Resources—News, Articles and Events

UPS Efficiency Ratings: A Study in Energy Savings

Reliable data center facilities use Uninterruptible Power Supply (UPS) systems to ensure that reliable and clean power reaches the critical load. Generally the entire critical load is served by online UPSs. These UPSs may be paralleled together for increased capacity or redundancy, or both. In a Tier Four facility, the UPS systems may be arranged in a 2N configuration in order to ensure that there is always an available path of UPS power to the critical load.

During the initial process of matching a facility’s desired reliability to the business’ actual requirements, the owner will often come up with a dollar amount per minute or hour that unplanned downtime will cost the firm. This amount is then considered against the costs of designing and constructing a facility of sufficient reliability to minimize the risk of this happening. Typically this cost includes facility construction and equipment cost, design costs, and occasionally maintenance costs. One cost that is not always considered, however, is the cost of efficiency of the UPS system itself.

Static UPS systems have efficiency ratings, which are a measure how much of the input electricity is actually available to the load after the overhead incurred by system electronics, power conversion and so forth. These efficiency ratings usually range from around 92% to 95%. Certain systems may be able to achieve efficiency ratings of up to 97% at or near full load.

The issue of UPS efficiency can be broken down into two separate problems:

  1. Different UPSs have different efficiencies.
  2. The same UPS has a different efficiency at a different load level.

This fact, combined with the fact that not all manufacturers publish their data for different load levels, can make this a complicated issue; but it is an issue that needs attention during the design process.

Consider the following example:

A firm is building a data center with 20,000 sf of raised floor space at a load density level of 120 W/sf. For simplicity we will assume that the facility will be fully populated with servers on day one. This building will have a critical load of 2400 kW (20000 sf × 120 W/sf). But this is the load on the UPS output side, not the input side. When one considers the UPS efficiency, it means that the firm is actually paying for the utility power on the input side of the UPS.

In the above scenario, and using an electricity cost of 0.10 $/kWh, the cost of UPS efficiency becomes apparent. If the firm is choosing between two UPS systems to service the load, one with a 95% efficiency rating and one with a 92% efficiency rating the difference in annual electricity cost between these two systems is just over $72,000. Over ten years the difference is over $790,000 in electricity costs alone! A smart firm, then, will look closely at the UPS ratings of different systems when making final decisions.

The Impact of Load Levels on Efficiency

In addition to choosing between different UPS systems, the same UPS system has different efficiency ratings at different load levels.

Consider again our previous example using two different UPS configurations. If the building were powered in a typical N+1 configuration, the UPS system, as a whole, would likely be running at 67% load. If the building were powered with a 2N configuration the loading level would likely be around 42%. (Both these numbers assume we are using 1000 kVA/900 kW UPS modules.) The efficiency of the first system, at 67% load, could be as high as 94%; the second system’s efficiency, at 42% load, would be 79-87%, using a typical double-conversion online UPS system1. Assuming the higher end of that range, the electricity cost difference between the two systems would be around $180,000 per year and over $1.9 million over ten years!

This would indicate that an intelligent system design would do as much as it can to appropriately balance UPS load levels with the needs of redundancy. Additionally, many UPSs would perform at the lower end of the efficiency scale when operating at very low loading levels, increasing the cost of a 2N arrangement even more.

Conclusion

Designing an effective data center entails balancing many conflicting and sometimes confusing design goals. In addition to system topology and other basic design requirements, the actual system components purchased and their operating efficiencies can make a significant difference to the long-term cost of a data center. This is most apparent in the UPS itself, where higher efficiency and optimization of load to match rated levels can save you literally millions of dollars over the long term.

1 “Increasing the Efficiency of UPS Technology”. Energy and Power Management, Oct. 1, 2006. Richard L. Sawyer.