The benefits of any manufacturing equipment cannot be fully achieved by plugging in, booting up and leaving it to run. Intellect is needed for each activity in order to ensure that you’re satisfied during use and output. Additive manufacturing, or 3D printing, is no different.
The documented benefits of additive manufacturing include design freedom, streamlined product development and potential savings of cost and time. The benefits, however, can only be achieved with experience and in correlation with scientific, engineering and manufacturing rigor, said Gene Kulesha, Senior Director of Advanced Engineering at Onkos Surgical.
During a panel discussion at OMTEC 2019, Mr. Kulesha repeated “scientific, engineering and manufacturing” a half a dozen times, always ending the phrase with words like “rigor,” “discipline” or “principle.”
These strong, strategic words were echoed by his fellow panelists, additive manufacturing experts in their own right, to reinforce that industry is still on an additive learning curve and knowledge is to be gained in handling equipment and processes.
Among the panelists were Gordon Hunter, Ph.D., Principal Manager, Material Science at Smith+Nephew and integral player in the company’s adoption of additive, and Andy Christensen, Founder of Medical Modeling, which he sold to 3D Systems. Mr. Christensen’s recent orthopedic experience has led him to invest in and serve on the board of Precision ADM. Prior to his role at Onkos, Mr. Kulesha led teams at Stryker that researched, developed and transferred the company’s flagship additively-manufactured implants.
The OMTEC conversation was led by Chuck Hansford, Director of Advanced Materials Processing at Tecomet, who has nearly three decades of experience with additive manufacturing. Mr. Hansford selected these early adopters and staunch supporters of additive manufacturing in orthopedics as panelists in order to shed some of the hype around additive and paint a picture of what it takes to experience success with the technology.
In this article, we recap their comments on important subjects like design, validation and the future of additive in orthopedics. Of note, they started the discussion by dismissing any debate between the use of subtractive and additive manufacturing.
Mr. Kulesha was to the point and said, “If you can make something subtractively, there’s no reason to make it additively.”
Hansford: We talk a lot about what additive manufacturing can do and how it can change the manufacturing environment, but we don’t talk about issues or concerns. What do you see as the major issue to enter the market?
Christensen: The cost of the machinery isn’t trivial. You’re going to spend a million dollars just to get your feet wet. Historically, that would have held a few people back.
There’s a lot to know about becoming production capable. You can get a one-off part to look good without knowing a whole lot, but controlling a process as you would control machining or some other processes takes effort. That is another reason why additive is still in its infancy, but growing.
Hunter: Once you’ve got the equipment, it has tremendous potential. The reality is a lot more complicated.
You have a powder that you’re spreading. The characteristics of that powder are really critical. Finding powder manufacturers that can produce a consistent and appropriate powder for the type of system that you’re using is important, and there are not very many of them out there. That’s developing, but it’s a pretty small market at the moment.
Then, how that electron beam or laser beam interacts with that powder is also critical. Once it hits the surface and starts to melt, there is vaporization of metals. That vapor can now interact with the beam. There are angle effects on how that beam interacts with the powder, so the center of your build is going to have slightly different characteristics than the edges of your build.
How do you validate all of that; how do you properly sample product? You don’t want to destroy all of your product in testing; you’ve got to make some samples with the product to test it. Validating all of that is incredibly challenging if you are pushing the design envelope toward the properties of the material.
A lot is made about how you can recycle this powder until it’s all used up; that’s not true. There’s a lot of spatter produced in these processes; that spatter enters into the powder. You have to sieve it out, condition it and re-blend it. Powder can pick up oxygen, nitrogen, hydrogen; it can pick up all sorts of things that change its characteristics. Plastic changes its melting point and its molecular weight as you heat it over and over and over again. There are limitations to how many cycles you can use, and how you can recondition the powder and continue to use it over and over again to achieve the kind of cost benefits that are promoted for the technology.
