Student Abstracts: Engineering at ORNL

A Thermal Mass Benefit Calculator. AARON KEEGAN (Warren Wilson College, Swannanoa, NC 28778) JAN KOSNY (Oak Ridge National Laboratory, Oak Ridge, TN 37831).

Studies dating from the energy crisis of the 1970's to the present day have shown that buildings with exterior walls constructed from massive materials can achieve lower heating and cooling energy demands than buildings with comparable walls of light-weight construction. For a given home, determining the quantity of savings that could be achieved from massive walls requires field experimentation and/or whole building energy simulation. These methods are often too time consuming to be included in the construction planning process. In an effort to reduce the time necessary to determine the potential energy savings of a massive wall system, a simplified calculation tool was created. Data was obtained from a massive wall system study performed by J.Kosny et.al. in 2001. Regression analyses were performed on multiple data fields including heating load, cooling load, electricity consumption, natural gas consumption, and total energy consumption. The regression equations were then programmed into a graphical user interface program written in Visual Basic. The result is a prototype application that allows users to compare the energy performance of four different massive exterior wall configurations to a wood frame exterior wall. Comparisons can be made for three model homes in 10 U.S. locations. The program will benefit from further development as more locations, home models, and climate control system options are added. The program will eventually be deployed on Oak Ridge National Laboratory's Building Envelopes Program website for public utilization.

Analysis and Calibration of Side Emissions for Development of new Control system. SCOTT STELLERN (Clemson University, Clemson, SC 29634) D. L. BESHEARS (Oak Ridge National Laboratory, Oak Ridge, TN 37831).

The Hybrid Lighting project combines the use of solar collection and remote source lighting. A hybrid light fixture combines fluorescent light with natural sunlight. This conserves energy and the natural light helps provide a friendlier atmosphere. To maintain a constant light level, a photometric Li-cor sensor is placed in the center of the room to monitor light levels and make adjustments. Presently, to measure the amount of light a fiber is emitting an integrating sphere must be used. A new system involving photometric light sensors attached to the fibers themselves would eliminate the need for an integrating sphere, which can be hard and time-consuming to use. The integrating sphere was used to measure light levels of a specific fiber, which was then compared to the measurements of a photometric Li-cor sensor attached to the side of the fiber. After comparing these two measurements, a formula was made to relate the measurements. It was discovered that the relationship was linear and that the side emission is a simple percentage of the total light emission. Now to monitor the amount of light a fiber is emitting, a simple side sensor can be attached to the fiber and a few calculations can be done. This will help monitor each fiber for defects and deterioration, which will help conserve energy, the ultimate goal of this project.

Application of Phosphor Thermometry to Fuel Cell Diagnostic. DUSTIN GARVEY (University of Tennessee, Knoxville, TN 37916) STEVE ALLISON (Oak Ridge National Laboratory, Oak Ridge, TN 37831).

Phosphors are most generally described as rare earth doped materials that emit fluorescence when properly excited. This fluorescence may have several temperature dependant characteristics. One of the more temperature sensitive characteristics is the decay time. Measuring and comparing decay times to known calibrations allows very precise temperature measurements to be made. This method is to be applied to fuel cells. The main requirements for this detector are a low cost and a measurement range from '40 - 150oC. In order to satisfy the cost aspect, light emitting diodes (LED) have been chosen to excite the phosphor. By bundling varying numbers of 400-micrometer diameter optical fibers and directing them toward an LED, it was found that fluorescence was obtainable from a YAG:Cr sample using each of the fibers contained in the largest bundle (35-fibers) for excitation. This result establishes that it is possible to use one LED for many individual thermal detectors (35 demonstrated), which lowers the cost even further. To lower detection cost, a photodiode (PD) should be used in the future to replace the more expensive photomultiplier tube (PMT). Because a PD has a smaller gain than a PMT, a phosphor with a very intense fluorescence must be used. Small diameter ruby spheres have proven to be the best candidate. They have emitted a 1.4 V fluorescence signal, which is significantly larger than the 0.35 V signal from the YAG:Cr. Using the various LEDs that were available in the lab (370, 395, 405, 430, and 450 nm), a limited excitation spectrum has been made of the ruby in order to choose the LED that maximizes ruby excitation. It was found that the ruby is most excited by a 405 nm LED. Next, varying resistances were used to measure the fluorescence. The larger the measurement resistance, the stronger the fluorescence and the more distorted the decay time. Through comparative analysis it was found that the measurement resistance could be on the order of 26 kÙ with no significant distortion in the decay time (±2%). The collected data represents the bounds that limit the design of the thermal detector and will aid in future design decisions.

Building Integrated Photovoltaics On Low-Slope Commercial Buildings. EDMUND BROWN (San Juan College, Farmington, NM 87401) WILLIAM A MILLER (Oak Ridge National Laboratory, Oak Ridge, TN 37831).

