Designing generator systems

When designing generator systems, such as engine-driven generator sets, design engineers must ensure that the generator selection, sizing, and the design of electrical systems are appropriate for the application.

By Tarek G. Tousson, PE, Stanley Consultants, Austin, Texas March 14, 2016

Learning objectives

  • Explain the basics of designing generator systems.
  • Outline the impact of load types on generator system design.
  • Analyze the codes and standards that pertain to designing a generator system.

When designing generator systems, engineers must ensure that the generators and the building electrical systems they support are appropriate for the application. Many questions must be addressed before starting the generator system design. Most of these questions are related to the application and the site operating conditions, which drives the selection of the different generator system components and their characteristics.

The answers to these questions will help engineers make decisions regarding generator sizing, loading sequence, paralleling, fuel storage, switching scenarios, and many other criteria.

Codes and regulations

The purpose of the generator system will determine which set of codes and regulations apply to the design. The applicable codes and regulations set the parameters for the design, performance, and maintenance requirements. NFPA 70-2014: National Electrical Code (NEC), NFPA 99-2015: Health Care Facilities Code, NFPA 101-2015: Life Safety Code, and NFPA 110-2016: Standard for Emergency and Standby Power Systems cover the essential aspects of those requirements. The design engineer must carefully consider all the different codes and regulations that will apply to the generator system design.

The NEC provides the key design requirements for a generator system. These design requirements are found in Article 445, Generators; Article 700, Emergency Systems; Article 701, Legally Required Standby Systems; Article 702, Optional Standby Systems; Article 705, Interconnected Electrical Power Production Sources; and Article 708, Critical Operations Power Systems (COPS).

Article 445, Generators, contains installation and other requirements for generators, such as location, marking, overcurrent protection, ampacity of conductors, and others.

Article 700, Emergency Systems, applies to those systems legally required to automatically supply illumination and power essential for safety to human life in the event of exiting places of assembly during normal power failure or an accident. Per 700.12, General Requirements, when normal power is lost, emergency system power is required to be applied in 10 seconds or less.

Article 701, Legally Required Standby Systems, applies to those standby systems legally required by municipal, state, federal, or other codes, or by any governmental agency having jurisdiction. These systems are intended to automatically supply power to selected loads (other than those classed as emergency systems) that could create hazards or hamper rescue or firefighting operations when stopped in the event of failure of the normal power source. Per 701.12, General Requirements, when normal power is lost, standby system power is required to be applied in 60 seconds or less.

Article 702, Optional Standby Systems, applies to those systems intended to provide an alternate supply (onsite-generated power) to selected loads, either automatically or manually, where interruptions to the electrical system cause serious discontinuation of operation.

Article 705, Interconnected Electric Power Production Sources, applies to synchronous generator sets that operate in parallel.

Article 708, Critical Operations Power Systems (COPS), applies to those systems required to continuously operate for reasons of public safety, emergency management, or national security. These systems are intended to automatically supply power to COPS loads in the event of failure of the normal power source.

Additional codes and regulations apply in some special applications such as fire pumps. NFPA 20-2016: Installation of Stationary Pumps for Fire Protection and NEC Article 695, Fire Pumps, provide guidelines for the generator system design. Per 695.7 [20:9.4], the voltage shall not drop more than 15% of the controller’s normal voltage at the fire pump controller under motor-starting conditions.

Early involvement of the permitting agency and the authority having jurisdiction will help the design engineer determine which set of codes and regulations apply to the generator system during the design phase. It will also help the design engineer understand the different interpretations of the codes.

Site operating conditions

The location of the generator system to be installed will have substantial impact on how the system is built and arranged. If the generator system is to be installed indoors in a designated room or co-located with other building system equipment, an open-type generator-set configuration can be used (see Figures 1 and 2). However, other concerns may arise. Generator sets and ancillary equipment must be accessible for operation and maintenance, building load-bearing capacity must be adequate to house the generator set(s) and ancillary equipment, construction must comply with applicable codes and regulations (noise, emissions, vibration, etc.), and room layout must satisfy manufacturer requirements for adequate combustion-air intake, fuel supply, ventilation, and exhaust. All building system equipment co-located with the generator system must be evaluated because it must be able to operate in the same environmental conditions.

