The Importance of Annual UPS Preventative Maintenance Visits
Uninterruptible power supplies are no different to other critical systems (cooling and fire suppression) within a server room or data centre environment. For a UPS system to provide no-break backup power on demand, it must be regularly inspected and maintained. Most UPS failures are down to poor battery health, leading to shorter than expected runtime and IT load downtime. There are other factors that can also limit the ability of a UPS to keep critical loads running and like battery health, they can only be identified and/or prevented from regular inspection and maintenance.
UPS Maintenance Contracts and Preventive Maintenance Visits
At least one UPS preventive or preventative maintenance visit (PMV) per annum is recommended by UPS manufacturers both inside and outside warranty. For some older systems and those in high security, military, and larger data centres sites, two visits may be required to meet maintenance schedules and service level agreements.
A UPS preventative maintenance visit can be performed as a one-off inspection (typically known as a health check). This is typically the case for third party systems being inspected for inclusion within a service or UPS maintenance contract.
More generally, preventative maintenance visits are included within a UPS annual maintenance contract to ensure the continuing health and availability of the UPS, including replacement of consumable items such as batteries, fans, filters, and capacitors as required.
UPS Maintenance Contracts
Most companies will offer a range of UPS maintenance contracts. These will differ in their level of comprehensiveness, response times offered (working or clock hours from 8 working hours to next day to 4 clock hours) and emergency cover periods (Mon-Fri working hours excluding bank holidays or Mon-Sun 24/7 clock hours).
|UPS Maintenance Contract||Platinum Level||Gold Level||Silver Level||Bronze Level|
|Preventative Maintenance||1 per annum||1 per annum||1 per annum||–|
|Emergency Response||4 clock hours||8 working hours||12 working hours||best endeavours|
|24/7 technical hotline||included||included||included||working hours|
|Battery Cover||extra cost||extra cost||extra cost||–|
|On-site Crash Kits||option||option||option||option|
|247 Remote Monitoring||option||option||option||included|
|Monthly Status Report||option||option||option||alerts reported|
|Individual Battery Testing||option||option||option||–|
|Load Bank Testing||option||option||option||–|
Whilst a typical UPS maintenance contract will cover 1 year (12 months), 3- and 5-year contracts may be available with a discounted price.
It is worth noting that one of the prerequisites for a maintenance contract is the site having an external UPS maintenance bypass switch as part of the UPS installation.
Preventative UPS Maintenance Visits
A site visit will be scheduled and may take place during or outside normal working hours. The more critical the site, the more often the UPS is inspected and maintained outside working hours, when workloads are less to mitigate the risk of power interruption and downtime, especially where the UPS is installed as a single rather than a parallel/redundant system.
Whilst UPS engineers will generally follow a planned preventative maintenance schedule, additional items may be included to meet client requests.
Typical areas and topics covered during a preventative maintenance visit include:
- UPS History and Event Logs: the visiting service engineer will download the UPS’ history and events logs and check them to see if there have been any underlying issues since the last service visit, with either the uninterruptible power supply or the site itself. Previous notes may also be shared with the engineer to provide comparison. The engineer will also check all configuration settings, including firmware version(s) and make sure the UPS is set to achieve its optimum performance. Where issues are identified, the UPS service engineer can change these later in the process.
- Operating Environment: operation in an environment that meets specification is paramount for the UPS and in particular its batteries (lead acid) to meet their working lives. The UPS service engineer will check temperature and humidity and record the results, clean the UPS (noting any dust or ingress issues), check the fans, and remove any dust. If there are fan filters fitted (for higher IP-rated cabinets) these should also be checked and replaced as a consumable item according to the fan manufacturer’s guidelines.
- Inspections: the whole installation is also checked for any signs of overheating or damage. This should include all the cabling to and from the UPS, the UPS bypass panel (where applicable) and any other physical connections inside or outside the UPS system include remote interfaces.
- Battery Set Inspections: next the batteries are checked for any physical signs of venting or corrosion. Their temperature is checked, and the date code of the batteries (or installation date) recorded to ensure that they are still within their working and design life.
