A COMPREHENSIVE, ANALYTICAL STUDY OF LOW POWERED COOLING IN HOT DESERT CLIMATES. Jennifer L. Borofka Clare T. Caron Kelly M. Kanz Stefan A.

Transcription

1 A COMPREHENSIVE, ANALYTICAL STUDY OF LOW POWERED COOLING IN HOT DESERT CLIMATES Jennifer L. Borofka Clare T. Caron Kelly M. Kanz Stefan A. Yanovsky University of St. Thomas School of Engineering St. Paul, Minnesota May,

2 ABSTRACT We have designed and tested a cooling system that will be implemented in Community Learning and Information Centers (CLICs) in Mali, West Africa. The project was approached from three perspectives: revising the existing evaporative cooler to decrease its power consumption; connecting the evaporative cooler to a geothermal duct that could pre-cool the incoming air; and add strategic shading to minimize the solar heat load on the building. All three approaches were experimentally verified on location in Mali and at UST. Continuous measurements of the roof and room temperature, air humidity, and air velocities over 5 days led to a final design that will decrease ambient air temperatures by 30 F. The final design will use only 146 Watts, consume 0.5 GPH of water and stay below the 50% relative humidity target. The entire system will cost within the $1000 budget set by USAID. It will be implemented in Ouéléssébougou, Mali this summer by USAID and the Geekcorps, a corps of information technology volunteers. 2

3 TABLE OF CONTENTS Title Page I. Contributions 1 II. Glossary 2 III. Mission Statement 3 IV. Problem Definition 4 5 Customer Needs 4 Product Design Specification 4 5 V. Concept Design 5 7 Concept Generation Process 5 Concept Alternatives 5 Details of Concept Alternatives 5 7 Selection Processes 7 Final Concepts and Rationale 7 VI. Preliminary Prototype Calculations 7 8 VII. Testing and Experimentation 8 13 Barrier Materials 8 9 Barrier Placement 9 Kangaba CLIC Overview 9 10 Shading Telephonic 11 Ground Temperature Geothermal Piping 12 Geothermal Adapter VIII. Data Analysis and Results Shading Telephonic 14 Ground Temperature 14 Relative Humidity 14 Geothermal Piping Conclusion 15 IX. Final Design Description Ouéléssébougou CLIC Overview Modified Evaporative Cooler Evaporative Cooler Fan Geothermal Piping

4 Placement of Piping and Cooler Shading Barrier X. Project Management Team Member Assignments Work Breakdown Structure Team Dynamics 26 XI. Budget XII. Conclusions and Recommendations XIII. References 30 XIV. Appendices Appendix A Preliminary Prototype Calculations Appendix B-F Hand Calculations (available on request) Appendix G Quality Function Deployment Appendix H Mali Packing List Appendix I Mali Testing Procedures Appendix J Kangaba CLIC Diary Appendix K Kangaba CLIC Details Appendix L Ground Temperatures Appendix M Relative Humidity Appendix N Geothermal Pipe Temperatures Appendix O Telephonic Temperatures Appendix P Air Velocity Appendix Q Other Temperatures Appendix R Kangaba CLIC Outside Air Temperature Appendix S Kangaba CLIC Roof Temperatures Appendix T Kangaba CLIC Ceiling Temperatures Appendix U Kangaba CLIC Room Temperatures Appendix V Ouéléssébougou CLIC Details Appendix W Ouéléssébougou CLIC Diary Appendix X Preliminary Project Management Appendix Y Work Breakdown Structure/Gantt Chart Appendix Z List of Materials Appendix AA Implementation Proposal i lxxviii i xxxiii xxxiv xxvi xlii xliv xlv xlvi xlvii xlviii xlix l li lii liii liv lv lvi lvii lviii lxi lxii 4

5 I. CONTRIBUTIONS Abstract By Jennifer L. Borofka Problem Definition By Clare T. Caron Concept Design By Kelly M. Kanz Prototype Design Description By Stefan A. Yanovsky Testing and Experimentation By Clare T. Caron, Kelly M. Kanz, and Stefan A. Yanovsky Data Analysis and Results By Clare T. Caron and Kelly M. Kanz Final Design By Jennifer L. Borofka, Clare T. Caron, Kelly M. Kanz, and Stefan A. Yanovsky Project Management By Jennifer L. Borofka Budget By Stefan A. Yanovsky Conclusions and Recommendations By Jennifer L. Borofka and Kelly M. Kanz Edited by Jennifer L. Borofka, Clare T. Caron, Kelly M. Kanz and Stefan A. Yanovsky 1

6 II. GLOSSARY Geothermal Cooling The process of running air underground to cool it prior to entering a building Relative Humidity The ratio of the amount of water in the air at a given temperature to the maximum amount it could hold at that temperature, expressed as a percentage Evaporative Cooler Nature's most efficient means of cooling is through the evaporation of water. A simple example is the cooling you feel when stepping out of a swimming poolthe water evaporates quickly from your body, taking heat with it. Evaporative coolers do essentially the same thing by drawing hot air through wet pads using a fan or blower. As water evaporates from the pads, it takes heat from the air with it, resulting in up to 30 degree cooler air being discharged from the cooler. [1] Psychrometrics The study of the physical and thermodynamic properties of air-water vapor mixtures 2

7 III. MISSION STATEMENT Our goal is to design a cooling system for Community Learning and Information Centers (CLICs) in Mali, Africa that minimizes energy usage and can be made cost effectively in Africa. 3

8 IV. PROBLEM DEFINITION Our goal is to design a cooling system to be used in Community Learning and Information Centers (CLICs) in Mali, Africa. The problem is that they are currently using evaporative coolers that run on an unreliable energy source, and the units are purchased and shipped from the United States which makes them expensive. A cooling system that can be manufactured locally to reduce the cost and also a system that might reduce the energy consumption are the most desirable requirements. Currently, there are 13 CLICs in Mali, with plans to build more in the future. They are small, single room buildings that are approximately 20 x 20 x 12. Each CLIC is different, but the average building contains about 8 personal computers with CRT screens along with a television, VCR, DVD player, radio, and LAN modem hub. They are constructed with 15 cm thick cement walls, and most have reinforced concrete roofs, while three have corrugated metal roofs. The interior walls and floors are tiled. At any time there may be up to 18 people in the room. Mali is a landlocked country located in West Africa. It has three main seasons which vary with latitude. The rainy season lasts from June through October. The cooler, dry season runs from October to February, and the hot, dry season runs from February to June. Our design addresses cooling during the hot, dry season. Monthly average high temperatures during this season range from 93 to 102 degrees Fahrenheit [2] and there is virtually no rainfall. Customer Needs Our main contact is Ruth MacDonald, who has worked with the Peace Corps in Mali, and remains in contact with many people who work directly with the CLICs. After our initial meetings, we established the following customer requirements: Little to no energy consumption Cool a one-room building Design should be able to operate in a very hot, dry and dusty environment It should be able to be manufactured in Mali with locally available materials Simple to use and maintain Keep material costs as low as possible (under 1000 USD) Ruth emphasized that the most important requirements were cost and energy efficiency. Product Design Specification We first examined the data sheets of the evaporative cooler that is currently being used in the CLICs. We determined the power consumption of this device which we used to set goals for our design. It uses 4.9 Amps at 120 Volts [1], so any reduction of this would be an improvement. We then conducted research on the energy load of the people, equipment, and environment on the CLICs. The load created by the computers can be up 4

9 to 1600 Watts (200 Watts per computer). The maximum load created by people is 2700 Watts (150 Watts per person). The load created by the environment is from solar radiation and is estimated to be 915 Watts per square meter on the roof. The dust content in the air ranges from 26 micrograms per cubic meter during an average day to 13,735 micrograms per cubic meter during a dust storm. Dust particle size ranges from 44 to 50 micrometers in diameter. During the dry season in Mali, the relative humidity is around 10%. By increasing the relative humidity in the room, the temperature decreases. We calculated that the relative humidity can be increased to 50% at 75 degrees Fahrenheit and still remain in the human comfort zone. To ensure that our design would remain low-cost, we researched what materials were available in Mali, and found that the materials were mainly limited to local blacksmiths and metal-workers. The Quality Function Deployment chart in Appendix G illustrates the relationship between the customer requirements and the engineering specifications. V. CONCEPT DESIGN Concept Generation Process In order to satisfy all of the customer s requirements, an inexpensive, efficient cooling system is necessary. Passive cooling is ideal, because it uses no energy. Concept Alternatives Passive Cooling Options 1) Shading 2) Geothermal Cooling 3) Architectural Modifications a. Insulation b. Night Cooling Options Requiring Power 4) Modified Evaporative Cooler 5) Thermoacoustic Cooling Details of Concept Alternatives 1) Shading Shading is the use of a fabric barrier to minimize solar radiation. Shading can reflect up to 80% of solar radiation. It requires little to no maintenance. The fabric can be easily retrofitted depending on the structure it is shading. The material can be simply removed during the rainy season. The structure can be adjustable to provide maximum shading depending on the angle of the sun. One negative effect of shading is that it blocks natural light from entering the building. It is unlikely that solar reflective fabrics are readily 5

10 available in Mali. A supporting structure must be designed for the fabric barrier. This structure is susceptible to wind and dust storms. 2) Geothermal Cooling Geothermal cooling involves running air through an underground duct before entering the building. The ground temperature is cooler than the surrounding air temperature in Mali. Geothermal cooling uses some power to force the outside air through the duct and into the building. A negative of geothermal cooling is that it may be difficult to keep the duct free of debris and dust. It may also be hard to clean the duct after installation. The cost of the duct can be relatively expensive. 3a) Architectural Modifications (Insulation) One architectural modification is the use of insulation in the structure to reduce the external heat gain. The insulation can be easily modified to fit each building. This architectural modification requires little to no maintenance. There is just a one time installation cost, and no cost to operate. One problem with this option is the availability and cost of insulation in Mali is unknown. Also, water damage may occur from possible leaks in the roof, resulting in maintenance issues. Insulation also reduces the internal overhead height, resulting in a smaller room. 3b) Architectural Modifications (Night Cooling) Night cooling is the process of adding an object with a high thermal mass to absorb heat during the day and exposing the thermal mass to cool night air, in turn releasing the stored heat. Night cooling would have been best implemented before the construction of the CLICs. It is difficult to retrofit an existing building. 4) Modified Evaporative Cooler A modified evaporative cooler would allow for a more efficient cooling system. Modifications that can be made to the evaporative cooler are to modify the water pump, modify the cooler pads, and modify the fan. The coolers are already installed in the CLICs, so the cost would only be for the changes made. The cost to presently run the coolers would be lowered, because the power used to run the coolers would be lowered. Downfalls to this option are that it still requires power and relies on foreign parts. 5) Thermoacoustic Cooling Thermoacoustic cooling is the use of sound waves to cool. Sound waves created by a loudspeaker cause a container of helium to expand and compress, which in turn produces heat. There are many problems with using thermoacoustic cooling. Thermoacoustic cooling is very new, expensive, and complicated technology. It would require the importing of compressed helium gas into Mali. Although this cooling option is 6

11 environmentally friendly, it still requires electricity. Finally, there is a patent on thermoacoustic cooling. Selection Processes The pros and cons of each option were evaluated in order to choose options which best fulfilled the customer s requirements. After the selection process there remained three options: Shading, Geothermal Cooling, and Modified Evaporative Cooler. Both of the architectural modifications were discarded since insulation and night cooling would be best implemented prior to the construction of the building. Thermoacoustic cooling was also discarded for a number of reasons. The technology is too new for the area in which it would be used. Final Concepts and Rationale Using a Trade-Off Matrix assisted in evaluating which of the remaining three options was the best. Design Options Modified Geothermal Shading Evaporative Engineering Specifications Cooler Lower Temperature from 100 F to 75 F Uses Supplies Available in Mali Cost Under $ Less than 4.9 Amps at 120 Volts Works 8 hours/day, 7 days/week Operate with 26 μg/m³ Dust Content in Air Culturally Acceptable Safe Pollution Free Total Table 1: Trade-Off Matrix Each of the three options generated fairly close totals in the Trade-Off Matrix. Calculations were then performed to determine how well the options can actually perform to lower the CLIC room temperatures. VI. PRELIMINARY PROTOTYPE CALCULATIONS All three options were investigated simultaneously. Preliminary calculations can be found in Appendix A. Analysis of each cooling method was performed in order to determine the best option for a preliminary concept. As demonstrated from the supporting analysis, each individual cooling method operating separately is not able to 7

12 sufficiently lower the inside temperature of a CLIC. Individually, the evaporative cooler, the geothermal cooling and the shading barrier lower the inside temperature to 107 F, 106 F and 119 F respectively. However, combining the three methods yields a final inside temperature of 92 F and it should be noted that this is a 35 F change in temperature from the initial temperature of 127 F. Type Temperature ( F) Evaporative Cooler 107 Shading Barrier 119 Geothermal Cooling 106 Combined Systems 92 Table 2: Preliminary Temperature Calculations Although a noteworthy achievement, the calculated inside temperature did not meet the preliminary goal of 75 F. There are several explanations as to why the original calculated inside temperature is too high and how a temperature closer to the goal can be achieved. In the calculations, the assumptions were very conservative. The predicted 92 F inside temperature is for the worst-case scenario. Calculations were performed at steady state, during the hottest time of the day with the most direct sunlight. Realistically, outside temperatures and solar radiation vary throughout the day. The temperature of the building structure will rise and fall, lagging behind the outside temperature and not remain constant. Additionally, the ground temperature applied in the analysis was the highest temperature measured by the USDA in the Mojave Desert. VII. TESTING AND EXPERIMENTATION Barrier Materials The first test experiment ran concerning the shading barrier was to test the effectiveness of various materials and colors of materials. The setup consisted of a halogen lamp at a fixed distance away from a heat flux meter attached to a small block of aluminum with various test barriers being inserted between the meter and the lamp. The results can be seen in Figure 1. All of the materials tested worked equally well to reflect the radiation given off by the light bulb with the exception of the radiant heat barrier, which reflected effectively all of the radiation. One noteworthy effect seen was the white fabric from the same sample material reflected 30% more solar radiation than the black material. 8

