Modulation matters - how to build all features great and small

Additive manufacturing (AM) gives us the freedom to build a vast range of part geometries, consolidating multiple conventional parts into a complex whole. Multi-functional AM parts often combine the structural with the intricate, mixing relatively bulky geometries with fine details. Building such parts cost-effectively means that our AM process must be both productive and precise.

How we deliver laser energy can significantly influence part costs and quality in laser powder-bed fusion (LPBF). Our parameter choices affect not only how the metal powder melts, but when combined with the local part geometry, they also determine how rapidly the melt pool cools and solidifies. Bulky features conduct heat away more effectively than smaller features can, and so our energy input must be tailored to suit varied local conditions. Using the right parameters in the right place is critical.

This article looks at how versatile laser modulation can help to match the mode of energy delivery to the needs of component features of all sizes. We will consider both continuous and modulated energy delivery, their advantages and limitations, We will see how it is often best to combine different techniques to produce AM components.

Productivity and precision

AM components come in all shapes and sizes, from the strong and solid to the fine and feather-light. They may use regions of bulk material to provide strength under load, or they may feature thin walls or lattice structures to reduce weight or increase performance.

Components that have been specifically designed for AM are often multi-functional, fulfilling the roles of several traditionally-produced parts. AM parts will often combine structural elements with complex details that deliver additional functionality, such as joining, fluid passages or heat transfer. Many use textured interior or exterior surfaces to enhance their performance.

Orthopedic implants are exemplars of this approach. Acetabular cups are now increasingly produced with solid interior load-bearing surfaces covered with a lattice exterior surface, whose role is to promote osseo-integration to produce a strong bond between the artificial implant and the patient's living bone.

It is quite common, therefore, for additive manufactured components to exhibit a wide range of section thickness, from solid at one end of the spectrum, to a very low volume fraction at the other. This 'spider' architectural bracket, shown in section (right), is a good example.

Of course, we want both precision and productivity from our build process as we produce these different regions of our component. Solid regions should be produced quickly and efficiently to minimise build time. The more intricate features, however, often require a more subtle approach if we are to achieve the precision, shape, surface quality and metallurgy that are essential to their function.

Heat dissipation drives parameter choice

LPBF is a thermal process. We input intense heat to melt each layer of powder and fuse it to the layers below. Solidification occurs as heat energy leaves the melt pool, with the majority flowing down through the solid metal below.

Heat dissipation is dominated by thermal conduction through the part that is being built and into the build plate. At the start of the build, there is good conduction down into the solid build plate, but as the build proceeds, the local part geometry becomes increasingly important.

If there is a good thermal pathway, then the substrate can be an effective heat sink, drawing heat away from the top surface of the part, enabling it to cool quickly between each layer. If the thermal pathway is constricted by the local part geometry, then heat will build up in the upper-most regions of the part as it is being built.

Another key factor is the time between laser passes. Thin wall sections feature short hatch vectors, so the laser moves back and forth across the same region frequently. This can result in intense local heating. Longer hatches in bulkier regions allow more cooling time before they are re-melted. Short layer times, sometimes encountered at the end of a build, can also result in a build-up of excess heat in the component.

Image above - part geometry affects the conduction of heat down into the substrate. Thinner sections are less effective at conducting heat, whilst shorter hatch lines result in more intense heating of detail features.

Impact of substrate temperature on melting behaviour

What are the consequences of this? In X marks the spot, I explored how process parameter choices must respect the melting behaviour of the material. Each alloy requires a different amount of energy to melt a given layer thickness, and too much energy will lead to keyhole porosity. This leaves an 'operating window' within which we can achieve successful melting and solidification. We should select an energy input that is somewhere in the middle of this region.

If the top of our part and the surrounding powder are pre-heated due to restricted dissipation of previous laser energy, then we will need to input less energy to create the required melting effect. Pre-heating reduces the power at which keyhole porosity will occur for a given scanning speed, narrowing our operating window. If we do not adjust our parameters, then we are likely to over-heat the alloy and form keyhole pores.

Impact of substrate temperature on solidification

Another consequence of an elevated substrate temperature is a reduced cooling rate, which affects the metallurgy. Slower cooling produces a coarser 'as built' microstructure. In Ti6Al4V, for instance, we see coarser alpha laths forming at higher build temperatures, whilst formation of the martensitic phase reduces with more build plate heating. This helps reduce residual stress, but also impacts on the strength and ductility of the 'as built' component.

Image above - faster solidification in Ti6Al4V produces a finer microstructure.

So, the way in which we input laser energy and the way in which heat dissipates both affect the component temperature, which in turn affects the melting and solidification behaviour of the next layer that we build. When the geometry changes, we need to adjust our energy input. Clearly, using the same parameters in all circumstances is not wise.

Using the same parameters in all circumstances is like driving with the throttle wide open - fine when the road is straight, but not so good when you reach some turns

So far, we have discussed how we may need to vary the power and speed of our laser to cater for different local part geometries, but not the finer detail of how that energy is delivered. In addition to selecting appropriate power and speed for each region of our part, we also need to consider laser modulation.

Continuous and modulated laser melting

Two main techniques are used to melt powder – continuous and modulated scanning - both of which are supported by Renishaw's RenAM range of industrial AM systems. Continuous energy delivery, as the name suggests, involves irradiating the powder using a continuous laser beam, which is moved back and forth across the surface of the powder bed to melt and then solidify the metal. The scan lines overlap, so that each successive pass of the laser partially re-melts the previous scan line and the layer below, creating a solid mass of welded material.

