Digital Alloys’ Guide to Metal Additive Manufacturing – Part 8
Powder Bed Fusion (PBF)
May 6, 2019
Powder Bed Fusion (PBF) is the most mature and widely used metal additive manufacturing process. PBF, which prints metal using a laser or electron beam to melt lines in powder, is able to produce complex parts that can’t be produced by conventional manufacturing. However, PBF has challenges, including high material costs; slow speeds; laborious post-processing requirements; and restrictions on material compatibility. Due to these limitations, commercial use of PBF is mostly limited to rapid prototyping or high-value part designs (e.g., light weighting and assembly consolidation) that leverage the unique geometries that can be produced while withstanding the high costs of the technology.
This post provides an overview of PBF, digs deeper into its advantages and disadvantages, and highlights applications for which it is best suited. Providers of PBF are highlighted in the below chart.
PBF was the first metal 3D Printing process, invented over 20 years ago by a German company, EOS. They called it DMLS™- Direct Metal Laser Sintering. Many companies (see chart above) have since built PBF printers, with most coining their own names for the process: SLM™- Selective Laser Melting, DMLM™ – Direct Metal Laser Melting, LaserCUSING™, and EBM™ – Electron Beam Melting to name a few. If you find all these trade names overwhelming, you are not alone. This was a key motivation for ASTM to create the PBF category – which incorporates all of these similar technologies.
PBF definition – ASTM (F2792):
“Powdered material is selectively consolidated by melting it together using a heat source such as a laser or electron beam”
How Powder Bed Fusion Works
PBF printing starts by spreading a thin layer of metal powder (typically 20-120 microns) over a thick metal print plate. The powder layer is then selectively melted (printed) by a heat source: a laser (L-PBF) or an electron beam (E-PBF). This first printed layer bonds to the print plate. Then, a new layer of powder is spread on top of the printed layer – this is called re-coating. The newly spread layer of metal powder is then printed. The process repeats until the full 3D part has been built up within the powder bed.
Here are example system architectures for L-PBF and E-PBF printers:
A few key elements of PBF systems influence material performance, design, safety, and economics.
The Core Printing Process
As with any manufacturing method, an understanding of the core process provides insight into its fundamental strengths and challenges. In PBF, the process involves the heat source, how it is directed and controlled, and the powder melt dynamics.
The heat sources in PBF machines range considerably in power and resolution. The power of the lasers used are 200-500 watts and the electron-beams are 1,000-6,000 watts. The higher-powered E-PBF process can melt metal faster but at lower resolutions. PBF’s resolution is a direct function of the diameter of the beam (referred to as spot-size), layer thickness, and the largest particles present in the metal powder. L-PBF has a beam spot size of 20-100 microns while E-PBF is typically 100-200 microns. The layer thickness is usually in this same range and both layer thickness and spot size are roughly proportional to the largest particle size (more on this in the materials section, below). These resolution parameters determine the minimum printed feature size and surface roughness. See below comparison of “as-printed” surfaces. L-PBF has an Ra surface roughness of 5-10 microns, E-PBF 20-25 microns.
The speed of a metal AM process is critical to its production cost and throughput. PBF systems are slow, as their print speed is limited by the re-coating step and the rate at which the layer of powder can be melted from top to bottom. This is analogous to the time it takes to cook a piece of meat. If too much heat is applied then the outside will burn before the inside is cooked. In PBF, if the beam’s energy density is too high then the powder is vaporized instead of melted. L-PBF systems print most materials at speeds of 10-50 cc/hr, while E-PBF prints at 50-90 cc/hr (highly material dependent). This is more than an order of magnitude slower than Joule Printing™, select Wire DED processes, and the claimed speed of emerging binder jetting technologies. Recent L-PBF printers attempt to increase print speed by using multiple lasers (up to 4). This can incrementally decrease the time to print each layer, but it doesn’t speed up the slow re-coating step. Adding lasers significantly increases the machine cost and complexity as well.
