Joe E.

MrPLC Member
  • Content count

    1593
  • Joined

  • Last visited

Posts posted by Joe E.


  1. It's been about 7 years for me since I set up a Prosoft PLX31-EIP-PND module to communicate between a ControlLogix and an S7-300. The software was free and easy (on the AB side). I had to get Prosoft support involved to get the Siemens side working (their instructions and samples were in TIA Portal while our system was Simatic Manager), but it worked very well once it was going.

    Be aware that it will work with an S7-300 with onboard Profinet but the add-on network module has 2 versions. One will work with this while the other won't. I don't remember the catalog numbers or have the ability to look them up, but the cheaper one didn't support I/O and that's how the Prosoft module shows up to the S7. The more expensive one was supposed to work but we didn't have any to test. This is all in the S7-300 world. I don't know how it would work to use it in Portal.


  2. So, a quick followup. I opened up my Win10 VM again this morning and tried to connect to the 5/05 but RSLinx couldn't get the COM port, said it was busy. The Keyspan utility showed it bouncing between idle and busy about every second. I ended up having to delete the driver in RSLinx and set it up again from scratch. Even so, auto-configure took several tries to work. I fired up my Win7 VM (same Win10 host) and it connected and worked out of the box with no issues. There may be an issue with the Keyspan driver (I'm using the latest one) or with how Windows 10 is handling things. I'll definitely be keeping my Win7 VM...


  3. My go-to USB-serial adapter has been the Keyspan (Tripplite) USA-19HS for over a decade. This thread made me realize that I'd never tested it in Windows 10, so I just did. It took a reboot and a few tries in RSLinx Classic, but I did get it to  connect to a SLC 5/05's serial port. I've used it in the past to connect to a Watlow F4 as well, but using Windows 7 on a different laptop.

    My software is in a Windows 10 virtual machine running on a Windows 10 host. The last time I used a Watlow controller was with Windows 7 VM on a Windows 7 host.

     

    1 person likes this

  4. 2 hours ago, JJH said:

    They haven't been happy with having to work with Ethernet-based encoders and for me, I think the latency with Ethernet communications isn't the greatest solution for trying to sync motion or be accurate. Obviously you obtain data from a servo drive in that fashion, but all the feedback is held in the drive. I honestly haven't used the CIP encoders much so I could have a gross misconception there but I often cringe when an idea comes of of controlling critical data/positioning via Ethernet.

    Interesting, and I agree to an extent. It depends on the speed of the movement and how busy the network is. One of the places I've used one was on a dancer arm riding on bimetallic strip to detect delamination and ripple. The dancer mechanism wasn't the greatest but the encoder did really well. The network had a ControlLogix PLC, 7 remote IO racks, 5 HMIs, 5 or 6 drives (VFD, DC), a PC (doing servo control), another PC doing data logging, and a SCADA PC. The strip was only running at about 40 ft/min, though, so it wasn't a super high speed thing.

    In a true motion control setting, I don't think I'd use them either.

    In your customer's application, the issue with 0-10V signal may just be inability to get an analog input module for the PLC. The last time I had a 5069 analog module quoted, the lead time was long and fluid.


  5. I would lean towards the PROX sensor and timer solution myself. Couple it with conveyor speed feedback (from an encoder or the feedback from the drive, or whatever) so you can tell for sure how much conveyor has passed by when the sensor turns off. If the drive is stopped and the sensor is on, you know exactly where the cleat is and may or may not be able to drop a carton. If the sensor is off, you can wait until the drive is at speed and then drop after a time after a cleat passes.

    Out of curiosity, what's their/your reluctance to use a CIP encoder? I've used them in the past and have had zero issues with them. That processor isn't motion capable, so you wouldn't be able to use the built in motion instructions but you wouldn't need to for this.

    For the incremental encoder, figure out how many counts there are between cleats and use a sensor to see the cleat. When the sensor turns on, zero the HSC count. That eliminates errors due to conveyor stretch and such and makes it independent of conveyor length.


  6. When I go to the downloads page, it shows that firmware 19.015 works for the 1769-L32E, both series A and B. You would need that firmware and v19.01.01 of RSLogix. I'm not sure why your screenshot shows the red X there.

    I created an empty project with the 1769-L32E at v20 and tried to change it to v19 but it wouldn't let me downgrade. I was able to create a new project at v19.01.00 (the version I have installed). You may have to do that and manually build the I/O Configuration and copy/paste or export/import the program.

    To get the firmware, go to their download center:
    https://compatibility.rockwellautomation.com/Pages/Home.aspx
    Click "Downloads by Product"
    Type in the model number in the "Search" window. Click on it, select your series (A or B) and click on version 19.015. Click the red "Downloads" button and it'll take you to a page where you can select files or choose firmware only. You'll have a "Downloads" cart icon at the top that you can click to start the download.


  7. 18 hours ago, PaulKim1003 said:

    Thank you @Joe E.

    It is 1769-L32 CompactLogix 5332E Controller (revision 20.19).

     

    You're already at the latest firmware version for the CPU. What minor version of RSLogix v20 are you running? That matters too. I'm running 20.04 but it's been a while since I needed that version. The latest is 20.05. If you're not sure, open the project and go to Help -> About RSLogix 5000 and it will tell you your version. I would also seriously consider @Otto Automation's suggestion of "down"-grading to v19. Version 20 has always been a....problem child.