If the generator system is to be installed outdoors, the generator must be provided with a weatherproof enclosure for protection from environmental conditions, such as rain, snow, corrosion, flooding, etc. The enclosure also will restrict undesired access and protect the generator from possible vandalism. Additional features, such as acoustical lining and silencer grade, may need to be specified when selecting the weatherproof enclosure to meet the permitted noise level. A walk-in weatherproof enclosure is convenient for performing maintenance in severe weather conditions. A double-wall subbase tank is an added feature to provide diesel fuel storage. Portable or trailer-mounted generator systems are required in cases where the generator cannot be permanently installed onsite or needs to be used for backup power at multiple locations.

Load characteristics

Load types and electrical characteristics of loads to be served by the generator system have a significant impact on the generator system size, system configuration topology, and complexity of designing the system. Design engineers must understand the load profile and assess the different design options to provide a reliable and economical generator system with a rating based on load profile and run time requirements.

Load types are linear and nonlinear. Linear loads include heaters, motors, and transformers. Nonlinear loads include computers, uninterruptable power supplies (UPSs), electronic lighting ballasts, electronic equipment, and variable frequency drives. Nonlinear loads typically introduce harmonics in the electrical system. Depending on the magnitude of harmonics and total harmonic distortion present in the electrical system, the generator set will be derated to reliably support the loads. Linear loads, such as large motors, have a high starting current (locked-rotor current) that requires the generator set to be oversized in some cases to overcome the demand.

The size of the load relative to the size of the generator system also can have significant impact on the operation of the generator system. Whenever a load is applied to or removed from a generator set, the engine speed, voltage, and frequency will experience a transient condition or a temporary change from its steady-state condition. If the connected load or load-block pickup is too large, the generator may not be able to start. The engine speed, voltage, and frequency may drift outside of the generator system operating limits and trip offline, or they may deviate beyond limits acceptable to the load and cause the load to trip offline. Thus, load characteristics and generator system response capabilities are important design parameters that must be considered.

After defining the load characteristics involved, a load profile must be developed describing the sequence of supplying generator power to the different load types and increments, and how often those loads are switched on and off.

Engine-generator system

The engine-generator system consists of many components that greatly affect the sizing and output performance of the generator set. The common key components to consider when selecting an engine-generator are the alternator type, the exciter type, the engine-speed governor, and the fuel type.

The synchronous generator alternator is the most widely used alternator type. It is the best fit for a backup generator application because the output power, voltage, and frequency can be easily controlled in a standalone (off-grid) application. The output power is generated by the applied torque on the generator shaft from the engine. The output voltage is controlled by the dc excitation current of the revolving field winding. The output frequency is controlled by the rotation speed of the field winding.

The exciter generates the current required for the field winding to establish the output voltage on the alternator winding. Output voltage is continuously monitored and regulated by adjusting the field winding excitation. There are many types of exciters. The most commonly used are the self-exciter and permanent magnet exciter. The permanent magnet exciter offers better response to motor-starting loads and lower order harmonics (see Figure 3).

The governor controls the speed of an engine-generator set to provide the proper frequency of the output power under varying load conditions. The isochronous governor, whether mechanical or electronic, is most commonly used with standby generator systems. The electronic isochronous governor provides more accurate speed response to load variations compared to a mechanical isochronous governor.

The frequency of the generator output power is calculated by the formula:

F = (P * N)/120

Where F = frequency (Hz), P = number of poles of the generator and N = speed of the generator (rpm). For example, a four-pole generator must rotate at 1,800 rpm to provide 60 Hz.

There are several fuel types from which to choose: diesel, gasoline, natural gas, and propane. Before selecting fuel type, design engineers should consider factors such as availability of physical space onsite, reliability of fuel supply, onsite storage limitations and associated hazards, compliance with codes and regulations, cost, and driven-generator performance.

