University Energy Audit
 

Kathleen Corr
Chris Johnson
Dan Lajiness
Karen Park
 

ENVI 490 Energy Group - April 15, 1998
 

1. Abstract

 
 Our group's focus was specifically on energy usage and conservation at Rice. Our first goal was to conduct an objective study of the present energy use and efficiency on campus. This has shown that over the last five years, Rice has added four major buildings to its campus, yet campus-wide energy consumption has decreased by nearly one-fourth over the same period of time. This reduction in consumption came mainly through replacing a water chiller used in the central plant, changes in the usage of the cogeneration equipment, and replacement of lights throughout campus with more efficient ones through the "Green Lights" program.

 In order to properly compare buildings to each other and to the university as a whole, all of the metered buildings were categorized into 5 groups: science, academic, administration, colleges, and miscellaneous. Our research showed that the science group uses 41.20/0 of the total campus utilities and has the highest energy costs per square foot, $1.58/sf. Although the miscellaneous group had the next highest costs per square foot, $1.16/sf, it only comprises 11.6% of the campus total. The administration group uses 8.0% of the total, the colleges 16.2%, and the academic group 23.0% and rank third, fourth, and fifth respectively in terms of costs per square foot. See Figure 1 and 2 in the Appendices section at the end of this report for graphical representation of these figures.

 Student behavior is the primary source for excess energy use in the eight residential colleges. To gain a measure of the students' individual energy consumption patterns, a survey was distributed among on-campus students. The collected data suggested that, overall, triple rooms were most efficient in terms of appliances per room and per individual per room. Singles used the least energy overall, both as a room and on a per person basis, yet these figures did not include the heating and cooling costs per square foot per resident in a room.

 Computers without a "sleep mode" function costs the university $20,000 per year. A student-awareness campaign would help reduce the excess energy expenditures. The campaign would encourage students that do not have sleep modes on their computers to turn off their computers when not in use, and encourages students purchasing new computers to factor a sleep mode into their computer decision.

 Refrigerators are one of the most energy~consuming appliances in a dormitory room, and older models are significantly more inefficient than newer models. The group compared yearly energy usage and costs between a new compact refrigerator and an older model compact refrigerator to find that Rice spends an additional $10 on older refrigerators in single rooms alone. The university can alleviate this extra cost and energy use by implementing a system where the institution forms a contract with an appliance firm to either sell or lease newer refrigerators to students at rates that would make newer refrigerators more appealing than older ones. The university could also purchase the refrigerators to directly rent or sell the appliances to students. The university can alleviate this extra cost and energy use by implementing a system where the institution forms a contract with an appliance firm to either sell or rent newer refrigerators to students at rates that would make newer refrigerators more appealing than older ones. The university could also purchase the refrigerators to directly rent or sell the appliances to students.

 The energy group investigated photovoltaics as a potential energy-saving and environmentally friendly option for Rice. From researching, the feasibility of solar panels on the Rice campus, the energy group found that significant savings in electrical costs can be obtained through the provision of solar energy. However, the high initial costs of implementing this option turns out to be much more expensive than these savings. Further investigation is recommended.

 In hot and humid climates, many building design strategies have been shown to effectively minimize energy use. These strategies are most cost-effective for new buildings, and concentrate on reducing the demand and load on the lighting and Heating Ventilation and Air Conditioning (HVAC) systems.

2. Introduction

 The Energy Group is a component of the 1998 environmental research class at Rice University. Our mission was to determine the current energy use at Rice and to investigate technological improvements to this current state that are both environmentally-friendly and cost-effective. The first half of the semester was devoted to the former part of our mission. The second half investigated such options as improvements to the current buildings and energy use on campus, the potential for solar power, and suggestions for the design for the new buildings to be built.

 The Rice campus has been undergoing an accelerated process of change over the past few years, including the completion of several new buildings. In addition, plans for spending over $300 million dollars for the construction of new buildings have been approved. As the campus prepares for such change, the effect of university action on energy use and efficiency must be considered.

 In order to visualize any such affects it was necessary to first obtain an understanding of how energy is presently used and distributed on campus. This was achieved with the help of several contacts within Facilities and Engineering Jim Riley, Operations Manager, James Munoz, Staff Engineer, and Greg Dulaney, Central Plant Engineer. By investigating the present systems, equipment, and operations involving energy the group identified the areas with large energy demand.

 Rice University Food and Housing (F&H) Director Bob Truscott suggested that excess energy use in the colleges comes pnmarily from student behavior. The most feasible way to evaluate individual student behavior was to survey a random sampling of the on-campus student population to determine the number of appliances and energy consumers in each room. Data pertaining to the behavior of the students would guide the group to areas where excess energy use was greatest, areas where the group could potentially have the greatest impact. Changing student behavior can be a cheap and fairly easy way to conserve energy.

 The energy usage for computers with and without "sleep modes" was also compared. Sleep modes are a function built into the computer so that when the computer is on, but not in use for a set period of time, the computer automatically turns off the monitor while slowing down the CPU processes to a near stop so that the computer is "sleeping." Sleep modes differ from screen savers since screen savers do not slow down the computer's functions.

