RS5000 multiple MSG message polling thru Ethernet

[Jeff23spl]

Hello everyone

Usually i poll each comm message 1 by 1 but in a particular project, i would like to have 2 different refreshing speed

Does it matter if more than 1 message get initiated at the same time on ethernet media ?

 

[daba]

Simply - NO

More exactly, there is a limit, but in your case, 2 messages, fine.

 

[Jeff23spl]

if i have 15 messages is it still fine ?

if i poll each of them each scan it may become a problem ? I sometimes get 0 as return values but isn't timed or i can't find any relationship with the sequence programming for these 0...

There probably be an easy way to make sure the is no collision or overloading ? How do you do that when there is a lot of comminucation message programmed ?

 

[daba]

if i have 15 messages is it still fine ?

if i poll each of them each scan it may become a problem ? I sometimes get 0 as return values but isn't timed or i can't find any relationship with the sequence programming for these 0...

There probably be an easy way to make sure the is no collision or overloading ? How do you do that when there is a lot of comminucation message programmed ?

15 messages should be OK, providing they are not all firing off simultaneously and rapidly repeating themselves.

YES : Do not re-initiate MSG instructions immediately they are complete. Instead, use free-running timers to initiate them at a much more relaxed rate, depending on the application. And I would also schedule the messages to fire off at a rate appropriate to the data they are transferring.

Example1 : If I am reading Analog data acquired in another processor, I would not try to read it faster than it is changing - look at the Real-Time Sample and Requested Packet Interval rates of the remote processor to determine their update rates.

Example2 : Production Report data, providing it is time-stamped within the data itself, probably could be read once a minute, or even longer, or be triggered by "Production Complete" instead of timers.

If you do require any "high-speed" data transfers, then set the MSG configuration to enable Connection Caching. This will speed-up the completion of the message, at the expense of a permanently used "connection".

Make the messages "large" - use UDTs to group data into a single tag, and poll the single tag rather than use multiple messages.

If the remote processor is Logix5000, consider using Produced/Consumed tags instead of messaging. Again, pack the data into "large" UDT tags (maximum 500 bytes of data for a produced/consumed tag).

A processor loaded with messages would need a greater "System Overhead Time-Slice. The default is 20%, but you would almost certainly need to increase this %age, at the expense of a slower user-code scan-time.

In Rev. 16 or greater, there is the option to recover unused time-slice and allow the processor to resume scanning the user-code - I would always check that option in any project anyway. I have seen this value at >80% on a project, but the option was enabled.

 

[Jeff23spl]

thanks
the clogix is talking to 5 micro1400 via 3 MSG at each