UPS: Planning for Power

Maintaining UPS batteries properly is essential for keeping systems operating as designed and protecting facility operations

By Thomas M. Divine III, P.E  

Along with the explosive growth in data-processing equipment over the last decade has come an unyielding demand for high-quality, continuous electrical power. Often, institutional and commercial facilities meet that demand with an uninterruptible power system (UPS).

For an organization to obtain the maximum benefit from the investment in a UPS, maintenance and engineering managers must select an appropriate system for a facility's critical load and then maintain it to ensure it is in proper operating condition.

Battery Maintenance

Only trained personnel should perform maintenance on UPS batteries, which generate voltages that are dangerous and can even be lethal. Battery racks and cabinets often provide little working space for connecting probes or tightening bolts, and unintentional contacts can easily happen. Sealed UPS batteries look similar to the more familiar and benign automobile batteries, which can make the danger easy to overlook.

The requirements of an effective battery maintenance program depend to a degree on the type of batteries that are installed.

Flooded-cell batteries, whose electrolyte is visible through the glass container, generally deliver higher performance for a greater length of time, but they have higher initial costs and advanced maintenance requirements.

Valve-regulated batteries, also known as sealed or maintenance-free batteries, have lower costs up front and require less maintenance than flooded-cell batteries. But they also have higher internal resistance and shorter life. Flooded-cell batteries can last 20 years, while the average expected lifetime of valve-regulated batteries is 7 years.

An effective battery maintenance program must include regular inspections, adjustments and testing of UPS batteries, with thorough records of all readings. Trained technicians should:

  • visually inspect batteries and racks monthly for signs of corrosion or leakage
  • measure and record the float voltage and current of the entire bank
  • note the electrolyte level in each cell
  • record the voltage and electrolyte density of selected battery cells
  • log the ambient temperature.

They also should verify that spill-containment materials are available, that emergency wash stations are operational, and that the battery-room exhaust system is functioning.

Quarterly maintenance typically includes monthly inspection items, in addition to recording the voltage readings for each cell and electrolyte temperature of selected cells. Annually, technicians should document intercell resistance readings for each cell connection and the internal resistance of each cell. Annual maintenance also involves re-torquing connecting bolts and measuring the exhaust airflow with remedial action, if required. They also should perform annual maintenance procedures after a high-current discharge.

Storage batteries have limited life, usually showing a slow degradation of capacity until they reach 80 percent of their initial rating, followed by a comparatively rapid failure. The number and depth of discharge cycles, ambient temperature and charging characteristics affect battery life. The combined effect of these factors is difficult to quantify, so managers need a means to determine when a battery is near the end of its useful life in order to replace it while it still works and before the critical load is left unprotected.

The only sure way to determine battery capacity is to perform a battery run-down test. The module is taken off line, connected to a load bank and operated at rated power until the specified run time elapses or the unit shuts down due to low battery voltage. If the observed battery capacity is 80 percent or less of its rated capacity, the technician should replace the battery.

Thermal scanning of battery connections during the battery run-down test will identify loose or marginal connections. This test is normally a manager’s only opportunity to observe the battery during an extended, high-current discharge. Scanning should take place during both discharge and recharge cycles.

The optimal maintenance interval for battery run-down testing is a matter of some debate. Testing is expensive and inconvenient, requires a large load bank, and requires removing a UPS module from service and exposing the critical load to a greater hazard of interruption.

Usually, the test must be performed during off-peak hours on a weekend. Managers understandably prefer to delay or avoid this test when possible. A reasonable testing interval is every two years until the battery reaches 85 percent of rated capacity, and annually thereafter. Some experts maintain that managers can avoid this test by rigorously monitoring the internal resistance of all cells and inferring remaining capacity from those measurements.

A battery monitoring system can automate many battery maintenance tasks, including electrical measurements and record keeping. The system routinely can perform voltage, current and resistance readings and can make the data readily available to an analyst.
Battery monitoring systems range in function from a simple hit counter, which records the number of discharge events, to highly sophisticated systems that continuously log electrical data and present it in graphic form. While these systems can reduce routine maintenance costs, they are quite expensive.

