Profinet diagnostics: organisational issues

Profinet is an established communication standard in industrial automation and the general development and installation of Profinet networks does not present a problem. However, if a diagnosis of the network is needed many users venture into new territory. Plant operators and maintenance need clear best-practice guidance on how to monitor Profinet networks during operation, how to reliably keep them up and running, and how to react quickly and efficiently if problems occur. A network acceptance test, for example, is often not included in the plant commissioning procedure. 

The reason for this lies in the profound changes brought about by the shift from traditional fieldbus systems to Profinet, or to TCP/IP-based Industrial Ethernet systems. 

It is necessary to clarify what is meant by the terms ‘diagnosis’ and ‘network diagnostics’. In this context, Softing Industrial Automation uses a comprehensive approach that not only covers fault localisation and correction in the case of failure, but also includes general action and measures ensuring proper operation of a Profinet network throughout the life cycle of a plant – from installation to commissioning and operation. 

TCP/IP is an open, widespread standard protocol. Users are looking to profit from this openness and the possibilities it offers. As a result, the conditions on Profinet networks frequently change in the field. For example, users regularly modify the network configuration and exchange the nodes connected to the network. The use of a TCP/IP protocol, therefore, also means that the boundaries between industrial automation and enterprise IT are becoming more blurred and permeable.

Even if a network can be kept stable, the configuration of the network and nodes is still complex. Regardless of whether faults in the network occur right at the beginning, during commissioning, or later on - after the replacement of a defective device during operation of the plant - they are often caused by configuration errors. 

Faults resulting from physical causes, in comparison, are far less frequent than in Profibus networks, for example. The lower protocol levels of Profinet communication are complex. Depending on the cause of the fault and the necessary analysis to be performed by the user, it may take in-depth knowledge of communication technology or IT to identify and correct the fault.

On the one hand, users of Profinet technology are faced with the technological challenge of adapting network diagnostics to the new conditions brought about by TCP/IP-based communication. On the other hand, users need to address organisational issues. These reach from routine checks of open networks that are basically running stably but are subject to changes, through to the complex technical troubleshooting of a persistent fault. The first question that needs to be answered is how can the wide range of diagnostic tasks best be fulfilled and who takes care of what tasks?

The following three example questions illustrate the organisational challenges facing PROFINET technology users:

1. Who do the maintenance staff call if they are unable to locate and correct a fault with their standard tools? A specialist from their own department? The in-house IT department? An external service provider? Are there any clear criteria by which the maintenance staff decide at which point external experts are to be called in?

2. How can users obtain a transparent, reliable acceptance test for a Profinet network as part of plant commissioning? Is acceptance testing performed by the plant vendor? Is it performed by the in-house planning department or is an independent third party called in?

3. To what degree does the planning department take the responsibilities, processes and tool selection for network diagnostics into account at an early stage? If this is not the planning department's job, then whose is it?

It is striking how differently these organisational questions are answered by Profinet technology users and how widely the assignment of responsibilities and tasks to different roles varies. Even within a single company, there can be great differences between individual production sites.

It is largely undisputed that maintenance plays a key role. The maintenance department is responsible for first-level support for the entire plant, which is now based on complex Profinet technology. Maintenance staff need processes and suitable tools that enable them to efficiently work on Profinet networks without having to ‘fly blind’. At the same time, it is neither necessary nor practical to have every member of the maintenance staff become an IT and communications expert able to detect and identify such difficult errors as an internal error in the device vendor's protocol stack.

Against this background there are a number of basic requirements that need to be met by network diagnostic tools. There is no ultimate all-in-one tool that can automatically perform all diagnostic tasks for any user in the best possible way. Instead, what can be used is a set of individual tools that each fulfill a specific diagnostic task in reference to the assigned responsibility. The other way round, users should select their tools with a view to the precise role and tasks they have been assigned.

Two working group initiatives in standards bodies are worth mentioning. In the first initiative, a Profbus + Profinet International (PI) working group, headed by Prof. Dr.-Ing. Karl-Heinz Niemann, revised and extended both the Profinet Design Guideline and the Profinet Commissioning Guideline (Profinet Design Guideline Version 1.14, Profinet Commissioning Guideline Version 1.36, both released in December 2014). In the second, the GMA Technical Division 6.15 (VDI/VDE), headed by Prof. Dr.-Ing. Jürgen Jasperneite, is currently preparing a guideline entitled ‘Reliable Operation of Ethernet-Based Bus Systems in Industrial Automation’. These documents provide many detailed descriptions accompanied by concrete recommendations, in particular also for Profinet network diagnostics.