1) Introduction
With the issue of redundancy, the important issues are: reading synchronizations, switching time between controllers, process status continuity, type of solution, cost of implementation and software changes needed for the redundant operation. These subjects can all be addressed through supervision simply by varying the focus.
Referring to the developed system, two types of redundancy are available for Siemens equipment. The difference between the two options relates to equipment and software that can be better analyzed by contacting the German supplier directly. For the purposes of this document, we need to look more closely at what was implemented.
Types of redundancy
As mentioned, there are basically two types of redundant Siemens systems: "Redundância por Software" (Software Redundancy) and the "Hot-StandBy".
Software redundancy consists basically of two CPUs running with the same programming process augmented by software that synchronizes the various CPU variables.
Siemens Hot-StandBy systems are more complex than this. Normally, as well as utilizing software, they also include equipment for the implementation of redundancy. However, the costs for this architecture can be prohibitive for some projects.
Further details on this architecture, equipment and costs can be obtained from Siemens.
2) Solution
This article is concerned with solutions for software redundancy. The equipment and networks required, as well as the existing E3 stations on the project, are described in figure 2.
This figure makes it possible to see that all nodes are linked through a Profibus network. This is possible because the S7-315 2DP CPU has a Profibus communication port available, in addition to an MPI port used by the CPUs redundancy software to synchronize the variables defined as redundant.
At each E3 Server station, a CP5611 board with the Simatic Net software is available, which can be licensed to the Profibus DB network. Refer to the "Configuração de redes Siemens" (Setting Up Siemens networks) document available at the Elipse website for tips on how to setup this network. The two stations with individual network addresses will work separately with no communication between them.
E3 Viewers, which are the system operation points, can be linked to any of the E3 Servers via Ethernet networks.
Figure 1: Redundant Architecture with the Siemens and the E3
3) Applications developed
An E3 application has been developed to detect which CPU is active. An internal tag and two tag events were used to detect the CPU return, one for each. There is no priority in operation and reading in E3, i.e., when a station is working as a master, even if the other one enters in operation (which shouldn't occur since there are guarantees that only one of them will be selected as the active server). In order for the E3 to always be on air, the two events tag makes it possible to work without problems with any variation in the CPUs. In these situations, it is possible to direct each tag to read from one or the other network CPU. This has a number of advantages:
Maintains the number of points to be read, which avoids duplicating the PLC reading and writing addresses, as the same tag can read from any network controller.
Engineering Time: Creating associations once between tags and points of use either on screens or in histories and formulas diminishes programming risks by reducing development and testing time.
Through modifying only the tags parameters, the connection switching time between CPUs is minimized. In addition, all required actions are managed in a single script.
In cases where E3 communication with the CPUs fails because of machine maintenance, or a change of controls among the PLCs occurs, an internal policy in the PCL program was developed which allows the writing of the value zero within the status of the two CPUs, remembering that only one of them has value one status (called master status). From this point forward, the process continues based on CPU status detection events.
All applications for either PLC or E3 are available for download on the Elipse Software website. In order to reproduce the previous example, the following equipment and software are required:
4) Conclusion
Redundant systems implementation, either in automation or supervision, has become more common today in project specifications. Offering tools and means for implementing this feature in new or existing projects is an important differential for the products which serve these functions.