NAMRC 33 Highlights Manufacturing Research
Challenges and opportunities presented at Columbia University
By Brian J. Hogan
Robert B. Aronson
"If you want to be around in the future, you will have to invest in innovation." That was the advice of Michael Idelchik, vice president of advanced technology programs, GE Global Research Center, in his keynote address at NAMRC 33 held recently at Columbia University (New York). Two things he strongly suggests to achieve that goal is to understand emerging technologies--for example nanotechnology--and make the investment in R&D.
GE has an impressive list of innovative projects, particularly in manufacturing. GE consists of 11 businesses--from aircraft engines to NBC to commercial finance. Six of those businesses are industrial and contain manufacturing facilities in 32 different countries.
GE has major research centers in the US, India, China, and Europe employing over 2500 researchers.
Idelchik offered a glimpse at some of the key technologies under development at GE:
Medical: Molecular imaging. This is the emerging field of being able to use medical imaging (e.g. magnetic resonance imaging and computed tomography) to see the biological functions that are the very genesis of diseases like Alzheimers and cancer.
Security: Advanced monitoring systems that can more readily detect explosives, biological substances, or narcotics. This work can produce a new generation of systems for baggage-screening, port and cargo inspection, and security checkpoints.
Water: Water for consumption and industrial purposes will be a growing world need as population increases and industrial economies emerge, particularly in areas where water is scarce or aquifers are becoming depleted. To utilize water from the ocean and brackish sources, researchers are developing new desalination systems and membranes. Along with products that may reduce water scarcity, GE is also developing products that assist in water analysis and treatment.
Sustainable energy: Coal now accounts for 60% of the fuel burned to generate electricity in the US and 90% in China, so it's critical that we develop ways to make fossil fuels cleaner and more pollutant free. Researchers are working on gassification systems that remove CO2 and pollutants from coal at high temperatures. This results in higher thermal efficiency and a reduction in greenhouse gases and emissions. Another potential fuel saver is a hybrid locomotive that uses brake energy to recharge batteries.
GE's wind energy business is expected to hit $2B in revenue this year. GE is looking at large-scale wind systems--3-7 MW--and offshore installations. More efficient generators and composite blades are in the works. Another renewable energy source GE is exploring is converting biomass to useable fuel gas. Idelchik stated that renewable and carbon-free energy sources have the potential to generate 30-40% of the nation's power needs.
Composites: These "made to order materials" will have a number of uses. For example, use of composite materials in jet engines reduces weight, adds reliability and increases "time on wing." Lower weight can translate to increased fuel efficiency and reduced emissions.
Industrial Inspection: The US is suffering from an aging infrastructure. There is, therefore, a need for better diagnostic equipment to make both flaw detection and analysis easier, particularly in the area of corrosion. This category of technology includes devices that utilize eddy currents, and optical and ultrasonic systems to inspect everything from manufactured parts to chemical plants and refineries.
Nanomanufacturing: Through nano research we will see a step-change in material properties that will impact all industries. There will be a new generation of specialized materials that will offer unprecedented strength, ductility, conductivity, or thermal properties. For example, work is being done on specialized nano particles that could help with the detection of heart disease.
In all these projects, GE specializes in taking a cross-discipline approach to tackling the largest technical challenges "It's amazing what can be achieved, when a group of people with diverse talents focus on a single problem," says Idelchik.
"Behind all these projects, manufacturing technology is a key driver. The ability to adapt manufacturing processes to new materials and processes is a core competency of any product company in the 21st century. The future of US manufacturing will hinge on our ability to innovate and lead. Those who don't lead in technology innovation, will only have price to compete on. And that is not a place anyone in manufacturing wants to be," he concludes.
Geoffrey Boothroyd, of Boothroyd Dewhurst Inc., a man long known for his innovative thinking in the areas of automation, assembly, and design, delivered the Founding Lecture. In reviewing his career, he noted that his best ideas came to him when he was doing nothing, or working on something other than his initial problem.
Early in his professional life, after looking at a number of challenges, he decided to work on converting the "black art" of part feeding to a systematic technology. One of his early interests was working with vibratory bowl feeders to feed and orient parts as a way to simplify assembly. This was initially to help manual assembly operations so that the parts were presented in such a way that the operators could easily pick them up.
"At first my work attracted very limited interest, particularly in the US, although other countries that were more interested in productivity gave me some attention," he said. "The issue is that much of assembly involves human labor, so when you improve assembly, you increase productivity. However, the assembly workers and the part checkers are a big part of the cost."
Boothroyd's work ultimately led him to develop a part-categorization system with three part types: those easy to orient, those difficult to orient, and those impossible to orient. This systematic look at parts then evolved into programs assisting design for assembly. Designers could use this information to simplify part assembly by avoiding or minimizing the use of impossible or difficult-to-orient shapes. In turn, this type of design reduced labor content, and improved productivity and part quality.
