Sign in to follow this  
Followers 0
gravitar

"All PLCs are the same"... What do you think?

32 posts in this topic

I've just started working for a company that employs a WIDE variety of PLCs. Some machines have AB, some Siemens, some Mitsubishi, some PLC Direct, some GE, even some brands I've never heard of, like Control Technologies Inc. Likewise, they have HMIs of just about every conceivable flavor. Some of the larger lines mix-n-match several brands.. One line I was working on has a Control Technologies PLC (w/ TCP Quickpanels) controling one thing, a SLC 500/Panelview controlling something else, a PLC Direct (C-more HMI) controlling yet another part of the machine, and a Mitsubishi controller with a Maple Systems HMI controlling yet another part of it (!!) Naturally I inquired into the reasoning behind this. One maintenance guy that had been with the company a long time, explained that in the beginning it was just one PLC, but every time they added functionality, they added another controller and it would always end up being vendors-choice or whatever seemed to be cheapest at the time. I then asked my boss, my boss's boss, and a fellow controls guy about this crazy situation. The answer I got from all of them was along the lines of "Logic is Logic", "They all do essentially the same thing", "You program one you've programmed them all", "Yeah, its a challenge, but I like challenges", etc. Now, to me this whole situation seems asinine. They think it's a badge of honor to try to cope with a PLC version of the "United Nations", but I'm not convinced. Am I just being lazy/ignorant or have I walked onto the set of the Twilight Zone?

Share this post


Link to post
Share on other sites
I'll be you a beer that not one of their programmings software packages is up to date (unless its something aquired withing the last year)

Share this post


Link to post
Share on other sites
If you see Rod Serling or Ray Bradbury coming to meet with Plant Manager I'd get really worried. The only way to combat this and get them to settle on a single or couple of brands is to quantify the cost of learning and troubleshooting each different system.

Share this post


Link to post
Share on other sites
maybe this is a blessing in disguise ... a great opportunity to learn new skills - in preparation for when it’s time to move on to another job ... personally, I’d start by making a complete inventory - with precise catalog numbers and firmware versions, etc., etc. of everything ... and of course a good backup copy of every program that’s currently being used ... ALSO ... don’t forget that many (most?) systems have “hidden” stuff such as DIP switch and jumper settings that don’t show up in the disk copy of the program ... write that all down too ... then I’d start going through the “easy stuff” first - looking at how “Brand A” performs the same BASIC control tasks as “Brand B”, etc. ... what are the similarities? ... what are the differences? ... I’d make notes of these - preferably on paper printouts of the various programs ... once the “surface layer” had been scratched, then I’d dig deeper and go into the more “advanced functions” - for “Brand A” and “Brand B”, etc. ... actually I’d make a checklist (based on the initial inventory) and check off the various features and functions of each platform as I went along ... and if you’re interested in “fixing” this company’s “issues” then I’d record the dates and times that I spent on each one of the items in my checklist ... my plan of attack would be to cover ALL of the platforms in a LIMITED amount of detail first - and then keep going back across the board for MORE detail ... that way I wouldn’t get hung up on just one system - and neglect all of the others ... whether the mix-and-match fiasco ever gets “fixed” or not, personally I’d try to make the best of the situation and get the most of this as a genuine “learning experience” ... just think how impressive the product and model list will look on your next resume ... Edited by Ron Beaufort

Share this post


Link to post
Share on other sites
Ron's list is a good starting point for dealing with this from a practical perspective. No company is going to change all of these PLC's just out of a sense of neatness. One thing I would add to Ron's list, however, is to identify which controllers are actually obsolete and no longer available as spares. These are the ones I would look at changing first (and I suspect you will find quite a few). Andybr

Share this post


Link to post
Share on other sites
Gravitar, I feel your pain also. At my plant we have (mostly) AB PLC's. Pretty much all the SLC's from 100's to 503's along with the micrologix lines. I want to start getting into Contrologix or compact logix but that's a whole separate issue. We also have 4 different types of Mitsubishi controllers, 3 idec controllers, 2 PLC Direct and a couple of Omron PLC's. These (other) PLC's that I just mentioned came installed in equipment that the company purchased and the PLC's are the types that the vendors used. One a side note, we also have 6 robots. (mostly Fanuc and Panasonics) there are about 3 different programming types for those.....Anyway..... At one point, I had a Idec PLC start to go crazy on a powder booth paint system. Well, I had nothing to get into the plc. So I went online to idec's website and downloaded a trial version of their plc software and got a comms cable from another vendor just to pull the program out and save it to look at. Unfortunately the software only lets you upload the program from the plc you can't download anything to one unless you buy the complete version, which my company did not want to do. So after looking at the program, which I could only understand about 1/2 of. I decided to put a micrologix in its place. I figured I could do what I wanted with this setup. All my ranting basically comes down to this. The other plc's we have (Other than Allen Bradley) we don't really have any software or hardware to do anything to, so if something were to happen to them we try to go back to the vendor for support and if that doesn't work its time to swap it out for an AB.

