Announcement

Collapse
No announcement yet.

Best Equipment for PLC

Collapse
X
 
  • Filter
  • Time
  • Show
Clear All
new posts

    #16
    190507-2222 EDT

    For what I believe is a good history of the early PLCs see
    https://www.automation.com/automatio...controller-plc

    What has evolved into the AB controller was invented here in Ann Arbor, more correctly probably Dexter or Saline, at a company called Information Instruments Inc (3-I). The name John Dute of 3-I rings a bell, but I don't believe I ever met him. I have some vague recollection that I may have talked with him by phone at some much later time.

    It should be noted that in the late 1960s there were no microcomputers. But, by the late 1960s there were digital integrated circuits that performed computation functions. Such as the 74181 and there were very limited memory chips. Memory chips of some good size, like 1024 bits, did not arrive until about 1970. The Intel 4004 was the first integrated digital computer. See
    https://www.intel.com/content/www/us...ntel-4004.html Prior to this were dedicated calculator chips. I believe the early Modicon used core memory to obtain non-volatile memory, one of the GM specifications.

    There were RTL (resistor transistor logic) chips starting in early 1960s. By mid 1060s the TTL design was created to reduce noise susceptibility. Reasonably available by about 1967 or 8 at not too high of a price..

    The 4004 did not exist when the first PLCs were developed. Thus, various simple integrated circuit devices were combined to do the required functions.

    The Intel 8008 came along about early mid 1970s, the 8080 followed, and then the 8086 about 1978.

    Early PLCs used a fixed time length scan cycle. You needed to know how the memory was scanned, or you could make logic mistakes. This should still be the case.

    In the past Modicon had a transitional contact, and AB did not. This was a very useful function.

    AB had a sequencer mode, as well as a combinatorial mode. For most machine control applications I think the sequencer mode is a bad choice.

    To learn PLC programming get a PLC, and its manual, and a book on logical circuits, then play.

    Also see https://en.wikipedia.org/wiki/Invent...grated_circuit

    .
    Last edited by gar; 05-08-19, 12:32 AM.

    Comment


      #17
      The inability to be able to save symbol and description names to the PLC was a major PIA with older AB.
      No documentation would be even a bigger PIA.

      Occasionally I have to go back to something I did years ago and have to retrace my steps as to why I did what. I hate to think about following up someone else.
      Tom
      TBLO

      Comment


        #18
        Simple inputs/outputs all one really needs to know is the mentioned ladder relay logic. One PLC to another may have some minor differences but all is based on that ladder logic.

        More complex logic functions, use of digital or analog input/outputs still probably is similar across the industry, but maybe a little harder to pick up for some people. But if you can read and even draw somewhat complex wiring diagrams, simple ladder logic used is easy to catch on to. You might get away with breaking logic rules in a hand drawing where the PLC will have certain rules that must be followed pretty strictly. Logic generally must read from left to right, one load per line is sort of common rule. Number of aux contacts per component is somewhat unlimited - compared to hard wired applications, software may still put a limit but is pretty high.
        I live for today, I'm just a day behind.

        Comment


          #19
          Automation Direct, you can download programming software free and write programs right now if you wanted.

          But unless you actually upload and run a program on an actual controller it is difficult to know how successful your outcome was. They have some pretty inexpensive controllers if you want to purchase one to play with. May only have 4-6 inputs 4-6 outputs - relay type, no special function digital/analog I/O's but they can be added on.
          I live for today, I'm just a day behind.

          Comment


            #20
            Originally posted by kwired View Post
            Automation Direct, you can download programming software free and write programs right now if you wanted.

            But unless you actually upload and run a program on an actual controller it is difficult to know how successful your outcome was. They have some pretty inexpensive controllers if you want to purchase one to play with. May only have 4-6 inputs 4-6 outputs - relay type, no special function digital/analog I/O's but they can be added on.
            Get a Cmore Micro HMI while you're at it and have even more fun. Ethernet connection ability for both is worth the money. You can have both programs running make changes as you need and see the results without leaving the kitchen table.
            Tom
            TBLO

            Comment


              #21
              As Jraef said, AB is definitely the largest in the US with Siemens at #2. At the ripe old age of 69, I've used many, TI, Omron, GE, Modicon, divelbiss, Westinghouse, AD, Siemens, AB ... our customers usually told us what to use on their jobs.

              There are subtle, and not so subtle, differences. Prior to Modicon and the transitional contacts, we realized them with 2 rungs of logic. I was quite confused at those 2 rungs when I first saw them. In old statement logic ...

              store not out 1
              out 2
              store out 1

              turns output 2 on for 1 scan. A pushbutton input is often involved ...

