This audio was created using Microsoft Azure Speech Services
Reliability and safety are two paramount components of a data center. At the same time, there is an inherent conflict between the two. High system reliability and performance is often achieved though selective coordination, which results in increased arc flash energy, a major safety concern.
Understanding the reliability versus safety relationship requires a basic knowledge of selective coordination. Enhanced data center reliability can be obtained through selective coordination, which refers to the selection and setting of over current protective devices (such as circuit breakers) in an electric power system so that the smallest possible portion of the system is de-energized in an overload condition. This ensures any over current event is cleared by the smallest circuit breaker in the system before allowing a larger line-side circuit breaker to operate on the fault.
Through this, the reliability of the system is increased by limiting the service interruption to only the circuit experiencing the problem and does not shut down a larger portion of the facility. As data centers push the limits of reliability, slowing down the tripping of breakers through selective coordination is essential.
The rationale for selective coordination is self-evident – clearing and isolating faults as quickly as possible without disturbing the unaffected portions of the system. However, this solution to increase reliability comes at a price; it results in an increased total clearing time of protective devices during an arcing fault, thereby causes an increase in arc flash incident energy.
Why is increased arc flash incident energy a problem? An arc flash is the release of energy through light and heat produced by an electric arc. Sufficient levels of arc flash energy can cause substantial damage both to workers and the equipment. There is a growing awareness of the risks associated with arc flash, as worker safety standards, such as the National Fire Protection Association (NFPA) 70E: Standard for Electrical Safety in the Workplace, are becoming increasingly strict. 
A few key factors affect arc flash incident energy: the available fault current at the equipment, the time taken by the upstream protective device to clear the fault, and the distance from the arcing source.
Therefore, data center design managers need to delicately balance reliability and safety. There are trade-offs to maintaining reliability that selective coordination offers and ensuring safety by reducing arc flash hazard risks. While industry standards have yet to be defined, such as the optimal device-to-device selectivity or the selectivity thresholds, new technologies are being developed to strike this balance.
One method is the use of zone selective interlocking (ZSI), which allows electronic trip devices to communicate with each other so that a short-time trip or ground fault will be isolated and cleared by the nearest upstream circuit breaker with no intentional time delay. Devices in all other areas of the system remain closed to maintain service to unaffected loads. Zone-selective interlocking eliminates intentional delay, without sacrificing coordination, resulting in faster tripping times. This limits fault stress by reducing the amount of let-through energy the system is subjected to during an over current.
While there is no absolute consensus on selective coordination standards or the perfect way to achieve the balance with arc flash hazards, one thing remains clear: to maximize reliability and ensure safety, optimized selective coordination and arc flash mitigation have to work in tandem.
If you would like detailed information on the balance between selective coordination and arc flash, please refer to the white paper, Selective Coordination vs. Arc Flash Requirements.
