Robots emerge as an alternative to the traditional welding positioner

Welding positioners are ubiquitous in heavy-duty metal fabricating operations. They are incredibly useful for maneuvering large weldments into position to make life easier on welders.

Meanwhile, welding robots are becoming just as common, even in small shops. Costs have come down in recent years, and they are easier to program. Given that many metal fabricating companies can’t find welders interested in working for them, they also are looking at robotic welding cells to alleviate bottlenecks in the welding department.

Even with the move toward robots in the welding department, positioners are still a part of the mix. Welding positioners serve the same function and a more demanding taskmaster, as welding robots don’t need bathroom or lunch breaks.

With more robots in metal fabricating shops, it was only a matter of time before someone asked a simple question: What if a material handling robot replaced the welding positioner in the welding cell?

Acieta, a robotic systems integrator, has been discussing this type of automated welding cell setup at the Fabricators & Manufacturers Association’s Annual Meeting for the past couple of years. As fabricators have sought more efficiency in their welding operations, they have been much more receptive to new approaches to automation. The FABRICATOR spoke with Pete Rogers, vice president of operations for Acieta’s Council Bluff, Iowa, operations, about this trend and the use of material handling robots as positioners for weldments.

The FABRICATOR: Does Acieta have any metal fabricating customers currently using material handling robots as positioners?

Pete Rogers: I would say we have a handful of customers right now that are doing it. It’s becoming more accepted. The labor shortage is driving a lot of it, to be honest with you. This is much more a part of the discussion today than it was five years ago.

FAB: What drove this interest in the use of material handling robots as positioners?

Rogers: In the past, if you had a heavier part, you used something like a sky hook to position it on a 2-axis positioner that could tile or rotate, and you would use welding robots in conjunction with that. The biggest downfall with those cells, although there are probably thousands of them in the U.S., is most of the time they’re not running because they take 20 to 30 minutes to swap the part off of the fixture and get the new one on.

So what we’re doing is taking that same concept where you have the welding robots, but we have a person wheel up a part to an area where a larger robot can reach it. The robot grabs the part, places it in front of the welding robots, removes it from the cell, and grabs another part. So you have a third robot in the mix. With that third large positioning robot, you are basically eliminating the time it takes to load and unload the fixture. It also gives you a little bit more flexibility in the part positioning because you have a 6-axis robot to put the weldment anywhere it needs to be.

FAB: Does the addition of a robot for weldment positioning complicate programming for such a cell?

Rogers: I wouldn’t say it complicates programming necessarily. It adds more programming.

Obviously, from an initial investment point of view, that is going to be a little higher because there’s essentially a third robot to program. Once that cell is set up, as far as maintaining or continuing to run the system, I would say there isn’t additional difficulty in running it.

FAB: With more fabricating shops of all sizes investing in automated welding cells, do you think this makes them more open to this new twist on automation?

Rogers: I believe that would be the case. So you’re starting out with your simpler, standard weld cells. Somebody is loading and unloading parts out of that standard cell, and shop management is seeing the cell turn out parts. That’s why we’re seeing more interest in this and why we’re going to market with the systems like this.

It’s really driven by the labor shortage. Now instead of having these guys who have to tend the cell and who have to load and unload with cranes, you can use them elsewhere in the plant.

FAB: Is the use of a material handling robot as a positioner more for shops that work with large weldments or can it also work with small parts?

Rogers: Technically, you could use this approach with parts of all sizes. You may not get as big a bang for your buck or as big of a return on the smaller parts, however. That’s mainly because one of the factors in determining return on investment is limiting the downtime of that welding cell. So, for example, a standard weld cell like we’re talking about only requires someone to grab the small part by hand, throw it into the fixture, and watch the automatic clamps lock the part down. Even with manual clamps, it would take less than a minute.

The real time benefit comes with the heavy parts. That’s where you see the positioning robots providing a quick ROI.

Now we do have some welding cells in which material handling robots load stationary welding fixtures, and then a robot welds the part. The material handling robot then grabs the part out of the stationary fixture and moves it to a pallet or something. In that arena, you are able to see more of a financial benefit because you’re eliminating the travel from the welding cell to the next station.

