How to maintain UPS to avoid data center business interruption

When ensuring greater resiliency in the data center, it is important for organizations to regularly maintain an uninterruptible power supply (UPS) that provides power protection for them. And its maintenance is aimed at minimizing risks and enabling UPS to operate in a safe and efficient manner. However, what if the act of performing maintenance itself constitutes a risk? What kind of countermeasures can the organization take? For example, the data center of a British airline failed due to human error in the summer of 2017. Human error is the main cause of problems during UPS maintenance. Engineers may press the wrong switch or execute procedures in the wrong order. Although it is easy to blame engineers for negligence in these situations, such errors are usually caused by irregular operating procedures, poor markings, and even inadequate training measures. By eliminating these problems at the beginning of the UPS installation, risks can be avoided. For example, if the installed UPS system is a power supply system composed of large-scale parallel UPS and complex switch cabinets, interlocking devices should be added to the design. These measures encourage users to switch in a controlled and safe manner, but are often excluded from the design at the beginning of the project to save costs. Equipment monitoring that runs around the clock also provides powerful protection functions and should be part of the organization's maintenance system. And rigorous training is also crucial. Simple measures can also make a difference. Disaster can be avoided by updating basic labels and switching diagrams in real time. It is recommended that the organization provide a clear documentation of the switching procedures. If on-site maintenance is critical, the maintenance staff will remind each other during implementation (two engineers will check the execution procedures before performing each action) will prevent most human errors. Any repair and maintenance using advanced technology may bring risks to UPS or switchgear, so it is necessary to reduce the number of repairs. And most of the problems (including the failure of electrical components) can be detected in advance by monitoring the heat of the electrical components. For example, if the connection point of an electrical component is not tightened, it will start to heat up and eventually fail in some way. The most effective solution for checking each connection is to use thermal imaging technology. Thermal imaging technology can identify potential problems. Monitoring equipment and capabilities All-weather equipment monitoring also provides powerful protection functions and should be part of the organization's maintenance measures. Strict training is also essential, while ensuring that the equipment supervisor engineer is competent for the job. The maintenance personnel of the organization should not be afraid to ask questions to the maintenance service provider. They have the responsibility to provide qualification certificates, which are related to the organization itself and its engineers, and always need to check the on-site handling. Strong maintenance measures should also ensure timely and effective response when UPS fails. Service level agreements need to be critical to the application. If UPS can be accessed only during normal operating hours, then such a maintenance contract is meaningless for UPS's 24/7 response. In other words, if the data center operates 24/7 and is critical to the business, then a 24/7 response is necessary. The organization needs to specify the composition of the response, which may be just a phone number or an engineer who went to the site to deal with it, and needs to review the qualifications and levels of the engineer. A review of the organization’s current UPS maintenance procedures will help identify and reduce the risk of critical operations that the organization may not have anticipated. Organizations can avoid accidents by conducting due diligence.