Need help with winding application VFD speed control

i am using two AB Powerflex 700s AC Drives. both have local encoders connected. one drive will be used on a winding conveyor and the other one will be used to feed a fiberglass sheet onto the the winding conveyor.

i am having trouble coming up with a way to calculate the outer surface speed of the material as it is being wound and the outer diameter is increasing. i would like to make sure that i slow down the winding conveyor as the outer diameter increases and match the speed of the feeding conveyor to the outer surface speed of the material being wound. i would like to keep the outer surface speed constant.

any help would be greatly appreciated!

I'm sure the mechanics of the system will dictate it, but would keeping the conveyor speed constant and putting the winder into a constant torque mode (as in maintaining tension) be an option here? just thinking out loud.

SilverLoop said:

i am using two AB Powerflex 700s AC Drives. both have local encoders connected. one drive will be used on a winding conveyor and the other one will be used to feed a fiberglass sheet onto the the winding conveyor.

i am having trouble coming up with a way to calculate the outer surface speed of the material as it is being wound and the outer diameter is increasing. i would like to make sure that i slow down the winding conveyor as the outer diameter increases and match the speed of the feeding conveyor to the outer surface speed of the material being wound. i would like to keep the outer surface speed constant.

any help would be greatly appreciated!

Click to expand...

Let me give this a try:
You have to set the speed of the winder to the speed of the incoming sheet. The speed at the outer diameter will slow inherently as the diameter increases. You do not have to slow it down.

Edit: And the outer diameter speed of the feeder will increase as its diameter gets smaller. You just need to run them both at the same speed in my opinion.

What says Peter?

automatic said:

Let me give this a try:
You have to set the speed of the winder to the speed of the incoming sheet. The speed at the outer diameter will slow inherently as the diameter increases. You do not have to slow it down.

Edit: And the outer diameter speed of the feeder will increase as its diameter gets smaller. You just need to run them both at the same speed in my opinion.

What says Peter?

Click to expand...

hmmm...okay i understand what you are saying. i believe the customer wants the feed conveyor to run at a constant speed and wants the outer diameter speed to be constant as well.

SilverLoop said:

hmmm...okay i understand what you are saying. i believe the customer wants the feed conveyor to run at a constant speed and wants the outer diameter speed to be constant as well.

Click to expand...

The speed is a constant.
Speed = distance / time. Whatever the size of the outer diameter distance (2 * pi * r) / time is constant if you are running at a constant rpm.

Is there a reliable way to get speed feedback right from the strip of material using a wheel and encoder?

Just curious, how long, thick, and wide is the sheet you are winding?

The speed is a constant.
Speed = distance / time. Whatever the size of the outer diameter distance (2 * pi * r) / time is constant if you are running at a constant rpm.

Click to expand...

I believe he wants the strip speed to be constant, which would make the winder speed variable, or slow down as the diameter of the roll of material increases.

The only other way i can think of is using a totalizer calculation. Every revolution the radius increases by one sheet thickness. Then calculate the rpm from that.

Gerry M said:

Is there a reliable way to get speed feedback right from the strip of material using a wheel and encoder?

Just curious, how long, thick, and wide is the sheet you are winding?



I believe he wants the strip speed to be constant, which would make the winder speed variable, or slow down as the diameter of the roll of material increases.

The only other way i can think of is using a totalizer calculation. Every revolution the radius increases by one sheet thickness. Then calculate the rpm from that.

Click to expand...

I was completely off on that one.

So it will be something like:
New RPM = Old RPM * New Radius/Old Radius

You can select a suitable time interval to read the new radius and calculate the new RPM.

You could use an ultrasonic sensor to find roll circumference. Knowing the roll circumference in feet divided into feed speed in fpm will give desired rpm`s for the roll.Through your roll gearing you deternmine desired motor hz. Then send your analog signal to your roll drive and set your min and max hz`s in your drive to match your min and max roll speed. Just some thoughts??

Gerry M said:

Is there a reliable way to get speed feedback right from the strip of material using a wheel and encoder?