Environment & Powder Handling
The fine metal powders utilized in PBF have significant quality and safety implications that affect how they must be stored, handled, and printed. Metal powders are sensitive to their environment due to their high surface area. This creates a quality risk because powders can absorb moisture, oxygen and other elements present in the air, affecting printability and the final material properties of the part. Metal powder is also a safety risk due to its flammability and potential for inhalation. Manufacturers who choose to use PBF must build and maintain expensive infrastructure, safety equipment, and careful procedures to manage these risks. The risks are exacerbated during printing because of the high heat loads, so PBF systems must utilize an inert gas or vacuum environment.
In PBF, thermal management is incredibly important for part quality. Tight thermal control is required to mitigate distortion and residual stresses caused by heat differentials within the build. One way to control temperature is to heat the build chamber. L-PBF systems use resistive heating to heat the print bed whereas E-PBF systems heat the build by completely scanning each layer with the e-beam before printing. L-PBF operates at bed temperatures of 200-500 C; E-PBF much higher at 500-1,000 C. The specific temperature is dependent on the material being printed. In L-PBF, the melted metal cools rapidly, resulting in small grain sizes and harder, more brittle material. This increases the potential for part distortion and residual stresses.
The high heat loads and long print times (sometimes over a week!) of PBF systems create a need for active cooling to regulate the system temperatures. The key concerns are that the laser, e-beam, or control systems overheat.
PBF process dynamics create specific metal powder requirements and limit which metals are compatible. Laser and E-beam systems both use fine metal powders produced through an expensive gas or plasma atomization process that outputs a wide range of particle sizes and shapes (aka morphology). PBF has tight requirements around the powder particle size and morphology. L-PBF is especially sensitive. The particle size required for L-PBF ranges between 15-60 microns, which drives up material costs. E-PBF uses larger particles, between 45-105 microns. These larger particle powders are also expensive, but cheaper than the tight range of particle sizes required in powders used by L-PBF. Metal powders are produced from wire so are inherently higher cost than the wire consumed by Joule Printing™ and DED processes. Powders are also more expensive to maintain for the safety and quality reasons listed earlier.
PBF alloy compatibility is limited by metallurgical considerations, the powder atomization process, and the rapid cooling rates (especially L-PBF). Weldability and absorption/reflectivity of the heat source by the powder are also important factors. That said, PBF still has a lot of material options. Those that are generally commercially available for PBF systems are listed in the below chart. There are more metals currently in development for PBF, including Invar, Tantalum, Iron, Niobium and other refractory metals.
PBF’s strongest value proposition is that it can unlock highly optimized, complex geometries not possible with conventional manufacturing. Despite this, it has many critical and nuanced part design rules that require experience to master.
As we outlined in our prior post on Design Rules. PBF design rules are driven by the fundamental physics of the process. They include 3 key elements:
1. Mechanics – How the feedstock, part, and energy are moved and controlled
2. Thermodynamics – Heat flow and thermal history
3. Metallurgy – The relationship between the printing process and the chemistry and crystal structure of the metal it produces.
One of PBF’s main mechanical design constraints stems from the re-coating process. The re-coater blade can impart significant forces that damage parts and cause a build to fail. This is referred to as “re-coater bump” and is a major consideration in part orientation and support design.
Another mechanically driven design element is the support structures required to build any surfaces greater than 45 degrees from vertical. Supports provide initiation points to grow overhanging surfaces and also have thermodynamics benefits (more on this below). The roughly 45-degree self-supporting angle rule is highlighted in the following image:
Large, uneven heat loads are applied during PBF (see our post on energy consumption) which result in significant residual thermal stresses within parts. These thermal stresses, if not accounted for, result in warping, cracking and/or delamination of a part.
Many of PBF’s design rules attempt to limit the effects of thermal stresses by constraining aspect ratios (the ratio between the length, width, and height of a part or feature) and wall thicknesses, adding radii on sharp corners, and optimizing part orientation within the build. Thermal stresses cannot be avoided entirely, so strong supports are always required to anchor certain features to the plate (or to the part itself). The supports also serve as a heat sink to pull energy away from the feature and avoid “burn through”, where the laser energy penetrates into the powder below the layer being printed. Burn through is shown in the prior image by the texture on the 30 degrees downward sloping surface.
