|
Manufacturing Metrology Acoustic and Vibration Research Predictive Process Engineering Sensor Interfaces And Networking
|
Manufacturing Metrology Division Contact: Kevin Jurrens Increasingly, advanced optical systems are designed around high-accuracy, aspheric optical elements. Measuring the figure error of generalized aspheres to the required accuracy needed by industry is a complex and unsolved problem. NIST worked with commercially available phase-measuring interferometers and demonstrated that software compensation for some systematic errors is possible. When no null lens is used, other issues limit the accuracy achievable with these instruments. In response, NIST initiated efforts to develop a next-generation instrument, which is currently being commissioned. The resulting measurement service will support the manufacture of lightweight, high-performance optical systems for space-based applications and multi-layer mirror systems for extreme ultraviolet and X-ray lithography. Research also is being carried out to provide the technology needed for the measurement of flat wafer surfaces, either in the free-form or as-chucked states. Specific goals are to provide interferometric measurements of flatness on as-chucked 300-millimeter diameter wafers and to develop and demonstrate infrared interferometric measurements of wafer thickness, thickness variation, and bow. This research is motivated by the decreasing lithographic depth of focus budgets, combined with larger silicon wafers, in lithographic applications. Conventional vacuum chucks, used to hold the wafer during processing, can introduce distortions in the wafer as well as thickness variations in the wafer itself. These combined effects may reduce the process latitude. Capacitance-based tools are used widely today for wafer geometry measurements, but they have some limitations. Optical techniques are being developed in a number of organizations, but initial intercomparisons show significant measurement divergence. Instrument developers and wafer manufacturers alike have expressed the need for calibration artifacts including calibrated optical flats as references for the instrument makers and reference wafers with mapped thickness variations for the instrument users. The optical metrology tools developed in this program will provide traceable measurements for 300-millimeter wafers at uncertainties compatible with future lithographic processes. Minimizing subsurface damage also is key to reducing fabrication costs and improving in-service performance for many optical and electronic components made from single-crystal materials. The lack of reliable means to measure subsurface damage increases fabrication time and costs, and it necessitates additional steps to ensure adequate performance. Consequently, damage incurred early in manufacturing typically is not detected until much later in the fabrication process. Although components appear fully functional, they often are badly impaired. This is equally true for glass, metal, and ceramic parts. NIST's approach is to define the physical nature and extent of the damage, using the best available method. Using NIST laboratory facilities, researchers apply one of several non-destructive evaluation methods to parts representative of ongoing fabrication steps. Optical microscopy—in one of its many forms—and X-ray topography are the methods of choice. Contact: Bob
Polvani Acoustic
and Vibration Research Our acoustic and vibration research supports a variety of industries and essential governmental functions, often leading to standards and to testing and measurement methods that improve industrial and scientific capabilities. Acoustic and vibration measurements underpin a broad spectrum of activities, including noise control and abatement, health and safety programs, product development, acceptance testing, condition monitoring, and object detection. Some of the economic impacts are very large. Acoustic measurements of new jet engine noise levels can have multibillion-dollar impacts. Vibration and acoustic measurements in product development in the auto industry are extensive. National goals in health and safety also are very strongly affected. NIST researchers use advanced signal-processing techniques to measure and characterize frequency-dependent sensitivities of transducers and instrumentation used in the generation or measurement of sound, vibration, and mechanical shock. For acoustic research, three anechoic chambers, one with a volume of 450 cubic meters, are available for measurements and calibrations of loudspeakers, microphones, acoustical arrays, and hearing aids. Pressure response levels of customer microphones are obtained from averages of comparison calibrations performed using two NIST standard reference microphones. These standard reference microphones are calibrated periodically by the reciprocity technique. Typical expanded uncertainties for pressure calibrations of customer microphones are 0.09 decibels or less at frequencies from 50 hertz to 7 kilohertz, and 0.32 decibels or less at frequencies from 7 kilohertz to 20 kilohertz. Free-field response levels of customer microphones are obtained by the reciprocity technique. Five measurement systems have been developed for vibration research. These systems produce highly linear, uniaxial, sinusoidal motion for the excitation of test or reference accelerometers. Accelerations are determined via fringe-counting interferometry, quadrature interferometry, minimum-point interferometry, fringe-disappearance interferometry, or reciprocity. These systems permit measurements and calibrations of the sensitivity of accelerometers by absolute methods or by comparison with NIST reference standards. The estimated expanded relative uncertainty is 0.5 percent to 4 percent over a frequency range of 2 hertz to 20 kilohertz and, depending on the frequency range, an acceleration range of 0.2 meters per second squared to 2,000 meters per second squared. Contacts: David