13 Temperature vs Time Temperature (Celcius) 8.00E E E E E E E E+01 Fabric Alluminum Fabric Separated Foil Separated Foil 0.00E Time (Seconds) Figure 1: Results of Shading Materials Barrier Placement The second experiment was designed to test the effect of separating the barrier from the barrier. For this experiment test were ran with a piece of radiant hear barrier or a piece of white fabric directly on a piece of aluminum and resting two inches above the piece of aluminum, while exposed to a halogen bulb. Temperature readings were taken on the piece of aluminum. According to the data, a fabric barrier that is separated from the metal surface is approximately as effective as the radiant heat barrier attached directly to the metal surface and the radiant heat barrier separated from the metal surface was the most effective. From this data the plan was to use the radiant heat barrier on the roof in Mali with an air gap of several cm between the barrier and the roof. Kangaba CLIC Overview The building in Kangaba containing the CLIC is 30 meters by 12 meters by 4 meters (length, width, height). The CLIC is bordered on each side by a room. The dimensions of the room containing the CLIC are 10 meters by 12 meters. The building has a double roof. The two layers are approximately two feet apart. The top layer of the roof is corrugated metal. The top of the roof contains bolts, approximately every three feet, sticking up about two inches above the corrugated metal. The bottom layer of the roof is a ceiling of metal tiles. The walls are approximately 15 cm thick and made of concrete. The floor is also concrete. In the front of the CLIC, facing the steps, there are vents that 9

14 cannot be closed. They are approximately two feet in height and are located at the top of the wall. Figure 2: Kangaba CLIC Storage Steps 1 CLIC 2 Classroom Windows Figure 3: Kangaba CLIC Geothermal piping Shading Type J and K thermocouples were used to record the temperatures in the CLIC and on the roof of the CLIC, as well as the outdoor air temperature. All thermocouples were connected to an Agilent Data logger and the Agilent software was used to collect the data. Locations of the thermocouples are listed in Table 3, below. 10

15 Channel Location 1 Room Air (1) 2 Roof Temperature 3 Ceiling (2) 4 Room Air High (2) 6 Outside Air 7 Ceiling by Door (1) 8 Shaded Roof 9 Unshaded Roof Table 3: Thermocouple Locations The room air was measured by two different thermocouples (channels 1 & 4) by hanging them from the ceiling of the CLIC. The ceiling temperature was also measured by two different thermocouples (channels 3 & 7) by attaching them to the ceiling of the CLIC. The outside air temperature (channel 6) was measured by attaching a thermocouple to a pole approximately three feet from the ground. Each of the three roof thermocouples (channels 2, 8, & 9) were attached directly to the top of the roof in direct sunlight. Approximately twenty four hours of initial control testing was done without use of the evaporative cooler and with the doors and windows of the CLIC closed. The shading barrier was added on Monday, March 21. The shading barrier consisted of light colored cotton fabric beneath a foil heat barrier. The addition of cotton fabric was necessary, because the bolts on the roof ripped holes in the foil. The fabric was fastened to each side of the roof by wrapping it around a cement overhang and pinning or sewing the fabric to itself. Rocks were then added along the length of the fabric strips in order to weigh down the barrier during strong gusts of wind. On Tuesday, March 22 the doors and windows remained open. On Wednesday, March 23 the doors and windows were closed for the majority of the time. There were classes going on inside the CLIC throughout testing, so it was difficult to consistently supervise the condition of the doors and windows. Total testing time was about three days. A complete test manual was created prior to arrival in Mali and was modified to the Kangaba CLIC upon arrival (Appendix I). Telephonic In order to determine the difference in the temperature of a single roofed building and a double roofed building, the air and ceiling temperatures were taken at various times at a local single roofed telephonic. The telephonic is a small building, approximately 3 meters by 3 meters. The building has telephone available for public use. Ground Temperature The ground temperature was measured by digging a hole 85 cm deep and placing thermocouples in the ground at 85 cm, 70 cm, 55 cm, 40 cm, and 25 cm deep, filling in the hole with dirt that was dug up. The initial plan of pounding a metal rod with several 11

16 thermocouples attached, as suggested by geologists, was unsuccessful. Since the climate in Mali is very dry, the ground was also very dry and rocky. In some places it was solid rock. Therefore, the process took several hours. Geothermal Piping The geothermal pipe was installed in Kangaba prior to arrival. The pipe is made of PVC, 20 meters in length, 20 centimeters in diameter, at a depth of 60 centimeters, and is located on the North side of the Kangaba CLIC. In order to determine the air temperature inside the pipe, a thermocouple was inserted at various depths inside the pipe at different times throughout the day. Upon completion of the evaporative cooler modifications, the evaporative cooler and geothermal piping were joined. Both air velocity and air temperature were measured at the inlet and outlet of the geothermal piping with the cooler fan running. Geothermal Adapter If the evaporative cooler was to use air that came through the geothermal cooling pipe, an adapter was designed. The first prototype tested can be seen in Figure 4. Figure 4: Geothermal PVC Adapter One of these was placed in front of each of the faces of the evaporative cooler. An envelope covered the support structure and extends to the evaporative cooler so that air could enter the evaporative cooler through the hole in the adapter. This design caused a significant decrease in airflow into the evaporative cooler and had to be redesigned. The evaporative cooler adaptor was modified to envelope the entire evaporative cooler. This structure was placed around the evaporative cooler and then covered in a plastic tarp. The seams of the tarp were covered with duct tape to ensure that it was airtight. An opening was cut for the evaporative cooler exhaust and sealed with duct tape. A 20 cm hole was cut in the back as the only way for air to get into the evaporative cooler. This hole was connected to the geothermal cooler by a flexible duct. The structure can be seen in Figure 5. 12

17 Figure 5: Geothermal Pipe Adapter On-Site VIII. DATA ANALYSIS AND RESULTS Shading The outdoor air temperature, recorded by channel 6, is shown in Appendix R. The final day of testing was the hottest day of testing. The period of time in which the outdoor air temperature was over 100 Fahrenheit was double that of any other day. The overall change in the temperature of the roof after adding the shading barrier was a decrease of 60 Fahrenheit. The roof temperatures are shown in Appendix S. Thermocouples 8 and 9 were added after two days of testing. The effects of shading are visible on the second day at noon. The instant the shading barrier was added, the temperature of thermocouple number two decreased by almost 20 Fahrenheit. Since the building had a double roof, the effect of the decreased roof temperature on the room temperature was not as dramatic. The addition of a shading barrier caused the ceiling temperature to decrease by approximately 15 Fahrenheit. The changes in ceiling temperature are shown in Appendix T. The room temperature decrease is marginal; approximately 2 Fahrenheit. The room temperature results are shown in Appendix U. The reason this temperature difference is so small has to do with many factors. The double roof damped out the large decrease in temperature caused by the shading barrier. The doors and windows were another uncontrollable factor in the testing. However, even though the temperature change is minor, the room still feels much cooler to humans because the heat radiating from the ceiling to the room has been greatly decreased, shown below in Table 4. This table shows the heat load on a person, not the actual temperature; the effect of evaporation resulting from human sweat has not been included. 13

18 Roof Type Without Shading ( F) Temperature Reduction with Shading ( F) Single % Double % Table 4: Room Temperature Comparisons Heat Load Reduction Telephonic The ceiling temperature of the telephonic ranged from 91 F to 141 F, with indoor air temperatures ranging from 85 F to 99 F. These temperatures, along with the times of measurement can be seen in Appendix O. Ground Temperature We took readings from the ground thermocouples at varying times during the day for 4 days. We found that the ground temperatures followed a fairly regular pattern over the course of a day which was expected. The temperature at 25 cm was generally the highest, while at 85 cm it was the lowest, though they didn t differ from each other by much. At the most there was a 5 degree Fahrenheit difference between the two points, and on average there was a 2.6 degree Fahrenheit difference. We found that the underground temperature was lowest around 14:00. It then increased to reach its peak around 08:00, and then began to decrease again. This data and a graph are included on Appendix L. Relative Humidity We found the relative humidity in Mali to be very low as we expected. Though our actual measurements were slightly higher than the 10% relative humidity that we used in our calculations, it is not enough of a difference to change our design. The relative humidity is still low enough for the evaporative cooler to be effective without affecting human comfort. This data is shown in Appendix M. Geothermal Piping We measured the air velocity and temperature of the air through the geothermal system connected to the evaporative cooler fan over the course of 30 hours. We discovered that the restricted opening of the pipe greatly decreased the ability of the fan to move air, which is what we expected. Appendix P shows our air velocity measurements. We used this information as a basis of comparison for our final design in order to find a fan that would provide equal or greater air flow through the geothermal system. The average exit air velocity from the geothermal piping was 350 feet/minute. Twenty four hours after the evaporative cooler was connected to the geothermal piping, the inlet temperature at noon was F while the outlet temperature was 91 F. Therefore, at the hottest time of day there was a 10 F decrease in temperature of air flowing at 350 feet/minute through the 14

19 geothermal piping. The geothermal system temperature data can be viewed in Appendix N. Conclusion The on-site measurements were used to guide the conclusion of the project. They influenced our final design. The shading barrier is not as effective when used on a concrete roof. The shading barrier was found to be more effective for buildings with single corrugated metal roofs. Even though the ground temperature was higher than anticipated, the results and data gathered from the geothermal cooling were still found to be useful in cooling the room. The geothermal piping was found to be effective at the hottest time of the day as it decreased the ambient air temperature by 10 F. IX. FINAL DESIGN DESCRIPTION Ouelessebougou CLIC Overview The CLIC in Ouelessebougou is a more common building layout for the Mali CLICs. It is located in a single room building which is much smaller than the Kangaba CLIC. The building is 4.5 meters by 9 meters. It contains a large computer room as well as a smaller office. There are 8 computers, a printer/copier, fax machine, radio, television, VCR, and DVD player. The roof is made of cement and has a border, 43 centimeters high, 21 centimeters thick, that surrounds the entire perimeter. The walls are 21 centimeters thick and made of stucco. Inside walls are 3 meters tall, floor to ceiling. Outside walls are 4 meters in height from the ground to the border around the roof. There are 5 widows, 109 centimeters wide, 112 centimeters tall. There are two doors, each 1.2 meters wide. Figure 6: Ouéléssébougou CLIC 15

20 S Doors, 1.2 m W 4.5 m 4 m Satellite dish (on roof) Fans Office/ Side room 2.3 m 6.1 m Figure 7: Ouéléssébougou CLIC 2.9 m The entire town of Ouelessebougou is run on a diesel generator. Most brown-outs occur during the hottest time of the day. At times, there is not enough power to turn on the computers. Three to four people use the internet for one to two hours per day. The computers are only turned on while they are in service. Most people come to use the photo copier. The budget for the Ouelessebougou pilot site is between $500 and $1000. Concentration on local products is recommended. Importation of fans is allowed. Powering with solar panels is unnecessary at this point. The design should be created with the dish in place. Only three of the CLICs in Mali have metal roofs, and the remaining have cement roofs. Modified Evaporative Cooler In an attempt to make the evaporative cooler more effective and cheaper to operate, modifications are needed. Since the evaporative coolers in Mali were not properly installed, they were not effective in cooling the rooms, and would eventually make the inside of the building so humid that it would be worse than without using the evaporative cooler. A permanent hookup to the geothermal cooling system that allowed access to the inside of the cooler to add water was designed. More efficient fans were found and mounted in the cooler to reduce power requirements. A passive system of soaking the pads was considered to lower power requirements, but rejected because the need for constant flowing water for the pads to be effective. Additionally alternative pads that could be locally manufactured were researched, but contacts in Mali working on the problem said that no alternative could be found. In order to modify the evaporative cooler the inside of the evaporative cooler was 16

21 stripped. Instructions on how to do this are found in Appendix AA. A sheet metal cover that surrounded the location of one of the side pads, and extended 2.5 inches away from the cooler was built. This box has a hole to connect to the geothermal cooling pipe and can be seen in figure 8. Figure 8: View of Modified Evaporative Cooler The hole is cut at a height that matches up with the geothermal pipe, coming in through the wall, when the cooler is resting on the floor. The pad was slid into place and this cover was pop riveted to the cooler, and the geothermal pipe was connected to the hole in the sheet metal. Both were sealed with caulk so the cooler would draw as much air as possible through the geothermal cooling system. In this design this pad would be the only pad in use. The face on the opposite side is used to mount the fans. To mount the fans in this side, the pad would be removed and a piece of sheet metal is attached in its place covering the entire opening. The sheet was attached using pop rivets and sealed airtight using caulk. The fans have a lip surrounding the cage that is used to hold the two pieces of the fan housing together using screws. To mount the fans inside the evaporative cooler, the holes for these screws were used. The sheet has two circular holes in it with diameters of 25.5 cm. The size of these holes matches the diameter of the fans, minus the lip of the cage, which is used to mount the fans. Holes were drilled in the sheet metal to match up with the holes for the screws in the housing. The fans were then mounted into the evaporative cooler using longer.3 cm screws. These fans were found to use 145 Watts together and still blow 240 CFM of air through a simulated geothermal pipe. This is a substantial decrease in power from the original fan (which used 303 Watts with the pump) and moved over twice as much air through the geothermal pipe as the original fan, which moved 105 CFM. 17

