Hi,

My company builds web handling equipment that in its simplest form, has an unwind station on one side and a rewind station on the other side. In between the unwind and the rewind, I use two driven rolls to pull the web at a constant speed from the unwind to the rewind. (In between the two driven rolls, the web is subjected to my "process"). I setup the two driven rolls in a master/follower relationship with one roll following the speed of the other roll. It is very important that the follower matches the speed of the master precisely at all times, otherwise my web will stretch, break, or develop slack that will adversely affect my process.

In the past, I have used DC motors with dedicated speed controllers and 4 quadrant regenerative drive boards. I am not 100% sure that the regenerative capability is really necessary for us, but most likely one of our vendors suggested it at some point in time and we just kept doing the same thing. This has always worked for us, but now some of our customers are requesting a wider speed range (on the low end) than what is typically within the realm of DC motors. Anyway, I am planning to make the switch to AC vector drives / motors so that I can get the speed range they want. (I am leaning towards Allen-Bradley PowerFlex 700S).

Anyway, as I have been thinking, reading, asking how to implement my master/follower relationship, I have become aware that I will have to determine whether I will need a dynamic brake resistor, and if so, what size I will need. I started reading Allen-Bradley's document PFLEX-AT001 (PowerFlex Dynamic Braking Resistor Calculator) and got stuck at the point where it says I will need to know my load inertia. I do not know how to calculate this in a web handling application, but my "gut feeling" tells me that it is probably low because (1) my web speed is very slow - my gearbox/belt reduction is usually such that 1750 motor rpm would equate to no more than 50 feet per minute in web speed (less than 25 rpm for a 8 inch diameter roll) and (2) because my driven rolls are constantly pulling the web against the resistance created by my unwind shaft. My unwind shaft is under tension control (using an electromagnetic brake) so excluding the inertia of the driven roll itself, the load (my web) is going to lose momentum almost immediately when my driven rolls stop pulling. If my reasoning is correct, I also have the benefit that there are no circumstances in which I would need fast deceleration. All I really need is for when the master drive slows down and/or stops, I need the follower drive to slow down at the same rate.

I was just wondering if anyone here has experience with AC Vector drives in web handling applications similar to mine, and if you have found dynamic braking resistors to be necessary (or useful). Any advice will be greatly appreciated.

Thanks in advance,
Paul

 

The dynamic braking resistor operates during deceleration phase doesn't allowing increase DC bus voltage above safety level. During operation the resistor simly connected to DC bus by switch (internal for driver). Most of modern amplifier has such option and it maybe used further during operation. Standard value of braking resistor is 25...75 Ohm with more than 500W power.

 

Typically what is done on web applications is that one drive is set up as the speed follower, then the second drive is put into Torque Control mode and is a torque follower of the first drive. In other words your speed command goes to drive #1, which has an output proportional to torque that is given to drive #2 as the torque command it must follow. That way you never have any extra tension on the web. This is typically accurate to about 1% using analog signals. If you go with the PowerFlex 755 using Ethernet and CIP motion commands, you can get better accuracy than that.

But yes you need the DB resistor package just in case, because if the system is given a decel command at any point in time, the kinetic energy in your web must go somewhere. So don't think of it only in terms of the DIFFERENCE between the two drives, that can be controlled via the torque follower method I just described. But if anyone EVER asks the web to slow down, it's inertia represents kinetic energy that must go somewhere. So you would size your resistors to dissipate the total energy in the system.

The alternative is to use a line regeneration module and put any excess energy back into the line. Rockwell doesn't offer than until you get to about 350A, so if your motors are each under 150HP, that's not viable for them directly. But you can also get a Bonitron Line Regen unit to go ahead of them. They offer a canned package to accomplish that.