Newest Issue

Research Papers

J. Manuf. Sci. Eng. 2019;141(3):031001-031001-15. doi:10.1115/1.4041949.

Short tool service life is always a major concern when milling hard materials, such as Ni-based superalloy. In the current research of tool life optimization in multi-axis machining of freeform surfaces, the focus is mostly on choosing suitable cutting parameters and better application of coolant. In this paper, aiming at averaging the tool wear on the entire cutting edge and hence prolonging the tool service life, we report a study on how to generate a multilayer toolpath with a varying tool lead angle for multi-axis milling of an arbitrary freeform surface from an initial raw stock. The generated toolpath is guaranteed to be free of chatter, which is well known for its detrimental effect on the cutting edge. In this study, we first experimentally construct the chatter stability lobe diagram, which reveals the relationship between the lead angle and the cutting depth. With the chatter stability lobe diagram as the major constraint, we then generate the machining toolpath by selecting a proper pair of the best lead angle and cutting depth along the toolpath. While the proposed algorithm currently is restricted to the iso-planar type of toolpath, it can be adapted to other types of milling. The physical cutting experiments performed by us have convincingly confirmed the advantage of the proposed machining strategy as compared to the conventional constant lead angle and constant cutting depth strategy—in our tests the maximum wear on the cutting edge is reduced by as much as 39%.

Commentary by Dr. Valentin Fuster
J. Manuf. Sci. Eng. 2019;141(3):031002-031002-13. doi:10.1115/1.4042108.

A growing research trend in additive manufacturing (AM) calls for layerwise anomaly detection as a step toward enabling real-time process control, in contrast to ex situ or postprocess testing and characterization. We propose a method for layerwise anomaly detection during laser powder-bed fusion (L-PBF) metal AM. The method uses high-speed thermal imaging to capture melt pool temperature and is composed of the following four-step anomaly detection procedure: (1) using the captured thermal images, a process signature of a just-fabricated layer is generated. Next, a signature difference is obtained by subtracting the process signature of that particular layer from a prespecified reference signature, (2) a screening step selects potential regions of interests (ROIs) within the layer that are likely to contain process anomalies, hence reducing the computational burden associated with analyzing the full layer data, (3) the spatial dependence of these ROIs is modeled using a Gaussian process model, and then pixels with statistically significant deviations are flagged, and (4) using the quantity and the spatial pattern of the flagged pixels as predictors, a classifier is trained and implemented to determine whether the process is in- or out-of-control. We validate the proposed method using a case study on a commercial L-PBF system custom-instrumented with a dual-wavelength imaging pyrometer for capturing the thermal images during fabrication.

Commentary by Dr. Valentin Fuster
J. Manuf. Sci. Eng. 2019;141(3):031003-031003-14. doi:10.1115/1.4042051.

Chemical mechanical planarization (CMP) has been widely used in the semiconductor industry to create planar surfaces with a combination of chemical and mechanical forces. A CMP process is very complex because several chemical and mechanical phenomena (e.g., surface kinetics, electrochemical interfaces, contact mechanics, stress mechanics, hydrodynamics, and tribochemistry) are involved. Predicting the material removal rate (MRR) in a CMP process with sufficient accuracy is essential to achieving uniform surface finish. While physics-based methods have been introduced to predict MRRs, little research has been reported on monitoring and predictive modeling of the MRR in CMP. This paper presents a novel decision tree-based ensemble learning algorithm that can train the predictive model of the MRR. The stacking technique is used to combine three decision tree-based learning algorithms, including the random forests (RF), gradient boosting trees (GBT), and extremely randomized trees (ERT), via a meta-regressor. The proposed method is demonstrated on the data collected from a CMP tool that removes material from the surface of wafers. Experimental results have shown that the decision tree-based ensemble learning algorithm using stacking can predict the MRR in the CMP process with very high accuracy.

Commentary by Dr. Valentin Fuster
J. Manuf. Sci. Eng. 2019;141(3):031004-031004-9. doi:10.1115/1.4042019.

