Tag intermittency

Hello mates,
ive been suffering some kind of error, if thats the case, in some of my tags. im using two datatypes affecting 8 retorts, the thing is a few days ago only one of the retorts started having some comunication problems with only a few tags and not all the time. the problem can last from seconds to couple of minutes and it affect the functionality but the comunication cut doesnt seem to affect the chart of the tags. there is no log input or error msg…just null value and thats it. there are some pics and thanks for the help in advance.
image
image
image

during the error the same tags in other datatypes arent affected and this has been working for about a year with no similar problem.

Someone doing online edits on that PLC?

not at all,the only one allowed is on holidays jeje. i have also detected that there are “two groups” of tag in the error secuence. sometimes one group, sometimes another and even both groups.

  1. What is the Load Factor on that PLC?
  2. What is the scan class rate for these tags?
  3. Have you been adding tags to this system?
  4. Are those tag scripts doing tag read and writes to the same controller?.
1 Like

1-sorry im not used to that term, ill ask tomorrow to the plc programmer but the plc is allen bradley SLC-5/05
2-default (1/s)
3- the tags from “LOADED_RECIPE” have been added and working for almost a year and the tags from “VALVLES” have been added and working for about 4-6 months
4-only read that im aware (im not the only one working in the system and the script is older than my experience, it hasnt been changed)

thanks for the help by the way, im fairly new to this and this isnt even my language (is harder to explain than i tought)

image
forgot to post this SS, the plc is the same but the tags come from another datatype

Load factor is from igntion gateway status page under devices, go to the details of the device. I think it’s renamed to “loading” now

While you're asking the PLC programmer questions, also ask what the System Overhead Timeslice is set for.

Also, how many tags are in this 1/s scan class?

Why are all of your tags yelling? :slight_smile:

hahaha that one is good. i dont like it either, not my choice

in the same data type, plc or the whole system?

load factor is arround 24% checked few times in a couple of hours and was 21-23-26-22

How many tags in the whole system are in the 1000ms scan class.

When you checked were you experiencing the issue with the tags or not?

Are you using Transaction Groups or any type so system.opc* calls?

6337 tags at default scan class

both with and without

no transaction group, and not sure but i think there is any system.opc call. many system tag write/read and query but i doubt it about opc

How many of those are for this SLC?

arround 550-570 within 2 data types

What firmware of the 5/05 is this? Some of the earlier version had issues with 10/100 connections and to fix things you had to specify 10mbps rather then auto. We had issues with that on other SCADA systems.

Also check the message servicing and comms servicing rate limiting bits in the SLC status file:

not really sure how can i check it but the processor type is :
1747-L551 5/05 16kmem OS501

no tag with that address and their value is
S:2/15 = 1 ( in all other plcs too)
S:33/7 = 0 ( same with others)

update:

we couldnt find the System Overhead Timeslice in this cpu, maybe its under another description

The thing is...all six cpu have the same configuration and tags and only this is experiencing this error, could

We were having a similar issue, just a different PLC, it turned out to be an issue with the network, one of the IT switches was having a hard time keeping up with the traffic.

We took Wireshark traces to help troubleshooting.