Printer Friendly

Computer modeling predicts tolerances of molded parts.

The growing importance of plastics moldings in complex mechanical assemblies for electronics, medical devices, and other high precision applications focuses increasing attention on requirements for dimensional control. Dimensional tolerances often are the crucial specification for the quality of molded parts.

In predicting part dimensions and their fluctuations (tolerances), one must consider many variables, including the resin properties and resin variability, the geometry of the part, the toolmaking quality applied in building the mold, and above all, the molding conditions and process fluctuations inherent in the equipment. Computer programs developed over the past ten years have made it possible to model the complex interactions among these many factors, thus allowing molders to accurately predict part dimensions and to quantify the relationship between control of the molding process and part tolerances. This article briefly reviews the causes of dimensional variability of molded parts, then presents the software requirements for analysis of part tolerances, and concludes with a practical example that illustrates the use of an industry proven software package.

The Origins of Dimensional Variations

If there were no shrinkage of the resin in the mold, every part would have the exact dimensions of the cavity, and dimensionally perfect parts without any tolerances would be molded. It follows that by selecting a resin with the lowest possible shrinkage and molding conditions that minimize shrinkage, a molder will produce parts with the tightest dimensional tolerances. (Obviously, this logic ignores the fact that some of the conditions favoring low shrinkage, such as a cold mold, may cause other flaws in part properties, such as high internal stresses.)

It also follows that dimensionally perfect parts without any tolerances could be produced if shrinkage from shot to shot were absolutely constant. This raises the question: What causes shot-to-shot variability in shrinkage?

If the resin is assumed to be consistent and uniform, and the molding conditions remain unchanged, then changes in shrinkage in a given mold must be attributed to inherent variability in the molding process in terms of injection (holding) pressure, melt temperature, mold temperature, and temperature variability at the inspection site where the parts are measured. This emphasizes the direct linkage between part dimensional tolerances and equipment performance and process control.

Software Requirements

The foregoing discussion indicates that an industrially useful software package for predicting dimensional tolerances in molded parts must satisfy the following criteria:

* Based on the resin and the geometry of the part, the software must define a range of processing conditions within which parts of acceptable quality can be molded.

* Within the defined range of molding conditions, and considering the inherent variability of the equipment employed, the software must select those molding conditions that will produce parts within the specified tolerances.

To meet these requirements, the software must define the relationship between the key processing variables and part shrinkage (for a description of software to analyze shrinkage for cavity dimensioning, see E.C. Bernhardt, ANTEC'89), and the software must consider the influence of part geometry, gating, and orientation with respect to flow. It is evident that this task demands a knowledge base of great size and sophistication.

To complete the picture, toolmaking tolerances must also be included. The relationship between part dimensional tolerances, shrinkage variation caused by process variability, and toolmaking tolerances is summarized in Fig. 1.

It is important to note that some parts may not be moldable and some tolerances may not be attainable within the above constraints. The software must flag such situations. Also, the specification of molding conditions for critical parts may need to be further refined, with the use of advanced finite-element flow-analysis programs that are able to integrate with the software described in this article (E.C. Bernhardt, et al., International Polymer Processing IV, 1989).

Practical Example

Developed over the last fifteen years, the TMconcept[TM] Molding & Cost Optimization (MCO) software package is designed as a practical working tool for any engineer with responsibility for a molding project. Its underlying concepts, based on multivariate regression analysis, were described by E.C. Bernhardt and G. Bertacchi at ANTEC'88. A detailed description of the software operation is beyond the scope of this article, but some of the key inputs and outputs are here illustrated, and the means by which it links part dimensional tolerances with processing conditions and process variability is demonstrated.

The part to be molded is the rear panel of a 5.25-inch cassette box, shown in Fig. 2. Two of the dimensions with tolerances are specified on the drawing; the material to be used is general-purpose polystyrene.

The variables that affect the attainable tolerances are listed in the first section of the Box. At the start of the analysis, this information is called up automatically from the two databases associated with the software:

The Materials Database contains data on the processing characteristics, including shrinkage values parallel (P) and transverse (T) to flow as well as shrinkage correlation errors, for a very broad range of commercial resins.

* The Plant Database includes detailed information on the variability of the processing equipment.

The software carries out the analysis on two levels:

1. Assuming ideal gate dimensions and perfect packing of all sections of the mold, the First Approximation (MC01) very rapidly computes the attainable tolerances for any given dimension, taking full account of the material and plant variables defined in the Box. These computed tolerances provide immediate information as to whether, and how easily, the specified tolerances can be met. The output can also be presented in a particularly handy graphic format (Fig. 3).

If the specified tolerances are significantly wider than those attainable, there is no need for concern. Conversely, attention is called at once to dimensions with specified tolerances that are too tight to be met under the indicated conditions. The software then lets the analyst evaluate what degree of effort, through tighter process control or better quality toolmaking, is required to meet these tolerances, or whether they are unrealistically tight.

2. When the specified tolerances must be met, Compute-Molding Conditions (MC02) determines the best molding conditions to produce parts of that quality. Required inputs include not only the part geometry, the critical dimensions with their tolerances, and the section in which the dimension is located, but also the precise location of the dimension with respect to the gate, its orientation with respect to resin flow, and all flow restrictions between the gate and the dimension that may impede packing of that section. Other key inputs that have a direct bearing on the degree of shrinkage include gate dimensions, holding pressure, and mold temperature.

The final output from the program (see the Box) provides a comprehensive record of all inputs and outputs with a clear definition of the molding cycle required to meet the specified dimensions. The Tolerance Classification indicates the relative ease with which the specified tolerances Fig. 3) can be maintained. For wide tolerances that are easy to mold (coarse molding), the user may consider widening the toolmaking tolerances to save on mold costs without impairing part quality.

The computer model described here is a powerful and reliable tool to define the influence of part design, molding conditions, and process variability on the ability to maintain tolerances. It allows the molder to accurately predict the molding conditions required to meet part specifications and to quote jobs with greater confidence.

(Tabular Data and Other Figures Omitted)
COPYRIGHT 1990 Society of Plastics Engineers, Inc.
No portion of this article can be reproduced without the express written permission from the copyright holder.
Copyright 1990 Gale, Cengage Learning. All rights reserved.

Article Details
Printer friendly Cite/link Email Feedback
Title Annotation:plastics design
Author:Bernhardt, Ernest C.
Publication:Plastics Engineering
Date:Oct 1, 1990
Previous Article:Recycling: implications of post-consumer plastic waste.
Next Article:Producing thermoplastic matrix sheet composites: two processes.

Terms of use | Copyright © 2016 Farlex, Inc. | Feedback | For webmasters