Our Sites

Better testing leads to better roll forming

A new bend test helps a roll former to predict and prevent a common defect

Anyone who works in precision sheet metal must deal with the reality of residual stress. One batch of sheet metal may cut perfectly under a laser, while another may bow slightly (or sometimes significantly) after the laser completes cutting and releases the piece from the nest. That bowing probably comes from residual stress. The same thing goes for bending sheet metal on the press brake. One batch may bend perfectly over a given die width; then in another, the radius and angle may be off by a few thousandths.

Roll formers live with residual stress too, and it becomes especially problematic when they roll form certain shapes. The residual stress can come from the previous rolling processes at the mill, including cold rolling. This even includes skin rolling, which involves less thickness reduction than other cold-roll processes. Full-hard rolling reduces the material thickness by 50 percent, while skin rolling reduces the thickness by less than 1 percent. That small amount of reduction means that skin-rolled stock can be ideal for subsequent cold working, like the bending found in roll forming.

But as most people involved in metal forming know, residual stress remains in such material. It’s a fact of life that complicates forming processes the world over, including Down Under, at Australian Rollforming Manufacturers (ARM) in Dandenong, near Melbourne. There technicians experienced a familiar problem: oil canning, regular dimples or waviness in the web, when working with certain coils to roll form sections with wide webs in between two flanges, a common profile for metal roofing and other roll forming applications (see Figure 1). The defect is named for the ripples seen in an oil can (back when car oil came in a can instead of a plastic bottle).

Engineers at Deakin University in Waurn Ponds near Geelong (across the bay from Melbourne proper), roll former ARM and South Australia machine builder FSK Engineering developed a new test to predict which materials would likely exhibit this oil-canning defect and which materials wouldn’t. The test method draws from some age-old practices in metal forming.

“The tester is basically a simple device that does what metal formers sometimes do to get a feel for the material,” said Matthias Weiss, PhD, the head of the roll forming research group at Deakin University. “They cut a strip, bend it between their thumbs, and feel the resistance.”

The Conundrum

Technicians at ARM were roll forming 13.1-in.-wide drawing steel between 0.045 and 0.047 in. thick. Altogether, forming took 12 passes to complete. The curved part of the profiles were formed aggressively in three passes, and the final passes formed two bends to turn the flanges upward. Emerging from the final pass, the roll formed sections tended to bow upward (see Figure 2).

To compensate, the team added a special tool to bend the section downward as it emerged from the final pass. By pushing the section downward ever so slightly, the tooling adjustment induced a little expansion in the flange and a little compression in the web (see Figure 3).

Problem solved, right? Not so fast. The team ran four test pieces from different batches of material. The first three emerged from the roll forming machine with no bowing whatsoever. The last, though, emerged with that familiar rippling and waviness—classic effects of oil canning.

This happened, they assumed, because the fourth batch of material had slightly different material characteristics, a common fact of life in metal forming and fabrication. But was there a way to predict this behavior without going through the expense of running a test piece?

The Test

As Weiss explained, the problem with conventional tensile tests is that they indicate average stresses over a cross section, though those stresses can vary significantly in part thanks to skin rolling or the effects from coiling and the resulting coil set. It would be ideal to know the actual range, not the average, for the yield stress—that is, the point at which the formed metal “yields” and doesn’t return to its original shape—and the variation of forming behavior across the thickness.

Figure 1
Note the subtle ripples in Coil B. These ripples are characteristic of oil canning, a common defect in roll forming.

(Note: In his research, Weiss and his colleagues use the term yield stress, which is related to yield strength and yield point. Yield stress is technically the amount of stress applied on a sample to deform it. The yield strength or yield point is a quantification of those yield stresses.)

So from here, the team developed a new device, what they call a “free bending tester,” which basically emulates the age-old practice of an expert roll forming tech bending a small strip of sheet metal (if it’s thin enough, of course) with his thumbs. Among other things, the device measures moment resistance, which is the resistance to forming during bending.

Think of a metal former pushing a thin strip with his thumbs, feeling the resistance, and then knowing through experience how that resistance will affect the shape of the final roll formed product. Now replace those thumbs with two clamps, the thumb’s pushing force with stepper motors, and the metal former’s experience with special encoders and software, and you get the testing system shown in Figure 4.

