- September 24, 2013
How do transverse direction variations in equipment or web properties affect roll structure?
Part 3 of this discussion covers machine direction (including radial position) and time dependent variations in rolls. Lastly we need to address transverse direction (TD) variations, especially since many roll and web defects have TD variations.
In accessing TD variations within a roll, we imagine dividing a winding roll in narrow lanes much like dividing a wide roll into narrow slit rolls.
Different TD lanes will have different average thicknesses and will build to differing diameters. Lane-to-lane of a wide winding roll or roll-to-roll diameter variations of slit rolls create surface speed variations. When the web sees increasing or decreasing speed, the web strain and stress will follow. Lower than average diameter lanes will have below average tension; above average diameters will have above average tension. Nipping will vary TD from both diameter variations and uneven nip loading. Lower than average diameter lanes will have below average nip load; above average diameters will have above average nip load.
Many factors in winding are non-linear, so small changes in thickness can create 10x to 50x higher stresses and strains in thick vs. thin lanes. The biggest non-linear factor in winding is the stack modulus. A stack of most materials loaded with minimal pressure is initially soft and easy to compress. However, as a stack is pressed into a tighter stack, it becomes more difficult to compress. Then, under extremely high pressure, the stack will have squeezed out all the voids of entrapped air, porous structures, or surface roughness, and you will reach a compressibility of the bulk material. This transition from spongy to full compressed stack will be non-linear when viewed on a pressure vs. strain plot.
In winding a roll with TD thickness variations, the thick lanes will be under higher tension and stresses for two reasons. First, the larger diameter lanes formed from thicker than average material will have high wound-on tension since the above average diameter will steal tension from the other lanes of the roll and have an above average nip load and nip-induced tension. Second, the higher tension from torque and nipping will form a stiffer stack in the thick lanes of winding. Due to higher starting tension and less compression to relieve the wound-on tension, the thick lanes of a wound roll will have non-linearly higher stresses and strain than average or below average thickness lanes.
These variations would create less problem if the web reacted elastically. If you look at the strain differentials within a roll, they don’t appear to exceed the material’s yield point. If the same strains occurred while wrapping a roller, the web would pass by undamaged. However, since the web is held under these stress and strain conditions for the long times of storage and transport, the web’s response is not elastic.
In materials science, creep is the tendency of a solid material to move slowly or deform permanently under the influence of stresses. If different lanes of a web were be strained at 0.5% vs. 0.2% (or 0% in loose lanes) for a short time period, such as when wrapping a concave roller, and the yield strain was 2%, there would be no permanent damage to the web and the lay flat or bagginess would be unchanged. However, if the web was held under the same strain differential conditions for long periods of time, such as within a wound roll, a portion of the strain would become permanent. For example, all lanes might retain 10% of their long-term train, setting in a permanent strain of 0.05% and 0.02%.
These may seem like small dimensional differences, but the length variations of 200 and 500 parts per million (200 micro-in. per in. or 200 microns per meter) are enough to create an obvious baggy lane. For a stretch material, this level of bagginess would be pulled out at modest tensions. For a high modulus material, such as bond paper or polyester, it would require 0.25–0.75 PLI/mil to pull out this bagginess (1.7–5.1 MPa).
Surprisingly, it would take more tension to pull out bagginess if there are only a few baggy lanes and more tension to pull out a web with a great percent of its width that is baggy. This makes sense when you realize that bagginess goes away by stretching the short lanes to a length equal to the long lanes. Therefore, more percent width of bagginess means there is less width of short material needed to be pulled to the baggy lanes length.
There are other transverse direction factors that may create cross-roll stress and strain variations, including:
- Baggy input web
- Uneven nipping left-right
- Uneven nipping due to core or nip roller deflection
- Uneven nipping due to core diameter variations
- Uneven tensioning from roller or core misalignment
- Wrinkles from a poor core start (quite common with flying knife auto roll transfer systems)
- Wrinkles wound into the roll, often in the center 80% of the roll width but not near the edges.
- Poor slit edge quality on one or both edges
- Uneven bulging or shrinkage of the winding core.
The result of TD variation in wound rolls is defects that may form in specific TD positions (e.g., near one edge, in the center of the roll’s width, aligned to a thick or thin lane, between thick lanes.
The combination of TD variations and natural changes in roll structure from core to outside layers may also create defects in specific combination of TD and radial positions.
Solving wound roll defects is complicated. However, by understanding these four fundamental areas of winding [1) wound-on tension, 2) profiled control of tension and nip load, 3) understanding the transitions to wound-in stresses, and 4) mapping TD variations onto your wound rolls] you have the tools to apply our best knowledge of the winding process to improve your product quality and yields.
Click here to read Part 1 of this series.
Click here to read Part 2 of this series.
Click here to read Part 3 of this series.
Web handling expert Tim Walker, president of TJWalker+Assoc., has 25 years of experience in web processes, education, development, and production problem solving. Contact him at 651-686-5400; firstname.lastname@example.org; www.webhandling.com.