Documentation of the benefit of Building Integrated Photovoltaic's (BIPV) effect on mitigating building cooling load and roofing system performance is sparse. Roofing contractors, architects, photovoltaic (PV) manufacturers, builders and utilities are asking for this data to increase their understanding of building system performance and to quantify building load energy savings. Creating a rating procedure to compare the energy efficiency of a roof that integrates PV equipment with the efficiency of a conventional insulated roof, is the objective of a project under way at The Building Technology Center (BTC) at Oak Ridge National Laboratory (ORNL). Two ongoing programs at the BTC provide a unique opportunity to conduct the research required to quantify the impact of BIPV on building cooling and heating loads. One program involves a partnership among the Federal Energy Management Program (FEMP), the Tennessee Valley Authority (TVA), British Petroleum (BP), the ORNL State Partnership Program, and the State Energy Office (SEO). An array of 72 polycrystalline panels are installed on the roof and south wall of the Envelope Systems Research Apparatus (ESRA). The second program involves the testing of a laminate amorphous silicon material on the ESRA - a roofing test facility with just about every material used in low-slope roofing installed on the roof. Extensive historical data is available from both these programs including climatic, roof and BIPV temperature and heat flow, and solar system performance data. This project is focused on the consolidation and validation of the data, from the above mentioned sources, into a single database and the creation and validation of the roof performance rating system. To create and test the rating system, this data is input to a BTC created roof system model that predicts heat flow and temperature within the roof. This model is called STAR - Simplified Transient Analysis of Roofs. Actual measured data is compared to the predictions of STAR to validate the rating procedure. The rating system has been tested and it performs well. The rating system can now be used to predict the performance of roofing systems and the impact of BIPV on building heating and cooling loads in various locations around the country. Student Name: Ed Brown School Student Attends: San Juan College Name of Mentor: William A. Miller, Ph. D. Division: Energy Science and Technology Program: Community College Institute

Characterization and Properties of Aluminum Lithium Alloy 2195 Friction Stir Weld. JACOB DAWSON (South Dakota School of Mines and Technology, Rapid City, SD 57701) STAN DAVID (Oak Ridge National Laboratory, Oak Ridge, TN 37831).

Aluminum-Lithium alloy 2195 is being used in constructing the external tank of the space shuttle for several reasons including a high strength to weight ratio. However, like many aluminum alloys the material is difficult to fusion weld. Friction stir welding is being explored as a possible alternative to fusion welding. Flow properties (mechanical properties) of the friction stir weld zones including the weld nugget, thermomechanically affected zone (TMAZ), and heat affected zone (HAZ), as well as the base material were studied in this experiment. The flow properties of each relatively small zone were determined through a new nondestructive and localized testing method called Automated Ball Indentation (ABI) testing. The HAZ exhibited the lowest yield strength and UTS. Flow properties of the weld nugget and TMAZ were very similar to one another. However, the strengths throughout the weld region were lower than the base material. Decrease in strength of the friction stir weld is mainly due to metallurgical changes that arise from the thermal cycle experienced by the material during the weld. As the strengths decrease through friction stir welding, the uniform ductility of the material is increased. Uniform ductility was found to be ~6% in the base material and ~12% in the weld area. Future work will include fracture toughness experiments as well as attempting to strengthen the weld through precipitation hardening. Fracture toughness measurements are of special interest at cryogenic temperatures required to store liquid hydrogen and liquid oxygen, which serve as propellants for the space shuttle.

Computational Modeling of Bifurcated Abdominal Aortic Aneurysms. JENNIFER LILLY (University of Tennessee, Knoxville, Knoxville, TN 37916) KARA KRUSE (Oak Ridge National Laboratory, Oak Ridge, TN 37831).

New developments in research have strived to model abdominal aortic aneurysms (AAA) by integrating computed tomography (CT) scans, segmenting software and finite element (FE) analysis. AAA has troubled the medical field in the past due to the lack in development of an accurate means of diagnosis. The high mortality rates in cases where the patient experiences a rupture have been the primary motivation in developing new and improved methods to determine where the point of rupture is likely to occur. Developing a proper means of estimating the distribution of stress in an AAA would be advantageous in predicting whether an aneurysm has a high probability to rupture. Fundamental to this process is creating a model that is appropriate for both the wall and the intraluminal thrombus found in AAA. Past case studies on AAA have neglected to include cases that account for the bifurcation in the aorta, the point at which the aorta separates into the two femoral arteries. Previous software has been limited in producing a viable model of the bifurcated aneurysm that would be capable of producing reasonable results in finite element analysis. New improvements in currently available commercial software have led to the more in depth investigational studies and applications. Various programs have been undergoing evaluations in order to determine which would be most efficient in segmenting the AAA and creating a representational 3D model of the bifurcation. At this time there is no conclusive evidence as to which program performs the tasks at hand superior to another. However, one particular program, Amira 3.0, has been utilized extensively and shows potential promise. This program successfully segmented the AAA from the patient CT data and created a 3D model. Currently, a model has yet to be created that is sufficiently smooth enough to be imported into the finite element program in order to create a mesh. It is expected that a model will be produced that is capable of creating a representable mesh of the surface of the model, at which stress analysis may be conducted in order to determine areas of significant pressure occurrences. Once these high-pressure areas are determined, the quantitative data will be compared with the known point of rupture from the vascular surgeon who handled the particular case in study. This comparison will aide in the determination of the accuracy and precision of the model and stress analysis.

Development of Modeling Techniques for Abdominal Aortic Aneurysms that Include the Bifurcation of the Aorta. STEPHANIE BARNES (University of Tennessee, Knoxville, Knoxville, TN 37916) KARA KRUSE (Oak Ridge National Laboratory, Oak Ridge, TN 37831).

