What may be the biggest build volume in additive manufacturing, at least for metal parts, is being done by Sciaky Inc. using a technology that combines an electron beam welding gun with wirefeed additive layering. This direct manufacturing method can make parts as large as 19 ft x 4 ft x 4 ft.
The term "direct manufacturing" is often used to indicate an additive manufacturing process that makes net or near-net production-worthy parts, not prototypes. It's being used for several aerospace applications, in particular making metal parts for aircraft. For example, we've told you about the partnership between Airbus and South African aerostructure manufacturer Aerosud to develop 3D printing methods for large aircraft parts made of titanium. That technology is a form of selective laser sintering (SLS) called laser additive manufacturing (LAM) that forms large, complex structures from titanium powders. The two companies did not disclose the build volume of the machine they are developing.
A large, finished titanium structure built for an aircraft application using Sciaky's direct manufacturing technology that combines an electron beam welding gun with wirefeed additive layering. This method can make parts as large as 19 ft x 4 ft x 4 ft. (Source: Sciaky Inc.)
Sciaky's direct manufacturing method has a faster deposition rate than the very fine layer deposition of powder metal beds, which are commonly used in SLS. In Sciaky's process, a fully articulated, movable electron beam wirefeed welding gun deposits metal layers on a substrate plate, Kenn Lachenberg, the company's applications engineering manager, told us. Metals include titanium, tantalum, inconel, and stainless steel. The machine can deposit anywhere from 7 lb to 20 lb per hour, depending on the object's shape and material. The process does require a small amount of post-processing finished machining (watch a video of the process below). Lachenberg said:
We've incorporated an electron beam, similar to the one we've used for conventional electron beam welding, with wire or feedstock placement to add material, a process that's also available with conventional electron beam welding. In additive manufacturing, since you're layering metals you have long campaign times and runs, heavy vapor loads, and a higher heat environment. So we rebuilt the electron beam welding machine to handle those issues and give it features that are more feasible for additive manufacturing, such as a closed-loop control system and a faster traveling speed.
We've reported on a somewhat similar process that NASA developed for making parts as needed on the International Space Station. Called electron beam freeform fabrication (EBF3), it has a much smaller build volume. The system uses an electron beam gun and a dual-wire feed. On the ground, it's created parts for the F-35 Joint Strike Fighter's vertical tails.
The build volumes of SLS and powder metal techniques are limited by the size of the bed to smaller parts. Even laser sintering systems or other electron beam systems may only create a net part of 1 cubic foot, Lachenberg told us. Because Sciaky's automated gun travels throughout most of the length and width of the chamber, the area where it can deposit material is much less limited.
There's also no real limit on the size of the machine: the current one is 25 ft x 5 ft x 5 ft, built to fabricate the wing box of an F-14 fighter jet. A typical build rate with titanium is about 15 lb per hour. "We could increase travel speed and wire speed to provide a greater deposition rate, but a lot of work has been qualified at that parameter set," said Lachenberg. The company is considering increasing both of those speeds, as well as increasing wire diameter, to speed deposition. Sciaky has worked on the development of the process since the late 1990s, when it did feasibility studies with Lockheed Martin. The company is also working on R&D direct manufacturing projects with the US Air Force and the Department of Defense.
Sciaky's machine and process were displayed recently at the Pennsylvania State University Technology Showcase on Additive Manufacturing. The event was sponsored by the National Additive Manufacturing Innovation Institute (NAMII), launched last August, as well as by DARPA's Open Manufacturing Program, and the Center for Innovative Materials Processing through Direct Digital Deposition.
Very interesting technology. I'm glad they are developing this and increasing the speed that they can produce parts with additive manufacturing. However, it looks like they have a ways to go before it makes sense to use for production. From the video it appears that they have to machine the entire surface of the completed part before it can be used. It seems like using this method for prototyping and modeling makes sense. It should be good for development and even limited production runs. I'm sure we will all be watching this technology evolve with great anticipation.
