A new wrinkle in blacksmithing is hailed as the third wave of the industry’s digitization
Metamorphic manufacturing, also known as robotic blacksmithing, is poised to bring about faster time to market, less material waste, more available materials, less energy used and more control, proponents said.
Blacksmithing technically began around 1500 to 1000 B.C. with the advent of the Iron Age. But techniques for forging, smelting and hammering metal existed for thousands of years before that. Now, with the aid of robots, blacksmithing is moving into factories where it is being hailed as the third wave of digital manufacturing after CNC machining and additive manufacturing.
“Imagine if a machine can act like a blacksmith does, squeezing and bending metal into shape and doing this at temperatures and with deformation that actually improve the materials’ properties,” said Glenn S. Daehn, professor of metallurgical engineering at The Ohio State University and a leader in developing metamorphic manufacturing.
Metamorphic manufacturing is not yet ready for primetime. But adherents claim key benefits and said the technique shows great promise. The immediate benefit will come using the same materials but replacing manual processes with automated processes tailored for robotic systems.
Potential use cases include speeding up production of critical forgings that are at the heart of airplanes, ships, automobiles and power-plant equipment.
Closed-die forging, which metamorphic manufacturing may replace in some cases, is a lengthy, complicated process, and much of the critical die-making work often is carried out in Southeast Asia, Daehn said.
Meanwhile, forgings for turbine engine and airframe components are produced mainly in the U.S. or western Europe, primarily because of the quality needed to meet critical performance requirements, said Howard Sizek, U.S. Air Force Research Laboratory liaison to the Air Force Rapid Sustainment Office.
“Dies don’t ship across oceans easily, and miscommunications or errors can dramatically slow the process,” Daehn said. “We envision a future that moves from closed die forging to robotically controlled open die forging that uses simple tools incrementally. This can dramatically speed the time to make the first component.”
The most compelling use cases are those in which small numbers of large parts are urgently needed, he said.
Daehn gave this example: “Often, a new component is needed in an airplane or ship because the old component failed due to fatigue or corrosion, and you don’t have the dies to make the part. One or two hard-to-source parts can keep a plane grounded for many months.”
At this stage, faster time to market is the most appealing feature of metamorphic manufacturing, Formlogic CEO Paul Sutter said.
His company, which makes precision parts, started with automating CNC machining. The company is now exploring adding robotic blacksmithing based on Daehn’s research.
“We’re finding the biggest demand is faster lead times,” Sutter said, referencing the company’s exploratory calls with 130 manufacturers. “We’re focusing on short runs of critical parts. Our goal is a 10× reduction in lead times—from 30 days to three days and from two weeks to one day.”
“Companies spend a trillion dollars on fabricated metal parts every year,” he said. “When a part breaks, you can get a replacement part but it will be a week or more before a vendor can deliver that part. With metamorphic manufacturing, we can do next-day parts.”
Shorter lead times targeted
ROI will depend partly on a manufacturer’s cost of having a machine idle waiting for a replacement part, Sutter said.
“The value of an urgent part is many times the cost of an urgent part,” he said. “If you look at urgent parts, the price is about 3× higher to get something in two days instead of two weeks,. Even if they pay us 3× more, that’s a fraction of their savings. When a production line is down, manufacturers may be losing $50,000 an hour. Getting back up is important. They want to get that part fast. We can serve those customers.”
Other companies are predicting similar shorter lead times by saving on new tooling.
“Now, for a company to get a part made out of sheet metal, we need investments in energy-intensive tooling and heavy machinery,” Machina Lab CTO Babak Raeisinia said. His robotic manufacturing-as-a-service company was established in 2019.
“Time-wise, it could take eight to 20 weeks to get to the tooling, then trials. With that kind of lead time, it’s prohibitively difficult for any company to jump in, especially new ones. The cost of tooling could be $100,000 to $1 million, depending on the complexity and size of the part. With the robotic sheet metal forming solution we are providing, if someone gives us the CAD file, we can get to the part in a matter of a week.”
Wide use imagined for new tech
Use cases abound in aerospace, automotive, medical parts and the oil and gas sectors, Raeisinia and Sutter said.
For the U.S. Air Force, one potential benefit of metamorphic manufacturing is maintaining existing aircraft, Sizek said.
One potential application would be making large aircraft skins.
“The problem we have is that, because the systems are older, the tooling and dies to make the parts on may not be readily available,” he said. “The suppliers may have gone out of business. Tooling has a long lead time and adds a lot of process time. This would give us a chance to do things more quickly without all the tooling.”
Since the process—forging—remains the same, manufacturers are more likely to have confidence that it works, Sutter said.
“The really powerful thing is, robotic blacksmithing is a traditional process. If you’re doing robotic forging, it’s still forging,” he said. “What’s different is, you can turn it around faster, making it more flexible. At the end of the day, the metallurgy is the same, which is important to customers. They want what works: the tried and true.”