 Upon examination of the data collected in the survey, it was discovered that personal refrigerators consume the most energy in a dormitory room, more than twice as much energy as the second-largest consumer, computers. The amounts of energy personal refrigerators use vary by model and size, some consuming substantially less energy than others. From the information gathered in the survey, it was also noted that, on average, rooms of four people have close to two refrigerators per room.

 Two different models of 1.7-cubic foot refrigerators were examined, and their yearly energy usage and costs were evaluated. The average expenditures and energy usage for both old and new refrigerators were calculated to determine how much money the university could save with newer, more efficient refrigerators.

 The energy group investigated solar technology as a potential environmentally-friendly source of energy for use on the Rice University campus. This technology, namely the conversion of solar energy into electricity done by photovoltaics, seemed ideal for Rice and the Houston area given the large amounts of annual sunlight and large energy use on campus. From researching the best options for setting up photovoltaics on campus, we found that Rice can take advantage of tremendous savings in their current cost of providing electricity on campus. However, the initial costs of implementing this technology on campus appears to be much greater than these savings. Using the research in this paper as a starting point, continued investigation and monitoring of the potential feasibility for solar technology on campus is encouraged.

 At Rice University, expenditures for energy totaled over $3.4 million for the 1997 fiscal year. Economic interests obviously make minimizing these costs a priority. Improving the energy efficiency of new buildings in the Houston climate in a cost-effective manner though building design strategies has been demonstrated for similar climates. In fact, a Florida study has shown that strategies for increasing energy efficiency in institutions in a hot and humid climate can result in a 43% reduction in energy use compared to current practices (FSEC BDAC 1994).

3. Methods

3.1 Present University Energy Use
 Our group worked closely with the University class Bake 302, Understanding Environmental Systems, and with Facilities and Engineering to gain a basic knowledge of the present energy systems at Rice. Mr. Riley, Mr. Munoz, and Mr. Dulaney also explained the various measures that Rice has undergone to decrease energy consumption despite the continual growth of the university. Most importantly, the group received a Microsoft Excel spreadsheet from Mr. Riley containing energy usage and costs data for each metered building on campus for fiscal year 1997. With this information the group was able to make several calculations for each building to quantify its energy usage and efficiency. As noted above the buildings were separated into five groups according to their type of energy use (i.e. research buildings with large equipment versus administration buildings with mainly offices). The buildings in each group are as follows:

Science and Research: Abercrombie, Anderson Biological Laboratories, Dell Butcher Hall', Old Chemistry, George R Brown Hall, Keith - Wiess Geology, Mechanical Engineering, Mechanical Engineering Laboratories, Physics Laboratories, Ryon Engineering Laboratories, and Space Science.

Academic: Alice Pratt Brown Hall, Anderson Hall, Baker Institute, Duncan Hall, Fondren Library, Herman Brown Hall, Herring Hall, Rayzor Hall, and Sewall Hall.

Colleges: The colleges are metered in groups, North and South. The north side includes Jones and Brown Colleges, while the south side includes Baker, Hanszen, Lovett, Sid Richardson, Wiess and Will Rice Colleges.

Administration: Allen Center, Facilities and Engineering, Lovett Hall, and the RMC.

Miscellaneous: Continuing Studies, Media Center, Copy Club, President's House, Gym, Mudd, Central Kitchen, Hamman Hall, Cohen House, Stadium, and the Campus Police

 The calculations used for comparison were, by building:

1. Total utility use as a percent of the campus and building group's total
2. Each separate utility use as a percent of the campus and building group's total
3. Total energy costs per square foot

3.2 Energy Consumption Within the Residential Colleges
 The most effective way to obtain data that reflected student activities regarding energy use was to conduct a survey of the student population. A survey was composed which evaluated the number of appliances per room and the number of hours of daily computer usage. The survey asked for the respondent's college, the number of people in the respondent's room, the number of various appliances in the respondent's room, the number of hours the respondent's computer stays on, and whether the computer has a sleep mode or screen saver (Appendix 1). The survey aimed to target at least 10 percent of the on-campus student population. Four hundred surveys were sent to a random selection of on-campus students via email, and 177 responses were received.

 Using the number of each kind appliance for every set of rooms, the number of appliances per person and per room in each room type (single, double, triple, etc.) were found.

 The daily energy use for each type room was found by using, the number of hours each appliance was used a day, and the kWh used by each appliance, located in Table 6. The yearly costs of energy for each type of room were then derived from the daily energy consumption rates, considering that the residential colleges are fully occupied only nine months out of the year. The figures for each appliance's energy use were summed to find a total energy use per room type. These energy consumption rates were put in terms of kWh, and multiplied by a cost of $0.05 per kWh to find the daily and yearly expenditures on energy (Table 3).