Tool-workpiece deflection is one of the major error sources in machining thin walled structures like blades. The traditional approach in industry to eliminate this error is based on modifying tool positions after measuring the error on the machined part. This paper presents an integrated model of cutting force distribution on the tool–blade contact, automatic update of blade static stiffness matrix without resorting to time-consuming finite element solutions as the material is removed, the prediction and compensation of static deflection marks left on the blade surface. The main focus of the paper is to compensate the deflection errors by respecting the maximum form errors, collision of tool/machine/workpiece, cutting speed limit at the tool tip, and ball end—blade surface contact constraints. The compensation has been carried out by two modules. The first module adjusts the tool orientation along the path to reduce the error by constructing an optimization problem. This module is computationally inexpensive and results in about 70% error reduction based on the conducted experiments. The modified tool path resulted from the first module is fed to the second module for further reduction of the form errors if needed at the violated cutter locations; hence it takes less computational time than the stand alone approach proposed in the literature. The proposed algorithms have been experimentally validated on five-axis finish ball end milling of blades with about 80% reduction in cutting force induced form errors.

Commentary by Dr. Valentin Fuster
J. Manuf. Sci. Eng. 2019;141(3):031005-031005-9. doi:10.1115/1.4042110.

Micropencil grinding tools (MPGTs) are micromachining tools that use superabrasives like diamond and cubic boron nitride (cBN) grits to manufacture complex microstructures in a broad range of hard and brittle materials. MPGTs suffer from a rather low tool life, when compared to other more established microprocessing methods. It was documented that when used on hardened steel workpieces, MPGTs suffer from a large amount of adhesions, mostly located at the pivot point of the tool. These adhesions lead to the clogging of the abrasive layer and ultimately in tool failure. Another problem this machining process suffers from is the formation of substructures (smaller channels inside the microchannels). The pivot is usually less prone to abrasive wear, has higher protrusion, and is therefore responsible for the deepest substructures. These substructures can easily take up half the depth of cut, obstructing the function of machined microchannels—it is one of the major flaws of this micromachining process. A micro-electrical discharge machining method (μEDM) can solve these issues by manufacturing a cavity at the pivot of these tools. A novel method that uses measurement probes to position the substrate above the μEDM electrode is implemented and a parameter study to determine the cavity manufacturing parameters is conducted for substrates with diameters < 40 μm. The goal is to demonstrate the first ever complete and reliable manufacturing process for MPGTs with a cavity and to demonstrate the advantages they provide in a machining process when compared to regular MPGTs.

Commentary by Dr. Valentin Fuster
J. Manuf. Sci. Eng. 2019;141(3):031006-031006-14. doi:10.1115/1.4041834.

The honeycomb sandwich structure has been widely used in the aerospace industry due to its high specific strength and stiffness. However, the machining defects of the aluminum honeycomb core (AHC) have become the key factor that restricts its application. In this paper, the defects' characteristics including the formation mechanism, distribution characteristic, and cutting process of honeycomb cell walls during AHC milling process were experimentally investigated. Furthermore, using normalized Cockcroft and Latham ductile fracture criterion and Johnson–Cook (JC) constitutive model, the numerical simulation of the AHC machining process was conducted concerning the entrance angle. It is indicated that six categories of milling defects are obtained and the quantity as well as distribution regularity of AHC milling defects are determined by the double effects of both the entrance angle and cutting force. Most of the surface defects of honeycomb materials were found concentrated in three regions, named by zones I–III, in which extruding, shear, and tensile deformation was mainly generated, respectively. Besides, the finite element simulation results also agree well with the experimental findings. Finally, a novel optimization method to avoid defects in the aforementioned regions by controlling the entrance angle of all the honeycomb walls during the cutting process was proposed in this paper. Meanwhile, the optimal control equations of the entrance angle for all cell walls were derived. This method was verified by milling experiments at last and the results showed that the optimization effect was obvious since the quality of the machined surface was greatly improved.

Commentary by Dr. Valentin Fuster
J. Manuf. Sci. Eng. 2019;141(3):031007-031007-9. doi:10.1115/1.4042053.

