Emergency Power Distribution Equipment

Emergency and standby power systems are designed to provide an alternate source of power if the normal source of power, typically the electric utility service, should fail. Reliability of these types of systems is critical and good design practices are essential.

Classification of Emergency and Standby Power Systems

• Emergency Power System: NEC Article 700 specifies electrical safety requirements for circuits and equipment that must operate to enable the evacuation of buildings where large numbers of people assemble, such as hotels, theaters, areas, and healthcare facilities. Circuits and equipment that provide emergency illumination are covered by Article 700. Examples of other systems to which Article 700 may apply include ventilation systems, fire alarm systems, elevators, fire pumps, and industrial processes where interruption of power could result in a serious safety risk or health hazard. For instance, if power interruption could result in a release of a hazardous material from industrial process machinery, the associated circuits and equipment could be subject to the provisions of Article 700. The systems are also classified as Level 1, critical to life safety. When power is lost, emergency systems are required to provide alternate power within ten seconds or less.

• Legally Required Standby Systems: NEC Article 701 specifies electrical safety requirements for legally required standby systems circuits and equipment that must operate when the normal supply or system is interrupted. Article 700 addresses equipment and systems that are needed to provide required illumination for building egress or power for equipment essential for safety of life, Article 701 specifies the requirements to provide power to aid support personnel responding to emergencies or supporting recovery from emergency events. For example, while emergency circuits under Article 700 powers lighting required to exit a building, legally required circuits may power lighting that enables responders to view controls for critical building equipment, such as controls for valves, transfer switches, power distribution panels, and other electrical or safety equipment. When power is lost, legally required standby systems are required to provide alternate power within 60 seconds or less.

• Optional Standby Systems: NEC Article 702 specifies requirements for the “installation and operation of optional standby systems, both those that are permanently installed in their entirety and those arranged for connection to a portable supply …. where life safety does not depend on the performance of the system.” [702.1, 702.2] Examples of systems covered by Article 702 include those in (1) residences provided to avoid inconvenience or discomfort; (2) business facilities installed to avoid interruption in the operation of revenue-generating equipment, and (3) warehouses where loss of refrigeration would result in product spoilage and business losses.

The Type refers to the maximum time that an EPSS can remain unpowered after a loss of the normal source. The types are listed in NFPA 110 Emergency Power Supply System Types NFPA 110 Table 4.1(b).

The Class of an EPSS refers to the minimum time, in hours, for which the system is designed to operate at its rated load without being refueled or recharged. The classes for emergency power systems are shown in NFPA 110 Emergency Power System Classes NFPA 110 Table 4.1(A).

The Level of an EPSS refers to the level of equipment installation, performance, and maintenance requirements. The levels for emergency power systems are shown in NFPA Emergency Power System Levels.

Generators are the most prevalent source of power for emergency and standby power systems. For most commercial and industrial power systems these are engine-generator sets, with the prime-mover and the generator built into a single unit. For reciprocating engines, diesel engines are the most popular choice of prime-mover for generators, due to the cost of the diesel engines as compared to other forms of power and the relative ease of application. Engine generator sets can also run on natural gas, however natural gas engines typically have slower response times, than diesel units.

Another option available is the turbine generator, typically powered by natural gas. Gas-turbine generator sets are generally lighter in weight than diesel engine-generator sets, run more quietly, and generally require less cooling and combustion air, leading to lower installation costs. Turbine generators typically are utilized in large capacity applications, when lengthy continuous operation is required and in combined heat and power (CHP) applications. Gas-turbine generator sets are more expensive and typically less efficient than diesel engine-generator sets. They have more complex controls and require longer starting times (normally around 30 seconds compared to the 10-15 seconds or less for diesels). The long starting-time requirement, cost, and lack of available small sizes (< 500 kW) makes the gas-turbine generators infeasible in most emergency and standby power applications.

Considerations for generator installations include the combustion and cooling air required by the generator and prime mover, provisions for the removal of exhaust gases, noise abatement, expected run time hours and emissions. These considerations can increase the installation costs, especially for diesel engines. Fuel supply must also be considered; building code and insurance considerations may force the fuel storage tank to be well removed from the generator(s), usually forcing the addition of a fuel transfer tank near the generator(s).

Engine-generator sets must be sized properly for the application. Several ratings exist for the output capability of an engine generator set. The continuous rating is typically the output rating of the engine-generator set on a continuous basis with a non-varying load. The prime power rating is typically the continuous output rating with varying load. The standby rating is typically the output rating for a limited period of time with varying load. The manufacturer must be consulted to define the capabilities of a given unit.

