Ausman

Seasons greetings to all! One for the creators in particular. Would it be possible to have a minimum and maximum put into the Linearisation function so that I don't have to do extra equations "externally" to achieve this? Nearly all the Linz I do are only active over a certain range, and I have to write compares to set the high or low figure that over-rides the output once the input ranges are exceeded either way. It would make things vastly easier if it could be done within the Linz function itself. Pls lend support to this idea so that it might get done! Or let me know if I have been dumb and missed an easy way of doing this! cheers, Aus.

Hi again all, One of my jobs has sprung up something unusual, and I am pondering solutions. I have a V230 on a site, and for various reasons it is 25m from what it is controlling, all done via multiple long shielded runs of multicore which works perfectly. Improvements are being considered which would create further length and practicality issues, so I may have to move the 230 direct onto the controlled item. But the 'easy remote control" aspect still needs to exist. I am wondering if there is a simple "repeater" function somewhere that I have missed. In that I simply hook up another V230 via canbus and it imitates the screen and keypad actions exactly as seen on the Master, without actually doing a lot of programming on the slave. Much like remote access works, but running as a direct link. I know I could set this up via programming on the slave (sadly in this case it is a fairly complex program), just wondering if there is a special function etc somewhere that I have missed that would make my life much easier. I'm also considering doing this with a dedicated little laptop or mini puter running remote access and a direct cat5 connection, but would prefer the real thing as it would likely be more stable and easier to use. If such a thing might be considered as a useful addition by the creators, how long till it appears?!!!! cheers, Aus

Hi Flex, I have 2 observations/suggestions. 1). Is the printout an actual copy of the program that supposedly worked? Or could it be something printed during development, that someone else has been told is a "copy of the program"? ie are you sure that the print is the exact code that was in the controller that worked? It all sounds a bit sus to me. 2). Why reinvent the wheel? It might be a good starting point to do things exactly as written and see how/if it runs. I'm sure you can get copies of older versions of programs and O/S's by asking nicely here. Write it as printed, put it all in and see how it goes. If it works then you are one step ahead. If it doesn't then you are a bit back to point 1), but at least you have a much better idea. No point working on something if your baseline is totally incorrect. Cheers, Aus

A few weeks ago I replaced a power supply that was playing up. I finally got to preparing invoicing etc and looked at the supply again to clarify something. I discovered that I had marked it with S/N. Huhh? Why have I marked it as wanting the serial number? Did I talk to the agents and they are replacing it? Don't remember doing this. Must have, but forgotten. Hmmmmmm. Thought about it some more, without luck, during finishing the invoice and then put the supply into my electronics recycle bin. During this process I managed to turn it around 180° and everything became clear! My marking of S/N became N/S....Not Serviceable!!! Did I really do this? Yep! Did I feel stoooooopid? Yep! But I thought some of you might laugh. Normally I mark things as R/S (rat poo...being polite) so this is the excuse I'm using! cheers, Aus

Hi again all, today I discovered, quite by accident, that during a Remote Access session you can use the main computer keypad to enter numbers, returns etc when viewing input screens. It makes it a fair bit easier than fiddling around with the mouse to "push" the buttons on screen. Worth trying and seeing how much you can do, in my case it is mainly V130s. cheers, Aus

Hi all, a bit long, but needed. Sorry. Well.....I have gone back to my stored copy of Win 7, and disabled win10 notifications and downloads. I have simply had enough of numerous problems with all sorts of things. The final straw was too much background activity going on reducing speeds to a crawl. Reached the stage where even the taskbar would lock up, and after a while "maybe" come back. Background services running at 50-91% for days and unable to be killed via any method....hopeless....and no way of finding what started them either....you drill down and get to "that process doesn't exist". I know that Msoft do a pretty good job at allowing for all the myriad permutations of a computer out there, and that I have unusual PLC & control programs with virtual serial ports on my system that might have got things in loops looking for things that aren't really there, but it was ridiculous. I don't have anything that unusual. I've wasted the last 4 days putting up with it, trying to accomplish real work whilst chasing problems that shouldn't exist in the first place. So I have gone back to the OS that I know I had really stable and fast. For the record, the instances were taskhostw.exe being called by something I couldn't find. Someone more knowledgable than myself could likely track them down, but I had a good try at it without luck. It seemed that all the calls originated from task scheduler. I would get a random start on an instance, then another, then another. At one time I had 11 of them running...as if there was a failure somewhere so another instance started again without stopping the first. I suggest you take a really good look at what is being called from scheduler, you will be amazed at the breadth. A lot of items are classified as "random start" after X time. I disabled a host of things I felt might possibly be the reason, and also others I felt were simply invasive/ridiculous, but the processes would eventually pop back up and continue running, even for days if left on, so my choices perhaps helped a bit, but didn't cure it. In 6 months or so when 10 should have matured considerably, I might try doing a clean install, rather than working from the upgrade. But I'm not that happy at present. I suggest that you periodically have a look at what is running via Task Manager. And oh yes...in 10 it displays this a bit simply, you might have to use procexp.exe or similar to really see everything. If you have a few instances of taskhostw running, watch them carefully....when my system was ok I believe that there was only one going. In hindsight I think that I had the evolving instances gradually building over a few weeks and only noticed them when they really started to hog CPU time. cheers, Aus

