Traverse Rewinding

[Peter Nachtwey]

Is there an algorithm I could use to handle this?

We got the basic code from Numerical Recipes in C. It isn't too bad in in or C++ but it would be a b!tch on a PLC using ladder. We enhanced the original code so that the user could specify position, velocity AND acceleration at the ending points. The normal 3rd order spline doesn't do this. We also added linear motion or constant velocity motion segments and this too is an expansion. For me this is just math but the part that takes a lot of time is the user interface. Windows programming is slow and tedious.

Believe me. It is easier to buy the technology than to create it yourself unless there is no value on your time.

 

[strantor]

If you get speed feedback from the nip and from the rewinder you can calculate roll diameter. This will let you determine your required traverser speed based on material width + gap.

What does roll diameter have to do with traverse speed? Traverse is measured in pitch; turns per inch along the length of the drum, has nothing to do with the diameter of the drum. Calculating roll diameter is definitely a good idea, for feedback to the torque loop.

Really??? Oh no... back when I was looking at this job I was told the hydraulic traverser wouldn't be an issue....

http://www.plctalk.net/qanda/showthread.php?t=84309

Who said that? It definitely wasn't me . Just poking fun, I think it was proved by more experienced people that I didn't know what I was talking about.

Been thinking about this...
Operator will enter Material Width, thickness, Reel Width, Line Speed (FPM) and Rewind Torque (%)
PLC will calculate the gap of each wrap.

The Rewind motor will need to control line speed based on the torque. (Torque is too high, then speed up nip roller).

That sounds backassward to me. In my experience, line speed is usually a constant, and the rewinder will speed up or slow down to maintain a set torque on the material feeding into it. In normal operation it will start out fast and slow down as it fills up.

The Traverse mixes everything up since it is a hydraulic controlled servo valve so speed is not adjustable (I don't think...)

Any thoughts...

is it a flow control valve or a pressure control valve?

 

[Critt]

What does roll diameter have to do with traverse speed? Traverse is measured in pitch; turns per inch along the length of the drum, has nothing to do with the diameter of the drum. Calculating roll diameter is definitely a good idea, for feedback to the torque loop.

I agree with that...

That sounds backassward to me. In my experience, line speed is usually a constant, and the rewinder will speed up or slow down to maintain a set torque on the material feeding into it. In normal operation it will start out fast and slow down as it fills up.

I agree with the speed of the rewinder, but what about the speed of the nip roller feeding the rewinder. What tells the nip roller the speed to run? Also, if the torque starts to get too high or low the nip will have to slow or speed up... If the rewinder starts and speeds up to the required torque and adjusts up/down as needed... what gives the nip the speed...

Is it a flow control valve or a pressure control valve?

I believe it is a flow control valve... (4-20mA)

 

[Critt]

We got the basic code from Numerical Recipes in C. It isn't too bad in in or C++ but it would be a b!tch on a PLC using ladder. We enhanced the original code so that the user could specify position, velocity AND acceleration at the ending points. The normal 3rd order spline doesn't do this. We also added linear motion or constant velocity motion segments and this too is an expansion. For me this is just math but the part that takes a lot of time is the user interface. Windows programming is slow and tedious.

Believe me. It is easier to buy the technology than to create it yourself unless there is no value on your time.

Thanks for all your info... I was looking at your website and trying to decide what to use for the controls.. we have a Compact Logix L33ER that we would like to keep....

The traverse distance is 0 to 16 inches max

 

[strantor]

I agree with that...




I agree with the speed of the rewinder, but what about the speed of the nip roller feeding the rewinder. What tells the nip roller the speed to run? Also, if the torque starts to get too high or low the nip will have to slow or speed up... If the rewinder starts and speeds up to the required torque and adjusts up/down as needed... what gives the nip the speed...
Y

Well, assuming the nip roll is the only thing between the unwinder and rewinder, YOU control its speed. It can be setup with a stupid simple V/F drive, controlled by a speed pot. The rewinder and unwinder are both slaves/followers of the nip roll. Nip roll is what sets the constant line speed.

If there's some process before the nip roll, like extrusion or annealing or something, and it demands its own constant speed then I would just remove the nip roll; as far as I can see, it wouldn't be needed, and would only cause problems. Or, if in that scenario there was some reason you HAD to use it, then it would have to be painstakingly synchronized with the speed of whatever upstream process.