Share this post


Link to post
Share on other sites
at the very root of this situation, there is something mildly philosophical going on ... (1) when we (meaning the folks who visit this forum) look at a piece of industrial equipment, we always see TWO things ... first, we see the machinery (sensors, switches, pumps, valves, etc.) ... second, we open the control cabinet and see the PLC that controls the equipment ... (2) when other people (meaning bosses, and operators, etc.) look at the same piece of industrial equipment, they see only ONE thing ... specifically, they see the equipment as one complete system - and the PLC never gets considered at all ... so ... when we “PLC people” consider a piece of equipment, we’re naturally very interested in the brand of PLC, the model of PLC, the software required to program or monitor the PLC, the type of cable required to go online with the PLC, and other stuff along those same lines ... and we tend to worry whenever all of those parts aren’t properly accounted for ... on the other hand ... when other people consider a piece of equipment, they’re primarily just interested in whether the equipment “works” or not ... and if the equipment happens to be “working” at the present, then all of that PLC stuff is of no interest to them whatsoever ... so from their point of view: “turn it on ... if it works, let’s make some widgets” ... in the context of this thread, I suspect that just considering the other person’s point of view might help us understand why the “powers that be” aren’t overly concerned with the situation being discussed ... at the same time, if we’re the guys in charge of the PLC end of things, it behooves us to look at the situation from an insider’s point of view ... it goes without saying who’s going to get the blame if “our” end brings things to a screeching halt ...

Share this post


Link to post
Share on other sites
Spot on Bob but dont you have some sort of holiday today Happy Holiday Rodney

Share this post


Link to post
Share on other sites
Programming is programming. Coding is much different because it depends on the tools and the are much different. A PLC is a tool the programming environment is a tool. Even the language is a tool. The question is where do you want to put your mental energy. Into getting the job done or learning tools. Personally, I get by with C and a little C++.

Share this post


Link to post
Share on other sites
I'd tend to agree with Ron on this one - What an excellent opportunity for you to learn a wide variety of systems. As far as management purchasing a hodge podge of brands because they're functionally "the same" - what a terrible idea - an exercise in poor decision making and an interoperability/support nightmare! The marginal savings will be significantly overshadowed by systems that aren't maintainable. Projects may need to be recreated, software/support purchased, etc. I don't agree with the opposite end of the spectrum (dictating a single brand as your corporate standard regardless of the problem), but this approach has to be the absolute worst. It's all about creating a workable "happy medium" that's consistent and consolidated, yet flexible enough to meet your needs.

Share this post


Link to post
Share on other sites
I agree with most people here... This is a good chance for you to examine the strengths & weaknesses of each PLC & HMI, but it's definitely a maintenance nightmare in the making if you don't have the right software & hardware to support each different machine. I'm a stickler for backing up programs, so I'd be doing what Ron B said... Make a list, check it twice and then go out for quotes on what it would take to replace each part (if it isn't obsolete already) along with lead times so you can decide what must be kept in a spare parts inventory. (Gasp! Shock! Horror! for management when you give them the list. This usually makes them think about cutting down the # of different controllers in the future.) I'm in a plant that has gone from 4 PLC makers to just AB over the past 5 years - mostly in the past 2 yrs thru obsolescence & sales of old machines. So now I have a bunch of spare PLC parts. Anyone know where I can sell them besides eBay? Edited by ssommers

