WEB EXCLUSIVE: Tension Control: Dancer Systems Defined

Published: April 16, 2009, By By Greg Cober, Warner Electric www.warnerelectric.com

One of the challenges in working with tension control systems is working with closed loop dancers. While dancers are a seemingly simple mechanism, there are complexities below the surface that can impact system performance significantly. This article will define the key parts of a dancer system and how they can interact to provide a system that works well or poorly.

A well-built, well-installed dancer working with a proper control can provide a flexible closed loop means of controlling tension in both unwind and rewind applications.

We will use a simple unwind system to discuss the various parts of a dancer system.(Please note that there are many possible dancer designs, but for clarity’s sake, we will restrict our discussion to the simple design in Figure 1.) As a web of material (or strand of wire) unwinds, it passes over a roller, around the dancer roll, and then over a third roller.

The dancer roll is allowed to float (rise/fall) as material feeds into the process. As the dancer roll rises and falls, the tension system is adjusting for minor changes in tension. In most well-designed systems, the dancer roll will rise and fall in small increments constantly. In short… it dances!

In a closed loop dancer unwind system, there are several key components: the dancer itself, a feedback potentiometer, a control, and the brake or drive. The goal in a dancer system is for the dancer to remain at the mid-point of travel. Pressure on the dancer establishes the tension for the web. The brake in the system controls how fast material is allowed to unwind. Brake pressure changes based on feedback from the dancer potentiometer to the control and from the control to the brake or drive.

If the process is consuming material faster than the unwind is releasing material, the dancer will rise (the loop under the dancer will become shorter). If the process is consuming material slower than the unwind is releasing material, the dancer will fall (the loop under the dancer will become longer).

In the simple dancer system, the dancer can rise and fall 30 deg around its midpoint as changes occur. As this rise and fall occurs, the feedback potentiometer will provide a signal back to the control (as shown in Figure 2).

The control in a dancer system interprets the feedback signal through a PID loop. It looks at the direction of change, amount of change, and rate of change. Based on these three variables, the control will adjust the force at the brake appropriately. A slow change in dancer position will result in a slow change in brake force. A rapid change in dancer position will result in a rapid change in brake force. Adjustments within the control tune the response to the level required by the specific product being manufactured. Some products require very responsive systems, some less so.

Correcting a Range of Conditions
The dancer in the system, as it dances, is correcting for a wide range of conditions. The unwind roll may be slightly out of round with the result being a fast/slow feed rate. There may be mechanical wear issues in the unwind stand, such as worn bearings, alignment issues, and similar mechanical inconsistencies. The material being handled may be highly extensible and prone to stretching. The process itself may have a start/stop function as might be found in a heat-sealing process. Finally, there is an acceleration and deceleration ramp as the machine starts and stops. All of these dynamic conditions, some of which are occurring simultaneously, conspire to change the tension of material being used in the process.

This ability to absorb changes and correct for them within the dancer loop is the great advantage of the dancer system. A second benefit of a dancer system is that the storage loop allows for faster roll acceleration and deceleration and allows for faster splice times than for systems that do not have the same storage capability, such as load cell systems.