Metropolitan Mechanical Contractors (MMC) provides services to design complete mechanical systems including HVAC, chiller and boiler systems, plumbing, sheet metal fabrication, and service and maintenance to a number of commercial facilities located in Minnesota, Iowa, North Dakota, South Dakota and Wisconsin. In business since 1963, the company has completed projects for the Mall of America, Target Center, Target Field and Xcel Energy Center. And that’s just in Minnesota.
When evaluating pipe welding automation systems for the fabrication shop, MMC was looking for a solution that would help achieve consistent, high-quality welds.
“The projects that we work on vary a lot,” says Corey Hagerty, pipefitting shop foreman at MMC. “Specific jobs typically require a percentage for X-ray and ultrasound testing. Sometimes a job is 100 percent X-ray, and sometimes 10 percent. What was important for us was to improve the quality and consistency of our welds.”
Path to automation
Novarc Technologies works with fabricators around the globe that are serving a number of industries but need an affordable path to automation. Novarc’s Spool Welding Robot (SWR) is a welding cobot (collaborative robot) designed specifically for pipe, small pressure vessels and other types of roll welding. It’s also specifically designed to provide pipe fabrication shops a smart solution that promotes safety, reduces costs, increases productivity and improves weld quality.
MMC purchased a Novarc SWR in 2022, and the move to automation has made a world of difference in weld quality and capacity. MMC appreciated the technical support, as well.
“The Novarc Academy was also a great help,” Hagerty says. “I walked through the training sessions myself, and I think that it took out a lot of the unknown and the pessimistic views that a lot of welders can have toward automation. The fact that they can override the SWR, interact with it and make adjustments on the fly is a huge benefit. I fully believe that’s the benefit of it over other semi-automatic processes.
“Before implementing the SWR, we typically had three to four guys welding, depending on the project, and we averaged from 60 to 80 factored diameter inches (FDI) a day,” Hagerty continues. “Currently, we’re doing 200 to 250 FDI a day on the SWR, and we even achieved 290 FDI in one day. The SWR has helped with our capacity on the larger projects, and that’s definitely where we have shined.”
Since bringing the SWR into the fabrication shop, MMC has been able to realize significant productivity gains: achieving 250 FDI per shift, increasing the shop’s pipe welding capacity, achieving zero percent repair rates and reducing the dependency on highly skilled welders.
But the SWR doesn’t abandon the human operator. Instead, it works alongside the welder, allowing those with less skill and experience to successfully take on a job that only senior welders could perform previously. And the results? Welds with greater precision, accuracy and speed, and a never before achieved balance of quality and productivity.
Besides improving weld quality, the SWR reduced MMC’s dependency on highly skilled welders due to the global labor shortage.
“We have an individual that was a pretty skilled welder in the field of stick welding that didn’t have a significant amount of experience in wire, but he just obtained a certification with wire, and we’ve been able to put him on the SWR,” Hagerty says. “Putting a junior-level welder on the SWR and achieving high-quality welds has been pretty seamless.”
Moving from manual welding to a collaborative process with the SWR also resulted in quality improvements. In North America, repair rates at pipe fabrication shops are typically around 3 to 5 percent. The SWR can bring this down to less than 1 percent by producing consistent, high-quality welds.
“The cost of repairs is typically between $800 to $1,000 per test, and with the SWR, we’ve achieved a zero percent failure rate,” Hagerty says.
Many leading players in pipe fabrication are looking to adopt robotic technologies that can incorporate AI systems in order to prepare for the future. To that end, Novarc has launched an AI vision system that uses algorithms to process vast amounts of data in a much more efficient way than any human can achieve. These algorithms are used to fully automate the pipe welding process. The NovEye vision system minimizes human interaction with the robot, and human errors and the associated costs are expected to decrease significantly.
Novarc has been working on these deep neural algorithms for the past several years. Now, with the videos and data collected by NovEye, temporal and spatial welding parameters are evaluated in real time, allowing adaptation to the environment and changes in the geometry, also in real time. These AI algorithms will help the SWR adjust on its own, much as a human welder or operator of the basic SWR would. With NovEye, the welder sets up the machine, moves the arm to the joint, hits the start button and walks away while the SWR uses the AI-powered vision system to adjust welding parameters on its own throughout the entire welding process.
And given the increasingly competitive nature of the manufacturing industry on the global level as well as the need for fabrication shops to deliver high quality and consistent welds, the shift to automation with the added benefit of AI is predicted to continue to accelerate.