As commercial and industrial customers become more and more reliant on high-quality and high-reliability electric power, utilities have considered approaches that would provide different options or levels of “premium power” for those customers who require something more than what the bulk power system can provide.

Recent surveys suggest that power-quality (PQ) and power-reliability related events cost U.S. industry nearly US$174 billion in lost productivity and profitability. These numbers indicate a clear need to bridge the gap between what the electric power system can provide and the level of immunity to power disturbances that electronic equipment must be capable of withstanding.

These incompatibility problems have become so prevalent that the Electric Power Research Institute (EPRI; Palo Alto, California, U.S.) and the Electricity Innovation Institute (E2I) have cosponsored a major research effort (referred to as CEIDS) with a long-range goal of improving the overall performance of both the electric power system and the electronic equipment connected to that system.

In the interim, while the CEIDS effort is moving forward, we have a major gap to fill in terms of keeping process equipment online and operational for electric power users.

This article overviews the hardware solutions that, when applied at the distribution system level, will supply the PQ performance improvements that enable the majority of commercial and industrial customers to increase their process uptime and subsequently improve bottom-line profitability and productivity.

What is Custom Power?

The term “custom power” has been coined to describe distribution system level power-conditioning products. This is primarily because the market for these multimegawatt power-conditioning solutions is nowhere near the size of the market for traditional facility-level PQ and reliability, such as uninterruptible power supplies (UPS) and surge protectors.

In many cases, a customized system design is required to properly integrate power conditioning to the specific distribution system where it is applied.

Custom power is formally defined as the employment of power electronic or static controllers in distribution systems rated up to 38 kV for the purpose of supplying a level of reliability or PQ that is needed by electric power customers who are sensitive to power variations. Custom power devices or controllers, include static switches, inverters, converters, injection transformers, master-control modules and energy-storage modules that have the ability to perform current-interruption and voltage-regulation functions within a distribution system.

Each custom power device can be considered to be a type of power-conditioning device. In general, power-conditioning technology includes all devices used to correct end-user problems in response to voltage sags, voltage interruptions, voltage flicker, harmonic distortion and voltage-regulation problems. While the term “power conditioning” has no voltage boundary, the term “custom power” is bound by the scope of IEEE P1409 “Guide for Application of Power Electronics for Power Quality Improvement on Distribution Systems Rated 1 kV through 38 kV.” This scope limits the location of a custom power device to a medium-voltage primary distribution system. Other than their kVA rating, the technologies used at medium and low voltage often are similar; however, custom power devices generally have load ratings in excess of 500 kVA. The main custom power configurations are:

  • Static series compensator (SSC) provides ride-through for sags by injecting a signal to offset the voltage lost during the sag.

  • Backup stored energy provides ride-through for sags and momentary interruptions by using stored energy (batteries, superconducting coil, ultracapacitors).

  • Static transfer switch provides ride-through for momentary interruptions and most voltage sags by quickly switching between two different utility feeders.

  • Reactive power and harmonic compensation devices provide dynamic compensation to the power system when temporary corrections are required to support varying loads and power-system stability conditions.

Compensation Devices

Momentary voltage sags and interruptions are by far the most common disturbances that adversely impact electric customer process operations. In fact, an event lasting less that one-sixtieth of a second (one-cycle) can cause a multimillion-dollar process disruption for a single industrial customer. These events don't even have to be interruptions, thus a transmission circuit fault 50 miles (80 km) away from an auto manufacturing facility can result in a process shutdown even if the voltage only drops to 70% or 80% of the system nominal. Several compensation devices are available to mitigate the impacts of momentary voltage sags and interruptions as described in the following sections.

Source Transfer Switch

Source transfer switches (STS) have been used for many decades to protect critical loads from power system interruptions. The traditional switch can be manually or automatically switched to the alternate power source to improve power system reliability. Typical times for the transfer are on the order of seconds to minutes, therefore, loads not supported by a ride-through storage device will shut down and must be restarted.

