Photoeye Make Breaks Logic?

[Chris Elston]

How do you guys program your photoeye make break checks? Where you watch for a photoeye to go on and off to verify that your part is present and your photoeye is not stuck on? I'd like to see some fresh ideas from other people.

[Big Country]

I just use the simple latch and unlatch. But usually when I use this its on prox sensor and I have an ejector that is removing the parts from a machine. Once this ejector has been up for x amount of time I shouldn't have any sensors on at this time and if I do then I know that there is a problem

[TConnolly]

Here is how I did it for a part detection photoswitch that must detect the removal of a processed part and placement of a new part.

[panic mode]

it depends but normally i base this type of checks on sequencer. every condition that comes on during cycle, must go off sometime during that same cycle. if for whatever reason input/condition doesn't change (stays always on, or always off), it get's detected within that one cycle (machine doesn't make few thousand of bad parts before somebody notices problem or consumer complaints start pouring in).

[Chris Elston]

This is my preferred method as well. Step 1 = check for part present Step 2 = do action Step 3 = check for photoeye OFF In some rare case of high speed machines and parts that "feed" well, the part behind the first part in an dead nest pick and place will actually advance forward and into position before step 2 is complete. In those cases, it's hard to make logic to make and break because part #1 was present, then at step 2 you are moving up out of a nest lets say, then the part #2 starts to move forward, before step 2 is complete. By the time step 2 is done or clear from moving part #2 is in position for pick up on the next cycle, then step 3 is evaluated and whoops! photoeye is still on....

[ECSI]

I would activate a timer when the part is detected. If the timer times out then the part hasn't cleared the area and you can shut down the conveyor, sound an alarm, or whatever else is required to stop the process. The timer setpoint would be a fraction of a second longer than what it would normally take for the photoeye to clear. If the conveyor speed is adjustable, then the timer setpoint would also have to change accordingly.

[IO_Rack]

We usually program a differential pulse from the photoeye. This will set a latch for rest of the process. The logic for the photoeye is set to 'ON' = 'Good'. A continuous 'no good' count will stop the process.

[Clay B.]

Ran into this problem myself. To get around this I made a seperate section of code just for the part there/gone and next part there. Then fed the condition back. Another thing I have done is on a really fast process I have an A part present and then a B part present so the action done to part A (orientation check) was not transefered to part B I did this by setting a bit that set orentation for part A then activate code to check for part be while I went thru the next steps with part A. If part was detected during this time then use B code. If no part detected then go back to A code. Thankfully I did not need 3 levels. At full speed part was present only .08 seconds and next part was .05 + seconds behind it. IF part was oriented wrong the my process could take 1.3 seconds. The part that tripped me up was the reset. I needed to know when I was done with part A and working on part B and when I was done with part B. I actually tried to watch this machine run parts while creating this code. Tossed my cookies after about 5 minutes. I then created a index pointer that loaded the sensor times into a bank of memeory locations. By seeing that the times where consistant I could see htat the sensor was working correctly and could create display and alarm code from it.

[rpraveenkum]

Photoeye Make Breaks Logic in rockwell samples file Read terms & conditions, click I accept http://samplecode.rockwellautomation.com/i....hcst?dID=52392

[Bob O]

That looks like it only the terms and conditions

[rpraveenkum]

Once you click I accept button you will get zip file download it contains the sample files

[Bob O]

All I get is the Terms and Conditions in PDF format. Something must be stripping it off. Bob

[rpraveenkum]

ZIP contains ACD file & PDF file ACD is the sample file pdf is the Terms and Conditions

[Bob O]

I don't recognize the .ACD extension so may be thats it but I would think that it would show up asking what I wanted to open it with. Could you post that file in PDF? Thanks, Bob

[TimWilborne]

ACD is for RsLogix 5000

[Bob O]

That could explain it. Thanks and have a good 2007.

[Chris Elston]

i don't have RS 5000, can someone open and PDF it? Photoeye_Make_Break.ACD

[TimWilborne]

PDF Photoeye_Make_Break.pdf

[Chris Elston]

Well GOOD GRIEF!!!! That's my own logic! http://forums.mrplc.com/index.php?autocom=downloads&showfile=12 Who put that in the Rockwell software downloads? Someone went through the time to convert my origional RS 500 program into a RS 5000, then uploaded it....geez....

[TimWilborne]

I didn't see your name or MrPLC on it anywhere . Guess no credit given

[Chris Elston]

I didn't do that in the orginal sample snipplets I did. That was download # "12". it's an old one....ah well, doesn't bother me, just suprised to see it up there.

[TimWilborne]

Must be an oldie but goodie

[Chris Elston]

Here is my new and improved make-break logic. Since the PDF above…new meaning about a years worth of run time on several machines…. I created a word to monitor my make-break sequence, I've been in habit of doing this because it saves on creating bits, eliminate bits, element busy work by not having to create a symbol and description for each sequence. Hence save on engineering time and make the company a profit...anyway that's for another day. My sequence is based on steps, think way back when you did BASIC in school step 10, step 20, etc... So the first sequence is step 10, the station started. I have a bit I call "OK" that starts each station. I know that the normal sequence has started. The sequence that does the actual work. Once the station started, the logic is indexed to step 20, check to see if the sensor is "ON". This condition for the way I write logic is a given because you have to have a part presents before the station is "OK" to start. Once the "On" condition is verified, somewhere in the cycle the sensor has to go "off". if it does, then step 30 is reached. Then finally, the actual station is complete and "DN" bit comes on and resets the make-break check logic. If during the station sequence the sensor did not go "on" and "off" or transition, then this make-break checks chokes up and stops at a step number somewhere. Now looking at the alarm PDF, you have a couple of choices. You can use a NOT EQUAL to "0", which monitors all conditions of the photoeye. Did it come on? Is it stuck on? Is it stuck off? Or you can do as I did and put EQUAL to step 20. If the check logic gets stuck at step 20, then the sensor never shut off. Hence use a timer to set a bit in an alarm word. So it's a little bit much of allot of logic, but it's what I've used and morphed into what I got today and seems to work well on machines I've used in the past. So it's been proven and seems to work. So that's why I was asking what others do, just to get a fresh idea and maybe revamp what I got now that I am using for make-break logic. make_break_ladder.pdf make_break_alarm.pdf