A means must be provided to switch the critical loads from the normal utility source to the standby power source. Several types of devices are available for this.

An Automatic Transfer Switch (ATS) is defined as “self-acting equipment for transferring one or more load conductor connections from one power source to another” [1]. Automatic-transfer switches are the most common means of transferring critical loads in Emergency Power Supply Systems (EPSS). An automatic-transfer switch consists of a switching means and a control system that senses both normal and emergency supply voltage and frequency. The major functions of an automatic-transfer switch include the following:

• Carry current continuously

• Detect power failures

• Initiate the alternate source (Send a start signal to an engine generator)

• Transfer load

• Sense restoration of the normal source

• Retransfer load to the normal source

• Withstand and close on fault currents – ATS Withstand Current Rating (WCR)

An automatic-transfer switch determines when an outage occurs and after an adjustable time delay (Typically one to six seconds in the event of a momentary outage) sends a start signal to the emergency generator. Upon sensing the generator has achieved acceptable voltage and frequency the automatic-transfer switch transfers the load to the alternate source. When the normal source returns and after an adjustable time delay the automatic-transfer switch controls sense the normal source voltage and frequency and transfer the load back to the normal source when it has achieved acceptable voltage and frequency levels. automatic-transfer switches are available in ratings from 30 - 4000 A, and up to 600 V [1]. Low Voltage Transfer Switches are listed to UL1008. There are options for medium-voltage transfer switches up to 15 kV that are listed to UL 1008A. Because automatic-transfer switches are designed to continuously carry the loads they serve, even under normal conditions, take care in sizing these so that the potential for loss is minimized. Similarly, pay attention to available short circuit current at each transfer switch so that proper Withstand Current Rating (WCR) capability is provided. Automatic-test switches with adjustable pickup and dropout setpoints and integral testing capability should be included.

Automatic-transfer switches can be provided in open-transition, closed-transition or delayed-transition modes.

Open-transition is the simplest mode of transferring load between two power sources. Open-transition operates in a “Break before Make” sequence which results in a momentary interruption of power, typically 30 to 50 milliseconds.

Closed-transition operation transfers between two acceptable power sources without interrupting power to the load in a “Make before Break” sequence. Closed-transition operation connects the normal and alternate sources together for a short (100 milliseconds or less) time when switching between two acceptable sources. Closed—transition switches are typically applied in critical applications to avoid interruption to load when switching between two acceptable sources. Examples include returning to normal after an outage, when load transfers are planned or in test mode. Take special care when transferring motor loads or high inrush current loads.

When power is disconnected from a motor it can become self-excited, delivering energy until they slow to a stop. In addition, if connection to a source were to occur when the motor is 180° out of phase with the source it is transferring to the motor and motor coupling could be subjected to potential damage. There are two methods that can be used to manage transfer of motor loads. One is inphase monitoring which measures the real time phase angle difference between the sources and allows transfer to occur when the sources approach synchronism. This approach is typically used with Open-transition transfer switches. A second method is a Delayed-transition mode of operation. Delayed Transition operation includes an automatic-transfer Switch with a center-off position for the switching contacts. In this mode of operation, the ATS disconnects all loads from both power sources for a brief user adjustable time period, allowing voltage from inductive loads to decay. At the expiration of the time period the ATS transfers to the alternate source.

Manual versions of transfer switches are also available. A one-line representation of an automatic-transfer switch is shown in Automatic-transfer Switch One-line Diagram Representation.

In critical applications when loads cannot be de-energized for long periods of time, combined Automatic-transfer/Bypass-isolation switches (ATS/BPS) are used to bypass the automatic-transfer switch and connect the source to the load via the bypass switch. This allows isolation of the transfer switch for maintenance purposes. Because emergency power circuits must always be available to serve life-safety equipment, NEC Article 700.5(B) specifies that “Means shall be permitted to bypass and isolate the transfer equipment.” Automatic transfer switches equipped with bypass isolation provide continuous power to loads when the ATS is removed from service for inspection, testing, and maintenance. Bypass/isolation Switch Application shows a typical automatic-transfer/bypass - isolation switch arrangement.