 Next, figures dealing with computer usage were calculated, such as the overall average number of hours one computer is left on per day was figured, and the number of hours computers were left on per day in each of the different rooms. The percentage of computers with and without sleep mode, along with the average number of hours computers with and without sleep mode were left on, was also computed (Table 4).

 The differences between the energy consumed by computers with and without sleep modes was then found. It was estimated that, on average, students use computers four hours a day. A computer in sleep mode uses 30 watts of energy, while a fully-running computer uses 150 watts (MGE). With these figures and the average amounts of time computers with and without sleep modes were left on per day, the amount of energy used per day by computers with and without sleep modes was estimated. Yearly energy use for computers with and without sleep modes was then calculated using the daily energy use numbers, and the costs were determined on both a daily and yearly basis (Table 6).

 The wattage for a 1.7-cubic foot compact refrigerator in a college room was found to be 70 watts. Since students are in school nine months out of the year, this older model refrigerator used 460 kWh per year. General Electric was also contacted and questioned about their newest model of a 1.7-cubic foot refrigerator. The company reported that this type of refrigerator used 300 kWh per year (GE interview), so in one school year the refrigerator uses 225 kWh.

 Based on informal information gathering, it was assumed that most rooms in the colleges had a refrigerator similar to the 460 kWh/yr refrigerator. The kWh used by each type of refrigerator per year for each type of room (single, double, triple, etc.) were determined by using the energy used per year and the average number of refrigerators in each type of room, obtamed from the survey. The costs for both refrigerators were determined for each type of refrigerator by multiplying the number of kWh used per year for each type of room per each refrigerator by $0.05/kWh.

 Using raw data from the surveys, the total number of refrigerators on campus were calculated, and from there those figures the total amount of energy used by all compact refrigerators for each type of refrigerator was calculated. The total university expenditures for each kind of refrigerator were then found.

3.3 Photovoltaics
 The idea of solar technology came up from brainstorming within the energy group as a means to introduce a environmentally friendly and cost-effective energy source on campus. Houston, being a frequently sunny city, seemed to be an ideal place to implement solar technology. In photovoltaic energy potential, Texas is ranked number one (Bandy, 4/14/98). Rice can be among the first to introduce the solar movement in Houston. After having conferred with the Facilities and Engineering Department at Rice, we learned that the bills that Rice receives from the Houston Lighting and Power Company is strongly influenced by peak energy use throughout the year. If these peaks can be shrunk by the accommodation of solar energy, then Rice can obtain significant savings by reducing the cost of their bills from HL&P (Dulaney, 4/3/98).

 In terms of the location of the panels around campus, we took into account that Rice puts a lot of emphasis on physical aesthetics of the campus. Because of this, we decided that the first place to put the panels would be on the flat roofs of the buildings.

 Oberlin College will be constructing a new Environmental Studies Center, a building that will be designed completely "green. Included in the design are energy-providing solar panels. The panels will be supplied for free through a contract with the NASA Lewis Research Center, located in Cleveland, Ohio. Both institutions will mutually gain from this exchange in that Oberlin will obtain free solar panels for use and NASA will receive experimental data collected from the energy output (Orr, 1/30/98). A similar exchange between Rice and the Johnson Space Center in Houston would be an excellent opportunity to bring the panels onto campus and to build a partnership between two highly prestigious research institutions.

 One of the initial steps in our research involved familiarization of solar technology by searching the Internet for information. From the web, we found that we could obtain extensive information from the US Department of Energy in the form of brochures and websites. Most of the information discussed how solar power worked and the types of options available. We also learned about the Million Solar Roofs Initiative presented by President Clinton in June 1997. This is a national goal for the United States to have solar energy systems installed on a million roofs by the year 2010 ("Electric Utilities are Going Solar"). The initiative is meant to encourage the development of cost-effective solar energy in comparison to fossil fuel energy (Bandy, 4/14/98). This is to be achieved by encouraging all solar energy projects which will "increase the market for solar energy and assist in promoting these projects" ("Million Solar Roofs", October 1997).

 Employees from F&E provided advice on the type of setup Rice would have if solar technology were to be implemented. The electricity created by the panels is produced as DC current. In order to use the energy as electricity in buildings, the DC current must be converted into AC current by an inverter. For the purposes of this campus, the most efficient option would be to connect the solar energy systems with inverters directly into the power grid of the buildings (Dulaney 4/3/98, Munos 4/3/98).

 We consulted with several solar energy companies by email and phone. Many of these companies were unresponsive. However, we obtained extensive advice from Tess Floyd of Southern Sol-Air Power. She gave us recommendations and estimates for the equipment we would need. She also gave us general costs including those of the panels themselves, installation, and inverters (Floyd, 4/6/98).

 To determine approximately the number of panels that the Rice campus could accommodate, we estimated how much square footage of flat roof space existed on campus by means of measurements on an aerial map of the campus. We designated 75% of the total flat roof space available for panels, leaving 25% for the perimeter of the roofs and walkways between the panels. Then, by knowing the size of one panel, we estimated the number of panels that could be put on the flat roofs of Rice University.