  8. Hmmm....

    So the Knowledgebase has a bunch of articles about that error happening upon other events (like going online, importing L5K, etc.) but not yours. I opened one of them and it gives a few solutions. They seem to refer to v20 a lot, so it may be a good idea to make sure you're running the latest minor revision of v20.

    Error when going online (access level: TechConnect)
    Solutions:
    1: Delete Harmony files (access level: Everyone)
    2a: Repair the installation (access level: TechConnect)
    (looks like the "normal" way to repair an installation from Windows add/remove programs)
    2b: Force software to run as administrator
    2c: Repair installation of FT Services Platform
    2d: Disable UAC (access level : Everyone)
    2e: Check system tray for WEBROOT SecureAnywhere. If it's there, rt-click and shut it down. Then go to Control Panel and repair RSLogix 5000 v20
    2f: CCleaner can cause it as well. Repair RSLogix install
    3: If you're using a Logix Echo Controller, uninstall FT Logix Echo and FT Linx, reboot, and reinstall Logix Echo
    4: Uninstall the problematic version of RSLogix/Studio 5000, reboot, reinstall

    Error when going to Who Active (access level: TechConnect)
    Upgrade to v20.05

    There's another article that specifically refers to the error when opening a specific project file but was linked to as a general fix for other situations (access level: TechConnect)
    Solutions:
    1: Hold SHIFT and CTRL when starting RSLogix (this restores all settings to factory default)
    2: Export ACD to L5K and then import it back into a new project
    3: Check KB article linked in steps 2a-2f above

    Lots of circular links where their articles refer to each other, so I may have some links crossed. Basically, I'd first upgrade to the latest minor rev of v20 before doing anything else. If that doesn't work, or if you're already there, try the repair installation, then Delete Harmony files. The rest of the "fixes" seem to be really long shots to me....

    What PLC processor are you using? Can it be upgraded past v20? That may ultimately be the best option altogether if you can.


  9. So, I found another KB article (access level: TechConnect)
    The max length you can send is 82 characters...but that may not be your issue.
    They provide sample code to split the message up into 82 character pieces that are sent individually.

    I'm by no means an expert in ASCII comms, so I'll have to punt on this one. I've done some of it in LabWindows and in an S7-300 but both have been a long time ago.

    Does the printer respond as expected to what you're sending? By printing or whatever you're having it do? You may not need to manually read the Rx buffer of the Tx instruction completes successfully. The Tx instruction may be verifying the ACK signals itself.


  10. I never tested it, of course, but the Prod-cons connection path I built offline didn't involve the network at all since they're in the same chassis. I just added the other processors into the chassis in the I/O tree of the project and pointed the consumed tag's source to the other processor in the chassis. If they were in different racks and on different non-routable subnets....I think a NAT device or router of some kind might work but I've never had opportunity to test one with IT being the way they are.

    Caveat: my only experiment with produce-consume tags didn't work well. I could only get it to work in one direction because one PLC was a 1756-L55M12 at v16 (max supported version) while the other was a 1769-L30ER at v20-ish (lowest supported version, I think). The -L30ER could consume tags from the -L55 but not the other way around. I saw reports of it working by lying to the lower version PLC about what was at the other end of the connection but I never was able to get back to it to test and found the flexibility of being able to edit things online made MSG instructions more attractive in that setting.


  11. I would think you'd be able to since they're in the same chassis. The routing would be through the backplane and wouldn't involve the network at all.

    As a quick test (I don't have the hardware), I made a project with 3 -L83ES processors in a 7-slot chassis and set up a produced/consumed connection and compiled it with zero warnings/errors. I would be really shocked if it didn't work.


  12. One way I've seen it work is to used a Bridged connection in your VM with DHCP enabled. Attach the bridged adapter to the host's NIC that's connected to the company network that provides the licenses. Then, and this is most important, cross your fingers. @pturmel's point about internet access for the VMs is a key troubleshooting step. At my previous location (different division of the same corporation), we could lend/borrow licenses between PCs but here it's blocked. I tried it once between 2 laptops that could ping each other but as soon as I pointed FT Activation Manager at the other laptop, our firewall locked it down so I couldn't even ping any more. It even gave the host laptop a popup warning about an attack. I have a very strong suspicion that you'll have to involve your IT folks in this.

    You *may* be able to install FT Activation manager on the host and have it point to the server, then have FTAM on your VM point to the host. I really don't think that will work reliably, though.


  13. Is this a new program or one you're taking over or modifying? If it's an existing program, look around for other places that may be writing the DispContrast variable. Could be in a program or in a screen tick event or someplace similar.


  14. Hmmm....looking at the manual, it says it's a read/write parameter. You say that writing to DispBrightness works but DispContrast doesn't? I don't have a screen I can use to test (lead times keep getting extended...) but if one works I'd expect the other to as well. Have you contacted their tech support? In my experience, they're good at resolving issues.


  15. If the system works for them, you may have to borrow a laptop from them to make the change. Who knows what special firewall/security/etc. software/settings are different between your PC and theirs. If even their laptops don't work...like @pturmel said, they need to sort out their source control system. It seems like a bit of a crazy kludge to use a Profinet via a 3rd party system to connect to the PLC. If it normally works and has recently stopped working, then whoever set up that system needs to support it. Or, I guess you could bill a nice hourly rate to try to sort it out yourself.