In Bypass/isolation Switch Application the bypass switch contacts bypass the automatic-transfer switch, and isolation contacts serve to isolate the automatic-transfer switch. The Bypass/isolation switch is typically manually operated. Bypass/isolation switches are available with a "make before break” feature allowing bypass of the automatic-transfer switch to be completed without disconnecting the load. Automatic-transfer/Bypass-isolation switches are available in Open-transition, Closed-transition and Delayed-transition configurations as described above.

Definition

An Uninterruptible Power Supply (UPS) is an electrical device that supplies temporary power to a load when the input power source fails. This differs from a standby generator in that the UPS provides near-instantaneous protection from power interruptions such that the load is not subject to any power interruption. Energy is provided via a stored energy source discussed later in this document. Continuous power is provided by an upstream generator / ATS source, or through resumption of utility power.

Application

UPS products can range from 200 VA single-phase up to over 1.5 MW three-phase, with a variety of voltages and frequencies available for global applications. They are typically installed in applications that require continuous power so as not to lose critical data, such as financial information, critical process information, sensitive electronics, and lighting, manufacturing, specific health care (MRI, CT, etc.) and other applications that are highly desired to keep up through a temporary power disruption. A byproduct of most larger UPS products is that they provide tightly controlled, highly conditioned power to those sensitive loads, by nature of the double conversion process.

Types of UPS

The three most common types of UPS are offline/standby; line-interactive; and double conversion. The type is typically indicative of the power level and criticality of the load, that is offline/standby are normally small VA capable devices, from 200 W - 2 KW single phase, line-interactive 500 W - 5 kW single phase, and double conversion from 5 KW single phase to the largest sizes of double conversion. The bulk of the discussion in this paper focuses on the three-phase double conversion market:

• Offline/standby: This product offers surge protection and battery backup, but is normally in a standby condition, allowing utility to pass through to the critical load during normal operations. These are most commonly used on a personal desktop, at point-of-sale systems in stores, and other small applications.

• Line Interactive: This product is similar to standby, with the additional feature of an autotransformer that adjusts to ongoing sag and surge conditions to provide conditioned power. These are a lower cost options to compared to double conversion and are typical implemented in small back office computer rooms, to keep small servers and network switches energized during brief outages.

  Line-interactive

  

  Line-interactive 2

  

• Double Conversion: The majority of large scale three-phase UPS in the world utilize double conversion systems. As the name suggests, takes input utility, converts it to DC via a rectifier and charges the Direct Current (DC) source while simultaneously converting back to AC off the DC bus via an inverter. This system provides highly conditioned and controlled AC power to more susceptible critical loads. A bypass static switch allows for continuous power to the load in the case of power train loss in the double conversion system. These UPS modules typically start at five - ten KVA single-phase and can be as large 1.5 MW. Additional system capacity can be achieved by paralleling multiple units together. Total system capacity is limited to paralleling bus and breaker sizes as well as limiting control capabilities. Rarely are system capacities see greater than 3.2 MW.

  Double Conversion

  

Use Examples

UPS systems are used in a wide variety of locations. In the home, personal desktop computers may be backed up by a small offline/standby system. In small business operations, the back-office server, phone switch, network equipment, and other business critical components may be protected by a line-interactive UPS. Typically, units are placed in the bottom of the two-post rack.

In larger applications, double-conversion UPS are used for a variety of purposes. Applications include lighting inverter support for emergency lighting, CT/MRI/PET scanning systems, data centers to back up the critical computing facility, and manufacturing processes to prevent production losses. Many times, a UPS is used to back up the controls portion of whatever process is occurring. This allows for the PLC, computer, or whatever is controlling, for example, the manufacturing process to stay energized through a temporary outage, and thus knows where the process left off until power is restored for the process to continue.

Codes and Standards

Uninterruptible Power Supplies are UL listed conforming to UL1778. Special application standards include UL924 for emergency lighting. Recent DC storage adoptions have seen the rise of UL1973 for lithium batteries, and large-scale fire testing to UL9540A test standards. DC Energy sources are directly coupled with the UPS. They are governed by International Building Code (IBC), International Fire Code (IFC), and National Fire Protection Agency (NFPA) standards, addressed later in this document.

• Transformer versus Transformer-less: Double-conversion UPS have seen a fundamental transformation over the last ten years. Prior generations of UPS included output transformers (a transformer in series with the inverter) for waveform shaping and electrical isolation. Some also included an input isolation transformer, either for galvanic isolation, or to allow for special rectifier configurations. With inverter technology improvements, transformers on both input and output have been all but eliminated. The only common remaining transformer is in UL924 lighting inverter systems at 480 V. Since lighting is many times at 277 V off a 480/277 V system, the UPS takes 480 V 3 W+G input and provides an output isolation transformer, 480-480/277 V 4 W+G, developing the neutral to support lighting loads.

