Web Handling & Converting Blog

Equal Diameter Myth

When I first started out as a designer of web machinery decades ago, I observed that top and bottom slitter blades were always different diameters.  Sometimes only slightly so like 10.0” and 10.5” respectively.  I observed similar things with calender rollers.  I asked some old timers why not just make them the same?  They answered that the rollers would be turning at the same RPM’s and damage to one would beat on the same spot and ruin its neighbor.

That didn’t seem right, so I put it to two different tests.  First, with FFT vibration analyzers, I noted a beating of imbalances.  This beat frequency represented the slight phasing of the heavy spot of one roller with respect to another.  The beat frequency was several cycles a minute for high speed machinery.  Then I put a chalk mark on the two rolls and found out the same thing.  The chalk spots quickly moved out of sync.

After a moments reflection this made sense.  Even if rollers were tracking perfectly, something like gear teeth, a mere 0.001” diameter difference would get 1” out of phase in only 1,000 revolutions, mere seconds on high speed machines or minutes on low speed machines.

So, I didn’t believe it and I still don’t believe all the claims made by the old timers which sounded plausible but had nothing but anecdotal evidence to shore up the conclusion.

I have learned as much by doubt as I have by curiosity, maybe even more.

Roller ID Best Practices

You and I have at least two names:  a first name and a family name.  Rollers should likewise.

The first name of a roller corresponds to a position in a machine.  It is used by everyone in a plant.  The first roller is #1, the second #2 and so on.  This numbering is physically lettered on the machine so people know what you are referring to.  It makes no more sense to say we got wrinkling on the little idler after the calender than it does to say to your boss that you observed a skinny operator do an unsafe practice.  Rather, it is #13 and Jose respectively.  Of course, as we’ve mentioned in the past, the threading diagram which everyone has paper and electronic access to, reflects this numbering system.  It is posted on the machine for threading, in the instruction manuals and on the conference room wall.  If you do a rebuild which adds rollers, you can subscript new rollers with a letter much like they do with highway exits.

The second name of a roller is a specific ID, much like your social security number, that corresponds to a specific component.  It is used almost exclusively by maintenance.  Each roller has a folder in which is filed a maintenance and machining history.  Idler rollers will have little activity.  On the other hand, process rollers may be ground and recovered and reworked as many as dozens of times in its history.  Ingoing and outgoing profile reports of diameter and TIR and dates allow one to monitor wear rates and perhaps predict the next required maintenance down.  These folders will quickly become quite fat.

Last Chance to Learn Web Handling and Winding

before the fall.  See you May 20-23 in Ontario CA.  http://www.convertingschool.com/courses/course.asp?coursecode=WH1

TopWeb

I very seldom endorse anything publically.  The exceptions I have made in the past were books because "A month in the plant is worth an hour in the library."

I will make another exception for TopWeb2 which is a polished commercial code for doing common web handling calculations such as for roller traction and drives, laminator curl, guiding, spreading, winding and much more.  Everyone doing serious design or troubleshooting should consider it.  A 30 day trial version can be downloaded from http://rheologic.co.uk/.  Those of you doing serious coating and drying must also check out TopCoat.

No, I do not make a comission.  I do not accept any comissions or gifts from suppliers.  This is just my honest recommendation.

AWEB08 - Day 3

Tim Walker gave an hour long talk on Size Matters: The Effects of Roller Diameter.  It did not occur to me until seeing his excellent compilation of just how pervasive diameter is in web handling physics.

Jerry Brown of Essex talked about What FEA Analysis Can Tell Us About Spreaders.  His custom package overlaid on commercial FEM code was used to quantify bowed roller and concave spreading.  His conclusions were that the CD tensions produced are quite short-lived.

Steve Huff of Imperial Rubber Products talked about What is Your Roller's True Shape and Why Does It Matter?  Conventional measures such as TIR and TAPER are not adequate to capture the shape or cylindricity of rollers.

We then had an hour long Sage On Stage where the expert panelists included the distinguished web handlers Dr. Dilwyn P. Jones, Dr. David Roisum Tim Walker and Jerry Brown.  This stump the chump included lively discussions of baggy webs, web handling certification and many other topics.

In the afternoon we had a shortcourse on winding shared by experts Dr. Dilwyn Jones, Dr. David Roisum and Tim Walker.

