Scary...Pic Heavy

[robh 12]

This is what I am getting ready to replace. This mess has been controlling one of our production machines for 35-40 years. I get to replace it with an Allen Bradley MicroLogix 1500 and PanelView 1000 color. Notice the drum switch, and the way the enclosure door opens up and is held with a stick! It has ran remarkably good considering. But I think the eagle timers and drum switch were obsolete before I was born.

[TimWilborne 108]

Wow, I'd say that one has paid for itself a time or two

[robh 12]

Yes it has. Despite the looks, we were having a tough time deciding to go ahead and do this right now. We are going to be replacing the hood and external frame work of this machine, and well, all of the electrical is attached to that and we don't have very good records of how the machine is wired other than chicken scratch drawings from over the years. Probably could have made 5 or 6 more years, but while were into it, might as well make it right!

[splicer480 0]

Good luck with the upgrade. Without having some detailed schematics, you've got your work cut out for you. One question: What does the Allen Bradley Interface screen tie into? Picture 1 Top right corner.

[Andy_P 7]

Best of luck with that job. No doubt you will be given a wish list compiled over the past 35-40 years of things people would like the m/c to do, but previously not possible / cost effective to do! I have had a hand in a few 'upgrade' jobs such as this. While there will undoutedly be a few headaches along the way, when it all finally falls into place the job satisfaction factor is pretty high. Andy.

[Clay B. 1]

Well as bad as that thing looks it does have several good things going for it. For one most of your functions fire off the drum timer. What that means is you can create a master timer and do compartive instructions off of it. Other smart move your making is using a touch screen. Forget a switch and you do not have to get out a roll of wire and a knock out set. As long as you have a hrough understanding of the process this will be a job you will see alot of benifits from. Know from experience how touchy those drum timers can be, and that one looks extra crispy.

[Guest]

thanks for the pictures ... please come back and post the final upgrade when you get through with it ... I KNOW that you won’t get hit by this particular bus, but your project reminds me of a story (supposedly true) of a similar upgrade to a “rough-cut” lumber mill ... BIG logs were pushed back and forth through a large saw chamber by hydraulic rams ... the whole thing had worked well enough for many years - and then the decision was made to upgrade the controls to a new-fangled PLC ... the old-timey relays and drum switch were replaced with an equivalent ladder logic program ... or so they thought ... things were fine for several months - until one day the power to the plant flicked off and then back on again while the machine was running ... unfortunately the PLC programmer had made use of common “seal-in” type rungs to mark the various “Log in Position A” and “Log in Position B” type conditions ... when the PLC came back up in the run mode, the “seal-ins” dropped out - the PLC promptly forgot where all the logs were - and it signaled the hydraulic rams to shove the next log into the ALREADY OCCUPIED saw chamber ... oops! ... this was told to me as a true story - and I have no reason to doubt it ... anyway, I always make it a point to retell it to my students right after we’ve covered Latches and Unlatches - and how they can “retain” their status even after a PLC power cycle ... the point is that some of those “old-timey” drum switches and relays were often very cleverly designed to do things that weren’t always immediately obvious ... writing a PLC program to incorporate those same features CAN of course be done, but the “trick” is to recognize what those features ARE in the first place ... which goes back to the quote from Clay B. ... anyway - good luck with your project ... please keep us posted ...

[robh 12]

I will try to post pics as I go. Have a meeting today to finalize the plans and get any of those last minute wants. This will be the second machine I have done here so it shouldn't be to bad. I have a real good understanding of the process and how the machine works. The I/O count isn't very big. 30 or so ac inputs, 20 outputs, 4 thermocouple, and 1 4-20mA in. That PanelView 300 and a MicroLogix 1000 replaced an old Cutler hammer keypad and D100 that was used to control some water valves on timers. It was the company's first crack at using a PLC to control something...a long time ago.

[PdL 71]

It's not always easy to catch all possible idiot proof scenarios when programming, but its unbelievable the above scenario was not simulated during testing... I always like to flick the power switch to test a programs power cycle "habits". After all that is what testing is also for right.

[Guest]

from PdL: I completely agree ... but the older I get, the more often I find myself “believing” the “unbelievable” ... the scariest part of what I see in this business that we’re in, is that many people think that just because they’ve successfully written a program to make the machinery “work” - that the job is now “complete” - and so it’s time to move on to the next project on the schedule ... the sad fact is that many “integration houses” (and other companies) are trying to do-the-best-they-can-with-what-they-can-get in the way of programmers these days ... in many (most?) cases there is no “wise old veteran” programmer available down the hall to oversee the job - and to point out things like the “what-if?” scenarios that SHOULD BE keeping Junior-the-new-kid programmer awake at night ... I’m not sure exactly whose fault that is - but it seems that I’m seeing more and more of it lately ...

[robh 12]

I stay awake more than I probably should some nights for that reason. There are so many what ifs that I think about and have very few people who understand why I would worry about it. But when I have a 36" x 48" x 12" stainless Hoffman enclosure ($$$$.$$), I for some odd reason worry about the holes that need to be put in it. Or when I have designed a system (with input from everyone), calculated 150% of the required I/O, sized the enclosures and conduit accordingly, approved it with the powers that be, and then sent out all the orders. After all my parts come in, I ask for one last hoo haa to finalize any unknowns that I may have missed, we the company decide to change a bunch of stuff and start to deplenish the remainder of my spare I/O and then wonder why I say the project has become a bit more complicated? I don't have a guy down the hall to run to. Thanks To Every One Here At MR. PLC! Helps me through the tough times. Sorry for the rant, rough week.

[paulengr 44]

My plant is riddled with this sort of stuff. It's pipe. 20 feet long. Nominal dimensions of 6-24". Sounds an awful lot like a lumber mill? Those drum timers are actually worse in that respect. Someone has to either hit a signal to make it reset or to manually turn it to "reset" it. I have gone back and specifically rewritten most of the code to clear out "states" on startup. The preferred method on power up is to clear out any "automatic/manual" things, and check all the sensors out. Same thing following an E-Stop condition. Wait for an operator to hit a reset or manually jog all actuators back to "fail safe" positions. Then based on sensor readings figure out whether there is "Pipe in Position A", "Pipe in Position B", etc. Now operators know that if something goofy happens (pipe falls off the production line, actuator fails, etc.), just hit the E-Stop or reset. The PLC will clear out all that "bad data" and attempt to start over. This has saved countless hours of troubleshooting and made it safer overall.

[Andy_P 7]

A customer of ours wanted to order a machine, that started up and performed some operations the moment the power was turned on. Without any operator interference at all, no start button press, nothing! I should say that the machine is nothing as large or scary as a saw-mill, but even so. My boss (not technically savvy about safety at all) was inclined to agree to this. I had to say I was not prepared to do it and would take no responsibility. It's a small company so there is a tendency to agree to whatever request a customer makes, however ridiculous, and in this case potentially dangerous. Happy to say that sense prevailed, and startup of the machine has to be operator initiated. The mind boggles. Andy.