Respected sir,
I came upon a TAPPI standard that tension should be 10-25% of yield strength.
1. Is there any scientific rule to setting these value or are they heuristic?
2. What rule shall we follow in printed electronics industry? I am especially interested in the allowed variation from the set point.
The 10-25% rule has been quoted more often than all other tension guidelines put together since it was first introduced in my Mechanics of Rollers book in 1996, and for very good reason. It works (for most web materials and most positions in most any web machines and conditions etc). So, by this evidence it is a heuristic. However, there are very sound mechanics and statistical bases for this ‘rule’ that I can not detail here. But one way to think of the upper end of the range is to have a safety factor of 4 on permanent web distortion to allow for other causes of tension such as bagginess (residual stresses/strains), poor drive control, roller misalignment and roller diametral variation to name some of the bigger sources.
Now to clear up a few misconceptions before they also go viral. This ‘rule’, no matter how often quoted and even if done so in standards setting organizations such as TAPPI, is not a standard in the most rigorous use of the word. It is not even a rule, but rather a ‘rule of thumb’ that works well enough for most situations where you have no other guidance (such as a new material, lamination, process etc). It is also easy to apply, requiring information that is almost always readily available in most plants. As a finer point, the ‘rule’ is based on breaking strength rather than yield for some truly practical reasons: it is easier to define/measure and it is close enough for government work (given the large design range that is applied).
However, there are well known exceptions. The biggest is in forming sections (such as the wet end of paper machines or the drawing of film, nonwovens and metals). In those sections the web is past yield (by intention), thus permanently distorted and is usually run in speed control because tension is indeterminent. Another is with very thin and very thick/strong webs and sometimes small machines where we don’t tend to pull as hard, perhaps 5-10% instead of 10-25%. So even though there are exceptions, the idea works pretty well for most things. For those wanting a better answer; economics must be incorporated. I illustrate this and many other ‘choices’ a plant might make for product/process settings in my venerable Optimization by Integrating Business and Engineering Models.
Looking even further you will also see other time tested ‘rules (of thumb)’ that I have published. One is tension control variability not to exceed 5-10% of setpoint as a machine control quality specification. Similarly on the mechanical end I have proposed (in the same roller book) standards for roller deflection which really are standards in the narrow sense that many machine builders have adopted them as internal standards and a few builders have been sued for breach of contract (for crappy machines) based on published reasonable expectations from our web industries. Finally, I am putting the finishing touches on a similar set of guidelines for roller alignment because the guidance here is all over the map, but generally nonexistent, unhelpful when even mentioned and almost always (economically) wrong. This will be first published in the AIMCAL Web Handling conferences this spring and this fall. My hope is that machine builders, erectors and maintenance departments will pay attention; though I am sure to be disappointed because we are too often too busy to do our homework and always reluctant to rock the boat because there are so many other issues demanding our attention.