Completing phase III of Chipmunk electrical packaging upgrade. YEVHEN RUTOVYTSKYY (Three Rivers Community College Norwich, CT 06360) VINCENT J CASTILLO (Brookhaven National Laboratory, Upton, NY, 11973)

At the Brookhaven National Laboratory (BNL) Collider-Accelerator (C-A) Complex radiation exceeding levels for safe human occupancy are produced as a standard by-product of normal operations. The "Chipmunk" is the radiation monitor that is used for radiation protection and up to 100 chipmunk monitors are located strategically around the C-A Complex during Beam operations. Since the chipmunk was designed in the early 80's, its electrical packaging is outmoded and needs to be updated. This update will enhance features of the old design that experience has shown necessary. These include independent circuit boards which will facilitate troubleshooting and repair along with a backplane in place of hand-wiring. This update is expected to curtail cost and reduced the manufacture cycle from 14-16 months to 3-5 months. The schematics for the project were drawn using MS Visio. My project is a culmination of a three-years effort by CCI students, who worked on the various electrical changes to the original model of the "Chipmunk". Testing and engineering critique of this and subsequent prototypes will lead to a model that is ready for mass production.

Computer Aided Engineering Analysis in the Development of the Liquid Helium eBubble Cryostat. GAVIN MCINTYRE (Rensselaer Polytechnic Institute Troy, NY 12180) MARGERETA REHAK (Brookhaven National Laboratory, Upton, NY, 11973)

Prior to the fabrication and experimentation, the eBubble cryostat undergoes structural, thermal, fluid dynamic and electrostatic analysis using ANSYS Classic and ANSYS Workbench. The analysis is important to the success of the experiment, because specific boundary conditions are set. To obtain proper data, no distortions can exist in the electric field. Depending on the analysis type, 2 and 3 dimensional models were constructed using computer-aided drafting software (Pro/Engineer and/or Mechanical Desktop), and imported into ANSYS. The structural and thermal models were developed using minimal symmetric sections of 3 dimensional drawings in ANSYS Workbench, while the electrostatic and fluid dynamic studies were done using an axisymmetric 2 dimensional drawing in ANSYS Classic. The structural analysis focused on the forces applied to the experiment and the material properties of the components. Stress analysis and deformation simulations were created to ensure that the combination of multiple bolt loads and an internal pressure of 1.05 bar did not surpass the ultimate tensile stress of 4.3 Mega Pascals (MPa). The applied forces created a maximum stress of 1.8 MPa, which is well in the acceptable range creating a minimal deformation of .29 centimeters in specific locations. Since the experiment is held at 4.4 Kelvin, thermal and fluid dynamic analysis determined any potential temperature loss that was within the appropriate range between various materials, as well as motion that could occur within the liquid helium due to temperature differentials. A few variations of electrostatic models were tested in order to pursue the best plausible electric field. The desired field would produce completely parallel electric potential lines, which will attract the freed electrons to the measurement components in a non-circuitous manner. The 2 variations consisted of copper rings with applied voltages or a printed circuit board with charged extrusions, both ranging from ground to -100,000 volts (V). The analyses, with the specific boundary conditions, proved that the chosen materials and designs functioned to specifications. The structural analysis proved to withstand all forces, and the liquid helium satisfied the thermal conditions, since the helium remained stationary. Although the circuit board had more of a variation in voltages, the rings produced a field with the most parallel potential lines, and a basic design was moved closer to fabrication.

Designing an optical system to extend the working distance of a surface scanning laser confocal displacement meter. MICHAEL KORNHAUSER (University of Rochester Rochester, NY 14627) PETER TAKACS (Brookhaven National Laboratory, Upton, NY, 11973)

A surface scanning laser confocal displacement meter, by means of a fast oscillating tuning fork and an optical lens system, measures a targeted surface's height to within .01 microns. In order to measure accurately, the working distance between the surface in question and the device must be 6mm with a range of .3mm in either direction. Extending the working distance to beyond 60 mm, while keeping the accuracy to within 1 micron and the range approximately the same, is necessary in order to examine the surface of a thick telescopic lens. Zemax is a computer program designed to model and optimize optical lens systems. After careful inspection of the scanning laser, it can be modeled accurately in Zemax and its working distance extended using an added lens system. From this, an actual lens system can be built which is very similar to the model created in Zemax. The working distance can be extended to beyond 60 mm, but there is no conclusive evidence as of yet as to whether the accuracy is sufficiently small. As for the range, it actually increased by a large amount and thus does not limit the ability to examine the telescopic lens.

