Rewinder Speed & Pressure

[Leonightiz]

Hi All… I work for a company that makes bags. Operators of the machines that make the bags need to take the bags from a cardboard box that the bags fall into as one long line of bags (between 350-4,000 bags/per lane. Each machine can have 1-8 lanes of bags. Each bag can be 1” wide to 20” wide.
Many times the customer will ask for these bags to be rolled onto a cardboard core which is about 3” in diameter. So, the box of bags will be on a table and few feet away is a “rewinder” machine that the operator threads between three rollers and then directly to a core with a side plate to divide each lane. Each “rewind” machine has a manual level to increase/decrease speed and pressure (from the front with pads that guide the bags).
The bag thicknesses vary but only have a few thicknesses: .0015”, .002”, .0025”, .003”, and .004”. It seems that the thicker .004” bags rewind pretty easily at whatever the speed and pressure is, but the lighter ones vary dramatically. There seems to be no set numbers to make use to make this process easy to do. It is obvious that the lighter the bags are, the easier it is for them to want to bounce up and down in the air make them hard to stay straight without going very slow, like 8-18 feet per minute. I would like to get them to go at least 25 feet per minute if possible, to free up time so they can work on the other packaging while it is running and not have to babysit this machine. I have noticed that a lot of times when the Rewinder is running okay, it’s seems like the speed is 2-3 times the pressure or vice-versa. Hopefully, I have explained this well enough for some input. Is there a formula for this? Something based on size and thickness of the bag type that I can figure out the best speed and pressure? The rolls are to be tight when done, but at the same time, the pressure can’t be too high or it has problems either tearing the bags apart where they connect or it simply is so tight it bunches the bag up and won’t keep them flat to roll nicely. Any ideas or help would be greatly appreciated. Thank you in advance.

[parky]

I have seen a number of similar systems, this is my take on it, firstly, a dancing arm is often used (this is usually chosen to be sized & weighted that gave the best results over the range of product) & the tension could be altered with a clutch i.e. 4-20ma signal, this often stopped snapping of the film, may or may not improve your system it certainly did on the one I did (not my idea the machine Mfg'r came up with it).
The other was applying the correct tension on the reel, this used a ultrasonic sensor to measure the diameter of the reel & altered the tension, this one was the reverse of yours as it was the feed off the roll rather than feeding on the roll.
The other thing was a trial & error on one machine having recipes for the types of materials i.e. thickness & size, initial settings were found, held in recipe , the system then calculated the move from the recipe as the reels reduced in size.
Ther was a trial on the dancing arm to measure the movement however, I was pulled off that job to do another so I never found out if it worked.

[drforsythe]

Regarding your question about formulas for figuring this out, I am not sure.

If you are interested in a hardware solution: We have a winder for a cast film line at our plant that uses a load cell on a roll that our film rolls over. That load cell is connected to a tension controller that controls the amount of slip on our winding clutches. The tension control system is made by Maxcess International. This system allows us to maintain a consistent tension on our rolls independent of speed or material thickness.

[drbitboy]

Winding and web handling is a big field. Oklahoma State University had a department dedicated to it, and they had commercial sponsors whom they would ask to direct the research topics: wrinkles; roll formation; etc. The benefit to the sponsors was exclusive access to research for a set time before the papers were published. I suggest you try to find that department via The Google and see if you can get access to their publications (e.g. I found this and this).. I would also be surprised if no one on this forum worked either for one of the sponsor companies at some point or has winder experience; they would have more information than me.

One thing I know is that you want to control the tension such that it decreases (linearly?) with the wound length, otherwise there can be roll formation problems (telescoping).

[JeffKiper]

Do a search in this forum for taper tension. You will find a few of my older post asking the same questions and getting some great help.

[Puddle]

Originally Posted by drbitboy

Winding and web handling is a big field.

It's almost an engineering discipline by itself. I've been looking into doing something similar to improve a process where the web is being wound onto a center-driven A-frame and it's hard enough to track down a supplier for load cells to mount a roller on, never mind getting it set up properly.