Diesel engines traditionally provide better transient and load response while natural gas engines are more environmentally friendly in terms of emissions. The development of spark-ignited (natural gas) industrial engines can now optimize the speed of these engines to make their transient response similar to that of diesel. Manufacturers are now producing natural gas engine-generator units that can meet the 10-second start-up requirement for backup systems traditionally associated with diesel engines alone. Typically, a natural gas engine-generator unit is physically larger than a diesel engine-generator unit with the same power output. Consequently, it will impact the physical space requirement for the installation. In addition, a natural gas unit would cost more than diesel when considering units with power outputs higher than 150 kW.

The selection between a natural gas engine generator and a diesel engine generator depends on the application, availability of fuel source, project-site conditions, and other project constraints as mentioned previously. If the application and project constraints require the use of a diesel engine generator, a fuel-delivery and storage system should be designed. In most cases for diesel fuel, a double-wall subbase tank is sufficient. However, in large generator units or applications that require a large amount of onsite fuel storage, such as COPS, the size of the subbase tank may not be convenient for operation and maintenance because the unit will sit too high off the pad. A more complex design of fuel delivery and storage would be more practical in this case. A fuel system with a bulk-storage tank, transfer pumps, and day tanks may be a more viable option. Fuel-transfer pumps must be powered by the generator system for continuity of operation during an outage.

Transfer switches

The basic function of a transfer switch in a generator system is to change the load connection from a normal (typically utility) power supply source to an emergency power supply source (see Figures 4 and 5). A manual transfer switch is either physically or motor-operated by qualified personnel to transfer between normal and emergency power when required. An automatic transfer switch (ATS) is motor-operated and is equipped with a controller to sense the availability and the condition of power sources (normal and emergency). The switch automatically transfers from one source to the other according to the preset program in the event of power interruption or scheduled generator system exercises.

When it comes to transferring from one source to the other, there are two types of transfer switches: open transition and closed transition. The most commonly used transfer switches are the open-transition type. The open-transition transfer switch will completely break the connection to one source first, move to a transition position, and then connect to the other source. Conversely, the closed-transition transfer switch will connect the load to both sources momentarily—less than 100 milliseconds—when transferring from one source to the other. The transfer switch is a critical component of the generator system. Design engineers must select the proper transfer switch or switches suitable for the application.

When the normal power source is interrupted or its characteristics (voltage, frequency, or phase rotation) are outside of set parameters, the transfer switch controller will continue to monitor the normal power source until the delay time has elapsed. The transfer switch will send a "RUN" signal to start the generator system and open the contacts to the normal power source, "Open Transition" position. After the generator system is running and reaches the emergency power supply preset parameters for voltage, frequency, and phase rotation within the programmed delay time, the transfer switch will close the connection to the emergency power source. If the emergency power supply fails to reach the preset parameters within the programmed delay time, the transfer switch will initiate an alarm. The transfer switch will continue to monitor the normal power source until it is within the preset parameters. When the normal power source is restored and a programmed delay time has elapsed with no change, the transfer switch will switch back to the normal power source and send a "STOP" signal to shut down the generator system after the programmed delay time has elapsed.

The operating time of the transfer switch, including the delay times to transfer from normal power source to emergency power source, must comply with the applicable codes and regulations. If the transfer switch is installed to serve an emergency system application when normal power is lost, emergency system power is required to be applied within 10 seconds per NEC Articles 700, Emergency Systems, and 700.12, General Requirements. Whereas if the transfer switch is installed to serve a legally required standby system application when normal power is lost, the standby system power is required to be applied in 60 seconds or less per NEC Articles 701, Legally Required Standby Systems, and 701.12, General Requirements. In some applications, such as fire pumps, design engineers must ensure when specifying a transfer switch that it is listed for the application service as required per NEC Articles 695, Fire Pumps, 695.10, Listed Equipment, and per NFPA 20, Chapter 10, Electrical-Driven Controllers and Accessories, 10.1.2.1 Listing.

