FEATURE
Issue: 

Printing parts

Laser additive manufacturing is an excellent tool for prototyping, but now the technique is being used to build production-quality parts, as Tom Eddershaw discovers

While the term 3D printing has certainly garnered a lot of attention in the media, industrial interest in additive manufacturing (AM), as it is known in the manufacturing community, is partly due to its ability to work with new materials. Metal composites, ceramics, reinforced plastics, and many other powder-based materials can now be processed for parts in a range of industries.

Visitors to the Laser Additive Manufacturing workshop in Houston, Texas on 12 and 13 March will get an idea of the range of AM technology available and the types of industrial parts being made. In terms of working with exotic materials, however, it is largely due to incorporating higher power lasers into the machines that allows production-quality parts to be built by AM.

Andrew Payne, doctorial candidate at the Institute of Manufacturing, Cambridge, commented: ‘The process of building with metal has evolved at the rate that the lasers have evolved, in the sense that we can now produce lasers that are capable of fully melting  a wide variety of metal powders. In their early days, fibre lasers weren’t powerful enough to process the higher melting point metals.’

Payne explained: ‘In the early attempts at metal powder bed building, the powder grains would be sintered rather than fully melted. Sintering is a semi-melting process whereby the grains adhere together but leave air gaps in-between. These “green” parts are then infiltrated with a liquid metal with a lower melting point that would fill the gaps, so you would have a fully dense part, but it wouldn’t be a single metal or even an alloy. You can still get functionality out of the part but it’s not the same as from a single metal.’

He continued: ‘Now, though, we have sufficient power to fully melt the powder, and therefore it is possible to produce nearly fully dense parts. It depends on your strategy, but many methods give above 99 per cent now.’

The cost of the process is still very high, however – prohibitively so for mainstream industry. AM is only really feasible at the moment for expensive, smaller volume production markets, such as motor racing, high-end sports cars, and the aerospace industry.

The technique might still be only used for niche applications, but the components made by AM are now functional, which wasn’t always the case. Kevin Lambourne, managing director and founder of Graphite Additive Manufacturing, recalled that when first used, the process was named ‘rapid prototyping’. He explained that at the start: ‘Most of the parts were one-offs; they weren’t functional and they weren’t accurate. But as other technologies were invented and materials started getting better, people began using them for production parts. A large part of the market still created prototypes, but functional manufacturing was beginning to come into play.’

Lambourne continued: ‘As they [AM companies] began to explore new markets and applications, such as in the aerospace and automotive industries, the perception of the technology changed and the phrase “additive manufacturing” was born.’

Whether it is called rapid prototyping, additive manufacturing or 3D printing, it still involves taking a computer-aided design (CAD) model, slicing it into thin sheets and then using a machine to manufacture each layer one on top of the other.

While there are a number of different methods for AM, three of the most common approaches are: stereo lithography (SL, more commonly known as SLA for stereo lithography apparatus), based on laser curing resin to build up the part; selective laser sintering (SLS), where the material is introduced layer by layer in powder form and the part laser sintered; and direct metal deposition (DMD), in which powder is sprayed on a point and melted by a laser, and then traced around the part design.

Lambourne stated: ‘It’s exciting where people are adopting it [AM], using it more and more in different areas, and we are here to help advise them which is the best material for their application. We may even help with the design process to ensure that the best result is achieved.’

Graphite works with carbon fibre reinforced polymer to provide parts that are rapidly processed, strong, and light. The company has made components for motorsport, aerospace and on the Bloodhound SSC, the 1,000mph world land-speed record car currently being built.

Lambourne said: ‘It’s not really a prototype anymore; it’s a production piece,’ referring to the AM parts produced for motorsports. ‘They may only need a few to take to a race and a few spares, or they may build dozens of parts. SLS produces a lightweight, tough material that’s temperature resistant.

‘The beauty is that they can produce quite complicated designs, which would take weeks to manufacture normally, quite easily and quite quickly. Within hours, literally within hours, they could have parts to take to the race track.’ That is not to say that all the parts for racing teams are manufactured in this way. For strong and structural pieces, such as the wings, wishbones, and most of the body work, carbon fibre is still better.