3.4 Design strategies for New Buildings
 There has been much research into design strategies which lead to more efficient energy use in new buildings, including government programs like Energy Star. Many of these strategies are unsuitable to applications at Rice University either because the research targets residential buildings or because the climate considerations are not appropriate. However, the Florida Department of Education sponsored research into design strategies for educational facilities in hot, humid climates which led to a comprehensive document (FSEC BDAC 1994). This research provides an excellent source of information and can serve as a model for planning and construction strategies at Rice. The simulations of building designs were run with software from the Department of Energy and Lawrence Berkeley Laboratory, DOE 2.JD.

 In addition to this study, the group consulted Dr. Gordon Wittenberg, a professor in the School of Architecture at Rice, about architectural design principles. For information concerning the energy consumption of buildings on campus, the group analyzed data from Facilities and Engineering on the consumption and cost of utilities. This provided a breakdown by building and by utility for the campus.

4. Results

4.1 Present University Energy Use
 Rice has undertaken various measures to reduce its energy consumption by an impressive margin, especially with the major improvements to the campus. The three most prominent energy-saving programs included rearrangement in how the cogeneration units are used for power generations upgrading a water chiller with a more efficient, non-chloroflourocarbon (CFC) based version, and participation in the "Green Lights" program, an Institutional Conservation Program (ICP).

 By changing how the power producing cogeneration equipment was run the university was able to save $1,216,964/yr. Rice also replaced an existing chiller with one which did not contains CFCs. This produced an annual energy savings of 5,694,000 kWh/yr, which equates to $284,700/yr saved. The university is currently in the process of retrofitting building lights with more energy efficient ones under the ICP "Green Lights" program.

 Rice has attempted to conserve energy in many other ways throughout the campus such as in building insulation and the installation of motion detectors in several buildings. The newer buildings on campus have relatively good insulation properties. This includes a glass tint, Eglass, which has a higher reflectance than normal glass to keep more of the solar radiation out. The windows are single paned because double paned windows are not cost effective in Houston (Mufioz 4-12-1998). However, the older buildings have significantly less insulation. When these buildings were built, energy was very cheap, and consequently not an important factor in construction. In most cases on campus, adding additional insulation to existing walls is impractical. In some cases, insulation has been added in the attics of older buildings to improve the resistance to heat flow.

 Increasingly, motion sensors have been installed in buildings to conserve energy. Motion detectors were included extensively in the new Baker Institute. Detectors were also installed in the Physics building and in some Geology building labs. The costs of these detectors range from $75 for the office units to $125-225 for larger rooms (because of multiple circuits). The typical office uses about 250 W, and estimates on usage lead to approximately a six year payback for these units (Mufioz 4-8-1998). The potential savings in classrooms and public areas of limited use-bathrooms, copy rooms, conference rooms-are even greater.

 A seasonal trend for the university (Fig. 3) shows that chilled water use is highest during the summer months and lowest in the winter. Steam is used most in the winter, November through March, and less in the summer while electricity use remains relatively constant throughout the year. A peak in electrical consumption usually occurs in September coinciding with the start of school.

 In terms of total utility costs as a percent of the campus total the science buildings use 41.2%, academics 23.0%, colleges 16.2%, administration 8.0%, and miscellaneous 11.6%. However, when costs are taken per square foot the ranking changes to science, miscellaneous, administration, colleges, and academics from highest to lowest respectively.

Science and Research
 Those buildings which had large equipment for research or large amounts of space devoted to labs were placed in this category. The science group as a whole makes up 26% of the total campus square footage but uses 41.2% of the total campus utilities. This group makes up 39.9% of the total electric costs, 46.1% for chill water, and 30.9% for steam. When considering costs per square foot the science buildings use the most electricity, $1.54/ft^2, and chilled water, $1.77/ ft^2 and consumes the second-most amount of steam. This group uses the most overall energy per square foot spending $1.58/ft^2. Therefore it is obvious that this group has a very important impact on the energy use on campus.

 Graphing the building utility cost per square foot (see Figure 4) shows that most of the buildings in this category exceed the campus averages by a significant amount, though less so for steam use. Examination of the calculations and charts for this group showed the George R. Brown, Old Chemistry, Biology, and Space Sciences buildings being disproportionately large energy consumers. The major energy consumption in this group is due to the ongoing research that requires once through air circulation. Rice has about 300 fume hoods in its science buildings. These fume hoods are one aspect responsible for the large energy costs in the research buildings because of the large amounts of fresh air they consume. However, closing the fume hood when it is not in use leads to savings of nearly $1000 per year-which is nearly $300,000 for the whole campus per year. Therefore, education programs aimed at faculty members engaged in research and at students in labs which use fume hoods can lead to phenomenal savings each year. The larger amount and size of equipment used for this research is another significant energy sink. From here, we can investigate specific energy usage aspects of these buildings in greater detail in the future.