Hi again all, I have a suggestion that would get around a lot of the issues involved here. Every single jigger in my life that writes to any card, be it SD, CF, TF etc.....all of them are innately able to do a format and prepare the card to the way they want it to be. This ability should be included within the PLC SD card options accessible via Info mode! It would make things so much easier. Instead of all the hoohah preparation we need to do with the card out of the PLC, we should be able to put the card in and let the unit prepare it itself. Such an improvement can't be that hard to implement, and the only thing that I can see it would need would be an extra password level be able to be set to prevent unauthorised formatting taking place. If this utility was available, it would be a very useful thing, as well as a work-around to the grief that is currently occurring. Cheers, Aus

Hi all, the process I use for the write is described at the end of my post "Write table to SD with date stamp?" where I also have a jpg of the procedure etc. This has worked perfectly, the card is permanently mounted into the V130 slot and is just there for logging and backup storage. I prepared the card as Alex has described in #7, the system is stable and on a ups. I upload the logs at the most a month apart, and then delete the block, so that at no time am I anywhere near the 64 limit. In deleting the block, though, I perhaps have repeated the procedure that Jerther describes in post #8 to get the problem. It is interesting that his problem is a file with only 4 characters to the name, same as mine has ended up...see the jpg. I don't exactly remember how the file appeared, it was just noticeable next time I logged in to do the upload, and then I couldn't do anything with it remotely. But it is ok in my uploaded data storage off site, so it was fine at one stage. cheers, Aus

Dear R. Thanks for your post. However, perhaps I didn't make myself clear enough. I can run all of my Unitronics programs fine, no problems at all. It is programming/maintaining other ancillary equipment that is part of my older plc installations that is the issue. Under Win 10, all of these programs appear to install ok without any error messages, but they don't work correctly no matter how much I tweak things. However, under XP they work fine straight off the install, on the same hardware. I only mentioned my issue in case someone else was having similar troubles. Cheers, Aus

Hi again all, further comments re some things not working ok. I have many older installations that used the most up to date technology at the time. They are not really 'out of date', just older than 5 years. I have found that Win 10 simply does not work ok with a host of the programs that are essential for maintenance of these existing installations. I am talking puck programming, interface setups, actuator programming etc. Although compatibility is meant to work, it doesn't. Nor does running things via a virtual machine. It must be a driver issue, but even though many of the manufacturers say that things will be ok, they aren't. I spent a frustrating day yesterday on a site, trying to get one program to work without any luck, using all sorts of tweaks. Silly me had not checked it before visiting the site...didn't twig that I hadn't used it since the 10 upgrade. I finally gave up and today I moved a few partition copies around, putting my copy of XP SP3 back onto the laptop. I was back in business immediately, so it is definitely OS related. I am going to now go down the path of always having my older laptop with me with a 'real' version of XP to use. It will have all my programs to use when directly hooking up to things, but won't be used for anything that needs internet access. Some time in the future I will experiment with a dual boot of XP and 10 on the newer laptop, but I have read contradictory reports on this, and can't afford the time at present. I think it should be fairly easy using EasyBCD etc. I guess the essence of this post is that if you are having trouble with something, give up now and use a known OS that worked ok before! cheers, Aus

Dear HighHTech, I suggest that you read my 26th August post a few above about all run as admin settings being lost during the upgrade. Might be the reason for your troubles. cheers, Aus

