Manufacturing Research Community Steps Up
Each year SME's North American Manufacturing Research Institute holds a conference and members report on the results of their labors; here are some highlights from NAMRC 38
By Brian J. Hogan
By Ellen Kehoe
The 38th North American Manufacturing Research Conference (NAMRC) had a homecoming of sorts at Queen's University (Kingston, ON, Canada) May 25–28. During NAMRC's first eight years, the late Professor William B. Rice of Queen's chaired the scientific committee, thereby significantly shaping the research agenda. NAMRC, however, had never before been hosted at the University.
NAMRC 38's opening session began with a tribute to Rice, as well as to Erich Thomsen and John Schey, both early supporters of NAMRC and pioneers in manufacturing research, who passed away earlier this year. Next on the agenda was a keynote panel discussion titled: Connecting Manufacturing Research to Manufacturing Engineering Education. Panel members were Steven R. Schmid, associate professor of aerospace and mechanical engineering, University of Notre Dame (South Bend, IN), Steven R. Hayashi, principal engineer and manufacturing platform leader, GE Global Research Center (Niskayuna, NY), Ron Bennett, director of the Minnesota Center for Engineering and Manufacturing Excellence (Mankato, MN), Scott Smith, professor, University of North Carolina–Charlotte, and Dianna Chong, vice president of assembly, factory and support technologies, The Boeing Co. (Chicago). Thomas R. Kurfess, professor and BMW chair of manufacturing, and director of Clemson University's (Clemson, SC) International Center for Automotive Research served as panel moderator.
"Bringing research into the classroom is time-intensive for faculty, but it's a must," observed Schmid. Discussions with students are very inspiring, and help develop the higher-order, critical thinking skills demanded in the workplace, added Hayashi. "We hire for problem-solving ability. We need to cross-pollinate teachers, professors, and industry."
All panelists agreed that fundamental knowledge is the first step. "Some basic technology [e.g., thermodynamics] is the same as it was 40 years ago. The challenge is how to apply it in new ways," said panel moderator Kurfess. "The mantra of 'innovate, innovate, innovate' means nothing if engineers can't manufacture the product," he added.
The tradition of recognizing outstanding research and individuals in manufacturing continued at NAMRC 38. The 2010 NAMRI/SME S.M. Wu Research Implementation Award honored Richard E. DeVor and Shiv G. Kapoor of the University of Illinois (Urbana-Champaign) for innovative research—beginning in the early 1980s—on mechanistic simulation models for machining processes. These models have benefited industry in product design, process planning, and machine-tool system design at conventional and micro scales.
The NAMRC 38 Outstanding Paper Award went to J. Karandikar, R.E. Zapata and T.L. Schmitz of the University of Florida (Gainesville, FL) for Incorporating Stability, Surface Location Error, Tool Wear, and Uncertainty in the Milling Super Diagram (SME order code: TP10PUB13).
Several 2010 SME Richard E. Morley Outstanding Young Manufacturing Engineers accepted their award at NAMRC: A. John Hart (University of Michigan), Raja Kountanya (Diamond Innovations), M. Ravi Shankar (University of Pittsburgh) and Mike Vogler (Caterpillar Inc.).
The following material was taken from papers representative of those delivered at NAMRC 38 by members of the North American Manufacturing Research Institute.
It's widely agreed that nanotechnology will have a significant effect in many areas of technology. In technical paper TP10PUB11, Performance of Graphite Nanoplatelet-Enhanced Fluid in Reduced Quantity Lubrication Centerless Grinding, researchers from the Manufacturing Research center at the George W. Woodruff School of Mechanical Engineering at the Georgia Institute of Technology (Atlanta) sought to evaluate the performance of a graphite nanoplatelet-enhanced fluid in reduced quantity lubrication (RQL ~100 mL/min) centerless grinding of Inconel 718 and Ti-6AL-4V superalloys as an alternative to flood cooling. Effects of graphite platelet diam and concentration on specific grinding energy were evaluated over a range of material removal rates.
Effects of graphite platelet diameter and graphite concentration by weight on specific grinding energy were investigated at three different removal rates in plunge centerless grinding of 70.94-mm diam Inconel 718 and Ti-6Al-4V workpieces. Experimental results were compared to results obtained using the same grinding fluid without any solid lubricant additive under flood cooling (about 5 L/min) and dry grinding conditions. Dispersion of graphite nanoplatelets in the oil was done using a Misonix Sonicator S-4000 with a probe diam of 12.7 mm.
