Jump to content

tmoulder

External Moderators
  • Posts

    196
  • Joined

  • Last visited

  • Days Won

    4

Posts posted by tmoulder

  1. I have enough gray hair to remember DOS and the concepts of "TSR" and "HIMEM". So I use high-number memory locations for certain canned subroutines like alarms and communications, and lower addresses for the system playground. Like Joe, I group things in 10s and 100s for station and process logic, ordered based on the workflow of the system.

    One exception to this, timers. Aside from timers assigned to those canned routines (which I try to avoid) I just use the "next operand" button. I'm rarely concerned with a particular timer and what it does, only with the effect on logic flow (alarms, for instance), and since there are fewer available timers than other bits, it makes no practical sense to reserve 10 timers for a station, then end up only using 3 or 5 of them.

    TM

    • Upvote 1
  2. Hi All!

    I'm working with Arkadi Samarov's 570 DHCP program and trying to pull and IP from my company servers. I had this working a couple of days ago, but when I went back to play further, it won't connect.

    Here's the funny part - the PLC is not showing a link. I plug it into my computer, and the link shows up, but the corporate router is a no-show, like the cable is not plugged in at all

    I'm trying to find out if a router can outright refuse a link - I thought this was a purely electrical signal, and had no dependency on IP or anything else, but looking online for the word "link" throws up too much static for a meaningful answer.

    Anybody see anything like this in the past?

    Thanks,

    TM

  3. Hi Damian,

    I had an experience last week that has some bearing on your situation. It was time to validate a couple of Omron-based machines as part of our pre-relocation inventory. I tried to get online with my linux/xp combo running CX-One, and failed completely. None of the comm drivers would work, serial or usb.

    I researched this ad nauseum, and found I was not the first to experience this problem, specifically with CX-One. Apparently Omron was late to the game supporting Win7, and alot of guys tried to VM XP to support their equipment. Omron provided no support for this at all, completely silent in fact. Some guys theorized you had to have Omron's drivers installed in the host, but I'm running Trio and PcanOpen both without any special setup, so unless Omron purposefully set it up that way, I couldn't imagine it.

    I tried every configuration I could think of, including making a separate XP service pack 2 VM and hacking the VM config file to simulate a 32-bit version. Still no love from Omron.

    In the midst of all of this, Omron began supporting Win7, with a laundry list of caveats, but I finally switched my host OS to Win7 and got CX-One operational. $#$#&&^ the %$^% off, because I had to lose linux to do it, but I also have to support this equipment, so there it is. I hate Omron.

    Long story short (I know, too late), the part you should know is that when I moved my usual VM from Linux to Win7, it went off without a hitch. Had to reassign the virtual CD-Rom manually, and reinstalled VMware tools as a precaution, but otherwise, smooth as silk - the main reason I went to a VM scheme in the first place.

    Of course, I use VMware Player (free). I've heard good things about the XP client in Win7 Pro, but I'm old enough to remember "Dos Mode" under Win98, and I find myself thinking about the compatability issues people had back then too, running what was essentially a stripped-down DOS VM. Microsoft is doing a better job these days, but virtualization is not their key revenue stream, so I have my doubts about it. And I seriously doubt you could move the instance you are running to another host OS in any way, shape or form.

    Best of luck,

    TM

  4. I've not seen this myself, but I would surmise you've got columns in your data table that contain more than one element. For instance, a column of MI, with 5 elements. When read, this column will load 5 sequential MI with the data from the table.

    Presumably, there's a limit to how many elements you can put into a single column, and you've run up against that. Look for such a column and try breaking it up.

    Best of Luck,

    TM

  5. So with my netbook reacting so slowly, I conscripted my wife's new laptop and started about setting up shop with it.

    It's a compaq presario CQ56 with a 2.6 Ghz processor. I cranked the ram up to 5 gig and installed linux mint 11, then added VMware Player and created a Windows XP virtual machine with 2.5 gig of memory assigned.

    After installing my various software, I fired up Visilogic 930 and it runs perfectly. This is particularly interesting, since a VM is traditionally poorer performance than a native install, but regardless, it does look like the problem lay in my computer, not the software.

    Thanks!

    TM

  6. Hi guys,

    Trying to set up a 1210 screen on my Acer netbook. Now, my trusty little Acer is not a bad computer - I'm running an Intel Atom N270 CPU at 1.6 Ghz, with 1.48 Gb ram. It's got Windows XP SP3, all the latest updates, and I keep my hard drive and registry scrubbed regularly with CCleaner.

    Nonetheless, Visilogic is running sloooooooow. I've noticed in the past it was a bit laggy (took a second or two for pop-ups to open and so on) when trying to program a 570, but the delay was tolerable, if not ideal. But the 1210 is, in fact, intolerable. The delay drags on for 5-10 seconds simply trying to select an object on the HMI screen.

    Perhaps I'm missing something. I tend to doubt that it's the hardware specs, given I can run Autocad on it without issues, and that's the most resource-devouring software I can think of. I tried disabling my anti-virus, but it made no difference.

    Thanks,

    TM

  7. Hi Damian,

    It seems like the capacity exists in Visilogic to look for these sorts of "gotchas". After all, if you try to use a CanOpen PDO or NMT, or a modbus command, the compiler checks to see if you put a buffer bit or busy bit ahead of the instruction.

    I really think the simplest solution (for us users) would be something in the OS that checks what HMI is displaying an ignores additional calls to that screen. Then it wouldn't matter how you called the HMI, and the problem would go away altogether.

    My two bits.

    TM

  8. Greetings Uni People!

    Okay, I'm trying to design a screen in a new V1210 project, my first time with an HMI bigger than a 570, and I discover to my dismay that I can't zoom in on the screen (cntrl-I does not work.)

    Now, this is a big issue for me. I don't have a 50" monitor and my eyes aren't what they used to be 10 years ago. To get things aligned on the screen "just right", I really NEED to be able to zoom in on parts of the screen.

    Am I missing something? I checked the menus, but no options there I could find.

    If zooming has been left out of large HMIs for Visilogic 930, I think it really needs to be put in.

    Thanks,

    TM

  9. Don't feel bad. I've done it, people I know personally have done it, people here have reported doing it - it happens alot.

    And on that note, to the Uni design gurus - I know, I know, we're supposed to be engineers and technicians and we should at least be able to read the little labels on the sides of the units. But we're still humans, and we get excited about our new toys and we goof up.

    Human nature being what it is, it's inevitable that people will continue plugging the enet cable into the expansion port, and I can count at least 20 hours (mine, another fellow I know, and now Rick) that have been lost on this.

    Sometime in the future, it might be a good idea to change the form factor of the expansion adapter away from RJ-45. Just sayin'.

    TM

  10. Greetings all,

    I've got a system with a V280 and 4 devices communicating over modbus TCP. Each requires a TCP connect, so each has it's own socket init, it's own IP Config block, it's own modbus busy bit.

    I'm trying to optimize the communication traffic pattern. In CanOpen, the PDOs are buffered, and comms are handled at the chip level. The socket arrangment in Modbus TCP seems to bear some resemblance to this.

    So here's a question:

    If I were to send a PHR message on all four sockets within the same ladder scan, what would happen?

    1. The socket traffic is independent and buffered in OS/hardware, ala CanOpen. Concurrent messages are not an issue.

    2. The first message gets through and the rest get dumped.

    3. Strange behavior and error messages.

    4. V280 screen displays "Self-Destruct Sequence Initiated". Evacuate immediately.

    Thanks,

    TM

×
×
  • Create New...