Cyber-manufacturing system (CMS) offers a blueprint for future manufacturing systems in which physical components are fully integrated with computational processes in a connected environment. Similar concepts and visions have been developed to different extents and under different names—“Industrie 4.0” in Germany, “Monozukuri” in Japan, “Factories of the Future” in the EU, and “Industrial Internet” by GE. However, CMS opens a door for cyber–physical attacks on manufacturing systems. Current computer and information security methods—firewalls and intrusion detection system (IDS), etc.—cannot detect the malicious attacks in CMS with adequate response time and accuracy. Realization of the promising CMS depends on addressing cyber–physical security issues effectively. These attacks can cause physical damages to physical components—machines, equipment, parts, assemblies, products—through over-wearing, breakage, scrap parts or other changes that designers did not intend. This research proposes a conceptual design of a system to detect cyber–physical intrusions in CMS. To accomplish this objective, physical data from the manufacturing process level and production system level are integrated with cyber data from network-based and host-based IDSs. The correlations between the cyber and physical data are analyzed. Machine learning methods are adapted to detect the intrusions. Three-dimensional (3D) printing and computer numerical control (CNC) milling process are used as examples of manufacturing processes for detecting cyber–physical attacks. A cyber–physical attack scenario is presented with preliminary results to illustrate how the system can be used.

Commentary by Dr. Valentin Fuster
J. Manuf. Sci. Eng. 2019;141(3):031008-031008-14. doi:10.1115/1.4042052.

Trochoidal (TR) tool paths have been a popular means in high-speed machining for slot cutting, owing to its unique way of cyclically advancing the tool to avoid the situation of a full tool engagement angle suffered by the conventional type of slot cutting. However, advantageous in lowering the tool engagement angle, they sacrifice in machining efficiency—to limit the tool engagement angle, the step distance has to be carefully controlled, thus resulting in a much longer total machining time. Toward the objective of improving the machining efficiency, in this paper, we propose a new type of TR tool path for milling an arbitrary curved slot. For our new type of TR tool path, within each TR cycle, rather than moving circularly, the tool moves in a particular way such that the material removal rate is maximized while the given maximum engagement angle is fully respected. While this type of TR tool path works perfectly only for circular slots (including straight ones), by means of an adaptive decomposition and then a novel iso-arc-length mapping scheme, it is successfully applied to any general arbitrarily curved slot. Our experiments have confirmed that, when compared with the conventional TR tool paths, the proposed new type of TR tool path is able to significantly reduce the total machining time by as much as 25%, without sacrificing the tool wear.

Commentary by Dr. Valentin Fuster
J. Manuf. Sci. Eng. 2019;141(3):031009-031009-14. doi:10.1115/1.4042188.

Face milling commonly generates surface quality of variation, is especially severe for milling of large-scale components with complex surface geometry such as cylinder block, engine head, and valve body. Thus surface variation serves as an important indicator both for machining parameter selection and components' service performance such as sealing, energy consumption, and emission. An efficient and comprehensive numerical model is highly desired for the prediction of surface variation of entire surface. This study proposes a coupled numerical simulation method, updating finite element (FE) model iteratively based on integration of data from abaqus and matlab, to predict surface variation induced by face milling of large-scale components with complex surfaces. Using the coupled model, three-dimensional (3D) variation of large-scale surface can be successfully simulated by considering face milling process including dynamic milling force, spiral curve of milling trajectory, and intermittently rotating contact characteristics. Surface variation is finally represented with point cloud from iterative FE analysis and verified by face milling experiment. Comparison between measured and predicted results shows that the new prediction method can simulate surface variation of complex components well. Based on the verified model, a set of analyses are conducted to evaluate the effects of local stiffness nonhomogenization and milling force variation on machined surface variation. It demonstrates that surface variation with surface peaks and concaves is strongly correlated with local stiffness nonhomogenization especially in feed direction. And thus the coupled prediction method provides a theoretical and efficient way to study surface variation induced by face milling of large-scale complex components.

Commentary by Dr. Valentin Fuster

Sorry! You do not have access to this content. For assistance or to subscribe, please contact us:

  • TELEPHONE: 1-800-843-2763 (Toll-free in the USA)
  • EMAIL: asmedigitalcollection@asme.org
Sign In