Determination of energy generated in an electrical system in case of an electrical accident and calculating the possible hazards and risks. GREGORY JIMENEZ (SUNY New Paltz New Paltz, NY 12561) SWAPNA MUKHERJI (Brookhaven National Laboratory, Upton, NY, 11973)

Unusual occurrences such as arcing and short circuit faults in an electrical circuit produce potentially dangerous events such as an Arc-Flash and an Arc-Blast. Both are equivalent to an explosion. Limited, Restricted, and Prohibited Approach Boundaries have been determined from the following electrical standards such as National Fire Protection Association (NFPA) 70 (National Electrical Code), NFPA 70E (Standard for Electrical Safety in the Workplace) and the Institute of Electrical and Electronics Engineers (IEEE) 1584. Electrical software, "Power Tools for Windows" by SKM Corporation is used to calculate the energy generated by an electrical system. We can also use this software to calculate the Personal Protective Equipment (PPE) needed for possible hazards and risks. By going out in the field with a qualified electrician, we were able to verify information needed to design more accurate one line diagrams of some facilities in Brookhaven National Laboratory. This diagram details the circuits, connection and all electrical equipment utilized. The information gathered is then applied to Power Tools for Windows software for calculations on electrical systems. This work is a new study at Brookhaven National Laboratory that will improve Brookhaven's safety standards.

Development of an Enhanced Base Component for the Brookhaven Atmospheric Tracer Sampler. JOHN CORNWELL (Duke University Durham, NC 27708) RAY EDWARDS (Brookhaven National Laboratory, Upton, NY, 11973)

The Brookhaven Atmospheric Tracer Sampler (BATS) is an integral part of the Perfluorocarbon Tracer Technology (PFT) developed at Brookhaven National Laboratory. The BATS is used to actively sample released PFT by pumping air through adsorbent filled tubes. It is comprised of two components: the lid, which contains the adsorbent tubes, and the base, which contains the pump and electronic control systems. Although the BATS lids are still state of the art, the electronic logic within the bases has become outdated over the past 20 years. The bases lack programmability and are prone to failure in the field. The main goal of this internship was to design a cost-effective replacement base with complete flexibility in terms of sampling periods as well as programmability through a wireless interface. The Netburner 5272 processor board was selected for its speed, data capacity, and also its easy-to-use TCP/IP interface with an available 802.11b wireless expansion card. This device was programmed to serve a dynamic web page and allow sample periods, the real time clock, and other settings to be programmed simply by navigating the website and filling out forms. The pumping system was also replaced with a miniature diaphragm pump paired with a flow meter to allow the Netburner to implement a feedback controller. This allows for precise volumes to be sampled. A working prototype was constructed with a new pump, flow meter, and processor. Almost every design constraint was met and the prototype includes functioning wireless programming, backup data storage, and a pumping system controlled through a negative feedback loop. Although the prototype demonstrates proof-of-concept of the improved base unit, more work still needs to be done. A custom circuit board needs to be designed to hold all of the electronics and everything needs to be mounted in an enclosure. After these improvements the device will be ready to be tested during an actual PFT release.

External Switch Fabric Deployment at Brookhaven National Laboratory. JEAN ROBERT JR. BRUTUS (State University of New York - Stony Brook Stony Brook, NY 11790) W. SCOTT BRADLEY (Brookhaven National Laboratory, Upton, NY, 11973)

