great deal of deforestation as noted by the International Energy Agency [1, 2, 4].
In order to understand the impact to the human element, we must first look at the added burden to the portion of the population who access light through the use of fuels. Depending on the fuel, between thirty minutes and seven hours a day may be spent in the pursuit of fuel for domestic use. This task is generally performed by the women and children of the household using time which could be better spent working or learning. Electric lighting allows families to read without having to strain or deal with the sound and smell from the kerosene lamps. Artificial lighting can also extend the potential workday so that family members may engage in additional income generating activities in the evening [2, 4]. There has been a great deal of research done to show that there is a positive relationship between increased educational opportunities made possible by electric lighting, resulting in higher lifetime earnings.
There are also significant health benefits and improvements to public safety. The byproducts of combustion are hazardous to the health of those exposed. Lack of proper ventilation in the areas of use allows for the buildup of carbon monoxide to levels many times higher than the World Health Organization (WHO) standards. It has been shown that, due to their increased exposure, women and children suffer most from these emissions. The WHO estimates that 2.5 million premature deaths of women and children may be attributed to breathing the fumes from burning biomass for domestic use [7].
Modernization and electrification are spreading to all reaches of the planet, but they are unable to keep up with the ever increasing demand. In many developing communities it is not feasible to connect these rural areas to the power grid, and power companies are hesitant at best to supply power due to the high risk of power theft [8, 9]. Thus, a self contained lighting system would seem to be the solution to the current problem. White LED’s have recently undergone a great technological advance and can now reach efficiencies of up to 100 lumens/watt, as opposed to 0.1 lumens/watt for the average flame-based lantern. Additionally, a 1-watt, white LED requires 20% of the electricity needed to power the most efficient and smallest compact fluorescent light on the market [4]. LED’s are also significantly more robust and have a factor of five to ten times longer lifetime than fluorescent lighting. These facts indicate that a self-sustaining lighting system centered on the use of white LED’s to provide illumination may be an answer to the developing world’s lighting woes [10].
The primary project goal is to develop and deploy efficient, high quality, and economically viable lighting systems for use in the developing world based on recent advances in LED technology. Furthermore, the proposed system should be designed such that it may be manufactured – at least in part – in the developing world. Additional goals set for this generation of the LED lighting system for developing nations were three fold. First, it was necessary to form contacts with sponsoring organizations with on-the-ground connections in the developing world. Secondly, the team strove to meet the requirements of their primary sponsor by preparing an alpha prototype and presentation for the National Sustainable Design Expo on the National Mall. Finally, the alpha prototype was to be tested and direction for future projects in this family was to be set.
Design process
As a unified group, the team worked to establish a set of customer needs. These needs were initially produced in team brainstorming sessions as it was very difficult to access the end user of the product. However once the first set of needs had been established, the team worked with two nonprofit organizations with representatives in Haiti – Sustainable Organic Integrated Livelihoods (SOIL) and H.O.P.E. – to refine these needs. Through these organizations, the team was able to administer a lighting use survey which helped to determine what the actual lighting need was and what was desired from an “improved” lighting solution. Fifteen needs were identified to create an effective lighting system that would prove to be more economical, environment, and health friendly. The most important of these needs were to: provide improved lighting levels and distribution, be more environmentally friendly than the kerosene lamps now in use, decrease the purchase and operating costs of the lighting system, and have the ability to be manufactured in developing nations.
From the established customer needs the team developed twenty-five engineering specifications or metrics. These metrics provided a manner in which to quantitatively assess how well the end product met the needs of the customer. The specifications were used during the concept generation phase of the project. By utilizing the set specifications the team was able to determine the ability of the ten initial concepts to perform adequately. These ten concepts were narrowed down to two primary concepts for further review by the sponsoring organizations in Haiti as well as the RIT faculty in attendance at the system-level design review. Of the two concepts show in Figure 1 and Figure 2, it was determined by both reviewing bodies as well as the team that the community bicycle charging system better met the customers’ needs than the photovoltaic charging system. In both concepts the primary module was a rechargeable and recyclable lighting unit that would be charged on a regular basis by the second subsystem. Also in both cases, the charging stations would be located in a community-central location and would easily interface with the lighting modules.
Figure 1. Community Bike Charger
Once the final concept was selected, the team split up into individual areas of expertise and concentrated on the separate subsystems.