Metallurgy is critical to metal manufacturing processes because of its large impact on the final material performance. The high cyclical heat loads in PBF can cause anisotropic grain structures (e.g. columnar grains) which degrade material performance. The PBF process is also susceptible to porosity issues. These are caused, in part, by suboptimal packing density and particle spatter during the printing process (you can see this in the video at the top of the post). Although these issues can be compensated for, in PBF it is very difficult to control the critical parameters needed for effective real-time closed-loop control. Therefore, printers need to be very tightly calibrated and print parameters finely tuned through iteration. Post-processing such as HIP-ing is often required for PBF to improve metallurgy issues such as grain structure and porosity.
PBF is most useful for intricate geometry parts smaller than a bowling ball, where the value of enhanced part design outweighs the high costs of PBF. The build envelope of most PBF machines is about a 250mm (10”) cube. Some recent, more expensive multi-laser L-PBF machines use a 400mm (16”) cube. Larger X and Y build envelope dimensions increase the number of parts that can be produced per build, but It is hard to use the full height of the PBF build envelope because parts must be anchored to the build plate. Although single parts can be produced to the PBF full build size, such parts are rare because residual stresses scale with part size, and because of the high cost of the process.
Manufacturing technology must be cost-effective. PBF is the most expensive category of metal AM. This is due mostly to its high equipment costs, slow process, and expensive materials. Another often overlooked cost factor is the infrastructure costs to integrate PBF into a facility. This is due to the previously mentioned dangers of powder metal processes. It is not uncommon that in addition to a $1M+ capital expense for a printer, there is another $1M+ of facility work needed before installing it. (Facility modification expenses were not included in the above pricing analysis).
PBF is a high-cost manufacturing process, so will only make sense for high-value applications where it can enable significant performance improvements.
Industries & Applications
Like most AM technologies, today PBF is primarily used for rapid prototyping but its biggest opportunity is in production of optimized, end-use parts. There are thousands of PBF printers in the field making prototype parts, where short turnaround times and rapid iteration are valuable. There are relatively few PBF printers being used for real production because of its speed, cost and complexity.
The production use cases are mostly found in Aerospace, Industrial, and Medical industries, where low volumes and high-value geometries can justify its high cost and complexity. Here are a couple of great production examples:
Aerospace – Fuel Nozzle
GE previously made this component from more than 20 conventionally manufactured parts, all assembled. The LEAP engine fuel nozzle can now be printed in one part, increasing its performance and reducing its weight. Assembly consolidation and light weighting are key to PBF applications in Aerospace and industrial.
Medical – Implants
PBF is a great fit for medical implants for two mains reasons. First, the digital workflow enables each implant geometry to be customized to the anatomy of the patient. Second, the rough surface and porosity that is inherent to the process is a benefit. It helps promote bone growth and integration.
PBF is a compelling solution for creating complex, valuable designs. However, it is slow, expensive and requires much expertise to implement successfully. PBF will continue to be used for very high-value parts but with its current economics, it addresses only a niche market. It is unlikely PBF’s applications expand significantly without large increases in print speed or reductions to machine and material costs. Given its two-decade maturity, it is unlikely that any of these cost factors will see dramatic improvement.
Newer metal AM technologies such as Joule Printing™, wire DED, and Binder Jetting solve many of the economic and scalability challenges of PBF. However, every solution has its pros and cons. These alternative solutions present some of their own limitations in design rules, part sizes and thicknesses, material compatibilities, and more. Just as in conventional manufacturing, where many different processes co-exist because of their strengths in different applications, this will continue to be the case for metal AM technologies too.
Please sign up to our mailing list to receive updates on future posts in our Guide to Metal Additive Manufacturing as we do a deep dive into each of these other Metal AM technologies:
We would like to acknowledge and thank the following individuals that contributed to this post:
John Barnes – The Barnes Group Advisors
Joris Peels – 3DPrint.com
Please check out other posts in our blog series:
- Part 1: Business Value
- Part 2: Application Criteria
- Part 3: Process Steps in the Workflow
- Part 4: Design Rules
- Part 5: Economics
- Part 6: Comparison of Additive Manufacturing & CNC Machining
- Part 7: Energy Consumption
VP Business Development
Digital Alloys is committed to providing the technology and expertise manufacturers need to use metal additive manufacturing in production — enabling them to save time, shrink costs, and produce valuable new product.