Abdominal Aortic Aneurysms (AAA) are a serious problem for the medical community; often the end result is the death of the patient, as there is no truly successful way to treat AAAs after rupture. The best way to treat an AAA is to prevent rupture. Developing a means by which the rupture point of an AAA could be determined could advance the procedural treatment of AAAs. Research has been done utilizing computerized tomography (CT) scans as the basis for creating a 3D model of the AAA, which can then be tested to determine the points of elevated stress and to predict potential rupture sites. CT scans are often taken of patients with AAAs to determine the extent of their distress; they provide visualization of the outer wall of the aorta, the inner lumen, and the thrombus, as one progresses down the aorta, which makes them beneficial for modeling purposes. The previous research proved very successful; the point of rupture that the model predicted coincided with the actual point of rupture. Current research focuses on the CT scans of a patient with an AAA that ruptured approximately 3 days after the scans were taken. The difficulty with this particular AAA is that it encompasses the bifurcation, where the aorta divides into the two iliac arteries, whereas the AAAs from the previous research did not incorporate the bifurcation. Previous methods for modeling the AAA did not allow for inclusion of the bifurcation; thus a method had to be devised to successfully model the bifurcation. Testing is in process to determine which of various programs would be most beneficial and accurate in segmenting the AAA and visualizing the bifurcation in a 3D model. Though it has not yet been determined which program will prove the most successful, one of these programs, Amira, has been utilized for further testing. The CT data was segmented and a 3D model was created employing Amira. This model was then imported into an NURBS geometry program, which was used to modify and improve the geometry of the bifurcation. The model was then imported into a geometry-meshing program, which created a mesh on the surface of the model; the surface proved meshable, which is an important requirement of the model. The next step includes the model being put through a finite element analysis to determine the points of high pressure in the AAA. These determined points will then be compared with the known point of rupture to assess the accuracy of the model and the modeling techniques.

Energy and Demand Savings from a Geothermal Heat Pump Retrofit in Military Residences in the Southeastern United States. ANTHONY FLORITA (University of Wyoming, Laramie, WY 82070) JOHN SHONDER (Oak Ridge National Laboratory, Oak Ridge, TN 37831).

At a military base in the Southeastern United States, air-source heat pumps were replaced with geothermal heat pumps (GHPs) in more than 2,000 family housing units. Because GHPs exchange heat with the earth instead of with ambient air, they have the potential to make significant reductions in energy use compared with air-source equipment. Ground temperatures do not vary as extensively as the temperature of the ambient air over the course of the year, which allows GHPs to maintain stable heating and cooling capacity and to operate more efficiently. The objective of this paper is to assess the design of the project and to determine the energy and demand savings that have resulted from the retrofits. Energy usage was calculated by correlating measured energy data with total degree days. The correlation between monthly extreme temperatures and peak energy demand was used to find demand savings. Linear regression techniques were utilized in both cases to identify comparable equations and find pre- and post-retrofit savings. The savings calculations are complicated by the fact that the available utility bills measure electrical use in the entire family housing stock, which includes about 4,000 homes. It was found that the GHPs reduced energy use by 30%, which compares well with other projects. Demand saving was on the order of 10%. One potential problem with the design is that the ground heat-exchangers may have been undersized in some residences. Anthony Florita University of Wyoming John Shonder Engineering Science and Technology Science Undergraduate Laboratory Internship (SULI)

Evaluation of Candidate High Temperature Phosphor Binders. NOAH BERGERON (Univeristy of Louisiana at Lafayette, Lafayette, LA 70504) STEPHEN W. ALLISON (Oak Ridge National Laboratory, Oak Ridge, TN 37831).

Fluorescence from phosphor materials offers a method of non-contact thermometry in hostile environments such as those found in high temperature regimes (1200-1700 oC). Phosphors are typically rare earth doped ceramics that emit light when excited. The intensity, rise time, decay time, and wavelength shift of this emitted light can be temperature dependent. If thermographic phosphors are applied to a surface, an excitation source and a method to characterize the emission are provided; it is possible to determine a temperature of the surface. The coating method involved in this study is the use of phosphor paints. These temperature sensitive paints are created by mixing phosphor with a binder material to form a sprayable coating that can be easily and economically applied to a large area. Given that decay time calibration data exists to 1700 oC, it is desirable to have phosphor paints that will survive to at least that temperature. The survivability of phosphor paint depends on characteristics of the binder. The goal is to discover binders that will allow phosphor paints to survive to high temperatures. Binders that survive to high temperatures will allow for the construction of non-contact measurement devices useful in environments that are not suited for more common thermocouple or infrared devices. For a phosphor paint to be useful at a certain temperature the paint must fluoresce when excited and decay time must be measurable. In this study, Cotronics Resbond 791, 792, 793, 794, and 795 ceramic binders were evaluated to determine their suitability for serving as phosphor binders with Y2O3: Eu. Post-thermal cycling spectral analysis was used to quantify wavelength and intensity changes in emission from UV excitation. Several of the paints utilizing their binders were able to survive temperatures of 1500 oC. It is hoped that the results of this research will lead to studies involving the use of phosphor/binder paints in rocket-based research.

Evaluation of in vitro Canine Abdominal Aortic Aneurysm via Finite Element Analysis. JOEL OUTTEN (Univesity of Tennessee, Knoxville, TN 37916) KARA KRUSE (Oak Ridge National Laboratory, Oak Ridge, TN 37831).

There is a great scientific effort aimed at quantifying endotension in abdominal aortic aneurysm (AAA). Finite element analysis (FEA) has been utilized to computationally model endotension, but there has been no comparison with experimentally determined data to validate stress measurements within the aneurysm sac (AS). A previous experiment entailed taking pressure measurements, via diaphragm pressure transducers, within the AS of three in vitro canine AAA specimens. Pressure measurements were taken at eight distinct locations within each specimen. The goal of this study was to generate FE models of the three canine abdominal aortas and compare the computational stress measurements at the transducer sites with the experimental data. Computed tomography (CT) images were utilized to construct the models, which were subsequently meshed into finite elements. The models consisted of three different sections: AAA wall, intraluminal thrombus, and the stent-graft. The ends of the models were fixed in the axial direction in an effort to simulate the experimental set-up, and a constant, uniform, intraluminal pressure was applied. Each canine model was run at intraluminal pressures of 85, 100, and 120 mmHg. Element sets were defined at the locations of the experimentally utilized pressure transducers, and an average stress was calculated for each set in the direction normal to the transducer diaphragm. The computational stress was only weakly correlated to the experimental pressure for one model, while the other two were uncorrelated. While FEA failed to correctly model the pressure measurements, this work presents a new method of FE validation for AAA modeling.