J. Evans (Vibration Research) NIST research on force supports a variety of industrial sectors, including the aerospace, automotive, weighing, and construction industries as well as manufacturers of materials and testing equipment. Our results also assist private-sector research and development efforts, and they contribute to the development of standards and improvements in calibration services. Force measurement capabilities include a unique, fully automated facility that is known worldwide. The force laboratories contain six thoroughly characterized dead weight machines with capacities of 2.2 kilonewtons (kN), 27 kN, 112 kN, 500 kN, 1.334 meganewtons (MN), and 4.44 MN. These machines provide a capability to realize force over the range from 44 newtons to 4.44 MN, with a relative combined standard uncertainty of 5 x 10-6. An additional hydraulic-based system permits force measurements up to 54 MN. Facilities have also been developed for electromagnetic compatibility testing and for certification of prototype force transducers in accordance with requirements of the National Type Evaluation Program. While the facility is normally maintained at 23 degrees Celsius, most of the deadweight machines are equipped with environmental chambers that cover a temperature range from -10 degrees Celsius to 40 degrees Celsius. Contact: Zeina Jabbour Research in the area of mass supports the realization and dissemination of the fundamental unit of mass, its traceability to the international standard, and the measurement of solid density. This research impacts a broad spectrum of industries, including the pharmaceutical, instrumentation, and nuclear industries. Further, mass is a fundamental unit and is key to the definition of derived units in both mechanical and electrical metrology. Thus, mass research potentially affects all manufacturing and all technical communities. Measurement capabilities include a state-of-the-art class 1,000 clean room with high levels of temperature and humidity control. This facility houses comparators that enable kilogram comparisons to the national prototype kilogram, achieving a combined standard uncertainty of less than 20 micrograms, and 10-kilogram comparisons to a combined standard uncertainty of less than 200 micrograms. The mass laboratories house a total of 20 balances that provide the capability for mass measurements over the range from 1 milligram to 1,100 kilograms and a platform scale for mass measurements up to approximately 27,200 kilograms. Three of these balances also enable susceptibility and solid density measurements of mass artifacts. Contact: Zeina
Jabbour Predictive
Process Engineering This research aims to develop the process models, methods, measurements, and standards needed to enable "first part correct" manufacturing capabilities. Due to the inherent complexity of manufacturing processes, process development is often ad-hoc and empirical. Manufacturing engineers typically rely on engineering handbooks or costly trial-and-error prototyping to specify process parameters, such as machining speeds, feed rates, and tool selection. Not surprisingly, the chosen process parameters often are far from optimal. One study indicates that less than optimal process selections cost U.S. industry an estimated $10 billion per year. A principal barrier to reducing these inefficiencies is the lack of access to validated, physics-based models of the manufacturing processes when key engineering decisions are being made. Process knowledge and data are necessary and useful throughout the product life cycle, from evaluation of product design through process planning, production scheduling, control, and beyond. The NIST Predictive Process Engineering program pursues key advances in process development, modeling, data representation, and metrology. First, research will establish a suite of rigorously defined representations for manufacturing process information. These representations will be based on first principles and will enable integration of process-related applications across extended enterprises. In support of this objective, the NIST Process Specification Language project is developing and standardizing a neutral representation format for exchanging process information among manufacturing applications. Related efforts focus on process capability information and on specification of data requirements for the integration of design and process planning applications. Second, the predictive process engineering effort will establish a publicly available set of validated, physics-based models of milling and turning processes to support next-generation industrial priorities in planning, analysis, optimization, and real-time control. Development and validation of these process models is supported through program efforts in process metrology to establish the measurement and validation test methods to obtain in-situ, real-time, or process-intermittent information about the milling and turning processes. Program results are highlighted and demonstrated through a prototype process integration framework to showcase the current status and capability and to illustrate the usefulness and value of incorporating predictive process models to improve the applications of design, analysis, planning, optimization, and real-time control. Contact: Rob
Ivester U.S. manufacturers face a multitude of challenges. These include global competition, demand for more complex parts with closer tolerances, smaller batch sizes, shorter time-to-market, just-in-time production, and effective coordination of production operations that are distributed worldwide. To survive, manufacturers must constantly look for ways to meet this "faster, better, cheaper" mantra of today's economy. Machine tools, such as milling machines, lathes, and grinders, play an important role in this quest. They are the key equipment used to manufacture parts and the tools and machines to make parts, such as molds and dies. Although integral to manufacturing sector performance, the machine tool industry is relatively small. (The entire U.S. machine tool industry, if it were a company, would rank only 307th in the Fortune 500, at $5.8 billion.) Items ranging from autos, airplanes, and refrigerators to computers and paper clips