IN2-H2O
Member
This is more of a statement of findings than a question and I encourage others to add their experiences as well. It took me quite some time to track down and resolve this issue and so I feel it best to post my experience here since I could not find a single reference here on the subject previously. Hopefully others will find this usefull.
We have over 35 HMI's,(Intouch 9.5, XP SP3),on our network talking to 22 different PLC's,(Contrologix L61/62's). I was hunting down intermittent network failures earlier this year where we would lose all communication to a group of HMI's from a single PLC. I could still ping all devices involved, network traffic appeared normal and no red flags were raised in the event viewers. At first we would run around and reboot all the effected HMI's but later realized that they would return all on their own in approx 8 minutes and the effort was wasted. We exhausted our network team for months and they kept putting it back on us as a software issue.
Well, it turns out they were right. I happened to do an unrelated PC upgrade in one of the trouble areas and the issue went away. Upon further investigation, I realized that I had loaded the new machine with DASABCIP V3.5 where it was running V2.5 previously. I reviewed the devices again and discovered two remote touchscreen HMI's were running V2.0 still. I upgraded them to V3.5 and we have not had another incidence since.
It seems that different versions arrange the packets in different sizes which can cause some errors when a PLC is being polled by clients running different versions. I am still not 100% sure what causes it to run for days or weeks without issue and then suddenly lock up but I do know that the system runs flawlessly when all HMI's that talk to a PLC, are running the same version, (Major revision # anyways).
Our entire plant is now running DASABCIP V3.5 and we have not had a single communication fault in several months.
We have over 35 HMI's,(Intouch 9.5, XP SP3),on our network talking to 22 different PLC's,(Contrologix L61/62's). I was hunting down intermittent network failures earlier this year where we would lose all communication to a group of HMI's from a single PLC. I could still ping all devices involved, network traffic appeared normal and no red flags were raised in the event viewers. At first we would run around and reboot all the effected HMI's but later realized that they would return all on their own in approx 8 minutes and the effort was wasted. We exhausted our network team for months and they kept putting it back on us as a software issue.
Well, it turns out they were right. I happened to do an unrelated PC upgrade in one of the trouble areas and the issue went away. Upon further investigation, I realized that I had loaded the new machine with DASABCIP V3.5 where it was running V2.5 previously. I reviewed the devices again and discovered two remote touchscreen HMI's were running V2.0 still. I upgraded them to V3.5 and we have not had another incidence since.
It seems that different versions arrange the packets in different sizes which can cause some errors when a PLC is being polled by clients running different versions. I am still not 100% sure what causes it to run for days or weeks without issue and then suddenly lock up but I do know that the system runs flawlessly when all HMI's that talk to a PLC, are running the same version, (Major revision # anyways).
Our entire plant is now running DASABCIP V3.5 and we have not had a single communication fault in several months.
Last edited:
