Injection molding is a widely used process for high volume production of thermoplastic resin parts. It can produce complex parts with different combination of plastics and fillers and has a high efficiency ratio between applied material and the end product.
Despite major advantages of injection molding, there are significant difficulties in controlling quality of the molded part. These difficulties arise not only from the complex dynamics of the process and unpredictable behavior of the material under temperature-pressure fluctuations, but also from the lack of consistent relationships between the material properties, machine parameters, mold geometry and molded part quality. There are several sources of variation in the injection molding process which include material properties, process machinery, mold design, environmental fluctuations, and human interaction. If these factors had precisely defined and deterministic effects on the process, the quality of the molded part could be easily controlled. However, there are many interactions between stochastic factors, resulting in frequent irregularities in process parameters so the ideal state of constant quality does not exist. Examples of such irregularities include mold and melt temperatures that can drift from their set values, changes in polymer properties from batch to batch, and the machine and tool performance, which alter as they wear. Besides these major factors there exists small natural fluctuations making the process even more unpredictable.
There are several different approaches to quality control and process optimization that are used in industry today,such as expert systems, continuous process control, regression modeling, and design of experiments. Unfortunately, none of these approaches deliver 100% quality assurance. Difficulties in regulation arise mainly from non-linear character of the process and interactions between inputs that make the relationships between machine parameters and part quality not obviously defined. At the same time, the number of variables adds difficulty in evaluation of the significant parameters. Thus, a more capable quality control system requires a process model that not only describes the relationship between the process parameters and the resulting part quality, but is also capable to learn and express the process nonlinearities and complexities. Because of the stochastic correlations between the process parameters and characteristics of part quality, such a model should also be adaptable to the changing process conditions.
To overcome the difficulty require the engineers have wide knowledge and humble attitude. It is a course of researching, and should be complex.
SO,injection molding is a complex process!