Power fluctuation could also create vulnerability to the build. We are in Memphis; we have thunderstorms. Every thunderstorm that comes through blips our machines. We can put big conditioners on the equipment, and we do, but if one of those blips comes through and it’s not detected, there’s a blip in our part.
Those are the sorts of things you have to think through as you’re designing the limits of your process.
Kulesha: That sounds pretty bleak.
Hunter: It’s not bleak! You have to be realistic about the challenges. I don’t want to make it sound bleak. Every process I’ve ever worked on has advantages and limitations. This is just another one. We have to recognize what those limitations are and account for them if we’re going to make a product that’s going into a human being.
Kulesha: There are pitfalls with this technology, as there would be with any other. I think they can be overcome with engineering, science and discipline. You’ve got to be sober about it; you can’t believe in the hype. It takes a lot of work and a lot of effort.
Another pitfall is that the equipment is relatively new. When I got involved 15 years ago, there were just a handful of manufacturers, and their equipment really wasn’t worth much. It was tough to manage and tough to validate, but it’s gotten significantly better, exponentially better.
You’ve got to apply engineering, as I said, discipline and resources and scientific rigor to what you do, how you do it, how you set up the equipment. You have to respect the way that the beam interacts with the powder. That was an early lesson that I learned a long time ago. It’s achievable, as long as the right people are working on it.
The way the material prints and comes out of the machine is unique. This technology very rapidly melts and then quickly quenches in a crystal structure. It builds all kinds of materials that have grain structures that are not recognizable by a lot of metallurgists. This yields one of two results: it limits where you can use the parts off the machine, or it requires some post processing. Whether that is hot isostatic pressing (HIP) or heat-treating depends on what you’re trying to do. It depends on the end game, and the end product.
As Gordon suggested, I think it’s important to know that there are some applications out there, like primary hip stems, where additive is not yet strong enough in terms of fatigue to be used for a solid stem. However, it is strong enough to be used in augments and acetabular shells, and some knee components. It takes an understanding of the equipment and potential pitfalls to design and fabricate parts with additive that are going to be strong enough for your application.
Christensen: From a business-case standpoint, the products that have come out, to date, are all fairly high-value with advanced surfaces or complex porous structures. To get to the next level, the machines need to be more controllable, faster, have more up-time.
A key is getting to the point where the cost becomes more neutral. If you could make your product or something similar the same way in a traditional technique, you’d still find it to be more cost-advantageous to do that. Getting to cost-neutral opens up a whole part of the curve that additive could tackle.
Hansford: Gordon, from a material science standpoint, please enlighten us on the concerns of sub-core porosity. Can we get rid of it? If not, how do we work around it? How do we design around the grain structure differences and potential sub-core porosity of the component?
Hunter: We don’t have all of those answers yet. We have a lot of good questions, and we’re starting to deal with it, but the nature of a powder process is that it generates porosity when consolidated. Historically, we have used HIP, cold isostatic pressing (CIP) or something similar to reduce that porosity or eliminate it in castings. It seems to work well in castings.
Our experience in manufacturing is that these very small pores are difficult to eliminate with traditional HIP and CIP. I don’t have to worry about that with bar stock; I don’t have porosity in bar stock. The grain structure is something that’s well characterized, well known, and it’s consistent from one end of the bar lot to the other end of the lot.
When we’re dealing with additive manufacturing, there are a series of wells. You get this fish scale structure, it does look martensitic in some cases; I don’t recognize some of the structures. I don’t necessarily understand what to do with them all the time. I have to put them through physical tests to figure out what they’re going to do.
In our experience with 17-4 PH technology when we were making instruments, the standard H900 tempering heat treatment does not work on additive manufacturing products the same way it does on rod or cast product. I don’t know why. We’re not the only ones who have noticed this.