His next area of research was to look into ways to reduce the number of parts in a product. Or, make one part do the work of two or more. Again there was initial reluctance to go along with this idea. Designers did not see the benefits because they were used to designing for manual assembly.
As automation became more critical to manufacturing, design for assembly helped unify product engineering and the popularity of concurrent engineering evolved in more advanced companies.
"But," cautions Boothroyd, "automation isn't the answer to every problem. The manufacturer must be product-sensitive for the most efficient results. For example, initially robots were added to many systems without a clear understanding of what they could do. Often they were not suitable for the type of assembly action required. There is a need to carry out a careful analysis and get a cost per part. This should include all aspects of the job."
In discussing research in general, Boothroyd noted: "I'm not telling the research community what to do. But there are obvious problems. Keep in mind the millions of manufacturing jobs that we have already lost in the last five years.
"Our exports are currently half the amount of our imports. And about 40% of the cars on US highways today come from abroad.
"To reverse this trend, our research should explore subjects of interest to manufacturers and cover a wider range of subjects. And above all you should emphasize work on problems of practical interest," he concluded.
Research projects discussed at NAMRC 33 covered a tremendous range of scientific investigation in 77 papers. In the following excerpts, we look at a few of those papers that represent the type of work being done these days in manufacturing research laboratories.
In a paper entitled A Statistical Mechanistic Model of Acoustic Emission Generation From Shear Zone of Machining, Satish Bukkapatnam of the School of Industrial Engineering and Management, Oklahoma State (Stillwater), and Ding Chen Chang, Daniel J. Epstein Department of Industrial and Systems Engineering University of Southern California (Los Angeles) look at the atomistic origins of acoustic emission (AE) waveforms from the primary deformation zone of a machining process. Based upon a statistical mechanistic framework, their model quantifies the release of AE from the annihilation of dislocations in the shear zone.
Schematic illustrates the concept of AE ray propagation
The US National Research Council says monitoring of the microdynamics of machining is important for precision at the mesoscale and for integrity assurance. Microdynamics results from interaction among plastic deformation, shear localization, microcracking, and contact mechanisms in machining interfaces (such as the shear zone). Machining dynamics has been shown by researchers to be nonlinear and quite likely chaotic. Outputs of AE from machining tend to occur in bursts and occur on a wide band of frequencies. Information sensitive to microdynamic anomalies is buried in bursts and transients. High-frequency AE waveforms--commonly filtered as noise--contain information on microdynamics.
In their paper, the researchers go through the mathematics underlying their work on detecting the annihilation of dislocations. Their model includes information such as the distribution of mechanical properties, macroscopic variables like strain, strain rate, stress, stress rate, and temperature near deformation zones. These data were obtained for machining operations performed at different cutting parameters (depth of cut, rake angle, and cutting speed, etc.).
After a series of experiments conducted on an EMCO F1-CNC milling machine, the researchers concluded that their model captures the significant patterns observed in AE waveforms from machining processes. They express the hope that this work will spur efforts toward finding more atomistic sources of AE, and lead to the effective use of AE to monitor high-precision machining and related operations.
Their future research includes plans to study the application of the model to the extraction of features to detect instabilities in ultra-precision machining and chemo-mechanical polishing. They feel that these features can result in the early detection of instabilities in ultra-precision machining before chatter can fully develop. This capability might eliminate many of the quality problems encountered when manufacturing high-precision and mirror-like surfaces.
Composite materials are becoming more important to aerospace, naval, space, and automotive companies. Skin temperatures in next-generation military and commercial aircraft may approach 177ºC. Because of their light weight and stability at elevated temperature, various titanium-graphite hybrid composite (TiGr) materials have the potential to solve this problem.
During these experiments with solid lubricant, surface grinding was done in a plunge mode.
In a paper entitled Machinability of Titanium/Graphite Hybrid Composites in Drilling, D. Kim of the School of Engineering and Computer Science, Washington State University (Vancouver), M. Ramulu, Department of Mechanical Engineering, University of Washington (Seattle), and W. Pedersen, University of Minnesota Duluth examine the challenges involved in drilling TiGr. Made of thermoplastic polymer-matrix composite (PMC) plies with titanium foils as the outer plies, TiGr composites are assembled by bonding the PMC plies and the foils to form a hybrid composite.
Drilling composite-to-metal stackups is difficult because of the different machining properties of metal and PMC. Problems encountered include tool wear, head-induced damage, hole size-and-roundness variations, surface texture, and titanium burrs. The researchers used a Bridgeport vertical mill with a Fujitsu Fanuc System 3M - Model A CNC controller. Fanuc DC servomotors were installed on the machine to move the worktable (X and Y axes and the spindle Z axis.