Modulated lasers operate in a slightly different way. Here the laser is turned on and off, creating a series of exposures, with a short gap between each one. Each exposure partially overlaps with the previous one. These can be formed into similar scan lines that efficiently move across the powder bed to solidify the bulk of the component.

Image above - continuous laser scanning (left) involves a series of overlapping scan lines, each formed with the laser operating continuously. Modulated lasers achieve the similar effect using a series of sequential exposures (right).

Image above - comparison of laser power delivery with time during a short hatch line in continuous and modulated modes.

The previous example shows the modulated exposures as overlapping circles. This is what we get when the time interval between exposures is sufficient to enable the laser spot to move between points and settle in its new position. As the time interval between pulses is reduced, there is insufficient time for this to happen and so the exposures elongate, eventually becoming a continuous exposure when the time interval is eliminated altogether.

Image above - impact of time interval between pulses on sequential modulated exposures

Melt track shape

So far, we have looked at things from above, but it is important also to consider the depth and profile of our melt pool in the vertical plane. In his article Viso profundum et latius, my colleague Martin McMahon explains how parameter choices affect the aspect ratio of our melt pool. Melt pools can be broad and shallow, or narrow and deep, depending on the choices of power and speed that we make. Our choice of laser modulation also has an impact here.

In continuous melting, the laser spot moves along the scan vector, dragging a long melt pool behind it. As it does so, the laser energy conducts both downwards and sideways. This produces an even melt track that is relatively wide and shallow, making continuous melting suitable for relatively high powers and speeds. It is the most productive energy delivery mode since the laser is switched on throughout each scan vector, making it ideal for rapid filling of bulk regions.

Image right - melt track produced by continuous melting. The laser energy creates a broad, shallow melt track that progressively solidifies as the laser energy moves on. Heat flows from the molten metal into the substrate and neighbouring powder and a solidification front follows behind the laser spot as the melt track cools.

By contrast, a modulated laser delivers energy in pulses with gaps in between when the laser is turned off. The length of these pulses and the time interval between them can all be varied. These short bursts of energy each create a small melt pool, which starts to cool and solidify as soon as the pulse is complete. After a time interval, the next pulse is delivered close to the previous one, creating a new melt pool and partially re-heating the previous melting region.

Image above - modulated melting generates small melt pools which cool between exposures and are subsequently re-heated by the next pulse.

This contraction of the melt pool between exposures tends to produce a melt track with a deeper aspect ratio than we see with continuous melting - i.e. the melt track is narrower for a given penetration depth.

The aspect ratio is also affected by the time interval between pulses. Where the gap between pulses is long, then each individual exposure forms a static local melt pool. As the time interval reduces, then these melt pools increasingly merge to form a wider melt track, becoming more and more like a continuous melting track as the interval tends to zero.

Detail features

A narrow melt track is useful when we are producing detailed features. Lattice structures, for instance, often require thin struts, which must be produced with a single melt track. Modulated lasers are ideal for this task, enabling the production of struts that are little thicker than the laser spot diameter. Varying the power and exposure enables us to control the melt pool width and depth, and thus the detail that we can produce.

Image above - an array of lattice struts produced with different combinations of modulated laser parameters. Higher powers and longer exposures produce thicker features. Struts as small as 120 um in diameter are producible with an 80 um laser spot.

Overhanging features

Where detail features are overhanging un-melted powder, we want to avoid excess penetration of laser energy that would lead to loss of definition. Here we want a melt track that is wide and shallow so that we minimise attachment of excess material on the down-skin surface. It is common to use lower energy for down-skin scans, with short modulated exposures being an effective way to limit energy penetration. See Viso profundum et latius for more details.

Residual stress

As I explained in Want to build accurate AM parts? No stress!, residual stress is a natural consequence of the rapid solidification and cooling in the LPBF process. As each melt track cools, it contracts and sets up shear forces between the layers. These stresses accumulate during the build and can lead to part distortion or even cracking.

Both continuous and modulated laser energy delivery modes result in a build up of residual stress. However continuous melting, especially when high powers and speeds are deployed, tends to produce higher stresses in the built component, as shown by the comparative cantilever testing shown below.

Image above - comparison of deflection of cantilever artefacts built in Inconel 718. The thin supports for each cantilever are severed, leaving the beam attached to the base-plate at one end only, revealing the residual stress accumulated during the build. Both parameter sets use the same layer thickness and energy input per layer, but the continuous parameters use higher laser power and scanning speed, resulting in a faster build rate. The trade-off for this increased speed is an 8% increase in residual stress. Note that this trade-off for different materials will vary.

For large, bulky parts, an increase in residual stress is unlikely to cause problems, but this may be critical for more detailed features.

Summary

We have seen how different parameter choices and energy delivery techniques each have their place in producing multi-functional AM components. Modulation matters when we need to combine productivity with precision. Continuous melting is great for speed, whilst modulated melting is often the best choice for details. Micro-structure and residual stress are also important considerations. A flexible process toolkit is vital when building all features great and small.

Renishaw's RenAM range of industrial AM machines supports all of these processing techniques, enabling both continuous and modulated energy delivery in the same build. In multi-laser machines, these techniques can even be deployed at the same time.

Parameters are open, so that users can choose the energy delivery strategy that best meets their needs, and take advantage of innovative processing techniques as they emerge.

Next steps

Visit www.renishaw.com/amguide for more education resources and to access downloadable versions of this and other LinkedIn articles by Renishaw authors.

Readers may also find the following articles useful:

• X marks the spot - how to find ideal parameters for your metal AM parts
• Viso profundum et latius - choosing the right melt track shape to suit your part
• Want to build accurate AM parts? No stress!