I believe it is a flow control valve... (4-20mA)

Then (AFAIK) it should be a speed control. Not sure why you wouldn't be able to control its speed.

 

[Peter Nachtwey]

The Traverse mixes everything up since it is a hydraulic controlled servo valve so speed is not adjustable (I don't think...)

Any thoughts...

The speed of the traverse is 'geared' to the encoder. Actually the encoder is used to index into the curve/cam/spline. This way if the encoder slows down, so does the traverse. This is basic.

 

[JRW]

Critt
Can you cam or even gear the winder encoder to the traverse?
The traverse would be a slave to the winder encoder
What functionality does the AB have?
Or the ABB drives ?
Study post 7

 

[Critt]

Not sure if I can Gear the hydraulic traverse in the AB controller... been playing with motion logic but I don't think that will work...

I found this about the hydraulic servo valve
"The spool position is proportional to the input current and with contact pressure drop across the valve flow to the load is proportional to the spool position.

 

[strantor]

I suspect that quote is missing some critical punctuation marks. I'm not 100% sure understand it, but I think it's proclaiming itself to be a flow control valve, and if that's the case, then controlling its speed should be pretty easy. Just vary the supply current.

 

[Critt]

Hello all...

I was onsite today and testing with the hydraulic position valve... I think I may have a huge issue
As most of you have suggested the servo valve controls flow, well you are 100% correct.
The higher the mA signal to the valve the faster it will travel to the end of the stroke.

If I give it a 4mA signal after it is extended it stays extended. I open the loop and the valve returns but at a slow set speed.

Will this valve work for my application... I need to traverse in both directions

 

[strantor]

do you have a part #?

 

[Peter Nachtwey]

Hello all...

I was onsite today and testing with the hydraulic position valve... I think I may have a huge issue
As most of you have suggested the servo valve controls flow, well you are 100% correct.
The higher the mA signal to the valve the faster it will travel to the end of the stroke.

If I give it a 4mA signal after it is extended it stays extended. I open the loop and the valve returns but at a slow set speed.

Will this valve work for my application... I need to traverse in both directions

A servo quality valve will allow the actuator to go both directions.
Is there a valve that switches the direction of flow? If so this is not a good hydraulic system design.
You should have a servo quality valve. Usually these kinds of valves require +/- 10 volts or +/- some current in the range of about 20-50ma. Positive voltages or current should make the actuator extend and negative voltages or current should make the actuator retract.

This is going to take a long time just trying to figure out what should be done. I know but you don't seem to want to take my advice. There are people that can put this together for you. This company can do it.
http://www.hydraulicmotorcontrols.com

 

[dcooper33]

I don't know if you are beyond the point of making design changes, kind of sounds like you are, but if possible I would take the advice from post #2 and switch to a servo-driven linear actuator for the Traverse winding.
As Peter has pointed out, what you are trying to do with the hydraulic positioner can certainly be done, but you're better off buying a controller made for the job or hiring someone to develop the software for you.
Controlling the Traverse position to maintain pitch is fairly simple with an electronic gear or jog to a speed reference from the winding drive.
You said you already have the Compactlogix, look into the Kinetix platform.

 

[Critt]

Strator - Servo valve - Part Number is MOOG Industrial Division, 62-110

Peter - I don't think we want to hire in for this job...(No money for that) We are however, looking at the possibility of purchasing one of your hydraulic controllers....

Its funny now how I post this before and everyone tells me that I am fine with the system... Now everyone is telling me I need to upgrade the traverse.... AHHH okay done venting.

 

[kamenges]

You don't need to upgrade the traverse hardware. You just need to control it right. Throwing different hardware at it won't automatically make it all better. There is knowledge involved in making the system work. That knowledge is very difficult to convey across a forum three paragraphs at a time.

You obviously undervaule your own time. Does it really make sense to spend 40 of your own hours to accomplish something someone else could do in 8 even if you paid twice as much for those 8 hours? The only exception to this that I can think of is if you are designing a product that you intend to make multiple copies of and you need to be able to replicate the concept. But it would probably still make financial sense to bring someone in to directly show you how to do the first one. There is alot the forum will miss because we can't look at what you have doing what it is doing. And you will likely not provide the pivotal piece of the puzzle because you don't recognize it as the pivotal piece.

Keith