could not be produced without metal-cutting and metal-forming machinery. Consequently, advances in machine tool technologies have highly leveraged impacts: productivity gains, declines in inventory requirements, and improvements in product price, quality, functionality, and energy efficiency. Manufacturers need accurate and reliable machine tools. Ideally, performance capabilities can be ascertained for a large variety of tasks and operating conditions. With this advance knowledge, manufacturers can determine the range and features of products that can be produced, the likelihood that the first and every subsequent part will meet specifications, and the efficiency and agility of operations. Together with industry and academia, the scientists and engineers of the Smart Machine Tools program are developing the metrology, smart sensor systems, applications, and standards needed to realize a new generation of smart machine tools with the intelligence to interpret and communicate performance capabilities in part-specific terms. Their work focuses on two key aspects: accuracy and reliability. In the area of machine tool accuracy, NIST researchers are developing new metrology concepts and improved standards to reduce the time and expertise required to specify, validate, and monitor machine accuracy. Machine tools have many sources of errors that can affect accuracy in complex ways. Some of these errors change over time due to wear and collisions and need to be assessed periodically. Therefore, non-intrusive methods that allow a machine tool to monitor and improve its accuracy are of particular interest, as are procedures that allow a machine to determine whether or not it can produce a specific part to the requested tolerances. Traditional quality control methods, which are based on trends in the inspection results of similar parts, are difficult to apply in an agile environment. NIST also is working on methods to estimate trends in machine errors from the inspection results for different parts. Other research includes development of in-situ traceable inspection procedures that allow the machine tool itself to determine the quality of its work. Together with industry, NIST also is creating standardized formats for machine tool performance data. This work includes developing tools to facilitate archiving of data and communicating it to manufacturing applications, such as e-commerce, quality control, machine programming, simulation, and maintenance. In the area of machine tool reliability, NIST researchers are working on metrology and standards for condition-based maintenance. Here, sensors and controller information are used to detect subtle changes in friction, temperature, and vibration patterns that signal the onset of failure in spindles, drive systems, and other key machine tool components. The work provides the foundation for efficient preventive maintenance and remote diagnosis of machine condition. Another research aim is to develop standardized procedures to specify the mean time between failure for machine tool, based on the mean time between failure for key machine components. Contact: Johannes
Soons Sensor Interfaces and Networking Sensors and actuators, commonly referred to as transducers, are used in a wide variety of products, such as automobiles, airplanes, home appliances, medical devices, and industrial machinery. Increasingly sophisticated, small, and capable, sensors are devices that measure quantities such as pressure, acceleration, flow, force, temperature, vibration, torque, position, chemical composition, and other process or environmental variables. The rapidly growing multibillion-dollar sensor market uses transducers extensively for industrial automation and process control, condition monitoring of machinery, building-system management, and intelligent highway applications. Automobiles, for example, have many sensors, ranging from accelerometers in airbags to gauges for oil and tire pressure. Enhanced capabilities are being introduced into transducer systems by integrating information, intelligence, digital communications, and Internetworking. Enhancements enable a variety of new and innovative applications of smart transducers. This includes connecting multiple transducers to a single digital network, which greatly simplifies and improves reliability and scalability. Development of cost-effective applications for smart transducers, however, has been inhibited by the lack of standardized interfaces for connecting transducer devices to microprocessors and field networks. It is not economical for transducer manufacturers to develop custom interfaces to support the multitude of networks and protocols currently in the marketplace. In response to this problem, NIST is working with industry and the Institute of Electrical and Electronics Engineers (IEEE), a voluntary standards organization, to develop a set of common interfaces for smart transducers. The objective is to solve the device interchangeability problem and to further progress toward universal plug-and-play compatibility between transducers and networks. The standard interfaces, collectively known as the IEEE 1451 series, will provide the enabling technology for seamlessly integrating and networking transducers in the distributed measurement and control arena. With IEEE 1451, transducers will have the intelligence to identify themselves and work with any type of industrial network and instrumentation system on the market, providing "plug-and-play" systems. Using the standards, transducer manufacturers design their devices using a common standard that allows interoperable connections to networks, computers, and instrumentation systems. With universal transducer networking capabilities, transducer data can pass easily from the device through the local area network and eventually to the Internet, thereby facilitating remote monitoring applications. The standards will solve interoperability problems, minimize the risk of technology investment and obsolescence, and accelerate the implementation of smart transducer technology. Contact: Kang
Lee
Date
created:
Dec. 17, 2001 |