Jump to content


MVP 2017
  • Content Count

  • Joined

  • Last visited

  • Days Won


Everything posted by Ausman

  1. Ausman


    So true Joe. That's why Neil's name was "Armstrong", he was flapping them up and down for ages holding the lander up above those rocks. 😉 I'm always amazed at how they managed to design and build everything in the first place, without any of our "modern technology". Did you see that TV doco/series about the builds and testing? Holy moly! cheers, Aus
  2. Joe is correct. Both must be in the Main routine as shown. But I'll also add that if you have controlled the outputs later in your ladder, then these controls over-ride previous decisions made by the logic. Given that you say you are new to ladder logic, this is a critical thing to understand. All your outputs should actually only have one ladder rung referencing each of them, and ideally this rung is active all the time. ie there is only one appearance of a physical output in the entire program. If you need to have multiple references to the same output, you do this in a variety of ways, the simplest being a number of MBs configured in that one rung so that when any of them are on, the output is on. Be careful calling outputs with subroutines, you can have situations where your output won't change state as you might think when you turn a subroutine on or off. In some ways you need to treat PLCs like the dumbest of dumb. You have to tell them everything, and in the correct sequence, to get them to work as you want. A bit like descriptive text. "Make toast and spread jam on it." A tiny bit of this translates into: 1 Open bread box 2 Locate bread in bag 3 Find the end with the closure 4 Remove closure blah blah blah etc etc And don't forget that all 4 of these actually involve far more bits of logic and actions to achieve.....I have simplified it a little to illustrate the point. (And try to make you laugh!) The entirely simple thing to make toast and spread jam would be an instruction set with many hundreds of individual instructions/references, all of them critical to achieving what you want. cheers, Aus
  3. Ausman