Teach, can I be excused?  My brain is full

AWEB08 - Day 2

Yours truly, David Roisum, gave an hour long tutorial on Troubleshooting Baggy Webs.  The premise was that the options are two: fix the bag in manufacturing or suffer.  However, troubleshooting is challenging because measurements are difficult, identifying the element responsible is difficult and changing the element that is responsible is difficult.  Nonetheless there is hope if you keep your wits about you and are brave.  You must commit, something like kick-starting an old Harley, if you hold back even the least you may get hurt.  Many people fear getting hurt so they just don’t try.  Perhaps they will be inspired by the half dozen case histories illustrating how troubleshooting might be done.

Jerry Brown of Essex Systems talked about the Anatomy of a Wrinkle.   He designed and built a pilot line up which he ran a latex belt.  The latex was extensible enough that one can see the geometry changes on a misaligned roller which can then be removed on a concave roller.

Michael P. Bria of MEGTEC talked about Web Vibration in Flotation Dryers.  He experimentally verified that the Chang vibration model was close, but did not properly account for web mass.

Brant Ross, Ph.D of MotionPort talked about Initial Application of Multi-Flexible Body Dynamics Software to Web Handling Simulation.  Software such as this has been used for complex dynamic mechanism design for some time, but here it was shown to be useful for modeling of roller runout and misaligned rollers.

Andre Icso of Circonix Technologies talked about Feed Forward Algorithms in Web Conveyance Applications.  To make a drive control system work well, the drive must calculated or infer as much as possible rather than rely heavily on the trimming from load cells or dancers.  Static and dynamic friction, inertia, speed changes and many other upsets can be feed-forword to get the correction started sooner.

Dr. Lee Fenney Sigmala Ltd talked about Intelligent Micro Servo Control - The Future of Converting?  Micro servos have many advantages over pneumatic actuators including reduced cost.  Four of them are now used in one model of automatic slitters.

William Gilbert of Siemens talked about Synchronized Drive Control in Web Handling Applications.  Servo and position controlled drives have revolutionized the printing and packaging industries.  In addition to much more precise and flexible registration control, they can eliminate drive shafts and accelerate varying inertias more consistently.

Pierre Nader of SPN Consultants talked about Equipment Optimization - A Comprehensive Approach.  This organization specializes in offsite analysis of man, machine and materials in a winding system.  It does so by piping mega-amounts of data from an array of sensors to computers for analysis.

Joel Oakes of Delta Industrial talked about Optimizing Registration in Converting Equipment.  Registration accuracy is determined by a number of machine factors such as encoder count, speed, tension, path length, gears, tooling accuracy and so on.

Bob Hosler of Keyence talked about Non-Contact Measurement in Web Handling Processes.  Inductive, 1D and 2D lasers can be used for a wide variety of measurements such as roller runout, nip gap, web profile and thickness, vibration and so on.

Darrell Whiteside of Maxcess talked about The New Technology of Closed Loop Tension Control.  Unwind tension controllers have advanced to include weightless load cell calibration, gain scheduling, out-of-round roll compensation and auto-tuning.

Mike Kordik of Beta LaserMike talked about Using Laser Doppler Velocimetry to Increase Profits and Reduce Downtime on Converting Lines.  Lasers can measure web length and speed to better than an incredible 0.05% without the contact or slippage risks of conventional roller or wheel based methods.

Teach, can I excuse myself?  My brain is full.

AWEB08 - Day 1

The keynote address on Winding Models - Using Science to Profit was delivered by Dr. James K. Good, PhD from the Web Handling Research Center at Oklahoma State University.  He showed us that winding models can predict the following defects:  air entrainment, blocking core crush, loose cores, starring, registration issues due to creep, tin-canning and more.  He also showed how the effects of gage variation and changes in moisture/temperature affect stresses inside the roll.  Finally, we were given a short treatment on winder classes and the affects of tension and nip.

Dilwyn Jones of Emral talked on Solutions for Winding Multi-Layer and Patterned Webs.  Here we learned how laminates, especially thick ones favored by the electronics industry, challenge winding and even web handling.  Also, product designs which use patterns at the micro and macro scales cause immense affects on winding.

Andreas Schwab of Corenso talked on Winding - Parameters and Measurements.  Here we learned how standard and non-standard quality measurements which can be used for failure prediction.
.
Scott Hansen of Goldenrod Corporation talked on Performance Comparisons in Air Shaft Materials...What Material Works Best for Your Application?  Here we were treated to how to use mechanics to predict whether aluminum, steel, carbon-fiber or composite shafts were the best choice for any particular application.  Each application differs in the importance placed on load carrying capacity and minimized deflection, critical speed and cost.