• Bypass:

  A second major design improvement has been in the static bypass switch technology. Higher efficiency requirements have driven more efficient modes, such as ECO or E-Conversion, whereby power is transferred to the load via the static bypass. In the event of a power outage, the UPS transfers back to double conversion mode. This requires the static bypass to be a 100% continuous duty rated component. Previous components were typically momentary duty rated with wraparound contractors.

• Power Factor

  : Power factor and power capability of UPS have been transformed in the last decade as well. In the 1990s, UPS were typically .8 power factor (pf). For example, a 500 kVA UPS could only provide 400 kW of usable power. In the early 2000s, these were improved to .9 pf (500 kVA/450 kW). Power electronics improvements have now made unity power factor the de facto standard, where kVA equals kW (kVA/kW).

• Input Voltages:

  Input versus output voltage of data center UPS remains a common confusion point. Incoming utility transformation is typically to 480/277 V main switchboard use. From here, lighting panels, mechanical loads, and the UPS are fed. While the other equipment (lighting and mechanical loads) might use the neutral and require it to be pulled, the UPS typically does not require, nor desire, a neutral to be run. Downstream of the UPS, in the data center, or point of use, most IT loads utilizes 208/120 V, or something similar. Therefore, a Power Distribution Unit (PDU) incorporates a transformer to develop the 208/120 V required. Thus, a UPS can be 480 V 3 W+G input and output, feeding the 480 V input of the PDU to provide appropriate transformation. This can be a significant cost saver in not requiring a neutral run, four-pole breakers for bypass operation, complexity of operation, and other issues.

  208 V is a common voltage for double conversion in the range between 10 kVA and 150 kVA, and frequently used in small to medium size data centers, as well as Medium Distribution Facilities (MDF) and larger edge computing deployments.

• Redundancy and Reliability: UPS design configurations are often described by nomenclatures using the letter “N” in a calculation stream. For instance, a parallel redundant system may also be called an N+1 design, or a system plus system design may be referred to as 2N. “N” can simply be defined as the “need” of the critical load. In other words, it is the power capacity required to feed the protected equipment. An N system is a system comprised of a single UPS module, or a paralleled set of modules whose capacity is matched to the critical load projection. This type of system is by far the most common of the configurations in the UPS industry. Schneider Electric’s White Paper 75 is available for further information on this topic: https://www.apc.com/us/en/support/resources-tools/white-papers/.

  In large data centers, reliability is a key metric. Mean Time Between Failure (MTBF) has historically been a key indicator of consistent performance. This metric measures the number of hours between component failures within a UPS that resulted in a load loss. Mean Time To Repair (MTTR) is another key metric that determines the amount of time in hours from onset of component loss to complete repair and return of that UPS to operation. While both metrics are key, in modern UPS topologies with modularized design implementations, MTBF values continue to extend while MTTR continues to be reduced dramatically. Schneider Electric’s White Paper 96 is available for further information on this topic: https://www.apc.com/us/en/support/resources-tools/white-papers/.

• Breaker and Generator Sizing: UPS input breaker sizing is an important topic. UPS sizes are stated for their output capacity. They are considered a constant kW device. As such, it is necessary to provide a larger input breaker than expected. This is due to the losses experienced through the UPS, as well as the additional load required on the input to charge the DC voltage source. A typical rule of thumb is to use a breaker that is 125% of nominal UPS output. UPS manufacturers provide detailed breaker sizing information in their literature.

  Generator and Automatic Transfer Switch (ATS) components must also be oversized to account for this additional input load, and to account for UPS input filters that can present leading power factors at low load levels. A 1.5 rating factor was common in previous generations of UPS topologies. With the introduction of power factor corrected rectifier assemblies, Insulated Gate Bipolar Transistor (IGBT) input technology, and other advancements, that generator rating factor can be reduced to 1.1-1.25 in modern UPS installations.

Direct Current Energy Sources

UPS modules can use a variety of DC energy storage systems to provide the necessary energy during outages. These vary from different battery technologies to flywheels. In large data center grade UPS, as the most common voltage is 480 Vac in/out, and the DC bus voltages is 480 Vdc to connect to the DC energy storage source.

