Printer Friendly

Gas injection molding of an automotive structural part.

Gas injection molding provides a solution to many problems associated with conventional hi h pressure injection molding and structural foam molding. It significantly reduces volume shrinkage, which causes sink marks on many injection molded parts. It also yields improved surface finish, which is not possible by conventional structural foam molding.

The technology and processing know-how are still in their infant stages, even though the initial attempts at gas injection molding were made in the 1970s. There are numerous gas equipment suppliers in Europe, Japan, and the U.S., but simply having gas equipment is meaningless unless part design, tool design, and processing knowledge are mastered.

Many commercial applications exist, but they represent only the tip of the iceberg of gas injection molding's potential. Once the process is widely developed and understood, many applications will arise that were previously either impossible by conventional injection molding or never seriously considered.

This article uses a case study of the successful development of a structural application to illustrate the fundamental principles and outline the limitations, benefits, and future applications of the gas injection process. Much of the information is necessarily proprietary, but we feel that enough details are presented to help readers attain a better understanding of the technology. Process Description The two basic types of processes are gas through nozzle (Fig. 1) and gas through runner or cavity (Fig. 2).

In both cases, the mold is partially filled with a "short shot" of plastic melt. The gas is introduced simultaneously with and/or subsequently to the plastic, or the plastic flow is completely stopped by a specially designed shutoff nozzle and a controlled volume of inert gas (usually nitrogen) is injected into the center of the flow. The combination of high melt surface tension and the lower viscosity of the hotter molten plastic in the center of thicker sections, such as ribs, confines the gas to form hollow areas in thicker sections of the part. The molten plastic that is displaced by the gas is pushed into the extremities of the tool, packing out the molded part (for details, see PE, July 1989, p. 35).

The outer surfaces of thicker sections do not sink because the gas has cored them out from the inside and the gas pressure holds the plastic against the mold surface while it solidifies. The sink in these sections takes place internally rather than on the exterior surfaces, eliminating sink marks on the part. Because the pressure used for final filling of the mold is confined to an area defined by the system of gas flow channels, the resultant force against the mold is relatively modest, and lower clamping forces are adequate.

The advantages of gas injection over conventional injection molding are: * Elimination of molded-in stresses, because of the low cavity pressure exerted by the gas; * Lower tool cost, and simplification in certain designs; * Reduced clamp tonnage/lower equipment cost; * Significant reduction of sink marks over ribs and bosses; * Improved surface finish; * Possible elimination of external runners; and * Part design flexibility, such as mixed thick and thin walls, box sections without movable cores, part consolidation, and larger, complex parts.

Limitations of the process include: * Longer development lead times, resulting from a poor understanding of part/tool design and process optimization for a variety of materials; * Difficulty in controlling multicavity (more than four) molds; * The injection molding machine should be capable of providing consistent, accurate shot weight; and * The part must have a vent hole on a nonvisible portion.

Each process, gas through nozzle and gas through cavity, has its own pros and cons. The former is good for symmetrical multicavity tools, but cannot be used with a hot runner system. The gas cannot be injected simultaneously, that is, plastic flow is stopped and then the gas is injected. However, existing molds can usually be used without much change for this mode, depending on part design.

Because hot runner systems can be used and simultaneous plastic/gas injection is possible, the gas through cavity or runner mode is the more versatile technique.

Part/Tool Design


Little has been published in this area. Empirically, simultaneous part, mold, and process design is understood to be important for the success of any gas-assisted injection molded part. A recently published paper (I. Baxi, SPI Structural Div. 1989 Conf., 18, 158), which describes design features for part, gate, and runner systems, is a good introduction. However, we believe that guidelines have yet to be established for different materials and for all geometries and wall thicknesses.

Some important considerations are: * The gas channel geometry is usually either symmetrical or unidirectional relative to the injection gate. * The balance for the plastic flow and gas flow from the gate is critical. Mold fill simulation (short shot) can be a very helpful tool for gate design and location. * A rule of thumb is that the width of a rib should be [is less than or equal to] 3x nominal wall thickness, and the depth (height) of a rib should be [equal to or greater than] 3x nominal wall thickness. * A gas flow channel must be continuous and should not loop back on itself. The plastic melt it displaced must have somewhere to go, and the material must be sufficient to pack out the mold. * Spillover space in the tool should be provided for fine tuning the flow balance to achieve the desired hollow channels.

Door Window Glass

Guidance Channels

The function of the guidance channels in an automobile door window is to guide the glass through up/down travel while withstanding the static and dynamic loads during the glass travel. The part requirements are high strength-to-weight ratio, dimensional stability, resistance to creep, stiffness, no warpage, and low wear.

