Life beyond the 3D print

…there is a tendency to believe that 3D printing should be used for all prototyping requirements

However, there is a tendency to believe that 3D printing should be used for all prototyping requirements. This overlooks the value that an experienced prototyping company can contribute to product development.

When it comes to printing, I believe that a little knowledge is worse than no knowledge at all. It is important to engage with a prototyping company and seek their expert advice based on the desired outcomes of your project; the hand skills of a modelmaker are as relevant today as ever.

Rapid-prototype technology has strengthened this skill, and seeking advice will only extend the possibilities. Such advice should be based on all services and techniques, and not just the technology a company possesses in its armoury – it is about using what is available in today’s market to achieve the best outcome. I will expand on some of these additional services and how they stand alongside 3D printing.

The invention of 3D printing is not as recent as many may think. A patent was first registered in 1986 by Charles Hull for the first stereolithography machine (or SLA machine as it is now often called).

The original output print, although extraordinary for its time, was somewhat difficult to use in many applications. It was brittle and of much lower resolution compared to what clients would expect now. Although the SLA machine remains very similar commercially, today the surface finish of the parts is flawless and the resins are more advanced.

A little knowledge

Ten years ago when I told someone I ran a range of printing machines they would look puzzled. Now it seems everyone knows someone with a 3D printer. With so many printers in the market place where does this leave the prototyping industry and the clients it relies upon? Some would say this proliferation of 3D machines has thrown the industry into confusion; others would confidently claim that for every prototype requirement there is a specific printer that can supply an exact solution. I feel there is a little truth in each.

It’s not just a need for speed – clients want more!

The 3D printer is fast – there is no other method of manufacture that will provide parts for fit and function testing or a space model quicker. There is no substitute for this at the start of any prototyping process. But what if a client wants more? What if a client is confident that the form now fits its application, but wants to know how colour will affect the look of the product? What if they want a gloss finish, a texture finish or rubberised touch points? All of this can be achieved with the skills of a modelmaker.

Splitting the model up from a client’s computer-aided design (CAD) into colours and finishes means the modelmaking process can begin. Each coloured part will be printed independently, then hand fitted to one another to create high-quality surface finishes with flawless colour breaks, without the need for masking. Once these parts are painted, the model can be reassembled, leaving a fully finished ‘block’ model. All of this requires the skills of a modelmaker.

Next it is necessary to consider user interaction with the model – how can clicks, button pushes and door closures be replicated in a realistic way, to gain feedback from users in human factors studies?

With advances in the painting field it is now possible to create simulated chromes and high-lustre metallic finishes. It is even possible to affix metal plating directly to a printed part, in copper, nickel or precious metals, giving a cold touch and a heavier part that denotes further quality.

Graphics can then be applied to the model to add finishing details. This can be done via a screen print, tampo print or the latest direct digital printing straight onto the product, saying goodbye to the fragile rubdown transfer. With this level of finish, it can be difficult to distinguish the model from the intended final product, and is something that simply cannot be obtained from even the best colour output 3D printers.

Achieving the correct fidelity to test interaction

Next it is necessary to consider user interaction with the model – how can clicks, button pushes and door closures be replicated in a realistic way, to gain feedback from users in human factors studies?

The technology is certainly available with 3D printing to achieve the dimensional accuracy required (if high-quality Viper SLAs are employed, and a build layer of 0.05mm in an epoxy-based resin). But this level of interaction often requires prototyping before the design team have entered the detailed design phase – a realistic mock-up is needed, to simulate just the interaction and not necessarily the final method of achieving it.

This is where once again the skills and experience of a hand modelmaker come into play – typically, an experienced modelmaker will have tackled these challenges many times before and will have a number of tricks up their sleeve to achieve a robust solution more quickly and at lower cost than it takes to build a CAD model and print the components.

What if you need more than one?

In order to run multiple tests at once, present in multiple locations or send copies to many interested parties, vacuum casting is the next logical step in prototyping. This process is a step closer to an injection-moulded part.

In this type of casting, silicone rubber tools are filled under vacuum with a polyurethane resin. The silicone tool is created from a 3D printed SLA master which is then hand-finished in all ways to replicate the surface finish of the intended final production part (just like in the space model). The castings will be processed through the silicone tool to create a self-coloured part in a mimic polyurethane resin, replicating what would be the intended injection-moulded material.

More accurately mimicking mechanical function

Vacuum casting not only makes a more robust model by removing the risk of paint wear on parts during handling; with both rigid and flexible materials available in the polyurethane range, all aspects of the product can be replicated including live hinges, compliant features, even twin-shot parts. Now the prototypes will display repeatability from one assembly to the next, delivering on both finish and fit.

The introduction of CNC (computer numerical control) machined metal inserts to the silicone tooling process means that very tight-tolerance parts can be manufactured akin to injection-moulded parts. These prototypes will be hard to distinguish from a final production assembly, and will demonstrate any function (even housing electronics, if required) to give all aspects of interface. With a production time from single-part CAD delivery to cast parts in less than five days, this process can be incredibly time-efficient.

The vacuum-casting process can be used not only for prototyping, but also for low-volume manufacturing. When volumes are expected to be in the region of less than 100 units per year, this method of manufacture can be practically perfect. The investment in tooling is low and part design forgiving. The parts do not need draught – it is possible to manufacture with negative draught and undercuts. Differential wall section will also not cause an issue. With these forgiving parts it can be a great way to get products out to customers with a smaller initial investment than injection moulding.

Sometimes simulation is not enough

Simulating injection-moulded products is not always good enough, and the actual material is needed to assess the longevity of the true function of a part under load without compromise. To achieve this, there are two options: actual injection moulding or CNC machining.

Rapid injection moulding has developed over the years and it is now common to make a soft aluminium tool or even steel tooling for a low-volume production. The challenges of injection moulding remain the same in “soft tooling” as in production tooling; draft is needed and side cores may also be necessary. Unlike the vacuum-casting process, there is no way around this. Simple parts can be made very rapidly but complex parts can be slow and expensive. This may be where CNC has the advantage.

“It is important to think outside of the ‘cube’ of 3D printing.”

CNC machining is a process that removes material from a block (as opposed to 3D printing, where material is added in layers to create the part). A block of the required material is presented to a cutting tool that is spinning at high speed, which then moves over the surface of the block, removing material in stages until the final part is produced. This process is normally a slower process than 3D printing, as parts tend to be machined one by one rather than in a cluster, as in the printing process. Parts machined from solid material have a consistent structure throughout, whereas printing processes such as selective laser sintering or fused deposition modelling produce a structure in the final intended material, but one made from fused granules and the other a printed strand of material; each inevitably changes the mechanical function.

CNC machining leaves a clean surface finish that can be polished to enable free movement of assembly. However, there are some limitations to CNC machining – some corners will require corner radii, where the cutter is not able to achieve sharp corners, and not all parts can be machined due to limited cutter reach or obstructed access. Again, advice from a prototyping company will inform the decision as to whether or not a CNC process is beneficial for a particular product.

Not just picture-perfect models

This article should demonstrate that it is important to think outside of the ‘cube’ of 3D printing. Whilst the impact it has had on speeding up the product development process is huge, it is only part of the picture when it comes to developing effective prototypes.

It is worth engaging with a prototyping company early on, to seek their expert advice based on the desired outcomes of your prototype. Naturally, given my position, I would say this, but it is amazing how much the hand skills of a modelmaker are as relevant today as they were 25 years ago.