• Battery Technologies: Battery technologies that are most commonly seen: Flooded Lead Acid, Valve Regulated Lead Acid, Lithium, and Nickel-Zinc. Each is discussed in the following paragraphs.

  Flooded Lead Acid (FLA) batteries are by far the most space consuming and most expensive choice, which also must include spill containment and room ventilation measures. They were the most common battery types up until the mid-2000s. Since then, they have seen a sharp decline in use, mainly due to size and cost. They are a 20-year product, with a common life expectancy of 13-15 years in UPS applications. They require quarterly maintenance to keep the electrolyte levels constant.

  Valve Regulated Lead Acid (VRLA) have been a common second choice to FLA for decades, due to space savings as well as significant initial cost savings. However, VRLA are a ten-year product, with three-five year life expectancy. They require semi-annual maintenance, and it is recommended to use a battery monitoring system to be alerted when they begin to fail.

  Lithium batteries (LiB) are a recent addition (2016) to the large UPS market. Initially at a significant price disadvantage to VRLA, volume of sales has driven their price point to or on par with VRLA. Due to their longevity of 15 service years, size and weight savings, they are a very attractive choice in modern data center and other designs. Recently, IFC/IBC and the NFPA have all provided updated standards (IFC/IBC 2018, and NFPA855), which require limitations on lithium battery technology. The most common lithium offerings have received UL1973 listings for stationary application and have passed large scale fire testing as required under those standards (UL9540A test standard).

  Nickel-Zinc batteries (Ni-Zn) are a relatively recent addition to battery technology used in UPS applications. They offer similar savings to weight and footprint compared to LiB. As this is an emerging technology, they come at a significant price disadvantage. However, they do not pose the same fire concerns, and can be placed in greater quantities without restrictions.

• Flywheel: In some applications, frequently seen in hospital installations, flywheels are also being used as an available energy source. Flywheels allow for very short ride through times of 20-40 seconds, just sufficient to get to generator on the UPS input. Flywheels are relatively expensive, yet use only a very small amount of space, rivaling LiB and Ni-Zn. They do require extensive maintenance periodically, such as bearing and oil replacements. Service support and availability needs to be considered when considering flywheels as an energy source.

• Warranted Life: Warranty for batteries has also evolved over the last 15 years. FLA typically carries a three - five year full warranty, with a pro-rated value over the remaining 15 years. VRLA commonly carry a three-year full warranty, without subsequent pro-ration. Previously, both technologies offered a cycle life warranty. This required a battery monitoring system to provide the required data. Cycle life warranties have been mostly eliminated in the current battery market. LiB and Ni-Zn commonly carry 10-year performance guarantees, with offerings for full warranty coverage.

• UPS Backup Time: UPS back up times vary widely. Specialty applications such as UL924 have a mandated 90 minute required run time.

  Smaller facilities are typically limited by space and cost, which leads to backup times such as five minutes or less.

  Data Centers are typically in a range of 5-15 minutes on the longer end. However, data centers are nearly always provided with a backup generator, and the actual run time on the UPS is in the 15-40 second timeframe, regardless of actual battery installation.

An additional design consideration is the battery technology used and its limitations, as well as the overall design of the electrical infrastructure. If a generator exists, consider minimal battery run time, as the additional run time is never needed. However, battery chemistries may limit just how short that time can be. A chemical phenomenon known as “coup de fouet” limits VRLA batteries to be not less than five minutes, unless a thin plate pure lead style of VRLA is used, then one - two minutes are common. LiB have a specific amount of energy available (one jar style with a specific amp-hour), therefore they have very specific run times available, from as little as two minutes to commonly five-seven minutes, depending on UPS KW. Ni-Zn will have similar considerations to LiB.

These LiB installations are typically referred to as using “Power” batteries. As described, they can discharge the entire stored energy rapidly (min. versus hours). Energy batteries, also LiB, on the other hand, allow to discharge the stored energy over a much longer period (hours vesus minutes). Just recently are these types of implementation considered to support additional use cases, such as Demand Reduction and Energy Arbitrage to reduce overall electrical utility charges.

Conclusion

UPS implementations are used to protect against power disruptions in many diverse applications. The two main reasons are to protect against financial losses or life-safety concerns. This includes data centers and hospitals or control systems and entire manufacturing processes. While those fundamentals of the different UPS technologies have not changed dramatically over the last decades, what has changed, due to advancements in power electronic components, are improvements in the overall efficiency, energy storage systems, and the reduction of physical size of the UPS. We have witnessed the use of mainly transformerless UPS in system designs, the use of Li-Ion batteries, and the shift from monolithic to modular UPS architectures. The most used UPS technologies are offline/standby, line interactive and double conversion. They range from a few hundreds of watts to multiple megawatts.

