A few key takeaways from FABTECH 2021 for tube and pipe

After a hiatus in 2020, FABTECH returned in 2021 and by nearly all accounts was a success. Although the attendance wasn’t as robust as it had been before the pandemic, more than enough attendees showed up ready to do business to make exhibiting worthwhile. Attendance exceeded 24,000 and “sold” signs were displayed on machines throughout the exhibit halls.

By the time FABTECH was held (Sept. 13-16), manufacturing and nearly every other industry had had some time to adjust to the fallout of the COVID-19 pandemic. However, adjust doesn’t mean recover. The shipping and trucking industries continue to struggle to keep up, some raw materials and core components are still in short supply, and the perennial problem—the lack of skilled labor in manufacturing—is more acute now than ever. Many manufacturers have given up on finding skilled labor and are willing to train unskilled labor, but the availability is still low and the retention rate is problematic.

A key message from FABTECH? Equipment manufacturers are getting more requests than ever for machines that require less setup and offer more automation, thereby reducing their reliance on labor, skilled or otherwise.

Going Digital

The digital revolution is nothing new. Manufacturing has been using CNC for decades. The earliest machines were a departure from their predecessors and led to a rethinking of the value chain. Machine setups were no longer lengthy or complex, and long part runs were suddenly a thing of the past.

Developments and advances in sensor technology, software, and storage capacity since that era have led to machines and entire systems that have far more capabilities and far greater capacities than they did at the beginning. For the most part, such changes were incremental, but the pandemic ushered in a new era.

“The pandemic taught everyone a lot, especially when one sick person leads to 25 absent employees,” said Robert St. Aubin, president Region Americas with Bystronic Inc., Hoffman Estates, Ill., speaking at FABTECH. “It led to astronomical growth in automation. For example, Bystronic sold more BySort parts sorting systems in the last year than in the entire history of the product.

“Some technologies have been available for years, but used very little,” he said, citing Microsoft Teams, a videoconferencing capability. The pandemic took hold early in 2020 and suddenly everyone was interested in Teams, Zoom, and other videoconferencing services. Likewise, start-to-finish automation and integration have been available for decades, but initially such systems were in the domain of large manufacturers. These days they are more affordable than ever, St. Aubin said, and cloud-based systems built on commonly available platforms such as Microsoft 365 contribute to ease of implementation.

Small shops have the potential to become medium-sized shops, of course. A shop’s journey depends a lot on how it plans for the future, and this is where the idea of scalability can facilitate growth.

“Every building has a unique size, shape, and layout,” said Brendon DiVincenzo, head of solutions for Bystronic. “This is why Bystronic’s storage tower system, ByTrans, is modular and scalable and can be combined with BySort for sorting nested parts.” Many fabricators understand that they have to think about today and well into the future, and they ask about scalability, he said.

Making Strides With Industry 4.0

After CNC, one of the next big things in the digital journey has been Industry 4.0. That term might sound like a fad or a buzzword, but it’s not that at all. Every piece of equipment on the shop floor has a status, whether it’s in use, idle, or stopped. Every length of tube or pipe likewise has a status, whether its raw material, work-in-process, or a finished good. Every production order has a status, whether it’s not yet started, working its way through the shop floor, or finished. Industry 4.0 is a system of sensors and software that captures all of this data, combines and filters and streamlines it, and generates a meaningful output that enables real-time decisions. This is a big bite for a small shop, but it’s here and it has been shown to work.

During a FABx Tech Talk, Jon Sobel, CEO/co-founder of San Francisco-based Sight Machine, a plant data and analytics firm, compared Industry 4.0 to Moneyball, the bestselling book and movie that described a statistical approach to managing baseball. Rather than focusing on each player’s individual strengths, the management of the Oakland Athletics looked at how each player contributed to the team as a whole. This perspective and its statistical underpinning allowed the small-budget (2002-era) team to compete against teams with much larger budgets.

Equipment manufacturer TRUMPF built a factory near Chicago to demonstrate the potential of Industry 4.0. It uses automated everything, and any customer can check on order status from anywhere in the world via mobile phone. Furthermore, the company’s integrated bending and cutting systems use a software system that runs the machines and continuously monitors the processes, always searching for inefficiencies.

Fewer Decisions, Fewer Actions, Fewer Errors

Progress in the digital revolution doesn’t have to come from a full-blown communication network that monitors every machine and coordinates all of the logistics. Often a seemingly small-scale technology can provide a substantial boost in productivity in a metalworking operation. It can be as simple as an RFID tag affixed to an upgraded component, one you can hold in your hand, combined with a corresponding machine and software that ties everything together.

Hypertherm, Hanover, N.H., unveiled its latest product, its Powermax SYNCseries, a new generation of Powermax models 65, 85, and 105. A key feature is a single-piece consumable torch, a cartridge that replaces the conventional five-piece torch. In examining how the consumable is assembled, installed, and used, the company found 20 opportunities for an error. The new consumable needs a quarter turn for installation, and that’s it. The operator no longer needs to match the machine’s settings to the consumable. An RFID tag synchronizes the consumable to the power supply.

