For many applications, backup power systems protect a company’s business investment. During periods of utility power loss, these systems secure ongoing revenue, retain customers, prevent product and information loss, and maintain a safe and secure environment. For hospitals, however, the impact of the emergency power system is literally ‘life and death’. During natural disasters, the emergency power system must allow the hospital to defend-in-place for multiple days until help can arrive. For these reasons, the design of hospital emergency power systems gets significant attention within codes and standards, as well the design community.
The essential power system in a hospital is comprised of multiple code-specific circuits connected back to a common generator system. The essential power system is comprised of emergency loads for life safety and critical load support.
Both life safety and critical circuits are classed as emergency systems, requiring power to be restored within 10 seconds, but for hospitals (assuming over 150 kVA) these circuits are separated onto different transfer switches.
The essential power system also includes essential equipment circuit(s), consisting of medical imaging, building systems for heating, air handlers, exhaust fans, smoke control and stair pressurization. These circuits are separated onto their own transfer switches and are required to transfer at appropriate time-lag intervals.
Additionally, many hospitals have non-essential equipment circuits that are classified as NEC 702 optional standby loads. One of the most common examples of this load type is building cooling which is often deemed necessary for hospital sustainability in many parts of the country. Many system designers include 60% of the chiller capacity onto the generator system.
In reviewing NEC 517, items like stair pressurization that are normally part of the emergency system are reallocated to the equipment circuit, and NEC 517.26 specifically claims the ability to amend NEC 700 emergency requirements. As a result, the requirements for emergency system sizing, selective coordination, and separation of circuits are all amended and applied differently in healthcare.
Selective coordination is the process of selecting and locating overcurrent protective devices, so the device closest to the fault trips, minimizing the impact to the entire system. Traditionally, in the first round electrical design, thermal magnetic breakers are utilized.
These breakers typically require a 3:1 ratio between breaker levels to support coordination. As the design progresses, it may be desirable to compress the resulting breaker range from source to load by using electronic LSI breakers. Additionally, these types of breakers have greater control of trip characteristics allowing the steps between breakers to compress to a 2:1 ratio.
Compressing the breaker range helps align the size of the generator breaker with the size of the first level of feeder breakers. For example, if a 1000kW generator has a 1600 electronic trip breaker, the largest equipment circuit feeder has the potential to be limited to 800 amps. To accommodate this constraint, most hospital projects result in multiple transfer switches per load type, creating the load separation needed to support selective coordination.
Selective coordination has been a much-debated topic since its implementation onto emergency systems and legally required standby in the 2005 NEC. The division and disagreement created through the NEC 700 & 701 adoption continues, and healthcare system designers and code committee members continue to push back on the tight requirements for selective coordination. Recent updates, including NFPA 99 220.127.116.11.2.1 and NEC 517.30(G), now allow healthcare design engineers the flexibility to not selectively coordinate below .1 seconds. These are more relaxed than those required in NEC 700 & 701 which requires coordination in a thermal magnetic breakers instantaneous range (typically to .01 sec).
The move to reduce coordination requirements is rationalized through a more holistic view of the electric power system, while balancing the competing protective design needs against potential – albeit less likely – failure modes. ANSI-IEEE 242 (Recommended Practices for Protection and Coordination of Industrial and Commercial Power Systems) summarized it this way:
“In applying protective devices, it is occasionally necessary to compromise between protection and selectivity. While experience may suggest one alternative over another, the preferred approach is to favor protection over selectivity. Which choice is made, however, is dependent on the equipment damage and the effect on the process.”
Putting it in the context of a hospital, bolted 3-phase faults generate the highest level of fault current, but the vast majority of the faults are low level, line to ground shorts. Systems that coordinate the bolted 3-phase fault during the first .1 seconds may actually result in a less safe system (much higher arc flash potential) and be less optimized for the typical electrical failure.
The multi-part article series exploring emergency power design considerations for hospital applications will continue in the next Codes & Standards article as part of the PowerConnect newsletter.