Our intention was to manufacture this part by conventional injection molding using reinforced engineering thermoplastics. The finite element analysis based on our structural requirements suggested a part design having many ribs and very thick wall sections. We realized that conventional injection molding of this design would create problems such as significant warpage, dimensional instability, and a long cycle time. The alternative design was a tubular structure, which can only be made by gas injection molding.

The advantages of gas injection over conventional injection molding were listed above. Its advantages over metal stamping are reduced weight, design flexibility, corrosion resistance, and parts consolidation. Along with designing and demonstrating the feasibility of manufacturing the guidance channels by gas injection molding, the objectives of the project were to compare the commercially available technologies, to optimize the process and define the key parameters, and to evaluate the gas/plastic flow analysis software.

The material must withstand the required load conditions, have excellent wear resistance to meet window cycling requirements, and have a low coefficient of friction relative to the guide pin attached to the glass. Based on these criteria, a 30% glass/mineral-filled, impact-modified polyester (PET) was selected.

In gas injection molding, it is important that the part be designed both for performance and the process manufacturability). The results of a PATRAN finite element analysis simulation with a 2.5-mm minimum wall thickness for the channels indicated that the proposed part design should function, without failure, when subjected to the loads listed in Table 1. The "Factor of Safety" based on the Von Mises theory of failure, i.e., the ratio of stress to ultimate material strength, is included with the FEA results in Table 2. No attempt was made to factor in the wall thickness variation, inherent in the process, into the model. To ensure a conservative result, the minimum specified wall thickness was assumed throughout the length of the "U" channel.

Gas Through Nozzle


The balancing of plastic flow by part design and gate design and dimension is critical in gas injection molding, in order to achieve uniform wall thickness throughout the part. Because the gas was planned to be introduced at the end of the plastic filling, a mold flow analysis (Moldflow, Australia) was performed to aid in gate design and location and selection of proper processing parameters. Figure 3 is an Isochrone, that is, a plot of the pressure drop of the material during filling of the cavity on which the shadings indicate different short shots. In this instance, there is no separation of flow until the material fills the flange area. At this point, the material flows around the through-holes, and knit lines form on the opposite sides, indicating the need to locate vents in these areas. The results of the mold filling analysis (Table 3) show that the calculated temperature of the material at 95% fill is fairly uniform (12[deg.]C range), indicating that the shearing of material will not occur and the gas injection process should be able to run at relatively low pressure.

The amount of time for the material to reach the "no flow temperature" is indicated in Fig. 4. This information was used to determine the time delay possible before the gas could no longer move the plastic melt. Similarly, different short shots were plotted on the 95% fill and 80% fill profiles to gain understanding of flow patterns, temperature profiles, pressure distributions, etc.-information used to establish the processing parameters.

The objectives of the experiments were to achieve completely hollow cross sections with uniform wall thickness in both subchannels, and part-to-part consistency. The use of an injection molding machine with a microprocessor control system is important in gas injection molding. The most important variables are injection time, shot size, melt temperature, gas injection speed, gas pressure, and delay time between the end of plastic injection and introduction of the gas.

Figure 5 shows the gate location at the center of the two subchannels and the lack of symmetry between the ends of the part. This indicates that after the short shot of plastic fills the subchannels, the gas will flow in two directions through the center of the subchannels (least resistance) and push the plastic into the corners. The corner with the least volume will fill faster, and there will be extra melt with no place to go. The cross section of this channel will then not be completely hollow. This problem, which did indeed occur repeatedly, was resolved by placing spillovers at appropriate locations to take up the imbalance.

Each part was weighed with and without the sprue and cut at different locations to measure the subchannel wall thicknesses. The dimensional stability and warpage were measured by special checking fixtures. The best channels (front and rear) were put on the door and cycling tests were performed. Results were as follows.

1. Wall thickness. Thickness devices such as ultrasound and IR were found to be unsuitable for hollow channels. The devices did not indicate voids in the walls or other areas arising from gas penetration. CAT scans were performed on a few parts at approximately 24 locations on each part. Figure 6, the CAT scan of one cross section, clearly shows the nonuniform wall thickness distribution at that location and the melt profile and very thin wall joining the two subchannels. Based on this information, the gate design was modified. Wall thickness variations depended on processing parameters and ranged from 2.5 mm + 1.5 mm -0.5 mm. Ultrasound thickness devices are currently being re-evaluated.

2. Despite much effort, a hollow cross section was never achieved on about I inch of the subchannel of the top section having the lowest volume on the end. We believe that the gas possibly crossed over through small ribs between the subchannels, disturbing the gas flow front, and causing melt to be trapped in between by the gas flowing in opposite directions.

3. Strength testing. Twelve samples from four different trials were tested under two methods. One method consisted of an 8mm pin in the channel loaded perpendicular to the guide. The other consisted of loading across the flat of the channel with an 8-mm rod. The average load at failure was 280 lbs. The average displacement at failure was 0.85 in. Both results met specifications.