The company says that the new consumable lasts about twice as long as its predecessor and provides a more consistent cut quality over its life. It removes the guesswork, so the operator doesn’t run the risk of replacing the torch too soon or too late. The SYNC system also has a data-logging feature that assists with troubleshooting, helping to uncover problems such as an incorrect air pressure setting.

Robots and Cobots

Data analysis isn’t just for conventional machines. It’s a handy tool for robots as well, said Joshua Leath, senior product manager for Yaskawa Motoman, Miamisburg, Ohio.

The company’s Cockpit interface, displayed at FABTECH, gathers streams of data from the power supply, the torch, and the robot, and turns these streams of data into a unified, sensible output that can be used for troubleshooting and traceability, he said. In some cases, it doesn’t merely assist with troubleshooting, but can actually identify erroneous welds.

Robots aren’t only used for machine tending and welding, he added. Another function for a robot is to use it as a weld positioner. This might sound like overkill, but it’s an extremely versatile, programmable, multiaxis positioner. Although it’s not inexpensive, the cost is comparable to that of some conventional positioners, Leath said.

A more recent development is a robot built to work alongside people. Sensors, programming, and failsafe development strategies eliminate the need for safety barriers, facilitating a collaborative role for this variety of robot: the cobot. “After 40 years of incremental changes in robotic welding technology, the cobot was a breakthrough,” said Mette McCall, speaking on behalf of Denmark-based Universal Robots. Also, the software has been getting easier to use.

“Initially it was faster to weld than to program a robot,” she said. Eventually it was faster to write the program, but these days, programming is a breeze.

“Some software is welding-specific, including proper welding terms and symbols,” she said.

However, the days of programming a welding robot or cobot might be numbered. In many cases, a machine operator merely puts the cobot into a teaching mode and moves the torch by hand. After the operator guides the torch from the starting point along the weld path to the end point, the cobot is ready to weld.

“The welder sets up the job and monitors the process, but he doesn’t have to do any programming,” said McCall. Mobility adds another layer of versatility to an already flexible machine, she said. Some small robots and cobots can be mounted on wheels so they can be moved from location to location as needed.

For pipe welding applications, Novarc Technologies, North Vancouver, B.C., introduced its Spool Welding Robot (SWR) in 2016 and displayed it at FABTECH. It was built around a cobot that was designed to be used specifically for pipe fixtured on a rotating chuck. The cobot, armed with a welding gun, moves up and down and forward and back as it performs the weld. Typically, a welder oversees the operation, adjusting the cobot during the welding process if he sees a problem or if the cobot has to contend with a subpar or inconsistent weld preparation, such as a bevel angle that falls outside of expected tolerance windows. The SWR can weld flanges, T-joints, and elbows.

This type of pipe welding traditionally has been the domain of skilled welders. The high-mix/low-volume work called for talented welders who could smoothly go from one job to the next and handle the variances in pipe prep and fit-up. The SWR is Novarc’s answer to the question many are asking: “Where will our company find the next skilled welder?”

“It helps to make the junior welder much more productive,” said Soroush Karimzadeh, CEO, Novarc Technologies. “With the SWR, a junior welder can weld as much as three or four highly skilled welders over an eight-hour time span—and with much better quality.”

Artificial Intelligence

The term artificial intelligence (AI) sounds like something out of a sci-fi book, and indeed the phrase has been around a long time. Research in AI goes back to the 1950s, and the field got a big boost in 1996 when IBM’s Deep Blue computer won a game of chess against then-reigning world champion Garry Kasparov. Although AI had been growing in fits and starts until that time, mirroring valleys and peaks in funding trends, the funding and growth have been smoother over the last 25 years or so, and researchers have made a lot of progress in manufacturing applications. Some examples were on full display at FABTECH.

For example, Path Robotics, Columbus, Ohio, uses AI to get around programming. The company has put together a welding booth outfitted with a 3D vision system, a robot equipped with a welding torch, and some very capable software. After a part is tacked together and fixtured in the booth, the software compares the assembly’s CAD file to the view from the vision system to find the exact location of the assembly in the booth; the software then determines the weld path and the robot gets to work.

“We’re effectively snapping those weld seams onto the actual weld topography of the part,” said Paul Boulware, director of applications engineering.

Novarc offers another example of AI in action with the neural algorithms it developed to help the SWR learn and adjust on its own. The result is NovEye, an AI-powered vision system that advances the SWR’s automated welding capabilities. The way it handles tack welds is an example. It’s not uncommon for any cobot-based welding system to weld along a root and go over tack welds, but the system operator has to alter the parameters as the cobot approaches the tack weld so the entire weld seam has complete fusion. The NovEye lessens that operator engagement because the vision system sees the tack weld and adjusts welding parameters on its own.

“The other part of this is the consistency,” Karimzadeh said. “We noticed that with some operators, the inputs could be different depending on who was running the machine or what the skill level of the operator was. The NovEye means a lot more consistency. Our algorithm is going to react the same way and in the most optimal way every single time.”

The strength of Novarc’s ability to learn is leveraged by voluntary access to some of Novarc’s customers’ data.

“We collect the data sets and use them to train the algorithm,” Karimzadeh said. Optimizing the algorithm helps to improve both the output of the robot and the reliability of welds.

These applications of AI might not make the news headlines like Big Blue did in 1996, but they sure are more practical on the shop floor that the ability to play chess at the world champion level.