The advent of 802.11 wireless networking has brought about the proliferation of user-installed wireless access points (WAPs) on the internal campus network of Brookhaven National Laboratory (BNL). Seen as a potential security risk, these WAPs are being migrated to an external switch fabric. These new external networks for wireless devices join numerous existing external networks that serve various purposes. These networks are logically out in front of the campus network firewall, but traverse the same physical links and devices as the internal campus networks. In a qualitative investigation, considerations were made based on: user needs; security posture requirements and fiscal resources. A physically separate and logically distinct switch fabric was designed to isolate inherently less secure external network traffic from the internal campus network. This external infrastructure of switches and trunks is interconnected with fiber optic cables due to the distances between buildings. The project consisted of the analysis, the design and migration processes. This involved configuring switches so they can efficiently communicate on the same network being the new external switch fabric. Once major components of the external wireless switch fabric were installed, partial migration of the old network system to the newly installed switches was executed and tested. With this upgraded system, the advantages were numerous: easy access to the new external wireless network using existing WAPs; wireless network and support equipment designed to sustain an expanding wireless network user community; external network allowing users access to a network disparate from the main BNL campus; newly deployed network allowing visitors easy internet access when on campus; eradication of potential security risk for the internal network and improved reliability. This project is a small part of major upgrades that are being done on the Brookhaven National Laboratory (BNL) campus network in order to enhance its network performance and security. Future work will continue until the external switch fabric is totally implemented throughout the campus and fully operational by the end of December 2005.

Fire Modeling Analysis of Laboratory Facilities. ALAN KOUCHINSKY (University of Maryland College Park, MD 20740) MICHAEL KRETSCHMANN (Brookhaven National Laboratory, Upton, NY, 11973)

The occurrence of a fire in a typical laboratory environment can be detrimental to the research being conducted, the everyday operations of the building, as well as the monetary value associated with the destruction of the facility. In order to develop an accurate fire model for a laboratory room, a quantitative investigation of the combustibles and flammable liquids present, is needed, as well as the dimensions of the room, the materials of construction, and openings for natural ventilation. These parameters will determine the worst case fire scenario for a specified laboratory. The more combustibles and flammable liquids present, the greater the potential of a fire occurring causing work stoppage for months and millions of dollars worth of damage. Using CFAST (Computational Fire and Smoke Transport), a DOE (Department of Energy) sanctioned fire modeling program created by NIST (National Institute of Standards and Technology), a calculation of the time evolving distribution of smoke, fire gases, and the temperature throughout the building during a specified fire is determined. Two research labs were chosen to be modeled based on the rooms being classified as a high hazard, due to the flammable liquids present and nonexistence of suppression systems. Room A had an area of 1040 ft2 with a ceiling height of 19 ft. Room B had an area of 506 ft2 with a ceiling height of 8 ft. For each lab, the worst case scenario was modeled, which meant there was natural ventilation allowing a constant air supply to feed the fire, as well as 100% of the flammable liquids stored in the lab were spilt on the floor. Based on the type of flammable liquids and the amount of gallons spilt on the floor, using a Fire Dynamics Tool created by the Nuclear Regulatory Commission (NRC), the heat output of the fire and duration was determined of the insipient fire. Comparing the results of each simulation over a 15 minute interval, Room A showed that the fire only had an immediate impact on the room of origin due to the high ceilings for the hot upper layer of gases and smoke to concentrate. Room B differed in that the hot gases and smoke was confined to the lower ceiling, which caused them to bank down and fill not only the room of origin but the adjacent corridor. This work is a starting point to help analyze the need for suppression systems or other means of protection due to the high risk associated with a specific room or overall building use.

Fire Modeling Analysis of Storage Facilities. SEAN VAZ (suffolk county community selden, NY 11784) MICHAEL KRETSCHMANNPE (Brookhaven National Laboratory, Upton, NY, 11973)

Brookhaven National Laboratory has numerous facilities with various fire hazards within. I was selected to observe a number of storage facilities on site to determine their fire load and to model the building to see what the outcome would be in a fire emergency situation. The buildings were chosen based on identifiable concerns with particular chemicals within the structure. I used a program called Cfast to create a fire model of these facilities to determine the ending result in a fire situation. This program uses information such as, size of the building, fire load, and most important the specific heat release rate of the main fire and combustibles in the room. I use this tool to tell me the temperature of the room and the actual wattage of the fire. By telling us how hot the fire becomes at different layers in the building, and the temperature of the opposing rooms, I can accurately say if there is extreme concern for a hazardous plume release or if the fire doesn't reach the flashpoint of the hazard in question. The Nuclear Regulatory Commission (NRC) has fire dynamics tools that were used to verify the results that were acquired from the Cfast model. After telling us if our fire model was very accurately formed we were able to then able to take the results of the program and make copious graphs to present our data.