    And to think it was all done using a V130! 😀 I think the really cute trick was the way it got there in the first place. cheers, Aus.
  4. How do you want to interface the keypad? Direct inputs? Modbus etc? cheers, Aus
  5. Just to clarify, ALL LEDS, not just those on the plc gear, ALL indicators, flicker? Are the pilots driven from the PLC call, or from the controlled contacts closing? Big difference. I'm also with Joe on his suggestion, had not considered this. cheers, Aus
  6. +1 to all that Joe says. I might be tired this morning, and about to say something stoopid, but aren't you calling the same network ID? ie need to change your slave to 2? I know you said ports termination ok, but have you got termination set on the drive as well as the plc? Need resistors at both ends. For your test purposes, you could also just be using MB1 without any of the MB2 involvement. Once the function is called, it runs through. If your call might be prevented from starting on the rising edge due to another modbus process going on, your logic is good. However, I would also keep in mind that I always leave at least a few scans b/n modbus finishing and the next call to ensure that the buffers have completely cleared. So this means that the logic is in some ways not so good! Paste from the help guides re ideally having a delay b/n modbus calls: <<< Slaves: Consecutive References Whether the MODBUS master is a Unitronics PLC or another device, if the master application size and system requirements allow, it is recommended to add a delay between consecutive references to slaves according to the table below. Delay (msec) between consecutive references to slaves; Minimum Baud rate = 9600 Minimum Recommended Jazz 20 40 M90/M91 15 30 V120, V2xx 10 20 V130 5 10 V350, V570, V1040, V1240 0 5 >>>>> cheers, Aus
  7. It definitely sounds like an incorrect power configuration issue. Amongst other possibilities, the flicker could be your power supply constantly approaching it's shutdown limit, but not quite getting there. I'd check a few things. 1). Are the 2 systems being powered from different phases of a 3 phase supply? I ALWAYS ensure that all controls are on the same phase. And associated with this, ensure that the neutral to your gear's location is ok as well, and the actual volts at your end are correct. 2). Have you got the modbus cable earthed at your end? If so, if you connect, but with the earth disconnected, does the problem go away? 3). If the problem persists after checking 1 & 2, measure the current draw on your power supply with a good quick acting DC clamp directly at the output, with your system running with and without the modbus cable connected. If you can't get a good enough DC clamp, put an analogue ammeter in the line. If you find that there is a significant difference, then there is definitely a power interaction issue. Tracking the difference down the various branches will help pinpoint things. cheers, Aus
  8. Yes, I do this, but the makers keep on changing their coding and systems to stop the access that lets you do this. It's a bit of a one step ahead type thing. You have to have the correct hack for the particular build/unit, otherwise it is bricked. Ahhh....but essentially you are telling the world your system's security. cheers, Aus
  9. Sorry All....Real work has prevailed over volunteer stuff. It is coming and you'll like it. cheers, Aus
  10. You keep on mentioning that you are using a VirtualPLC. Where did you get this? More information please. cheers, Aus
  11. But they constantly change their code to try to foil this, such that even the latest hack might not work on a new unit. I have a few lying around used as paperweights! But I wouldn't even want them sending anything out at all. If you can hack so that they work directly within your network simply via the router, that is ok, and is possible. We'll see these things controlling major supply generators soon, for management's convenience! Without the Hack. 🙄 cheers, Aus
  12. But don't forget that Visilogic has the annoying habit of showing red power lines (implying active) on a subroutine that isn't being called. It is a quirk that takes a while to get your head around, and has been the subject of much discussion. It essentially shows the plc's state of anything in that sub, but it isn't necessarily happening. One solution is a self-resetting count in the first rung, to confirm that the sub is indeed being called whilst you are watching it. Another way of doing this, but note that there are currently no safety interlocks relating to physical reads from contactors if that is important....it is purely logic based. cheers, Aus
  13. Ok, there may be a bit of misunderstanding here. I am talking about doing it remotely. I'm assuming that everyone here is talking about doing it using a card. Hmmm. Just realised that you could likely do this remotely, then read the card file remotely as well. So c'mon gang, if this is the case, why didn't anyone pipe up during recent asks about "how to get all info out of the plc into a new one?' But also, and I know I'm being stupid this morning, how do you do it if you don't have a unit with a card? cheers, Aus
  14. Sorry...other things on the go. Almost there...will be worth the wait. "Elegance Extraordinaire Monsewer" Well...maybe. Certainly will get your brain cells firing. Cheers, Aus
  15. So you have GPM at anything up to 1000. Your problem is just Maths. If you want the average flow per second you have to first set up the logic to do this, based on how many reads you do per second from your sensor. An accurate method is to constantly add all the raw data values read over the second, then divide them by how many times the reading has been done, derived from another count monitoring the actual read. Once you get that result, you run that through a similar Linearize and then divide that result by 60. This logic would be set up on a self-resetting loop basis, starting again when the time period is reached. For slightly better accuracy you could adjust the Linearise to convert things based on a per second read...ie 1000/60 = 16.666rpt. For example you could use 16666 as your @20mA amount and divide the result by 1000. Think about this a bit, and hopefully you will understand why this will be a slightly more accurate read. cheers, Aus