Future Combat System Combat Identification Speed of Service Model. LAUREN HATCHELL (Louisiana State University, Baton Rouge, LA 70803) GLENN ALLGOOD (Oak Ridge National Laboratory, Oak Ridge, TN 37831).

DARPA's Future Combat System (FCS) Integrated Support Team has developed a Combat Identification (CID) notional architecture for Increment I FCS. Combat identification is the process of identifying entities in the battle space for purpose of fratricide reduction and increased combat effectiveness. CID solutions must include ground-to-ground, ground-to-air, and air-to-ground elements. In support of the air-to-ground element - Fixed Wing to Ground CID - a 'speed of service? model has been developed using queuing and renewal theory to provide an analysis of latency and quality of service in the FCS communication network. The application of queuing and renewal theory eliminates the conditional dependencies of one node to another, reducing model overhead. Latency is described in terms of Time-to-ID based on the following model ? propagation time, transponder processing, interrogator message identification and reformatting, data links to command and control (C2), C2 message identification and reformatting, human decision, and data links to shooter. The model also allows for a re-task decision at the C2 level. Algorithms have been developed that provide closed- and open-form probability density function solutions. The closed-form solution is derived by convolving input functions, while the open-form solution is derived from a simulation that yields an array of data points. In the future, the model will allow for the input of specific platform variables (aircraft, radar, sensor). Although originally developed for Fixed Wing to Ground, the speed of service model can be extended to ground-to-ground and ground-to-air situations. The model is flexible in that it will allow for many scenarios, network configurations, and sensor-shooter links. Subject matter experts have been consulted to determine reasonable input functions for a non line-of-sight (NLOS) situation. Results of the Fixed Wing to Ground NLOS simulation fall within a thirty second Time-to-ID requirement, but in a re-task scenario Time-to-ID exceeds the requirement by as much as twelve seconds. However, uncertainty in the network configurations may lead to longer latency times. Future plans include acquiring more information on the network links to confirm an NLOS Time-to-ID latency. Intelligent agent software will be developed that will reduce latencies and make the information more useful by abstracting large amounts of information and analyzing the quality of service of the networks.

Gas-Phase Mercury Separation Involving Room-Temperature Ionic Liquids. CHRISTOPHER HARRIS (Texas A&M University, College Station, TX 77843) DAVID DEPAOLI (Oak Ridge National Laboratory, Oak Ridge, TN 37831).

The discovery of toxic mercury in many areas throughout the United States has recently increased scrutiny of coal-fired power plants, the largest source of anthropogenic mercury emissions. Coupled with the pending EPA regulations on mercury emissions, a pressing need exists for efficient mercury removal technology. Previous experiments have proven room-temperature ionic liquids (RTILs), "green solvent" salts that possess negligible vapor pressure and are molten below 100° C, to be effective chemical agents for catalysis, metal extraction, and gas separation. The goal of this research was to experimentally determine the suitability of mercury-specific RTILs for mercury removal from a gas stream. The bench-scale experiments pass redox-generated mercury vapor past a semi-permeable membrane coated with RTILs. A time-of-flight mass spectrometer detects changes in the temporal response of the mercury signal as mercury is removed from the gas stream and absorbed by the RTIL. Preliminary results indicate that two of the ionic liquids studied create up to a thirty percent mercury reduction of mercury in the gas, while one reduces values in excess of ninety-nine percent, suggesting the further evaluation of their use for mercury removal. Future research will be focused on RTIL optimization, application, and mercury and ionic liquid recovery. This research is part of an ongoing collaborative effort to reduce toxic contents from flue gas and other emission sources by using ionic liquid separation technology.

Improving Existing Homes for Energy Efficiency. KIMBERLY HULVEY (University of Tennessee at Chattanoga, Chattanooga, TN 37403) THERESE STOVALL (Oak Ridge National Laboratory, Oak Ridge, TN 37831).

Energy efficiency is a growing concern for people around the world. Conserving energy will not only save consumers money but also insure a better environment for the future. Energy is transferred in the form of heat through walls and windows of homes by conduction, convection, and radiation. Conduction is energy transmitted between two areas by the transfer of kinetic energy. Convection is heat transfer by fluid movement, and radiation involves waves of energy. To make homes more energy efficient, these methods of heat transfer need to be minimized. Existing products are being tested to help consumers select the best energy efficient products for their homes. The goal of this research is to increase ventilation and decrease infiltration. Ventilation is the process of providing adequate indoor air quality. Infiltration is the movement of outdoor air in and out of the inside of a building through inadvertent openings. Tests were done on an 8 ft x 8 ft wall with a window to determine thermal resistance using the Rotatable Guarded Hot Box test facility. Various insulation materials were applied to this wall and window in order to determine the best products for certain climates. Air leakage was characterized by regressing airflow rates against pressure differences. This characterization can then be used to predict infiltration for various environmental conditions. The test results were analyzed and used to develop a building simulation model employing the DOE 2.1E program. This model then estimated whole-building energy consumption for three sizes of homes in ten cities across the United States. Heating and cooling savings will be analyzed from this output. A summarized form of this data will be included in a consumer's guide to improving existing homes for energy conservation and savings. Previous work done on windows was also reviewed and will be included in the consumer's guide.