Experimental results indicate that drilling conditions play a major role when drilling TiGr composites. Titanium chips were long and continuous at low speeds, and became shorter and stiffer as feeds increased. Polymer composite chips were continuous at low feeds and became dust-like chips as feed increased. When looking at surface texture, matrix smearing and fiber pullout were found in the TiGr holes. Pit depth resulting from fiber pullout depended on the manner in which cutting load was applied, and relative angle between fiber orientation and cutting direction.
Hole damage was similar at the entrance and exit of the drilled holes. Titanium burrs and delamination between titanium and PMC occurred at the hole's exit as well as its entrance. Hole-size error was minimal when feed was nearly equal to one ply thickness. Feed rates exceeding one ply thickness resulted in deep fiber pullout with a negative effect on hole quality. Finally, both thrust force and torque increased linearly with feed at all four of the drilling speeds employed during the experiment.
Conventional grinding operations employ flood lubricant, largely due to the intense heat generation inherent to grinding. In a paper entitled Elimination of Cutting Fluids in Grinding: An Investigation on the Application of Solid Lubricants, V. Radhakrishnan, Director-Research, Amrita University (Coimbatore, Tamil Nadu, India), and S. Shaji, Department of Mechanical Engineering, Government Engineering College (Trivandrum, Kerala, India) report that, if solid lubricant can be correctly applied to the grinding zone, it could be an effective alternative to conventional flood coolant.
Despite the well-known "air barrier" that limits entry of coolant to the grinding zone, and film boiling caused by heat in the grinding zone, flood coolant still plays an important role in grinding by providing excellent wheel cleaning and bulk cooling. Unquestionably, however, many environmental and fluid management issues are associated with conventional coolants.
In their experiments, the researchers used high-temperature solid lubricants, specifically graphite, calcium fluoride (CaF2), barium fluoride (BaF2), and molybdenum trioxide (MoO3) mixed with water-soluble oil and general-purpose grease to form a paste. The experiments were performed in a Blohm 6.5-kW horizontal spindle surface-grinding machine. Solid lubricant in a paste form was loaded into a cylinder and pushed onto a rubber wheel, freely run by the grinding wheel, that transferred paste to the grinding wheel's working surface. Flow rate was calibrated for different pressures placed upon the cylinder piston. Surface grinding was done in a plunge mode.
Speaking in general terms, the trend of the results obtained with the solid lubricants was the same for all lubricants examined. Detailed performance studies were done and compared to a normal grinding operation. Process parameters related to wheel-workpiece friction, such as tangential force, grinding force ratio, temperature, surface finish, and residual stress all improved. But wheel loading due to the absence of any effective removal of swarf was a major hindrance.Researchers state that, if proper application of the solid lubricant to the grinding zone can be ensured, with some means to avoid wheel clogging, solid lubricant application could emerge as an alternative to the use of flood coolant in grinding.
Crack initiation during hemming and bending is detected by a new method and optical system developed by a team of researchers--S. J. Swillo, G. Lin, S.J. Hu, K. Iyer, J. Yao, and M. Koc--from the Department of Mechanical Engineering at the University of Michigan (Ann Arbor), and W. Cai of the Manufacturing Systems Research Laboratory, R&D Center, General Motors Corp., (Warren, MI).
To detect crack initiation during hemming, the cumulative lengths of microcracks are plotted against the total number of microcracks. The total number of microcracks and average microcrack length can be used to determine the initiation of cracks and to evaluate hemline surface quality. The average lengths of microcracks corresponding to macrocracks are found to be nearly constant, regardless of hemming process conditions. Average microcrack length from each captured image of the hemline surface is used as an index for surface-quality evaluation.
Two identical monochromatic CCD cameras were used in the experimental setup. One is positioned perpendicular to the hemline surface to monitor surface quality, while the second records hemming height from the side view. Two 100-W white-light reflectors are used as light sources. At selected hemming stages, images of deformed specimens are captured and saved with a frame grabber and built-in image acquisition software for further processing.
After image processing the microcracks in the hemline can be counted and the length of each microcrack measured. Measurements demonstrate that the number of microcracks first increases, then decreases after a critical point, while the cumulative length of microcracks keeps increasing. The critical point is explained by the transition of microcracks, through propagation and conglomeration, into macrocracks (material failure at the hemline surface).
The procedure was demonstrated with the hemming of 6111-T4 aluminum alloy. Measurement results were correlated with human visual inspections. Researchers believe that the total number of microcracks, and the average length of microcracks, can be used to determine crack initiation and to evaluate hemline surface quality. Because average microcrack length corresponding to crack initiation is nearly constant, it may be used to characterize hemming quality. The researchers expect their proposed methodology to be applicable to general bending processes.
For more information on NAMRI, or to obtain papers from NAMRC 33, go to the NAMRI Web site at www.sme.org/namri.
This article was first published in the August 2005 edition of Manufacturing Engineering magazine.