  16. Hi Denis, although a bit off topic, I have always been under the impression that Vision's clone works from the file on the PC. Are you saying that it incorporates all "adjustable" elements ("all settings") by consulting with the PLC during the process? My clones don't do this, but if it is possible I'd like to know about it given that the Vision forum often answers questions about how to easily duplicate everything. If there is a simpler way than what is currently advised this would be good to know. cheers, Aus
  17. I haven't done a save from new for quite some time, so that might explain it. But I would love @Saragani or someone else on the team to explain things. I have not put Version Swapper into place yet on this new build, either. Just 9.8.65 at present. cheers, Aus
  18. Hi all, started something new yesterday, worked on it for a little while, then had to close and save. Came back to it 12 hours later and up came the update pop-up which surprised me a bit. Have a look and you'll see why. I am curious about why it came up, given it was on the same system that created it. The only difference appears to be the Version Code. Many times I get the pop-up when it is an older program, but I've never had this happen before. I have recently shifted onto 7-64 Pro on this puter, instead of 7-32. A totally new build, so no possible older references at all, so that makes it even more odd. cheers, Aus
  19. That's OK. I often pop over to the Rolls Royce forum and ask a question about my Trabant. cheers, Aus
  20. I know, I know, taking longer than I wanted. Other things cropping up. I am doing a subroutine (export/import) to show examples. Hopefully able to do this homework tonight. cheers, Aus
  21. Hi all, it has been the subject of many topics, but a consistent issue is mounting problems with snap-ins. Can the creators please truly-ruly consider implementing the following: 1). Possibly incorporate screw fittings, depending on where you might go on points 2 & 3. I recall someone finding that there was now extra plastic, or pcb clearance allowed, in later versions. But I don't recall the logical arrival of screws actually happening. 2). The main issue that I see is that the pins need to be made slightly longer. This surely isn't a hard thing to implement on the PCB line. Everyone who uses them has always noticed that the engagement seems to be on the last few mms, if that. And as well, they always seem to be a very easy insertion. Think how hard it was/is to plug in an IDE hard drive......this should in theory be at least the same. Longer pins would help the issue, or maybe a better clamping female? 3). Although the principle of the Snap-In is just that...ease of snapping it in......perhaps a better and relatively easy alternative would be a clamped ribbon? I understand that this would bring all sorts of production issues and compatibility into play, but maybe it would be possible to have both types of connection in place so that users can choose whichever suits their hardware. In the world of PCB assembly and component costs, this would add very little to the cost of the unit once the board changes needed was amortised. Every PLC arrives with basic comms connection cabling, so that OEMS now have gazillions of them lying around. Having a ribbon in a snap-in box would be much the same sort of thing and just an accepted part of the package, so that the user could fit what suits. Once the change was initiated, it would gradually become the norm. Perhaps the cable could be just like an IDE ribbon, connecting existing pin layouts without much change at all. 4). Creators need to consider that each time someone has an issue due to the current poor implementation, that is likely to be one (or more) less customer in the future. "Useless plc......jams up for no reason....bin it and go somewhere else....what a waste of time and money....tell Fred to stay clear of them". cheers, Aus
  22. +1. Do this, and then when reseated also apply strips of some sort of woven cloth adhesive tape (gaffer tape) to the seams to ensure they stay together correctly. Also, your photos show all the solenoid coil cabling running up next to what looks like a power supply. Are you sure that all solenoids have clamping? If so, where is it? Ideally it is in the coil plug itself. Anywhere else is just letting the spike into adjacent lines. With this layout I also ponder 4 things: 1). The use of the V700 looks to be overkill given the number of I/Os actually in use on the snap-in. 2). Given all the "inductive" activity going on in the enclosure, I'd be putting a correctly earthed metal screening panel around the plc as neatly as possible. This isn't hard to do well and be made for ease of access in the future. 3). Is the door panel earthed correctly, with a flexible cable properly attached? 4). Is the plc itself properly earthed? It is a bit hard to see in the photos if this has been done...it looks like it hasn't. cheers, Aus
  23. I'm working on my "methods" pros and cons essay. Might take a bit longer. 🐢 Almost done but I'll revisit it a few times before posting. Stay tuned! cheers, Aus
  24. We now have mods pretty much everywhere the sun is ON. Perhaps even where it is ON or OFF for months. As well, some mods often have irregular hours resolving some conundrum....and distract themselves from what they should be doing by checking the forum!!! A sort of literal cup of coffee. Do the glass half full bit....think of it more as the orange bar of knowledge! 😀 cheers, Aus
  25. Good. And good info Dan, I was thinking along the lines that absolutely nothing was happening. I'd like to pin this with a slightly changed name if no-one objects, as I always remember previous topics that have this sort of thing mentioned during discussion, but can never find them easily. It is the sort of thing that needs to be prominent for users in "distress". cheers, Aus
  • Create New...