I see dancing rollers as a more basic way of maintaining tension and because they can move around, the control behind them doesn't have to be as precise.

Originally Posted by drbitboy

One thing I know is that you want to control the tension such that it decreases (linearly?) with the wound length, otherwise there can be roll formation problems (telescoping).

That's pretty much what I'm trying to iron out of this machine as the current (lack of) control results in telescoping on larger rolls.

[JeffKiper]

We gad a center rewind that was on loadcells. It controlled a lot better than the loadcell winder we had.

[GaryS]

Keep it simple

Use load cell dancer to measure the tension on the web
Use the web tension to control the torque on the winder
No clutches needed
The tension is the feedback for torque control
Just make sure you set the winder speed high enough to wind the core, that will be the highest speed as the material builds up I the winder the speed will go down but the torque will go up
I have done this many times

[kamenges]

Originally posted by GaryS:

Use the web tension to control the torque on the winder

This is application specific. It works much better on high speed winders than low speed winders and on winders for web with a high modulus of elasticity. The OP is hoping to get to 25FPM. That is low speed. Assuming the OP doesn't want to throw more motor at this than he needs to, that means a lot of gearing.

Low speed means high gear ratios
High gear ratios means increased transmission losses.
Increased transmission losses means loss of torque feel at the spindle.
Loss of torque feel at the spindle means compromised torque command-to-tension linearity.

Its hard to tell from the description but it looks to me like there is nothing putting tension into the web as it exists the box. Thin webs without tension will walk, wander and flop around. There is not enough cross-machine dimensional stability to resist cross-machine bending. It is why loop control works great for steel sheet but not so much for 1 mil plastic. Push on the side of the steel sheet loop and it won't move. BLOW on the side of the plastic sheet and it moves.

If the web isn't controlled coming right out of the box there isn't much else you can do.

Keith

[drbitboy]

Winding Machines: Mechanics and Measurements, written by Dr. David Roisum and Dr. James K. Good.

[GaryS]

To answer your question
I have found over the years of doing controls systems that torque control is the best way to go
for both winding and unwinding applications.
But there are some additional things you need to keep in mind when doing the controls.
When doing winders you need the higher speed with a smaller core diameter and lower the speed reference and the diameter goes up. In all cases the speed reference need to ne be faster than needed to match the web speed. I like to keep my speed reference just a little above what is necessary to wind the material that way if the web tares out, it does not go into runaway until I can bring it to a stop.
The tension is best detected using a load cell dancer system. You should also measure the diameter of the to set the rollup speed.
As the windup roll size increases the torque required to maintain the web tension needs to increase because the weight of the roll will increase.
Years ago the standard was to use a dancer following the roll diameter and try to match the speed to maintain the web tension. It didn’t work very well you had to always keep adjusting the system as the product changed.
For unwinder’s you need to set the speed reference to be slower than you need and you must be able to rapidly and smoothly shift between motoring and braking depending on the weight of the material roll and the web tension you need. A regen drive works well for this ( here’s where a common buss drive configuration works well ) if the roll is heavy you may need to drive the roll as some speed to maintain the web tension as you unwind the roll, but as the roll diameter goes down the you will have to shift to braking mode again to maintain the web tension. You must also be able to stop the roll quickly in the event the web tares out.
Here again you need to know both the web tension and roll diameter to do the job correctly .
Here again torque control is the best way to go but with an unwinder control you need to control the torque in both the motoring and braking mode.
High gear reduction in order to run at low speed is really not a problem. Modern vfd’s can run at very low or very high speeds. In braking mode they could go down as low as 1 or 3 hz and high speed the can run at over 150 hz. And in motoring mode they can run from 1-2 hz to 120 hz without any problems.
I have run them in that range many times in that range the only recommendation is that if you think you may need to run the motor over 3500 rpm then have the motor balance and rated for the higher speed.