Sign in to follow this  
Followers 0
rajsiyer

Comm. breakage with Advntys STB on Ethernet

26 posts in this topic

Hi guys, My current project is one where ALL IOs are on remote Advantys Islands. We're using Unity pro + Premium PLC with only 2 ethernet modules ETY-4103 to pass logic on to the remote I/Os located in plant on an ethernet link. Yes we're using I/OScanning with the communication folder showing the ethernet link as correctly configured. Things work most of the time. However, there's a break in commns once in several hours, with the loss of communication lasting a few seconds!. We took this problem to Schneider who had us upgrade the NIP-7122 module on the island from version 7.20 to 7.21. The problem of communications breaking still continues. There are about 24 I/o modules on the remote STB island all with the required power distribution modules etc. I've kept the repeat rate setting in Unity pro I/O scanning at the default value of 60ms. How does one go about addressing it? Above all how reliable is Advantys? is it worth going by Advantys despite its lower costs or are In-rack I/os more reliable and thus more cost effective in the long run? Regards, best wishes, Raj S. Iyer

Share this post


Link to post
Share on other sites
Hey there! I have the same problem with one of my five STB in a plant, i run it on 30ms to get pulses (yes i need the fast scan) the STB is connected with a 10" CAT5 cabel to a switch and goes from there to another switch aprox 250m away, were its connected to the mainswitch to communicate with the PLC. All ther other STB runs great but this one is really bugging me. im watching the 10base LED on the STB and it should flash all the time but it dosent. it flashes like it should for maybe 15s then it stops, then it comes back again. Im thinking that maybe the cabel is to long from PLC to STB, runs through 2 switches and 250m of cabel. There is also 3 Schneider NSX that communicates through a TSX Gateway modbus->modbusTCP on that for end, but they works great. But with an lower IOscan rate. do anybody know the max cabel lenght for modbus tcp, and what to do? put an reapeater of some sort? rajsiyer: whats your distance in cabel from PLC <-> switch <-> STB ?

Share this post


Link to post
Share on other sites
nip manual http://www.global-download.schneider-elect...03688k01000.pdf stb installation manual http://www.global-download.schneider-elect...02947k01000.pdf good manuals to have, shows lots of do's and dont's Calle it would seem in your setup the cat5 cable length is definitely a problem Edited by cHud

Share this post


Link to post
Share on other sites
Hi Svensky! As CHUD rightly says.. any ethernet on CAT-5 must not exceed 100m. This is true even for your LAN/Internet cables as well. I think you have to go for a ethernet/fibre-optic media convertor + Optical fibre cable to cover 250 metres! What you will need are like this ethernet/optical convertors - 2 nos (one for each end of the optical fibre link) optical fibre cable - 250 mts. breakout boxes + pigtails set - 2 sets(one for each end of the optical fibre link) Your existing ethernet switches will directly interface with the media convertor. Regards, best wishes, Raj S. Iyer

Share this post


Link to post
Share on other sites
Thanks to both friends. Please look at my I/O scan settings Is it OK? I've not changed the defaults. Raj S. Iyer STB_settings.bmp

Share this post


Link to post
Share on other sites
That value means how often you want send a request after a time-out to check when a node is newly reliable in consequence of a fault for instance. In the figure you send I didn't see any mistake.. just I notice that one slave is through a gateway with address 2. That gateway may interfere if not well configurated (might be, not sure). Which one goes on fault before others? Or is the network goes down? What about other nodes in the network? What about switches configuration (if any conf.)?

