Regulator Tuning – Request for Metrics

At this time, we do not have good statistics for the speed and tension regulator responses used in web handling.

Today I am asking for help in compiling some metrics for how speed and tension regulators are tuned in the web handling industry. The purpose for this information is to publish it for the industry through an AIMCAL conference as well as other web handling forums.

For purposes of this study, I am not interested in your web product. It is important for the industry to know the design line speed and web width. My hypothesis is the web width and line speed will have very small impact on regulator response. Optionally, please provide short word description of the tension zone (ex. Cast roll, calendar, cooling roll, slitter, unwind, winder, or if you are shy, zone 1, 2, 3).

For both speed and tension (or dancer position) regulators, I am interested in the regulator response or Bandwidth (BW) in radians per second. That is one over the time constant (time to 63% of the final setpoint). I am also interested in the % overshoot and % undershoot.

To start the process, I will provide regulator responses for 10 tension zones I am familiar with. I defy you to determine the owner of the equipment listed in the chart. You may be able to guess the web product.

Thank you in advance for your help.

E-mail me when people leave their comments –

You need to be a member of Association of International Metallizers, Coaters, and Laminator to add comments!

Join Association of International Metallizers, Coaters, and Laminator


  • I agree with comments below

  • Drive vendor instruction manuals describe a method to use to tune speed loops. Engineering must determine speed loop tuning requirements based on knowledge, design and/or experimentation.


    Your question about existing system speed loop response assumes people who designed/tuned the original system knew what they were doing. My OEM experience, clearly demonstrated that drive vendors do not send technicians with extensive experience. New engineers fresh out of school are typically placed in these positions to gain experience. Often, after startup, there was significant arguing about operational issues. Drive vendor argued issues were caused by poor mechanical design. While, OEMs argued operational issues were caused by poor tuning and axes coordination. As a result, this OEM decided to bring all web handling design/tuning/startup in-house.


    To properly setup a web transport system, engineering must look at the entire system and what is being transported. If web is sheet steel, two nips cannot be operated as speed loops because they will be too tightly coupled via the web. Speed loops will fight each other. Special configurations are required to allow this system to transport sheet steel. If web is changed to polyethylene, speed loop coupling via web is non-existent. Therefore, separate speed loops are a viable solution. Now, the question is how to tune these loops. If upstream speed loop is tuned weak, downstream speed loop is tuned very hot and speed command to each speed loop is a step, web will probably break because of excessive speed differences between these two nips. If speed command is ramped or s-curved with long acceleration times, speed differences during acceleration will be reduced. With sufficiently long acceleration times, web will not break. If both speed loop regulators are tuned to the same response, differential speed between nips will drop to zero independent of the rate-of-change of the speed command. Thus, in a web handling system, tune all speed loops as hot as practical. Then, re-tune all speed loops to the same response as the weakest speed loop.

This reply was deleted.