By Sean Craig, global vice president of Product Management and R&D, Maxcess

Differential winding is a tool to help achieve control in web winding. It is the process of winding each independent roll at the speed and tension appropriate for each diameter roll and entering web length. It achieves this by varying the torque applied to the core on which the product is wound and allowing the roll to slip through reduced torque when tension in the roll increases above the set point.

Believe it or not, your web has multiple personalities, characteristics and properties. Every web is unique. No matter what web is being run, that web has properties and characteristics that affect the quality and structure of your finished roll.

Of course, you account for this by applying the TNTs of winding – tension, nip and torque – in your process to ensure the finished product meets the rigorous expectations of your customers (see Figure 1). You understand the Jack Daniel’s principle of tension (see Figure 2), so you ensure a hard start, a smooth transition and a soft finish to the roll. You may even have a nip or lay-on roll in place to ensure entrained air is removed and that the rolls are packed adequately. Yet, despite all your efforts, the finished package for your customer, the wound roll, is unacceptable. The symptoms: loose edges, baggy lanes, telescoping or dishing rolls, starring, loose or excessively tight rolls.

A cause of wound roll defects

The reason all your efforts are for naught is that the cause of these defects lies not in the upstream process and controls but in each individual roll as it is building. Caliper or gauge variation in the web, combined with the properties of the material that is running, allow the effect of this variation to rear its ugly head.

Consider what happens to a web in any given process. You start with a parent roll of material that you hope meets the quality standards expected of the supplier. But even so, that roll is quite diverse in its makeup. Depending on the material, across the web, there will be variation in thickness and quality. Then, the roll is run through your process where it may be coated, laminated, printed on, embossed or any number of other processes. As it comes out the other end of the process, you then slit it into multiple webs and wind it up into packages, also called rolls, designed to meet customer expectations. At this point, you have taken that web in all of its diversity, personality and character and split it up into individual rolls. You expect that each roll will behave the same. However, due to the subtle differences in the portion of the web running into each roll, as the roll builds, those differences begin to affect how the roll is building and lead to differences in wound-in tension for each roll.

Eliminating these defects requires more than just adjusting your tension. You need to understand what effect your diverse web is having on how your roll builds and what factors need to be controlled to compensate for this effect.

Controlling web tension begins with a taut web. There is little to no tension in a slack web. Once the web is taut, tension in the web increases the tautness to a desired set point. On a solid web, equal tension is maintained across the web by ensuring the web is pulled and wound up uniformly. In a multiple web condition, the requirement is the same. However, you now have to ensure each roll is being pulled uniformly – or to put it another way, you must make certain that you are taking up the same amount of material in each roll. For example, Figure 3 shows that in order to wind an equal amount of .0019 in. thick material on roll two and .0020 in. material on roll one, roll one will eventually be at a greater diameter than roll two. The result is that the revolutions required to build each roll uniformly begin to differ. If you maintain the rpm required to wind roll two, roll one will become increasingly tighter. If you maintain the rpm to wind up roll one, roll two will become increasingly loose.

Now introduce the variable of controlling tension. Tension is equal to the torque required to rotate the roll divided by the radius of the roll. If roll one is building at a different rate than roll two and has a different radius, it will be impossible for the tension between the two rolls to be equal. Therefore, to accurately overcome the effects of caliper variation in the web, you must not only control the speed of each roll relative to its position in the wind cycle but also control the tension in each entering web lane to ensure the desired wound-roll tension required for your material. And because these effects differ roll to roll, you must control these two variables in each roll.

Differential winding is a tool to help achieve that level of control. It is the process of winding each independent roll at the speed and tension appropriate for each diameter roll and entering web length. It achieves this by varying the torque applied to the core on which the product is wound and allowing the roll to slip through reduced torque when tension in the roll increases above the set point.

Differential shaft technology

Differential shafts, also called slip shafts, allow each roll to slip on the shaft as the roll is building. Unlike a lock-bar shaft that transmits enough torque through an expanding element to ensure none of the rolls slip, a differential shaft applies only enough torque through some expanding element to maintain the tension required at a given roll diameter.

