How Do We Move from Metal Rapid Prototyping to Metal Additive Manufacturing?

Experts weigh in on their efforts to push metal 3D printing to widespread adoption.

There are many factors that will play into the ultimate shift of 3D printing as a technology for rapid prototyping to one of end-part manufacturing, but the one area of additive manufacturing (AM) that may stand in the way of mass adoption of 3D printing within the larger manufacturing supply chain is that of metal AM.

Metal AM has the ability to produce intricate, streamlined components with physical properties that can sometimes exceed those of parts manufactured by traditional means. Consequently, the technology has the potential to completely shift the way that we fabricate critical components. With it, we can create lightweight objects with unique geometries capable of decreasing material waste and energy consumption.

The paragon example now used to demonstrate the power that metal 3D printing can bring to manufacturers is the 3D-printed fuel nozzle for the LEAP jet engine, developed through a joint venture between GE and Snecma called CFM International. By redesigning the nozzle for AM, CFM was able to consolidate 18 different components into a single part. Moreover, the redesign has been estimated to reduce fuel costs and, therefore, CO2 emissions by 15 percent. Though the first LEAP engines are only just now being delivered to their first customer, they have already become GE Aviation’s best selling engine, with more than 6,000 confirmed orders from 20 countries, valued at more than $78 billion (U.S. list price). If this same potential can be achieved by redesigning legacy parts or creating entirely new designs for metal 3D printing, it’s possible to bring entirely new levels of innovation across industries.

CFM International’s 3D-printed fuel nozzle reduces part count from 18 to just one. (Image courtesy of GE.)

CFM International’s 3D-printed fuel nozzle reduces part count from 18 to just one. (Image courtesy of GE.)

If that’s the case, then why doesn’t every manufacturer go out and purchase a metal 3D printer? ENGINEERING.com spoke with a number of specialists in the field to determine what it is that is holding back metal 3D printing. More importantly, these experts are all working on overcoming these obstacles in their own ways, potentially speeding up the widespread adoption of metal AM worldwide.

 

Predictability and Repeatability of Metal 3D Printing

Before printing can even begin, the engineer or designer usually has to craft the CAD model specifically for the AM process that will be utilized. For instance, with powder bed processes, support structures must be incorporated into the design in order to prevent residual stress caused by the build up of subsequent layers from warping the end part. Therefore, the object must be oriented in a particular way within the print bed to minimize the number of supports and stress, which may severely limit the design options.

Preparing a model for printing may see several iterations of an object printed at various orientations before the part comes out of the machine within design specifications. In the case that a design needs to be shifted over to a different machine type, this may have to be performed all over again, given changes in size, energy type and more. This trial-and-error approach means wasted time, material and money. For this reason, 3DSIM, a spinoff from the University of Louisville, is currently beta testing software to allow for a simulation approach over a purely empirical approach to preparing objects for printing.

After roughly eight years of research, Brent Stucker, co-founder and CEO of 3DSIM, has begun commercializing physics-based simulation technology for metal 3D printing. At the moment, this commercialization comes in the form of two programs, exaSIM and FLEX. While exaSIM is directed more towards machine operators and generating a print preview for the optimal placement of support structures, FLEX is for the research and development of new materials or parts for metal 3D printing.

Both programs, however, rely on the same advanced mathematics and physical principles, which 3DSIM suggests would require 5.7 x 1018 years to solve on a 16-teraflop computer without their software. Chris Robinson, director of application engineering at 3DSIM, elaborated on how these programs operate, saying, “We input material parameters for a particular material. We go in and calculate a temperature prediction for what’s happening as the laser interacts with the powder at every scan vector at every layer of the part. Then, we can go in and calculate shrinkage, distortion, stress, microstructure, porosity, surface roughness and other calculations.” Robinson further explained that calculations are based off of the chemical makeup of a powder, at what point the material solidifies, how much laser energy it absorbs, how much it reflects, and in which direction the energy is reflected.

This information can then be used with exaSIM to generate the optimal orientation and support structures for a print to minimize errors. With FLEX, new material chemistries can be simulated, rather than created in small batches and tested in the physical world on actual machines. Validating parts is made that much easier as well, as they can be simulated before being manufactured.