Just curious, how long, thick, and wide is the sheet you are winding?



I believe he wants the strip speed to be constant, which would make the winder speed variable, or slow down as the diameter of the roll of material increases.

The only other way i can think of is using a totalizer calculation. Every revolution the radius increases by one sheet thickness. Then calculate the rpm from that.

Click to expand...

this application only involves one winding...it is not being fed by another roll that is unwinding. i also only have the two drives and their respective encoders to work with. the sheets will be of differing thickness and width so i need to come up with something that will work with each sheet as it is being fed.

This may just be a terminology thing, but when people talk about 'winding conveyors' I picture surface driven winders. So, SilverLoop, what do you have? A winder with the motor driving the center spindle or a winder with a belt that rides on the outer surface of the winding roll?

Keith

kamenges said:

This may just be a terminology thing, but when people talk about 'winding conveyors' I picture surface driven winders. So, SilverLoop, what do you have? A winder with the motor driving the center spindle or a winder with a belt that rides on the outer surface of the winding roll?

Keith

Click to expand...

actually what i have is a metal "spool" called a mandrel that has a rod in the center and the rod is attached to a motor that spins the mandrel. the feed conveyor feeds a fiberglass sheet onto the mandrel that is perforated and a vacuum is being pulled to hold the fiberglass onto the mandrel during the winding process.

OK, so you are center driven.

This kind of depends on how accurate you need to be. Since you say that all you have are the two motor encoders I assume you don't have any direct or indirect tension measurement deveices (load cell roll or dancer) between the feed conveyor and the winder.

With only this information you are in a bit of a tight spot. If your transmission torque losses are relatively low compared to your tension torque you may be able to simply use a torque limit on the drive to control tension. If you do this you can use the mandril RPM relative to feed conveyor speed to calculate diameter and modify your torque limit accordingly. This can be a little rough but may be good enough for what you are doing.

The totalizer calculation that Gerry M refers to can work. But with nothing to correct for caliper variation the diameter may ultimately have enough error to cause you a problem, especially if you calculate larger than what you actually have.

My guess is torque limiting may be your best bet if you can't use some sort of direct tension measurement.

Keith

Using

conveyor material speed / winder motor rpm

you can calculate a diameter that is accurate to 1/4 or 1/8 of an inch.

kamenges is too fast for me! and more in depth, too. what he says is correct. if you could or need to do it, load cell feedback will give you better wound roll quality by providing excellant tension control.

Tension control would be easiest for most controllers

What says Peter?

Click to expand...

I have avoided saying too much about winders even though there has been many threads about them. There are people that make winders and have the forumulas. I like to borrow from the experts instead of re-inventing the wheel when I can.

Some controllers can calculate the gear ratio between the conveyor and winder drive on-the-fly every scan ( .25 to 1 ms ). This sounds like what silverloop needs.

Can the Powerflex drives do calcuations?
The calculations:

ccps * cscale = 2*pi*radius*wcps/wcpt
radius=radius+thick*wc/wcpt

Where:
ccps is the conveyor counts per second
cscale is the conveyor position units per count
wcps is the winder counts per second
wcpt is the winder counts per turn
thick is the material thickness
wc is the winder counts per controller update

Now these formulas are pretty close and many would be happy with the results but you guys should know that I like to use the best algorithms available and that requires calculus. Using calculus provides EXACT positions and velocities. See this link.
http://www.apicsllc.com/apics/Ie_ias03/IAS4p5a.htm
There have been many posts lately about winders. Everyone should know this topic has been beat to death.
Some motion controllers can do these calculations on-the-fly every controller update.
It helps when the two servos can be control from one controller.

The easy way is to due tension control and avoid the calcuations but many don't do tension control right. You should keep the tension constant as the roll gets bigger. This means the torque must increase as the radius increases. Remember that torque = force x radius and it is the force ( tension ) you are trying to keep constant. Can the drives you are using change the torque on-the-fly? Another problem is that the torque applied by the motor must overcome the friction which means the net torque applied by the motr may not be the net torque that provides tension ( force ) on the material being wound.