Additive manufacturing (AM), also known as 3D printing, is quite literally one of the most innovative technologies revolutionizing manufacturing today, in terms of both industry “buzz” and thermal properties. Unlike subtractive manufacturing methods such as machining, the growing range of AM technologies creates components directly from a computer model, adding material only where needed. Wohlers Associates, a leading independent consulting firm focused on these technologies, is forecasting that the value of the worldwide AM market will grow to more than $10.8 billion by 2021, up from just $2.2 billion in 2012. That rapid escalation, however, isn't the result of hobbyists buying desktop 3D printers that cost a few hundred dollars.

Binder jet 3D printing techniques are often used to create tough, wear-resistant metal parts like these. (Credit: FLIR)

Growing numbers of high-tech organizations are pioneering AM technologies to use in applications ranging from product development to specialized manufacturing in fields such as architectural design, aerospace components, and medical implants. NASA has even sent two different 3D printers, designed to operate in zero-G, to the International Space Station.

Additive manufacturing allows for far greater design flexibility, decreased energy consumption, and a faster time to market. AM parts, however, can be subject to quality issues, thermal stresses, and distortions that are difficult to diagnose. Studying the process and its thermal characteristics with an infrared (IR) camera can help manufacturers make in-situ corrections needed to improve their product quality and avoid disruptions to production.

Taking Control of the Process

Today, the variety of AM materials being experimented with is expanding rapidly to include substances as varied as cement, carbon-fiber reinforced thermoplastics, and living cells. For now, most 3D printers are based on either metal- or polymer-deposition technology. The U.S. Department of Energy's Manufacturing Demonstration Facility (MDF) at Oak Ridge National Laboratory (ORNL) in Knoxville, Tennessee is one of the world's leading centers of AM research. Their work in the development of new and improved AM processes requires the ability to monitor these processes closely, evaluate new materials, and understand the reasons for process failures. The quality of the parts produced can vary widely, often depending upon the interaction between the manufacturing process characteristics and parameter settings. Additively manufactured parts are subject to a variety of quality issues. Too often, process parameters are set using trial-and-error techniques; such methods take time, money, and can be highly subjective and material-specific.

The ability to monitor processing equipment, materials, and in-process part temperatures quickly and accurately is crucial to AM research. Typical contact forms of temperature measurement, including thermocouples, resistance temperature detectors, and thermistors, would be difficult or impossible to use effectively. In contrast, IR cameras offer a high-speed, non-contact form of temperature measurement, and provide the data accuracy necessary to correlate process parameters and in-process temperature data with measures of finished part quality. Instead of using trial and error to set process parameters, MDF researchers employ infrared cameras to monitor the effect of changes to printer settings or to the materials used. The cameras help to identify the source of quality issues, such as part porosity, delamination, and thermal stresses, to name just a few.

MDF's ‘Customers’

Researchers at the MDF work closely and share information with dozens of AM equipment developers, manufacturers, and end-users, as well as government agencies like NASA and the U.S. Air Force, to explore new approaches and enhancements to various 3D printing processes. MDF researcher Ralph Dinwiddie has cooperated with many different 3D printer companies.

“With one of these companies, we were measuring the extrusion temperatures of experimental materials they were developing, as well as the temperature of the previous layer to understand the optimum temperature to promote adhesion,” said Dinwiddie.

Another recent partnership included a joint development project to create a system capable of printing polymer components up to 10 times larger than previously producible, and at speeds 200 to 500 times faster than before. The system incorporates the design and technology from the partner's laser platform, including the machine frame, motion system, and control, with an extruder and feeding system developed by the MDF. The technology prints objects as large as the chassis and skin of a car, or sections of a small house.