Metal 3D Printing Data

3DSIM is tackling the issue of material properties in terms of the physical environment of the print chamber, but what if you don’t even know what material to choose? Or what machine to print something on? According to Zach Simkin, co-president of Senvol alongside Annie Wang, the lack of data regarding AM may prevent traditional manufacturers from adopting 3D printing in their operations.

As Simkin elaborated, “Organizations don’t have a full understanding of how materials are going to perform. They don’t have a full understanding of repeatability. There’s a lot of variability from machine to machine, from operator to operator.” Generating that data, then, is part of a larger barrier to understanding the diverse variables involved in AM.

Simkin and Wang started Senvol as a means of providing this data, establishing the Senvol Database with over 1,300 different machines and materials for AM that can be searched by a wide variety of parameters. Through a new partnership with material intelligence company Granta Design, users of Granta software can now match Senvol’s material data for 3D printing to Granta’s material data.

AM materials from the Senvol Database plotted with CES Selector. (Image courtesy of Granta.)

AM materials from the Senvol Database plotted with CES Selector. (Image courtesy of Granta.)

About the partnership, Simkin said, “The data that does exist in additive has sat in a vacuum to date. You can look at additive data, but it’s hard to compare to non-additive data, which is what it really comes down to because additive is really just one tool in the tool shed to produce parts. For the first time ever, through our partnership with Granta, engineers can start to compare additive data to material data from conventional processes to figure out where additive stacks up and where there might be advantages or disadvantages to using it for manufacturing.”

Metal 3D Printing Is Labor-Intensive

Metal AM is far from an automated process. The advanced training that machine operators may receive to handle such complicated equipment may be seen as an advantage when it comes to education and expanding human knowledge. However, the labor required to manage those machines could be holding some manufacturers back from adopting metal AM as a mass-production technology.

Concept Laser’s AM Factory of Tomorrow consists of different modules for printing a part and removing the part. (Image courtesy of Concept Laser.)

Concept Laser’s AM Factory of Tomorrow consists of different modules for printing a part and removing the part. (Image courtesy of Concept Laser.)

At the “formnext powered by TCT” show in October 2015, Concept Laser unveiled its plans for what it calls the AM Factory of Tomorrow, a modular platform for automating metal 3D printing as much as possible. In addition to the module that performs the actual printing, the company is developing a station for automatically loading metal powders into the print area and a station for removing the print from the bed and retrieving unused powder. These systems will go into production at the end of 2016, with delivery anticipated in early 2017.

A more extensive AM Factory of Tomorrow would see multiple areas set up, with powder carried either manually or robotically from storage to build job preparation stations and then to 3D printers and finally to part removal and post-processing stations. (Image courtesy of Concept Laser.)

A more extensive AM Factory of Tomorrow would see multiple areas set up, with powder carried either manually or robotically from storage to build job preparation stations and then to 3D printers and finally to part removal and post-processing stations. (Image courtesy of Concept Laser.)

This is only the beginning for the AM Factory of Tomorrow, however. To create a fully automated 3D printing factory, part manufacturers would conceivably store all of their powders in a specialized environment in their facility and make use of a robotic assistant that would automatically carry the powder from this area to the printing room, delivering the material to the powder handling machine.

Additionally, multiple 3D printers, powder handling stations and part removal modules would be lined up in such a way that excess powder might be delivered, through interior channels connecting all of these modules, to another machine for 3D printing. Other robotic assistants would also carry finished parts to post-processing stations, consisting of milling machines that would clean up the part. Concept Laser is currently working on introducing post-processing capabilities into a dedicated module next year.

Daniel Hund, head of marketing for Concept Laser, described the inspiration for this technology, “This technology came from prototyping. This means that you have standalone machines. There was no need for mass production in the beginning … It’s not really designed for mass production because it’s very time consuming. So, the first step was to get it faster, so we added more lasers to produce parts more quickly. The problem is, the part is made more quickly, but you still have to perform all of this manual labor. We then approached it from another perspective to perform all of these steps automatically.” Hund added, “We’re not thinking of the process itself, but the whole environment.”

