Press Die Safety

[blueheeler 0]

We are using a SLC500 with a HSCE module and a photo eye embedded in the die to monitor the 360 degree stroke of a punch press. The PLC checks to see if a part is ejected shorted after the bottom of the stroke and to see if the part is jammed at the top of the stroke and stops the press if it determines there is a problem. We seem to have a greater problem with die crashes since the stroke rate has been increased. The stroke rate is 100 strokes per minute. Are there higher speed PLCs available now that will do a better job than the older SLC500s? I would be interested in any other suggestions on this subject. Thanks.

[TimWilborne 108]

Probably more important is how are have you written your program? Are you just addressing your HSCE in your ladder?

[JeffKiper 9]

Tell us what is going on. What re the symptoms? Does the strip hang in the feed section before the cut off? What type of cut off is it? A cut and carry, shear and fall, punched out like a slug? What type of press? Mechanical, hypraulic, gear, servo. What was the speed of the press? What is the stroke lenght of the press? What type of feeder do you have ? servo, air, mechanical gear (hitch) Does the die have pilot pins. Whay have you all tried Air blowoffs, stripper pins? Adjusting the feed advance? does the die have a fringe notch?

[Leitmotif 0]

What I think Jeff is getting to is Is your problem in the PLC or in the press? What if the machine itself cannot go any faster? The mechanism has to have enough time to do its work - speeding up the PLC will not improve this. Dan Bentler

[JeffKiper 9]

That is where I am trying to get to. I have done several speed upgrades and you can make it work. We just need a few details to help him figure out where the problem is.

[RandyB2004 0]

If it is not a press problem, what series 500 are you running? 5/01, 5/02....etc. The higher series do scan faster. Or maybe you need to go over your ladder and see if you can optimize it to make it scan fast. Rookie suggestions from a rookie programmer.

[Armadillo852 25]

100 strokes per minute, but what is the stroke length? I am thinking there isn't enough time to process it. If there was, I am sure AB would have come out with a die protection package already. Like they do for the Clutch Control. Or you really need to look at a Link Electric or a Toledo Integrated Systems. These are designed to be used at high speeds and low speeds. Just a thought. Link Electric Toledo Integrated Systems Edited 23 May 2010 by Armadillo852

[JeffKiper 9]

if you look at AB and search for Omega system. You will see that AB did die protection with the SLC family. They opted to use the Logix going forward because they could lock routines and not the entire processor. Toledo also uses AB PLC as well. Link does a great job with hardware side of things but interfacing with their hardware sucks. read "Blackbox" The HMI side of their equipment could use some updating. I like Link from the overall system standpoint compared to Wintress but Wintress HMI kicks Links backside I have done several homemade die protection systems using SLC. One of the problems are that people try to use them for everthing else as well. So once you setup a press to run 50 SPM at a 8" stroke and you thing everythingis great it is until you increase ram velocity to a point that the scan time becomes a problem.

[blueheeler 0]

Thanks for the suggestions everyone. Is there information available on arranging the program for faster scan times? The HSCE is used in the ladder logic. It also operates the hitch feed. It has several subroutines and production rate calculation subroutines have also been added. The press makes brads (finish nails). It hammers a head on the end of the wire band and then cuts the brad strip off. I'm ignorant of the mechanics of the press and will have to do some investigation to answer some of the other questions.

[JeffKiper 9]

What has been added to this system since it worked correctly? You said a hitch feeder. Those hitch feeders that I am use to using are driven off the crank by a throw. But you said this HSCE drives it so can you explain?

[TimWilborne 108]

Maybe I'm a little confused...maybe the program isn't written the way I would...but what are you using the HSCE for then?

[JeffKiper 9]

I ASSuME that he is using it as a PLS. I guess that is like everything else let's add a rate counter to it, then what about adding a PLS, and a , and a ,and a. You know you been there and done that. TW how would you do it? Look at the position in 1 degree increments and use LIMit to fire a output?

[TimWilborne 108]

I think I'm missing something. Pretty much we have a prox to see the top of the stroke? We're looking for it to be made and not made and then made again to check some logic in a normal scan? Why not just have it connected to a regular input? My modern day press experience is limited but I'm not seeing where the HSCE is actually being used

[JeffKiper 9]

Maybe it is me that is missing something. I assumed (probably where I went wrong) that he was using the HSCE to count pulses off of the crank to handle the Programmable Limit Switch (PLS) functions of a press control. Then using a standard input to bring back the Photo Eye for part ejection. I still use Top Dead Center sensors to verify that the press is still rolling and we haven't lost the pulses from the HSCE. TW Where you thinking of a hydraulic or mechanical press?

[TimWilborne 108]

Mechanical Your idea of what is going on makes more sense. Most of them I have seen has rotary switches or proxes at various degrees instead of an encoder. He mentioned the TDC prox so I just guessed that was what was hooked to the HSCE. I guess well have to wait form more clarification from the OP to be sure

[blueheeler 0]

Let me try to clarify what we are doing. We use an encoder, TDC prox switch and HSCE module to monitor the position the press crankshaft is in. The solenoid for the pnuematic hitch feed is turned on at a certain position (degrees) and off at another. The die safety has a part ejected program and a part jammed program. The die safety part ejected program must see the cut off part cross the photo eye in the die within a certain window defined by two degree settings. The press stops if a part is not ejected. The part jammed program must not see a part using the same photo eye during another window defined by two degree settings. This window is near the top of the stoke and stops the press if a part is jammed. The photo eye sensitivity is adjusted by the operator to see the actual part as it falls and not the header that enters from the side and hammers a head on the wire strip that is the next brad strip ejected. I hope this helps everyone understand the problem.

[JeffKiper 9]

So what has changed? We know the SPM has but, what else since it last worked right? What type of hutch feeder do you have? I have never seen one that had air on it for feed advance. A make and model would be cool so I can see what it is. Im always up for leaning something new. Did you check your PM?

[blueheeler 0]

It is an in house made hitchfeed using two pnuematic grippers and one push cylinder. And it operates from one solenoid valve. When the solenoid is energized the rear gripper closes, the front one opens and the push cylinder feeds the band of material. The oposite happens when the solenoid is de-energized. The die safety part of the circuit never worked perfectly. How could it? It depends on several adjustments by the operator. And the 10 presses are all set different. I thought improving the scan time might improve the accuracy of the settings. If the press is turning 90 strokes per minute and the scan time is 10ms, then the crankshaft passes thru 1 degree per 10ms scan. If the processor takes more than one scan to process the ladder program, then the settings are not that accurate. How many scans does it take to process a 200 rung program broken up into 5 subroutins and read the outputs from the HSCE module?

[JeffKiper 9]

The scan time will vary with programming stlye. If you have a loop in the program it will vary. There are ways to program that make scan time good and even more ways to make it bad. If you can post you program we might be able to help.