Not only do variable speed drives (VSDs) run motors at the right speed to optimise energy use, they can also detect excessive motor temperatures and extreme load conditions, improving reliability. However, they can only make this contribution if they are themselves reliable, explains Anthony Rawson, Service Manager for ABB Drives and Controls.
So, what causes failures in VSDs? Some 80-90 percent of failures are the result of excessive temperatures or unsuitable environmental conditions.
These in turn can be the result of blocked cabinet filters, failed cooling fans and contamination from dust, dirt or chemicals. Other factors include moisture, incorrect installation leading to poor ventilation, earthing and motor and cable faults.
Mitigating these failures comes down to maintenance, which has three strategic forms: corrective, preventive and predictive.
Corrective maintenance is essentially fixing it when it breaks. Because no routine maintenance is carried out, the risk of VSD failure is higher. This also entails high costs, both in labour and time and unexpected lost production. Corrective maintenance is simply not sustainable or cost effective.
Preventive maintenance is based on interventions at fixed time intervals or a set number of running hours. This is done in order to prevent potential breakdowns. Regular tasks should include changing the filters in the motor control centre (MCC), vacuum cleaning the heat sink and changing VSD cooling fans. As the only moving part in the VSD, these are particularly susceptible to failing.
Preventive maintenance carries much less risk of breakdown but costs may be higher than necessary as maintenance may be carried out more frequently than is actually required.
Predictive maintenance uses data and statistics to predict abnormal running or abnormal conditions. It seeks to determine the optimum time to conduct preventative maintenance. Although it carries a slightly higher risk of breakdown than purely preventative maintenance, it achieves the lowest overall cost.
Predictive maintenance can be particularly effective when it is based on remote condition monitoring. Smart drives that can monitor both themselves and connected components such as motors can communicate with systems that can analyse this data. This is used to predict where problems might occur and to recommend actions to prevent them.
Just how effective predictive maintenance can be was demonstrated recently at a remotely monitored plant that uses a VSD to run a pump. A temperature difference between two power modules was detected. On investigation, this was found to be due to a build-up of dust near a heat sink and a cooling fan not running at the correct speed. The situation was corrected, saving the plant up to five hours of unscheduled down time that would have cost 500,000 US dollars.
Keeping VSDs reliable is the key to maintaining the reliability of the plants they operate and predictive maintenance is the most cost effective way of achieving this.