David Kazmer

New Directions in Design for Manufacturing

This paper gives an overview of research that is expanding the domain of design for manufacturing (DFM) into new and important areas. This paper covers DFM and concurrent engineering, DFM for conceptual design, DFM for embodiment design, DFM for detailed design, design for production, platform design for reducing time-to-market, design for system quality, design for life cycle costs, and design for environment. The paper concludes with some general guidelines that suggest how manufacturing firms can develop useful, effective DFM tools.

A Self-Energized Sensor for Wireless Injection Mold Cavity Pressure Measurement: Design and Evaluation

This paper presents the modeling, design, and experimental validation of a self-energizedsensor system for pressure measurement in the injection mold cavity using ultrasound asthe information carrier. The sensor extracts energy from the polymer melt pressure anddiscretizes the pressure information into ultrasonic pulses for wireless transmissionthrough the mold to a remote receiver. Analytical and numerical models are presented forthree constituent components of the sensor: the energy converter, the threshold modulator,and the signal transmitter. Quantitative results were obtained to guide the parametricdesign of each constituent component. Simulations and experimental studies have vali-dated the functionality of each individual component, as well as the sensor as an inte-grated unit. In addition to the injection mold pressure measurement, the sensing techniquedeveloped is applicable in a broad range of process monitoring applications where highpressure fluctuations occur. @DOI: 10.1115/1.1767850#

An Economic Design Change Method

New product design as well as design revision to remedy defects is complicated by an inability to precisely predict product performance. Designers often find that they are confident about the performance of some design alternatives and uncertain about others. Similarly, alternative design changes may differ substantially in uncertainty, potential impact, and cost. This paper describes a method for including the effects of uncertainty in the evaluation of economic benefits of various design change options. The results indicate that the most profitable change option sequence depends not only on relative costs but also on the relative degree of uncertainty and on the magnitude of the potential design defects. The method demonstrates how design change alternatives can be compared using the engineering design of a beam. Finally, the validity of some common engineering change heuristics are discussed relative to their associated, quantitatively determined limits.

A Comparison of Statistical Process Control (SPC) and On-Line Multivariate Analyses (MVA) for Injection Molding

Abstract Manufacturing process automation is often impeded by limitations related to automatic quality assurance. Many plastics manufacturers use univariate statistical process control (SPC) for quality control by charting the critical process states relative to defined control limits. Alternatively, principal component analysis (PCA) and projection to latent stuctures (PLS) are multivariate methods that measure the process variance by the distance to the model (DModX) and the Hotelling t-squared (T2) values. A methodology for robust model development is described to perturb the manufacturing process for process characterization based on a design of experiments; best subset analysis is used to provide an optimal set of regressors for univariate SPC. Four different statistical models were derived from the same data set for a highly instrumented injection molding process. The performance of these models was then assessed with respect to fault diagnosis and defect identification when the molding process was subjected to twelve common process faults. Across two hundred molding cycles, the univariate SPC models correctly diagnosed five of the twelve process faults with one false positive, detecting only eighteen of twenty four defective products while indicating two false positives. With the same molding cycles, PCA and PLS provided nearly identical performance by correctly diagnosing ten of the twelve process faults and detecting twenty three of the twenty four defective products; PCA indicated two false positives while PLS indicated only one false positive.

Incorporation of Phenomenological Models in a Hybrid Neural Network for Quality Control of Injection Molding

Abstract Injection molding is characterized by complex dynamics, which makes quality difficult to control. This is because the exact relations among the machine inputs, material properties, and molded part quality are not known precisely. Hence, the existing models for quality prediction have a limited accuracy and difficulty in application to general molding applications. This article investigates the integration of analytical process knowledge and artificial neural networks as a solution for quality prediction of molded parts, with accuracy increased toward quality control targets of three defects per million (60). This article describes the hybrid system based on the neural network and process knowledge, then compares its performances with conventional neural models for the prediction of the injection pressure.

Feasibility Analysis of an In-mold Multivariate Sensor

Abstract The initial design of a novel multivariate sensor is described for the measurement of melt temperature, melt pressure, melt velocity, melt viscosity, and mold temperature. Melt pressure and temperature are respectively obtained through the incorporation of a piezoceramic element and infrared photodetector within the sensor head. Melt velocity is derived from the initial response of the melt temperature as the polymer melt flows across the sensors lens. The apparent melt viscosity is then derived from the melt velocity and the time derivative of the increasing melt pressure given the cavity thickness. The feasibility of the envisioned sensor is then analyzed using a production-grade mold that is instrumented with commercial piezoelectric pressure sensors, infrared pyrometer, and thermocouples. Several predictive models of part weight are developed using multiple regression of data obtained from a design of experiments to evaluate the capability of the envisioned multivariate sensor. The results indicate a correlation coefficient, R2, of 0.79 for a model based on the machine settings, 0.80 for a model based on a cavity pressure sensor, 0.90 for a model based on the multivariate sensor, and 0.98 for a non-linear model based on the multivariate sensor. The implication is that multiple orthogonal sensing streams with high fidelity models are necessary to provide automatic quality assurance sufficient for fully automated plastics manufacturing.