Experimental results point to several advantages of the nanoplatelet-loaded fluid that could increase process productivity, reduce abrasive consumption, and reduce grinding fluid costs. Although nanoplatelet additives are expected to increase the cost of the grinding fluid, the reduction in fluid used in RQL (a minimum of 50% reduction in fluid volume) relative to flood cooling will yield a net decrease in grinding-fluid expenses.
Also, when grinding the Inconel 718 and Ti-6Al-4V alloys, RQL using cutting oil enhanced with 2% graphite nanoplatelets reduced specific grinding energy levels by up to 19% for Inconel and 15% for titanium alloy relative to flood cooling over the experimental range of material removal rates. Consequently, the graphite-nanoplatelet fluids may allow more aggressive material removal without causing thermal damage.
Finally, RQL using the cutting oil enhanced with 2% platelets increased the G-ratio by up to 73% for Inconel and 39% for titanium alloys over flood cooling. Higher G-ratios corresponding to lower wheel wear rates lead to a reduction in the wheel-dressing frequency and a higher part output.
Researchers from the Manufacturing Systems Research Laboratory, General Motors R&D (Warren, MI), presented a paper titled Development of Gun-Drilling MQL Process and Tooling for Machining of Compacted Graphite Iron (CGI). In the paper, designated TP10PUB9, it's noted that CGI is a type of cast iron with a graphite morphology lying somewhere between ordinary grey cast iron and nodular ductile cast iron. It is well-known that CGI offers great advantages in terms of performance, weight, cost reduction, and quality improvement for automotive applications. Unfortunately, tool life has been significantly lower in CGI, and especially in deep holes using standard gun-drill geometries.
Research at GM's Manufacturing Systems Research Laboratory looked at the machinability of CGI using Minimum Quantity Lubrication (MQL). Tool wear and wear mechanisms were analyzed, together with hole quality and part temperature. This research study's aim was the development of a highproductivity gun-drilling solution using carbide tooling, to approach the processing time of conventional cast iron. Gun-drilling was used to produce oil-gallery holes. Research was done with 12-mm-diam singleflute carbide-tipped gun drills with internal coolant at 55 bar on both CGI and CI material. Cutting speed was 80 m/min, and chip load was 0.1 mm/rev/tooth.
Temperature measurements made using an IR camera demonstrated that the MQL process provided no benefits for deep-hole drilling for large-diameter holes. Actually, drill life with high-pressure internal coolant was better than that achieved using MQL. Drill point temperature was significantly higher, and resulted in delamination and abrasive wear before the drills used with internal coolant exhibited such wear.
Three drill designs were studied and all failed by the same mechanisms; all were acceptable in terms of flank wear. One test involved drilling 400-mm-long holes. Holes were drilled from both sides of the workpiece, so feed length per side was about 200 mm (L:D ratio was about 16). Hole quality was very good due to small diameter variation and small mismatch of the two sections drilled from opposite sides. Cutting speed of 60.80 m/min was found optimum for coated carbide drills. A more stable oil holes at the margins will help reduce margin wear.
Researchers at the University of Illinois (Urbana) and TechSolve Inc. (Cincinnati) collaborated to prepare technical paper TP10PUB10, entitled Failure Mechanisms Encountered in Micro-Milling of Aligned Carbon Fiber Reinforced Polymers. While carbon fiber reinforced polymers (CFRPs) have many applications, the fiber-failure mechanisms that occur during machining at the microscale are not understood.
The failure model proposed by the research team suggests that the failure mechanisms vary as a function of fiber orientation relative to the direction of tool cuttingedge motion. Micromilling slotting tests were conducted to validate the model. The tests were done on plates of layered, resin-infused carbon fibers with 60% fiber by volume. Each was approximately 180 µm thick, and were cut from a large 3-mm-thick composite panel into 10 x 10-mm samples to fit on the machining test bed. A 396-µm diam, single-fluted end mill with an edge radius of 1–2 µm was chosen for the study. A single-fluted end mill was chosen because it allows precise control of chip load and eliminates effects due to tool runout.
Slotting experiments were done with an axial DOC of 80 µm and feed per tooth values of 2, 4, 5, 6, 7, 8, 10, and 18 µm. Four different fiber orientations relative to feed direction were examined. For every cutting condition, a 10-mm-long slot was machined. A new tool was used for every cutting condition to ensure that the machining responses were not confounded with tool wear effects.
