Cut to length problem

I recently aquired a project that requires a cut to length application using a pneumatic cyclinder, plc, and a 2500 ppr encoder to control the cut. When I aquired the project they were having problems maintaining the length, so I rewrote the code and now the length control is +- .115. The problem that I am having is that when they speed up or slow down the length becomes short or long. Then they level out the speed the length is right back. Is there any way to compensate for this speed up and slow down issue.

The problem that I am having is that when they speed up or slow down the length becomes short or long.


i had the same problem before ,

but let ask you som thing is the feeding system depand on
the rotery encoder or how ?

is it a motorized feeding or a pnematic cylinder ?

give more details, i might be able to help you .

You didnt show the method used to obtain the length. With a variable speed machine the math involved will change dependent on the variables involved. Sounds to me like its possible your math equations are not setup properly to obtain the set length regardless of speed setting.

Show the code for that part and see what everyone thinks.

There is not enough detail in your question to give more than a general answer. What does the pneumatic cylinder do? Is the material stopped when you make the cut, or do you cut on the fly? What brand of PLC? Are you using a high speed counter module to read the encoder?

The general answer is yes, you may be able to add some form of speed compensation and improve accuracy. The details of how to do it and whether it will be worth the effort depend on the details of your system.

steve is right ,

if you are using the encoder to count pulses and then when cv=pv
cut the sheet , in this case you can control the length by adjusting
the HSC values ,
but if the pnematic cilinder should be opened until the buffer
stop the movment , then you should use timers to control the
feeding and the cutting like i did in our machine .

Sorry for the lack information about the system. The control system uses an Aromat plc with a high speed counter card in the back plane. The HSC monitors a 2500 ppr Quad encoder that tracks a ply of paper that is being wound on a mandrel of a core winding machine (this is a seperate machine). The operator enters the desired cut length ( from 0 to 500 inches)into the HMI and starts the machine. Then when the plc has determined that the correct amount of product has passed the saw blade it sets an output on the plc, that controls a pneumatic cyclinder (for in out control)attached to the blade to make the cut. This cut is made without stopping the machine. One other piece of information that is vital, is that the mechcanical head assembly that the blade is attached to is also pneumatic and a roddless cyclinder. This cyclinder is used to let the blade float, and when it enters the product it matchs the speed. During a constant speed the control system maintains the length, but during the speed up and slow down it varies.

Calculation:

Ply Width/Mandrel circumference=Cosign

Product Cut length/Cosign=Web Length

Web Length*pulses per inch of web= cut setpoint

In addition to a variation while the line is accelerating/decelerating, I would expect to see a difference between the cutoff length at constant 100 FPM and constant 500 FPM.

Presumably, you are maintaining a constant air pressure to the cylinder and the distance between the blade's rest position and the edge of the web is constant. Given that, it will always take the same amount of time for the blade to reach the web and start cutting. But at different speeds, that constant time represents a different amount of paper passing by a fixed point. Your speed compensation should add or subtract from the cut setpoint. At higher line speed, you command the cylinder to extend at a slightly lower preset.

I don't know the details of the Aromat HSC, but many brands include a 'counts per time' value in the data that the HSC reports to the PLC. That would be the basis of your speed compensation. Also, if the HSC has direct-acting outputs that aren't subject to the PLC's scan time, you can use that to eliminate additional variation.

this is the problem of pnematic ,

every time the pressure is diferent , because the speed is
adjusted by adding or reducing pressure in pnematics ,


i think that you will get over this problem
by aplying a pressure switch , when they increase
the pressure ( speed up)the program will decrease the value of the
HSC ,for example . and when they reduce the pressure (slow down )
then the program will increas the value of the HSC,



maybe im wrong , but in my machine it just worked.

Be a little careful with your speed compensation. If this is a continuous web process this is a case of recurring errors. That's why you only see the problem at speed change.
As Steve correctly points out the reason for the error at all is that the saw takes the same time to move regarless of speed but the amount of web traveling under the saw before the start of the cut is different. But here's the tricky part. Let's assume that the cut point changes by 0.25" in a given speed change. So the product length of the product run during the speed change is off by 0.25" (too long if you sped up, too short if you slowed down). However, the thing to keep in mind is that, assuming you don't change anything, at the point of the cut because the piece jsut being cut off is 0.25" too long, the piece whose leading edge you just cut is 0.25" too short. But, assuming that you don't change speed again, you will be 0.25" too long when you cut the trailing edge of the same sheet. So the sheet length will come out correctly and will stay correct at any given stable speed.

The thing to watch is where you reset your counter. If you can reset at some point that you know the saw is in contact with the material then you can successfully use Steve's suggestion to subtract counts as the speed increases. But the important thing is to make sure the counter reset and the saw move output are not logically tied together. You need the saw move initiate to change but the counter reset point cannot.

Keith