Said Sutter: “The whole idea of 3D printing has been, ‘Traditional processes are slow and complicated. Let’s invest in new processes like 3D printing.’ Our approach is, ‘The old processes all work but they’re slow and expensive.’
“Our focus is using traditional processes with better lead times based on autonomous manufacturing. Let’s automate them with machine learning and robots. Our goal is to accelerate delivery times for all kinds of operations.”
Global supply chain dependence noted
Metamorphic manufacturing can help the U.S. lessen dependence on global supply chains, Daehn said.
“In the United States, we have led the development of the machine tool industry,” he said. “More and more of that has shifted offshore. We have gotten in the position where if our global supply chain is shut down, we’re in trouble. If we aren’t able to make what we need, we put ourselves in a precarious position. Metamorphic manufacturing gives us the opportunity to put the next generation of equipment here in the United States—so we can make what we need in a time of crisis.”
Modern manufacturing supply chains are generally multi-faceted, involving many stakeholders.
Instead of multi-faceted supply chains, manufacturers using robotic manufacturing will be able to consolidate their entire process not only within one facility but within one metamorphic manufacturing system, Raeisinia said.
Having a long, spread out supply chain can contribute to security and reliability issues, he said. “If there are any issues, such as a geopolitical issue or natural disaster, it’s hard to get access to the materials or the products that you need. A single node will impact the entire supply chain.”
“When we talked with small manufacturers, they were impacted more with the COVID situation,” Raeisinia said. “A lot of their operations were outsourced to China. When the shops closed in China, their businesses were directly impacted.
“With a robotic system like ours in place, they would not be impacted at all. You can minimize the number of individuals that need to be present through built-in controls that minimize the need for human intervention. This way, you’re not dependent on having parts shipped from abroad. Production can go on.”
For example, when making a car body panel, stamping, trimming, joining and hemming processes are generally carried out at different stations within an automotive factory, he said.
“Concentrating multiple of these operations or nodes into one reduces lead time. You can make the part, hem the part, have multiple parts and join them—all within one robotic cell, Raeisinia said. “As opposed to multiple operations/nodes, your operation is focused on one single node.”
Part quality said to be improved
Part quality also is better, Sutter and Raeisinia said.
Sensors integrated into the metamorphic manufacturing system mean that quality control happens in real time while a part is being made, Raeisinia said.
“You don’t need a separate quality control system; the system is your quality control,” he said. “You don’t have to wait until the end of the process and then discard the parts that don’t meet your quality-control standards. You also have the luxury of adjusting the process in real time based on the feedback you get from the sensors.”
The part is fully validated before the first cut is made, Sutter said. “We can get higher quality and better precision because of repeatability. The key for us is we’re doing the whole thing in simulation first. We simulate everything to make sure it’s going to work, then we make it right the first time. That’s why it’s faster.”
Using conventional sheet metal manufacturing technologies, 30 percent of the sheet metal can go to waste or recycling, Raeisinia said.
“With metamorphic manufacturing, you have the ability to exert a higher degree of local control and real-time monitoring of the process,” he said. “Scrap waste comes down. For our metal forming process, we can use blanks that are generally 10 percent smaller than what is used in processes like stamping.”
Although there’s interest, the Air Force has not adopted metamorphic manufacturing technology—and likely won’t until more issues are addressed and solved, Sizek said.
“Computation is going to play a huge role in this,” he said. “You’re going to have to be able to predict not only the path you need to make the part but you have to have some prediction of the part’s form, fit and function as it comes off the tool: what may happen if you do heat treatment, what if you get distortion? You need to be able to tweak the part when it doesn’t quite perform as you thought it would perform. Understanding the material’s response to these processes and the material’s performance in use are all really important to put this technology into play.”
As with other recent manufacturing technology developments, robotic blacksmithing needs to be plug and play.
So far, that is not the case.
For example, the technical know-how needed just for the sensors can require a specialist to come in and train staff, Raeisinia said. “At times, that’s a hindrance,” he said. “Within the condensed time lines, you don’t have that luxury. The systems need to be usable out of the box.”
Another area that needs more attention is standards, Daehn said. “We should be spending more time with standards than we are. We’re not looking to certify an exact process but rather to certify a system. Right now, we don’t have a lot of resources to do this work. We are always looking for resources to accelerate technology adoption.”
More accurate modeling is needed to ensure manufacturers are getting the material the models indicate. “Model, but verify,” he said. “Every now and then, the model can miss something that is there in real life.”
Improvements are needed in both computer hardware and software so that calculations can be used in real time for control, Raeisinia said. “We know how to measure the temperature but we need to know how to measure temperature fast in a surgical, localized area,” he said, providing an example. “It’s not a science challenge; it’s an engineering challenge.”
“The algorithms are already out there,” he said. “The only work that remains is the integration of machine learning with the hardware and simulations, something that we are currently doing at Machina Labs.”