New 3D Printing Developments on the Horizon

Adrian Bowyer, Director at RepRapPro Ltd and creator of the RepRap self-replicating 3D printer, and Mark Cundle, Head of Technical Marketing at RS Components, consider a few future possibilities for 3D printing such as making electro-mechanical devices or embedding graphene-based electronics, bringing about a liberation for designers and product developers.

 

3D printing has the capability to revolutionise product development by directly making a solid object from a 3D CAD model. 3D printing essentially works in two dimensions and builds up into the third, with successive layers of material such as plastic or metal being laid down in different planar shapes.  And this is creating new possibilities in mechanical design. While it is certainly true that CAD systems are getting much better at seeing if mechanisms will function, they will never be quite as good for a designer as the physical reality of holding the designed object. That part of the early product development process, the rapid making of prototypes that have dimensions of a few millimetres up to a few hundred, has been a key driving force behind the development of 3D printing technology.  It is still one of its major uses today across many industries.

More importantly, 3D printing is moving away from being a niche technology previously available only to larger companies to one that can be used by small businesses or even individuals.  RepRap machines, for example, are low in cost, and are suitable for desktop use.  They employ fused-filament fabrication, which is one of the most versatile among the half-dozen or so 3D printing technologies in that it is easy to extend to work with multiple materials.  And RepRaps 3D-print their own plastic parts, so when you have one RepRap you can freely use it to make more of them, reducing costs even further. Also, while 3D printing was originally too slow and too expensive for volume production, its new low cost, and the consequent possibility of running many 3D printers in parallel, is allowing it to be seen more and more as a feasible final manufacturing method. Another new freedom is the possibility for the increased customisation of products – the ability to make one bespoke device or object for one customer easily and quickly – as opposed to the traditional volume manufacturing of identical products. As the cost of 3D printers comes down with the integration of new technologies and the increasing ability to print with new materials, new possibilities will present themselves to designers in many different fields.

 

New possibilities and ‘things inside other things’

An example of something that 3D printing enables is the making of objects which are literally impossible to make in a single piece using other technologies such as the traditional injection-moulding process (which, of course, still remains mankind’s most important mass-manufacturing method). What 3D printing allows, quite simply, is the making of things inside other things. An elementary example of this would be a pea-whistle that could be made with the (plastic) pea already inside, whereas conventionally such a whistle is made of steel in parts and is then brazed together with the pea inside via a second process step.

The reason that making things inside things is difficult or impossible for a traditional tool is because of the number of dimensions (in the mathematical sense) in which a computer-controlled manufacturing machine such as a CNC milling machine has to work. Whereas a 3D printer is working in two dimensions at a time, the CNC machine has to handle five dimensions: the three directions in space, plus - entirely independently - the rotation of the cutting tool about two directions (rotation about the axis of the cutting tool does not count).  This profoundly affects the complexity of the required algorithms. As a consequence, most traditional manufacturing machines are not constrained by their own physics, but by what we, or our computers, can tell them to do. In contrast, the 2D algorithms driving 3D printers can instruct the machine to make absolutely everything of which the 3D printer is physically capable.  3D printers can also make things that are impossible to make using equipment that employs a cutting tool, because getting access actually to perform the cutting is an inherently tricky problem. However, a note of caution must be made: if a production process does not involve a 3D printing machine in the end, designers used to 3D printing will need to exercise some self discipline to avoid designing products that cannot easily be made with the ultimate manufacturing machinery.

 

Multiple materials

Today, we are primarily 3D-printing objects made from a single material – usually a thermoplastic; and while it’s certainly a useful ability to make objects inside objects, the number of cases for which we require this using only one material is probably not enormous. However, the move to 3D-printing machines that can deal with multiple materials can mean this ability becomes very significant indeed. As an example, consider a mechanical object that also requires some electronics.  The conventional process for this is to make a PCB that will sit within a cavity inside the mechanical object, followed by assembly including attaching the wiring and so on. But once machines can print electrical conductors – and some more expensive 3D printing machines can already do this – then there is the potential to print electrical connections inside the mechanical parts with embedded 3D-printed wiring.

The ability to design electromechanical devices in an all-in-one process liberates the designer from the necessity of accommodating the rather restrictive geometry of a PCB and can allow the production in one shot rather than via two completely independent and separate processes.  This should enable a significant reduction in time-to-market and production costs. So, the nature of the process of embedding things inside things means that conductors can be printed within mechanical parts and along twisted routes around bearings, for example. The caveat here is that we are not dealing with a high-frequency design where long (and hence inductive) pathways could be problematic.  3D printed electronics, even if they incorporate conventional integrated circuits, are likely to be physically bigger than their PCB equivalents, at least initially.  On the other hand, they can occupy all three dimensions, whereas a PCB is always a 2D device.