 Seasonal trends were also graphed for each building by month for fiscal year 1997. The seasonal trends generally followed the overall campus trend fairly closely for each utility.

Academic
 The academic building category describes a group of buildings on campus that are used for academic and teaching purposes and contain minimal facility for scientific laboratory use.

 The academic buildings use 23% of campus energy but comprise 28.4% of the total square footage on campus. In terms of utility usage by electricity, steam, and chilled water, the academic buildings use between 21% and 27% of the campus totals.

 Time trend plots illustrate that the energy consumption is consistent with the campus average. The Baker Institute and Duncan Hall stray from the average, with generally high usage for the Baker Institute and a decreasing trend in usage for Duncan Hall. This may be explained by the fact that both are new buildings and have not seffled into a consistent usage pattern yet. Alice Pratt Brown also differs for steam use with a much flatter trend than the others.

 When looking at the graphs describing utility cost per square footage, one may notice that Alice Pratt Brown (music school) is consistently distinguished from the rest of the buildings. It represents the highest usage for electricity, chilled water and steam. This large difference is attributed to the high humidity control necessary for this building.

Colleges
 Rice's residential colleges are metered in groups, instead of individually, like the other buildings on campus. When data is available which breaks the colleges up into sections, the colleges are usually partitioned into the north and south sides of campus. Chilled water is also the only utility which is metered by building in this group.

 The colleges comprise 23.0% of the square footage and use 16.2% of the campus energy. Hanszen has the lowest cooling cost per student at $137.64 compared to Sid Rich at $263.91 per student. Steam costs per student: North is $72.94, South is $93.77. Electricity costs per student: North is $459.23, South is $436.21.Total utility costs per student are $459.23 in North and $436.21 in South.

 Among the residential colleges, there are currently four different systems in place. Two colleges, Brown and Jones, use a four-pipe Fan Cool Unit (FCU). Four colleges - Baker, Will Rice, Wiess, and Hanszen - use a two-pipe FCU. The other two colleges use extremely inefficient Air Handling Units (AHU). Lovett College uses a coil-reheat system, and Sid Rich uses a mixing box to maintain the proper temperature. Of the four systems, the two-pipe FCU is the cheapest to operate, but has a range of temperatures where it is ineffective at achieving comfortable temperatures (65-70 oF). The two-pipe system allows only chilled water or steam to be flowing through the pipes at any time. The four-pipe system achieves a higher degree of comfort at a slightly higher cost because it can use steam and chilled water simultaneously. The other two systems are highly inefficient and waste large quantities of chilled water. The reheat coil requires cooling the outside air and then reheating it to the desired temperature. The mixing box requires the mixing of steam and chilled water all year round to achieve the proper temperature. The chilled water costs for each system vary greatly, from Hanszen, which has the lowest cost per student to Sid Rich which is nearly twice as high and represents the highest cost per student

 For the other utilities (steam and electricity), the colleges are metered by North (Brown and Jones) and South (the other six colleges). The difference in these utility costs can partially be attributed to the air handling systems described above. Steam use per student is 22% lower in the North colleges because of the excessive waste of steam in the two AHU systems in the South colleges. Electricity use is 11% higher in the North colleges, due to the use of electric dryers in these colleges, while most of the colleges in the South colleges are natural gas. Overall, the total energy utilities costs show only a 5% difference in per student costs.

 Seasonal trends shows utility use generally following the campus trends. Electricity however has its peak in February. The electrical consumption also is generally higher than that of the university as a whole. From December until July the steam use of both groups remains higher than Rice's average consumption.

Administration
 Administrative buildings are characterized by a large number of offices devoted to administrative duties for the university.

 As noted before, the administrative buildings occupy 8.4% of the total square footage on campus. In terms of utilities use, the contribution is even smaller. Electricity represents the largest utility expense for these buildings at 8.0% of the total. Steam and chilled water are a smaller portion of the total campus use (about 7.6%).

 The cost of electricity per square foot is similar to the campus average for electricity, $1.27. Considering the other utilities, the Rice Memorial Center exceeds the other buildings and the campus average, probably due to the larger public areas in this building.

 Analyzing the seasonal use of utilities also helps to characterize the administrative buildings. The electricity use tends to reflect the average campus use closely. The chilled water use reflects the pattern of the overall campus use, except the trends are more extreme (i.e. higher highs and lower lows). The steam use, however, shows a dramatic deviation from the campus average for Lovett Hall. December and January show a jump in steam use for this building much more than the increase in the campus use.

Miscellaneous
 The miscellaneous building category is comprised of eleven buildings which did not fit one of the other categories well. Many of these buildings are smaller or used only for special events or specific purposes.

 In all, these buildings comprise 14.2% of the square footage on campus. However, the total energy use makes up only 3.9% of the electricity use and 7% of the steam use on campus. The miscellaneous buildings consume more steam per square foot, $2.31 /sf, than any of the other buildings, and the second-most electricity per square foot.

 Because of the wide range of buildings included in this group, it is difficult to determine trends within the category.