Hi Flex, I've been waiting for someone more knowledgeable to suggest something, but looks like it is worth me giving my pearls of wisdom. It might be an easy way for you to solve your dilemma. I have trouble with Version Swapper when moving to a newer version of Visilogic. My way of getting around this is that during each new upgrade install, I delete all the current Unitronics entries via remove programs. After the new installation finishes, I copy the entire Visilogic_C folder and paste it into my dedicated "Unitronics old versions" storage with the relevant name change to what version it is. I can then easily reinstall it into the new Version Swapper by copying it back as a virgin system and letting Swapper pick it up from there, never touching the source folder. As I don't do this too often, I may have forgotten some of my process, but it is the essence of what I do. Hope it helps. cheers, Aus

Hi all, is there a way I can delete incorrectly named files, or change the names of files being remotely read via card explorer? 1.0.3.17 working on a V130 in this case. I can find no way of doing this. For some reason, some log files I use have corrupted their names and I cannot delete them because I get the "file name not in 8.3 DOS format" message. It's a bit of a catch 22, because I can't change the name either. Other than doing it manually next time on site, is there a way? Could this be incorporated in future editions, to allow for such situations like simply changing names, which doesn't exist at present etc. Or is it simply that the PLC can't handle such a request for file changes like this..ie it can only work with the 8.3 format and you're stuck if it isn't correct? cheers, Aus

Hi all, perhaps some of you might be in the same boat as me. I have a number of older programs that are crucial. Prior to the upgrade I was on Win7, and had been able to get them running fine using the "disable visual themes" setting in the compatibility tab. If I had this off, all sorts of weird things would happen with buttons, drop down menus etc. It was amazing how far reaching changing the setting to on or off changed the way the programs worked. On moving to Win10 the programs would not run correctly. So went to compatibility and huh?!#*, disable visual themes is not there! I hunted everywhere for a solution to this seemingly silly microsoft decision. After many weeks of looking, I have found that the capability still exists, just isn't shown. Stumbled across this solution for Win 8, but it works fine in 10 for me. Scroll down to the post by Morten near the bottom, with a little registry work showing. https://social.technet.microsoft.com/Forums/en-US/b49de36e-d36f-4f7c-a63e-b70c49c86efd/compatibility-administrator-and-the-old-setting-disable-visual-themes?forum=w8itproappcompat Thankyou Morten! Working perfectly again. Ahh Microsoft, why do you do such silly things? Huh? cheers, Aus

Hi Flex, Occasionally I have trouble wrapping my brain around how Unitronics timers work. When I first started years back, the brand I mostly used had many types of timers available, and adjusting to 3 has been "interesting" at times. Timers have their place, but sometimes I end up using a dedicated counter which often does the job much better and is easier to interpret than some convoluted way of manipulating timers. With the counter you can make it go both ways with ease and manipulate a host of things very simply through multiple compares that adjust, stop, restart, and see where the count actually is. cheers, Aus

Very well said, Keith. Hope it is taken on board for all you guys in the US. Even better if they could make it multi-voltage. cheers, Aus

Hi Steve, Re the hunting issue, if you are down to variations bigger than you like occurring, and you are only moving your outputs by single digits in that time, then you are a bit restricted in what you can achieve. That said, a number of suggestions: I don't use PID etc when large load variations can occur.. have found it good for roughly constant stuff...but hard with big variables. So I do much like what it sounds like you do. I have had luck in introducing small time delays into the output equations, be it small number of scans based or fixed interval. Another thing to check is that the drive actually recognises minor control voltage variations. I have seen many drives not react to tiny changes, only adjusting once things go past an internal reference to a previous input...essentially they have their own interference filter on their control inputs. Often there is a setting change to get a reaction no matter how small. Also worth a check with a meter during operation to make sure your control change is reaching the drive properly, as well, given the use of the signal isolator....might be doing the same thing with filtering. From memory those drives also have inbuilt ramping control up and down (memory=age=hazy!!) so I'd make sure that isn't interfering somehow as well...you'll likely have to turn it off completely and do all the ramping via adjustable rate maths in the plc. If the drive has inbuilt ramping and it is on, I can definitely see it giving you problems due to the lag involved. Hope my little bit helps, cheers, Aus

Hi Joe, I know it's not the answer you are looking for, but if real estate is the issue you might get a relatively easy result using a multi input signal condtioner. It would be a toss up on extra wiring and complexity vs ability to use less I/Os. A quick google brought up this pdf: http://www.omega.com/Auto/pdf/he-x_series.pdf. I'm sure there would be others around with more inputs available, but regardless even this 8 channel one would use much less space as you would likely get 3 conditioners on the rail space of one DI8. Or perhaps you could use a gang of pheonix singles or similar, or a mini relay module to do much the same thing? cheers, Aus

