IEEE 3006.7-2013 pdf download IEEE Recommended Practice for Determining the Reliability of 7×24 Continuous Power Systems in Industrial and Commercial Facilities
4. Special terminology and equipment for 7×24 facilities
Some special terminology has been developed for 7×24 facilities to help in describing systems, determiningcapacity,and various other issues involved with continuous operation.
4.1 Special terminology for 7×24 facilities
4.1.1 N + 1 designation for equipment
n order to keep critical distribution systems in operation, there are often more of the key componentsprovided than what are required based on system capacity, Redundancy, in general, means creation of newparallel paths in the system structure to improve its reliability. In a system with redundancy, there are moreinfrastructure components for the critical equipment (such as generators and uninterruptible power supplyUPS] modules) provided than are required to support the total load.
When discussing redundancy of a system, it is common to refer to what is required as “N” (for number). lfa facility has two of a particular component, and both are required to carry the critical load, N = 2. If a thirdcomponent was added, the redundancy would become “N+1.” Figure 15 shows an N+1 design in whichN = 2 for the generators and N = 4 for the UPS system.
In order to raise the reliability of the system beyond what is possible by adding redundant components, aredundant system is often added. If one system is required to carry the critical load and two completesystems have been provided, that is referred to as “2N.” Figure 16 shows a 2N design in which N = 2 forthe generators and N = 4 for the UPS system.
4.1.2 Capacity and load density of the data center load
A common metric used to discuss the capacity and load density of the data center is watts per square footin mmetric units, it is watts per square meter. Calculating the watts per square foot is the total usable UPscapacity in watts divided by the total square feet of the data center raised floor (for facilities that have one)or white space (room for IT equipment, often with a white or light colored floor). lt varies to some degreehow the area of the data center is calculated. The most common method is gross square feet of the datacenter rooms.
The other method is in watts per rack where the total usable UPS capacity in watts is divided by the totalnumber of racks that are planned for the space.
Which method is used is not as important as making sure that all comparisons between similar facilities usethe same method. Usable UPS capacity is defined as how much of the actual UPS nameplate rating has been designated in the design. Many owners and engineers designate 90% to 95% of the nameplate ratingas usable UPS capacity as a best practice.
For data centers designed during the 1990s. 50W/A to 75W/f was common, In the 2000s, 150W/A%to200W/f’ is much more common. There are data centers that have been designed for significantly higheroad density, 200W/ft but that requires special cooling methods to remove sufficient heat from the ITequipment.
4.1.3 Single point of failure (SPOF)
A single point of failure is any one component, piece of equipment, or function that by failing causes thewhole system to fail.
Consider the failure mode loss of fuel for an automobile. Most vehicles have only one fuel pump, so thefuel pump is a SPOF. If the fuel pump fails, the automobile stops. (There can be multiple failure modescausing the fuel pump to fail.)
Utility power can be a SPOF if there are no local generators. For facilities that have a local generator, aslong as the utility power is available, the generator is not needed. Therefore, when a site has both utilitypower and a local generator, the generator is not a SPOF. (Protecting electronic equipment, which can onlygo 10 ms to 20 ms without power, generally requires a UPS system to provide power while the generatorstarts.) The local generator could be a SPOF only if there is no utility power (the generator is the onlysource of power for the site)