It is imperative to determine if the generator system is a separately derived system. A separately derived system will require a four-pole transfer switch to switch the grounded circuit conductor (neutral) connection between the normal source ground and the separately derived generator system ground (refer to NEC Article 250.30, Grounding Separately Derived Alternating-Current Systems).

Transfer switches can incorporate features such as service entrance-rated, overcurrent protection, audible and visual alarm notification, and bypass isolation. With the recent changes to the codes, installation of a docking station is becoming a more common design option to provide a safe, reliable, and easy connection for load banks to load-test the generator system as required and also to connect a portable generator as needed.

Paralleling

A paralleled generator system uses multiple generators to form a large-capacity generator system. Paralleling multiple generators is achieved by synchronizing the output of the generators and connecting them to a paralleling switchgear common bus. Synchronizing the output of the generators requires all of the paralleled generators to have the same voltage, frequency, and phase rotation. To simplify the design of a paralleled generator system, identical generators should be used-or, at the very least, generators should have the same output rating and alternator pitch.

If paralleling of dissimilar generators is required because of existing onsite conditions, the design of a paralleled generator system becomes much more complex. Each generator configuration must be evaluated and dissimilar components, such as speed control, voltage regulation, and alternator, must be retrofitted to match.

"Pitch" is the term used to define the mechanical design characteristics of the alternator. Paralleling a generator of 2/3-pitch alternator design with a generator of 5/6-pitch alternator design will result in circulating neutral currents. The circulating current will be disruptive to protective device operation and may damage alternators.

In addition, the electrical system loads must be classified into emergency, legally required standby, and optional standby loads (see Figures 6 and 7). The electrical system should be configured in such a way that classified loads are separated, and the generator sets are sized so that one generator can serve the emergency and legally required loads.

A paralleled generator system can provide several advantages over a single generator system. The main advantages of a paralleled generator system are reliability, availability, expandability, and fuel cost savings. Reliability is established in a two-generator system (2N configuration, full redundancy) by sizing each generator to solely supply the loads. If one generator fails to start, the second generator will start and supply the loads. Reliability in a multigenerator system (N+1 configuration, limited redundancy) is established by installing one additional generator. Maintenance can be performed without interrupting the availability of the generator system because one generator can be removed from the system to undergo scheduled or unplanned maintenance while the other generators are available to supply the loads. Initial installation can be limited to a minimum number of generators matching the load requirement, and when the load requirement increases, the paralleled generator system can be expanded by adding generators. Fuel cost savings are realized when supplied loads vary, and paralleling switchgear adjusts for running the minimum number of generators to avoid fuel inefficiency of lightly loaded generators.

If a paralleled generator system is selected, design engineers must ensure that applicable codes and regulations are addressed and, if necessary, consider different paralleling control-system designs for improved reliability.

Generator sizing

The design engineer should learn how generator-sizing calculations are performed, whether by hand or using software. Most generator manufacturers provide generator-sizing software to assist design engineers and generator system specifiers. Although the generator-sizing software is a very handy tool, the design engineer must evaluate the load and performance characteristics before selecting one generator set over another. In addition, it should be noted that generator rating results are manufacturer-specific and may need to be derated for ambient temperature, altitude, and harmonics. Voltage dip and frequency response will vary between generators from different manufacturers.

To perform manual generator-sizing calculations, the following information is required for each load:

  • Load starting information: starting kilowatts (SkW), starting kilovolt-amperes (SkVA), and starting power factor (PF)
  • Load running information: running kilowatts (RkW), running kilovolt-amperes (RkVA), and running power factor (PF).

For motor loads, this information can be derived from nameplate data: horsepower, efficiency, locked-rotor kVA/horsepower, motor-starting PF, and running PF. In addition, nonlinear load characteristics would be required to appropriately size the generator alternator and select the optimum exciter type. The generator loading sequence will determine how the SkW, SkVA, RkW, and RkVA are summed to find the generator’s total SkW, SkVA, RkW, and RkVA. The generator is subsequently selected to meet the minimum RkW, RkVA, SkW, and SkVA required from the manufacturer’s generator specification sheets (see "Generator sizing examples" in the online and digital versions).