4.2 Energy Consumption Within the Residential Colleges
 The data showed that for every individual in a room, the number of shared appliances such as TVs, stereos, VCRs, phones and microwaves decreased linearly as the number of people in the room increased from one to four. The number of personal appliances, such as clocks, lights, printers and computers remained fairly constant, though the number of computers and printers per person increased in double rooms only to then remain fairly constant as the number of people in rooms increased afterwards.

 Triple rooms appear to have fewer appliances not only per person, but also per room. As one would expect, computers, lights, stereos and clocks increase almost linearly with an increasing number of people. However, the number of TVs, VCRs and refrigerators remained similar in doubles and triples but jumped in quads. When looking at the numbers of each appliance per person, a small increase can also be seen among these appliances in quads. It is interesting to note that a triple might draw less energy from appliances than a quad on a per person basis.

 When the total energy costs per room are examined on a per person basis, a single draws the least energy from appliances alone. The cost then increased by approximately 50 percent for a double. Costs in a triple are nearly double the costs of a single. The per person costs of a quad is only $11 per person more than the per person cost of a triple, while per person costs in doubles and triples differ by $30. The cost of energy per person in a six-person room then drops near the level of a three-person room, although there are only two rooms of six on campus.

 From the data pertaining to student computer sleep modes, computers without sleep modes cost the university about $13,000 more per year than computers with sleep modes. This cost is the result of students leaving their computers without sleep modes on for extended periods of time when they are not in use. This figure does not consider the increased amount of cooling needed to compensate for the heat given off by computers left on for extended periods of time. Unfortunately, software is not available to give a computer the sleep mode; it must be installed upon manufacturing. The cost inflicted upon the university could be eliminated if students simply turned off their computers when not using them.

 The university spends approximately $20 to operate every compact refrigerator on campus; with newer refrigerators, this number could be cut in half. For a quad, $40 is spent each year supplying the refrigerators with power to run; with newer refrigerators, this figure could be reduced to $20 per year.

4.3 Photovoltaics
 Floyd suggested that we use 120W panels, considering the size of our energy usage. One panel measures 3 feet wide and 4 feet tall. From our rough measurements from the campus aerial map, we determined that there is approximately 610,781 ft^2 of flat roof space on campus. 75% of this, which we took as usable space for the panels, is 458,086 ft2. With the given panel size, we estimated that 38,174 panels can be put on the Rice campus. For 120 W panels, a maximum of 4,581 kWh can be produced. With an average cost of energy of $70/year/kWh (Dulaney, 4/3/98), this output can save Rice approximately $320,662 a year just from peak shavings. To approximate for a maximum output during 6 hours a day, the panels would produce 27,486kWh a day. At $.05 per kWh, Rice could save approximately $501,620 a year.

 In terms of initial costs, however, Floyd told us that the panels would cost approximately $6 per watt. With 120W panels, 38,174 panels would cost about $27,485,280. Installation costs are usually $10-15 per watt. Assuming $10 per watt, the installation cost would be about $45,808,800. The cost of the inverters are generally $250 per panel. However, Floyd said that Rice would probably get the inverters for half-price because of the scale of the installation (Floyd, 3/12/98). The inverters would then cost $4,771,750. Overall, the total initial costs to retrofit the entire campus add up to $78,065,830. These costs do not support a realistic payback time period.

 The results from investigating options through the NASA in Houston found that solar panels are not directly accessible from the Johnson Space Center. However, the NASA Lewis Research Center has a Photovoltaic Branch in their Power Technology Division. Their goals include the research and development of solar technology and the monitoring of "a number of developmental efforts performed by private contractors and universities" (Clark, 4/13/98). Further contracting through them may provide a potential means for an exchange between NASA and Rice University.

4 Design strategies for New Buildings
 Energy expenses represent a significant portion of the total operating costs for Rice University. In 1997, energy expenses were over $3.4 million. The hot and humid climate in Houston means that one of the largest energy consumers on campus is due to air conditioning. In addition, electricity use represents an enormous expenditure. In fact, expenditures for electricity and air conditioning represent 94% of all energy expenditures. Of the electricity consumption, lighting systems in educational facilities make up about 32% of annual energy consumption, or about $700,000 at Rice. Lighting systems are also responsible for 15% of the cooling demand for HVAC systems-$156,OOO for 1997 at Rice.

 Targeting HVAC and lighting achieves the highest energy savings (FSEC BDAC 1994). Cost-effective design and planning strategies should be aimed at reducing the consumption and load of the HVAC and lighting systems. This strategy led to buildings 43% more efficient than current educational buildings in Florida, where the climate is similar to that of Houston.

 The HVAC system is responsible for providmg ventilation, supplying cool conditioned air, controlling moisture. The demands (loads) on the HVAC system come from both internal and external sources. The major internal peak loads come from occupants (29%), lights (15%), and equipment (6%). External peak loads are primarily due to infiltration/ventilation (36%) and glass radiation/conduction (9%) (FSEC BDAC 1994). The air conditioning consumption can be reduced by reducing these loads and by using more efficient HVAC equipment.