Post-Print Quality of Metal 3D-Printed Parts

When metal AM parts are displayed at exhibitions and trade shows, passersby may wonder at the beautiful components and their complex shapes and moving parts. What may not be immediately evident is that these objects have gone through significant post-processing before they’ve hit the showroom floor.

A 3D-printed drill bit for the oil and energy market, produced by the DMP 320 3D printer from 3D Systems. (Image courtesy of the author.)

A 3D-printed drill bit for the oil and energy market, produced by the DMP 320 3D printer from 3D Systems. (Image courtesy of the author.)

Depending on the 3D printing method, metal 3D-printed parts are heat treated to relieve stresses and then cut off of a print platform and closely refined with a CNC machine, which cuts away the rough surface finish and clears away channels. Additional actions like electro-polishing or tumbling might be implemented to further improve the overall finish of the component.

Though companies like Matsuura and Sodick have introduced CNC capabilities into metal 3D printing systems to combine the benefits of both technologies and, ideally, streamline post-processing, a new firm out of Israel called XJet has developed a platform that may eliminate many of these additional steps altogether.

Rather than use a laser beam to fuse metal powders, XJet’s novel technology relies on jetting nanoparticle inks and a support material to build up metal parts layer by layer. A heating element passes over each layer of a print, bonding the metal together, and once a print is complete, the object is placed into a conventional furnace for final sintering. At this point, the support material is entirely removed.

NPJ is capable of 3D printing metal parts with layer thicknesses as fine as 1 micron. (Image courtesy of the author.)

NPJ is capable of 3D printing metal parts with layer thicknesses as fine as 1 micron. (Image courtesy of the author.)

Avi Cohen, markets development manager for XJet, explained in a recent interview with ENGINEERING.com that, because this support material burns out completely, part designers are not limited in the geometry of their designs. The aforementioned issues of orienting a part and placing proper support structures are no longer relevant. In turn, XJet’s NanoParticle Jetting (NPJ) process can produce moving metal parts in a single print. The fine resolution of the technology, due to the submicron size of the metal particles involved, also eliminates the need for further refinement of an object after printing.

In turn, Cohen believes it can achieve what metal 3D printer manufacturers are really after when addressing the problem of post-processing: the mechanical properties of a part. As Cohen elaborated, “The most important thing is the mechanical properties. We are in AM, in production. Not prototyping. Every metal part needs to be a metal part. You should get what you ordered. Is it repeatable? Is it reliable? Is it predictable? Is it acceptable? All of these things need to be addressed. Once you are able to provide a machine with predictable results, you will be able to produce parts with the desired mechanical properties.”

The company will begin taking orders for their machines at the beginning of 2017, and if the technology lives up to its promises, NPJ will likely produce metal parts with the finest resolution and surface finish in the industry.

Safety and Certification of Metal 3D Printing

The NPJ process may be capable of bypassing many of the post-processing steps required by other metal 3D printing technologies. Cohen, however, was also quick to highlight the safety issues associated with metal AM, which he sees as one of the barriers to widespread adoption.

With powder bed processes, machine operators are required to deal with reactive metal powders that can start fires or cause irritation to the skin, lungs or gastrointestinal system. For this reason, they wear gloves and respirators or handle the material through a sealed chamber. XJet believes their technology to be safer, as the nanoparticle ink is stored in cartridges so that a machine operator never has to handle reactive powders. 

Albeit, not every part manufacturer will be purchasing a system from XJet. To ensure that manufacturers, operators and facilities meet certain standards, the oldest company in the safety and certification game has entered the 3D printing space. Established in 1894, UL has made its logo synonymous with the safety certification of just about every modern piece of technology. In addition to providing testing, certification and consultations, UL also trains professionals on the safe use of AM equipment through the UL Additive Manufacturing Competency Center (AMCC) in Louisville, Ky.