Share this post


Link to post
Share on other sites
When I started at the plant I work at now, we had very little variability in terms of the PLC's, but they are/were all PLC-5's. Very expensive to maintain, and communication capability constantly rears its head whenever you leave the simple world of 4-20mA or discrete IO. Programs were stored willy-nilly all over the place, and there were multiple variants with various amounts of documentation, all more or less incomplete. The HMI side of things was a total mess. I had to go somewhat medieval (thou shalt have ONE VERSION) as far as the documentation and storage of everything went. From the HMI side of things, I first tied together all the projects that were actually running on the same system so that from the user point of view, things were unified. Then I converted a few others until we were down to just one standalone operator HMI, one SCADA/HMI, and some deviant crap developed by IT that I'll never get control over (therefor, not support). Then we upgraded the licenses, got the current bug patches, etc., and upgraded the hardware for the SCADA. We settled on a single new platform for plant floor PC's (and then changed it three years later of course!) I've also slowly done screen cleanups. The guys before me were mostly interested in one of two versions of the HMI development. Version #1 looks like a 1970's control panel. Every screen is a grid of rectangles with a small misspelled cryptic text description underneath a number and a function key. Version #2 looks like the first time one of these guys discovered how to change colors and browsed through the HMI libraries so of course everything is very misfunctional but 3-D looking. There's no consistently on button placement so you sort of search for everything. Oh, and on the PC-based ones, they popped open subwindows for everything so you had to constantly reboot machines when operators managed to open a few hundred windows after a few weeks! Needless to say, most of this crap has been junked. I still haven't really updated every screen yet for consistency, and it is frequently very hard to figure out where all the screens are at (operators have the same trouble!) but over time it's getting better. A couple weeks ago was the first time in about a year that we had crashes where I stumbled into another nest of creating dozens of windows on a machine. Even when staying with a single PLC, we had brand mania in terms of parts. Did you know that you can get catalogs of sensors with hundreds of different models from multiple vendors? Does it make sense that on every project, you actually try to exercise a good fraction of those choices? By the way, all drives are the same. They all come in different sizes, but reliability, ease of use, product support, troubleshooting, all the same. Interfacing to a PLC is similarly simple and easy. So buy whatever is cheapest. We've also settled on just one or two vendors and models here but it's taking a while. Same thing with motors.

Share this post


Link to post
Share on other sites
Best place to sell them.(see link below) I have and have nothing but good thing to say about how they handle it. http://www.plccenter.com/default.aspx

Share this post


Link to post
Share on other sites
Sounds like an opportunity to learn a lot...or to pull your hair out and get a one way ticket to the loony bin.

Share this post


Link to post
Share on other sites
/rant on I never have understood the problem with learning a new PLC system. To me they are all just like different remote controls for TV's. Physically they all appear different, but all have the same control buttons and do the same basic functions. Everyone seems to get hung up on, "It's not _____." (Fill in the blank with that persons favorite vendor.) Ladder logic is ladder logic. Function blocks are function blocks. The only thing that is different is the programming GUI. Once the GUI is learned and the nuances of minor differences learned, what is really the difference between brand A and brand B? IMHO, not a lot. Some GUI's make programming a bit easier, some make debugging a bit easier, but I have yet to find any one vendor who does a great job at all of the things a programming package needs. Most people know AB. Why? Because they have the largest market presence in the States. In Europe it's Siemens. In Japan, Mistubishi. Each of those brands is the "best" according to people in those markets. Why? Because that brand is the leader. /rant off Back on topic I can't say I would be happy trying to keep all of that up to date, but it isn't the end of the world. The first major plant I worked in didn't have prints or any set standards. What it did have was the best learning environment in the world for instrumentation. I learned more in two years than I would have in ten if they had been standardized.

Share this post


Link to post
Share on other sites
I agree that it is a terrific learning opportunity. Working for a distributor with only one brand of controls means that I learn those controls in much depth. However, don't ask me for a comparison to other PLCs, because I have very little knowledge of them! While I enjoy my work, I wish I could learn other PLC systems. That being said, I think the main issue you will run into is support ($$$) from management to keep up with all the different systems. Forget training. You'll be lucky to get copies of all the programming software, until each of the machines break at least once! I really like the approaches listed above for cataloging hardware. Then you at least know where you stand.

Share this post


Link to post
Share on other sites
Greetings rlp122 ... first of all, let me say that I’m not trying to argue with you - but I have a difference of opinion on some of the comments that you made ... here is one major point that I personally disagree with ... most of the students in my PLC Boot Camp classes (99% or more) are NOT “programmers” who live with the PLC day in and day out ... instead, they are invariably “maintenance technicians” who are forced to deal with the PLC only “occasionally” as part of their duties in keeping their plants’ machinery running ... tracking down the logic which controls a malfunctioning field device is one very common example of the types of tasks they're trying to master ... trust me ... to THESE guys, even MINOR differences between how various platforms handle their ladder logic programs can present MAJOR stumbling blocks when the machinery is down and the boss wants it fixed again - quickly ... as I said in the beginning, I certainly don’t want to start an argument with you - and I fully respect your right to hold your own opinion ... my main reason for writing is that MANY “beginners” read these forums looking for information as they move into the field of PLCs ... I thought it important for them to know that there are some of us in the trade who hold opinions which are quite different from the ones which you've put forth ... specifically, ALL “ladder logic” is NOT always the same ... Edited by Ron Beaufort

