Help with cut to length program

Hi again,
I'm using a micrologix 1500bwa, a 2500ppr automation direct heavy duty encoder for length measuring, and a 2500ppr automation direct medium duty encoder for shear press position. Both encoders are single ended(totem pole) using only A&B channels. Both increment/decrement fine.
The problem I have is in controlling the shear presses position. It is constantly slipping past the high preset value. I've included the rss program. You will see that HSC0 latches the press coil and the HPR of HSC1 should unlatch it.
I think the problem might be a weak/leaking air operated clutch/brake, but would like some input on how I could improve the reliability and consistancy of the current control program.
Thanks!

Hello, Loundeb2,

I can't read your RSLogix file (my version is too old). If you could print it out as a PDF file, then anyone one the forum could read it, and you will get lots of comments. I might even have some useful hints. :site:

One thing you might consider to help offset a weak brake is to set stop a few numbers below the current stop, then pulse the coil until you arrive at the correct number. In other words, create a "creep" mode, so that you run at full speed until a certain value, then creep the rest of the way at a slower speed by pulsing the coil until you get the exact stop point. A "creep" mode or "dribble" mode is often used for this purpose in many other applications.

Loundeb2,

I would be remiss if I did not mention this famous Cut-to-Length thread where a machine was developed with Paula Stephens.

http://www.plctalk.net/qanda/showthread.php?t=11015&highlight=cut+length

Don't start reading this unless you have about 3 hours. It went on for months, but the end result was good.

Here is the Program in PDF Format.

Stu.....

Thanks

I did start reading PStephens thread, but couldn't relate to the program, plc, etc. so I stopped reading by page 4.
I did move the values around, but it would constantly cycle twice to between 2900 and 3000 pulses before unlatching and stopping. When i set a value under 600, the press just cycled constantly and would increment skyward. I tried using LIM and GEQ instructions in place of the HPR, same results.

You have an output O:0/1 on both rung 1 and 3. Not a good idea in my limited experience.

I did move the values around, but it would constantly cycle twice to between 2900 and 3000 pulses before unlatching and stopping. When i set a value under 600, the press just cycled constantly and would increment skyward. I tried using LIM and GEQ instructions in place of the HPR, same results.

Click to expand...

Yes, like Tom said, fix the double-coil error on Rungs 1 and 3. Rung 1 will never do anything, because it will be overridden by the value of the output on Rung 3. Just combine the two rungs together by making Rung 3 a parallel branch of Rung 1, with only 1 O:0/1 output.

Now, to create a "Creep" mode will take more than moving values around. EDIT: Rung 13 is the key. There, you have a number called L9:2 HSCO Length Preset. What you want to do is subtract some value (whatever it takes to stop the press, then creep up to L9:2. Say it is 5 or whatever length the press is over-shooting. Then create a new number, call it "Press High-Speed Stop" (PHSS). Set Press High-Speed Stop = L9:2 - 5. Then you modify rung 5. First, substitue an EQU insruction with the new PHSS for the old HSC:0/HPR. Then, create new parallel JOG branches for Rungs 5 and 6, where you Latch and Unlatch the Press Clutch Coil until the real Programmed Length is reached.

Now, I don't know the details of your press. You may be able to use the existing JOG rungs to jog without latching and unlatching the Press Coil. You will need some method to isssue jog commands. Perhaps a Down Counter to count from 5 (or whatever your creep mode is) to 0, pulsing the jog function each time until you reach the Preset Length.

The above is a "fix" to make up for your failing press brake. Ideally it is always best to repair the equipment instead of jury-rigging fixes. However it could be argued that a creep mode may should have been in the original program.

He has something bigger going on here. When he set the press rotation stop count to 600 the press should have stopped after 1/3 revolution (his base count was 1900 counts). So either his scaling is completely screwwed up and or the press count is never getting below 600 so the oneshot is not retriggering.

Keith

Loundeb2, two questions:
1- How is the coupling between your encoder and material?
2- Is material pulled always at the same speed, or it has some kind of profile, i.e. accel, constant and deccel.?

Please correct what tom@pton mentioned on #6.

IMHO, the first thing you need to do is get the thing to consistently go through a single cycle. Until you do that you can't correct anything else.

The way you code is written, if the press doesn't stop on the top limit at the end of a cycle, you are in trouble. The press count will likely not reset since the press may pass the top limit switch in less time than to takes to reset the counter.

As a test, replace T4:1.DN in rung 8 with HSC:1.HPR and see if you can get the press to go through a repeating cycle. Yur press may not stop in the correct position and it may drift but we can attack that as a separate issue. Just confirm it goes through a single cycle and you can adjust the stroke.

Keith

Reply

widelto said:

Loundeb2, two questions:
1- How is the coupling between your encoder and material?
The length measuring is taken at the entrance of the material into the rollform by the heavy duty AD encoder using an AB 3.85" diameter wheel contacting the material.
2- Is material pulled always at the same speed, or it has some kind of profile, i.e. accel, constant and deccel.?
The feedrate can be varied, but runs at a constant speed set by the operator for the entire run, usually around 50FPM, relativley slow.

My problem isn't the length measuring HPR latching the the shear press output, but more the shear press encoder HPR not unlatching it. The shear encoder is directly face mounted to the end of the press shaft via a coupling.
Thanks.

Click to expand...

Updated

Hi and thanks for all those who took time and interest to help me out.
I've attached an updated rss zip (sorry, I can't create pdf's). As you can see I've separated AUTO mode and JOG mode using JSR's. I'm able to get the press to cycle one top and stop at TDC with HSC1 preset to 925. I rebult the air valve and closed the leaks as well as adjusted the hub brake.
I still wonder why a complete rotation wouldn't be over, say 2300 pulse counts for a 2500ppr encoder?
Maybe I'll try a 360ppr encoder and see what that does.

Thanks, Lou

Lou:

Here it is converted to PDF. I use PDF Creator, it sets up as another Printer. It has a reasonable license, that allows commercial use.

http://sourceforge.net/projects/pdfcreator/files/


Stu....

loundeb2 said:

Hi and thanks for all those who took time and interest to help me out.
I've attached an updated rss zip (sorry, I can't create pdf's). As you can see I've separated AUTO mode and JOG mode using JSR's. I'm able to get the press to cycle one top and stop at TDC with HSC1 preset to 925. I rebult the air valve and closed the leaks as well as adjusted the hub brake.
I still wonder why a complete rotation wouldn't be over, say 2300 pulse counts for a 2500ppr encoder?
Maybe I'll try a 360ppr encoder and see what that does.

Thanks, Lou

Click to expand...

I´ve modified your program. I´ve added B3:1, and changed subroutine 2,3 and 4. Check your encoder cable, power supply etc, changing to 360 ppr is not going to solve the problem.

To create pdf docs use cutepdf it´s free and easy to use. Download it from http://www.cutepdf.com/