- Physical Measurements: these are taken using a calibrated instrument such as a DMM and a Current clamp and compared with the UPS front panel to determine if the display is accurate. The UPS service engineer will also check the Input/Output Voltage(s), Current(s), and Frequency, along with Neutral Current(s) and Neutral/Earth Voltage(s). The battery DC voltage will be checked both before and after a battery test. A battery test serves several purposes; it will check the actual battery capacity which helps determine if the unit will achieve the autonomy needed from the batteries on a mains power supply failure and provides an opportunity to check the recharge current(s). Critically, UPS service engineers should also check the ripple voltage and current which is a good indication of how well the DC Capacitors are performing.
- Switching: an online uninterruptible power supply spends most of its life in normal operation, with the inverter section powering the critical load. During a preventative UPS maintenance service visit the unit may be switched from normal operation to static bypass, from static bypass to manual bypass, and from manual bypass to external manual bypass (where applicable). This will test the internal static switch and the mechanical switches including the UPS maintenance bypass switch.
- Internal Inspection: at this point a service engineer can examine the UPS in more depth as it is isolated, and the critical load is powered directly from the mains power supply. The UPS engineer can check the torque and tightness of all connections and a more detailed inspection of the PCBs and the DC & AC capacitors.
- UPS Config: during a preventative maintenance visit, a service engineer can change system firmware or configuration settings, clear any service or battery alarms, and delete the history and events logs if required. Once the service engineer is happy that the unit is in good order, he or she can bring the unit back to normal operation without loss of power to the load.
- Non-intrusive service: it is not always possible to carry out all the above functions during a preventative maintenance visit. At the very least a UPS engineer should do a full visual inspection and take all the essential measurements. This is a non-intrusive service, and the UPS system remains online throughout the maintenance visit and the load always protected by connection to the UPS inverter.
Regular service visits and inspections will help ensure that the uninterruptible power supply reaches its targeted working life. A typical single or three phase online static UPS should be capable of at least 10 years of services when maintained in accordance with the UPS manufacturer’s recommendations, attended by certified UPS engineers trained by the manufacturer and where manufacturer supplied, or recommended consumables and parts are used.
A Typical UPS Maintenance Schedule
Every UPS system has consumable parts that must be replaced before their failure. A typical UPS maintenance schedule over the working life of a UPS system is outlined in the table below.
|Visual inspection of electrical connections||1 Year|
|Check of alarms on front panel display||1 Year|
|Check of DC filter capacitors||1 Year|
|Fan control & cleaning||1 Year (*)|
|Check of alarm History||6 Months|
|Cleaning of magnetic parts||1 Year (*)|
|Cleaning of electronic boards||1 Year (*)|
|Check of input voltage||1 Year|
|Check of output voltage||1 Year|
|Check of bypass functionality||1 Year|
|DC capacitor replacement||8 Year|
|AC capacitor replacement||8 Year|
|Cooling Fans replacement||4 Year|
Note (*) where UPS operate in particularly dusty environments, the frequency of inspection may be increased.
For more information on Preventive Maintenance Strategies for Data Centres, read this article by Schneider Electric:
If a UPS system is to be available 24/7 and provide instantaneous backup power when the mains power supply fails, it must be maintained as part of a UPS maintenance contract. A contract that simply provides emergency response will not provide the level of cover provided. Any UPS maintenance contract must include at least one comprehensive annual preventative maintenance visit.
Consumable parts must be replaced when identified and corrective actions taken in response to any issues identified by the UPS service engineer. Only then can a site be certain that its uninterruptible power supply will be in the best possible health and ready to protect critical loads from unplanned momentary power outages, short power interruptions and more long-term mains power supply failures.
The infrastructure systems within a server room or data centre are classed as either critical, essential or non-essential. The three most common critical infrastructure systems are power, cooling and fire suppression, followed by environmental monitoring and security systems. Regular inspections and preventative maintenance are required to ensure that these infrastructure systems are reliable and available 24/7.
Server rooms should be designed to provide a secure and managed environment in which to run critical IT servers and network infrastructures. Facilities vary in size from IT closets to computer rooms, Edge to regional and even hybrid datacentres. Whatever the size of the setup, each will have a core set of systems that require regular inspection and maintenance in order to ensure their uptime and resilience.