22 Figure 9: Fan Mounting The face of the evaporative cooler that originally had the exhaust for the cooled air is used to mount the controls for the fans and pump. The on/off switch box was removed from the cooler. The wires for the original fan were removed, but the wires from the plug and to the pump are still used. An outlet with two sockets was attached to the top face of the on/off switch. The two fans are plugged into the outlet and the outlet is connected to the on/off switch that controls the fan. Any exposed electrical wire was covered using electrical tape. The box was mounted to the metal lip that used to be attached to the fan housing so that it is flush with the outer surface of the evaporative cooler. This face was then covered with a piece of sheet metal with an opening for the switch, attached with pop rivets, and sealed with caulk. Figure 10: Switch Mounting 18

23 One of the panels remains removable to allow the users to add water to the cooler, and to allow for replacement of the pad if necessary. For this purpose the panel opposite the original evaporative cooler exhaust was designed to be removable. The bag of pad material was removed from its metal frame and a piece of sheet metal was cut that would cover all the openings on the frame. The sheet metal was attached with pop rivets and sealed airtight using caulk. The frame was then slid back into place. When in place this panel would restrict air flow enough to cause air to be drawn through the panel attached to the geothermal cooling system, but could be removed to allow a person to replace a pad from the inside of the cooler, and add water as needed. To make sure that the water did not leak out of the cooler a small piece of sheet metal was used to cover the opening on the bottom of the evaporative cooler. This sheet was also attached with pop rivets and sealed using caulk. Evaporative Cooler Fan Figure 11: Side Panel Removed Since one of the goals of our project was to reduce the amount of energy consumed by the evaporative coolers that the CLIC s currently posses, we looked at the components of the cooler to see what we could change. The two components that require energy are the water pump and the fan. The pump only required 29 Watts, but the fan needed 360 Watts to operate, so one of the possibilities was to replace the existing fan with a more power efficient one. We wanted to reduce the power consumption without compromising much on the volumetric air flow rate. To test the air flow, we built a pipe out of sheet metal with an opening that was restricted enough to simulate the drop in pressure that our underground pipe in Mali would produce. This drop in pressure is due to the geometry of the pipe (i.e. friction due to the number of bends and length of the pipe), and is related to the dimensions of the pipe as shown in equation 1.1. We started with a 6 diameter pipe, and found the diameter of the restricted opening with the following calculations: 19

24 1 2 L 1 2 Δ Po = ρ Vo = (6kb + f + 1) ρvd (1.1) 2 D 2 4 d 2 2 Dd V = o Vd (1.2) Do Δ P is the pressure drop, ρ is the air density, k b =0.8, f=0.02, L is the length of Where the pipe in Mali, and D d is the diameter. We can substitute (1.2) into (1.1) to get: D D d o 4 = 6k b + f L D d + 1 (1.3) Knowing the diameter of the pipe in Mali, D d =20 cm or 7.87 in, we solved for the diameter of the restricted opening, D o and found that it equaled 4.7 in. We then fit a cardboard piece with a 4.7 inch hole to the end of our test pipe. We attached the fan to the opposite end of the pipe and measured the air velocity through a small hole in the pipe near the fan. We used the air velocity to calculate the volumetric flow rate. The first fan that we tested was a 6 diameter, 14 Watt, Air King brand fan that was rated to provide 1637 CFM. We found that this information was inaccurate however and also that the fan was very sensitive to the restricted air flow of the pipe. It only produced about 74 CFM through the pipe. After looking at several other fans, we purchased a Honeywell fan for $14.99 that requires 57 Watts of power, and provides between 800 and 1000 CFM on its own. To ensure that we would get enough air flow through the pipe and the pad, we purchased two of these fans to install in the cooler. These were the most efficient fans that we could find for a low price that provided enough air flow for our application. We connected these two fans to the existing switch by simply adding an outlet inside the cooler where the two fans would remain plugged in and turned to the on position. We then wired the outlet to the outer switch so that when the switch was turned on, both fans would be turned on high. Geothermal Piping For the original testing in Mali, PVC piping was used as the ground piping for the geothermal cooling. While in Mali, the group researched additional materials to supplement the imported and costly PVC piping. Clay pottery piping was found to be a viable replacement material for the geothermal ground piping. Currently, these clay pipes are manufactured and used in Mali as rain gutters on housing. The decision to implement this material for the underground piping is based off of several factors. First, several producers manufacture the current clay rain gutters locally. The manufactures are easily able to custom make the piping to the specific diameters, thicknesses and piece 20

25 sizes necessary. Contacts in Bamako cited an estimate of $15 to $20 for a 20 cm diameter, 50 cm length piece of piping. Clay pottery is also more environmentally friendly and will biodegrade with time. Figure 12: Geothermal Piping Design The clay piping also possesses technical benefits over some manmade material, such as PVC. For instance, the thermal conductivity, the rate at which heat flows through a material, is approximately five times greater for clay pottery versus PVC. A higher rate of conduction or heat flow is preferred in this situation since it is desired that the material will easily transfer the heat of the warmer air to the cooler surrounding earth, thus cooling the passing air. Material Thermal Conductivity, k (W / m-k) PVC 0.19 Fired Clay 1.0 Table 5: Thermal Conductivity Values for Materials Placement Location of Geothermal Piping and Evaporative Cooler The greatest factor in determining the location for the placement of the geothermal piping is the movements of the sun. In order to insure the coolest ground temperatures available, the optimal location for the piping is situated on the east side of the CLIC building. This conclusion was reached through researching the sun s daily and yearly 21

26 paths at Bamako s latitude using an applet model of the sun s path sponsored by the Australian National University. At N, the sun s path travels for eight months of the year, September to April, in the southern hemisphere. This is known as the dry season in Mali. For the remaining four months, May through August, also known as the wet season, the sun travels in the northern hemisphere. Therefore, the southern wall of the CLIC receives exposure from the sun for eight months and shade for four months. The opposite is true for the northern wall. Figure13: Sun's Daily Course in July (left) and in January (right) In addition to traveling between the southern and northern hemispheres throughout the year, the sun travels daily from the eastern hemisphere to the western hemisphere. This daily movement has the greatest effect upon the placement location of the geothermal piping. The eastern ground and side of the building experience the morning sun during the coolest daylight period. During the afternoon, the hottest period of the day, the eastern ground and wall are located in the shade while the western ground and wall are exposed. Figure14: Sun's Daily Course and Noted Shading at 9 am (left) and 3 pm (right) during September 22

27 Therefore, the ideal ground location for the geothermal piping is on the northeastern side of the CLIC. Shading is realized throughout the day in this location during the dry season, the time period being addressed in this project. In addition, the eastern portion receives shading during the afternoon, the hottest period of the day. For the pilot site, the ground east of the building is free of landscape. Additionally, this is the best site for the placement of the evaporative cooler in the Ouelessebougou CLIC. The eastern wall is uninterrupted, lacking windows or technical equipment. Shading Barrier Figure 15: Building Diagram for Ouelessebougou CLIC The shading design for the Ouelessebougou CLIC does not include shading of the roof since it has a single cement roof. However, shading of the doors and windows is an option that could be used to block much of the solar radiation. The shading material is light colored cotton, which is easily found in Mali. The material is sewn around two curtain rods located at the top and bottom of each window, shown in Figure 16. The curtain rods are fastened to the outside of the building using hooks and screws, commonly used with all curtain rods in Mali. Prior to sewing, the material for each window measures 138 centimeters in height by 130 centimeters in width. The shading design for the doors is very similar to that of the windows. However, rather than the curtain rods hanging on the building, they will be placed on the top and bottom of each door. There are two doors in the Ouelessebougou CLIC; however one is a double door. Prior to sewing, the material for the single door measures 229 centimeters in height by 118 centimeters in width. The material for the each of the double doors measures 229 centimeters in height by 59 centimeters in width. The total material used is 14.4 square meters. Depending on the width of the fabric used, the material may need to be pieced together. An 8 centimeter seam will be sewn on both the top and bottom of each piece of fabric in order to insert the curtain rod. Since the windows already have curtains on the 23

28 inside of the building, shading on the outside may not provide a significant difference, therefore shading may or may not be used on the Ouelessebougou CLIC. Figure 16: Window Shading Barrier X. PROJECT MANAGEMENT Team Member Assignments The Mali Cooler team worked together to make group and project decisions, but we quickly learned that it was necessary and time efficient to assign and delegate sections of the project to individuals and smaller teams. During the fall semester, our team elected Stefan Yanovsky as the team leader. Stefan s role was to lead and manage the project by organizing team meetings, delegating assignments and serving as a team representative. All team members played a vital role in the many processes of the year long project including preliminary and conceptual designs, final design, testing and experimentation, prototyping, onsite experimentations and modifications leading to a final design. The team members worked individually and on teams to complete the numerous tasks involved in a project of this scale. The individual tasks and contributions of the group members are listed below: Jennifer Borofka: Gantt Chart and WBS, hand calculations of testing and experimentation of evaporative cooler, prototyping of shading barrier, construction of geothermal ground stake, designed and created experimental protocol for testing in Mali, gathered and tested equipment for Mali, created CLIC user survey, organized notes from Mali, rework and modifications for geothermal duct. Clare Caron: FEMLAB model of geothermal system, testing of evaporative cooler, gathered and tested equipment for Mali, researched low-power alternatives to evaporative cooler fan, Photovoltaic Cell panel research and testing, conducting air flow experiments, analysis of ground temperature data. 24

29 Kelly Kanz: Mali shading data analysis, tool assimilation and packing for Mali, construction of geothermal ground stake, Solid Works drawings of initial prototype, wrote test procedures for Mali, data analysis of roof temperatures, final shading design for Ouelessebougou, brochure, poster, construction of shading structure in Mali. Stefan Yanovsky: Psychrometric calculations, solar barrier calculations, geothermal cooling calculations, construction of a geothermal pipe to evaporative cooler adapter, barrier material tests, evaporative cooler modifications, Solid works modeling of final prototype. All: Background research, concept generation, concept and design selection, testing and design modifications in Mali, packing for and unpacking from Mali, building prototype of modified evaporative cooler, implementation proposal for CLIC pilot site, all other deliverables. Work Breakdown Structure (WBS) Microsoft Project was utilized to organize, schedule and manage the progress and completion of the CLIC cooler. The final Work Breakdown Structure (WBS) can be found in Appendix X. Initially, the WBS was very broad and vague during the start of the fall semester as the team was still in the preliminary planning stages of the project. The preliminary project management breakdown can be viewed in Appendix X. As the project progressed and approached milestones, such as a final conceptual design or testing and experimentation, the team would meet to brainstorm and discuss the necessary tasks and subtasks of the upcoming milestone. The team leader would either delegate specific tasks to team members or individuals would volunteer for assignments. The WBS helped the team to organize team member assignments and also provided a timeline and schedule for the project. The Mali Cooler team held group meetings on a weekly basis. During these meetings, each team member discussed his/or her progress on individual assignments. Action items from previous meetings were reviewed. The meetings where designed so that, when necessary, an individual could alert the team that he/or she needed additional help from the team on a certain assignments or of any conflicts. From individual reports, the team leader would assess the progress of the project at the meetings and determine the next steps. Action items would be created and assigned. In addition to the meetings consisting of the four group members, the team also met with advisor Dr. George on a weekly basis. During these meetings, Dr. George would update the team on recent news and information regarding presentations or approaching deadlines. Likewise, the team informed the advisor of progress of the project. At this time, Dr. George would make recommendations to the group. These meetings proved very useful to the team and project. Dr. George was very willing to help and assist the Mali Cooler team, for instance, by providing guidance, additional resources and contact 25

30 information. Similar to other meetings, actions items were assigned to insure clarification and completion. Team Dynamics Overall, our team performed very well together with little conflict. In part, this was due to the combination of personalities of the team members. Some members were more dynamic and instrumental in leading and organizing the team and project. They would step in when leadership was needed to complete an assignment or delegate tasks. Others were willing to follow leadership and receive tasks assigned by team members. Additionally, team members were interested and excited about the project from the beginning. In selecting a project, group members wanted to participate on a project which focused on the chance to use engineering to help others, particularly in developing, third-world countries. Team members had a strong desire to see the project come to completion and succeed. XI. BUDGET For project research and development, funding came from two sources; an $800 grant from the USDA and $700 from the Ireland fund for a total of $1500. A breakdown of how this money was spent can be seen below. Of this money $1275 was spent on various materials for testing. For the final product, the initial goal was to keep the cost per unit to under $100. This number was fabricated by the team as a reasonable cost for an individual CLIC owner to invest considering the economic condition of Mali. While in Mali, the team met with Ian Howard of Geek Corps, and Dennis Bilodeau of USAID to discuss the project. They explained to the team that in order to install a functional cooling system in the Ouelessebougou CLIC a budget of around $1000 was reasonable. Of this cost the majority would be dedicated to the geothermal cooling duct. For the original test in Kangaba, PVC piping was used at a cost of $665. While in Mali alternatives to PVC were researched. Ceramic rain gutters that were locally manufactured were seen as a local alternative to the PVC. The cost of this material was still high at $10 to $15 per 50 cm piece, but the thermal conductivity of the ceramic, the fact that it is a local material, and the new budget made this a reasonable choice. Materials for modifying the evaporative cooler consisted of sheet metal, rivets, caulk, fans, wire and an outlet. The total price of these materials was around $100. The only other costs that would be associated with installing the cooling system in the Ouelessebougou CLIC would be labor. A quote to put a hole in the wall for the geothermal pipe, to dig and lay the geothermal pipe, and to construct the evaporative cooler came to $35. 26

