Field buses
Continuously monitor Profibus installation
Fieldbuses do not necessarily remain in perfect condition after installation. They can also 'age', so that production downtimes can occur due to bus malfunctions. This situation can be avoided by continuously monitoring the networks.
At almost 150 years old, DS Smith's Aschaffenburg paper mill is one of the oldest in Europe. It mainly produces corrugated base paper with a low grammage - and has been doing so from 100% waste paper since 1990. The machine used here produces up to 1400 meters of paper per minute with a width of 7.53 meters. As a result, the Aschaffenburg paper mill can deliver 400,000 tons of paper per year.
This performance is made possible by a high and consistent level of automation in the regularly modernized mill, which, in addition to the paper machine itself, includes the waste paper processing areas, biological treatment plant, water treatment plant, waste incineration plant and the gas and steam turbine power plant. These areas are combined under a central process control system. One key to this is the continuous exchange of data, which is handled via Profibus.
Even if this fieldbus stands for robust digital data transmission, it must still be taken into account with regard to day-to-day operation that, regardless of a flawless installation, the operating reserves for fieldbus communication can progressively decrease over time, ultimately leading to serious communication faults. This can be caused, for example, by EMC influences, defective connectors or a faulty bus termination, such as those caused by severe oxidation. Sudden production downtimes and the associated financial losses are the result.
The Aschaffenburg paper mill was also affected by this problem. However, DS Smith did not want to accept such failures and decided back in 2009 to improve the situation in the long term by continuously monitoring the status of the Profibus networks. The first step was to define the requirements for this. The solution had to be able to be used without impacting on the paper mill's existing systems and not cause any additional work for employees.
With this in mind, DS Smith intensively tested various devices for continuous Profibus monitoring over several weeks and ultimately opted for the 'TH Link Profibus' solution from Softing. This is installed stationary in a control cabinet and offers controller-independent access to the corresponding fieldbus networks without the need for special configuration tools. Even during operation, the components can be integrated into a system without any feedback. In particular, there is no need to change the bus addresses or adapt the control program.
The fieldbus monitoring module can be integrated into existing systems without feedback and requires little additional space for installation in the control cabinet.
© Softing Industrial AutomationOne of the main advantages is the permanent network diagnostics via passive monitoring of Profibus telegrams. The cycle time is measured and critical events are detected - including the number of telegram and message repetitions, restarts of fieldbus devices, device errors and diagnostic messages. This data allows early conclusions to be drawn about the current status of the bus communication and impending errors while the network itself is still working perfectly. Individually definable limit values - such as the upper limit for permitted telegram repetitions - can also be used to quickly derive a summarized 'good/bad' statement about the operating status of the system from this data. All individual data can be retrieved via an integrated web server connected to a central process control system - eliminating the need for on-site checks. Based on this data, the solution described also offers intelligent troubleshooting assistance and statistics for optimizing the network configuration. Finally, a particular advantage is the automatic sending of an e-mail message as soon as the observed overall communication status changes from 'good' to 'bad'. Wolfgang Groß, who as an employee in the electrical, instrumentation and automation technology department at the Aschaffenburg paper mill is responsible for the use of the solution described, comments: "For me, the e-mail notification of an impending fault is an important function. This means that I always have a precise overview of the current communication behavior in all systems, even in the office environment, and don't have to determine the current status of the individual Profibus networks on site."
A key criterion for DS Smith's decision to use TH Link Profibus was that this device is not an active participant in the network that changes the communication behavior, but that it only passively listens to the telegrams sent. The first devices were installed immediately after the test in September 2009. As early as November, 15 more devices were ordered, so that all Profibus networks of the paper machine and the waste paper processing and biological wastewater treatment plant areas could soon be continuously monitored. The result: "Thanks to the continuous monitoring of our fieldbus network, we were able to reduce the downtime of the paper machine by half because we were alerted to weak points at an early stage and were able to rectify them in good time before the system came to a standstill," recalls Joachim Gutjahr, the head of the electrical, I&C and automation technology department at the Aschaffenburg paper mill. "This enabled DS Smith to reduce losses by a high five-figure sum in euros."
The possible error patterns
Based on the possibilities of the current Profibus standard, 'TH Link Profibus' recognizes error patterns of a decreasing operating reserve. These can be
The time stamp in the diagnostics list allows conclusions to be drawn about the reasons for a Profibus slave failure.
© Softing Industrial AutomationShort and sporadic device failures
Profibus uses a sequence of request and response messages for data exchange. If the Profibus master does not receive a response to a data request, it repeats sending the request message to the slave a configurable number of times. If these telegram repetitions also remain unanswered, a diagnostic request message is sent in each subsequent cycle until the slave responds. The master recognizes from the message that the station is not ready for operation that the station has failed and then sends the device parameters and the configuration information. The slave responds to the subsequent diagnostic request message that the device is ready for operation.
This entire communication is logged - with time stamps - in the TH Link Profibus and displayed in the diagnostics list. The time between the failure to respond to a data request and the successful response from the device can be used to determine the cause of the error. As a Profibus device usually takes between three and ten seconds to restart after a failure, a duration of less than one second indicates that the telegrams were not exchanged correctly, which often indicates a loose contact or an electrical or electromagnetic fault. If the failure is sporadic or if only one bus subscriber is affected by the failure, a loose contact can be deduced as the cause of the fault.
At an optimally set baud rate, the majority of the bandwidth is used for data exchange.
© Softing Industrial AutomationNon-optimal allocation of bus parameters
The exchange of messages in the Profibus network is characterized by two time variables: On the one hand, the cycle time determines the time interval between the sending of individual request messages to the various slaves; on the other hand, the baud rate determines the speed at which data is exchanged in the network. For communication to be as robust as possible, both variables must be in a suitable ratio to each other. It is not ideal to exchange data at the highest possible baud rate, as this makes it more prone to errors. Instead, it is better to set the baud rate so that around 70 to 80 % of the cycle time is required for the net data exchange. With this in mind, it makes sense to check whether the (pre)set baud rate is appropriate for the data volume in the Profibus network.
During continuous monitoring of the Profibus network, the telegram repetitions to the individual slaves are logged.
© Softing Industrial AutomationMissing termination and electrical or electromagnetic interference
A sharp increase in repeat telegrams within a few minutes indicates reflections in the network. If the repeat telegrams are limited to one or a few subscribers, it is highly likely that a missing bus termination is the cause of the communication fault. If, on the other hand, the repeat telegrams are evenly distributed across several or all subscribers, there is probably an electrical or electromagnetic fault. In this case, the number of repeat telegrams is generally not so high.
When using diagnostic repeaters, TH Link Profibus supports the user with a plain text display. This contains the diagnostic information sent by the repeater to the control unit on detected errors in the bus physics, such as reflections or short circuits. All possible texts of the GSD file are stored in the link for this purpose. In the event of an error, these are displayed together with dynamic information - for example the distance of the detected error.
TH Link Profibus displays diagnostic messages in plain text together with dynamic information.
© Softing Industrial AutomationDeterioration of bus communication
Two threshold values for the number of telegram repetitions can be defined in TH Link Profibus. One monitors their number within a period of 15 minutes, the second within a period of 24 hours. If these threshold values are exceeded, a corresponding alarm is displayed. Exceeding the 15-minute threshold may indicate damage to the line, while exceeding the 24-hour threshold is an indication of a general deterioration in bus communication, for example due to corrosion.
Authors:
Dr. Christopher Anhalt is Senior Product Manager Diagnostics at Softing;
Georg Suess works in Operational Marketing at Softing.


