Standby and Emergency Power Systems can be configured in customizable bus configurations including single isolated bus, segmented bus, common utility/emergency, main-tie-main bus and ring bus. Several operational modes, including open transition, closed transition and soft load operation can be provided. Configurations include single engine applications and multiple generator paralleling applications. Examples of some of the most common arrangements are shown here.

For Emergency Power Systems with a single alternate power source, NEC Article 700.3(F) requires a means of connecting temporary or portable power, an example is shown in Provisions for Connection of a Temporary or Portable Power Source.

Multiple Generator Paralleling System Configuration

Multiple engine generator paralleling paralleling/synchronizing switchgear refers to the controls and equipment required for connection of multiple sources, usually generators to a common bus and/or a utility source and the load control necessary for the specific application in the event of a power outage. Utilizing multiple generators inherently provides redundancy and a higher degree of reliability by design compared to a single larger generator. N+1 configurations can be provided for capacity and maintainability. For example, a three-engine generator N+1 Emergency System is designed so that any two-engine generators can support the full system load. If one engine generator were to stop functioning, the full load is still supported by the remaining two engine generators. In these applications, system priority load control for load adding and shedding is implemented to so that the most critical loads are connected to backup power when required. Generator paralleling systems are available for a low voltage system up to 600 V and medium voltage systems up to 15 kV. An example of a typical low voltage multiple generator paralleling system with multiple automatic transfers is shown in Low Voltage Multiple Generator Paralleling System with Multiple Automatic Transfer Switches.

Synchronizing: Paralleling/synchronizing engine generators requires additional controls and attention must be paid to the engine generator selection. An engine typically is provided with the same kW size rating and with the same pitch. Voltage regulators and speed controls should also be matched. An automatic synchronized device is required for each engine generator. Synchronization matches a generator’s speed. frequency and voltage with another source, typically another engine generator. Synchronization is necessary when connected generators together to control power surges, avoid reverse power conditions, reducing electrical stress on generators and switchgear and reduces mechanical stress and damage to prime movers. Several conditions must be met to synchronize sources:

• The number of phases must be the same.

• The direction of rotation must be the same.

• The voltage amplitudes must be closely matched (±5%).

• The frequencies must be closely matched (±5 Hz).

• The phase angles must be closely matched (±5º).

Automatic Priority Load Control: In paralleling swtichgear applications loads are assigned priorities so that they are added to the Emergency or Standby Power System in priority order. The highest priority loads are added first following by second highest. Typically priorities are sized in blocks to match the capacity of a single engine generator. For example, a two-engine paralleling system as shown in Low Voltage Multiple Generator Paralleling System with Multiple Automatic Transfer Switches with two 1000 kW engines would have two priority blocks each sized to the capacity of one of the engines. If load shedding is required due to bus overload or engine loss, the lowest priority loads are shed first. Load control can be provided by operating prioritized automatic transfer switches or electrically operated circuit breakers. Automatic transfer switch load control is most common in low voltage applications. Medium voltage applications typically utilize electrically operated circuit breakers for load control. Prioritized load can be added in blocks or steps as shown in Prioritized Block Load Control and Prioritized Step Load Control.

Modes of Operation: Emergency and Standby Power Systems as shown in Low Voltage Multiple Generator Paralleling System with Multiple Automatic Transfer Switches can operate in open transition or closed transition modes. In the example shown open transition or closed transition automatic transfer switches as described in System Arrangements can be provided. Closed transition transfer switches provide no interruption to load when transferring between two acceptable sources. Closed transition operation occurs on return to the normal source after an outage and during transfer switch test mode. Closed transition operation occurs at each automatic transfer switch as shown in Low Voltage Generator Paralleling System with Closed Transition Automatic Transfer Switches.

