A good bending strategy makes life easier for downstream workers
October 12, 2012
Freeing a bending bottleneck can backfire if it simply creates a severe bottleneck downstream. Bending throughput may increase, but overall throughput--the metric that really matters--may not. That’s why the best bending automation considers not only the bending department but also the throughput of the entire plant.
Bending remains one of the most labor-intensive operations in contract metal fabrication. If a job doesn’t require welding, chances are that bending is the bottleneck, and many shop managers are looking to relieve it.
Automated bending can help. But if you’re exploring automation options, there’s a right place and a wrong place to start. The conventional practice is to start with press brake uptime. How often are your press brake rams moving up and down, actually forming good parts? Many might think it’s more than 50 percent of the time. On closer examination, though, most press brake departments are lucky if they reach 30 percent.
Questioning and analyzing current press brake uptime is vital, of course, but it isn’t necessarily the first question to ask. Say you analyze your bending operations, improve them, and get machine uptime to sky-high levels. The bottleneck is gone, and the bending areas are productive. That’s great. But here’s the catch: Your company may not be making more money because jobs aren’t shipping any faster. The bottleneck simply moved downstream.
You can chase the worst bottleneck, free it, and then chase another, and the next one, and the one after that. That approach has its merits, but it also has pitfalls. If you consider one operation in isolation, you often get unintended consequences. You may increase your bending capacity to an amazing extent, only to find a larger bottleneck in welding or painting. Conversely, if you invest in a large paint line or more welding cells, you may increase potential throughput, but the bending operation can’t feed enough parts downstream to take advantage of the new capacity. Most important, if you think of improving each department in isolation, you may miss out on some unique opportunities. Improvements in one area can also improve operations in other areas.
Try thinking of sheet metal cutting and bending as service operations. They “serve” internal customers downstream: hardware insertion, welding, grinding, painting, assembly, packaging, and shipping. Ideally, improvements upstream should make life easier for operations downstream.
When attempting to improve your bending department, it makes sense to consider not only the actual bending operation, but also the entire manufacturing process. Take a look at all of your internal customers—those downstream processes. How many different colors does your powder coat line run? What can help your welding, assembly, and shipping operations become more efficient? In short, what would help them ship more jobs in less time?
If you take this approach, make sure you look at the entire part flow. Say you simply concentrate on welding (your slowest operation), try to improve it, and use that optimized operation to set the drumbeat for everything up- and downstream. Part flow may become much smoother, especially after reducing batch sizes and streamlining your changeover practices with clearly marked tooling and setup sheets. But here again, you can fall into the trap of looking at a process in isolation. You assume that the parts welders receive, as well as the manner in which they receive them, can’t be changed.
What if the number of welding operations could be reduced? What if all parts for an assembly could be cut and fabricated at the same time, on the same machine, and transported on one cart to the assembly area? For a contract fabricator, that’s single-piece part flow at its best, and you may not be able to achieve it without rethinking upstream operations. This includes your bending department.
One high-mix, low-volume manufacturer punches and bends 3,000 components a day that together make up 50 kits. They then are sent to joining and assembly departments. These are extraordinarily complex assemblies, and to process them the company invested in a very high level of automation.
Sheets are transported from a tower to a punching center, which cuts a variety of parts on each nest. After punching and being separated from the nest, parts are sent automatically to an inline panel bender, which bends one part after another, even if each part is completely different. The panel bender has a changeover time that’s faster than the automation bringing components to the work envelope. A part is bent and then manipulators change out tools in seconds, just in time for the new part to arrive.
The panel bender can do this because of the concept behind the process, which is fundamentally different from how a press brake operates. In a press brake, the die opening determines the bend radius. You change your die opening width, and you change your bend. In a panel bender, the part is introduced to the blank holder tools so that the flange to be bent protrudes on the other side. Next, bending blades from above and below the workpiece fold the metal. The blades can interpolate in X and Y to follow the natural arc of the bending flange (if it moved in a straight line between the start and end point, it would scrape against the material).
These bending blades can be manipulated to bump-bend very complex geometries that quite often allow the part designer to assign multiple bends to a single part, thereby reducing the number of total pieces in an assembly. It may be next to impossible on a press brake to bump-bend a complex curve so it meets precisely with an adjoining flange, but a panel bender can make the form in less than a minute.
For most panel bending operations, the motion of the bending blade—not the shape of the tools—determines the final bend angle and radius. That’s why a panel bender can change over so quickly. In most instances, the only elements that change between jobs are the segmented blank holder tools so that they match the part’s required bend lengths (see Figure 1, Figure 2 and Figure 3).
A typical bending operation might employ several groupings of blank holder segments to form a finished part. An 8-inch flange being bent up between two adjacent, previously bent flanges requires a group of segments just wide enough to hold down the flange being bent, without interfering with those previously bent forms. To form each bend, the machine reads the program, and manipulators group certain segmented tools together to accommodate the part geometry and bend sequence. The entire changeover takes place in less than 8 seconds, while the manipulator is repositioning the part for the next bend. That means zero setup time.
Panel benders also gauge parts differently. A press brake operator slides a part against the backgauge behind the tooling. He makes the first bend and then may slide the previous bend against the backgauge to make a return flange. This means subsequent bends are gauged off of previous bends. If initial bends are off just a little, that error compounds itself in subsequent bends. The panel bender, on the other hand, references all bends from initial notches on the blank edges, not against previous bends. The workpiece remains under the machine’s control. The manipulator moves the part per the part program and does not need to gauge current bends on previous ones. That’s how it avoids those common tolerance stackup problems.