              IMO, you can debug a machine with all but Siemens with the same general knowledge; they, more than the others, integrated "computer code" language into their ladder logic, and it was somewhat "different". Some had only unidirectional counters and timers; accomplishing what the designer needed id done differently. Some had reset of timer and counter capabilities.

              Buy one of the basic units from any; free programming software is a plus. AD's low cost HMI is a plus. Play a little. Talk with your customers' techs when you are having lunch ... they will probably be glad to discuss their systems applications ... and they KNOW theirs is the best <g>.

              Having taken and taught classes, I promise that only by doing a real job will you get detailed knowledge. But a little knowledge will give you the tools needed to learn more and accomplish jobs.

              Comment


                #22
                Originally posted by kwired View Post
                Simple inputs/outputs all one really needs to know is the mentioned ladder relay logic. One PLC to another may have some minor differences but all is based on that ladder logic.

                More complex logic functions, use of digital or analog input/outputs still probably is similar across the industry, but maybe a little harder to pick up for some people. But if you can read and even draw somewhat complex wiring diagrams, simple ladder logic used is easy to catch on to. You might get away with breaking logic rules in a hand drawing where the PLC will have certain rules that must be followed pretty strictly. Logic generally must read from left to right, one load per line is sort of common rule. Number of aux contacts per component is somewhat unlimited - compared to hard wired applications, software may still put a limit but is pretty high.
                No! Really No!
                There are a few languages that are being used in PLC programming.
                The Ladder Logic (LAD) language was invented by US engineers, the most prominent was “Dick”
                Morley. This language is more ubiquitous in US and in some English-speaking countries.

                An anecdotal account was, he invented PLC when he was drunk.
                Ladder language is considered easier to learn for electricians who are familiar with RELAY LOGIC.
                They don't have to learn Fortran or C++ coding . . . maybe a little bit of binary .

                For people who have background in electronics and computer systems, the Block Function Diagram (BPD) logic is used since it requires less space in computer display. Descriptors are often different. from LAD
                Since it requires (formal )intensive training-- it is also difficult to debug. . . but for geeks this is not a problem. You really have to know what you're doing to attempt program modification.

                The other language which is globally used is Statement Language (STL) introduced by Siemens in 1957 and popular in European countries.
                It doesn't show how lines are connected. . . it only shows the statement expressed in Boolean Theorem (binary).

                Think STL as a highly detailed specs where everything is spelled out . . . but in codes to minimize space.

                US didn't catch up until 1967 when Modicon (Gould) introduced their PLC. Morley dropped out of engineering school at MIT because he did not want to learn German.
                In 1957, there were no solid state components, so, the Germans were using germanium rectifiers and vacuum tubes.

                In the early seventies when Gould (Modicon) was just barely getting their feet wet in the PLC arena, I was working as rookie engineer for a global household product mfg company.
                In the company's drive to modernize their ageing- late fifties technology. . . they sent me (being the youngest in the gang) to train with the Modicon engineers to jump onto the bandwagon-- into this new-fangled wonders of automated production.

                Reliance and AB came later like the Japanese Fanuc. . . small units for CNC.

                At the time that we were getting involved in this technology—GM was already (pretty much ) enjoying the benefits of Dick Morley's idea.
                I did some improvements to my employers' dated equipment until I decided to bail out after 4 years.
                That didn't put a smile on my boss's face.


                And to answer OP's query where to find a good place to train, depends on what sector in the electrical industry to work for -- without having to pay for your schooling (which is quite expensive). The best place to work is in industrial environment. Commercial jobs would also help like Amazon's automated fulfilment centers. But, if you can afford it-- then go for it.


                At the time when I trained for it—it cost the company over $8000 for almost a month's training.
                Lots of automation also in modern machine shops running CNC machines.

                Money is good if you know where the money is.

                Also, the projected benefits that you expect to reap-- cannot be found in residential jobs, unless you are building fully automated homes like Bill Gates'.

                Comment


                  #23
                  190508-2125 EDT

                  By 1951 and before there were point contact transistors, and copper oxide and selenium rectifiers. In the summer of 1952 I made a point contact transistor in my Brooklyn NY YMCA room. Used a 1N34 diode as the basis. Put two point contacts on the germanium and fused them with a piece of paper as separation. Sometime I might see if it still works.

                  By 1954 one could buy a junction transistor for about $ 4.50 in 1954 dollars. General wages might be in the range of $ 1 per hour.

                  By about 1955 you could buy a Delco power transistor. I used some in a power supply in 1956.

