Wind-Up Applications Using Brushless DC Motor


Thread Starter

Chris Ayers

This turned into a wall of text, but bear with me. I'm not very experienced with controls and perhaps it is a simple solution:

So I have been working on the design of a semi-automatic label applicator. I'm starting from scratch, and my goal is to look at existing designs that are on the market and come up with my own design which eliminates much of the mechanical complexity (and cost) through use of controls. In my research I've found that the take up of the backing paper can be accomplished by using only a servo motor, that is, no additional sensors are required.

The design I have in mind would involve a supply reel (passive), guide rollers, a peel plate, nip/drive rollers (which is doing the precision work and has it's own motor), and a take up reel (with a separate motor).

So roughly speaking when the device receives input to dispense a label, it advances the webbing a certain amount until a label sensor detects an new gap between labels. This is actuated by the main drive/nip rollers. Simple enough, just advance at whatever speed is required, and perhaps some ramping up/down too keep things smooth.

The take up reel is a bit more tricky, however. We want minimal slack between the web that is between the drive roller and the take up reel. There could even be a roughly constant tension along this distance. Keep in mind that at this point the precision task has been accomplished and now the take-up reel just needs to keep things neat.

So obviously the take up operation could be done many different ways, but I want to avoid unnecessary sensors and components. I know of (at least) one product that accomplishes this with just a servo with its three hall-effect sensors measuring rotor position. From the manual: "The control current set by the system, in accordance to the state of operation, directly controls the rewind's tension."

The solution is not as simple as constant (angular) speed because as material accumulates on the reel, the diameter will increase, thus increasing web velocity. Keep in mind that we know what the web velocity is since it is being directly controlled by the drive/nip.

Similarly the solution will not be as simple as constant torque (current) since as the diameter increases tangential force will increase resulting in things like webbing tearing. Obviously there is wiggle room and and a range of tensions will be acceptable, but we are talking about the reel advancing from something like 180 degrees per label at the beginning to something like 60 degrees or less per label as material accumulates. (Diameter would range from about 3" to perhaps 9")

So does anyone know how they accomplish this with just the servo motor (as well as a setup similar to what is described above)? To me it seems like you need to know the diameter of the reel to make it happen. Perhaps they are estimating diameter based on webbing thickness? But webbing thickness varies and this assumes a tight winding on the take-up reel. But maybe that gives you a close enough approximation.

Another question: I originally wanted to have only the windup motor controlling the advancement of the webbing, thus eliminating the need for two motors. During my research I found that unfortunately this idea was patented ( Basically the patent says it's a label applicator (known art) but with only one drive where the drive is controlled by feedback from servo motor speed and webbing speed (given by encoder on peel plate). I mean, how can that be patented? It seems so obvious to me.

Anyway I'm thinking the same thing can be achieved by instead using an ultrasonic sensor to measure reel diameter to help control speed/torque of a servo motor. I think this would get around the patent (?), but it seems like it might be less accurate and there is no way to make sure the labels are coming out at the right speed (no encoder measuring).

Anyway let me know what you guys things, thanks much for the help!