Modeling of Electronic Components and Software with Reliability Prediction Methods. SAMIR BHARGAVA (California State University Long Beach Long Beach, CA 90840) TSONG-CHU (Brookhaven National Laboratory, Upton, NY, 11973)

Electronic components and software are present in any proposed digital system, and can be effectively modeled in conjunction with reliability prediction prior to realization of such a system in hardware and software. One will examine methods of determining reliability of electronics using physics of failure and statistical methods. Physics of failure focuses on design aspects and external stress conditions. Statistical methods rely on past experience with similar components and stress conditions to predict future performance. A modified statistical model can capture the effects of processes on hardware reliability, such as, part quality, infant mortality, environment, design, reliability growth, manufacturing, system management, induced, wear out, and no defect. One will discuss methods of evaluating the reliability of software, and strategies for software reliability improvement in all phases of the software life cycle. Future work will focus on the application of reliability prediction methods for the development of mission critical digital systems, such as, nuclear reactors.

Modeling of Electronic Components with Reliability Prediction Methods. SAMIR BHARGAVA (California State University Long Beach Long Beach, CA 90840) TSONG-LUN CHU (Brookhaven National Laboratory, Upton, NY, 11973)

Electronic components are present in any proposed digital system, and can be effectively modeled in conjunction with reliability prediction prior to realization of a system in hardware. One will examine will examine the methods of determining reliability of electronics using two different statistical methods. Specifically, Military Handbook (MIL-HDBK-217F) and Prism can effectively model electronic components failure rates. Military Handbook relies on past experience with similar components and stress conditions to predict future performance. Military Handbook examines the effect of die complexity, case temperature, thermal resistance, operating supply voltage, active thermal design maximum current, maximum power dissipation, junction temperature, package design, environment, quality level, and improvement of design, on the failure rate of a component. Prism applies the effect of system level parameters, such as, environment, operating temperature and year of manufacture, as well as, component level parameters for computing junction temperature using four different methods of calculation. Due to more detailed input specifications at the electronic component level, Military Handbook generates more plausible results in capturing the differences among different electronic components within the same category. Prism grouped electronic components within the same category to have identical risk assessment, and applies junction temperature calculation methods at the component level. Prism can quickly model the effect of varying system level parameters to gauge the effect of the system on failure rates. Military Handbook generally results in a more conservative estimation of component failure rate and may not fully capture improvements in electronics reliability. Prism may oversimplify results for component failure rate by lumping too many electronic components in the same category, for example, all microprocessors and memory.

Pressure Distribution Simulation for ERL Loop Vacuum System. MICHAEL TIEN (Hobart and William Smith Colleges Geneva, NY 14456) H.C. HSEUH (Brookhaven National Laboratory, Upton, NY, 11973)

The Energy Recovery Linac (ERL) is a R&D project to test the principles of electron cooling for the Relativistic Heavy-Ion Collider (RHIC). The major components of the ERL loop are currently being designed and fabricated. Ultrahigh vacuum is required for the ERL loop vacuum systems to reduce the electron-residual gas scattering, and to minimize gas migration and particulate contamination to the super-conducting radio frequency (RF) accelerating cavity. The effect of the quantity, speed, and distribution of ion pumps on pressure distribution in the ERL was modeled using a computer simulation program called Vac-Calc. The simulation indicates that the designed pressure of low 10-9 Torr can be optimally achieved with 20 l/s ion pumps positioned approximately every two meters.

