The global site of the UK's leading magazine for automation, motion engineering and power transmission
29 March, 2024

LinkedIn
Twitter
Twitter link

Condition monitoring: a powerful tool for checking drive health

17 September, 2021
Monitoring the condition of motors and drives can help to maximise their uptime and reduce operating costs.

Condition monitoring of motors and drives need not be an expensive way of producing complex data. Rohan Beck, Siemens’ product manager for simple motion and drivetrain digitalisation, explains to Drives & Controls that it can offer valuable insights into machine operation at a fraction of what it cost previously. 

For many years, a common maintenance practice for drives and motors has been “run to failure”. Assets are allowed to continue operating until a breakdown occurs, after which reactive maintenance is performed. The operation in which the asset is being used then has a period of unscheduled downtime while the reactive maintenance is being performed. It then returns to normal operation.

However, run-to-failure does not maximise the life of the assets, and the reactive maintenance often requires replacement parts. If suitable spares are not held in stock, an emergency replacement will be needed when a drive or motor fails, leading to additional downtime, especially if there are stock availability or logistics issues.

The most common alternative to run-to-failure and reactive maintenance is preventative maintenance, where pre-planned downtime is scheduled in advance and used to perform maintenance activities on assets, with the aim of lengthening the assets’ service lives and preventing breakdowns and unplanned downtime.

The major drawback of this approach is the difficulty of setting appropriate maintenance intervals. It is easy to over-maintain a system, leading to excessive expenditure and use of resources. On the other hand, if the scheduled maintenance is not frequent enough, then preventative maintenance becomes similar to a run-to-failure approach, with all of its downsides.

Even if the maintenance schedule is appropriate, the stochastic variation in methods and timing of failures may lead to breakdowns and unpredicted downtime, regardless of the preventative maintenance schedule. Much as we may want them to, breakdowns do not follow our schedules!

A third approach to maintenance, is predictive maintenance, which aims to limit maintenance interventions – and associated downtime – to times when it’s necessary to prevent an upcoming failure. When implemented correctly, this approach gives the benefits of preventative maintenance – such as minimising unplanned downtime, and maximising the service life of assets – while removing the possibility of over-maintaining, and thus maximising the uptime of the asset. It can be considered as a lean / just-in-time approach to maintenance.

Condition monitoring is a key component of predictive maintenance systems. To schedule maintenance interventions when they are needed, it is necessary to monitor an asset’s “health”, in order to predict when a breakdown is likely, and preferably also to receive some indication of how to correct or prevent the issue. This is the “predictive” part of predictive maintenance.

Moving parts

Predictive maintenance is suitable for any asset that needs to be maintained, so is ideal for drives and motors, and for any machine with moving parts. It involves installing some form of condition monitoring that can give information about the state of the machine. The monitoring system then records the “healthy” state of the asset, when everything is operating as it should. The operating asset is then monitored continuously, comparing its operating parameters with those recorded as the “healthy” state. Any deviation from the “healthy” state – whether abrupt, or gradually over time – indicates the need for a maintenance intervention, which can then be scheduled.

On large sites, companies can set up condition monitoring on every piece of equipment and view the health of an entire fleet, but there can also be great benefit from installing condition monitoring on one key asset, upon whose health a process depends and whose failure may cause a catastrophic bottleneck in production. As with most things related to condition monitoring and predictive maintenance, it depends on the details of the application.

Condition monitoring of machinery has traditionally relied on vibration analysis, because wear and imbalances on bearings, shafts, rotors, gears, and other moving parts, causes unusual variations in the vibration patterns in an operating machine. Monitoring machine vibrations and recording and analysing the patterns, can therefore help to identify defects and possible failures.

Vibration analysis and other forms of dynamic monitoring are still common methods of condition monitoring, but they are far from being the only option. Another traditional type of condition monitoring is performance monitoring. This uses observation and performance trending, and derives an indication of a machine’s health from its output and existing manufacturing performance measurements. Similarly, an indication of health can be determined by monitoring an asset’s electrical performance using power signature analysis, surge testing, motor circuit analysis, and similar tools.

External approaches to condition monitoring include thermography (thermal imaging), ultrasonic monitoring, acoustic analysis, radiography, oil analysis and tribology, laser interferometry, and electromagnetic measurement.

Every condition-monitoring application is unique to the asset being monitored, and the needs of the organisation using it. Depending on the application, one or more of these types of analysis could be used.

Monitoring vibration levels in motors can help to identify defects and avoid possible failures.

Smart drives

When it comes to monitoring rotating assets, modern drives are incredibly intelligent devices. For motors controlled by drives, significant amounts of useful data are already available to the maintenance engineer, who often needs only a way to access it and monitor changes over time. Drive parameters such as output voltage, DC link voltage, power, torque and rotor speed, if made accessible and easy to monitor, can all give an indication of the health of the machine, and monitoring how they’re trending can indicate potential impending failures.

For example, Siemens’ Sinamics Connect 300 is a device that can collect parameter data (including power, current, speed, torque, motor temperature, converter temperature, converter state, and fault codes) from up to eight Sinamics drives, and upload it directly to Siemens’ MindSphere Industrial Internet of Things (IIoT) platform. It is then presented as time series data, with the ability to set up text or email alerts for when a value goes out of pre-defined bounds.

Further analysis can be performed on the data using various applications available on the MindSphere platform. This has the added bonus of being accessible from anywhere in the world as it’s an IIoT platform. But this is only one way of making use of the data already available in drives. There are various other techniques for performing electrical monitoring of drive systems using parameter values from the drives themselves.

For a more in-depth analysis of the condition of a motor, vibration analysis may be appropriate. Another MindSphere technology is the Simotics Connect 400 – a wireless device that attaches to the housing of an induction motor. It uses onboard accelerometers and other sensors to monitor the health of the motor via its speed, vibration, and temperature, and communicates these directly to the MindSphere platform.

While these and similar devices are easy to install and retrofit onto existing drives and motors, some condition-monitoring systems are a bit more complicated to set up. A more in-depth vibration analysis system, such as Siemens’ CMS range, requires piezoelectric vibration sensors to be installed on the motor (and gearbox, if applicable). This would probably involve drilling and tapping holes in key locations, and would need advice from a vibration analysis specialist as to which locations are most appropriate.

Whether you need an expert to interpret condition-monitoring results depends on what is required in terms of interpretation. With vibration monitoring, for example, it is possible to determine upcoming failure modes for a particular bearing in a machine. This level of detail requires the involvement of a vibration monitoring expert – or, at least, someone extremely familiar with vibration analysis tools such as Siemens’ X-Tools software. However, many applications of condition monitoring do not require this level of detail, and only need warning of an impending failure in time to arrange for a maintenance engineer to inspect the asset and identify the problem.

Once the condition-monitoring system is set up, and the standards of healthy operation recorded, all that is needed is for someone to observe the state of current operation and compare it to the state of healthy operation to identify any significant deviations or trends away from the healthy state.

With the levels of automation available today – from the simple ability to set up alerts when a parameter goes outside of a pre-defined boundary, through to automated trend monitoring and maintenance scheduling – it’s now easier than ever to perform condition monitoring.

Whether you need an expert to interpret condition-monitoring results depends on the application and the results you require.




Magazine
  • To view a digital copy of the latest issue of Drives & Controls, click here.

    To visit the digital library of past issues, click here

    To subscribe to the magazine, click here

     

Poll

"Do you think that robots create or destroy jobs?"

Newsletter
Newsletter

Events

Most Read Articles