There are two primary types of pneumatic differential shafts – core-slip and core-lock, also called external slip and internal slip shafts. Core-slip shafts (see Figure 4) rely on the core to slip about the diameter of the shaft. Since the core is allowed to slip radially, these shafts also employ some form of core stop to prevent the core from slipping laterally. The primary advantage of the core-slip design is greater load-carrying capacity, and they are more readily available for odd size cores and larger diameter cores. The tradeoff with this design is that some dust can be generated from the slip action, depending on the quality of the core. Tension control also is heavily dependent on core quality.

Core-lock or internal slip shafts (see Figure 5) have some mechanism or cartridge that locks onto the core, and the core and the cartridge rotate about some base shaft that transmits a radial force to control tension. Because the cartridge locks to the core, these designs do not require a secondary stop to prevent the core from moving laterally. The primary advantage to this design is that the core generates no dust. These designs typically provide the best quality roll builds. The tradeoff of this design is that it has a limited load-carrying capacity due to the inner shaft diameter, and they typically are available only for the more common core sizes.

Both designs use the same principles for controlling the tension in the wound roll. The focus of this paper is on pneumatically controlled differential shafts where air pressure provides torque and, subsequently, tension control. Therefore, these systems are under “torque control” where the required torque (tension) is achieved by adjusting the tangential force between the core/cartridge and the tension elements. This tangential force depends on the contact pressure (created by air pressure) and the dynamic coefficient of friction, which means the roll and shaft must be slipping relative to each other.

Basic differential shaft operation

Core-slip differential shafts: These designs have two or three air circuits: one air circuit controls the tension segments, one air circuit raises the core stop tray and one controls the individual core stops in the core stop tray. This final function also can be done mechanically. The basic procedure to operate is:

1. Position cores and lock in place mechanically or with air pressure.
2. Raise the core stop tray with an air valve or through a rotary air connection.
3. Begin winding.
4. Activate tension segments with air pressure from the rotary air connection.
5. Increase air pressure for tension control with air pressure from the rotary air connection.

Core-lock differential shafts: These designs typically require only one air circuit to control tension via air pressure underneath the cartridges. The basic procedure to operate is:

1. Position cores.
2. Begin winding.
3. Increase air pressure for tension control with air pressure from the rotary air connection.

Controlling tension and speed in a differential shaft

Over-speed and differential winding

Differential shafts use the principle of slipping the roll relative to the shaft, combined with air pressure to help control tension. But this slippage is equally important for controlling the speed of the roll relative to the shaft. As tension, air pressure and over-speed are considered, what you will see is that none of them operate in isolation to solve the challenges presented by the web, but rather they must be considered in relation to one another.

As mentioned earlier, rolls building at different diameters rotate at different speeds. By allowing one roll to slip relative to the next, you can ensure the rolls build at their correct rotational speed. Furthermore, by creating a system in which every roll is dynamic, slipping relative to the base shaft, you ensure optimum and equalized tension control. With differential shafts, this is done by over-speeding the shaft relative to the building rolls. This is accomplished by applying only enough air pressure to maintain the torque necessary for the desired tension, which then allows the shaft to slip inside the core, and by driving the shaft a small percentage faster than the fastest rotating roll.

Theoretically, the over-speed could be as low as the difference in required speeds between the web strips. However, if you try to position the over-speed very close to this point, there would be times that some cores would stop slipping and travel at shaft speed. This will cause an instability problem (jerking) because there is a difference in the static (or breakaway) coefficient of friction and the dynamic coefficient of friction. The static value always is greater than the dynamic, so once a core stopped slipping, the tension would have to build up in that particular strip to the point where the static coefficient of friction was overcome. Once it slipped, there would be a “snapping” effect (like snapping a rubber band) that caused that particular web strip to go loose for an instant. The more extensible the web material, the worse this effect would be.