Much the same principle can be applied to pneumatics and hydraulics. It will be possible to combine the printing of hard materials to make the pathways that allow the flow of fluids, together with soft materials such as silicone to form one-way or stop valves, for example.  These might use the electrical circuitry described above to activate them.  3D printing is available already for silicone – and it is not too difficult to design a printing head on a low-cost 3D printer that will extrude what we might more commonly recognise as bathroom sealant. So, if a machine made by 3D printing contains a pumping system which is combined with electromagnetics and electronics, then we have an extremely sophisticated single-shot method of making items such as micro-reactors for doing chemical engineering a few millilitres at a time.

Addressing electrical conductivity specifically, RepRapPro is currently working on printing low-temperature-melting metal alloys. It is already possible to obtain 3D-printable plastics that are loaded with materials that enable electrical conduction.  But these have quite high resistivity, and so cannot be used easily to make high-current machines such as solenoid actuators and stepper motors.  But working with metals that have really good conductivity will mean anything that can be done with conventional wires or printed circuits can be done with the deposition of metal alloy.

 

Into the mix

Another key technique in development is the ability to mix materials together – and this is not so far away. This comes with the restriction that there are some materials that just do not mix together well, but there are many mixable plastics that can be printed that are actually quite soft and flexible and many others that are hard and rigid. And if they are mixable, then by changing the proportions of the mixture it is possible to print an object that has any desired degree of stiffness at any geometrical position inside the object. This means the creation of an object that is hard in some places and soft in others, and anything in between. It is very difficult to make an object with graded mechanical properties in a single shot by any conventional engineering manufacturing process.

A similar trick can be done with heat conductivity by printing a combination of metal and plastic.  Here these would not be mixed, but printed side by side microscopically in differing proportions to get different regions of good and bad heat conduction.  Combine that with the ability to print paths for fluids that evaporate and condense, and the ability to print porous wicks, and then heat pipes can be added to the heat-flow design.  This again, as always, would be 3D printed in a single process.

(As an aside, graded mechanical properties are a trick widely used by living organisms, which have control at the microscopic scale as everything is built from the cells upward. An example is a person’s head, which obviously includes a rigid bone skull yet a bendable cartilage-based nose. Bone and cartilage are made from similar materials, but in different proportions mixed together to give different mechanical properties.)

RepRapPro is currently looking at developing a nozzle – which could be available as early as mid-2015 – that will allow the mixing of different materials together to achieve combined mechanical properties. In addition to this, and perhaps trivially from an engineering perspective, but much more importantly from a design one, the mixing of multiple colours of plastic delivers the ability to print in an infinite range of colours using the CYMK (Cyan, Magenta, Yellow and Key/black) process, much like an inkjet printer. However, the fused filament fabrication process used by the RepRap printers will not offer sharp transitions in colour such as can be achieved in a photograph produced by an inkjet printer.

 

New logic?

Looking a little further into the future, there is also great potential for integrating semiconductors into the process. To handle metals, as mentioned previously, RepRapPro is looking at what is essentially an inkjet printer depositing metals in tiny droplets. And once the right inkjet technology has been developed, then printing graphene is a next possible step. Without going into the details of how this can be achieved within a 3D printer, graphene is a highly promising material that is actually quite easy to manipulate.  Technically, graphene is a zero-gap semiconductor, and both P- and N-types have been made.  So the potential is there to print actual electronic devices as well as the electrical connections between those devices.

However, there will be limitations in device resolution. For example, the current resolution – or the repeatable accuracy of the nozzle/head movement – of the RepRap machine is 0.1mm and the filament extruder is much larger still, at roughly 0.5mm wide.  The inkjet will be able to deposit droplets of material that will be much smaller than this. So eventually the limitation will become the accuracy of the machine's movement, which today at 100 microns is still very large compared to a transistor on an electronic chip. But assuming once more that the design does not use high frequencies, the size of a piece of electronics is only critical when it needs to be packaged. If an individual transistor ends up being a millimetre or so across, this may not be a disaster as there should be plenty of available volume in a mechanical product in which to distribute the electronic devices. As discussed earlier, there are no longer the constraints of having a PCB among a whole bunch of mechanical devices, which liberates the designer from the process of having to make the electronics compact. Having said that, clearly a high-complexity product that requires the processing power or logic circuitry of hundreds of thousands of transistors will be distinctly more difficult to pack together – even if there are three dimensions rather than the two traditionally involved in chip or PCB layout. So, making electronic devices is perhaps a few years away yet, but it is certainly on the horizon.

 

All in good time

The future for 3D printing is full of possibilities – mixing plastics to make materials with different mechanical properties and different colours in a relatively low-cost printer is extremely close; likewise the printing of metal alloys. However the most important idea is combining different materials in a single process – and this is where inkjet and fused filament fabrication (FFF) based technologies are ideal. We’ll see a growth in the number of materials being made available over the next year or so. Silicone is already here and many people are already working with clays and ceramics, and also metallised ceramics and many different resins.  Any material that is available in the form of a paste is simple to manufacture with using fused filament fabrication. In the next few years, 3D printing can become a single-shot process that will truly liberate electronic and mechanical designers and enable them to develop new and innovative creations and products that were previously all but impossible.