 Lighting systems are a vital part of energy conservation not only because of the large electricity consumption, but also because of the demand lights place on the HVAC system. All of the electricity consumed by the lighting system is either converted directly into heat, or first to light and then to heat. This heat contributes 15% of the peak load on the HVAC system (FSEC BDAC 1994) Efficient lighting strategies aim to minimize the operating time of lights and to maximize the efficiency of lighting systems.

 In terms of the overall design, the most effective means of increasing energy efficiency are achieved through orientation, windows and glazing, insulation, and configuration. The most effective means of achieving efficiency within the building are through lighting systems, proper HVAC sizing, and electric fans.

Orientation: The orientation of a building has important impacts on both the HVAC load and the lighting load. To maximize daylight potential - and minimize the load on the lighting system - windows should face north and south. Solar heat gains can be reduced by minimizing the wall area and windows facing east and west. These considerations imply a building which is elongated in the east-west direction. Buildings with non-optimal orientations typically suffer a 3% increase in total energy consumption (FSEC BDAC 1994). The results of this are apparent on campus, also. In comparing two academic buildmgs, one (Lovett Hall) with the axis aligned north-south, and the other (Allen Center) with an axis aligned east-west, we see that Lovett Hall is 5% more costly per square foot for cooling. Since Lovett Hall walls dominantly face east and west, it is subjected to much more infrared radiation, leading to a greater heat gain than the Allen Center.

Windows and glazing: Windows combined with lighting systems selection can have cost-effective impacts on reducing energy consumption. The primary energy-related role of windows is to provide daylight-this energy conservation potential can exceed the consequent thermal penalty (FSEC BDAC 1994). Windows should be selected to reject heat from the sun to reduce localized overheating, while providing daylight to reduce the need for electrical lighting. In Houston, double-pane windows do not provide a cost-effective window. Instead, windows should be chosen with a spectrally selective glazing or windows with a low shading coefficient and high visible transmittance. In order to encourage people to use windows for ventilation, windows must be easy to operate.

Shading: Shading is used to block the direct solar radiation in order to reduce heat gain, glare, and localized overheating. As with windows, the primary role of shading is to provide quality daylighting to reduce the electrical lighting (FSEC BDAC 1994). Exterior shading provides the most effective way to reduce heat gain, and can reduce the need for interior blinds. In addition, shadmg can also increase daylighting, by using light shelves. Light shelves are placed part way down the window to shade the majority of the window and to reflect light into the upper portion of the window. This strategy has been employed in the plans for the new Environmental Studies Center building at Oberlin College (William McDonough). Horizontal window shades on the south facade can also be used effectively, though they do not have the lighting benefits of light shelves. On the east and west sides of buildmgs, shading is often difficult to accomplish. For these sides, it is beneficial to take advantage of the large number of trees on campus, and locate trees near the building to provide shading.

Insulation: Though insulation is often a cost-effective solution in residential buildings, it is less effective in institutional and commercial buildings because they are dominated by internally generated loads (FSEC BDAC 1994). According to the study done in Florida, the only insulation increases that showed overall savings were in roof insulation. This is particularly applicable to Rice, since many of the buildings are constructed with red roofs, which absorb more radiant energy than lighter colored roofs. In walls, building codes were sufficient. Additional insulation can trap heat inside the buildings when outdoor temperatures drop and heat loss would be beneficial.

Building configuration: Configuring the spaces within the buildmg to take advantage of daylighting can lead to an efficient HYAC strategy. This design strategy involves clustering spaces with similar annual occupancies and similar daily occupancies leads to efficient HYAC operation. Also, planning buffer zones on the east and west perimeters to minimize heat gain.

Lighting Systems: With the above considerations in mind, an appropriate and efficient lighting system can be chosen. Daylighting provides a very cost-effective way to minimize the HVAC and electrical lighting loads. When planned from the beginning, daylighting is a very low cost strategy to reduce energy use. However, no savings can be achieved if interior lights can not be dimmed or turned off when natural light meets the desired level of lighting.

 In order to reduce energy consumption from lighting, Rice should use more efficient systems, install automatic or manual controls that limit the time lights are on, and provide mechanisms which dim electrical lights when sufficient natural lighting is present.

 Each component of the lighting system consumes energy, not just the lamp. Therefore, a system should be selected which will maximize efficacy - the ratio of lumens output to energy input. Also, energy can be saved by installing automatic controls in community spaces which no one feels responsible for, such as bathrooms, copy rooms, conference rooms, etc. Pay-backs are quickest in areas of occasional use and multiple fixtures, such as conference rooms. In addition, continuously dimming electronic ballasts with photosensors in daylit zones can achieve the greatest savings from daylighting (FSEC BDAC 1994).