31 Costs per CLIC DESCRIPTION Quantity Cost Shading Barrier Light Colored Cotton Fabric 14.4 m 2 $45 Curtain Rods 16 $32 Modified Evaporative Cooler Lakewood 57-Watt Fans 2 $35 Sheet Metal 1.18 m 2 $ Volt Power Outlet 1 $1 Caulk $5 Rivets 40 $5 18-Gauge Wire $2.5 Female Disk #22 Connectors 1 $2 Labor $15 Geothermal Ground Cooling 20 centimeter diameter Clay 20 m $600-$800 Piping Labor $20 TOTAL COST $812.5-$ Table 6: Costs Per CLIC Budget Status Project Budget $800 (USDA) + $700 (Ireland Fund) Spent $1275 Evaporative Cooler $350 PVC Pipe for Shading Structure $20 PVC Pipe for Evap. Cooler $25 Cement Blocks $5 Fan $50 Inverter $45 Tarp $5 Flexible Duct $20 Batteries $20 Solar panel $65 Geothermal PVC in Mali $665 Sheet Metal $50 Fans $35 Outlet $1.00 Wire and Connectors $5.00 Rivets and Caulk $10.00 Tarp $

32 XII. CONCLUSIONS AND RECOMMENDATIONS The issue of cooling in hot, dry climates, such as Mali s, is a difficult problem and one not yet solved by mankind. The hot, dry environment with intense solar radiation creates temperatures in the 100 s with even greater temperatures inside buildings. The CLICs have a very limited electrical supply supported by an unreliable energy grid. The customer desires that the product can be manufactured in Mali and, therefore, locally serviceable. The technology available in Mali and the financing for the cooling system is very limited. Expensive, foreign, and energy consuming cooling methods such as air conditioning, and energy sources such as solar panels are not options. The modified evaporative cooler, geothermal cooling and shading barrier options presented best address the specifications by the customer. We have designed and tested a cooling system that will be implemented in the Ouelessebougou Community Learning and Information Center (CLIC) in Mali, West Africa. Mali has a very hot and dry climate and is in a very underdeveloped part of the world. The evaporative coolers currently being used in the CLICs are power consuming, inefficient and expensive. The project was approached from three perspectives: revising the existing evaporative cooler by examining its power consumption and changing its fan; connecting the evaporative cooler to a geothermal duct that could pre-cool the incoming air; and add strategic shading to minimize the solar heat load on the building. All three approaches were experimentally verified on location in Mali and in the engineering laboratories. A 20 m duct was buried at a depth of 0.5 m before our arrival in Mali. Actual ground measurements and inlet and outlet air temperatures were measured over a 3-day period. Results indicated a 10 F drop in incoming air to the evaporative cooler at the hottest time of the day. A radiant heat barrier was installed on the roof of a large Community Learning and Information Centers (CLICs). Roof and room temperature measurements, air humidity, and air velocities were monitored over a 5 day period examining both the building baseline and building with intervention. Results showed a decrease in roof temperature of 70 F, significantly altering comfort levels inside the building. Data and statistical analysis of continuous 10-channel measurements over 5 days led to a final design that will decrease room temperatures of 20 F from ambient, using 146 Watts, keep relative humidity under 50% and consume 0.5 GPH of water. The final design recommendations will be implemented in Ouéléssébougou, Mali in summer 2005 by USAID and the Geekcorps, a corps of information technology volunteers. Recommendations One of the biggest improvements our team could have made is to have spent less time on the earlier stages of the project and more on the preliminary design and the testing and evaluation that accompany the preliminary design. It is recommended that the design requirements and the conceptual designs have earlier deadlines in the fall semester. 28

33 Traveling to Mali was the most useful event of the spring semester. It enabled us to perform on-site testing as well as to become more knowledgeable about the project as a whole. The final project was to be completed less than two months after returning from Mali. This caused the remainder of the project to be quite rushed. It is recommended that future trips are planned over J-term versus Spring Break. This would enable the team to spend more time on the design portion of the project instead of waiting all year to perform necessary tests. 29

34 XIII REFERENCES [1] AdobeAir, Inc What is Evaporative Cooling? 10 December [2] Climate-zone.com September [3] Dewitt, David P. et al. Introduction to Thermal Systems Engineering. John Wiley & Sons, Inc 2003 [4] Kuehn, Thomas et al. Thermal Environmental Engineering Third Edition. Prentice Hall. New Jersey [5] Lienhard, John H. A Heat Transfer Textbook. Prentice-Hall, Inc. [6] Moran, Michael J., and Howard N. Shapiro. Fundamentals of Engineering Thermodynamics. New York City: John Wiley & Sons, Inc.,

35 Preliminary Prototype Calculations Appendix A It should be noted that the following in Appendix A are the preliminary prototype calculations from the fall semester report. Calculated Room Temperature with No Cooling As a point of reference for all further analysis, a calculation was performed to determine the temperature, in degrees Fahrenheit, of the inside of a CLIC, at 12:00 noon, with no form of cooling being used. To calculate the temperature inside of the building there are three factors that need to be considered: the load caused by of solar radiation on the building, the load of the air surrounding the building, and the load of the computers and people inside the building. A heat transfer problem that takes these factors into account was set up to calculate the inside temperature of the building. The first load that was calculated was the load caused by solar radiation. At any given time there is 433 BTU/h*ft 2 of solar radiation that hits the earth s atmosphere. The amount of this solar radiation that reaches the earth s surface depends on latitude. At the equator, about 63%, or 273 BTU/h*ft 2 reaches the earth s surface according to Thermal Environmental Engineering [4]. As a point of reference, the location of Bamako, the capital of Mali, at degrees north latitude was used to approximate the amount of solar radiation that reaches the surface throughout the country. According to the Thermal Environmental Engineering [4], the incident solar radiation that reaches the surface is equal to the amount of solar radiation reaching the surface at the equator multiplied be the sin of the latitude angle ( β ) or I s = I dn sin(β ). From this equation the amount of incident solar radiation hitting the surface at 12:00 noon in Bamako, Mali would be 264 BTU/h*ft 2. In addition to the direct solar radiation, there is an additional amount of solar radiation which is scattered downwards by molecules in the atmosphere. This diffuse solar radiation can be up to 10% more than the direct solar radiation, meaning that the CLIC roof surface would have 290 BTU/h*ft 2 of total solar radiation. To calculate the temperature inside the building with no form of cooling a steady state energy balance on the CLIC was setup. As stated in the problem definition, the temperature in Mali is between 93 F and 102 F, so for all of the calculations an outside temperature of 100 F was used with 18 people generating 512 BTU/h each and 8 computers generating 683 BTU/h each inside the CLIC. To calculate the effect of the external loads, the building was broken into two sections, the roof and the walls. Figure 6: Heat Transfer through Roof For the heat transfer through the roof, it was assumed that the entire roof would remain at a constant temperature. The equation set up is i

36 shown graphically in Figure 1. There is solar radiation that hits the roof and heats the roof. This heat is dissipated through the top and bottom by both radiation and convection (Heat in= Heat out) Q = Q + Q + Q + Q. (1) solar convtop convbottom radvtop radbottom Area is constant on both sides of the equation so it is omitted from this portion of the analysis. The value Q solar is equal to the total solar radiation striking the surface ( I s ) multiplied by the emittance of concreteε. For both radiation values, Q rad is equal to the difference in temperature between the roof and its surroundings multiplied by the associated heat transfer coefficient or Q rad = h (2) rad ( Troof Tinside) For the convection values, Q conv again equal the difference between the temperature of the roof and its surroundings multiplied by an associated heat transfer coefficient or hconv ( Troof Tsurroundin gs ). The value of Tsurroundin gs on the top of the roof is equal to the air temperature or 100 F, and for the bottom of the roof it will be the room temperature. For the value of the heat transfer coefficients, the top of the roof combined heat transfer coefficient was estimated to be 5 BTU/h* F*ft 2, and for the bottom the heat combined heat transfer coefficient was 2 BTU/h* F*ft 2. The total combined equation for the heat transfer across the roof is ε I = h ( T T ) + h ( T T ) + h ( T T ) + h ( T T ) (3) * s convtop roof air convbottom roof inside radtop roof outside radbottom roof inside Of this total heat transfer, only the values for the bottom of the roof will affect the internal temperature. For calculating the heat transfer through the sides of the building, it was assumed that the effect of solar radiation on the walls was negligible. The equation set up is shown graphically in Figure 2. The hot air heats up one side of the wall via convection. The heat then conducts across the concrete and heats up the cooler air via convection. The equation for this setup is Figure 7: Heat Transfer through Walls Q ΔT =. (4) Rtotal The Δ T value is the difference between the air outside the building and the air inside the building. The R total value is the sum of the thermal resistances created by both ii

37 convections and the conduction. For convection, the thermal resistance is the inverse of the associated heat transfer coefficient (h) multiplied by the total exposed wall area (A) 1 or. For conduction, the resistance is the thickness of the wall (L) divided by the h conv A L conductivity (k) multiplied by the area (A) or. The total combined equation for heat ka transfer across the walls is Q walls ( Toutside Tinside ) =. (5) 1 L h A ka h A out inside To find the temperature inside the building, the heat transfer from the walls and the roof must be combined with the heat given off by the people and computers and the heat that is brought in by ventilation. Since no energy is created or destroyed, the sum of all these values must be zero. So the equation for this situation is 0 = Q + Q + Q + Q + Q. (6) walls roof people computers ventilation Q walls is given above. Q roof is the radiation and convection on the bottom of the roof. Q people is 512 BTU/h multiplied by 18 people, the maximum amount of people possible. Q computer is 648 BTU/h multiplied by 8 computers. Finally Q ventilation is the mass flow rate of air multiplied by its specific heat and the difference in temperature between the air and the internal temperature or m& c p ( Tair Tinside ). It is assumed that there are 5 air changes per hour (opening doors, leaking through building) which translates to a mass flow rate of 1350 lbm/h. The combined equation is 0 Tout Tin = + hconvbottom ( Troof Tinside ) + hradbottom ( Troof Tinside ) + mc & p ( Toutside Tinside ) + n people Load person + ncomp Load (7) comp L 1 + K h convout 1 + h convin In this equation there are two unknowns: the temperature of the air inside the building and the temperature of the roof. As a result, this equation alone is not enough to solve for the temperature inside the building. The equation for the roof temperature is needed. An equation for roof temperature is given in the analysis for the heat transfer across the roof. This equation also has the same two unknowns. To find an internal air temperature, a roof temperature is first guessed and put into the combined equation and a value for internal air temperature can be solved. Next, take internal air temperature calculated and solve for the roof temperature, using the roof heat transfer equation. This will reveal a more accurate roof temperature. This iterative process is repeated until the values converge. Using this method an internal temperature of 127 F was found. iii

38 It is worth noting that this calculation is for a steady state situation. This calculation simulates a situation where it is 12:00 noon all day long and the temperatures have all come to equilibrium. This is a worst case scenario and this temperature would in reality never be achieved. This is a reasonable assumption for a preliminary calculation in the absence of data about the change temperature and solar radiation throughout the day. Having calculated the temperature of the room with no form of cooling, the next step taken was to calculate the effect the three design options individually would have on the room temperature. First the room temperature was calculated if an evaporative cooler was being used, second if only air was being circulated into the room after running through the geothermal cooling system, and third if only a solar barrier was used to reduce the effect of the solar radiation. Calculated Effect of Evaporative Cooler on Room Temperature In performing the calculations for air temperature exiting the evaporative cooler, several assumptions were made. The air entering the cooler was assumed to be 100 F, or the same temperature we assumed the outside air to be. The air entering the evaporative cooler was assumed to be at 10% relative humidity. The water being used in the evaporative cooler was assumed to be 80 F and 100% of the water pumped (2.4 gallons/hour) was evaporated. The equation used is graphically represented in Figure 3. For this setup the mass flow rate of air is broken up into the mass flow rate of dry air and the mass flow rate of water vapor. Because mass is conserved, the amount of air, vapor, and water entering the system must be equal to the amount of air and vapor that leave the system m & + m& + m& = m& + m&. (8) a v w 1 v 2 a In addition energy is conserved. 0 = m & a ( ha h + m& v hv + m& whw m& v hv. (9) 1 a ) The enthalpy (h) values in the second equation have a temperature value associated with them, which makes it possible to solve for the temperature leaving the evaporative cooler. The first step to solving these equations is to find the total mass flow rate of air into the fan. According to the specifications given on the manufacturing company s web site, the fan is rated for 3000 ft 3 /min. To receive the mass flow rate, simply multiply this value by the density of air. For 3000 ft 3 /min, the mass flow rate would be kg/s. Fan ratings are generally higher than the actual fan performance, so Figure 8: Water in Evaporative Cooler Figure 9: Testing Mass Flow Rate iv