The principle at work here is similar to what you experience while driving in icy conditions. If you proceed slowly and do not lose traction, the friction at work between your tire and the ice is static. However, once you begin to slip, you have far less traction (friction) between your tires and the ice. To put it another way, in that same scenario, it takes more torque to break your tires loose when you are not slipping than when you have begun to slip. Here, the tension required to counter the torque from the shaft in order to break the roll loose to obtain a differential effect is greater (thus the resultant spike) than the tension required to counter the torque when the roll already is slipping in order to maintain tension.

Air pressure and tension control

Because a differential shaft is a tension-control device, the principles of controlling the tension in the web through a differential shaft are the same as those used to control tension in the web under any other process. To maintain constant tension (T) in a given application, torque (TQ) must be increased as the roll diameter (R) builds (see Figure 6). For a differential shaft, this increase in torque typically is controlled by air pressure to the shaft. Other means of increasing torque or controlling tension are axial thrust against the side of a core and magnetic hysteresis, but these are outside the scope of this article. To control tension, there must be some form of feedback to the shaft resulting in a change in air pressure corresponding to the desired tension. If this is controlled by measuring some element of the building roll, such as its diameter, and correlating that to a desired tension and pressure to the shaft, this is known as “open loop” control. If this is controlled by directly measuring the tension in the web and correlating that to a pressure output to the shaft, this is known as “closed loop” control.

Open-loop tension control

For an open-loop system (see Figure 7), there must be some form of measurement that controls air pressure into the shaft. The most common is the use of an ultrasonic sensor measuring the diameter of the roll, giving feedback to a controller which provides an output that is converted to air pressure via a current-to-pressure transducer. An alternative is to use a follower arm with a sensor.

Closed-loop tension control

For a closed-loop system (see Figure 7), there must be some means of measuring tension directly and converting that feedback into air pressure to the shaft. The most common way to do this is to use a load cell to measure tension in the web with feedback to a controller, which provides an output that is converted to air pressure via a current-to-pressure transducer. An alternative is to use a dancer system.

Air pressure to the differential shaft

Differential shafts require continuous air to maintain constant tension in the building roll. In cantilevered shafts, a standard or dual-port rotary union will suffice. The same air source can supply both shafts in duplex winders, or they can be split and supplied separately. For drop-in shafts, the rotary union needs to have a quick-disconnect air fitting to enable the shaft to be removed from the machine. Safety chucks also can be used for differential winding with designs that incorporate a rotary air circuit through the center of the chuck to engage an air connection in the shaft. These often are very simple and affordable solutions for drop-in shaft configurations. Air pressure to the differential shaft can be controlled manually, through the machine interface or through the tension control system.

Because the differential shaft now is controlling the tension in the web, the pressure to the shaft will change as the roll diameter builds. However, this is not a linear relationship. As the roll builds, the weight of the roll also contributes to the friction between the tensioning elements and the inside of the core or cartridge. As a result, the amount of pressure required to achieve the desired tension actually will plateau and then eventually decrease, depending on the density of the material. This effect usually is greater on core-slip differential shafts since the core is the primary contact point for the tensioning elements and the resultant coefficient of friction is greater. The diagrams in Figure 8 illustrate this principle.

An important consideration related to controlling air pressure to the differential shaft is the effect of speed and pressure on heat and dust generation from the shaft. Differential shafts act very much like a clutch, whether they are a core-lock or core-slip design. Core-slip designs rely on friction between the core and the tensioning element, and excessive pressure and speed difference between the core and the shaft can lead to dust and heat buildup. While dust is less of a concern on a core-lock design, heat buildup still is a concern. Therefore, it is important to manage the speed of the shaft relative to the speed of the roll.

Part 2 will discuss speed/motor control for differential shafts, as well as provide models for speed control, inverse diameter and tension control. 

For over 30 years, Sean Craig has been involved in applying web-management solutions to the converting industry. As global vice president of Product Management and R&D for Maxcess, Sean leads the new product innovations and value engineering efforts for global teams located in the US, Europe, China and India. He is associated with the Association for Roll-to-Roll Converters, TLMI and other industry organizations and has written numerous articles for Converting Quarterly and other publications. Sean can be reached at 380-833-7500, email: SCraig@Maxcessintl.com or www.maxcess.com.