The team cut rectangular samples and clamped them onto the two arms, which then bent the material. The tester incorporates a special encoder embedded in a stepper motor that drives the arms. That encoder measures the amount of bending and the curvature, and a load cell measures the moment. The data is fed into a computer, where software interprets the data (see Figure 5).

The Results

Analyzing the moment when just enough force is applied to deform the material permanently, known technically as the elastic-plastic transition, the team found that a number of different material parameters affect exactly how this transition takes place, including whether the bending took place with or against the coil set.

In fact, when bent with the coil set, the material that experienced the oil canning (Material B) had a yield stress of nearly 350 MPa, more than 100 MPa higher than the other materials (see Figure 6).

“This indicated significant aging in the material as well as the effect of coil set; these were not picked up in standard tensile tests,” Weiss said. “While aging is a useful means of strengthening a material, it is undesirable during forming, often leading to kinking and stretcher-strain markings.”

Meanwhile, the yield stress when bending in the opposite direction of coil set happened to be low for all materials. This is a bit counterintuitive, but it does follow the common phenomenon that engineers refer to as the Bauschinger effect. Named after German engineer Johann Bauschinger, this effect states that while more cold working of a material does increase the tensile yield strength, the local initial compressive yield strength after cold working is actually reduced. According to sources, though, the most important finding is that the bending test uncovered a range of yield stresses.

How Oil Canning Occurs

Back to the roll forming setup, the tool designed to compensate for the bowing induces slight elongation in the flanges and compression in the web—that is, the web becomes ever-so-slightly shorter, lengthwise. As Weiss explained, if this compression occurs uniformly, no buckling occurs. But if this compression doesn’t occur uniformly, the web buckles as it compresses and ultimately produces the ripples and waves of an oil-canning defect.

So what causes nonuniform compression? The bend test itself doesn’t reveal a definitive answer, but it does reveal that material with a higher yield stress, as measured when bending in the direction of the coil set, may be more susceptible to such buckling and hence be more susceptible to oil canning.

Figure 2
When 13.1-in.-wide drawing steel was being formed into this profile, the roll formed sections tended to bow upward, sometimes more than a half inch over a 118-in.-long section.

Predicting Problems Before Roll Forming

The trick is knowing which materials may be more susceptible to such defects before they are set up on a roll former, and it is here that this bend tester may help. The one used for this application was a pre-production prototype, and at this writing, a design for commercial versions is in progress.

Eventually such testing devices could be integrated with a die system installed at the beginning of a roll forming line. As Weiss put it, “The operator could cut bend test samples at certain intervals to fully monitor the process.” Theoretically, this could flag changes in material properties and spur tooling adjustments to prevent or at least minimize problems, be they oil-canning defects or anything else.

According to a paper written by Weiss and Henry Wolfkamp at ARM, “Whether or not a coil is suitable for a line is a decision an engineer needs to make based on the data available. [We] suggest that a simple bending test provides much improved information when compared with the traditional tensile test and could greatly assist this decision-making.”

As with most things in metal forming, the more information a technician has about a certain material, the better he or she can predict how that material will perform. In the old days, a metal former would test by just getting a feel for the material. Today testing systems and software are uncovering the science behind the craft.

Learn the basics of roll forming operations

FMA is offering a two-day seminar will teach those new to the industry the history, terminology, and proper processes involved with roll forming. Attendees will learn the definition of roll forming and understand the basic principles every operator needs to be effective.

A combination of interactive classroom learning and easy-to-understand exercises will help them learn or update skills needed to produce quality products while reducing errors and downtime. FMA’s expert instructors will guide attendees through the basic concepts of using measuring equipment, interpreting blueprints, developing die skills, and working safely.

This FMA Roll Forming Certificate Course is scheduled for Oct. 6-7 in Pacoima (Los Angeles), Calif. For more information or to register, call 815-227-8257.

About the Author
The Fabricator

Tim Heston

Senior Editor

2135 Point Blvd

Elgin, IL 60123

815-381-1314

Tim Heston, The Fabricator's senior editor, has covered the metal fabrication industry since 1998, starting his career at the American Welding Society's Welding Journal. Since then he has covered the full range of metal fabrication processes, from stamping, bending, and cutting to grinding and polishing. He joined The Fabricator's staff in October 2007.