In the model developed by researchers, it was proposed that while machining CFRP composites at the microscale, carbon fibers oriented at 90 and 45° to the direction of motion fail mostly in crushing/compression, while buckling dominates fibers in the 0° orientation and bending the 135° orientation.
Chip morphology validates the proposed model. Chips generated at the 45 and 90° orientation show small, fragmented chips that indicate compressive failure, while chips collected for the 0 and 135° orientations have fibers longer than the feed-per-tooth, indicating buckling and bending (tensile) failures. Observed delamination patterns support the proposed failure modes. Buckling failure at 0° orientation results in what researchers call negligible delamination, while the bending failure at the 135° orientation results in the highest positive delamination (failure due to entire sections of fibers being pulled from the matrix, along the top edge of the slot). Both 45 and 90° orientations show low positive delamination because the failure mode was crushing (compression).
Cutting forces were found to be 40% higher for the crushing-dominated failure as compared to bending or bucking failures. In the bending mode, the tool induces a moment that aids fiber failure, thus reducing cutting force.
Friction stir welding (FSW) is flexible and simple. Invented in 1991 at The Welding Institute (Cambridge, UK), FSW was initially applied to light metals like aluminum alloys; there are a few real-world examples of its use in welding steel materials. Researchers from Shinshu University (Nagano, Japan) and Chiba Institute of Technology (Narashino, Japan) say there are no extant examples of the use of FSW to weld cold-worked die steel. In their paper Effect of Tool Rotation on the Joint Strength of Cold-work Die Steel by Friction Stir Welding (TP10PUB12), the researchers describe FSW on die steel used for molds.
A FSW tool made from a WC-based material (K01) was used to butt-weld 10-mm-thick die steel plates at a constant welding speed of 1.0 mm/min, and at tool rotational speeds ranging from 250 to 1350 rpm. The specimens used were 40 x 30 x 10-mm blocks of SKD11 annealed cold-worked die steel. The micro Vickers hardness (HV) of the material was HV238. The FSW tool had a 10.0-mm shoulder diam, 5.0-mm probe diameter with a left-hand thread, and a 1.5-mm-long probe.
At a tool speed of 720 rpm, smooth, consistent semicircular patterns were formed. At a tool speed of 1350 rpm, however, cracking and spalling occurred, leading to a poorquality weld. The tool-speed range in which good quality semicircular weld patters were formed was from 720 to 935 rpm. At 250 rpm, FSW was impossible, because the tool broke. At tool speeds of 375, 540, and 1350 rpm, cracking and spalling was observed in the joints.
After welding, the appearance of the samples was observed with an optical microscope. For the metallographic analysis, micro Vickers Hardness tests, and tensile tests, the joint was cross-sectioned through a plane perpendicular to the welding direction.
For a welding speed of 1 mm/min and rotational speeds of 720 and 935 rpm, a joint depth of about 2 mm was achieved with good overall joint quality. The FSW temperature was about 1273°K, equivalent to the quenching temperature of SKD11. Heat penetration and a consequent state of quenching can be seen around the joints. The weld surface was discolored black because shielding was not employed. For the 935-rpm sample, the average grain size in the stir zone (SZ) was about 3 µm, and in the heat-affected zone (HAZ) was about 5 µm.
Also, the SZ and HAZ in all conditions yielded hardness values in excess of the quenching hardness of SKD11. For the 720 and 935-rpm specimens, breaking stress was comparable with annealed SKD 11.
Breaking stress proved equivalent to that of the base material, with fracture occurring between the SZ and base material. Therefore it can be concluded that the fracture strength of the material in the weld is higher than that of the base material. Strain for the 935-rpm specimen was smaller than that for the 720-rpm specimen, because the increased tool rotational speed led to increased welding heat input, and thus increased brittleness. Because both strains are less than 0.01, both samples are regarded as brittle.
All of the 98 papers presented at NAMRC 38 (by researchers from 15 countries) are available in the 800-page hardcover Transactions of NAMRI/SME, Vol. 38, 2010, (SME order code BK10PUB1). SME members can purchase this volume for $100, nonmembers for $130. For more information or to place an order, telephone SME Customer Service at 800.733.4763, 8 am.5 pm Eastern Time, Monday.Friday, or go to www.sme.org/store, and follow the prompts. Papers mentioned in this article have been made available as Technical Papers, which are free to all members. Technical papers can be purchased by nonmembers for $15 each. To find the papers cited above, search by title or by number.
This article was first published in the July 2010 edition of Manufacturing Engineering magazine.