- October 01, 2003, Timothy J. Walker, TJWalker & Assoc. Inc.
Spreader rollers are important tools for web handlers with two main applications. First, they can spread a web after slitting, creating a gap between slit strands. Second, they can ensure the web is laterally taut, removing or preventing wrinkles.
I've been laying the groundwork for this discussion on how spreader rollers spread. Over the past two months, I have covered the parallel entry principle (PEP) and the beam-like nature of web bending. Using these two concepts, let's dissect the commonly misunderstood mechanism and limitations behind two popular spreader rolls: the bowed roller and the expanding-surface roller.
Both rollers feature a surface of uniform cross-roller diameter that moves laterally, expanding and contracting, as the roller rotates. The spreading mechanism seems obvious: The web enters the roller at a given width, rides the roller's surface as it shifts laterally, and leaves the roller either taut (in the case of a single wide web) or spread with uniform gaps (in the case of multiple slit strands). Amazingly, on-roller spreading is a secondary effect. I would like to convince you that most of the spreading occurs before the web touches the spreader roller.
To learn more about how spreaders work, I recommend highly a milestone paper presented at the 1997 International Conference on Web Handling. Ron Swanson of 3M presented “Testing and Analysis of Web Spreading and Anti-Wrinkle Devices.” In Ron's paper, he evaluated ten different rollers, empirically demonstrating their ability to prevent wrinkling and spread a slit web.
Ron's simple spreading test started with a 12-in.-wide web, under controlled tension and speed, tracking from an upstream idler, across a 24-in. web span, and over the target spreader roller. A razor blade slit the web down the middle, upstream of the spreader roller. A roller's spreading effect was quantified by measuring the slit gap that developed between the two web halves.
Two observations from this test firmly disproved on-roller expansion as the dominant spreading mechanism. First, for both the bowed roller and surface expanding roller, the web spreading was greater than the roller's surface expansion (about twice as much). Second, the slit gap opened up prior to contacting the roller.
How can a spreader roller move the web before it even makes contact? The answer lies in the parallel entry principle. The gap develops because the two halves are responding to the differing surface vectors of the left and right side of the spreader roller. The left side of a bowed roller angles to the left, the right side to the right.
The expanding-surface roller works on the same principle. The center of the roller surface vector is in the machine direction. The left edge of the roller surface vector is angled to the left, the right angles to the right. Again, each side of the web will be displaced by the parallel entry principle.
How much will these rollers spread the web? The lateral motion of a web created by the angled roller surface vector will be equal to approximately two-thirds of the angle (in radians) times the entry span length. Therefore, the web will spread more with more angle and longer entry spans. The strong effect of entry span on bowed and expanding-surface roller spreading commonly is overlooked.
Will the web spread and achieve parallel entry? Only if there are roller traction forces great enough to bend the web. Beam bending will determine if the traction forces are great enough to bend the web as far as the parallel entry. A common mistake for both these spreader rollers is to set them up with large surface vector angles, thinking more is better. Setting up a spreader beyond the traction limits will not spread more but will abrade and polish your spreader roller to an early retirement.
Next month: how a flexible-spiral spreader roller works, plus wrapping up our discussion on spreading mechanisms.
Timothy J. Walker has 20+ years of experience in web handling processes. He specializes in web handling education, process development, and production problem solving. Contact him at 404/373-3771; firstname.lastname@example.org; tjwa.com