Share this post


Link to post
Share on other sites
Thank you for all the insight you've all provided. I agree, it'll be a good learning experience if nothing else. Hopefully I'll at least be able to figure out what hole to plug the programming cable into on all these different PLCs! (Who would've figured that you program a GE PLC through an unmarked hidden compartment on the power supply.. I felt pretty foolish a couple weeks ago when I figured this one out!) Anyway, in my opinion, it's true, "Ladder Logic is Ladder Logic", regardless of brand. But I suppose you could say that about the languages we speak and write tooc, couldn't you? English.. Spanish.. French.. German.. They all use nearly all the same symbols. They all branched from the same origin. They're all used to accomplish the exact same tasks. So it should be a trivial matter to master them all, shouldn't it? But wait.. what about all the regional/national customs that you would have to master too? Well I'm running into that too. Like I said, I wouldn't expect to "program" the power supply. I didn't realize that when you're online, you have to sit and wait for the ladder to "update" before you REALLY know what's turned on and what's not, and sometimes you have to wait a LONG time (even though it's on ethernet.. go figure). I wouldn't have guessed that, even though the HMI and PLC are programmed withy the same software package, you can't do both at the same time. I think Ron hit it on the head when he said that the decision-makers probably don't even think about the PLC. They see a machine, a machine makes parts, and that's that. I'll try to get a handle on what's installed where and what software is needed for it, as a first step. Whether or not I'll have any luck making anyone consider that there's another way, is anyone's guess!

Share this post


Link to post
Share on other sites
This used to be true. It is no longer true. There is a very good reason why most major PLC vendors are pushing their newer lines as "PAC" (programmable automation controller) rather than PLC systems. Older PLC's are programmed in an interpreted language roughly 1 step away from assembly language. The newer PAC systems are mostly derived from "C", Verilog, or some other similar higher level language. It makes a huge difference in how the system is implemented. The symbols look the same but that's where the similarity ends. If your idea of a PLC is a smart relay akin to the stuff that Automation Direct sells, among many others, then I'd agree. A smart relay is a smart relay is a smart relay, regardless of whether you program it in ladder logic on a PC or using a very small LCD screen and some cryptic op codes. But the moment that you start running a major process or multiple processes, things are very different. It gets worse these days since it is becoming more and more advantageous (not just for cost reasons) to ditch the central chassis concept and go partly or mostly for distributed I/O. In that environment, again, ladder logic is not any other ladder logic. The degree of gyrations necessary to get an AB PLC-5 to work with DeviceNet will make your head spin. In contrast, it is trivially easy with ControlLogix. Same thing comparing virtually any of GE's older PLC's to their PAC system. The biggest hassle these days with PLC's is spending time trying to handle all the translation and interfacing issues. There still isn't really a de facto field bus which is universally accepted everywhere (Modbus comes darned close). Interfacing to PLC's and anything that has a serial port or an Ethernet port is always a long, frustrating exercise at some point. It is getting better, but only by fits and starts. Furthermore, you've also got the field bus world. With some field buses, the I/O level has gotten to the point where it has enough capability that you can actually literally load logic directly into the I/O and you can at least in some cases dispense with the PLC altogether. There is no longer a central processor, so the last major single point of failure (the PLC) disappears altogether. The major detractors are that the system is complicated, expensive, and slow. So for now, it remains the bastion of certain specialized systems.

Share this post


Link to post
Share on other sites
Good! I was hoping someone would step in and point that out. Today's PLC's are more like computers than the older systems. IMHO ladder logic isn't always up to the task at hand. I have disassembled my share of AB programs and trying to dig through all of the possible ways that the programmer puts things together is very difficult. Usually it's more than difficult due to the lack of documentation of the program. Ladder Logic is not just ladder logic any more. It still shares basic elements, but even these elements do not work in similar manners. This is why I stated "...nuances of minor differences learned..." It's the nuances that make the differences, There is little that can't be done with any PLC on the market. It's just a matter of how much pain there is to do it. I still think that this might be one of the best learning experiences that the OP could have. It will be an experience with plenty of headaches, but it if you can get through it it will be worth it's weight in gold for your next job.