Evolving Materials, Evolving Techniques

A FLIR SC-8200 IR camera mounted on the ARCAM A2 electron beam system at the Manufacturing Demonstration Facility. (Credit: FLIR)

Although researchers at the MDF are currently experimenting with multiple AM technologies, Dinwiddie is concentrating on four of them:

  • Fused Deposition Modeling (FDM) uses a heated nozzle to melt and deposit a thin filament of thermoplastic material into a two-dimensional pattern. After each layer is complete, the build platform sinks and another layer is applied until the object is complete.
  • Large-Scale Polymer Deposition heats polymer pellets to near-molten temperatures, then extrudes them layer by layer onto an out-of-the-oven build platform.
  • Laser-Blown Powder Deposition uses inert gas to spray metal powder into a melt pool created by a high-power laser beam.
  • Electron Beam Melting manufactures parts by melting metal powder in successive layers that are bound together using a computer-controlled electron beam.

Dinwiddie works with a wide variety of plastic and metallic materials in his research, including combinations of high-strength thermoplastics and carbon fibers, a nickel alloy, titanium, and others. Each melts at a different temperature and interacts with previous layers in different ways, so monitoring and controlling temperatures accurately at every stage of production is essential.

Tools for Temperature Monitoring

Close-up of a FLIR SC-8200 IR camera mounted on the ARCAM A2 electron beam system. (Credit: FLIR)

Even before focusing on AM research efforts, Dinwiddie used his first highperformance IR camera to help continuous fiber ceramic composites developers to understand how the materials conducted heat. The effort allowed the manufacturers to optimize their manufacturing processes.

“We're often using multiple cameras to acquire temperature information with various points on the printer, like the extruder tip, the heated chamber that encloses the printer, the extruded material itself, and the previous layer of extruded material,” Dinwiddie said. “We need the ability to make and record temperatures at high speeds and to calibrate these cameras with our own black-body source.”

Dinwiddie says “windowing,” which reads out a smaller subgroup of pixels on the IR detector, is also vital to research. By recording a smaller subarea of pixels on the detector, Dinwiddie reduces the number of pixels per frame. Windowing allows the camera to send out more frames per second, achieving faster frame rates.

“My work also demands a lot of flexibility in terms of lenses,” Dinwiddie said. “For example, I've used telephoto lenses, wide angle lenses, standard 50-millimeter lenses, as well as telescopes, microscope lenses, and a macro lens. I've also used extension rings so I can focus much closer than I'd normally be able to do.”

In his AM research, Dinwiddie uses both high-speed, midwave infrared (MWIR) cooled cameras, and lower-resolution, longwave infrared (LWIR) uncooled cameras. The differing capabilities of the cameras make each device particularly suitable for specific sets of tasks.

The uncooled cameras, for example, are compact and can be mounted easily on a polymer 3D printer to monitor the temperature of the extruder tip and/or the extruded material. Their thermal sensitivity of less than 50 mK allows the cameras to distinguish between minor variations in temperature. For tasks in which high-speed temperature measurements are crucial, the cooled cameras’ windowing capability enables the faster frame rates necessary.

Although each new AM system or material presents its own set of characterization challenges, some common tasks include real-time detection of porosity while the parts are being printed with the e-beam systems. A pore typically appears on a thermal image as a dark spot. The cameras have also proven especially useful for “dialing-in” the correct processing parameters needed to prevent the formation of pores when working with new metal powder formulations. For polymer 3D systems, Dinwiddie often measures the temperature of each layer of a part as it is applied, in order to study the effect of the temperature of a previously deposited layer on the bond strength between layers.

The cameras also measure the temperature of the build chamber and monitor the thermal gradients in the part itself as it cools. With many polymer materials, uniform cooling helps reduce distortion in the finished part, which is why some 3D printers have a heated build chamber to slow the cooling of a part's outside edges.

Correlating Temperature with Quality

Once parts are completed and cooled, MDF researchers typically characterize their quality, analyzing their microstructure, strength, residual stresses, composition, thermal conductivity, etc., using X-ray tomography and other techniques. The data is correlated with the temperature data acquired from the IR cameras to gauge the effect of process variations on finished product quality.

IR cameras have proven their value in advancing a wide range of emerging AM technologies, giving materials scientists the accurate results they need to fine-tune materials, equipment, and process parameters. Such refining of the AM process will help the industry meet its expected rapid growth in the coming years.

This article was written by Chris Bainter, Americas Business Development Director — R&D/Science Segment, FLIR Systems, Inc. (Wilsonville, OR). For more information, Click Here .