Payoff Dancer Control

TimWilborne

Lifetime Supporting Member
Join Date
Oct 2005
Location
Roanoke, VA
Posts
2,210
I'm working on an existing payoff of a wire machine. It has a motorized payoff controlled by a Control Techniques Commander SE AC Drive using a dancer arm mounted potentiometer connected to the drives speed control. Very basic control, when the dancer arm is down the speed is 0%, midway is 50%, and all the way up is 100%. The dancer arm also has a pneumatic tensioner

It works OK but as spool gets lighter or the tension is minimal, it struggles to keep a consistent tension.

I've worked on machines that have servos for dancer controls, but I don't have a lot of experience with using AC drives to do this. My initial reaction was to change it out to servo control but this setup almost works. Does anyone have any experience with AC drive dancer applications? This is a very no frills drive so I'm wondering if I couldn't upgrade the drive to make it do what they want.
 
Most of the AC ones I have worked on had the pot but they also had a brake on the unwind, so as the spool got smaller you could increase the tension... I have seen even ones that had a longer journal sticking out of the unwind with a leather strap and a weight hanging down, they just needed a little (stop the free spin) and others had a 4-20 running to calipers/disk setup and could be controlled manually or auto but either way it sounds like you need a brake.

EDIT: Also seen one that had a leather strap tied to the dancer and a spring, as the dancer raised up the strap would lesson tension as the dancer lowered it would also give more tension, the spring would absorb the shock, they worked great.
 
Last edited:
Interesting idea. I was trying to figure out how to get the motor to hold it back, I didn't think of adding mechanical drag to it. I should be able to experiment with that without too much trouble. Thanks.
 
We have 30 small de-spoolers unwinding wire from (36" dia) spools, I am now in the process of converting from the control scheme that you are currently using to one that's based on a PID loop.

The dancer arm/potentiometer method you describe is problematic at best with intervention needed by the operator to trim (adjust) the base speed as the spool is used up. Our big problem was that this control method would accelerate and decelerate the spools every time the dancer moved up and down, which is 60 to 100x per min. Balancing the acceleration so that it isn't excessive but still enough to be responsive to the arm movement was impossible and the operator always needed to adjust the speed, which never ended up being right. Also, The excessive amount of accel/decel would cause the wooden spools to wear the drive rollers in short order.

The goal is to keep the spool moving at a constant rate while using the dancer arm to adjust the speed as needed. Basically the controller should be trying to hold the dancer at it's center position and the input from the dancer should adjust the de-spool speed accordingly to keep it centered. It's slightly more complicated than that but this is what web tension controllers do for you. In my case the web controllers are more than I could spend (and overkill for my app) so I used a small brick plc and analog input/outputs along with a PID loop to do the job.

My app is using a sub fractional DC drive with analog input for speed control but there should be no reason it couldn't control an AC drive. I don't see any reason for a servo in this kind of application.

Cheers
Ken
 
+1 for the leather strap and wingut as well as the PID to keep the dancer centered. If you need to take it a step further, you can modify the PID control to act as a ratio detector so that the PID only acts to make adjustments to a scale factor applied to the line speed. This works only if the line speed is known (or controlled) by the PLC running the drive and will make for a system that can be stopped and started at any position of the winder (or payoff reel) with minimal ramp times and minimal disturbances to dancer position/tension.
 
The line speed is not known and the current drive is not capable. I didn't build the system but I really want to go in there with a good working solution the first time.

I've used the PID loops on basic AC drives. Would that drive be more than capable along with some type of drag system for the payoff?
 
You don't say why you think the current drive is not capable.

You do mention that it struggles keeping tension as the spool is used or tension is low. What is the symptom or problem? Drive not slow enough, get jerky at low speed, etc.

My guess is that your control loop (like ours) is always "behind the curve" and the reaction needed for it to keep up is more than the drive can handle or that it's not in control enough of the spool speed to prevent jerking or bottoming out of the tensioner.

Keeping the spool at a constant speed is optimal and the PID loop does this as long as the drive can keep up to the line speed + overhead to "catch up" or accelerate along with the line as needed.

I found that the current control method you're using is always reactionary and drive "tuning" issues will only exacerbate problems.

Cheers

Ken
 
The line speed is not known and the current drive is not capable. I didn't build the system but I really want to go in there with a good working solution the first time.

I've used the PID loops on basic AC drives. Would that drive be more than capable along with some type of drag system for the payoff?

If it has been almost working with just a dancer applied as a speed reference, then using a drive built in PID is likely to make it better as long as there are not frequent and rapid stops and starts. Having the PID in the VFD can be better since the feedback signal is not delayed by going through a PLC. The last time I worked on a Control Techniques VFD...well... It had me cursing all firmware designers... it was a very low end drive and I don't recall the model but there's no excuse for how poorly designed the keypad/display was to use just to make some basic parameter changes. We actually had to call tech support to walk us through some of the settings. I never call tech support. It has to be a catastrophe for me to call for help.

