Process Control Networks

D

Thread Starter

DanD

What are the differences between Process Control networks and system management and the similar functions within a Corporate IT installation?
 
A typical process control system needs values updated several times a second, with each data item being relatively small. It is better to repeat the last value than to report an obviously spurious item. If a data item is missed, it is no big deal. Messages must be received within tight time limits after being sent, so delays cannot be tolerated. Process control network traffic consists of very many small messages which are time-critical. If a process control network goes down, the plant generally goes down with it.

Generic management or corporate traffic can consist of very large messages (eg historical logs, reports, etc) and these are not time-critical - as long as the boss gets his printout within the time it takes for the office boy to get to the printer room and back, there is no problem. A management network can go down without affecting the operation of the plant.

The two applications have very different requirements and cannot be mixed on a single network.
 
I agree that a typical process control network has both time critical and non-time critical requirements. However, we always knew that and most process control network protocols can handle both types of transmissions quite well. For example, Foundation Fieldbus both H1 and HSE handle time-critical control communications with an arbitrator, and allow non-time-critical transmissions with a publish subscribe mechanism.

EtherNet/IP, PROFINET, EtherCAT, and Ethernet/PowerLink all allocate bandwidth reservation for time-critical traffic, and allow the remainder of the time for non-time-critical communications - all within the same network.

> The two applications have very different requirements and cannot be mixed on a single network.
 
Top