Daniel Cain of Tekkote talked about Taking a Look at Razor Slitting.  This is first talk I’ve ever heard that was exclusively devoted to razors.  The practical thesis was the economic optimum blade material which gave the best sum of blade and blade change costs.

Luis Teixeira of Dienes talked about Web Running Slitting Changes Based on Vision System Output Signals.  Here a vision system was the feedback for slitter positioning.

Mark Daun of Preco talked about Advances in Laser Processing of Flexible Packaging Materials.  This was another first.  The first paper I’ve ever seen dedicated to the application of slitting, scoring, perfing and contour cutting.

Tim Walker talked about Why Isn't Your Slitter Running.  Here, rewinder productivity is analyzed in terms of the dozen or so activities an operator must perform.

Neal Michal of Kimberly Clark talked about Practical Considerations for Zero Speed Splice Unwinds.  I knew accumulators had many challenges, just not so many challenges.

Kelly Robinson of Electrostatic Answers talked on Electrostatic Charge Control in Slitting Operations.  Static generation and removal has always been a bit of mystery.  The new take home here is to remove static immediately after it is formed on the correct side.  If you wait even one roller, it may transfer to the other side of the web.  Then, removal of static is next to impossible.

Teach, can I excuse myself?  My brain is full.

AWEB08 - Anticipation

Just can’t wait!  Over 100 attendees from all over the world will be converging on Minneapolis today for the 2’nd Applied Web Handling Conference hosted by AIMCAL.  No wonder; with 30 papers, 2 keynote addresses (your’s truly is doing one on baggy web troubleshooting) and 2 shortcourses our brains are gonna get so filled.

Web Drive Instructors Needed

We are assembling a team to teach web drive school beginning in ’09.  The course would be made available to the public and targeted for the non-drive engineer who must work with drives or the drive engineer in training.

What the web world most needs now are web drive instructors.  Most plants have scores of drive points, often of different makes and models, and have difficulty knowing if they are working well and if not what to do about it.   Many plants are in the process of buying or upgrading drives and do not know what drives are supposed to do.  Too many people are writing or altering drive programs who simply don’t know what they are doing.  This has resulted, in my opinion, in drive performance being worse now than it ever has been in my nearly three decades of experience.  This degradation in performance is despite large improvements in processing speed and programming tools.  What we need is formal public instruction to take advantage of the mostly quite capable equipment.

There are only three qualifications for a web drive instructor:  familiar with web handling, at least a decade of field experience with demanding web drives, the willingness and ability to teach.  What will be disqualifying is any discussion of a particular make or model of drive.  A rough outline of some of the subjects would include:

1.  The web’s view of drive requirements:  forming, setting tension, minimizing tension variations

2.  Drive control strategies:  load cell, dancer, speed/draw, torque, torque differential, helper

3.  Drive control considerations:  feed forward and inertia comp, friction, no-(web)load torques

4.  Control block layout for common drive situations

5.  Special drive requirements:  payout, thread, stall tension and take-up

6.  Special driven element requirements:  rollers, bowed spreader rollers, laminating nips, cs winders.

7.  Types of drives:  mechanical brakes and clutches, DC, AC Vector, Servo

8.  HMI (Human Machine Interfaces):  what should be displayed and how

9.  Drive data acquisition and fault monitoring

10.  Summary

We especially would like instructors who are retired or work for integrators, but would also welcome the drive OEM’s.  Email me or Craig at AIMCAL if you or someone you know might be interested.

Winding Problems

Many problems show up at the winder.  It is the first chance you have to take a good look at the product.  However, just because you can see a problem at a winder does not mean it did not have an upstream cause that was not visible.  Even if you have a problem that was MADE on the winder, it may not be a winding problem.

Simply stated, winding problems have winding solutions.  By that I mean that the problem can be noticeably reduced by changes in operation, such as changing the TNT’s of tightness, or design (different arrangement) or maintenance (better alignment for example).

Some defects made on winders are not winding problems.  For example, baggy webs made by stretching over gage bands may not be a winding problem.  Same thing with blocking.  That is, provided that, you are already winding as loose as possible (before loose defects dominate rather than a limit of the machine).  Reducing winding tightness will just change one problem for another in this example.

Winding problems have winding solutions.  If there are no winding solutions, then it is almost certainly a product design problem.  This material may be great for end use, but is not windable.  So, if you make gage bands that the customer does not object to, fine.  However, that product must be put in a box rather than wound.  Any winder run with any settings will ruin that material for that black and white example.