Ausman

Or some variations out of this one...."have we finished yet?" : https://www.youtube.com/watch?v=mJ_x_1N6B20 1:40 is one of my favourite bits in the entire series. cheers, Aus

You are showing serial comms, yet are talking TCP/IP. cheers, Aus

In crucial operations I have a deliberate time check on data being updated correctly. Every nominated time period, I place a number well outside the normal range in a register, having the system ignore it for that small time, and then see if it gets changed. If it doesn't get changed within a given delay, the alerts come up. This is actually a very easy thing to implement as you only need to check one piece of data, not a full vector, and isn't based on any reads of system stuff that Alex has rightly suggested. It is based on actual results. You could do it to an entire vector if you felt it needed...I've never found it necessary. I should mention that all my slaves are routinely called in a varying sequence, with a given maximum time between calls, which you definitely have to relate to the slave's checking time periods. cheers, Aus

Hmmm.....more justification for Joe T's constant advice to get a simple scope for checking things? It would have likely picked up the differences that a digital meter possibly missed. All depends on the meter's specs and useful settings. I also have an older fully analogue meter that has proven invaluable and dead accurate over the 35 years of well looked after use. And it doesn't use batteries! Related aside.....has anyone else ever laughed at how progress has made things more complex? Instead of me having a car radio that gets the signal direct from a transmitter, I now have to have my phone on, using mobile internet to receive a favourite "radio station". The phone then uses bluetooth to connect to the car system that processes the signal from the phone and gets it to the speakers. 😖 That's such excellent progress, isn't it?!! In the overall scheme of things how much "total" energy is wasted by this method? The list is endless. cheers, Aus

Also, if you have a look through the pinned topic http://forum.unitronics.com/topic/6229-ongoing-13-topics-almost-worth-pinning/ you will find a very good explanation by Joe T. cheers, Aus

Visco (and Kratmel?) it would be intriguing to know the complete reason for this behaviour. It almost seems like the supply was cycling on it's overload system very rapidly. ?? Visco, can you drop some connections with a smaller supply in place and see if your sensor readings become stable with far less draw on the supply than before? cheers, Aus

I'd be changing your strategy. Use 2 (or more) tables that work under FIFO. Power up bit compares the time to a certain 24 hour period, and if appropriate makes the other table become primary, sends out the just changed over from one, and once the send completes clears the table ready for next time. Before clearing, it could also store it into the SD for safe keeping as a backup. The power up bit could act via some sets/resets that don't have any power up values, that then dictate appropriate other actions. Or they alternate using (some?) Toggle Coils. cheers, Aus

Being XP, did you do the install using Right click, Run as Admin? This is a must. If not, do a complete uninstall using something like Revo portable. Further to Kratmel's points, you can run multiple instances of the same version, but you need Version Swapper to run multiple versions. As an aside, I have never tried, but I would think that with Swapper in use, multiple instances of the loaded version could be done, if appropriate copies were in place. I know Flex has a large range of versions under Swapper, so maybe he could try this on one version just to see? cheers, Aus

Hi Joe, I'm assuming these stops/questions relate to things NOT covered by all the various options under Project/Properties? cheers, Aus

Further to Joe's advice 2 back, another method I found today that directly links to the offending program is the simple variation of hold the shift key and right click the invisible program's taskbar icon. Left click Move and the arrow keys will work, along with the mouse once it appears. cheers, Aus

I'm assuming the +24 input you mention is linked somehow to the output that controls your "plungers". What I was getting at is that you should always try to monitor the very final result of the output turning on. If the input comes from a gizmo that measures an actual flow from your plunger that's great, but if it simply turns on by perhaps using an auxiliary contact on a contactor then that isn't a guarantee that the action is happening. All that is monitoring is that one switch in the process has turned on, which doesn't necessarily mean the plunger is going. cheers, Aus

Yep, I only drew up a quick example and it runs all the time.......! One other thing Neil, "only require the counters to activate when I get a 24V input" I often harp on having inputs relate to the desired result of an output. All too often I see output/input relationships NOT relate to the actual desired result occurring. The input simply monitors some component in the supply chain downstream of the output, which doesn't work if there is a failure further down the line. If at all possible, monitor the final result of the output.....which you may already be doing. cheers, Aus

Hi Neil, although your Net 8 may work, it is not the best practice way of doing it. We often say on the forum that Unitronics does not charge for rungs/nets/whatever you want to call them. Break up all your MB 41, 42, 43, into separate ones etc. You could do every action needed for each of the bits in one rung instead. One way to do this is below. There would also be another net(s) to reset the counts, linked to a particular screen and button push that would reset the count to 0 and also reset MB40 etc. cheers, Aus

Haven't looked at your program. I often say on the forum that people are far too focussed on "Timers". Counts often perform much better, by simply using compares on a rolling count that increases, and finally resets/restarts, by whatever means you decide is applicable. From what you say, with only one timer active at a time, you would likely be far better off using this method. Counts are the forgotten simple tech, Timers can tie you in all sorts of knots if you aren't careful. Counts let you see what is going on very easily, interacting timers not as much. Blah Blah. cheers, Aus