Figure 2. Community Solar Charger
Physical Design:
Light Module
The light module proved to be more particular than was originally thought by the team and as such required several iterations of customer and advisor feedback followed by periods of redesign before the final prototype configuration was selected. Once the team had decided upon the final design, which utilized a recycled can as the body of the module, additional work was done to determine the construction and composition of the remainder of the module. The current design not only makes use of “trash,” but is also compact and simple to construct. All design work was done using the Computer Aided Design(CAD) program SolidWorks. This program was used to aid in the design of the light module as well as create engineering drawings for manufacturing. The physical design of the module was used to dictate the physical constraints of the components to be designed or selected by other members of the team. The final design of the light module may be seen inFigure 3.
Figure 3. Light Module Exploded View
Basic thermal analysis was conducted on the heat sink plate to ensure that the system would not exceed the maximum operational temperature for the chosen LED. The system was modeled as a circular fin (with the fin area corrected for the fact that the plate isnot actually a complete disk).
(1)
Equation 1 was used to find the efficiency of the fin, where r1 and r2c are the inner (LED) and corrected outer radii, K and I represent modified Bessel Functions of the first and second kind, and C2 and m may be defined as follows.
(2)
(3)
Where h is the convective coefficient of the environment surrounding the fin, k is the conductive coefficient of the fin material (aluminum), and t is the thickness of the plate. After obtaining the sufficiency, the resistance and temperature drop across the fin were found using Equations 4 and 5.
(4)
(5)
Calculations for the thermal analysis were run using highly conservative values for the independent variables of the equation. This allowed the team to prove that even under the worst conditions, the heat sink would dissipate heat effectively and the LED would survive. Using an input of 2.5 Watts from the LED with a very low convective coefficient of 4W/(m2K), the temperature drop across the heat sink was calculated to be 106°C. Thus even with these very cautious values the module could run in a 45°C environment without concern of damage to the LED. Further numerical data was gathered to determine the actual temperature of the LED connection and will be discussed in the testing section of this paper.
Power Module
Unlike the light module, the power module was relatively straight-forward and could make excellent use of low-cost, off-the-shelf parts for the prototyping process. Special consideration was given to the mechanical interface of the bike and generator. In the final prototype design, the rear tire of the bike simply drives the resistance roller from the trainer, which in this case no longer serves to add resistance to one’s daily workout but rather to transmit the power of the userto the generator. The generator is attached coupled with the shaft of the friction roller and then is connected via electrical conduit to a power conditioning circuit in the power module housing. The overall final power module design is shown in Figure 4.
Figure 4. Power Module Design
Electrical Design:
Light Module
The design of the lighting module electrical circuit had two main points to consider, one was the need to create a constant current to flow through a single LED. This is due to the fact that an LED although very efficient, has an exponentially increasing current-to-voltage (I-V) characteristic curve, meaning that a very small change in either the voltage or current across the diode will result in a much greater increase in the current or voltage respectively. For this reason, circuitry is necessary to ensure that the current and voltage will not change much across the LED if at all. The other point to consider was that of allowing the circuit to be modified at the user level to allow the LED to shine at different brightness levels (low and high). This was thought to be useful for the customer since having the ability to use the lamp at two different brightness settings makesthe light module more versatile and power efficient (in the low setting). Once the LED was selected it was realized that it had a maximum forward current of 1A, and at this setting it would output approximately 100 lumens of light intensity.
It was then realized that the power source for the lighting module would be four Nickel-Metal Hydride (NiMH) cells connected in series giving anominal input voltage of 4.8V (VIN). From there it was necessary to find a way to drop down the 4.8V to approximately 3.0V while providing a constant 1A maximum current. For this a constant current buck converter was selected (Linear Technology P/N: LT3474). This chip was to be used as a buck converter to drive a high power LED. The chip also has the ability to dim the brightness of the LED using Pulse-Width-Modulation (PWM) or using a simple voltage divider. PWM was decided to be too complex and would require the use of many more components, an important factor since space inside the lighting module in limited. The chip’s data-sheet gives the following formula for calculating the LED current based on a varying voltage at the VADJ pin:
(6)
From this equation, a voltage divider was created using two different resistor values (one for low brightness and one for high brightness) coming from opposite poles of a switch going into the VADJ pin, and then a common resistor going from VADJ to ground (GND). The switch to be used is a double-pole single-throw (ON-OFF-ON) which will take VIN from the batteries and either leave it floating (in the OFF state) or bring it into the input pin of the chip (to turn the chip ON) and also across either one of the voltage dividers to create two different brightness settings. The only other components on the board are various diodes and resistors used for power switching/regulating, also various capacitors and inductors used to level out the power at certain points in the circuit (mainly at the output to provide constant current across the LED).