Multiple Generator Paralleling System with Segmented Bus and Tie Circuit Breaker

Tie circuit breakers can be employed in Emergency and Standby Systems to achieve connection of loads in ten seconds for Emergency Systems per NEC Article 700. In this configuration the highest priority loads are split between the two bus segments. In Segmented Bus Configuration the tie circuit breaker 52T is normally open. When the transfer switches sense a normal power outage all engine generators are signaled to start. Individual generators are concurrently connected to isolated segments of the bus without the need for synchronizing first, bringing multiple engine generators online simultaneously and connecting the highest priority loads within the ten second requirement. As additional engines become available, they are synchronized with the connected generators, connected to the bus and the lower priority loads are transferred to the emergency system. After all the generators are connected and all the transfer switches have transferred their loads to the emergency system the tie breaker can be signaled to close by the emergency system Master Controls. Because the tie circuit breaker 52T is connecting sources together a synchronizing device is required across the tie circuit breaker to synchronize the bus segments prior to it closing. Configurations of this type are common in healthcare applications.

Medium Voltage Main-Tie-Tie-Main Configuration with Multiple Engine Generators

A medium voltage Main – Tie – Main or Main – Tie – Tie Main system can integrate an emergency/standby system into the overall design. More Complex Medium Voltage Main-Tie-Tie-Main system with Multiple Generators Configuration illustrates the integration of a multiple engine paralleling system into a Main – Tie – Tie – Main configuration. There are different variations of the configuration shown below, however all are more complex than a single common bus or a segmented bus with a single tie circuit breaker. The additional complexity however adds additional redundancy to the system design by introducing a second utility connection.

In this application the ties, shown as main breakers 52GM1 and 52GM2, are normally open and the utility circuit breakers 52U1 and 52U2 are normally closed feeding loads on their respective buses. Utility sensing via a protective relay is provided on each utility circuit breaker. Size each utility to carry the system full load. In the event one utility were to go offline as detected by the utility protective the system master controls open the offline utility circuit breaker, close both 52GM1 and 52GM2 allowing the remaining utility to power the entire facility load.

Additionally, the engine generators can be started in the event one utility is lost and operate as reserve capacity in the event the remaining utility were to go offline. In the event the remaining utility were to go offline, the engine generators are already running and available to assume load. Take care when running engines without load connected to avoid wet stacking. In this type of sequence, with one utility available, the engines can be run for set period of time, typically 20 minutes. At the expiration of a time delay and with the remaining utility serving the load the engines are shutdown.

In the event both utilities go offline as sensed by their respective protective relays, the system master controls open both utility circuit breakers 52U1 and 52U2 and the engine generators are signaled to start. When the first engine generator is connected to the dead bus circuit breakers, 52GM1 and 52GM2 are closed, and the highest priority loads are connected via operation of electrically operated feeder circuit breakers shown as 52F. As subsequent engines are synchronized and connected additional loads are connected via their respective feeder circuit breakers, until all loads are connected to the emergency bus. If load shedding is required due to a bus overload or engine loss the lowest priority loads are shed first.

When normal power is restored, several options are available to retransfer load back to the normal source. Retransfer sequences can be open transition, closed transition or closed transition/soft load. In the example shown in More Complex Medium Voltage Main-Tie-Tie-Main system with Multiple Generators Configuration all retransfer sequences occur between the respective utility and main breakers as follow:

Open Transition Retransfer: When normal power is restored main circuit breakers 52GM1 and 52GM2 are opened and utility circuit breakers 52U1 and 52U2 are then closed in a “Break before Make” operation, assuming facility load. The engine generator circuit breakers are opened, the engine generators enter a cooldown period and then shutdown. Provide electrical Interlocks so that the respective main and utility circuit breaker pairs from both cannot be closed at the same time.

Closed Transition Retransfer: When normal power is restored main circuit breakers 52GM1 and 52GM2 remain closed. Both utilities are synchronized with the live bus via additional synchronizing devices provided for each utility. Once in synchronism circuit breakers 52U1 and 52U2 are then closed in a “Make before Break” operation. Closed transition overlap time is typically 100 milliseconds or less. Both main circuit breakers 52GM1 and 52GM2 are opened and each utility assumes facility load on its respective bus segment. The engine generator circuit breakers are opened, the engine generators enter a cooldown period and then shutdown. Additional synchronizing controls are required at each utility for closed transition operation. Benefits of closed transition operation include no interruption to loads when transferring between two acceptable sources or during system test modes.

Soft Load/Closed Transition Retransfer: Soft load sequence is similar to the closed transition sequence however the overlap time is extended, allowing load to be ramped off of the generators on onto the utility. Interconnection can be approximately 30 seconds to several minutes. Benefits are similar to closed transition but also include less wear and tear on circuit breakers and UPS systems. Utility company approval is required for soft load operation and may require specific utility approved protective relaying at each utility circuit breaker. Synchronizing devices are required at each utility circuit breaker.