4. The simple mold flow simulation was of no help in the attempt to understand phenomena such as nonuniform wall thickness distribution, melt flow pattern, and optimum injection time and pressure.

Gas Through Cavity


The front guidance channel tool was modified for this process. Figure 7 shows that the gate is now located at the bottom part of the left subchannel and the gas injection point at the bottom wall of the part near the plastic injection point. A 30% glass-filled thermoplastic polyester (PBT) was used for process optimization. A long-glass-filled polyester was also tried, and we plan to evaluate other engineering thermoplastics.

Initially, the right side subchannel was found to be completely hollow, while the upper portion of the left side subchannel was completely solid. In an attempt to understand this phenomenon, a short shot analysis without gas Fig. 8) was run to follow the flow behavior of the plastic melt as it filled the cavity. The left side subchannel (near the gate) was observed to fill faster than the right side subchannel (opposite the gate). Though there was little difference in the relative flow fronts, once the gas was introduced there was a great difference in obtaining the through-hole. Figure 9 shows that the gas takes the path of least resistance, that is, the subchannel having the least plastic, which is in this case the right subchannel. Introduction of the gas therefore reversed the flow fronts, making the right subchannel completely hollow and the left subchannel only partially hollow. To balance the plastic flow front, the gate size and location were modified, and the wall connecting the two subchannels was tapered to have a larger opening on the right side.

Because of the number of variables that control wall thickness uniformity, we decided to run a design of experiment (DOE) using the Taguchi Method. The key processing variables and their ranges are shown in Table 4. Although our analysis of cross-sectional wall thickness and location thickness uniformity is incomplete, observations indicate that a uniform cross-sectional thickness distribution was achieved (Fig. 10).

Because it does not predict plastic/gas flow, conventional flow analysis software was of little help in reducing the number of costly, time-consuming, trial-and-error tests. Once the gas starts flowing into the cavity, pressure and temperature fields and filling patterns change dramatically. We have begun evaluating new software from AC Technology, Ithaca, N.Y., that simulates the cavity filling of polymer melt and subsequent or simultaneous injection of gas through the sprue and/or cavity. This software also takes into account the heat, mass, and momentum interactions of the gas and polymer melt. It can predict the gas penetration and the resulting skin thickness of the gas domain in addition to regular output such as temperature and pressure fields at specified times and the required clamping force.

One objective was to compare the results predicted by the software to the actual experimental results. Figure 11 shows the predicted polymer melt front at the beginning of gas injection, when the cavity is about 80% filled by the plastic melt. The predicted normalized skin thickness distribution is given in Fig. 12 together with the melt and gas front advancement pattern. We are working on effectively utilizing such results.


Dimensionally stable, warp free, and lightweight window guidance channels that meet structural performance requirements were successfully produced. For this achievement, we received an Award of Excellence for Product Design and Application at the 1990 Conference of the SPI Composites Institute.

Each process, gas through nozzle and gas through cavity, has its own benefits and limitations based on part design complexity, type of mold, number of cavities, tolerance requirements, cost, equipment life, and necessary modifications to the injection molding machine.

The part must be designed for the process, but at this stage the engineering know-how for part design, tool design, and process optimization is not fully developed. While conventional flow analysis simulation was of some help, new plastic/gas flow analysis software showed more promise in reducing development time.

DOE indicated that plastic shot size, melt temperature, and gas delay time are the most important variables affecting the wall thickness uniformity and distribution. Other significant variables, in order of importance, were gas injection pressure, plastic melt injection speed, and gas injection speed (which depends on the type of equipment). Of course, the part design and the dimensions and locations of the plastic gating and the gas pin are equally important.

It is worth mentioning that for each process many developments are under way for improved control of gas volume, pressure, injection speed, and time. The new gas equipment will be more sturdy, to minimize day-to-day production problems and to lengthen production life. The growth of this technology will depend heavily on cooperation among equipment suppliers, product and tool designers, and plastics molders. TABULAR DATA OMITTED
COPYRIGHT 1991 Society of Plastics Engineers, Inc.
No portion of this article can be reproduced without the express written permission from the copyright holder.
Copyright 1991 Gale, Cengage Learning. All rights reserved.

Article Details
Printer friendly Cite/link Email Feedback
Title Annotation:Processing
Author:Shah, Suresh; Hlavaty, David
Publication:Plastics Engineering
Date:Oct 1, 1991
Previous Article:Molded-in-color TPO bumper covers reduce cost and enhance design.
Next Article:Additives: maintaining quality & appearance under pressure.

Terms of use | Copyright © 2017 Farlex, Inc. | Feedback | For webmasters