ABOUT THE AUTHOR

Tarek Tousson is senior electrical engineer at Stanley Consultants. His expertise is motors, generators, and UPS systems, and he has 20 years of experience designing electrical power distribution systems for mission critical facilities and other types of buildings.


Case study: Portable backup generator system for a well pump

The design of a portable backup generator system for a confidential water supply cooperation was required to run a water well pump during normal power interruptions. The water plant’s pump and other miscellaneous loads were served from an existing motor control center (MCC). The MCC was fed from a main service disconnect located near the utility transformer. Solid-state reduced-voltage starters (SSRVSs) with external bypasses were used to run the pumps. A programmable logic controller was used to control pump operation. The normal sequence of operation is that when there is a demand to produce water, a start signal is sent to run both pumps. The pumps are stagger-started by delay timers that intercept the run-enable signal to the SSRVS of each pump.

After completing the field work to document the existing electrical system, a design was developed to install the generator system. The proposed generator system design addressed sizing the generator and ATS, sequence of operation, modifications to the existing control system, and equipment layout. Sizing the generator was the crucial challenge because of the nature of the load to be supported and the limitation of the market’s available products. The portable generator had to be capable of supporting both the pump and the miscellaneous loads. It also had to accommodate being mounted on a gooseneck trailer so that it could be pulled with a regular pickup truck and not a trucking rig.

To meet those requirements, a diesel generator with permanent-magnet-generator (PMG)-type excitation was selected to provide the best response to transient loading during motor starting and an improved recovery response. An ATS was installed downstream from the main service disconnect and upstream from the MCC. The ATS was integrated with a generator docking station to provide a safe, reliable, and easy connection for the portable generator. The generator docking station was equipped with an auxiliary wiring connection, which was used for control signaling to send a run signal to the generator in the event of normal power interruption or scheduled generator exercise. The connection also was used to remotely monitor the generator status (fuel level, coolant temperature, etc.). In addition, the connection provided auxiliary power for the portable generator’s battery charger and block heater.

The existing control was modified to prevent both pumps from running when the electrical system is fed from the generator. This was done by installing lockout relays, which intercepted the run-enable signal to the SSRVS of both pumps. The lockout relays were engaged in the control sequence after the ATS was in the emergency source position. A preference switch was provided to select which pump would run on generator power. The preference switch bypassed the lockout relay of the selected pump.

After construction was completed, the portable generator start-up test was performed. The portable generator tripped on overload and failed to start the pump. Another attempt to run the pump on the portable generator was performed, and again, it failed.

What went wrong? Was the generator improperly sized and incapable of running the pump?

By reviewing the generator-sizing calculations and the motor-starting capability of the selected portable generator, it was evident that field verification of the SSRVS-implemented setting, controls, and the portable generator accessories would hold the answer to why the portable generator failed.

The field verification revealed:

  • The SSRVS current limit and ramp-up time were not set properly, resulting in high starting current beyond the generator’s motor-starting capability.
  • The portable generator exciter was the self-exciter type, not the specified PMG type, which limited the motor-starting capability.

After resolving both deficiencies, the portable generator start-up testing was performed successfully. In this case, commissioning would be a more practical approach than start-up testing, which is limited to verifying operation of provided equipment per the design and manufacturer specification and does not address the system operational requirements. Commissioning, which is often confused with start-up testing, is the process of verifying that all equipment and components of a system are designed, installed, tested, operated, and maintained according to operational requirements.


References

NFPA 20: Installation of Stationary Pumps for Fire Protection

NFPA 70: National Electrical Code

NFPA 99: Health Care Facilities Code

NFPA 101: Life Safety Code

NFPA 110: Standard for Emergency and Standby Power Systems covers