Additive manufacturing (AM) is changing how we think about making parts from metal. There are many challenges to the widespread adoption of AM fabrication, and these challenges create interesting research opportunities with the potential for real impacts on manufacturing.
My research focuses on laser powder-bed-fusion AM processing and the challenge of determining how metals react when exposed to a moving laser beam. The beam power and travel speed determine the amount of melted material (i.e., melt pool size). This determines the features we can produce on a part and how fast we can build parts.
Certain combinations of power and travel speed cannot be used, however, because they lead to parts that are not fully solid. Engineers need to understand these limits as they design and manufacture AM parts. My research is focused on a particular kind of defect referred to as “melt-pool balling” or “melt-pool bead-up.”
Bead-up of the melt pool is characterized by the melted material solidifying as a line of droplets instead of a continuous layer. This leads to parts that are not completely solid and therefore unusable. Bead-up tends to occur at high values of beam power and travel speed, so it limits how fast we can build parts. This limits AM productivity and the types of parts that can be built.
One way around this is to incorporate multiple lasers to enable faster build rates. However, more lasers mean more parts and potentially more maintenance issues. The maximum number of lasers will also be limited by space constraints. If we can find strategies to control bead-up, we could enable faster build rates. But to do this, we first need to understand when and why bead-up occurs.
The occurrence of bead-up is typically attributed to a Plateau-Rayleigh (P-R) instability, which predicts that a long column of fluid will tend to break up into small droplets. You can see this yourself if you slowly pour out a glass of water from a few feet up. Near the top, the water column is smooth, but it will break up as the water falls. There are many examples in nature of the P-R instability (e.g., the size of raindrops, the droplets of water that form on a spider web).
We can even engineer the effect to make inkjet printing possible. In the context of AM, the melt pools are hypothesized to act as a fluid column that breaks up into droplets. The mathematics of the P-R instability have been applied to predict when bead-up of melt pools will occur. However, this analysis is only an approximation because melt pools in AM are not isolated fluid columns.
We still need to have a better understanding of when the transition occurs and the factors that influence it. We also want to find ways to shift the transition to higher powers and travel speeds so that we can enable increased build rates.
We are currently studying effects of metal composition and AM process variables on the occurrence of bead-up. We are using stainless steels as test materials because we can control properties of the liquid by adding elements that cause a metal droplet to spread out. The presence of these elements tends to make the bead-up more frequent.
Investigations also involve how to best measure the transition from a normal melt pool to a beaded melt pool. Just recently, some collaborators obtained real-time video of bead-up via synchrotron radiography.
Finally, we are investigating new ways of controlling beam power and travel speed to make melt pools shorter and reduce the occurrence of bead-up. The increasing interest in AM methods has opened up many new options for manufacturing with metal. As we in the academic world research fundamental questions like applying the P-R theory to an AM melt pool, we hope to create solutions that will open up new applications for these developing technologies.