Paul Bates, general manager of UL AMCC Services, described how they’ve carried over their track record of training and certification to the world of AM, saying, “One of the key aspects of certifying products is the question of who made it? And did they know what they were doing? When it comes to a part that is made through a welding process, the welder of that part will have a welding certification. And that’s something that follows that product through its manufacturing process.” Bates explained that, as UL establishes its presence in the 3D printing industry, they are applying a similar approach to AM. The operator of the equipment will be certified, having proven their skill and consistency in running a machine. That way, it will be possible to trace and track the quality of 3D-printed parts. 

Bates added that, given the relative newness of AM, not many individuals have been formally trained in the use of the technology in a way that is trackable or certified, which may make it more difficult to implement widespread adoption. For that reason, the company provides training programs at the UL AMCC in the United States and at the Global Additive Manufacturing Center of Excellence in Singapore.

As far as the machines are concerned, Bates said that they are sometimes certified, but that certification might be applied to a facility running the machines instead. As he explained, “In some cases, we have a service where we do facility assessment, where someone is developing a center and they may not have had one before and we ensure that they meet all of the local codes and requirements, particularly when they are very unique, as is the case with AM. We can come in and validate the facility, make sure it’s set up properly and that safety procedures are in place. We can then put a UL note on that facility so that when it’s time to launch a program there, they know that the facility is safe, that it’s good for the operator, it’s good for the products that are produced there, and it’s a safe environment. We really are focused on safety—not just for the finished parts, but for the people doing the work.”

Cost of Metal 3D Printers

The most obvious obstacle to the widespread adoption of metal 3D printing is cost. Every industrial metal 3D printer costs upwards of $100,000, and the materials with which they print are more expensive than metals typically used in manufacturing, due to the fact that they are often atomized into a fine dust with, ideally, spherical dimensions. Unless a business has the cash to invest in a technology that may be completely new to them, train or hire staff to operate the machines and set up the infrastructure to support these systems, they may be hesitant to jump into metal AM.

Metal 3D printer manufacturers are well aware of the cost factor associated with their technology. In order to accelerate the adoption of metal 3D printing, they’re taking various steps to make metal AM systems more attractive to invest in.

Britain’s Renishaw has developed a very straightforward approach to introducing their printers to businesses, yet it is one that seems less prevalent in the industry. Peter Kootstra, a service and support engineer for Renishaw, explained that the company leverages its worldwide Additive Manufacturing Solution Centres, which feature machines in-house.

As Kootstra said, “Companies can come to us, spend time with an applications engineer, look at a part and maybe redesign it specifically for AM. We can then let them rent some time on a machine. They can prove out the process, find some use out of metal AM and ultimately see the value in it. From there, it’s a lot easier to make a case for that initial investment.” The hands-on experience will provide evidence for how a new method for manufacturing a part might save money over the long run.

Concept Laser may still be developing its modules for the AM Factory of Tomorrow, but Additive Industries, a newcomer to the space, has already developed beta units for its highly automated MetalFAB1 3D printer and aims to deliver series production units at the end of the year. The company is attempting to make the upfront investment in metal AM more attractive by increasing the productivity of their machines.

Every step of the process, from introducing the powder to printing, cleaning and heat treating the part and finally removing it from the print bed has all been automated through individual modules. More than that, while all of this post-processing is taking place, a new print task can be initiated. Additive Industries CEO Daan Kersten suggested that the most expensive parts of the system, the lasers and the optical components, operate continuously, ensuring optimal productivity.

All of this combined allows for the system to pay for itself more quickly. As Kersten explained, “If you look at what total cost means, it’s material, which is hard to decrease because the material is part of the design, and we can bring it down somewhat if we broaden the material specifications. Labor is important. We’ve eliminated that totally. The machine hourly rate is determined by the productivity. If you look at the cost price per part with our machine, it is substantially lower than our competitors’ machines. That is how we contribute to the further expansion of this technology in manufacturing.”

These are just some of the obstacles to the widespread adoption of metal 3D printing. It’s obvious from these interviews, though, that those obstacles may not last for long. By this time next year, it may not be likely that we will have automated 3D printing robots that are safe and produce perfect parts straight out of the print bed. However, five years from now is another story altogether.