39 an experiment was performed to find the actual mass flow rate of the fan. The experiment, shown in Figure 4, involved taking velocity and temperature measurements at various points along a grid at exhaust of the fan. Using the temperature to find the density of the air and multiplying this value by the velocity and area of each square on the grid, would result in the mass flow rate for that portion. Adding up all the values gave the total mass flow rate for the fan. According to the experiment the mass flow rate of the fan was 32% of the rated mass flow rate, or kg/s. Knowing the mass flow rate of the air, the next step is to calculate the mass flow rates of the dry air and the water vapor based on the 10% relative humidity. The relative humidity value, φ is equal to the vapor pressure of the water divided by the vapor pressure in a saturated condition, or = P v φ. (10) Psat The vapor pressure was calculated to be lbm/ft 2. Vapor pressure can be related to the ratio of the mass flow rate of the vapor to the mass flow rate of the dry air by the equation mv Pv ω = =.622*. (11) P air atm Pv m The mass flow rate of dry air is kg/s and the mass flow rate for water vapor is kg/s going into the cooler. Knowing that the pump operated at 2.45 gallons/hour the mass flow rate of water into the cooler is kg/s. The mass flow rate of the dry air remains constant, so the mass flow rate of the vapor out of the cooler is equal to the mass flow rate of water and water vapor into the cooler. Now all the mass flow rates are known and the temperature can be found. Since the temperature of the fluids, air, water, and vapor are known, their enthalpies can be referenced. According to Introduction to Thermal Systems Engineering [3], the enthalpy values for air, the water vapor at 100 F and liquid water at 80 F are BTU/lbm, BTU/lbm and BTU/lbm respectively, meaning that the right side of this equation was completely solved: m & h + m& h + m& h = m& h + m& h (12) a a1 v1 v1 w w a a2 v2 v2 To solve for the temperature of the air, various enthalpy values corresponding to a temperature were guessed for the air and water vapor leaving the cooler, until both sides of the equation were equal as seen in Table 1. v

40 Guessed Temp Right side Left Side Table 7: Air Temperature The temperature of air leaving the evaporative cooler would be about 80 F. Using the equations given earlier, the relative humidity of the air leaving the cooler would be 40%. Knowing this value a new internal room temperature could be calculated. The original equation used to calculate the room temperature was slightly modified to do this. The m& c p ( Tair Tinside ) was changed so that the mass flow rate was increased to 4320 lbm/h to account for the mass flow rate of the fan and the T air value was changed from 100 F to 80 F. The calculated temperature in the room, using an evaporative cooler as the only method of cooling, is 107 F. Calculated Effect of Geothermal Cooling on Room Temperature The next cooling method looked at was geothermal cooling. The effectiveness of geothermal cooling depends largely on how cool the ground temperature is. Using a long enough tube, the air that enters the building, will be cooled to the ground temperature. Thus there is a limit to the cooling that this method can provide. The first step to making any calculations for geothermal cooling required finding the ground temperature in Mali. Dr. Tom Hickson of the UST geology department recommended contacting the USDA soils division to request international soil information. Unfortunately, there is no soil data available for Mali, so a close approximation here in the United States was used. The USDA soils division recommended using the Mojave Desert soil conditions as most similar to those in Mali. For the Mojave Desert, the ground temperature at 50 centimeters deep ranges from Fahrenheit. For our calculations we used the higher number, 76 Fahrenheit, as a worst case scenario. Knowing the ground temperature the next step is to run calculations to see how long the pipe must remain underground for the air to reach the ground temperature. Several assumptions were made to make this calculation. The pipe diameter was assumed to be 0.5 ft. The problem was set up as a one dimensional fluid flow problem. This means that the pipe had no bends in it. The walls of the pipe were assumed to have a uniform temperature that was equal to the ground temperature of 76 F. For this situation the exit temperature can be found according to the equation T out ground in ground _ h* Al m* c p = T + ( T T ) e. (13) For this equation, A l is the entire surface area of the tube in contact with the air and depends on the length of the tube A l = π * D* l. The mass flow rate of the air for this equation was assumed to be equal to that of the mass flow rate of the air in the vi

41 evaporative cooler equation. At this point the _ h value is the only unknown besides the exit temperature. Knowing the mass flow rate, a Reynolds Number can be calculated: m Re = 4 μπd (14) where μ is the dynamic viscosity of water and D is the diameter of the pipe. After getting the Reynolds Number, the next step is to calculate the Nusselt number f _ *Re*Pr Nu = 8 (15) f 2/3 [ * *(Pr 1) 8 where f.3164 = 1/ 4 Re (16) and Pr, the Prandtl number, is equal to 0.7. Finally, the Nusselt number can be related to _ h by the equation _ h = _ Nu* k D (17) where k is the thermal conductivity of air or BTU/h*ft* F. To calculate the optimal pipe length and diameter, all of the previous equations were put into a spreadsheet. Pipe length and the resulting exit temperature of air have an asymptotic relationship. Selected data can be seen in Table 2. Pipe Length 40 Pipe Diameter 0.5 Ground Temperature 76 Exit Temperature CFM Table 8: Pipe Length & Exit Air Temperature vii

42 A pipe length of 175 ft was required to get an exit air temperature of F. This number was put back into original room temperature equation to see what the new room temperature would be using geothermal cooling as the only method of cooling down the room. The m& c p ( Tair Tinside ) was changed so that the mass flow rate was increased to 4320 lbm/h to account for the mass flow rate of the fan and the T air value was changed from 100 F to 77 F. The calculated temperature in the room, using geothermal cooling as the only method of cooling, was 106 F. Calculated Effect of Shading on Room Temperature The third method looked at was shading. Since a large portion of the heat gain in the building is a result of solar radiation, the temperature inside the room can be significantly altered by reducing this load on the building. The setup for the calculations dealing with shading is graphically represented in Figure 5. For this situation, Q solar is the solar radiation originally calculated in the first calculation of this section. This hits the barrier and heats it up. The barrier gives off its heat through radiation and convection on both sides. The balanced energy equation for this problem is Figure 5: Solar Energy α * Q = 2* Q + 2Q (18) solar emit conv where the absorptivity of the barrier is α. The solar radiation given off by the barrier is related to the objects emissivity (ε ) which is assumed to be 0.9, the temperature of the object ( T obj ) and the Stefan-Boltzman constant (σ ) by the equation Q rad 4 * σ * Tobj = ε. (19) The convection is calculated as discussed previously to be Q conv = h ( T T ). (20) * obj air The temperature of the object depends on the objects absorptivity and the heat transfer coefficient between the object and the air. This temperature translates into an emitted radiation according to the equation emit 4 * * Tobj Q new = ε σ. (21) viii

43 Barrier temperatures and emitted radiation were calculated using various heat transfer coefficients and absortivities and can be seen in Table 3. h α T Q emit h α T Q emit h α T Q emit Table 9: Shading Calculations A barrier absorptivity of.9 and a coefficient of 2 were select as reasonable values for situation in Mali, base on wind conditions and available materials for a barrier. For this situation the emitted radiation would be BTU/h* ft 2 or W/m 2. This number was put back into the original equation to see what effect using shading as the only method of cooling would have on the internal air temperature. For this calculation the original room temperature equation was used with BTU/h* ft 2 being substituted for the original 290 BTU/h* ft 2. The calculated temperature in the room, using solar shading as the only method of cooling, is 119 F. Calculated Effect of Combined Systems Since none of the three options were able to get the room temperature to the desired 75 F, a calculation was run to see what effect all three methods would have on the temperature inside the building. For this calculation the radiation hitting the roof of the building with a solar barrier was BTU/h* ft 2. The air entering the evaporative cooler was set to 77 F, as if the air that had gone through the geothermal cooling. Using the already cooled air, the moisturized air would leave the evaporative cooler at 60 F. This 60 F air and the mass flow rate of the evaporative cooler were then used for the air ventilation into the building. Combining all three methods the temperature inside the room was calculated (Table 4). Toutside Tinside Aroof cp # of Load of person # of Load of computer Is (Incident Solar Radiation) # of Air exchanges Awalls (F) (F) Troof (F) (ft^2) (BTU/lbm) people (BTU/h) computers (BTU/h) Emissivity BTU/h ft^2 per hour (ft^2) Tair 60 Equation 1 E*Is = htop(troof-tair) + hbot(troof- Tinside) Troof = Equation 2 Tin= Table 10: Room Temperature ix

44 Appendix B-F Hand Calculations available on request x

45 Quality Function Deployment Appendix G Engineering Specifications Customer Requirements Cool Temperature of Building Little/No Energy Consumption Operate in Hot/Dusty Environment Manufacture in Mali Easy to Fix Easy to Use Maintain/Lower Water Usage Low Cost Reliable X Lower Temperature from 100 F to 75 F Less than 4.9 Amps at 120 Volts X Operate with μg/m³ Dust Content in Air X Less than 2.45 GPH Water Usage X Cost Under $100 X Works 8 hours/day, 7 days/week X Uses Supplies Available in Mali X xxxiii

46 Mali Packing List Appendix H Quantity Equipment S/N Packed? 2 Omega Digital Thermometer 77JY0159, 77JY0153 Jenny Carry-On, Green Suitcase 1 Omegaflo Anemometer HHF615M Jenny 1 Velocicalc Anemometer Black Duffle 1 Digital Data Logger US Kelly Carry-On 2 Multiplexers HP34901A 20Ch., HP34901A US , MY Clare Carry-On, Black Duffle 2 Omega Temperature/Relative Humidity (Small) , Green Suitcase 1 Omegascope Handheld Infrared Thermometer T Green Suitcase 1 Mastech Multimeter MY Green Suitcase 2 boxes Plastic Bags Green Suitcase 1 Yellow Flag Green Suitcase 1 Wire Cutter Green Suitcase 1 Wire Stripper Green Suitcase 2 Data Logger CDs Jenny Carry-On 1 Metric/English Tape Measurer Green Suitcase 1 Masking Tape Green Suitcase 2 Duct Tape Jenny Carry-On 1 Cardboard for Evaporative Cooler Black Duffle 6 9V Batteries Green Suitcase 18 AA Batteries Green Suitcase 2 Extension Cords Jenny Carry-On, Green Suitcase 3 Power Converter (220 V) Jenny Carry-On, Green Suitcase (2) 2 Evaporative Cooler Plastic? Evaporative Cooler PVC Pipe Black Duffle, Cardboard Tube 1 Soldering Iron Green Suitcase 1 Soldering Wire Green Suitcase 1 roll Type K Thermocouples Green Suitcase 3 rolls Type E Thermocouples Jenny Carry-On, Green Suitcase (2) 4 Disposable Cameras Jenny Carry-On 1 Digital Camera Stefan 1 Hammer Black Duffle 1100 sq ft Radiant Heat Barrier Black Ski Bag 1 package Safety Pins Green Suitcase 3 Heat Flux Meters Jenny Carry-On, Green Suitcase (2) 20 Clamps Black Duffle 4 Flathead Screwdrivers Green Suitcase 2 Small Flathead Screwdrivers (for multiplexers) Green Suitcase xxxiv

47 3 Phillips Screwdrivers Green Suitcase 1 Outdoor Thermocouple Stake Cardboard Tube 19 Graph Paper (backup for Data Logger) Green Suitcase 1 box Colored Pencils Green Suitcase 1 Pencil Sharpener Green Suitcase 40 Data Sheets Green Suitcase 1 Pocket Knife Green Suitcase 3 Safety Glasses Green Suitcase 1 Jump Drive Stefan 1 Rebar Cardboard Tube 1 Kill A Watt P Green Suitcase 1 Scissors Green Suitcase 1 Clip Board Green Suitcase 2 Solar Panels Jenny Carry-On 1 Duct? 2 AutoCAD Programs Black Duffle xxxv

48 Mali Testing Procedures Appendix I Saturday, March 19 Set Up and Initial Measurements 1. Place Type J/K Thermocouples in locations a. Ground thermocouples i. Dig 85 centimeter depth hole in ground ii. Insert thermocouples every 15 centimeters (at depths of 85, 70, 55, 40, and 25 centimeters) iii. Refill hole with dirt b. Indoor thermocouples i. 4 total 1. all connected to data logger 2. 2 on ceiling 3. 2 measuring room air c. Outdoor thermocouple i. 1 total 1. connected to data logger ii. Outside of building in shade iii. On stake 1 meter off of ground d. Roof thermocouples i. 3 total 1. connected to data logger 2. 2 under shading barrier (when shading barrier is installed) 3. 1 on roof in direct sun e. Geothermal thermocouple i. 2 total 1. 1 at air inlet 2. 1 at air outlet, before entering evaporative cooler ii. Measure using Omega handheld device 2. Take initial control measurements a. Take readings from thermocouples (2.a. 2.e.) i. Ground & Geothermal thermocouples 1. Every 8 hours ii. Data logger thermocouples 1. Every 10 minutes b. Determine relative humidity using Omega Handheld Temperature/Relative Humidity Meter i. Indoor usage 1. Perform every day in center of CLIC 2. Turn on to percentage relative humidity setting 3. Hold probe in air approximately 5 ft off ground for 3-5 minutes 4. Record ii. Outdoor usage xxxvi

