It is well known that winding looser (soft) will reduce the frequency and severity of some types of wound-roll defects. Examples would be ridges, webs stretched into bagginess over ridges and corrugations. In the case of ridges, areas of the web that are ever so slightly thicker than their neighbors stack up into ridges faster than the adjacent thinner areas. A ridge that is a mere 1/1,000 larger in diameter than the rest of the wound roll may be visible, leaving you open to criticism from your customer.

It does not matter at all whether this feature is a functional problem. Perception is reality. If your feature is more prominent than the competition's, your product will be viewed less favorably in that regard. Also, the customer could attach some other undesirable behavior to the feature, even if there is no real physical connection. It is equally possible that the ridge could damage the web by stretching it into a baggy lane at that location.

The corrugation, also known as a rope or chain mark, also has a caliper error as a root cause. Here, however, the error that causes the defect is a high-low or especially the high-low-medium gage error pattern across a narrow width. Typically, the corrugation is only 1/2 in.-4 in. wide. The same gage error over a wider band may not trigger the corrugation. However, there is a distinct difference between the ridge and the corrugation.

The corrugation is almost always a nip-induced defect. This means that of all of the three TNTs (tension, nip and torque), the nip is by far the most damaging with this corrugation. Thus, we would aggressively reduce nip load until we ran into some other limit or trouble and only then spend modest efforts on web tension.

One common misconception is that the gage errors stack up proportionally. This does not happen even on the stiffest of materials. All rolls have some tendency to self-level. Thus, the high gage area will wind tighter there and thus bring the bulge down a bit, and vice versa on the low gage area. For this reason, a 1/100 gage error may cause a 1/1,000 roll diameter bulge. As a practical matter, however, it does not matter whether the root problem is 1/100 or 1/1,000. It is still too small to be picked up by most lab or online gage measurement systems.

It is ironic that the QA or gauging system is often unable to measure, much less control, the very problem that it should be trying to prevent. Not only is the gauging system too insensitive, it is also too coarse. Corrugations and ridges are so narrow as to often fall between the bins or samples.

How then do we know if we have a gage error? Simple, the wound roll will tell you. Diameter variations across the width, with the exception of tin canning and a few other situations, are all the proof one usually needs. If you need more, roll hardness variations as measured by Rhometer, Schmidt Hammer or other device is usually most sensitive to gage variations. Sometimes, special sampling techniques can boost the resolution of lab instruments to the point of detecting problematic weight or gage errors.

Back to the question of whether you can hide a defect by winding looser. The answer is no. There simply might not be any defect to begin with unless you wind tight. In other words, a gage error in a web is not necessarily a problem with a web in manufacturing, converting or end use. It is a gage error in a tight wound roll that is the problem. Thus, the best way to view defects like this is with the following generalizations:

1. The most sensitive measure of a level web is a level wound roll.
2. The fussiest customer for a level web is the winder.
3. Winding loose does not hide defects; it prevents them.

David Roisum, Ph.D., Consulting Technical Editor, 920/725-7671, DRroisum@aol.com, www.roisum.com