FAB: Does a fabricating company gain flexibility with a welding cell setup that has welding robots and a material handling robot that is used as a positioner?

Rogers: Yes, especially when you look at more traditional welding cells that require manual loading and unloading of parts when you are using the 2-axis positioner. If you are changing parts, there’s a tooling changeover time. You have to pull the weld fixture out of there and put a different one in. In another robotic cell that comes to mind where we were doing two different parts, we used an automatic tool changer on that material handling robot. So we designed two different grippers that also acted as fixtures. Based on which part was going in the cell, the robot picked up the gripper/fixture and moved that part and was able to pick up the other gripper/fixture in a few seconds. There was extremely little changeover time from part number to part number.

So we’ll use tool changers in that instance if we’re doing multiple parts. What really becomes your constraint is the number of tools you can stack inside the cell and still have room to operate.

FAB: What are some of the unforeseen benefits for a fabricator adopting this type of automated welding setup?

Rogers: If a shop has multiple parts and fixtures in a welding cell, sometimes it doesn’t necessarily grasp the fact that the fixture changeover time is almost eliminated. That’s a big benefit. It allows the shop to become leaner.

That means you don’t have to build 50 of these parts now just to make it worth changing the fixture over, which might take an hour. You can build one or two, change the fixtures, build one or two, change back, and still get more production through. The ability to move closer to the Holy Grail of one-piece flow through the facility is closer to reality.

Manual tooling changeover, whether it’s welding or a press brake, really limits how frequently a fabricator will change over tooling from part to part. It also drives how many parts they batch before they change over.

So that’s one of the bigger benefits associated with this type of welding automation. Shops don’t have to have as much inventory, and they can reduce the work in progress.

We have one automated system that picks from the laser cutting machine, bending in a press brake, and then welding at the end. It is kind of a mini factory. The shop actually saved about 80 percent in terms of space when comparing this new automated setup with the old configuration of equipment. And it’s a 6,000-square-foot cell. It’s not a small footprint.

FAB: Are there any limitations for an automated cell like this?

Rogers: When we’re talking about the larger weldments and the heavy-duty robots moving these parts around, the weldments have to be tacked up fairly well, and you’ve got to have really good part consistency. If you were to put these large weldments on a stationary and beefier 1- or 2-axis positioner, it allows for a more inconsistent part. To maintain part consistency on these inconsistent parts, touch sensing or vision or laser tracking could be used, but that works only to a point. I can’t fix everything.

So you have limitations that are governed by the upstream processes. We’ll ask customers, “Are your parts consistent going into welding?”and they reply, “Yeah, they’re perfect coming out of welding.” Then we have to point over to the welding area and ask why the welder has two sledge hammers over in his cell. Obviously, you probably wouldn’t need those if the parts were consistent.

The customers that are willing to make the investment in improving their upstream processes really increase their throughput at the end of the day.

FAB: What type of individual should be put in charge of an automated welding cell like this? Is it an engineer who programs robots or perhaps a welder looking for a change?

Rogers: That’s a great question. What we really preach is finding an internal champion. Find that person who’s been working at the facility for 20 years and is maybe wanting to do something other than manual welding. Get them on board and put them in charge of operating the cell. The system will run itself. That person in charge of the cell, however, will be useful in adding another part to the weld cell mix and doing some of the troubleshooting should a problem arise. If you can get that operator training and have them involved with the robotic systems integrators who are doing some of the initial programming setup, that shows the operator how everything works.

So the operator is already familiar with the system before it hits the floor. The fabricators that are willing to make that investment find a huge amount of success.

FAB: Do you see a day when robotics have advanced enough that material handling robots can be used in the same cell with human welders, much like collaborative robots working side by side with humans on an assembly line?

Rogers: There’s no reason why you couldn’t use the collaborative robot technology to put a part in place for a welder. That way the welder isn’t being relied on to get the part in the correct position or having to find the right fixture to ensure the part is in the correct position.

I see that coming. What we need to see happen in this instance is the volume of collaborative robots must continue to climb, which will drive the price down. Right now they would be awfully expensive, but if you could eliminate 20 different positioners, maybe they aren’t that expensive.