The Journey from ‘Art to Part’ …and how to avoid the many pitfalls

Graham Webster, director at Plastic-IT, discusses the injection moulding route on making a high-quality part for medical devices.

On the face of it, the process of injection moulding is apparently simple – make a mould, force some molten plastic into it and let it cool, open the mould and ship the product. However, those reading this will know that is an oversimplification – so how complex is it?

First let’s look at why we use this process – which is now the largest manufacturing method in the world. The common denominator is that it is done to make profit – at least, that is the initial objective.  The ‘long and winding road’ is full of issues that detract from this primary aim, and here we try to identify them and provide knowledge-based information on how they can be avoided. Whether we are making a tiny, close tolerance part in an ‘exotic’ engineering material or a high-volume packaging or medical disposable product, every facet of the process must contribute to the creation of a ‘margin’ - financial or altruistic.

At the outset of this journey there are many difficult decisions to be made. The first is to determine which of the 10,000+ commercial grades of polymer we need to use for the application we are considering. All injection mouldable thermoplastics will melt at between roughly 90°C and 400°C so generally they can’t compete with metals on a heat basis. Most are good as electrical insulators whereas metals are mostly good conductors. Thermoplastics may be very stiff especially when combined with reinforcing fibres of glass or carbon (19,000MPa - 60% Glass filled polyamide 6) or very flexible and even elastic. Some are completely transparent and others totally opaque. Some burn readily, others self-extinguish if the ignition source is removed. Most can be coloured – some better than others. A few are resistant to UV and Ozone (weatherability) whilst others are poor.

Determining which is the optimum polymer - the grade of that polymer and other additives such as lubricants, flame retardants, etc. to choose, is a technical and commercial minefield for anyone who is not a specialist.

Until the performance criteria for the product are determined it is impossible to specify the polymer type and grade. Consequently, designing the geometry of the part for its fit, function and aesthetics, must be done in concert with choosing the polymer. This should be the expertise area of the product designer but as has been illustrated, it is a complex task that few people can take on with certainty. Once a mould is made there is some opportunity to mould different materials for evaluation but due to the likelihood of these polymers shrinking by different amounts during the moulding process, dimensional accuracy will be compromised.

Eventually after several iterations, some CAD emerges, and a polymer type and grade is identified - so what next? Frequently it is a call to a mould maker because surely all that is left to do is tool the part and mould it? This is often what does happen, but the outcome is almost never right first time.  Consequently, the costs will start to rise as the mould trials and subsequent mould rework cycle develops.

The better way is to prototype it. Today additive manufacturing can produce a 3D object direct from the CAD in a polymer something ‘similar’ to the injection moulded intention. However, it delivers little more from a production engineering perspective because at this point there has been no evaluation as to whether the geometry can be satisfactorily moulded. To do this we turn to a CAE moulding simulation product such as Moldflow.

In comes another knowledge-based technology that baffles the majority. You may be surprised to discover that the technology has been around for more than 40 years. In skilled hands it will simulate the total moulding process and determine the final sizes and shape of the part post moulding by using the CAD model and a very comprehensive data base of polymers. If you had a crystal ball that would tell you the six numbers in next Saturday’s lottery, you would surely use it. CAE tools, expertly used, are just like crystal balls – telling you precisely how well the CAD would mould – without having to make a mould.

Due to the aforementioned complexity of developing a toolable geometry (that satisfies fit and function requirements in an economically viable polymer that can be moulded), NOT taking advantage of science-based predictions from a CAE tools in 2021 is madness.

Frequently we see this application of CAE referred to as a DFM report whereas of course ‘Design for Manufacturability’ (DFM) is, by definition, a ‘design process’ and not an error checking process to be used just before steel is cut and performed by the mould maker.

For a moment let us revert to understanding the product development process. It is unlikely that the exact Widget that we are wanting to make as an injection moulded part has not been made before, so there can be no pre-existing knowledge of the detail of its manufacturability at the outset.

It is impractical to expect one person to be knowledgeable in so many diverse technologies, so it is best to be a collaborative process of teaming together experts in their field as early in the project’s life as possible. Instead of this being a serial process, today with CAD, CAE and on-line meetings the iterative process of getting from Art to Part can now easily be a true team effort.

The process must start from the product designer as it is normal here that the brief or requirement is distilled form the end user. These are always ‘fit and function’ specifications and probably also aesthetic ones too. The minimisation of cost is also a pre-requisite. Today the process should be one of collaboration between the product designer, the materials expert, the tooling engineer, and the manufacturing engineer who also take data, advise, and feedback from CAE engineers who can evaluate strength, dimensional accuracy and mouldability. 30 years ago, the term “Concurrent Engineering’ was coined; today it’s called Concurrent Knowledge based Part Production or CKPP for short. It is a specific methodology following concurrent engineering concepts specifically for plastic injection moulding. When adopted, the outcome is parts right first time at optimum cost in the shortest time frame.

Shouldn’t this be the methodology that you adopt on every project?