Nite_Owl, thanks for your comments. Sciaky says it's working with Lockheed to develop this technology further, but it is being used in real aircraft production environments, not for prototypes. For some OEMs, the ability to make parts this big in one pass at a reasonable rate of speed outweighs the value of making much smaller parts at a faster speed and bolting them together.
I'm going to guess that the additive technology makes sense when the traditional "subtractive" CNC technology needs to remove more than about 50% of the source material. We've been watching NOVA's Battle of the XPlanes for a few years in our materials class and "Bulkhead 270" for Lockheed's F-35 JSF is a particularly intricate component made of titanium alloy.
The finished 300 lb part is whittled down from a 5-ton slab of the alloy (10,000 lbs). So grinding 24/7 for weeks to obtain a part that is only 3% of the starting slab sounds like this additive method would be a smart route if the resulting material properties are appropriate...
My thought was that it might make sense to either cast or drop forge the part. You could also cast smaller pieces and weld them together and then do the final machining, but hey that would be old school. Don't get me wrong, I love additive manufacturing and I can't wait for this technology to become mainstream. Just think what inventors can create with machines like that. Glad to hear that Lockheed is involved. Very exciting...
I guess it depends on the material. Titanium is pretty tough to machine compared to say aluminum, so it might make sense to remove up to 80% of the material with aluminum while titanium maybe only 40 to 50%. If they cast the part in section, since it's large, and welded the pieces together, then only minimal machining would be required.
I suspect that heat treating and stress relieving would be required for a lot of parts made using this method of additive manufacturing. With the exception of the bottom-most substrate, the completed part is basically made up of layers of weld. I would think warping and stress fractures might be a problem.
I was blown away by how fast that machine could weld, though. That electron beam welding gun is awesome!
Aerospce parts are high value and low volume. Raw material costs for Titanium alloys can be $35/pound and more so reducing the volume of chips made makes great economic sense. As WilliamWeaver points out, a 300# part can start from a 10,000# blank, making $339,500(est.) worth of titanium chips.
Forging such large titanium parts also has issues, requiring very specialized, high tonnage machines, of which there are not many, along with finished machining and past heat treatment(s).
These parts are also highly engineered with the fabrication processes needing to not only achieve the desired net shape, but also the desired strength and performance in critical areas. I wonder if this build up process allows more control of the finished material properties in specific areas of the finished part, since they are effectively working with smaller "building blocks" of material.
I think JimRW nailed it. The big deal isn't just how much material must be wasted before this technology becomes viable--it already *is* viable because of what material is being wasted and how much it costs: incredibly expensive titanium. That's why various methods of building parts from titanium are being used that don't include machining, or only include a small amount of post-processing, such as AM in different flavors, and powder metal (PM) methods. The second major factor is size of those parts, the fact they are structural and must meet high performance standards, and the difficulty forging & machining them. Good point about control--I don't recall that mentioned by the folks at Sciaky but it does seem intuitively obvious.
I don't know. It looks like they are removing 30 to 50% of the material they added in the final machining process. A lot of expensive chips on the floor, plus machining cost. With casting, even in sections, the waste would be very low and the final machining would be minimal. The sintered powdered metal flavor of AM would also waste less and might reduce or even eliminate final machining, but takes a lot more time.
The best thing about this technology is that you can go from design to development to testing to manufacturing very quickly. If demand out paces your capacity, you could shift to other higher volume production methods. This would be perfect for custom part production.
@NiteOwl_OvO: I agree with you that the best thing about this technology is the ability to prototype. You could make a part like this as a forging or as a casting, and get much closer to net shape, at a much lower cost, but you'd have to invest in tooling. You could also weld the part out of titanium plate. That wouldn't be cheap, but it might be chaper than 3D printing, at least for now.
NIST's new five-year strategic plan for its Material Measurement Laboratory lists additive manufacturing materials development as one of the main areas it will support by developing measurements, data, techniques, and models.
Focus on Fundamentals consists of 45-minute on-line classes that cover a host of technologies. You learn without leaving the comfort of your desk. All classes are taught by subject-matter experts and all are archived. So if you can't attend live, attend at your convenience.