Share this post


Link to post
Share on other sites
You know.. the potential is there for me to learn a lot.. but I must confess, I'm on my fourth week in this place and I'm still not really sure what I supposed to be doing. So far I've just been watching from the sidelines, doing clerical/office work, calling vendors, writing memos, etc.. they keep on telling me about this job and that job that needs to be done.. Great, I think.. I'll actually get to wire something up or make a change in a program, and be able to show that I'm actually good for something. My boss keeps on telling me that "my time is much too valuable" to do an electrician's job, and they'll just hire a contractor to do that. AARRGGH!! I hate to come right out and tell the guy "BUT I'M NOT DOING ANYTHING!" but the fact of the matter is, I'm not making very good use of my time or the money they're paying me to be there. Probably the only things I've done that are worthy of a controls guy are to argue with maintenance about the safety features they aren't implementing, and argue with another controls guy about how a simple, passive sensor manifold is NOT a remote I/O block :) The project manager asked if we could do something and he had the nerve/stupidity to tell him we couldn't because "those fixtures were wired up with a special Remote I/O to DeviceNet adapter.." Give me a break dude, if you're going to lie to the guy, at least make it believable! But anyway, I'm quickly gaining the reputation as the new guy that likes to make trouble for everyone,so I'm wondering if I should just start keeping my mouth shut and collecting my paycheck..

Share this post


Link to post
Share on other sites
I'm going through the same situation; in the packaging department we have production lines with AB, Mitsubishi, and Omron PLC's. my platform is mainly AB, but last week one of Omron PLC's lost the program and the whole line stopped and the worse case is that we do not have the software; my boss ask me to buy the software and upload the program fron another controller, I ordered it but it will take about a week, fortunately for me I'm using this time to get to know about Omron conrollers.

Share this post


Link to post
Share on other sites
I think your problem is that your co-worker HAS been believable at some point. Evidently, he hasn't had someone around to call his bluffs. If you truly know how to make it work the way the project manager wants it to work, speak up and expect to do it - which is what I think you want anyway. I only bluff in special cases... like when the company president comes around asking what that blinking yellow light is for and I only have a faint idea. I found out 2 hrs later that it really meant was that the pH in a 5 gallon bucket was out of whack because several years ago they stopped using an old waste water treatment system but couldn't figure out how to remove the pH sensor without removing the whole controller and that stupid blinking light still actually had a real meaning for another part of the water treatment system that was still working. Yah... My bluff - blamed it on the outside waste water contractor and then fixed it just before he came in. (I really didn't know about the bucket! And will be reprogramming the controller to stop looking at the pH sensor later this week.) And then collect my paycheck!

Share this post


Link to post
Share on other sites
this may work short term but eventually you need to be comfortable at work. i know that i can't shut up for too long. i would try thinking about ways to get involved and show some results. maybe "i've seen those, let me try something quick..." etc. it's a shame.... for example we are looking for someone new to join our team but it's been tedious work of screening candidates who only look good on their own resume but don't show any interest, experience or know even basics. good luck

Share this post


Link to post
Share on other sites
I can attest to this. The best part is as you cross from brand to brand your skills will improve. You will find that you can understand the various "nuanices" of each make gets easier to understand. I have worked with everything you have listed plus some others. I would not say I am an expert on any perticular brand but I can muddle thru most code on each system. The really cool thing now is you can get manuals online for most systems and this helps alot. This is the first thing I always do when facing a new system. Find a manual. The internet is your best tool here. Once I have the manual I then find software and cable. I try to have a backup copy of the program as quick as I can. Now I get machine specific. I locate or create my own drawings of the electrical controls and find out how the machine is supposed to work. Now I see how the code matches up to my drawings and ideas of operation. By knowing what the machine is supposed to do and knowing how it is wired goes a really long way in understanding how the code is working. Side Note: From this type of work my programing has improved because now I have seen more than one way to get the job done. Irony: Now I work for an OEM. Got the job because I can work in multiple platforms and brands. We let the customer specify the PLC.

Share this post


Link to post
Share on other sites

Create an account or sign in to comment

You need to be a member in order to leave a comment

Create an account

Sign up for a new account in our community. It's easy!


Register a new account

Sign in

Already have an account? Sign in here.


Sign In Now
Sign in to follow this  
Followers 0