Fans: Ceiling fans offer an opportunity for substantial savings at Rice. Fans allow for increased air circulation and improved comfort, and allow thermostat temperatures to be set significantly higher. Air motion raises comfort temperatures 2-6 degrees F higher than without fans (FSEC BDAC 1994). In administrative buildings, ceiling fans lead to moderate savings, but in classrooms, fans lead to very substantial savings (FSEC BDAC 1994). The air circulation would significantly decrease the demand on the HVAC system, and help prevent the sudden demands for air conditioning that classrooms can cause.

Conclusions

 Educational programs should be implemented to raise awareness and concern among students and faculty on campus. Education programs aimed at both faculty and staff to raise awareness about energy issues on campus can be an effective method to overcome inefficiencies and wasteful behaviors. Both the Facilities and Engineering and Food and Housing staff have stated that the most effective way to become more energy efficient is through education (Mufioz 4/3/98). Education represents perhaps the most cost-effective manner by which the university can increase energy efficiency in this time of growth. Education programs aimed at fume hoods alone can result in phenomenal savings. In addition, findings from this study can be distributed to faculty and students in the form of brochures. Presentations and a public website will also allow greater access to this information.

 Educational programs should also be targeted at residential colleges, which account for over 16% of energy costs on campus. The greatest and quickest impact in conservation in the colleges will come through the education of the students about these issues and how they can help. The survey conducted helps to detail the extent of wasteful energy practices by students living on campus.

 In its study of the university's present energy consumption, this group found some areas of high energy consumption. The majority of this high energy use is related to either once through air circulation or humidity control. Therefore continued investigation into improvements in the present systems and/or making use of newer, more efficient technologies and designs could benefit in a great deal on energy savings.

 From the data collected, triples seem to use appliances most efficiently, both per room and on an individual basis. However, further investigation is necessary to determine heating and cooling costs per square foot for the various room types, both as a whole room and on an individual basis. These costs should then be considered with the appliance figures to determine the most energy-efficient type of room.

 As for the savings that could be achieved if students without sleep modes on their computers turned off their computers when not in use, students are probably not aware that their actions produce such a cost. A student-awareness campaign would probably be the easiest way to correct this problem. The campaign would ideally target both students shopping for a new computer, so that they are informed that a sleep mode is most beneficial, and students without sleep modes, so that they are aware that just turning off their computer when not using it saves significant amounts of energy and money.

 The university spends tens of thousands of dollars each year supporting compact refrigerators in the colleges, most of which are older, passed on from one generation of students to another. There are several options the university has to cut these excess costs. The school could form a contract with an appliance company so that the company either sells or leases new refrigerators to the students at appealing rates. The university could also purchase refrigerators itself and sell or lease the newer models to the students, as do other institutions such as Brown University (Brown is Green).

 Unfortunately, our calculations are not optimistic for the future of solar technology on the Rice campus. However, the benefits are still impressive. Future investigation into potential ways to alleviate the costs of implementation may be worth looking into, such as the Million Solar Roofs Initiative and communication with the NASA Lewis Research Center. The Department of Energy has a program for matching grants for solar projects, as do many photovoltaic and research facilities (Bandy, 4/14/98). Federal and/or state tax breaks may also help to alleviate costs. Rice may be able to take advantage of some financial aid due to their non-profit status. Despite the cost issues, the potential for solar technology should continue to be monitored seeing as the cost of implementation has been decreasing with time ("Electric Utilities are going Solar"). Also, the costs for solar power are front-loaded as implementation costs. After installation, only nominal costs will be required for maintenance.

 The use of solar energy does not have to be limited to the buildings on campus. Other energy-consuming items on campus that could benefit from solar power are the electric carts used by F&E, F&H, the campus police, and the escort service. The energy group envisioned a charge-up station for the carts whose electricity would be saved as battery power straight from the panels.

 The medical center in Houston is planning on retrofitting a parking garage that will be fully powered by photovoltaic technology. Their plan is to prepare and educate themselves about photovoltaics so that when the technology becomes more accessible and popular, they will already understand how it works for them and what their best options are (Bandy, 4/13/98). Rice may adopt a similar attitude and action plan by trying out photovoltaics on a small area or one building first. Installation onto the new buildmgs on campus may be cheaper than retrofitting onto the old buildings. The first step in introducing photovolatics on campus may be with the new colleges.

 The energy group's investigation in implementing solar technology at Rice University is only the beginning. There are still many options to investigate further. Hopefully, future interested students and faculty at Rice will continue to search for a feasible way to bring clean and efficient energy on campus and can use this research as a starting point for that ultimate goal.

 Minimizing the load and demand on the HVAC and lighting systems can be a very costeffective way to increase energy efficiency and reduce energy expenditures. In order to be effective, the university must commit to energy efficiency and plan a strategy early in the design stages. It requires commitment of time and resources to achieve the benefit of savings which will compound over the entire life of the building. A high level of communication between the architect, engineers, and end users is necessary to achieve maximum efficiency and savings.

 Rice can make significant changes to their current state of energy use and costs. The research that the Energy Group has performed provides a solid base for further investigation into the types of things that Rice can do to improve their energy consumption and financial status.