49 1. Perform every 8 hours 2. Same location as outdoor thermocouple 3. Turn on to percentage relative humidity setting 4. Hold probe in air approximately 5 ft off ground for 2-3 minutes 5. Record c. Measure outside environment air velocity with Omegaflo anemometer i. Perform outside of building away from structures ii. Every 8 hours iii. Select the 30 m/s velocity range setting iv. Hold probe approximately 5 feet off ground v. Record an average air velocity d. Determine exit air temperature and relative humidity of evaporative cooler using Omega Handheld Temperature/Relative Humidity Meter i. Turn on to temperature setting ii. Place probe at exit vent of evaporative cooler iii. Record temperature iv. Switch to relative humidity setting v. In same location, hold prove for 3-5 minutes vi. Record vii. Perform every hour of operation 3. Complete digging trench and installing piping (finished prior to arrival) Figure 10: Geothermal Piping 4. Put together evaporative cooler/geothermal connection xxxvii

50 Sunday, March 20 Continue Control Testing Figure 11: Evaporative Cooler/Geothermal housing connection 1. Complete previous day s activities and continue control testing 2. Record control measurements a. Take readings from operational thermocouples (2.a. 2.e.) i. Ground & Geothermal thermocouples 1. Every 8 hours ii. Data logger thermocouples 1. Every 10 minutes b. Determine relative humidity using Omega Handheld Temperature/Relative Humidity Meter i. Indoor usage 1. Perform every day in center of CLIC 2. Turn on to percentage relative humidity setting 3. Hold probe in air approximately 5 ft off ground for 3-5 minutes 4. Record ii. Outdoor usage 1. Perform every 8 hours 2. Same location as outdoor thermocouple 3. Turn on to percentage relative humidity setting 4. Hold probe in air approximately 5 ft off ground for 2-3 minutes 5. Record c. Measure outside environment air velocity with Omegaflo anemometer i. Perform outside of building away from structures ii. Select the 30 m/s velocity range setting iii. Hold probe approximately 5 feet off ground xxxviii

51 iv. Record an average air velocity v. Repeat every 8 hours d. Determine exit air temperature and relative humidity of evaporative cooler using Omega Handheld Temperature/Relative Humidity Meter i. Turn on to temperature setting ii. Place probe at exit vent of evaporative cooler iii. Record temperature iv. Switch to relative humidity setting v. In same location, hold prove for 3-5 minutes vi. Record vii. Perform every hour of operation 3. End control testing Monday, March 21 Begin Cooling 1. Implement shading barrier 2. Plug evaporative cooler into power inverter during daylight 3. Turn on evaporative cooler 4. Take measurements a. Take readings from thermocouples i. Ground & Geothermal thermocouples 1. Every 8 hours ii. Data logger thermocouples 1. Every 10 minutes b. Determine relative humidity using Omega Handheld Temperature/Relative Humidity Meter i. Indoor usage 1. Perform every day in center of CLIC 2. Turn on to percentage relative humidity setting 3. Hold probe in air approximately 5 ft off ground for 3-5 minutes 4. Record ii. Outdoor usage 1. Perform every 8 hours 2. Same location as outdoor thermocouple 3. Turn on to percentage relative humidity setting 4. Hold probe in air approximately 5 ft off ground for 2-3 minutes 5. Record c. Measure outside environment air velocity with Omegaflo Anemometer i. Perform outside of building away from structures ii. Select the 30 m/s velocity range setting iii. Hold probe approximately 5 feet off ground iv. Record an average air velocity v. Repeat every 8 hours xxxix

52 d. Determine exit air temperature and relative humidity of evaporative cooler using Omega Handheld Temperature/Relative Humidity Meter i. Turn on to temperature setting ii. Place probe Location? iii. Record temperature iv. Switch to relative humidity setting v. In same location, hold prove for 3-5 minutes vi. Record vii. Perform every hour of operation 5. Plug evaporative cooler into wall power at night Tuesday, March 22 through Thursday March 24 Continue Cooling and Gathering Data 1. Take measurements a. Take readings from thermocouples i. Ground & Geothermal thermocouples 1. Every 8 hours ii. Data logger thermocouples 1. Every 10 minutes b. Determine relative humidity using Omega Handheld Temperature/Relative Humidity Meter i. Indoor usage 1. Perform every day in center of CLIC 2. Turn on to percentage relative humidity setting 3. Hold probe in air approximately 5 ft off ground for 3-5 minutes 4. Record ii. Outdoor usage 1. Perform every 8 hours 2. Same location as outdoor thermocouple 3. Turn on to percentage relative humidity setting 4. Hold probe in air approximately 5 ft off ground for 2-3 minutes 5. Record c. Measure outside environment air velocity with Omegaflo anemometer i. Perform outside of building away from structures ii. Select the 30 m/s velocity range setting iii. Hold probe approximately 5 feet off ground iv. Record an average air velocity v. Repeat every 8 hours d. Determine exit air temperature and relative humidity of evaporative cooler using Omega Handheld Temperature/Relative Humidity Meter i. Turn on to temperature setting ii. Place probe at exit vent of evaporative cooler iii. Record temperature xl

53 iv. Switch to relative humidity setting v. In same location, hold prove for 3-5 minutes vi. Record vii. Perform every hour of operation xli

54 Kangaba CLIC Diary Appendix J Saturday, March 19 Travel to Kangaba Unpack equipment Make sure all equipment is working and ready to use Check out CLIC Come up with alternative ideas o Connecting geothermal piping to evaporative cooler and building o Dealing with double roof Set up data logger and install software Take few initial measurements o Indoor and outdoor temperatures o Ceiling temperature o Indoor and outdoor relative humidity Meet Summa Reddy, Peace Corps volunteer Find cook Get acquainted with town Sunday, March 20 Set up all thermocouples (indoor and outdoor) Get holes dug for ground temperature testing Insert thermocouples into holes and take temperatures Begin testing at 10 am o Initial control testing without evaporative cooler and doors and windows shut Initial setup of shading barrier on roof o Discover bolts on roof o Rip heat barrier o Buy fabric in town for shading Meet Leslie, Peace Corps volunteer Monday, March 21 Finish control testing Set up shading barrier o Fabric on bottom o Heat barrier on top o Rocks to hold down material o Fabric blows in wind, put down more rocks xlii

55 Tuesday, March 22 Add thermocouples to shaded and unshaded sections of the roof Set up evaporative cooler contraption Room conditions: windows and doors open Wednesday, March 23 Room conditions: Doors and windows closed for the majority of the time Packed up xliii

56 Kangaba CLIC Details Appendix K Double roof o Corrugated metal on top o Metal ceiling tiles on bottom layer o Layers are approximately 0.6 m apart Metal bolts protrude approximately 5 cm above corrugated metal roof o Located approximately every 0.9 m Building is much larger than expected o Complete building is approximately 30m x 12m x 4m (length, width, and height, respectively o CLIC portion is approximately 10m x 12 m Steps Storag e CLIC Windows Classroom Geothermal piping Evaporative cooler is not in window o Located and operated on floor of CLIC In the front of the CLIC, facing the steps, there are vents that cannot be closed, approximately 0.6 m in height at the top of the wall The ground is rock, so the ground temperature pole did not work The walls are stucco, approximately 20 cm thick xliv

57 Ground Temperatures Appendix L Ground Temperature at Varying Depths Temperature (Fahrenheit) cm 70 cm 55 cm 40 cm 25 cm /20/05 0:00 3/20/05 12:00 3/21/05 0:00 3/21/05 12:00 3/22/05 0:00 3/22/05 12:00 3/23/05 0:00 3/23/05 12:00 3/24/05 0:00 Time Ground Temperature (Degrees Fahrenheit) Date and Time 85 cm 70 cm 55 cm 40 cm 25 cm Hours 3/20/05 12: /20/05 16: /20/05 17: /21/05 7: /21/05 12: /21/05 15: /22/05 8: /22/05 12: /22/05 14: /22/05 18: /23/05 7: /23/05 10: /23/05 12: /23/05 14: /23/05 16: /23/05 18: xlv

58 Relative Humidity Appendix M Relative Humidity (%) Date and Time inside outside notes 3/19/05 14: /20/05 9: /20/05 17: /21/05 8: CLIC occupied 3/21/05 12: /22/05 8: occupied with doors and windows open 3/22/05 9: in side room 3/22/05 18: /23/05 7: xlvi

59 Geothermal Pipe Temperatures Appendix N Geothermal Pipe Temperatures Date and Time Inlet ( F) Outlet ( F) Notes 3/22/05 12: temp increased with fan speed 3/22/05 12:30 94 after running for 15 min 3/22/05 14: /22/05 15: /22/05 16: /22/05 18: /23/05 7: /23/05 10: Omega 3/23/05 10: TSI 3/23/05 12: TSI 3/23/05 12: Omega 3/23/05 14: TSI 3/23/05 14: Omega 3/23/05 16: TSI 3/23/05 16: Omega 3/23/05 18: TSI 3/23/05 18: Omega Pipe Temperature Readings Before Fan Installed Date and Time Entrance Temp ( F) Exit Temp ( F) 3/20/05 8: m m 82 1 m 92 1 m m m 84 2 m 96 2 m m m 83 3 m m 84 4 m 83 Ground Near Pipe Date and Time Temp at 50 cm ( F) Notes 3/21/05 7: /22/05 8: /22/05 12:00 Loose connection? xlvii

60 Telephonic Temperatures Appendix O Telephone Room Measurements (Degrees Fahrenheit) Date Time Roof Temp ( F) Inside Temp ( F) Outside in Shade ( F) 3/22/ /22/ /22/ /22/ /22/ /22/ /22/ xlviii

61 Air Velocity Appendix P Air Velocity (ft/min) Pipe Inlet Pipe Outlet Date Time Outside 3/20/ /20/ /21/ /22/ /22/ /22/ /23/ /23/ /23/ xlix

62 Other Temperatures Appendix Q Other Readings (Degrees Fahrenheit) Date Time Measurement Description 3/19/ Ceiling temp 3/19/ Surface ground temp 3/19/ Inside floor temp 3/19/ Outside air temp 3/19/ Inside air temp 3/22/ Side room temp These are some of the initial temperature readings we took on our first visit to the Kangaba CLIC before we set up our test instruments. l

63 Kangaba CLIC Outside Air Temperature Appendix R Outside Air (ch6) /20/05 0:00 3/20/05 12:00 3/21/05 0:00 3/21/05 12:00 3/22/05 0:00 3/22/05 12:00 3/23/05 0:00 3/23/05 12:00 3/24/05 0:00 Time Temperature (F) li

64 Kangaba CLIC Roof Temperatures Appendix S Roof Temp (ch2) Shaded Roof (ch8) Unshaded Roof (ch9) 40 3/20/05 0:00 3/20/05 12:00 3/21/05 0:00 3/21/05 12:00 3/22/05 0:00 3/22/05 12:00 3/23/05 0:00 3/23/05 12:00 3/24/05 0:00 Time Temperature (F) lii

65 Kangaba CLIC Ceiling Temperatures Appendix T Ceiling 1 (ch3) Ceiling by Door (ch7) /20/05 0:00 3/20/05 12:00 3/21/05 0:00 3/21/05 12:00 3/22/05 0:00 3/22/05 12:00 3/23/05 0:00 3/23/05 12:00 3/24/05 0:00 Time Temperature (F) liii

66 Kangaba CLIC Room Temperatures Appendix U Room Air (ch1) Room Air High (ch4) 82 3/20/05 0:00 3/20/05 12:00 3/21/05 0:00 3/21/05 12:00 3/22/05 0:00 3/22/05 12:00 3/23/05 0:00 3/23/05 12:00 3/24/05 0:00 Time Temperature (F) liv

67 Ouéléssébougou CLIC Details Appendix V March 25, 2005 Contacts clicoulessebougou@clicmali.org Madame Yattara Oumou Sow-Genante, Manager. Sowoumou2000@yahoo.Fr Abdoulahe Sanogo, Trainer. Sanogoahd@yahoo.Fr CLIC Specs S Doors, 1.2 m W 4.5 m outside 4.0 m Satellite dish (on roof) Fans Office/ Side room 2.3 m 6 m 2.9 m Roof o Cement o 43 cm high border around entire perimeter o 21.5 cm thick Walls o Stucco o 20.3 cm thick o Outside: 3.9 m height (including edge around roof) o Inside: 3 m floor to ceiling Windows o 5 total o 109 cm wide o 111 cm height From our observations, the best location for the evaporative cooler and geothermal piping would be on the east side of the CLIC o Shading o No obstructions (open wall) lv

68 Ouéléssébougou CLIC Diary Appendix W Meeting with Ian Howard, ESC/Geekcorp ihoward@netdotworking.com Entire town run on diesel generator Prefer electronics to run on solar panels During hottest time of day is when most brown-outs occur o At times, not enough power to turn on computers Budget for pilot site: $500 to $1000 Can import fans Don t worry about powering (solar panels) Concentrate on locally made products 3 to 4 people use internet for 1 to 2 hours per day o Computer on only during service o More people come in to use photo copier Exhaust heat from computers? Create design with dish in place Look into using fewer UPS s to run computers 3 CLICs have metal roofs The remaining CLICs have cement roofs lvi

