June 2021

18 | Plant & Works Engineering>> www.pwemag.co.uk June 2021 Maintenance Matters Focus on: Condition Monitoring on the tasks and degree of manipulation of parts required for each task. Different robots can therefore be subject to widely different stresses, but the most highly stressed units may be only a small number of the total number of robots in a production line. These tend to be “bottleneck” robots dealing with, for example, the processing of complex parts or manipulation of components through many different orientations. A robot cannot be designed for the worst- case scenario, as it simply would not be price competitive. It is therefore necessary to instead determine which factors reduce the performance and operational life of the robot to identify which ones are under the most stress. Measuring wear and tear Some of the key parts of a robot are the gear boxes that play a major role in the movement of the various axes. There are several issues that can reduce the life span of a compact gearbox, one of them being the duty factor. This is a very important measure, as it reflects the overall activity of the robot. The robot’s duty factor is calculated as the time that at least one axis is moving divided by the total production time. There is also an axis duty factor, which is calculated in the same way but on an axis-by-axis basis. The effect of duty factor varies depending on the robot application. For example, arc welding has a high duty factor, but requires low speed and low acceleration, producing low wear on the gear box. Spot welding has a low duty factor, but with medium speed and a high to very high acceleration, wear is correspondingly high. Inter press loading has an extreme duty factor and very high speed and acceleration, resulting in extreme wear. Previously, it has been difficult for users to determine whether key parts such as gearboxes are becoming worn or in need of replacement. The result has been that problems either go unnoticed until the robot fails or the company carries unnecessary stocks of parts or is unable to source the right parts when they are needed. This leads to disrupted production while the robot is offline. A data-based approach to maintenance This challenge is met through condition monitoring, a process which constantly gathers data on the operation and performance of a device to assess its maintenance needs. Designed for customers with large fleets of robots, ABB has launched a new Condition- Based Maintenance service, enabling the creation of a customized preventive maintenance schedule. This can be for either individual robots or fleets of robots, as used in a car plant. Based on real-time operational data, it allows automotive robot users to optimize productivity and minimize downtime. Condition-based maintenance works by gathering motion data on the operation of each robot to help identify any potential issues that could affect performance. These include duty factor, speed, torque, and gearbox wear. The data doesn’t exist in isolation – instead it is compared against a 14-year-old database of other ABB robots across the globe to calculate when or if a particular robot is likely to develop a potential fault or failure. Monitoring also minimizes the likelihood of premature failure and extends the Mean Time Between Failure (MTBF) rate, as well as prolonging the operational life of the robot. Insights maximise performance Based on this information, robot users gain the insights they need to create a preventive maintenance schedule to help keep robots in good working order and to help maximize performance. The service can advise whether any remedial action is required, either a repair or a replacement of any affected parts. This allows spare parts to be purchased at the right time, ensuring they are available but not taking up capital and space by being held in stock for extended periods. As well as better budget management, this ensures the car plant has the correct resources on hand and can plan in advance when the repair needs to be performed, reducing the risk of unplanned downtime. Deciding on exactly which preventive measures to take is achieved through a report provided for each robot. This includes a color- coded summary table that clearly indicates the status of each axis and the overall health of the robot indicated by the calculated joint usage score. Also included is data analysis, individual maintenance recommendations, conclusions, and rating of the system. Two levels of analysis are provided. Level 1 gives a factual overview of the customer’s installed base and identifies the most stressed robots in the plant. This gives the customer a detailed overview of how robots are used in their plant. The Level 2 analysis further investigates the robots selected at Level 1, giving the customer a detailed knowledge of their most “stressed” robots. This prevents the risk of unexpected gearbox breakdown in production and gives recommendations about how to deal with these “stressed” robots. Using this data, the customer can then design an appropriate maintenance schedule - ABB can also help in this if needed. The Level 2 report also helps define the budget for spare parts and supports the plant’s strategy to upgrade its robot fleet. By making maintenance and repair more predictable, condition-based maintenance helps automotive plants eliminate unexpected robot downtime caused by failures or delays in obtaining spare parts. Through identifying which robots are over-utilized compared to others in a production line, users can also better understand exactly which robots may have an increased risk of component failure. Keeping on the road to improvement With drivers demanding ever more bespoke models and options, ensuring that robots continue to work for the maximum amount of time is key to maintaining the extreme flexibility of the modern car plant – condition-based maintenance is a vital tool in ensuring that car manufacturers stay competitive in a rapidly changing industry.