Investigation of a Rotating Arc Spark Plug (RASP) for use in Stationary Power Lean Burn Natural Gas Engines. MARK BENNETT (University of Kentucky, Lexington, KY 40506) JOHN WHEALTON, JOHN ANDRIULLI (Oak Ridge National Laboratory, Oak Ridge, TN 37831).

As our nation's energy demand increases, less efficient power plants may be forced back into operation and result in a negative environmental impact. Natural gas is a relatively ‘clean’ fuel source while alternative technologies are being investigated. The Rotating Arc Spark Plug (RASP) has potential benefits in helping the Advanced Reciprocating Engine Systems (ARES) program achieve efficiency goals by reducing cycle-to-cycle ignition variation, improving ignitability by increasing arc volume, reducing electrode erosion by the moving arc sites, and improving ignitability by means of higher plasma electron temperature. In theory, an axial magnetic field imposed on a radial gap spark plug will cause the radial arc to rotate circumferentially by the Lorentz force. The amount of rotation depends on the magnetic field strength (produced by high temperature permanent magnets), the current amplitude, and the duration of the spark. To investigate RASP potentials a high-speed camera capable of capturing 10,000 frames/sec is used to observe a single arc from each RASP tested. The spark plug arc can then be viewed at various points in time to determine amount of rotation and rotation velocity at pre-combustion pressures. The RASP will then undergo thermal testing in a natural gas test engine at various loads. Upon successful thermal testing an in-cylinder combustion pressure gage will be used to compare the RASP design to that of a conventional spark plug by measuring the pressure pulse variation from cycle-to-cycle. Monitoring the combustion pressure pulses and other performance parameters will also allow for optimization of engine control systems. If the RASP design shows less pressure pulse variation, increased combustion pressures, then the possibility of a leaner air/fuel mixture may be used to produce the same amount of power with higher fuel efficiency and lower exhaust emissions. Future work will include research into new materials for longer lasting spark plug electrodes, higher temperature magnets, and comparison testing.

Modeling of Antenna Structures for Implantable Monitoring Systems. CHARYMAR CINTRON (University of Puerto Rico, Mayaguez, PR 00714) PAUL D. EWING (Oak Ridge National Laboratory, Oak Ridge, TN 37831).

The interest in knowing how electromagnetic (EM) waves and fields affect our bodies has become a popular field of study during the last few years due to the fast proliferation of cellular telephones and other electronic devices. Stemming from this point, the purpose of this project is to model different kinds of antenna structures for an implantable monitoring system. These antennas will be implanted in a simulation model of a human body to determine the radiation and efficiency of the antenna as well as the specific absorption ratio (SAR) of the surrounding tissue. The different antenna structures that have been modeled are the monopole antenna, the electric dipole antenna and the loop antenna. In order to develop these models, a software named REMCOM was used. This software uses the Finite Difference Time Domain method (FDTD) to predict the performance of the radiating devices and provide an accurate prediction of the interaction of EM fields with biological tissues. Simulations will be performed using antenna structures implanted in a human abdomen model with a focus on antenna transmission efficiency and SAR. The greatest challenge of this work has been to struggle with the complex nature of human tissues and with the reduced size of the antennas. This work is part of a DOE-funded research program developing implantable electronics for real-time tissue perfusion monitoring.

Monocular Visual Servo Tracking Control of a Wheeled Mobile Robot. ERIC HOLCOMBE (Clemson University, Clemson, SC 29632) WARREN DIXON (Oak Ridge National Laboratory, Oak Ridge, TN 37831).

Motivated by the desire to navigate a wheeled mobile robot (WMR) without the requirement for multiple sensors or cameras, a new control strategy was recently developed that utilizes a single camera for autonomous navigation. The controller uses multiple views by a single camera to recover depth information through a Euclidean homography (a geometric relationship). Performance of this controller was previously demonstrated via a MATLAB simulation. The goal of the current project is to construct a proof-of-principle robotic testbed and implement the control theory. Preparations for the testbed included integrating specialized camera and computing with the WMR, building a target, and implementing the simulation in C/C++ for execution on the QNX real-time operating system. Converting the simulation code involved making a server program for live image processing and calculations and another program that would make calculations from recorded images. Additionally, code that employs the kinematic aspects of the controller was written and designed for execution by a program created for directly controlling robots with variable tracking and tuning capabilities. For the server code, an algorithm was developed to scan image frames for the necessary image feature information. The server program would then use the image feature coordinates to compute the translation and rotation information that was then transmitted to the main WMR computer. The results of this project are that a control strategy was demonstrated to force a WMR to navigate along a prerecorded image trajectory via feedback from a monocular camera system.

Partial Discharge in Voids in Epoxy as a Function of Pressure. DANIEL DESCHENES (California State University, Fresno, Fresno, Ca 93710) ISIDOR SAUERS (Oak Ridge National Laboratory, Oak Ridge, TN 37831).