69 Preliminary Project Management Appendix X Preliminary Research Testing/ Experimentation and Design Contact and meet with advisor and client and Mali Peace Corps Brainstorm Plan project/set up Research current deadlines CLICs Evaluate researched options Prototype Testing Final Design Final Solid works Modeling Ship necessary supplies to Mali Address Design Problems QFD Buy parts Travel to Mali Rework/ Modifications Gantt Chart Research Mali Tradeoff Matrix Build prototypes Install Research Alternative Energy Pick 2 or 3 best options Combine 3 prototypes to one final prototype Solar Panels Preliminary Design Dig Trenches in Mali Research Cooling Methods Solid works Design Thermoacoustic Hand Calculations Meets Requirements (general performance) Evaluate temperatures, mass flow rate, humidity, power consumption, etc. Modified Evaporative Cooler Preliminary Experimentation Research (see attached WBS Passive Methods in Appendix H) Geothermal Cooling Contact CLICs for approval Architectural Modifications Shading Barrier lvii

70 Work Breakdown Structure (WBS) Appendix Y Testing and Experimentation 45 days 1/3/2005 8:00 3/3/ :00 Shading Barrier 10 days 1/10/2005 8:00 1/21/ :00 Designed Experiment 5 days 1/10/2005 8:00 1/14/ :00 Stefan Ran Experiment 5 days 1/17/2005 8:00 1/21/ :00 Stefan Evaporative Cooler 20 days 2/7/2005 8:00 3/3/ :00 Test Normal Function 1 day 3/2/2005 8:00 3/2/ :00 water consumption 1 day 3/2/2005 8:00 3/2/ :00 Jenny relative humidity 1 day 3/2/2005 8:00 3/2/ :00 Jenny power consumption 1 day 3/2/2005 8:00 3/2/ :00 Clare Testing with designed connector/adaptor 1 day 3/3/2005 8:00 3/3/ :00 water consumption 1 day 3/3/2005 8:00 3/3/ :00 Jenny relative humidity 1 day 3/3/2005 8:00 3/3/ :00 Jenny Retrofitting 11 days 2/7/2005 8:00 2/18/ :00 Designed Hook-up 5 days 2/7/2005 8:00 2/11/ :00 Stefan Built Hook-up 5 days 2/14/2005 8:00 2/18/ :00 Stefan Femlab 33 days 1/3/2005 8:00 2/15/ :00 Improvements on existing code 1 day 1/3/2005 8:00 1/3/ :00 Clare Code for varying length, radius and air velocity 1 day 2/13/2005 8:00 2/13/ :00 Clare Converging solution 1 day 2/14/2005 8:00 2/14/ :00 Clare Experimented with varying geometries 1 day 2/15/2005 8:00 2/15/ :00 Clare Prototyping 31 days 2/2/2005 8:00 3/15/ :00 Shading Barrier 31 days 2/2/2005 8:00 3/15/ :00 1st Design 5 days 2/2/2005 8:00 2/8/ :00 Kelly/Jenny Evaluation 3 days 2/2/2005 8:00 2/4/ :00 Kelly/Jenny Construction 3 days 2/4/2005 8:00 2/8/ :00 Kelly/Jenny 2nd Design 12 days 2/7/2005 8:00 2/21/ :00 Kelly and Jenny 3rd Design 21 days 2/15/2005 8:00 3/15/ :00 Design 21 days 2/15/2005 8:00 3/15/ :00 Kelly and Jenny Set-up 1 day 3/3/2005 8:00 3/3/ :00 Kelly/Jenny Testing 9 days 3/3/2005 8:00 3/15/ :00 Kelly/Jenny Modifications 9 days 3/3/2005 8:00 3/15/ :00 Kelly/Jenny Evaporative Cooler 6 days 3/3/2005 8:00 3/10/ :00 Redesigned connector housing 1 day 3/3/2005 8:00 3/3/ :00 Stefan Built connector housing 5 days 3/4/2005 8:00 3/10/ :00 Stefan Geothermal Construction 7 days 2/25/2005 8:00 3/7/ :00 Solid Works Drawings 1 day 2/25/2005 8:00 2/25/ :00 Stefan Approval 1 day 2/25/2005 8:00 2/25/ :00 Contact Dennis Bilodeau 1 day 2/25/2005 8:00 2/25/ :00 Stefan Purchase of Piping and Trench Construction 6 days 2/28/2005 8:00 3/7/ :00 Contact Dennis Bilodeau 1 day 2/28/2005 8:00 2/28/ :00 Stefan Contact Mohamed Ag Acharom 1 day 3/7/2005 8:00 3/7/ :00 Jenny Confirm Digging 1 day 3/7/2005 8:00 3/7/ :00 Stefan Construction of Ground 5 days 2/25/2005 8:00 3/3/ :00 Kelly/Jenny lviii

71 Thermocouple Stake Design 1 day 2/25/2005 8:00 2/25/ :00 Kelly/Jenny Construction 4 days 2/28/2005 8:00 3/3/ :00 Kelly/Jenny Research Alternative Methods 13 days 2/25/2005 8:00 3/15/ :00 Solar Panels 13 days 2/25/2005 8:00 3/15/ :00 Design 4 days 2/25/2005 8:00 3/2/ :00 Clare/Stefan Testing 8 days 3/4/2005 8:00 3/15/ :00 Clare/Stefan Travel 44 days 1/31/2005 8:00 3/26/ :00 Prepare for Travel 33 days 1/31/2005 8:00 3/15/ :00 Created Itinerary 5 days 1/31/2005 8:00 2/4/ :00 Stefan Create Packing List 1 day 2/2/2005 8:00 2/2/ :00 All Gather and Test equipment 1 day 2/16/2005 8:00 2/16/ :00 Clare/Kelly/Jenny Design and Create Experimental Protocol for testing in Mali 17 days 2/21/2005 8:00 3/15/ :00 Kelly/Jenny Create Excel Spreadsheets for usage in Mali 1 day 3/7/2005 8:00 3/7/ :00 Kelly Editing 17 days 2/21/2005 8:00 3/15/ :00 Kelly/Jenny Ordering of Equipment 8 days 2/28/2005 8:00 3/9/ :00 All Create Survey 1 day 3/9/2005 8:00 3/9/ :00 Jenny Bamako 9 days 3/18/2005 8:00 3/26/ :00 All Purchase Cotton Hangings 1 day 3/18/2005 8:00 3/18/ :00 All Foil 9 days 3/18/2005 8:00 3/26/ :00 All Research Pad Material 9 days 3/18/2005 8:00 3/26/ :00 All Kangaba 6 days 3/19/2005 8:00 3/24/ :00 All Installation in Mali 2 days 3/19/2005 8:00 3/20/ :00 All Control Testing 2 days 3/19/2005 8:00 3/20/ :00 All Testing and Collecting of Data 4 days 3/21/2005 8:00 3/24/ :00 All Research Pad Material 6 days 3/19/2005 8:00 3/24/ :00 All Rework/Modifications 20 days? 3/28/2005 8:00 4/22/ :00 All Unpack 1 day 3/30/2005 8:00 3/30/ :00 All Address Design Problems 20 days? 3/28/2005 8:00 4/22/ :00 All Organize Data from trip 4 days 4/1/2005 8:00 4/6/ :00 Kelly/Clare Organize notes from trip 3 days 4/1/2005 8:00 4/5/ :00 Jenny Solid works Drawings of CLIC 4 days 4/1/2005 8:00 4/6/ :00 Stefan Shade Windows and Doors 15 days 4/4/2005 8:00 4/22/ :00 Research Materials 15 days 4/4/2005 8:00 4/22/ :00 Kelly Design 15 days 4/4/2005 8:00 4/22/ :00 Kelly Geothermal Duct 15 days 4/4/2005 8:00 4/22/ :00 Ducting Material 15 days 4/4/2005 8:00 4/22/ :00 Jenny Justify location 15 days 4/4/2005 8:00 4/22/ :00 Jenny Evaporative Cooler 20 days? 3/28/2005 8:00 4/22/ :00 Wall Location 15 days 4/4/2005 8:00 4/22/ :00 Stefan Geothermal Connection 15 days 4/4/2005 8:00 4/22/ :00 Stefan Pad location and water distribution 1 day? 3/28/2005 8:00 3/28/ :00 Stefan Fans 15 days 4/4/2005 8:00 4/22/ :00 Clare Solar Panels (optional) 15 days 4/4/2005 8:00 4/22/ :00 Write Implementation Report for CLIC 10 days 5/9/2005 8:00 5/20/ :00 All lix

72 Evaporative Cooler Modifications 10 days 5/9/2005 8:00 5/20/ :00 Stefan/Clare Geothermal Ducting 10 days 5/9/2005 8:00 5/20/ :00 Jenny Location of Ducting and Evaporative Cooler 10 days 5/9/2005 8:00 5/20/ :00 Jenny Shading Modifications 10 days 5/9/2005 8:00 5/20/ :00 Kelly Brochure and Poster 16 days 4/22/2005 8:00 5/13/ :00 Brochure Draft 2 days 4/22/2005 8:00 4/25/ :00 Kelly Revisions 5 days 4/25/2005 8:00 4/29/ :00 All Poster 5 days 5/9/2005 8:00 5/13/ :00 Jenny Final Paper 19 days 4/22/2005 8:00 5/18/ :00 All Individual Sections 6 days 4/22/2005 8:00 4/29/ :00 All Draft to Dr. George 10 days 5/2/2005 8:00 5/13/ :00 All Editing 4 days 5/13/2005 8:00 5/18/ :00 All Presentation 6 days 5/6/2005 8:00 5/13/ :00 All lx

73 List of Materials Appendix Z Description Quantity Light Colored Cotton Fabric 14.4 m 2 Curtain Rods 16 Lakewood 57-Watt Fans 2 Sheet Metal 1.18 m Volt Power Outlet 1 20 centimeter diameter Clay Piping 20 m Caulk Rivets Gauge Wire lxi

74 Implementation Proposal Appendix AA IMPLEMENTATION PROPOSAL FOR OUÉLÉSSÉBOUGOU CLIC, MALI: USAID/MALI University of St. Thomas, Minnesota May 2005 lxii

75 FINAL DESIGN DESCRIPTION Ouéléssébougou CLIC Overview The CLIC in Ouéléssébougou is a more common building layout for the Mali CLICs. It is located in a single room building which is much smaller than the Kangaba CLIC. The building is 4.5 meters by 9 meters. It contains a large computer room as well as a smaller office. There are 8 computers, a printer/copier, fax machine, radio, television, VCR, and DVD player. The roof is made of cement and has a border, 43 centimeters high, 21 centimeters thick, that surrounds the entire perimeter. The walls are 21 centimeters thick and made of stucco. Inside walls are 3 meters tall, floor to ceiling. Outside walls are 4 meters in height from the ground to the border around the roof. There are 5 widows, 109 centimeters wide, 112 centimeters tall. There are two doors, each 1.2 meters wide. Figure 1: Ouéléssébougou CLIC S Doors, 1.2 m W 4.5 m 4 m Satellite dish (on roof) Fans Office/ Side room 2.3 m 6.1 m Figure 2: Ouéléssébougou CLIC 2.9 m lxiii

76 The entire town of Ouéléssébougou is run on a diesel generator. Most brown-outs occur during the hottest time of the day. At times, there is not enough power to turn on the computers. Three to four people use the internet for one to two hours per day. The computers are only turned on while they are in service. Most people come to use the photo copier. The budget for the Ouéléssébougou pilot site is between $500 and $1000. Concentration on local products is recommended. Importation of fans is allowed. Powering with solar panels is unnecessary at this point. The design should be created with the dish in place. Only three of the CLICs in Mali have metal roofs, and the remaining have cement roofs. Modified Evaporative Cooler In an attempt to make the evaporative cooler more effective and cheaper to operate, modifications are needed. Since the evaporative coolers in Mali were not properly installed, they were not effective in cooling the rooms, and would eventually make the inside of the building so humid that it would be worse than without using the evaporative cooler. A permanent hookup to the geothermal cooling system that allowed access to the inside of the cooler to add water was designed. More efficient fans were found and mounted in the cooler to reduce power requirements. A passive system of soaking the pads was considered to lower power requirements, but rejected because the need for constant flowing water for the pads to be effective. Additionally alternative pads that could be locally manufactured were researched, but contacts in Mali working on the problem said that no alternative could be found. In order to modify the evaporative cooler the inside of the evaporative cooler was stripped. Instructions on how to do this are attached. A sheet metal cover that surrounded the location of one of the side pads, and extended 2.5 inches away from the cooler was built. This box has a hole to connect to the geothermal cooling pipe and can be seen in Figure 3. lxiv

