Monitor wear and detect damage at an early stage12.09.2022
Permanently monitor data lines in production
Not only motors and pumps are subject to wear and tear in production plants, but also the data lines of the machine and plant networks including the cables and plugs. Constant alternating bending stresses, shocks, oxidation and corrosion allow aging and wear to run unobstructed. But while the mechanical wear and tear can be perceived analogously with the sensory organs, the wear and tear of a data line only becomes noticeable in the extreme case: failure. To counteract this, intelligent managed switches are required, with which the physical condition of the data line is digitized and thus visible.
The question of availability and the knowledge of the plants and their condition in the company directly affects production and productivity. With maintenance approaches such as predictive maintenance, premature wear can be detected and unexpected downtimes and the associated costs can be reduced. The awareness that a mechanical element like an engine can wear out is commonplace, but it is missing when it comes to the network: Because the network infrastructure with data lines and plug connections is also exposed to the environmental factors in production. Oxidation and corrosion on the plug connections are usually the results. Added to that are mechanical loads such as bending stress or torsion, for example on travel routes or in robot technology. Lines and structural components are therefore subject to wear and aging - a gradual process that is not immediately apparent.
The current VORTEX report (evaluation of service calls on machines and system networks) clearly shows that too little attention is paid to data lines during the operation of a system. Until commissioning, the cables are considered as a component and complex test reports and certification evidence are created according to CAT. However, all tests only relate to the cable, not the interaction of cable and socket. In addition, these tests are "offline" and do not say anything about the line quality in function of the system. Not to mention the degree of wear and tear or the interference reserve.
Conversely, the many data lines in machines and systems are to be regarded as components subject to wear and classified as such in the maintenance and repair concept. The material costs are negligible, but the consequences of a failure can hardly be estimated. The data lines thus represent an asset that needs to be protected and preserved through appropriate condition monitoring.
Predictive maintenance with managed switches
Failure-oriented maintenance and the replacement of defective parts does not go far enough here - the damage has occurred. Since the fact is known as such, the lines are subject to a life-dependent exchange at certain intervals in the sense of preventive maintenance. This avoids failures and seems to be becoming the standard. In view of the constantly declining resources of material and personnel, this strategy is no longer sustainable and very expensive. And the question remains unanswered as to whether the lines should have been replaced at all or whether an earlier point in time would have been even better. In terms of predictive maintenance, a key to maintaining the network is therefore the status-oriented exchange of the data line based on available data in the form of quality parameters..
This is possible with intelligent managed switches. They are the central point in systems with a high level of automation and Ethernet-based real-time communication. Managed switches have basic intelligence: they control where and on which port the information is to be sent. They recognize the neighbours, monitor the working speed at the ports, check utilization and availability and check whether discards (rejected telegrams) occur. In addition to this information on data transport, a managed switch with intelligent integrated sensors can also digitally record information on environmental variables (e.g. temperature), the line, cable quality and EMC load (shield currents) and thus important information as a usable added value for maintenance or the central maintenance tool deliver. This shows the status of the network in its entirety and allows a binding statement about security, reliability and stability at any time.
Managed switches make the network status visible
In addition to wear and tear from the outside, there are also events inside the switches that are sporadic, uncontrollable, and non-reproducible but affect the network and disrupt communication. Acyclic data communication events can lead to load peaks and thus to overflow (discards) of the internal memory, which nobody notices without monitoring. Although this does not always have a direct effect on the functioning of the machines, the events are harbingers of tomorrow. There is a reason for overloaded switch ports: the performance class of the switch may not correspond to the required bandwidth. Companies then have to take action, replace switches or schedule or limit the application. However, these events cannot be investigated under normal circumstances because no diagnostics are available. It reacts with a switching off and on, but the cause of the error remains in the dark. A central monitoring software, with the help of which the status of each individual switch port is permanently read out (e.g. via SNMP protocol), analyzed, graphically prepared and displayed on the web, enables permanent monitoring of the communication quality and network status. Changes and abnormalities are immediately visible and the simple display in the typical traffic light colors points the maintenance technician to the possible weak point in a targeted manner.
Another factor is electromagnetic compatibility (EMC): Shields protect data lines from magnetic influences, but they are repeatedly subjected to potential equalization currents that can get into the devices. They lead to disruptions and irregularities in data traffic and can damage or destroy the electronic devices in the machine's control system. It is important to identify these sources of interference and to recognize them at an early stage in order to take countermeasures (improvement / optimization of the equipotential bonding). A sophisticated measuring circuit can monitor the current level of shielding current and save the values and the frequency response if it is exceeded. A snapshot is created internally for the event (e.g. voltage peak) in the internal event memory. In this way, events are recorded and are permanently available for analysis, the causes and connections for EMC faults become comprehensible. A warning is issued if the leakage current exceeds predefined threshold values.
In addition, the alarm management of the switches warns of exceeding the device temperature - and thus the temperature in the control cabinet - and monitors the constancy of the 24V supply voltage. Last but not least, the mentioned functionalities of an intelligent managed switch make a significant contribution to security, because the first step in network security begins with knowing the status of the network.
The benefits of Network Monitoring
The network monitoring and diagnostic functions of the managed switches allow a complete overview of systems and network sections. Sudden, non-reproducible events can be tracked and the cause determined: communication events are saved, graphically visualized in bar charts and their quality value is shown. A target/actual value can be superimposed for comparison. The performance of each line can be displayed as follows: An integrated web interface allows full access to the monitoring and diagnostic data without additional software. If an anomaly is noticed, initial data is already available - without manual measurement effort. The data from the network can be integrated into the existing maintenance tools or analyzed using powerful software. This gives organizations a tool to extend their maintenance approaches to the network.
Managed switches are not only used for data transmission in production facilities. They can also provide network monitoring and diagnostics. This goes beyond making unexpected events visible or analyzing shield currents: The mechanical wear and tear of the data lines can also be mapped with managed switches. They thus offer a comprehensive overview of the status of the network and are a key component for comprehensive predictive maintenance.
Author: Karl-Heinz Richter, Managing Director of Indu-Sol GmbH