Thanks Alexander for the reply and I apologise for my delayed thanks...have been away. Hmmm......Pheonix........ I bet you buy them far cheaper than us little people can! cheers, Aus

Hi all, The upgrade to Win 10 went ok, but I subsequently had an issue doing downloads via the net...version 9.7.55/0. My ticker stopped working for a while as the site was hours away! All looked ok until the very last moment of the transfer where it usually goes to the run box, instead up came the big error window that essentially says to send the info to Unitronics! Tried twice with the same result. Didn't take a screen shot as too worried as I had to get the system going again asap. Quickly discovered that the 10 upgrade had turned off all of my run as admin settings....every single one of them in every program & shortcut I have it set on. Worked fine after turning it back on, but it is definitely something to be aware of. Thankyou Microsoft.....not! Just as an aside, the upgrade changed my date formats, language etc to US, and I had to manually shift it back to Aus. Here's hoping they don't do something else silly as they evolve the release. cheers, Aus

Hi all, yet again I have had a dud screw/cage in a terminal plug I was fitting up on a 120. It is annoying when the screw won't even pick up a tiny amount of pressure, let alone what I normally tighten things to on bootlace pins without any trouble at all. And before you all harp about ham-fisted I must be....I'm not, so shhhhhh! Definitely a dud and a big problem when it is the only one you have onsite and you are faced with a 300km round trip because of a badly made screw/thread worth maybe 1/2 a cent! Over the years I have sourced a few different types as spares, but even though they all are supposedly to a standard, their quality of fit/match to the Unitronics ones is not quite right. Mostly just a tiny bit tight on the shell. I have found this post : http://forum.unitronics.com/index.php?/topic/882-vision-v130-replacement-io-pluggable-terminal-blocks/?hl=terminal+blocks but what I really want is for the creators to reveal their supplier so that I can get some spares direct that I know will fit correctly. Or anyone else to suggest a block they have found that fits perfectly. And perhaps we could start putting combo slot/phillips heads into them as well? Hmmm? cheers, Aus

Hi again Marius, that sensor gives you a digital output, and as it is 5V I guess you were planning on using the AI in 0-10v mode to read when it goes on and off as it would be able to read that voltage level directly. However, I think you will likely get time interval problems doing this. My suggestion would be to interface a small SSR into the mix and put the output from the SSR into a dig input like I have already explained. Given you are in Aus, have a look at this. I have used them very successfully when I have had to read from a very low level LED driver into a 24v dig in. I think it would work ok, but no guarantees! http://www.jaycar.com.au/Electromechanical-Components/Relays-%26-Accessories/Solid-State/PCB-Mount-Solid-State-DIL-Relay/p/SY4090 If you can afford it, I would still go with a dedicated industrial 24v inductive. Will probably save you headaches & $s in the future due to the innate durability, and then the extra $ are well worth it. cheers, Ausman

Hi Marius, Like Alex says, more info needed. But for now I am assuming that your "vent" is actually some sort of blower with blades. I do this a lot and my routine is a good quality inductive from Sick or similar. You get a straight forward 24v digital output that you read using one of the high speed counter inputs to allow for any times the count might be an issue related to scan times, unless it is going to be fairly slow and you know for sure that every "on/off" change will be read correctly at every speed it runs at. You then use a normal DI. I always mount the sensor so that it is reading something from the slowest moving part of the blades. Most of the time I can easily achieve a mounting to read from the steel bolts in the hub. Perfect for good, error free inductive operation. I have had a few odd times where I have had to make a special end cap for the motor shaft that mounts as an extension on the motor fan end, and is a series of lobes that the inductive reads. You can often buy this sort of thing off the shelf. Once you have the inductive reading ok you can work the maths any way you like to arrive at your rpms. This really relates to how many pulses you are going to get per second, to enable good accuracy. You accumulate the pulses into a count, and then at whatever interval works best you do the maths and reset it to zero to start again. So the more pulses per second the better, as you can get a faster responding measurement. cheers, Aus

Hi Lawry, That's a useful idea and a good workaround. I have been working on the older system again and it still impresses me with it's very useful features. cheers, Aus

Hi Cara, Giving this a bump! Any answers? cheers, Aus