77 Figure 3: View of Modified Evaporative Cooler The hole is cut at a height that matches up with the geothermal pipe, coming in through the wall, when the cooler is resting on the floor. The pad was slid into place and this cover was pop riveted to the cooler, and the geothermal pipe was connected to the hole in the sheet metal. Both were sealed with caulk so the cooler would draw as much air as possible through the geothermal cooling system. In this design this pad would be the only pad in use. The face on the opposite side is used to mount the fans. To mount the fans in this side, the pad would be removed and a piece of sheet metal is attached in its place covering the entire opening. The sheet was attached using pop rivets and sealed airtight using caulk. The fans have a lip surrounding the cage that is used to hold the two pieces of the fan housing together using screws. To mount the fans inside the evaporative cooler, the holes for these screws were used. The sheet has two circular holes in it with diameters of 25.5 cm. The size of these holes matches the diameter of the fans, minus the lip of the cage, which is used to mount the fans. Holes were drilled in the sheet metal to match up with the holes for the screws in the housing. The fans were then mounted into the evaporative cooler using longer.3 cm screws. These fans were found to use 145 Watts together and still blow 240 CFM of air through a simulated geothermal pipe. This is a substantial decrease in power from the original fan (which used 303 Watts with the pump) and moved over twice as much air through the geothermal pipe as the original fan, which moved 105 CFM. lxv

78 Figure 4: Fan Mounting The face of the evaporative cooler that originally had the exhaust for the cooled air is used to mount the controls for the fans and pump. The on/off switch box was removed from the cooler. The wires for the original fan were removed, but the wires from the plug and to the pump are still used. An outlet with two sockets was attached to the top face of the on/off switch. The two fans are plugged into the outlet and the outlet is connected to the on/off switch that controls the fan. Any exposed electrical wire was covered using electrical tape. The box was mounted to the metal lip that used to be attached to the fan housing so that it is flush with the outer surface of the evaporative cooler. This face was then covered with a piece of sheet metal with an opening for the switch, attached with pop rivets, and sealed with caulk. Figure 5: Switch Mounting lxvi

79 One of the panels remains removable to allow the users to add water to the cooler, and to allow for replacement of the pad if necessary. For this purpose the panel opposite the original evaporative cooler exhaust was designed to be removable. The bag of pad material was removed from its metal frame and a piece of sheet metal was cut that would cover all the openings on the frame. The sheet metal was attached with pop rivets and sealed airtight using caulk. The frame was then slid back into place. When in place this panel would restrict air flow enough to cause air to be drawn through the panel attached to the geothermal cooling system, but could be removed to allow a person to replace a pad from the inside of the cooler, and add water as needed. To make sure that the water did not leak out of the cooler a small piece of sheet metal was used to cover the opening on the bottom of the evaporative cooler. This sheet was also attached with pop rivets and sealed using caulk. Evaporative Cooler Fan Figure 6: Side Panel Removed Since one of the goals of our project was to reduce the amount of energy consumed by the evaporative coolers that the CLIC s currently posses, we looked at the components of the cooler to see what we could change. The two components that require energy are the water pump and the fan. The pump only required 29 Watts, but the fan needed 360 Watts to operate, so one of the possibilities was to replace the existing fan with a more power efficient one. We wanted to reduce the power consumption without compromising much on the volumetric air flow rate. To test the air flow, we built a pipe out of sheet metal with an opening that was restricted enough to simulate the drop in pressure that our underground pipe in Mali would produce. This drop in pressure is due to the geometry of the pipe (i.e. friction due to the number of bends and length of the pipe), and is related to the dimensions of the pipe as shown in equation 1.1. We started with a 6 diameter pipe, and found the diameter of the restricted opening with the following calculations: lxvii

80 1 2 L 1 2 Δ Po = ρ Vo = (6kb + f + 1) ρvd (1.1) 2 D 2 4 d 2 2 Dd V = o Vd (1.2) Do Where the pipe in Mali, and Δ P is the pressure drop, ρ is the air density, k b =0.8, f=0.02, L is the length of D d is the diameter. We can substitute (1.2) into (1.1) to get: D D d o 4 = 6k b + f L D d + 1 (1.3) Knowing the diameter of the pipe in Mali, D d =20 cm or 7.87 in, we solved for the diameter of the restricted opening, D o, and found that it equaled 4.7 in. We then fit a cardboard piece with a 4.7 inch hole to the end of our test pipe. We attached the fan to the opposite end of the pipe and measured the air velocity through a small hole in the pipe near the fan. We used the air velocity to calculate the volumetric flow rate. The first fan that we tested was a 6 diameter, 14 Watt, Air King brand fan that was rated to provide 1637 CFM. We found that this information was inaccurate however and also that the fan was very sensitive to the restricted air flow of the pipe. It only produced about 74 CFM through the pipe. After looking at several other fans, we purchased a Honeywell fan for $14.99 that requires 57 Watts of power, and provides between 800 and 1000 CFM on its own. To ensure that we would get enough air flow through the pipe and the pad, we purchased two of these fans to install in the cooler. These were the most efficient fans that we could find for a low price that provided enough air flow for our application. We connected these two fans to the existing switch by simply adding an outlet inside the cooler where the two fans would remain plugged in and turned to the on position. We then wired the outlet to the outer switch so that when the switch was turned on, both fans would be turned on high. Geothermal Piping For the original testing in Mali, PVC piping was used as the ground piping for the geothermal cooling. While in Mali, the group researched additional materials to supplement the imported and costly PVC piping. Clay pottery piping was found to be a viable replacement material for the geothermal ground piping. Currently, these clay pipes are manufactured and used in Mali as rain gutters on housing. The decision to implement this material for the underground piping is based off of several factors. First, several producers manufacture the current clay rain gutters locally. The manufactures are easily able to custom make the piping to the specific diameters, thicknesses and piece lxviii

81 sizes necessary. Contacts in Bamako cited an estimate of $15 to $20 for a 20 cm diameter, 50 cm length piece of piping. Clay pottery is also more environmentally friendly and will biodegrade with time. Figure 7: Geothermal Piping Design The clay piping also possesses technical benefits over some manmade material, such as PVC. For instance, the thermal conductivity, the rate at which heat flows through a material, is approximately five times greater for clay pottery versus PVC. A higher rate of conduction or heat flow is preferred in this situation since it is desired that the material will easily transfer the heat of the warmer air to the cooler surrounding earth, thus cooling the passing air. Material Thermal Conductivity, k (W / m-k) PVC 0.19 Fired Clay 1.0 Table 1: Thermal Conductivity Values for Materials Placement Location of Geothermal Piping and Evaporative Cooler The greatest factor in determining the location for the placement of the geothermal piping is the movements of the sun. In order to insure the coolest ground temperatures available, the optimal location for the piping is situated on the east side of the CLIC building. This conclusion was reached through researching the sun s daily and yearly lxix

82 paths at Bamako s latitude using an applet model of the sun s path sponsored by the Australian National University. At N, the sun s path travels for eight months of the year, September to April, in the southern hemisphere. This is known as the dry season in Mali. For the remaining four months, May through August, also known as the wet season, the sun travels in the northern hemisphere. Therefore, the southern wall of the CLIC receives exposure from the sun for eight months and shade for four months. The opposite is true for the northern wall. Figure 8: Sun's Daily Course in July (left) and in January (right) In addition to traveling between the southern and northern hemispheres throughout the year, the sun travels daily from the eastern hemisphere to the western hemisphere. This daily movement has the greatest effect upon the placement location of the geothermal piping. The eastern ground and side of the building experience the morning sun during the coolest daylight period. During the afternoon, the hottest period of the day, the eastern ground and wall are located in the shade while the western ground and wall are exposed. Figure 9: Sun's Daily Course and Noted Shading at 9 am (left) and 3 pm (right) during September lxx

83 Therefore, the ideal ground location for the geothermal piping is on the northeastern side of the CLIC. Shading is realized throughout the day in this location during the dry season, the time period being addressed in this project. In addition, the eastern portion receives shading during the afternoon, the hottest period of the day. For the pilot site, the ground east of the building is free of landscape. Additionally, this is the best site for the placement of the evaporative cooler in the Ouéléssébougou CLIC. The eastern wall is uninterrupted, lacking windows or technical equipment. Shading Barrier Figure 10: Building Diagram for Ouéléssébougou CLIC The shading design for the Ouéléssébougou CLIC does not include shading of the roof since it has a single cement roof. However, shading of the doors and windows is an option that could be used to block much of the solar radiation. The shading material is light colored cotton, which is easily found in Mali. The material is sewn around two curtain rods located at the top and bottom of each window, shown in Figure 16. The curtain rods are fastened to the outside of the building using hooks and screws, commonly used with all curtain rods in Mali. Prior to sewing, the material for each window measures 138 centimeters in height by 130 centimeters in width. The shading design for the doors is very similar to that of the windows. However, rather than the curtain rods hanging on the building, they will be placed on the top and bottom of each door. There are two doors in the Ouéléssébougou CLIC; however one is a double door. Prior to sewing, the material for the single door measures 229 centimeters in height by 118 centimeters in width. The material for the each of the double doors measures 229 centimeters in height by 59 centimeters in width. The total material used is 14.4 square meters. Depending on the width of the fabric used, the material may need to be pieced together. An 8 centimeter seam will be sewn on both the top and bottom of each piece of fabric in order to insert the curtain rod. Since the windows already have curtains on the lxxi

84 inside of the building, shading on the outside may not provide a significant difference, therefore shading may or may not be used on the Ouéléssébougou CLIC. Figure 11: Window Shading Barrier lxxii

85 Attachment 1: Instructions for connecting the outlet to the evaporative cooler switch 1. Original switch has three connections: Switch High Hot Low 2. Using 18-gauge wire for all connections. 3. Connect the hot wire on the outlet to the high connector on the switch. 4. Connect the white wire on the outlet to the neutral wires of the pump and switch. 5. Connect the hot wire of the power cord to the hot connector on the switch. 6. The low connector on the switch is not used. 7. Connect all the ground wires of the power cord, pump, and outlet together and to the metal casing of the switch. 8. The outlet remains on the inside of the evaporative cooler with both fans plugged in and set to high. 9. When the switch is turned to high, both fans are turned on. When the switch is turned to off or low, both fans are turned off. 10. The pump remains connected to the switch as it was originally. lxxiii

86 Attachment 2: Instructions for Modifying the Evaporative Cooler 1) Remove the four screws on the face plate and then remove the face plate 2) Remove the two screws holding the panel in place and open the panel and open panel lxxiv

87 3) Remove the green ground screw and cut the wires attached to the black box 4) Remove all wire connectors and separate the wires lxxv

88 5) Disconnect all the wires from the switches and remove clips attached to wire clusters 6) Open side panel 7) Pull wires into main housing lxxvi

89 8) Remove 4 screws holding fan motor in place and remove fan 9) Remove 12 screws holding exhaust port in place and take off port 12) Pry motor housing away from evaporative cooler frame lxxvii

90 13) Attach the outlet to the top of the switch housing. Connect the wires to the switch housing as described in the wiring section. Drill pilot holes and mount the switch housing to the left lip of exhaust opening with the face of the switch flush with the evaporative cooler. 14) Remove all panels from the evaporative cooler except for the one on the right side (when facing the exhaust port). Remove the metal bar, cage and panel from the panels. Cut a piece of sheet metal 58cm X 43 cm. Line the edge of the panel with caulk. Place the panel over the vents on the metal evaporative cooler pad frame and allow the caulk to dry (4 hours). Drill holes 1.25 cm in from the edge the same size as the rivets (1cm recommended) and rivet the sheet metal to the panel, three on the short sides, four on the long sides, and line the edges with caulk to ensure an air tight seal. lxxviii

91 15) Cut piece of sheet metal that is 28cm X 38cm and a hole 7 cm in from the left and top that is 3.5cm x 11 cm (big enough to allow the access to the switch) with a notch in the bottom for the power cord. Line the edges of the sheet metal with caulk and place it over the exhaust port with the hole over the switch and the power cord coming out of the hole and let the caulk dry. Drill holes 1.25 cm from the edge, rivet the panel to the evaporative cooler frame (3 on the short side 4 one the long) and line the edges of the sheet metal with caulk. lxxix

92 16) Cut a piece of sheet metal to the dimensions shown. Fold it in on the red dotted line and out on the green dotted line (90 degrees). Cut a 20cm diameter hole in the panel that will match up with the geothermal pipe when attached to the cooler and the cooler is resting on the floor. lxxx

93 17) Line the edges of the sheet metal box with caulk and place it over the installed pad and frame (the right side when facing the exhaust) and allow the caulk to dry. Drill holes 1.25 cm from the edge, rivet the box to the evaporative cooler (3 on the short side 4 on the long side) and seal all edges and opening with caulk. 18) Cut a piece of sheet metal 48.5cm x 56 cm. Cut two 25 cm holes in the sheet metal, one centered 15.5 cm from the top and 15.5 cm from the right, and one 15.5 cm from the bottom and 15.5 cm from the left. Separate the front cage from the fan by removing the screws. Drill four.3 cm holes in the sheet metal around the 25 cm holes that match up with the holes on the lip of the cage. Line the edge of the sheet metal with caulk, cover the face of the left panel (when looking at the exhaust) with the sheet metal and allow the caulk to dry. Drill holes and rivet the sheet metal to the evaporative cooler (three on the short sides four on the long) and seal the edges with caulk. Mount the fans on the sheet metal, with the front cage on the outside and a rear cage and the fan on the inside using longer.3 cm screws through the holes on cages and the holes in the metal. Plug the fans into the outlet. 19) Cover the hole in the bottom with a small piece of sheet metal and seal it with caulk lxxxi

94 20) Remove the black plastic spout from the top inside of the evaporative cooler. Fill the two holes of the spout that are over the sides without pads. Allow caulk to dry and replace the spout to the top inside of the evaporative cooler 21) Replace the panel that is covered in sheet metal to the rear side of the evaporative cooler (when facing the exhaust). lxxxii