SLC5/05 OPC Sporadic

Hello,

I have RSLinx OPC Test Client tied to an OPC Topic linked to a SLC5/05 PLC (Series C/9). There are about 2300 tags in the test client (essentially simulating what the HMI is reading).

When I connect the test client to RSLinx, the data comes in. Immediately, many of them go to 'bad' and in RSLinx the Invalid Packets counter increases. The data comes back and goes away frequently, but always a different packet (i.e., N52:100-104, F36:0-254, ...). During this time I can ping the PLC at 1ms with no delays, so it doesn't appear to be a network issue.

There are other devices connected to this PLC that are polling the same or similar tags that have no problem. Example: Proficy iFIX polling through ABR driver to RSLinx (I believe an API-level connection through the driver channel). This connection IS slow at times and could be indicative of the same issue, maybe masked.

I've verified that all tags are accurate to the datatable in the PLC and the Optimized Packets represent the tags properly as well.

What else do I need to look at in order to figure out what is happening?

Oh yeah, S:2/15=0 and S:33/7=1

Unsure what to look for next, but has been a frustrating experience

Just as a test have you lowered the number of tags ? I would try 1000 and see if they all stay up or get another OPC to test and see if its RSLinx giving you the issue, but either one I think you need to try and narrow it down, I dont know of a real way of troubleshooting this.

geniusintraining said:

Just as a test have you lowered the number of tags ? I would try 1000 and see if they all stay up or get another OPC to test and see if its RSLinx giving you the issue, but either one I think you need to try and narrow it down

Click to expand...

I had it narrowed down to 8 tags previously because it 'appeared' it was a slow-connection issue. The data is taking a long time regardless of this problem, but I think the problem is the real issue. I can see packets sent / packets received taking about 5-7 seconds if I watch the Optimized Packets window.

I will try a few less tags again and see what comes of it.

38 tags - each optimized packet takes 3-5 seconds. No invalid packets yet, but could happen still.

Another thing that happens from time to time is that some of the SLC5/05's in the Ethernet Driver will get a red x, then every once in a while they will come back and think they are controllogix. When that happens - communication, of course, stops entirely. Not sure if that is related, but wanted to throw it out there.

Yes your losing your COMs on the SLC (the Red x), RSLinx has a diagnose screen that may help, try right clicking on the PLC

geniusintraining said:

Yes your losing your COMs on the SLC (the Red x), RSLinx has a diagnose screen that may help, try right clicking on the PLC

Click to expand...

I did look at the station diagnostics previously. I do not see anything there that looked odd except:

Replies timed out: 436 (stable)
Carrier Sense Errors: 758751017 (increasing)
Deferred Transmission: 5746984 (increasing)

I plugged one in and this is mine.... do you have a spare 5/05?

I do not have a spare SLC5/05.. But I can look at others on the network.

I am using a VPN and RDP to look at a system remotely. My screen looks similar except I have replies received and the 3 items in my post above. Otherwise identical.

Any chance of swapping this one for another in the plant? it will at least tell you a network or PLC issue.

geniusintraining said:

Any chance of swapping this one for another in the plant? it will at least tell you a network or PLC issue.

Click to expand...

Unfortunately not. It is a 24x7 plant and this is also one of the central pieces of equipment. I have heard that *maybe* 20+ devices communicate to this PLC also. So - would be pretty catastrophic to remove it.

I dont know enough about Ethernet coms to give you much advice (I sell USB cables ) but I think you need a spare regardless

Another shot in the dark... try downloading another OPC

Kepware has a demo and see if you lose it on there also, I am 90% sure you will then you will need to look at hardware, PLC, cables and switches that are in the loop

https://www.kepware.com/en-us/products/kepserverex/

I'm actually testing with Kepware (my company is a Rep for PTC who owns them). I've also tested with Kepware's OPC Quick Client and RSLinx Quick Client and the ABR driver mentioned previously which hits the RSLinx API. Been fairly thorough on that front.

Carrier Sense Errors: 758751017 (increasing)
Deferred Transmission: 5746984 (increasing)

Click to expand...

These are VERY strong indications of a damaged port or cable, or a misconfigured port (like a duplex mismatch).

Those values should be (and stay) zero on any ordinary Ethernet switch with an undamaged cable and connectors.

I will update when I find out more, but I have put the IT folks on this one for now. 20+ connections to this PLC would not cause this, would it?

I've heard it has that many connections - is there a way to verify this in RSLinx or the HTTP interface?

amueller said:

20+ connections to this PLC would not cause this, would it?

Click to expand...

Sounds like to many to me... But Ken would know, I did find this https://rockwellautomation.custhelp.com/app/answers/detail/a_id/587712

I know I would try and limit DH+ to no more than 10 nodes but Ethernet I would think is able to handle more than DH+, I also think it depends on the data you are sending/receiving, if all of your 20 nodes are passing 2300 tags... thats A LOT of data