The material will to a large extent determine the laser parameters, with polymers requiring lower laser power to fully melt the powder, while sintering metal needs much higher powers.

SLA machines typically use a UV laser to cure the epoxy resin. Lambourne commented: ‘There have been three main types over the years, and have changed as laser development has progressed. The original small SLA machines used a helium-cadmium (HeCd) laser which only produced around 30mW. The first of the large frame SLA machines used an argon ion laser and these produced around 200mW, but were very inefficient and wasted a lot of heat and energy.’ All of Graphite’s current SLA machines are fitted with solid-state lasers, which are much more efficient and the power varies on machine size between 100mW and 1W. He went on to explain that for laser sintering machines running plastic powder, the company uses a high power CO2 laser to melt the powder, which typically ranges in power from 60 to 100W. Laser sintering machines melting metals require even more energy and typically use a Yb-fibre laser from 200W to 1kW.

Jon Blackburn, section manager of the laser and sheet processes department at TWI in the UK, explained how the metal sintering machines manipulate the beam with a galvanometer scanner and therefore the beam quality of the laser is important.

Blackburn said: ‘You tend to have quite long focal lengths, so to maintain a small spot size you need a good beam quality. This means you tend to be limited to very high beam quality fibre, disk or CO2 laser sources.’

He continued: ‘For DMD, you typically don’t need as high power densities, and consequently don’t necessarily need to use laser sources with excellent beam quality. This enables you to consider diode laser sources, as well as fibre, disk and CO2 laser sources, for DMD.’

Dr Bhaskar Dutta of DM3D Technology, a direct metal deposition company based in Michigan, USA, said that ‘different lasers are used by different companies. CO2 is an old work horse in industrial lasers, while diode and fibre lasers are gaining popularity very quickly in the cladding market.

This is because metals absorb more diode or fibre laser energy than from a CO2 laser due to their shorter wavelengths, so the efficiency is better. However, we have found that for certain applications a CO2 laser is still better due to less dilution of the substrate or the parent part.’

By choosing the laser to suit the materials, the efficiency of the machine will be improved. Dutta explained: ‘Each material has its own absorptivity; copper or aluminium absorbs less laser energy, while steel, Inconel, titanium or cobalt alloys absorb more. We use a completely different set of process parameters that depend on materials and application. These proprietary recipes and the properties of DMD materials are part of our database.’ Payne, whose PhD project is based on improving the efficiency of AM processes, explained that there is no one set of parameters, as the optimum setup depends on both the size of the powder grains and the composition of the material. So, to give an example of the balance that has to be met, smaller grains are easier to melt, but if they are too small then static electricity can be a problem, affecting the powders flow characteristics to the detriment of the process.

Payne described the delicate balance when designing an AM system and how careful consideration should be given to the thermal stresses from the laser that could damage the part. For example, when trying to build a part with an overhang, the thermal conductivity of the un-melted powder in the previous layer is different to the melted powder. ‘This difference in conductivity leads to differential cooling which can cause the part to curl up and cause deformation or even failure if the raised part is knocked by the roller,’ he said. The part has to remain in a fixed position for the layers to build up accurately.

The three main laser parameters are power, spot size and scanning speed. These three things combined essentially determine the energy density – how much energy is put in per unit area. They are delicately balanced and a lot of work has gone into calculating the optimum set-up.

Building high-end parts with complex geometries is where additive manufacturing is most suited, because to make these parts by traditional means would be too costly and take too much time. And the components now being made with AM are production parts rather than prototypes.

As the laser processing parameters become better defined for different materials, the use of the technique is certain to grow.

About the author

Tom Eddershaw is a technical writer for Electro Optics, Imaging & Machine Vision Europe and Laser Systems Europe.

You can contact him on tom.eddershaw@europascience.com or on +44 (0) 1223 275 478.

 Find us on Twitter at @ElectroOptics, @IMVEurope, @LaserSystemsMag and @ESTomEddershaw.