Search Results

Lean manufacturing has greatly changed the way many companies produce products. Lessons learned in lean manufacturing are being applied to many non-manufacturing situations including design, logistics, and transactional processes. Such tools can be applied to metallurgical laboratories which support daily production or design and development. Four key principles of lean management and nine tools are presented with specific examples demonstrating how these lean tools can be successfully applied to a metallurgical laboratory operations. These principles and tools have been successfully implemented by the author in a captive metallurgical laboratory over the last year resulting in reduced cycle time, higher productivity, less variation in analysis methods allowing for direct comparison of data over time, and improved quality.

Seals are commonly made by using o-rings in many liquid and gas filled system. Since the introduction of mini-concentric fuel pressure regulators mounted on fuel rails, issues have arisen related to assembly resulting in damage to the o-rings. The problem of damage o-rings can result in fuel leaks. The purpose of this study is to determine the optimum conditions in design/process to eliminate damage to the o-ring while maintaining sealability. A total of 14 variables were chosen to be examined in this study, 9 design related and 5 process related. To maximize the efficiency of testing, an orthogonal array was used to structure the experiment. After completing the study, assembly fixturing contributes 80% to successful assembly.

As many underhood products have shifted from metal to plastic designs over the last decade, often parts break during assembly due rough handling conditions and less than ideal manufacturing practices. In particular assemblies often crack during screw torquing at the mounting tabs. The goal of this study is to determine how different mounting tab designs compare in strength. Designs with an external rib on the perimeter of the mounting tab behave differently than designs without a complete or lacking an external rib on the perimeter of the mounting tab. A positive correlation was found between knit line cross-sectional area and mounting tab strength for designs with an external rib on the perimeter of the mounting tab. A positive correlation was found between tab thickness and mounting tab strength for designs lacking an external rib. Material type was found to impact mounting tab strength.

Often electrical components are encapsulated in a plastic material after assembly. The goal of this study is to determine what variables are most important in reducing potting variation and identify the key machine parameters which can be used to make adjustments to the potting process. To maximize the efficiency of testing, an L18 orthogonal array was used to structure an experiment. Hose temperature, orifice size, and pressure were found to be the most significant control factors studied in this experiment. Shifting from the initial settings for these factors to the recommended settings should increase the S/N of the potting process by 14.53db. Motor speed was found to be the most significant variable for adjusting the mean of the process. The noise factors induced in this study were found to be a significant source of variation. Filters can shift the mean potting material applied by 25% over their planned usage life. Moreover, new filters induce more variation than old filters.

Plastic underhood components often crack when fastened to mating components. The goal of this study is to determine what variables are most important with relation to this problem. To maximize the efficiency of testing, an L18 orthogonal array was used to structure the experiment. Experiments were conducted using a computer-controlled electric driven screwdriver and a manual torque wrench. Control factors were adjusted by using hardware as specified for each trial. The torque value at failure was recorded for all samples. The “larger the better” S/N equation was utilized for data analysis in this study. Optimizing the design and process was found to increase the S/N by 6.9dB when verified experimentally. Fastener washer thickness, joint fit-up, and screw setting speed contribute to the gain by 46%, 28%, and 27%, respectfully.

When commercializing squeeze casting and semi-solid metalworking processes, component producers looked to conventional die casting to identify potential defects and control component quality. Several defects were expected including cold shuts, cold flows, flash, drags, warping, and gas entrapment, just to name a few. Efforts were taken to avoid these defects. New defect types, however, have surfaced unique to these high integrity die casting processes. Contaminant veins and phase separation are presented. Although, squeeze casting and semi-solid metalworking have proven to be successful, component producers have been reluctant to report defects for fear of giving these emerging processes a bad reputation. Nonetheless, these defects must be understood to avoid future problems.

Many manufacturers utilize statistical methods to control product characteristics during processing. Often traditional x charts are utilized. This method is appropriate when examining characteristics with one degree of freedom. For features with two or more degrees of freedom, traditional x charts may be misleading. Techniques for controlling characteristics with multiple degrees of freedom, however, have not been developed. The purpose of this paper is to present a method of statistical analysis using radial type charting techniques for characteristics with multiple degrees of freedom. Included in this paper is a detailed discussion of the steps to prepare and interpret radial type charts. These techniques are applied to true position measurements as a specific example.

Process data curves are plots of process characteristics as a function of time over one cycle. Statistical analysis of these curves has not been feasible in the past due to lengthy calculations compounded by large data sets. This paper presents a statistical analysis method for process data curves using today's computers. This method involves collection of process data, descrete models creation, and the calculation of control limit curves. When plotting process data with control limit curves, variation can be identified and improvements can be made to the process. This technique of statistical process control can be applied to numerous process.