The system I improved had the PID in a PLC and was "okay" as long as the starts and stops were really long ramps. This occasionally would lead to a machine that would oscillate if it was restarted with a nearly full roll and the operators knew some mechanical cheats to workaround that issue. The change to a ratio detector made that issue go away and improved production by changing ramp times from 120 to 10 seconds iirc. I was dealing with rubber products and relatively slow line speeds compared to some of the wire winding applications I have seen.
 
Last edited:
I have programmed a wire winding machine. The AC motor payoff ran with a simple dancer. It was programmed using a PLC and some simple scaling. A range was made to where no adjustments were made. If the dancer was outside of the range then the output to the drive would increase or decrease depending on the orientation of the dancer. The more outside the range the more the response. The payoff dancer then fed the takeup dancer. The takeup dancer had a weight that helps with the tension and any jerkiness in the spool winding.
I hope this helps you out.
Regards,
 
Most of the time on these setups either you have a brake or if you use the drive to control it then you base the PID off of line speed, roll diameter and trim to make it run without issues

So either a live line speed or good calculated line speed is needed and sensor to detect roll diameter in which roll diameter decrease becomes your trim into the drive directly or PID controlling the drive speed.

From your description it sounds like some kind of brake would be the easiest to implement and cost-effective solution
 
You don't say why you think the current drive is not capable.
The current drive isn't capable of PID. Only a basic speed reference so in the best circumstances the dancer bounces up and down while it runs.

If it has been almost working with just a the dancer applied as a speed reference, then using a drive built in PID is likely to make it better as long as there are not frequent and rapid stops and starts.
These make long runs once setup. I think you've about sold me on changing the drive out. I'm thinking about a Powerflex 525 since it has a "Trim Control" mode.

I hope this helps you out.
That does, thanks!
 
I've spent several years doing controls in a wire plant. Most of the payoffs there received a speed signal from the machine to payoff off to tell it line speed. The dancer signal then trims the line speed. I usually had good look scaling the line speed to .6 to 1.5 and multiplying that number by the line speed. When you are in the middle of the reel, the trim should be around 1.
 
I have seen this many times the unwinder speed goes from min to max speed and back with the dancer waving from min to max like it trying to flyaway. It looks like everything is out of control. Well it is out of control.
I have never seen any fo those system ever work well no matter how well it tuned. I think it is dangerous for any personal in the area I have seen them actually pull the unwinder stand across the factory floor endangering everybody in the area.
The problem is that the dancer is doing all the control of motor speed. Under this circumstance you will never get control.
As stated above you need to have the drive receive a speed command from the line speed to be used as a base speed for the motor with the dancer connected and scaled to add to the command speed base on the dancer position
It should be scaled to a max range of 50% of full speed 25% would be better. Ideally you want to keep the dancer in the center position at all times, but as we know that not going to happen in this case because you need the dancer offset to set the speed. The best way I have found to control the unwinder with a PLC set the speed from the line speed and add to it the offset of the dancer position. Over the full run the motor seed will change to keep the dancer in the center position. It’s a bit challenging to do it this way it requires a some calculations and timers but it does work well.

If you need to maintain tension on the material then you will need a regenerative drive that can be controlled seamlessly from full torque forward to full torque reverse, in this case the dancer would be used to control torque
 
It hasn't been said here that I saw, but when using a dancer to tension a wire, I much prefer to use the dancer to control the drive in torque mode, not speed mode. A PID in a PLC is generally good enough for this, or a stand alone card if you like, or one in a drive.


Why torque mode? You are controlling tension. Tension is directly relatable to torque, and far less likely to get into heavy oscillation. It also eliminates a control loop (speed loop) in the drive. I took a bunch of payoffs and takeups on wire drawers out of speed mode, and put them in torque, and eliminated blowing fuses daily.
 
It hasn't been said here that I saw, but when using a dancer to tension a wire, I much prefer to use the dancer to control the drive in torque mode, not speed mode. A PID in a PLC is generally good enough for this, or a stand alone card if you like, or one in a drive.


Why torque mode? You are controlling tension. Tension is directly relatable to torque, and far less likely to get into heavy oscillation. It also eliminates a control loop (speed loop) in the drive. I took a bunch of payoffs and takeups on wire drawers out of speed mode, and put them in torque, and eliminated blowing fuses daily.
If running in torque mode, do you need a speed reference from the main line?
 

Similar Topics

Hi, We have a bag in box machine with a few payoffs that unroll foil. The inverters are PowerFlex 525 and the tension rollers mechanism is...
Replies
4
Views
788
Have some equipment we built a few years ago that we utilized the micrologixs 1400 plc and allen bradley powerflex 525 drives in. Will refer to...
Replies
10
Views
3,179
anyone have a idea on who makes a laser type analog output sensor that could be used to detect the position of fabric and trim a AC drive...
Replies
10
Views
2,746
Hi guys, We have a customer that is using a Yaskawa MPiec Motion Controller with some remote I/O to do a bagging application. They are using a...
Replies
8
Views
3,043
Hi all, I will try as best as I can to draw you the mental picture. I am winding flexible plastic film with a gauge of 60 micron and a web width...
Replies
3
Views
3,415
Back
Top Bottom