Share this post


Link to post
Share on other sites
Thanks for your suggestions friend, I'll go to site next week and find answers to your questions. There's a redundancy issue here as well. Above all I'm left wondering whether this Advantys system is reliable and sound. Is it a good idea or bad? What is your experience? More of my headaches will be passed on to folks soon enough:-) Thanks a million Raj S. Iyer

Share this post


Link to post
Share on other sites
Redundancy?? What kind of? Advantys is a cost-effective system and I think is always a good idea. Anyway when they are used in a redundant system (hot-standby) they are not so reliable (due to ethernet connection by the way). That's why Schneider is moving (in such a systems) on M340 system of remote I/O (ready I suppose on next year). In the end I think is a good system. On the other hand in your case I see you have old software version (unity) so you can't appreciate some new features that help you integrating advantys I/O scanning and variables in Unity. Anyway maybe my judgment is biased... but it's what my experience says. I'm waiting for any news from site.

Share this post


Link to post
Share on other sites
Thanks friend, for your assurance. Fact is... we are new with advantys. It is certainly cheaper. Problem with this project is that our customers are fed up with comms. breakage so they've got a huge dispute with us over reliability. They have invested good money in Hot redundant PLC system, where 2 Premium hot standby plcs are networked through switches to a few separate Islands of Advantys. The Ethernet link & CPUs are redundant, while the Advantys is not. I worry about the success of this project as the customer has been forced to bypass the automation system and run the plant in Manual mode. Nobody is happy with our work and our billings are disputed. Thanks for helping out. I Genuinely appreciate your help (all of you)and taking interest in my problem. Ah... The latest Unity pro we're using is v4.1. You mean even this is old??? Regards, best wishes, Raj S. Iyer

Share this post


Link to post
Share on other sites
Is out Unity V5.0.. That version has many features that help tha advantys integration/configuration. From I/O Scanning page is possible to open automatically the Advantys configurator so Unity automatically import variables and configures all I/O scanning needed lines. And more over... Anyway with Hot-standby system you have to well configure the advantys modbus time-out (directly in the NIP by using webserver). So because during a swap of the CPU naturally the ethernet network take more time to re-configure (close socket, open new socket, etc.). During this time you have to configure the advantys in order to wait for that time before goes on comm fault. In such a way during a swap of the CPU all the outputs will be freezed to the last status. I don't remember the name of the parameter (it is a time) that you have to increase, but it is related to the modbus request time-out or something.... For your info there are advantys with two ethernet ports STBNIP2311 Ciao

Share this post


Link to post
Share on other sites
We also have had issues with an advantys island dropping out of run mode. This site has an Omron PLC as master, with 8x islands on Ethernet/ip. We have been unable to determine the exact cause of the fault as yet, so did had a band-aid of a remote reset via a digital output. One thing we have read, is that a certain type of din rail must be uses, and that there could also be firmware compatibility issues between the NIP and I/O modules on the same island. We have not had a chance to check this yet, but may be something else worth looking into.

Share this post


Link to post
Share on other sites
To the Italian, S. African, Aussie & Swede, it is update time, my friends... I tried juggling the repeat rates etc. No luck. Then I changed the switch to switch cables from Straight to crossed. These ordinary $20 D-link network switches actually don't seem to care! They work the same way (most of the time! as before). We use the common variety of D-Link's fast ethernet 10/100 mbps because the connexium switches recommended by Schneider are a whopping $ 800 /= each. The significant difference regarding the Connexium switch is that it is a managed switch. In our experience with various systems at different installations, the ordinary & cheap D-Link ethernet switch has never failed us. Not even once, despite lying unattended, even in very dirty panels! The reliability of the cheap ethernet switch does not seem to be a factor in the present problem:-). I'll have a go at it one more time. Ideas in my head are as follows. 1. Since this is supposed to be a HiPER ring topology, at least one managed switch is required. Here we plan to use a standard CISCO managed switch, a 24 port thing, costs $250/=. This cisco switch will replace one of the D-Link switch as shown in the diagram. Please take a look at the attachment.Network Schematic-Problem.ppt 2. I've no idea how to configure the CISCO switch for managing network redundancy, similar to the Connexium switch. I believe there have to be similar features. Any body who has managed to configure a generic managed switch in place of a connexium switch, may please share his experience here. I request you all to look at my diagram & also the connexium cabling guide for reference.managed_Switch_Connection1.pdf Thanking you, I remain, with best wishes... Raj S Iyer. Edited by rajsiyer

Share this post


Link to post
Share on other sites
Be sure to check - ALL switches will likely have to support HIPER ring, not just one. Cisco may have something else, but most likely you will have to replace all switches in your ring with models that support redundancy. Sadly, just replacing one usually won't work. If you figure out a way to get true redundancy by changing only one managed switch with fail-over times as quoted, please post back with your results!