Managers have a variety of options with regard to battery maintenance. They can elect to perform all maintenance tasks with in-house personnel, hire outside specialists, or perform some tasks in-house while using contract personnel for less frequent or more specialized maintenance.

UPS Maintenance

UPS modules are designed to provide maximum power in minimum footprint; consequently, maintenance spaces are generally cramped. UPS design varies considerably among manufacturers, and specialized knowledge is necessary to identify inspection and maintenance points within the unit.

Routine UPS maintenance consists of a variety of inspections, measurements, calibrations and preventive actions. The technician shuts down the affected module for these procedures, and remaining modules – or, in non-redundant systems, a standby generator or the local electric utility – provide power to the load until the module returns to service.

The maintenance team inspects the interior of the unit for corrosion and heat damage, records and adjusts the battery-charger float voltage, calibrates metering and protection functions, tightens power connections, cleans the module, and performs other unit-specific maintenance activities as recommended by the manufacturer. If the manufacturer’s service group maintains the module, it will implement engineering change notices while the module is out of service.

During the battery run-down test, technicians should perform thermal scans on internal power connections and components to identify poor or marginal connections. Scanning should be repeated during the recharge cycle to ensure that rectifier components are adequately scanned.

Selecting a UPS and developing an effective maintenance program is a complex endeavor that requires detailed analysis, specific knowledge of available systems and equipment requirements, and a thorough understanding of facility goals and constraints. Maintenance and engineering managers can get assistance from equipment manufacturers — especially with regard to specific maintenance requirements. Or they can engage an independent consultant to help weigh the costs and benefits of equipment selection, sizing and configuration, as well as to develop a maintenance plan that provides system reliability and longevity within the facility's budget.

5 Factors: Selecting the ‘Right’ UPS

When selecting an uninterruptible power system (UPS), maintenance and engineering managers must consider the following factors that can make or break the success of the system:

  1. Load size

    The size of the critical load determines the capacity of the initial installation. The UPS must have adequate capacity to reliably serve the critical load and additional loads, without immediate expansion. The excess capacity of a UPS will depend on the facility's plans for expansion of the supported load.

    In general, capacity should be 150-200 percent of the initial installed load. For small critical loads involving a single computer or a few racks, a single-phase desktop or rack-mounted UPS might be the optimal solution. For larger critical loads, such as data centers, freestanding three-phase modules generally are installed.

  2. System reliability

    System-reliability requirements will determine the configuration of the power system. Very high requirements will lead to a system with multiple UPS modules and multiple battery banks. The system also should have at least one redundant module so it can reliably serve the load if one module fails or undergoes maintenance.

    A single UPS module with a static bypass switch can serve loads with lower requirements to provide utility or generator power during periods when the module is down. The consequences of a power failure tend to dictate reliability needs. If an outage would result in lost revenue, the failure to meet contractual obligations, or lost customer goodwill, it is appropriate to install a redundant system.

  3. Battery run time

    The battery run time of a UPS is the length of time the UPS can reliably supply power to the critical load after input power has failed. Run time usually is defined as the length of time required for connected data-processing equipment to save data files and shut down in an orderly fashion, along with a margin of safety. Typical battery run time is 15 minutes.

    Batteries are heavy and can present a large dead load to a structure, so managers must make sure a structural engineer reviews the proposed installation to determine if modifications are necessary to support the load.

  4. Future expansion

    Requirements for future expansion affect UPS configuration and determine space requirements for future modules and battery banks. Depending on the timing of the expansion, it might be more economical to install a single module and add modules as needed, rather than installing a single, larger module.

    Managers who intend to install more capacity later should consider the electrical infrastructure required to support the maximum load, and they must carefully guard spaces allocated for expansion to ensure those spaces are not filled with other equipment.

  5. Budget constraints

    Budgetary constraints play a key role in determining the final UPS design. Often, a system that satisfies other considerations simply will be too expensive to implement, and some functionality or system reliability will have to be sacrificed to keep costs in line.

— Thomas M. Divine III

Contact FacilitiesNet Editorial Staff »

  posted on 7/1/2003   Article Use Policy

Related Topics: