Processing Effect

Processing is extremely important in regard to tolerance control; in certain cases it is the most influential factor. The dimensional accuracy of the finished part relates to the process, the accuracy of mold or die used, and the process controls, as well as the shrinkage behavior of the plastic. A change to a mold or die dimension can result in wear arising during production runs and should thus be considered.

The mold or die should also be recognized as one of the most important pieces of producdon equipment in the plant. This controllable, complex device must be an efficient heat exchanger and provide the part's shape. The mold or die designer thus has to have the experience or training and knowledge of how to produce the tooling needed for the part and to meet required tolerances.

A knowledge of processing methods can be useful to the designer, to help determine what tolerances can be obtained. With such high-pressure methods as injecdon and compression molding that use 2,000 to 30,000 psi (13.8 to 206.9 MPa) it is possible to develop dghter tolerances, but there is also a tendency to develop undesirable stresses (that is, orientadons, etc.) in different direcdons. An example as to how tolerances can change using the same process control and injecdon mold can be related to the amount of plasdcs used to fill and pack the mold cavity.

The low-pressure processes, including contact and casting with no pressure, usually do not permit meedng dght tolerances. There are excepdons, such as certain RPs that are processed with little or no pressures. Regardless of the process used, exercising the required and proper control over it will maximize obtaining and repeadng of close tolerances.

For example, certain injection-molded parts can be molded to extremely close tolerances of less than a thousandth of an inch, or down to 0.0%, particularly when filled TPs or filled TS compounds are used. To practically eliminate shrinkage and provide a smooth surface, one should consider using a small amount of a chemical blowing agent (<0.5wt%) and a regular packing procedure. Results are such that literally no change in density will occur (no visible bubbles, etc.) and the product surface will be at least as smooth if not improved. For conventional molding, tolerances can be met of ±5% for a part 0.020 in. thick, ±1% for 0.050 in., ±0.5% for 1.000 in., -0.25% for 5.000 in., and so on. Thermosets generally are more suitable than TPs for meeting the tightest tolerances.

The dimensions of the product design have to be converted into the dimensions of the mold cavity, taking the respective shrinkage into consideration. For that reason, the final decision of what kind of plastic will be used has to be made beforehand. Very often, the mold maker requires half of the tolerances permissible for the part in question for the job. This, in most cases, is not justified. Today's accuracy of metalworking permits tolerances as low as one tenth of those on the drawing, assuming the latter being reasonable for plasdcs.

Under difficult circumstances, it has been proved to be a good pracdce if cridcal dimensions are kept smaller in the mold first, and then being revised after a test run under producdon condidons thus permitting machining the cavity if required. In any case, close tolerances should be applied on such dimensions only as direcdy related to invariable mold dimensions. Any other dimension, which is related to a mold dimension in two different mold parts, should allow a generous tolerance.

Economical production requires that tolerances not be specified tighter than necessary. However, after a production target is met, one should mold "tighter" if possible, for greater profit by using less material. Many plastics change dimensions after molding, principally because their molecular orientations or molecules are not relaxed. To ease or eliminate the problem, one can change the processing cycle so that the plastic is "stress relieved," even though that may extend the cycle time. Also used is heat-treat, the molded part based on experience or according to the resin supplier's suggestions.

Theoretical efforts to forecast linear shrinkage have been limited because of the number of existing variables. One way to solve this problem is to simplify the mathematical relationship, leading to an estimated but still acceptable assessment. This means, however, that the number of necessary processing changes will also be reduced.

The parameters of the injection process must be provided. They can either be estimated or, to be more exact, taken from the thermal and rheological layout. The position of a length with respect to flow direction is in practice an important influence. This is so primarily for glass-filled material but also for unfilled thermoplastics. The difference between a length parallel to (0°) and perpendicular to (90°) the flow direction depends on the processing parameters. Measurements with unfilled PP and ABS have shown that a linear relationship exists between these points.

Regarding this relationship, when designing the mold it is necessary to know the flow direction. To obtain this information, a simple flow pattern construction can be used. However, the flow direction is not constant. In some cases the flow direction in the filling phase differs from that in the holding phase. Here the question arises of whether this must be considered using superposition.

In order to get the flow direction at the end of the filling phase and the beginning of the holding phase (representing the onset of shrinkage), an analogous model was developed that provides the flow direction at the end of the filling phase. For a flow with a Reynolds number less than 10, which is valid regarding the processing of thermoplastics, the following equadon can be used: A& = 0. For a two-dimensional geometry with quasistadonary conditions, this equation is valid:

Instead of the potential 0, it is possible to introduce the flow-stream function \ff for a two-dimensional flow. The stream lines (= constant) and the equipotential lines are perpendicular to each other. To express this, the following Cauchy-Rieman differential can be used:

A differential (two dimensional/quasi) equation has the same form as is used for a stationary electrical potential field,

as it can be realized with an unmanded molding out of resistance paper and a suitable voltage.

To control the theoretically determined flow with respect to the orientation direction, a color study was made. The comparison between flow pattern, color study, and analogous model is shown in Figs. 6.8 and 6.9. For a simple geometry the flow pattern method describes the flow direction in the filling phase as well as the holding phase (Fig. 6.8).

This description changes when a core is added and the flow is disturbed (Fig. 6.9). In this case the flow at the beginning of the holding phase differs from the flow pattern as it is shown in the color study as well as in the analogous model. Even the welding lines are broken in the holding phase so that at this place another flow direction than that in the filling phase is found. With further measurements this influence has to be tested by using more-complcx moldings. Available are computer software programs that provide guide lines to melt flow behavior in the mold cavities (Chapter 5).

0 0

Post a comment