Hospital Applications and Configuration

Hospital Emergency Systems are code driven and have very specific requires. Applicable codes are NEC Article 700 “Emergency Systems” and Article 517 “Health Care Facilities, NFPA 99 “Health Care Facilities and NFPA 110 “Emergency & Standby Systems. NEC Article 700 requires emergency systems to be designed to automatically supply power for exit lighting, fire detection and alarm systems, elevators, fire pumps, and public safety communications systems and may also provide power for ventilation where it is essential to maintain life.

The emergency system is classified as an Essential Electrical System and is described in NFPA 99 as “A system comprised of alternate sources of power and all connected distribution systems and ancillary equipment, designed to allow for continuity of electrical power to designated areas and functions of a health care facility during disruption of normal power sources, and also to minimize disruption within the internal wiring system.” Essential Electrical Systems are also divided into two types. Type 1 are Critical Care Spaces and Type 2 are General Care Spaces. Additionally, there are areas of a hospital that are classified as Non-Essential were an Essential Electrical System is not required. These areas include waiting rooms, general lighting and non-critical service equipment.

NFPA 99 assigns a risk category to each space within the healthcare facility based on the risk associated with a loss of the power distribution system serving that space. A summary of risk categories from NFPA 99, Chapter 4 is shown inFrom NFPA 99, Chapter 4. Most hospitals are Risk Category 1.

Essential Electrical Systems for hospitals are separated into three distinct branches as described in NEC Article 517.32-34 as follows:

• Life Safety Branch: The Life Safety Branch of the EES provides power for lighting, receptacles to those functions or warning systems that are required to allow building occupants to safely leave the building in an emergency. Transfer switches feeding the Life Safety Branch must be automatic and must be non-delayed. Emergency power must be supplied to the life safety branch within ten seconds of a normal source power outage. Typically, these loads are served by automatic-transfer/bypass-Isolation switches. Often these switches are provided with Closed Transition features. Wiring for the Life Safety Branch is kept independent of all other wiring.

• Critical Branch: The Critical Branch serves loads that either have immediate impact on patient well-being or are essential to the clinical functionality of the health care facility. Transfer switches feeding the critical branch must be automatic. Emergency power to the critical branch must be supplied within 10 seconds of a normal source power outage. Typically, these loads are served by automatic-transfer/bypass-Isolation switches. Often these switches are provided with Closed Transition features. Wiring for the Critical Branch is kept independent of all other wiring.

• Equipment Branch: The Equipment Branch serves loads for major electrical equipment required for patient care. The equipment branch of the EES consists of large electrical equipment loads can include chillers, compressed air systems, exhaust systems and sump pumps needed for patient care and basic facility operation facility. Transfer switches feeding equipment loads are configured for delayed connection to the emergency system.

An example of an Emergency System for a Hospital Application is shown below in Hospital Essential Electrical System with Life Safety, Critical and Equipment Branches.

Essential Electrical Systems are required to have two independent sources of power, the normal utility source and a backup generator or multiple paralleled generators. When normal source power is lost the generator(s) must be started and Life Safety and Critical Branch Automatic Transfer Switches must transfer to the emergency system and provide power to their loads within ten seconds of the normal source power outage. Life Safety Branch transfer switches must transfer to the emergency source immediately upon sensing the availability of emergency power without a delay. Life Safety automatic transfer switches are always considered Priority 1 loads. Critical Branch automatic transfer switches may have a delay to allow the Life Safety automatic transfer switches to connect to emergency power first but must be connected to emergency power within ten seconds of a normal power source outage. Equipment Branch automatic transfer switches can be delayed and are always considered a lower priority than Life Safety or Critical Branch ATS. Automatic transfer switches are required for equipment loads that serve suction.

Hospital Essential Electrical Systems have specific testing requirements as follow:

• NFPA 110 Chapter 8 specifies that an Essential Power Supply System (EPSS) including its transfer switches "shall be exercised under load at least monthly”.

• System testing is required 12 times a year, at intervals not less than 20, or more than 40 days – All ATS must be tested monthly.

• The essential electrical system must be maintained to supply emergency power within 10 seconds of loss of normal power. If the ten second criteria is not met during regular testing, the organization must have a process to confirm on an annual basis that the ten second criteria can be met. Joint Commission requirement based on NFPA 99.

• Tests must be run in accordance with NFPA 110 Requirements.