Super-Conducting Undulator development. JEAN CHRISTIAN BRUTUS (State University of New York - Stony Brook Stony Brook, NY 11790) JOHN SKARITKA (Brookhaven National Laboratory, Upton, NY, 11973)

Mechanical engineers and scientists from Brookhaven National Laboratory have been collaborating to design high-quality radiation insertion devices such as wigglers and undulators to achieve the required parameters necessary to perform state of the art research at the National Synchrotron Light Source (NSLS). However, as the goals of science advance, engineers have to create new devices to meet the scientific demands. The elliptical super-conducting undulator is one such new art device. To produce this undulator, various design prototypes for winding, cooling and manufacturing methods must be developed. For the design, we used Autodesk Inventor Computer Aided Design (CAD) software. This CAD software helps create and document the design from which the prototype will be manufactured. Computer Aided Manufacturing (CAM) software was used with a Computer Numerical Control (CNC) machine to automatically manufacture components of the undulator. GibbsCAM software is used to convert CAD files into codes that the CNC machine can read to fabricate a part. After learning the program, we have come to the point where an initial design has been prototyped out of wax. This method prevents damage to tools should there be errors in the codes. Improvements in the shape and different winding and cooling techniques are still being considered. When the undulator is completed, its performance will be evaluated in the X-ray ring of the NSLS. This work constitutes the initial phases of development for a new technology that will be incorporated in the construction of the NSLS II at the Brookhaven National Laboratory.

The Consequences of Demolishing the Main Stack. SABRINA THOMPSON (State University of New York - Stony Brook stony brook, NY 11575) ALAN RAPHAEL (Brookhaven National Laboratory, Upton, NY, 11973)

The environmental cleanup effort at Brookhaven National Laboratory is addressing the decommissioning of the Brookhaven Graphite Research Reactor (BGRR) and the High Flux Beam Reactor (HFBR). One component of the decommissioning is the main stack. The main stack was built in 1947 to support the BGRR's air cooling system and was used in that capacity until 1969 when science mission of the BGRR was terminated. When the HFBR was built, it was situated to take advantage of the existing stack; the HFBR operated from 1965 to 1996. Several other buildings have been connected to the stack over the years. Since neither the BGRR nor the HFBR is in operation, there is no longer a need for a 320-foot stack. The other, smaller buildings connected to the main stack could be accommodated by a smaller stack, if necessary. By researching historic drawings of the area surrounding the stack, we were able to find out which buildings had been connected to the stack. Drawings of the buildings' ductwork and heating, ventilating, and air-conditioning (HVAC) air flow systems were analyzed. Then field verification was performed at the different buildings to assure which buildings were currently using the stack and to verify existing conditions. After analyzing the drawings and existing conditions at the different buildings, we have come to find out which buildings are currently using the stack. Buildings which do not require air discharge will have their disconnections designed. We have found that Building 801 is the only building that really needs a stack to meet air discharge regulations. The plan is to design one or two smaller stacks for building 801.

The Construction of a Drain Trench to Prevent Floating in Building 911. JUAN CRUZ (Bronx Community College Bronx, NY 10452) ALAN RAPHAEL (Brookhaven National Laboratory, Upton, NY, 11973)

During a big storm, the sewer system around Building 911 gets so overwhelmed by all the water that the entire facility floods. The terrain around Building 911 is shallow, such that the facility is lower than the surrounding landscape. As a result, in an event of a storm, water from the buildings on high elevations, such as the Graphite Research Reactor and nearby facilities drain down to the area around Building 911. To address this problem, first, we need to survey the land. Surveying was done using a laser level and a measuring rod. There are 'Rim' elevations stored in our archives. We use these as a reference; we then used the measuring rod to find elevations on the surrounding area. We surveyed an area of 7440 squared feet. Then, we found the elevations of the parking lot in front of Building 923. This gave us an idea of how deep our pond could be and if we could have a slope for the water to run to the pond. We found the pond could be 5 to 6 feet deep and that there is a slope in the parking lot. Using a pay loader, we are going to dig a 19,200 cubic feet hole next to Building 923. We used the excavated soil to fill a hole next to Building 811. We will use this hole to store all the gallons of water that run from the parking lot to Building 911 and 912. The water will go into the pond via trench drain. We designed all the drawings for this project using AutoCAD. The drawings show pretty much how everything will look and work. These drawings will then be given to the contractor and the Environment Protection to be reviewed. Once the hole is dug we would have a place to store the storm run-off in an area of 48,611 square feet. With the help of the "basin" we are hoping to lessen the effect of storm water running into Building 911 and 912.