Panel benders can handle many part geometries, but not everything on the planet. The area behind the blank holder tools, where the blades perform the bending, is called the throat, and its depth limits the flange heights you can produce. A machine with a throat that’s 8 in. deep can’t produce a 15-in. flange. These days panel benders have throats deep enough to form flanges up to 10 in. high. To position the part for bending, the panel bender also needs that part to have a flat surface. In rare instances, certain forms made on the bottom of the sheet in the punch press, like stiffening ribs, can present problems. Still, design for manufacturability (DFM) can eliminate these problems. And such DFM efforts often are well worth it, considering the speed of the panel bending process.
Panel benders can handle sheet up to 0.125 in. thick, depending on the machine model. This is much thicker than was possible several years ago, but of course it’s nowhere near the thickness capabilities of a heavy-duty press brake.
Regarding part size, a panel bender’s blank holder tools can be only so narrow, and a manipulator needs to be able to hold the part. This limits how small parts can be. A decade ago most panel benders could handle parts down to 8 by 20 in. Now they can handle parts as small as 5 by 10 in., though they often can produce even narrower parts if the bender has an integrated shear. In this case, after the bending blades form the flanges, the shear blade rises to cut off the narrow part from the larger sheet. The cut part is sent down a chute, while the panel bender continues bending the remaining flanges.
For small and thick parts, press brakes remain the best option, but whether to automate depends on the application and volume. Robotic press brakes aren’t speed demons. A heavy-payload robot certainly can bend a large, thick plate faster than a human can. But for most applications, a robot actually works and performs tool changes at about the same pace as a press brake operator. The only difference is that the robot is in constant motion. If set up properly, a robot never stops working throughout a shift. Also, the more tool changes you have, and the more parts the robot cell processes, the greater the collection of end effectors a robot will need. This can make the operation complex.
A robotic press brake can be ideal for certain applications (see Figure 4). Say you have a relatively high-volume batch order for several large, heavy parts. The robot can work throughout the shift and produce more parts in less time—and it does something that’s arguably even more important. Workers need not break their backs lifting big parts anymore. You don’t necessarily need a robot for this. Lifting devices attached to the press brake bed can help. But the robotic setup gets workers away from the entire bending area. In this instance, the robotic press brake makes the operation both more productive and safer. That’s not a bad combination.
Many shops increase bending capacity by adding another press brake. It’s a straightforward upgrade, and a shop can increase its capacity incrementally. The investment meets expected demand, and all is well—as long as the company can find an operator for that additional press brake.
That last point has complicated the decision-making process for many. A panel bender still requires skill to operate, of course, but it increases the productive capacity of your current operators. It can produce as many parts as several conventional press brakes can, but it needs only one skilled operator—a big positive, considering how many great press brake operators are reaching retirement age. The decision to invest in panel bending may be more about making best use of available skill, and less about eliminating labor.
And labor availability isn’t the only consideration. By having a panel bender, you may be able to avoid hiring several operators, but if you integrate such capacity-increasing technology without thinking about those internal customers (downstream workers), you may not realize its full potential—say, if bent parts simply pile up by your welding or assembly cells. But what if a panel bender allows you to form complex parts that you couldn’t otherwise (at least within a reasonable cycle time), or makes welding and assembly much faster, because all parts arrive at the right time? These benefits reduce overall manufacturing time, and that is what really matters.
Consider that previously mentioned manufacturer with 3,000 pieces of various sizes and shapes grouped into 50 kits for assembly. In this application, the bender manipulates a large sheet, bends several flanges on one end, slices those flanges off with its integrated shear, and repeats this process until all narrow parts for a particular assembly are formed. It then forms the flanges for the remaining blank. Within a few minutes, the automated machine makes all the parts necessary for an extremely complex assembly. That kit then flows to downstream operations.
That said, was this investment necessary? At specified intervals, the fabrication department delivers stacks of kits, ready for assembly. And, yes, they might sit for some time as assemblers work to make the final product. To balance operations a little better, would it have been better perhaps to deliver only half the amount of work-in-process (WIP) for just a portion of the assembly? After all, a central tenet of lean manufacturing is to reduce excess inventory, and this includes that WIP between operations. And if the downstream department doesn’t need all components delivered at once to keep busy, was that advanced automated setup a worthwhile investment?
In this case, yes, and it’s because of another tenet that’s especially prevalent in high-product-mix manufacturing: variability. With thousands of different part numbers, subassemblies, and components, the company juggles a lot of work. And downstream, the variability increases even more. If, say, a large group of manual press brakes sends work in small batches downstream, it makes it easy for some small component to go missing. Many people may be performing many different tasks. And a missed part can halt production, force expedited jobs into the fabrication department, and wreak havoc on the production schedule.
Yes, transporting complete kits to the assembly department does increase WIP sitting before the assembly stations. But it reduces variability, because all parts arrive at once, and this allows assemblers to build the components as quickly as possible, without stopping to go on a part hunt. The large amount of WIP isn’t piled haphazardly, but instead organized in simple, consistent kits. Such organization is vital because every kit is different. And because all those parts are produced in one place by one extremely productive automated system, kits rarely if ever have missing parts. The result: More parts ship in less time.
This shows how automation helps serve internal customers downstream. If bending operations make internal customers’ jobs easier, they will ship more products in less time, which in turn will improve your bottom line. After all, if you aren’t improving that, your business really isn’t improving.