                  I believe Chrysler had an all transistor radio by 1957, and about this time introduced the alternator with silicon rectifiers. Ford was using radios with vacuum tubes at 12 V and transistors for the power amplifier. Thus, eliminating the vibrator.

                  By about 1960 the first working integrated circuit was built. See https://en.wikipedia.org/wiki/Invent...grated_circuit

                  By 1964 there was an integrated circuit operational amplifier. See https://www.computerhistory.org/sili...is-introduced/

                  My recollection is that the 702 was about $35 in 1965. I believe I have an opamp with a date code around 1965.

                  By the late 1960s an individual could afford to buy RTL and TTL circuits.

                  By 1971 one could do a lot of things economically with integrated circuits.

                  .

                  Comment


                    #24
                    I've heard the anecdote from Morely that he conceived of it all when drunk (actually it was during a hangover FROM being drunk the night before). But his timing doesn't exactly jive with known facts. GM had a project to modernize their production line controls away from relay logic to something that could be re-used when there was a model line change. In those days when you were building the 1964 to 1966 model cars, you gutted the 1961 to 1963 production line to create the 1967 to 1970 production line. When using relay panels, it involved a TON of work by both electricians and engineers, with a lot of troubleshooting etc., resulting in delays in execution. Westinghouse was selling a system called "Numalogic" that were solid state logic "modules", but you changed programming by moving wires on the front. So they were still "relay logic" just using solid state relays. A-B and Sq. D all had similar products as well, none of them were actually all that popular though.

                    So in 1968 GM created a "contest" for their Hydromatic factory (automatic transmissions) to come up with a better way, based on a paper and speech one of their engineers had given at a Westinghouse convention describing these problems at GM factories. The participants were GE, who hired Bedford Associates (Morely's company) as a consultant, Westinghouse and Allen Bradley, who's team was headed up by Otto Strugger. The first to present them with a working model would win the contract for $1 million worth of the devices. PART of that program was for this system (no name for it yet) to be easily programmed by ELECTRICIANS. So from that very beginning conceptualization by GM the programming was REQUIRED to look like a relay ladder diagram. All participants had their own similar versions, but it was not referred to as Ladder Programming until some time later.

                    Morely beat Strugger by only about 2 months with his Model 084 "Sequential Machine Controller, or "SMC" and won the first contract, the 084 name was because up to that point he was "0 for 84 attempts" to make it work. A-B then released their "Programmable Logic Controlle", aka the "PLC" 2 months later and got most of the contracts at GM and Ford and Chrysler after that because they already had very deeply established relationships at all of those companies. A-B vigorously defended the use of the trademarked term "PLC" for years until the trademark expired. So for most of that time Modicon, who became a separate company (when GE decided they didn't want to market this SMC) could not refer to the 084 or 184 as a "PLC", even though technically they released theirs first.
                    __________________________________________________ ____________________________
                    Many people are shocked when they discover I am not a good electrician...

                    I'm in California, ergo I am still stuck on the 2014 NEC... We'll get around to the 2017 code in around 2021.

                    Comment


                      #25
                      Originally posted by Fulthrotl View Post
                      AB. it's still the de facto standard in industrial control PLC's[...]
                      in the United States/North America. In Europe it's Siemens. There are numerous other players out there too but by far AB and Siemens are the big dogs.

                      Regardless, all PLCs use one or more of the IEC 61131-3 programming languages (Ladder Logic, Mnemonic Instruction, Sequential Function Charts, Structured Text, and Function Block Diagram). Unless you know you will be working with Brand A or Brand B equipment I would say it doesn't matter what equipment you learn on as long you learn those languages (well almost no one uses Mnemonic Instruction).

                      Most installations I've seen have a preferred brand and/or series of PLC but there are usually a couple of PLCs hiding in the corner that don't match the others because they are legacy or part of a turnkey system so you better be prepared to learn other systems as well.

                      Comment


                        #26
                        Step 7

                        As stated above the two big ones you'll see in industry are AB and Siemens. It is just a personal choice really

                        My personal choice would be to obtain TIA Portal V14 or above for Siemens S7. I find it easier to navigate and use then AB software. You can get a copy of it on Ebay for about 400 bucks I think. The TIA V14 that I have you can run a simulator for the S7-1500 and the 1200. So if you just want to play around you can set up the simulator for an idea that you have and write the code, and work out any bugs without ever having to be connected to the plc and you can see the program response. If you can get your hands on the Professional instead of the Basic then I would get the Pro. It allows you to program beyond just the S7-1200. But for learning purposes you can grab a S7-1200, TIA Portal 14 and you can learn to do quite a bit of stuff with that and the internet.

                        Comment

                        Working...
                        X