Design of ultrasonic transmitters with defined frequency characteristics for wireless pressure sensing in injection molding

This paper describes a new mechanical wireless data transmission technique using ultrasonic waves as the information carrier for on-line injection mold cavity pressure measurement. Ultrasonic transmitters with specific frequency characteristics were designed, modeled, simulated, and prototyped for pressure data retrieval from an enclosed machine environment, as well as for sensor identification in a sensor matrix configuration. The effects of the front layer and bonding layer of the transmitter on the overall sensor frequency characteristics were investigated, using an equivalent circuit model. The optimal layer thickness was determined for the design of transmitters with specific dominant resonant frequency and narrow bandwidth. Experimental results were in good agreement with the analysis, thus confirming the design approach.

Numerical simulation of phase separation of immiscible polymer blends on a heterogeneously functionalized substrate

The spinodal phase decomposition of an immiscible binary polymer blend system is investigated with numerical models in two-dimensional and three-dimensional (3D). The effect of the elastic energy is included. The mechanism of the evolution of the phase separation is studied and the characteristic length R(t) is shown to be proportional to t(13). In the case when the phase separation is directed by a heterogeneously functionalized substrate, the increase in the characteristic length is divided into two stages by a critical time. The R(t) approximately t(13) diagram can be fitted with a straight line in both the first and second stages. The slope of the fitting line significantly decreases after the critical time. The compatibility of the resulting pattern to the substrate pattern is also measured by a factor C(S). It is observed that there is also a critical time in the evolution of the compatibility for the cases with and without elastic energy. The critical time of C(S) is identical with the respective critical time of R(t). The lateral and vertical composition profiles functionalized substrate is observed with the 3D model. The difference mechanism of the cases with and without elastic energy is discussed.

In-Situ Shrinkage Sensor for Injection Molding

Dimensional consistency is a critical attribute for injection molded part quality and is highly dependent on the polymer morphology, the thermal expansion, and various processing parameters. The dimensional shrinkage can be estimated by knowing the pressure-volume-temperature behavior of the polymer but with limited accuracy. There are various process monitoring systems available in the market; none of which has the capability of directly monitoring and controlling the real time shrinkage and part dimensions online. With a view to measuring in-mold shrinkage, a button cell type in-mold shrinkage sensor was developed, validated, and compared against the traditional shrinkage prediction and estimation methods. The shrinkage sensor consists of an elastic diaphragm instrumented with strain gages connected in a full bridge circuit. The sensor is placed beneath the movable pin that is protruded into the mold cavity and remains in contact with the sensor diaphragm. The sensor diaphragm is deflected due to the melt pressure acting on the pin into the mold cavity and is retracted back toward its original position as the melt solidifies and shrinks away from the mold cavity wall. The sensor signals acquired during each molding cycle were analyzed to validate the sensor performance in a design of experiments as a function of packing pressure, melt temperature, cooling time, and coolant temperature. The regression results indicate that the shrinkage sensor outpeforms cavity pressure transducers and other methods of predicting the in-mold shrinkage. For polypropylene, the shrinkage sensor is able to measure the shrinkage to an average accuracy of 0.01 mm for a molded part with a nominal thickness of 2.5 mm. The coefficient of correlation, R 2 , between the sensors final positions to the final part thickness was 0.921 for the in-mold shrinkage sensor. Other dimension prediction methods had lower correlation coefficients.

A Performance-Based Representation for Engineering Design

A design representation is developed to model multi-attribute systems utilizing multidimensional clipping and transformation algorithms. Given a linear system characterization, three types of supporting information is generated for the decision maker: (1) a function matrix that describes the performance attributes dependent upon the decision variables; (2) a decision space that corresponds to the feasible decision set that meets performance requirements, and; (3) a performance space that represents the feasible performance region and the Pareto Optimal set. The analytical method developed for solving these feasible spaces is described for a linear system model. A case study is presented to demonstrate how to utilize the representation to locate a feasible solution and proceed to the desired trade-off of multiple attributes. Moreover, the potential incorporations of the representation with other influential design methodologies are discussed.