In fact, several universities have now started to come up with modifications to 17-4 PH to accommodate additive manufacturing. These are levels of complication. I don’t mean to make it sound bleak at all. I just mean we have to be realistic that this technology is very different than anything else we do. We need to step our way through it very carefully and understand it at each stage, because there are surprises that continue to come up. I’m still struggling with eliminating the porosity.
We need another generation. We’re on the steep learning curve of additive manufacturing. We’ve got some pretty good devices now, and good control on the powder. The next step is to develop advanced sensor systems and control systems that can interrogate the point where the beam is interacting with the metal or the plastic. Several of the innovators in this field are working on those things; they’re still pretty primitive. They’re going to get more sophisticated with time.
It’s going to take a while for this industry to settle down into a technology for broader applications than just porous products or instruments.
Hansford: Validation was mentioned. I know it’s a big buzz word in the industry. What does validation mean? What does that look like? Often when we talk about validating additive, everybody runs scared.
Kulesha: Process validation is obviously big. It takes time, effort and profound understanding of the technology. You’ve got to respect and scrutinize it, and approach it with, again, scientific rigor, engineering, discipline. With additive, you need to run design of experiments (DOE) on the equipment.
When you buy the equipment, the first period of time is discovery, followed by a little bit of research, then development. Experimental stress tests are run. Users should set lasers at various levels; same with the gas flow over the build chamber. Play with your wiper height to make sure that doesn’t affect anything.
Look at old powder. Look at new powder. All of these experimental parameters combine into a rather big DOE. Run it, let it go, see where your data ends up. See what your levers are that really do affect the process once it is running normally. Then you can proceed to process validation, your operational qualification where you know you set your worst-case parameters to the worst-case levels.
Run a big component, small component, old powder, new powder and see what happens. Only then do you have a validated process. The process should be run over time, letting the equipment go for months at or over hundreds or thousands of builds; that’s your performance qualification. See where it ends up, then track it.
In addition to all of that validation, it’s very wise to run on a per-build basis with witness coupons so that you can check the integrity and quality of the build in terms of chemistry, mechanical properties, incipient porosity and your design porosity.
It’s a big effort, but it really doesn’t differ from any other manufacturing technology that you have to validate, like plasma spray, thermal spray, casting, forging. The difference is that these technologies, especially casting, have been around for thousands of years. Forging has been around for maybe 200 years. All of that validation work has been done already, and people have forgotten about it. Additive is still a relatively new technology that has only reached more mature phases in the last five to 10 years. But this is not any different than what these other manufacturing technologies went through.
Additive is a more complex technology; there are more variables. As I said, there are lasers, gases, powder that varies over time. But I think, again, if you just apply discipline and rigor to it, it can be validated.
Hansford: I’ve read that additive is the future of manufacturing. Where does additive lie in the manufacturing space today?
Christensen: We all see those headlines. We’re still talking about parts, though, and I’m interested in talking about fatigue. A lot of what we’ve seen are non-fatigue sensitive parts in applications; those are low-hanging fruit. We are moving in the right direction. A lot of spine parts and some tibial trays have fatigue characteristics.
The number of orthopedic implant parts produced with additive is in the single digits. I don’t see it being 100%, and there is no reason to think it needs to be. It’s a tool that has certain purposes, and we’re figuring out where it’s good and where it’s not good. Could it impact 10%, 20%, 30% of parts? Potentially. As we get speed, costs go down and we gain reliability, it will become more on par with other technologies and you’ll be able to reduce time to market, tooling cost and inventory cost, potentially.
Kulesha: Porous structures made with additive are becoming ubiquitous, especially in the spine area. As those markets grow...if all cementless knees are going to be additively printed, there could be a significant percentage of additively-manufactured orthopedic components in the future.
In patient specific, customized products, additive is uniquely suited to make these odd shapes for oncology patients and secondary or tertiary revision patients. I think that’s going to be a lot better, and much faster, than machining or casting something. In my opinion, that’s a given.
Carolyn LaWell is ORTHOWORLD's Chief Content Officer.