Partial discharges (PD) are small surges of current, which occur generally in manufacturing defects in an electrical insulation system when the applied voltage exceeds a threshold value. Such undesirable discharges can over time deteriorate the insulation and shorten the lifetime of equipment resulting in a potentially catastrophic, costly, and unforeseen failure. Therefore, understanding PD and its causes can help diagnose problems and predict the expected lifetime of a dielectric. For insulation used in High Temperature Superconducting power equipment applications such as cables, transformers, and fault current limiters, PD is thought to be the primary degradation mechanism of the electrical insulation. One of the more common defects in a dielectric insulator is a void that forms as a bubble in the curing of an epoxy dielectric. To study the onset and pattern of PD, and to better diagnose electric insulation in general, two samples were prepared with artificial voids embedded in dielectric epoxy between parallel plane electrodes. The samples were also designed in such a way that the pressure in the voids could be varied. The PD onset was then investigated as a function of pressure in the void using Nitrogen and Sulfur Hexafluoride gas. Using a commercial digital PD detector with a sensitivity range from as low as 1 Pico-Coulomb(10-12 C) to several Nano-Coulombs(10-9 C) we were able to capture the PD pattern. Different PD onsets are observed as the pressure is varied for the two systems. The onset voltage generally follows the Paschen curve dependence on pressure for each gas. These samples were then modeled using the program FemLab to better determine the electric field of the sample and how this field affects PD. In order to help clarify the PD signals originating from the sample, and the significance the pressure in the void has on the PD, the general pattern of PD will be discussed, along with the importance of the dimensions of the void and how they affect the electric field in the sample. Some suggested mechanisms for the observed PD patterns will also be discussed.

Quantitative comparison of optical efficiency and spatial intensity distribution of dual hybrid luminaire designs. DAVID REED (Virginia Polytechnic Institute and State University, Blacksburg, VA 24061) JEFF MUHS (Oak Ridge National Laboratory, Oak Ridge, TN 37831).

Sociological influences have advanced the concern for more practical methods in energy consumption. Hybrid Solar Lighting incorporates the direct power of natural solar energy and the widespread energy efficient florescent lighting systems to deliver an advanced form of lighting that offers many benefits to its end users. First generation luminaires were designed to show project feasibility, but as the second-generation solar collector evolves, the need for more efficient luminaires has become an important area of interest. The hybrid luminaires that delivered the integrated light were composed of modified commercially available fluorescent light fixtures to accept the fiber optic solar light source. The inceptive luminaires consisted of two light emitting fibers placed strategically inside a traditional fluorescent light fixture, but furthering developments made use of a single light emitting fiber placed in a shallow, space saving design. The new design used a single 3M LF180EXN D side emitting fiber and a system of stranded fibers that utilized the remaining ambient light to the fullest extent. Testing was conducted in the ORNL Illumination Test Cell using a Radiant Imaging Pro-Metric Camera to measure the amount of light reflected off a removable wall coated in Spectralon reflective paint. Efficiency was determined by the totaling up the amount of flux measured off each side of the removable wall. Test results have shown promising results for our new design. The spatial intensity distribution results were less then desirable, but the optical efficiency indicated considerable improvement. The initial goal of the new design was to maximize the efficiency of the natural light, but our research has lead to more optimistic anticipation of progressive designs.

Research on Analysis of Integral Pressurizer Design in support of Nuclear Energy Research Initiative (NERI) Project. HORACIO VELAZQUEZ (Texas A&M- Kingsville, Kingsville, TX 78363) DR. GRAYDON L. YODER JR., PH.D., P.E. (Oak Ridge National Laboratory, Oak Ridge, TN 37831).

The International Reactor Innovative and Secure (IRIS) reactor is a new, pressurized water-cooled nuclear power reactor design. The reactor coolant system design incorporates a pressurizer, steam generators and eight reactor coolant pumps, all contained inside the reactor pressure vessel. The coolant system and its pressurization system are the main components used to improve efficiency and safety over existing reactor designs. Two potential Integrated Primary System Reactor (IPSR) pressurizer designs are being analyzed for this research project: a steam pressurizer, and a nitrogen gas pressurizer. These two integral pressurizer designs differ from most conventional types, because they incorporate a much larger interface more closely coupled to the reactor primary system. In the steam pressurizer design, the large interface between the saturated pressurizer and the subcooled primary coolant may lead to significant heat losses from the pressurizer and require a larger heater power to compensate. In the nitrogen gas pressurizer system, nitrogen diffusion from the pressurizer into the primary coolant can lead to significant quantities of dissolved nitrogen. Some postulated accidents have the potential to release this nitrogen as gaseous bubbles that could negatively impact heat transfer in the core. Engineering calculations examining heat transfer, mass transfer and diffusion coefficients are being evaluated in order to determine the operational specifications to avoid these negative impacts. In addition to analytical calculations, a computational fluid dynamics model (CFD) of certain parts of the reactor, using the FLUENT code, will be created to determine the operational characteristics of the reactor. These calculations will combine thermal fluid and mass transport to produce an accurate simulation of the reactor system. The ongoing research in this project will combine the analytical calculations and CFD modeling techniques, implemented by the FaST team, to help understand details of the pressurizer operation.

Ring Injection Dump Thermal Transient Analysis. JASON REAGAN (Pellissippi State TCC, Knoxville, TN 37933-0990) MARK WENDEL (Oak Ridge National Laboratory, Oak Ridge, TN 37831).

The Ring Injection Dump (RID) is part of the Spallation Neutron Source used as a place to stop stray hydrogen atoms that were not captured into the Accumulator Ring. The kinetic energy of these particles is deposited as heat in the RID. A recent steady-state thermal analysis model built in CFX5 has shown that at 200 kW, the bulk temperature of the concrete structure of the RID will exceed 65 °C in some areas, with localized points reaching 91 °C. This violates the upper limit specifications for concrete used in nuclear safety related structures. For this reason, a transient analysis of the RID was performed to characterize the response time to load changes and to verify the amount of load credit taken in the steady-state model to account for system shutdown periods. A new, less detailed model was created with HEATING 7.3. It was calibrated so that its maximum concrete temperatures agreed with the more detailed CFX5 model. The transient was then run for a twenty year simulation, with an assumed ten week duty cycle and an annual shutdown period of ten weeks. The analysis revealed an overall time constant of more than one year. Because the peak temperature in the final years of the transient agreed with the steady-state model, the load credit used in steady-state analysis was verified.