For your other request about jumpers etc. find the correct information for your plc in all the literature available in various areas of https://unitronicsplc.com/support-technical-library/ Vision Hardware. cheers, Aus

engahm, don't do the O/S update just yet. I would not use 9.8 as your Visilogic program. The O/S in the plc will likely match the program you have on your laptop. Changing the O/S may make the program not work correctly. Going on O/S to Visilogic versions, you should be trying to use 9.4.0 which is not readily available, but you can likely get by on 9.3.1 which is here: http://www.unitronicsplc.com/Download/SoftwareVersions/Visilogic/VisiLogicSetup_931.exe Change your Visilogic installation to that version and see how it goes doing Joe T's suggestions. I think you are still going to have comms issues, as it looks like you may have an adapter that doesn't do all the serial communications correctly. Ensure that you have the wiring as follows, which is in the Help file anyway. cheers, Aus

Denis, are you saying that on each button press the required result is being shown on screen? In Vision this is always shown as a dot. If Uni doesn't do this as a default I am thinking it should be shot to the top of the Creator's urgently needed modification list. Being only vaguely familiar with UniLogic, I perhaps incorrectly recall that the ability to have screen areas "brought to the front" etc is available? This might be a way around the issue, by essentially having a "mask". cheers, Aus

Thanks Kratmel, as Joe says there's a lot of great info in that AB pdf. cheers, Aus

Hmmmm. I have great mental pictures of you restlessly prowling around the house waiting for midnight to tick over. Coffee jitters, keys jangling in hands, can't wait to push the right one in and turn it! cheers, Aus

Bumping an old thread for comments from all on something new. Recently I had to replace an old IDEC DIN rail supply and due to both urgency and enclosure depth limitations I had limited purchasing options. Ended up the best was a Mean Well but it did not have an earth. It was an HDR-30-24 and it looks like the entire range of this series is the same configuration, being double insulated. I haven't actually purchased a DIN rail supply like this for a few years, so maybe earths got dropped a while back. Being double insulated makes me intrigued about best practice regarding commoning of the negative, and also earthing practices etc, as it is in some ways now a fully isolated, floating supply. Love to hear all the viewpoints on this "new" variation. I can't find any information on the MW site regarding best practice, but feel sure that running it not connected to earth is going to upset a host of reading types that rely on miniscule amounts of electricity. In this particular case it doesn't matter, but I see the potential for it to be an issue. cheers, Aus

In having a quick squiz around the web I found this which might do the job. Have never used it, nor am endorsing it, nor am paid by them blah blah. And they even have stuff ready made for Unitronics gear. cheers, Aus https://www.nushield.com/dayvu_info.php

HI Skof, for the reasons and mucking around already stated by Joe and Kratmel, I'd be changing to analogue actuation. Most valves have the actuator as a separate device that is very easily changed, often only a few bolts. The various actuators are not expensive. In your case I feel it would be much cheaper to do this instead of going to extra costs and complexity, yet still let you use UniStream. The valves you already have may be unusual in that the actuator is integrated into the valve, but I feel sure if you looked at the maker's info sheets you'd find that going the analogue path would eventually be much cheaper and let you continue with UniStream. I've yet to see an actuator that can't be separated from the valve body with the body remaining sealed in the flow line. Kratmel's good observations on how steppers can lose track of where they are is important even if things are 24/7. It is a good idea to incorporate a home every now and then if you can. With analogue actuators, most of them also come with An out so it is easy to get the valve to go to where you know a flow rate will be close, and then fine tune it with the flowmeter (assuming pressures and mediums are constants) to better remember that position etc. 3 point control also works ok if an external flow meter is in use, in that the closed loop lets you get there accurately, but I have found it is much easier on all the mechanicals to do this with analogue due to the "softer" nature of the control. I've had Analogue actuation last many, many years, whereas other methods are generally much shorter life. That may not match other people's experiences, but it is my gained knowledge from years of doing it. Also, it is well worth remembering that any sort of actuator does have "slop" in the gear train that gradually increases over usage. Finally I'm intrigued that your motor will "break the valve." All the stepper driven stuff I work with cannot overpower the mechanicals enough to break things....that's the way they "home" at whichever end is chosen. cheers, Aus

Alex, to me it looks like engahm is simply wanting to revert Visilogic back to a previous state, yet changing the PC language back to English has not done this, even though the initial language change to russian caused the problem. Engahm, have you tried a complete uninstall/reinstall? cheers, Aus

I've always had limited success doing this, in that serial (and other) comms would never work properly due to the awkward "internal" interfacing involved. But by all means try! On Joe's comment about the programming computers, I also have some systems that are complete standalone dedicated for a particular job. They are not connected to anything, hence have no security risks. But they are the only way I can get some older things to work properly, now that "modern" thinking has over-ridden perfectly good stuff. cheers, Aus

Tozo, I'm wondering whether it is simply the normal delay in the Remote Op process. Have you tried waiting a second or more b/n actions? Windoze only needing single tap does lend weight to this ask. cheers, Aus