Share this post


Link to post
Share on other sites
rajsiyer in a ring topology all switches must "speak" the Hiper-ring protocol so all switches must be managed. Anyway I don't think the problem is to find in the switches because they are normally transparent to the PLC communications and others. Did you check what I suggest you in my previous reply regardin modbus time-out in the NIP??? Did you assigned the correct monitored ETY in the cpu configuration??

Share this post


Link to post
Share on other sites
Hello, have you solved the break in communications problem you got ?! Thanks

Share this post


Link to post
Share on other sites
Yes,& thank you for asking .. Considering the interest on my problem at this forum, I should have answered my friends long ago. Sorry. Nothing was wrong with the system. The problem was with the optical fibre cable link. The fibre/Ethernet convertor on one end was failing intermittently due to accumulation of grain dust inside. We replaced it with a new D-LInk convertor and problem was solved. Raj S. Iyer

Share this post


Link to post
Share on other sites
So will you retract the statement that advantys is unreliable?? Glad you got the issue resovled. I have found in 99% of cases I have issues on networks being canopen ethernet , modbus or whatever, is there is something wrong with the cabling or the installation was done incorrectly. Very, very rarely is it equiptment failure especially in the beginning

Share this post


Link to post
Share on other sites
Sure thing Afrikaner, I retract my words. I will even go further to say that I'm a believer in the Ethernet RIO idea once again. Now Advantys even has modules with 2 ports!! I'll look forward to using them. Thanks again, for helping out. Raj S. Iyer

Share this post


Link to post
Share on other sites
Pay attention the two ports are useful for "ethernet daisy chain" not for a ring topology!! nor for a redundant system!! For those configurations you should focus on Modicon M340 "remote I/O" (actually PRA module, we will see in the future)

Share this post


Link to post
Share on other sites
The STB NIP 2311 which has two ethernet ports supports RSTP, suitable for use in a ring topology.

Share this post


Link to post
Share on other sites
billysmithde2 I'm quite sure of what I said. But if you have any official documents could you post the link???

Share this post


Link to post
Share on other sites
Hi, I'm using a modicon PLC with some STBs NIP2212 in a big gigabit optic fiber ring topology network !!! Everything is working fine ! But when testing the redundancy of the ring network (by disconnecting network segments) and the packets have to travel by the longest side of the ring (lot of network nodes) the communications are lost !! Through packet monitoring, we believe that the default TTL value (time to live) of 10 is not enough (more than 10 jumps) !!! I have contacted the local schneider support and they said that I can't change the TTL of TCP packets :O !!! Which is the best way to solve this problem !?! Thank you !!

Share this post


Link to post
Share on other sites
Read this: pdf doc # EIO0000000051 http://www.schneider-electric.com/corporate/en/support/automation-and-control-library/download/download-documents-iframe.page?URL=http://www.global-download.schneider-electric.com/repositorySchneider/index.nsf/DisplayProductDocumentation%3FOpenAgent%26L=EN%26App=Schneider%26p=606%26c=43,201,203,204,304

Share this post


Link to post
Share on other sites
Someone else beat me to the documentation. It's a fairly new feature. I have tested the following: M340 with NOC0401 module STB2311 ConneXium managed switch All with RSTP.

Share this post


Link to post
Share on other sites

Create an account or sign in to comment

You need to be a member in order to leave a comment

Create an account

Sign up for a new account in our community. It's easy!


Register a new account

Sign in

Already have an account? Sign in here.


Sign In Now
Sign in to follow this  
Followers 0