Robotics CAD Modeling: Using SolidWorks™ software. DARREN NORRIS (Roane State Community College, Harriman, TN 37748) CRAIG BRADLEY (Oak Ridge National Laboratory, Oak Ridge, TN 37831).

The Remote Systems Group at Oak Ridge National Laboratory is preparing a conceptual design of the remote handling equipment system for a new nuclear waste treatment facility. Because of the high levels of radioactivity expected in the treatment facility, the process components must be maintained and operated remotely. These design efforts are to develop: design requirements for the remote handling equipment, design concepts for the equipment, a maintenance plan, and a cost estimate for the remote handling system. Designing this equipment will help to shape the Archimedes Filter Plant for Archimedes Technology Group, LLC, a private company in San Diego. One of the deliverables for this project is to develop 3-D solid models of the remote handling equipment. This involved learning to use the state-of-the-art 3-D Computer Aided Design (CAD) software, Solid Works™, to prepare the 3-D models of the robotic and remote handling equipment. These models will help engineers design and plan the layout of the robotics equipment going into the Archimedes waste processing facility.

Spectroscopic Exploration of Spark Plug Erosion. KATHERINE WOODY (Middle Tennessee State University, Murfreesboro, TN 37132) JOHN WHEALTON (Oak Ridge National Laboratory, Oak Ridge, TN 37831).

When voltage is sent to a spark plug in an engine, the voltage causes a spark in the gap at the end of the spark plug. When this spark occurs there is erosion at the site of the spark. Spark plug erosion is caused by several factors, including the material of the spark plug and the pressure of the surrounding gas. This erosion is studied by spectroscopic analysis of the light emitted by the sparks. Through study of sparks in several gases, including argon, nitrogen, and instrument air, it was possible to determine whether a certain line in the spectrum was caused by an interaction with the surrounding gas or gases or by the impact with the electrode. This allows for more direct quantification of erosion per spark. Certain new types of spark plug, particularly a Rotating Arc Sparkplug (RASP), are thought to cause less erosion than traditional spark plugs, as they more evenly distribute sparks across the electrodes. Also, there are continuing studies on the RASP dealing with design and erosion.

Spectrum Analysis of a Spark Plug. CARL ENG (State University of New York at Stony Brook, Stony Brook, NY 11794) JOHN H. WHEALTON (Oak Ridge National Laboratory, Oak Ridge, TN 37831).

The photon emissions generated by the arc on a spark plug may be useful indicators of the materials that are being ablated off the spark plug. In theory, identifying the wavelengths of light emitted by the excited electrons in the spark plug material as it is being ionized within the spark plug gap would give information on the erosion of the spark plug. Specifically, the magnitude and rate of erosion could be determined on a spark-to-spark basis. Furthermore, the intensity of the spectra from the arc may reveal the ignition capabilities of the arc as some of the observed spectra are believed to characterize the energy of the ionized gas contained in the volume of the spark plug gap. Using a test chamber to vary pressure, spark plugs were run. The light from the spark was focused (via a fiber optic cable and lenses) into a spectrometer. Images of the spectra were then transferred to a computer by a charge injection device (CID) camera. Initial experiments were run on spark plugs designed for operation in a natural gas engine, which focused on a band of wavelengths ranging from 300nm to 750nm. In that range, two distinct peaks at 392.22nm and 395.69nm were observed on the used spark plugs that were not seen on the new spark plugs. Work is ongoing to identify the observed spectra. In addition, experiments are being conducted to establish a correlation between the other observed spectra and possible gas transition states within the arc. An examination of a larger spectrum range is being undertaken with the employment of spectrometers with sensitivities in the ultraviolet and infrared ranges of the electromagnetic spectrum. Initial data has yielded results worthy of further analysis.

Telomere and Chromosome Segmentation and Editing Tool. JUAN AGUILAR (Richard J. Daley College, Chicago, IL 60652) TOM KARNOWSKI / KEN TOBIN (Oak Ridge National Laboratory, Oak Ridge, TN 37831).

The Image Science & Machine Vision (ISMV) Group performs Applied Computer Vision Research & Development to address areas of biomedical, industrial and national security. The goal of this project is to develop a software tool to analyze cellular images in a biological imaging environment. The main development environment used in this project was Microsoft Visual Basic 6.0. The Telomere and Chromosome Segmentation and Editing Tool performs a set of complex algorithms and generates computational graphical data, in order to analyze images to locate the chromosome and telomeres, and estimate its volume. The data collected can successfully be saved and opened in graphical displays in the program. The ISMV Group scientists successfully tested many aspects of the program and reported several bugs as well as initiated new features for the software. The Telomere and Chromosome Segmentation and Editing Tool will provide valuable information to a wide range of biological researchers. The Tool will be used to study the chromosomes. This information is important because it will used to help determine why people develop cancer.

Test for Disruptions in ControlNet Communications due to Electromagnetic Interference. COLIN FINUCANE (University of Kentucky, Lexington, KY 40502) PAUL WRIGHT (Oak Ridge National Laboratory, Oak Ridge, TN 37831).

