Ethernet-APL is ready – now let’s get started. Diagnostics from Ethernet APL networks take on an important role in day-to-day practice. Users who until now have relied exclusively on their multimeters now need to have a rethink.
Ethernet has proven its worth over many years in other branches of industry, yet users in the process industry will have to adapt to new tasks. This is mainly because when a network problem occurs, usually you rely on the IT department or simply restarting the device or system.
Using the IT department is still a valid solution, of course, but you’ll struggle with a restart. Plus who wants to bother the network specialist with every problem? It is much better if users are in a position to solve it themselves.
To bust one myth straightaway, a carefully installed Ethernet-APL network won’t be the cause of any errors during operation, since the technology is very robust if it has been installed correctly.
However, there will be smaller problems from time to time – primarily at the beginning.
Troubleshooting in Ethernet-APL networks is much easier than in traditional installations, but the multimeter is no longer the right instrument to do it with. In the future, these methods – although not fully impossible – won’t be a good option. Consequently, all those involved will have to learn about it.
An IT technician has to deal with process technology; a maintenance technician will have to learn some IT skills. Relevant training and measures regarding qualifications are currently being developed by the fieldbus organisations.
Detecting errors more quickly and intelligently
Users’ lives are made easier thanks to the fact that Ethernet-APL provides additional diagnostics data, providing the opportunity for process optimisation, such as using smarter data analysis by integrating artificial intelligence or expert models for diagnostics.
At the same time, innovative maintenance concepts can be implemented, since the field devices are able to automatically report required maintenance more easily and quickly, among other benefits.
Diagnostics with Ethernet-APL allows for significant time savings. A service technician used to have to switch back and forth between the device and control room three times in some circumstances.
In the field, the device was measured, where necessary modified with need to wait and see whether the diagnosed errors really were the cause for the fault.
Now central diagnostics is possible, since you can look into the device via the web server at the engineering station, which is particularly interesting for hazardous areas.
What errors are there?
A distinction must be made here between IP diagnostics, i.e. errors in communication, and troubleshooting on the physical layer.
Typical IP errors include:
- Address conflicts caused by transposed digits or multiple usage of IP addresses
- Communication errors such as spontaneous terminations or attempts to reinstate communication again and again (retries, etc.)
- But also a broadcast storm, which may lead to a network overload, either because a device is defective or it is actually a network attack by a DoS attack.
As Ethernet-APL is actually a standard Ethernet, standard tools which have been tried and tested over many years are available for IP diagnosis, sometimes which are even free of charge. These can look down to the protocol level. Here, therefore, it may actually be worth contacting the IT department.
Errors on the hardware side
Errors on the hardware side can be detected with physical layer diagnostics:
- Network overload due to too many participants or lines that are too long in the network
- Cable errors caused by wire breakage or short circuit. The shield connection is challenging, i.e. if the cable shield is damaged during connection, as this generally only becomes noticeable after a while in operation.
- Signal damping due to cable aging. It is recommended to observe the quality of the cables to prevent this.
- Interfering signals, such as noises, signal distortion or reflections on conductors, also indicate problems in the network.
- Device errors and failures can also be diagnosed, which can even be used for further recommendations for action, such as condition monitoring or predictive maintenance.
Many devices are already fitted with internal diagnostics, or will be increasingly in the future, which allows a failure to be predicted.
Since diagnostics and asset management data can be transmitted in real-time over large distances, new concepts such as NAMUR Open Architecture (NOA) or the Open Process Automation Standard (O-PASTM) are possible.
Ethernet-APL is highly robust as regards network load. In order to control systems and processes in the process industry safely, the network load should ideally be below 10 percent.
Looking at the star topology shows that for process-typical cycle times of 400 to 500 ms, the load is below five percent even for 250 field devices. Sufficient reserves are therefore available.
For trunk/spur topologies, which have been designed for max. 50 field devices in practice, the network load is even below one per cent, assuming the same conditions. Nonetheless, we will examine what can happen in extreme conditions and how you should react.
Diagnostics for physical layers
Up to now, the NAMUR Recommendation NE 123 (Service and maintenance of the physical layer of fieldbuses) has been consulted for physical layer diagnostics. This recommends consulting the four parameters of noise, signal level, jitter and unbalances for troubleshooting and evaluating the fieldbus.
If you transfer this now to Ethernet applications, you come up against several limits. Looking at the noise pattern of a fieldbus cable with 31.25 kbps and a 2-wire Ethernet cable (full duplex communication) shows the difference.
Only experts can detect an error on an Ethernet-APL cable from the noise. This form of troubleshooting is less suitable for practical application.
However, there are simpler options: The advantage of Ethernet-APL is that it is a point-to-point connection. The power switch is therefore a port connected to exactly one field switch or the field switch is connected to a port with one field device.
Furthermore, the diagnostics in Ethernet networks has been established for a long time and is available as standard. This makes troubleshooting for Ethernet-APL significantly easier.
Field switches become diagnostics centres
As the field switches sit centrally in the network, they in fact are party to everything happening in the network. As completely normal participants in the network, diagnostics information is transmitted via PROFINET as standard, for example. Therefore, an error occurring is noticed immediately, which is particularly useful for commissioning.
The service technician can access much more information at their convenience on a small display on the field switch from R. STAHL. What’s more, it is possible to look into the network and the physical layer in a deeper and more detailed way via a web server. This is where information can be found about voltage, voltage level, current or redundancy problems, and more.
LEDs indicate the cause of the problem. The state of the relevant device functions on the trunk or spur is shown on a diagnostics dashboard in the form of a traffic light. A blue LED, in line with NE 107 (Self-Monitoring and Diagnosis of Field Devices), signifies the need for maintenance, for example.
The expected lifetime of a field switch is calculated using a continuous check of operating and ambient conditions, such as temperature change or switch on/switch off processes, allowing maintenance information to be provided to the operator in good time.
For example, this can be used to recommend replacing a device to prevent failures which lead to unplanned shut downs.
Ethernet opens up new possibilities
Furthermore, two new diagnostics parameters are particularly interesting for users:
- The SNR (signal-to-noise ratio) assesses the technical quality of the signal and informs the user about the the network quality, likewise in the form of a traffic light.
- The TDR (time-domain reflectometry), a type of cable radar, also provides support. Here, the port looks into the cable and checks the run length and reflection of communication rates. It can do so to determine whether there are interruptions or interference, such as a short circuit, in the conductor. This process is highly efficient; the TDR especially can locate such a line fault to within 0.3 metres accuracy.
New diagnostics concepts in progress
Many manufacturers are currently working on diagnostics concepts for their devices. For example, R. STAHL is planning to provide diagnostics data for its field switches via OPC UA in a cloud or edge gateways.
This is also interesting with regard to the NOA concept. If the PA-DIM model supplies the semantics for it, very capable diagnostics is available.
In conclusion, Ethernet-APL generates detailed diagnostics data in a very straightforward way, opening up a diverse range of options for detecting errors more quickly and improving processes.
The challenge will be to ensure that the data is also used appropriately and does not disappear into a cloud. If, however, you use the new tools, there’s nothing preventing untroubled operation with Ethernet-APL.
STAHL South Africa
Tel. +27 11 608 3120