ControlNet is a system that enables programmable logic controllers to communicate over RG-58 coaxial cable. Allen-Bradley ControlNet systems are being used in many controls applications at the SNS. It is necessary to ensure that ControlNet systems are capable of communicating in the presence of Electromagnetic Interference. A major source of EMI at the SNS will be the pulsed power sources for the RF Systems. These pulsed power sources known as High Voltage Converter-Modulators utilize relatively new technology. The effects they will have on ControlNet systems using RG-58 media had not been studied. Tests were conducted with ControlNet in which the cable was arranged to maximize the effects from Converter-Modulator produced EMI. Communications were monitored over the network with counters on ControlNet software, counters in the controller's program, and with an oscilloscope for EMI influences on the signal. The tests were performed for the same period of time with the Converter-Modulator both off and pulsing at 120 kV. The results with the Converter-Modulator running did not vary significantly from those with it off. No bad information was received at either end of the RG-58. These results were expected because the ControlNet system is designed to be resistant to EMI. The results indicate that there will be no problem with properly installed systems. The actual systems will have more nodes communicating on their ControlNet networks in the presence of additional Converter-Modulators pulsing at 140kV. Further tests should be conducted with the actual systems when they are installed.

Testing on Op-amps. DERRICK GREEN (Georgia Institute of Technology, Atlanta, GA 30332) CHUCK BRITTON (Oak Ridge National Laboratory, Oak Ridge, TN 37831).

Op-amps are used in almost all electronics presently. However, many op-amps do not perform exactly as they are needed to. Therefore, many times the electronic devices which require their use are not always as accurate as they need to be. The op-amp under test was specifically designed to have as little as possible of the input-offset error encountered by most op-amps. In order to run the series of tests that were needed to have conclusive results, the computer program Visual Basic was needed. The computer software told the voltage input device when and by how much to increase the voltage and also where to start and begin. Essentially, the input device was controlled remotely from a computer to do the specialized things which this series of tests required. The testing involved running voltages through five different op-amps labeled D2, D4, E3, E4, and E5 at different temperatures. The voltage ranged from zero to 3.3 volts and was increased in increments of 10 millivolts with a 0.2-second delay between each different voltage. At each voltage, five readings were taken with 0.2 seconds in between each reading and then the average was calculated. An offset reading was also taken each day prior to any readings that might be taken. To find the offset, the input voltage was subtracted from the average actual output voltage. Once a sufficient offset was obtained, the input vs. offset data could be calculated. The Agilent 34970A Data Acquisition/Switch Unit was used to control the input voltage and then take the resultant output readings and feed them to the computer software which then put the readings into a spreadsheet in Microsoft Excel. The software did all of the configuring of the 34970A automatically each time. The HP 6236B Triple Output Power Supply supplied voltage to the board. This voltage registered on the Fluke 79III voltmeter as 3.308 volts DC for most of the readings, but it dropped to 3.307 volts DC for many of the readings at the end, as noted. For the variation of temperature, the Delta 9023 oven was used. The temperature range was from room temperature, which with the door on would range from 32-33 C, to 150 C. Overall, the op-amps tested extremely well. The amount of error was usually no greater than 2 millivolts. Moreover, this error was uniform for the voltages that the chip is supposed to be used for.

Tracking and Monitoring of International Shipping Containers. BENJAMIN HUEY (University of Tennessee, Knoxville, TN 37996) PAUL D. EWING (Oak Ridge National Laboratory, Oak Ridge, TN 37831).

The international shipping container industry faces many challenges when transporting goods across the oceans. Due to the large amount of cargo on a ship, only about one in four containers are searched by United States Customs agents for illegal articles when they enter the country. These uninspected containers could potentially contain counterfeit merchandise and pharmaceuticals, illegal drugs, illegal firearms, nuclear, chemical, and biological weapons, or even terrorists who want to bypass immigration. The RF & Microwave Systems Group at the Oak Ridge National Laboratory is working on a solution to give shipping companies better monitoring capabilities of their cargo. A wireless system is being developed that will provide real-time temperature, position, and door seal integrity data from each container to receivers on the ship or at the shipyard. This system will use an advanced radio-frequency spread spectrum algorithm, as well as Code Division Multiple Access (CDMA) and Time Division Multiple Access (TDMA) technologies to allow the system to gather data from a large number of containers. Small scale prototyping and testing have demonstrated that the system is affective at providing position and simulated temperature and door seal data at power levels below 125 mW over a range of about one quarter of a mile in line-of-sight conditions. This was somewhat better than expected by the design team; however the system’s performance was not as impressive in a non-line-of-sight environment where one transmitting tag was moved behind a building. Large scale operation will be tested by modeling the system using MatLab™ and Simulink™ software packages. This will allow testing of the spread spectrum and multiple access techniques, while taking into account a wide range multipath losses and interference variables without the investment of producing a large number of prototypes.

VEMPS Wiring Harness. STEPHEN BERNARD (Pellissippi State, Knoxville, TN 37920) SHEAN HUFF (Oak Ridge National Laboratory, Oak Ridge, TN 37831).

This project, for the FEERC department, involves designing and build a wiring harness that will connect the Vehicle and Engine Management Prototyping System (VEMPS) Machine up to a Mercedes four cylinder diesel engine. This device controls the Mercedes Engine. The computer in the lab sends directions to the VEMPS in digital signals and the VEMPS then in-turn translates the digital signals into analog signals and relays the information to the Mercedes engine. At the same time the reverse is happening, the VEMPS receives analog sensor signals from the engine and translates them into digital signals to send to the lab computer. Specifically it translates data like engine oil temperature, engine water temperature, crank shaft rotation, cam shaft rotation, The mass air flow sensor (MAF), the fuel rail pressure sensor, engine coolant temperature, intake air temperature etc. In addition to communicating directly with the Mercedes engine, the VEMPS communicates with other controllers, which in turn communicate with the engine. They are the injector drive unit (IDU), and the Throttle control. The IDU is a controller that allows us to tune the fuel injection timing and duration. The throttle controller regulates the intake throttle to maintain engine airflow. Student's Name: Stephen Bernard School Student Attends: Pellissippi State Technical Community College Mentor: Shean P. Huff Division: FEERC Program: CCI