Within the last decade, technology advances have improved the speed and performance of the STS, enabling their use for both reliability and PQ applications. Solid-state (static) switches now can be used for the source transfer operation, thus decreasing the switching time and allowing for a more seamless transfer of load from one source to the next. The detection and switching time is virtually sub-cycle (eight milliseconds or less) allowing supported loads to be supported without the need for a restart. This section focuses on the newer technology available to utilities, specifically the static STS, and will address traditional automatic transfer switches, high-speed vacuum-switched transfer systems and hybrid systems.

Power-System Requirements

Regardless of the specific STS technology employed, certain criteria must be met for proper operation of the devices. For an STS to be effective in protecting critical loads from power-system disturbances, the two feeders (or sources) must be independent of one another. For example, a system disturbance on the preferred feeder should not cause the alternate feeder voltage to fall out of desired limits. If this were the case, when a disturbance occurred on the preferred feeder, the STS would be transferring the load to a feeder of poor-quality power. This scenario would result in the load being unprotected from the disturbance.

Another obvious requirement of the system is that both sources have the capacity to supply the critical load individually. If the STS transfers the load to a feeder with lacking capacity, the voltage level may drop considerably, resulting in a voltage sag. Synchronization of the two sources is not required but is recommended. Typical STS systems can transfer the load when the two sources are out of phase, but particular loads may be affected by the phase change (motor loads). Therefore, both sources should be in phase to allow a more seamless transfer of the critical load.

Static STS Circuit Topologies

Medium-voltage static STS systems usually are used to provide facility-wide protection. An entire facility can be transferred to an alternate source without experiencing equipment failure. Medium-voltage static STS systems currently are available up to 35 kV at 35 MVA.

Hybrid Source Transfer Switch

Another STS option is a hybrid static switch in parallel with a vacuum switch. During normal operation, the preferred-side vacuum switch conducts, thus supplying power to the load. When the need for a transfer arises, the vacuum switch opens and the appropriate thyristor is gated. The opening of the vacuum switch produces an arc voltage, which, in turn, forward biases one of the preferred-side thyristors. Once this occurs, the load current begins to conduct through the preferred-side static switch. The load is then transferred to the alternate source similar to the standard static STS. Once the alternate-side static switch picks up the load, the load is transferred to the alternate-side vacuum switch. This method increases efficiency to almost 100% and eliminates the need for cooling devices.

High-Speed Mechanical STS

The increased cost of medium-voltage STSs has led some manufacturers to reduce the cost of the device by replacing the traditional thyristor with a vacuum switch. Although less expensive and more efficient (at approximately 99% efficient or greater) than the thyristor-controlled switch, the transfer time is longer. Typical transfer times associated with the High-Speed Mechanical Source Transfer Switch (HSMSTS) are on the order of 1.5 cycles with no crossover time (paralleling of the two sources). Therefore, this approach is only a viable solution if the particular load in need of protection can withstand a 1.5-cycle system disturbance.

Static Series Compensators

The purpose of a SSC is to mitigate the effect voltage sags and interruptions have on a sensitive customer loads. An SSC is a waveform-synthesis device that uses series-connected power electronics connected directly into the utility primary distribution circuit through a set of single-phase insertion transformers. A series compensator can be configured to use the primary feeder to derive the energy required to boost the output voltage to the protected loads. This is referred to as a line energy supply or LES configured system.

LES systems may incorporate energy drawn from the incoming affected line. In this system configuration, when the voltage of one or more phases of the incoming supply drops below a preset threshold, the series compensation device injects a controlled amount of voltage into the affected phase or phases to boost load voltage to a more suitable level. The load, therefore, is buffered from the disturbance.

LES is an alternative to a stored energy supply (SES), where the injected energy is provided from some form of onboard, precharged energy source such as DC energy-storage capacitors, flywheel energy storage, superconducting magnetic energy storage or batteries. An SSC can be configured to operate as a standby compensator in which the inverter is not actively in the circuit until triggered by a PQ event that requires action to restore the incoming source voltage to acceptable quality. Alternatively, the SSC may be continually online providing voltage injection during idle conditions that can offset voltage drop caused by a sudden increase in load current through the series-insertion transformer.

Static Voltage Regulators

A traditional step-voltage regulator is a regulating transformer in which the voltage of the regulated circuit is automatically controlled in steps by means of taps, without interrupting the load. This transformer can boost or buck the voltage supplied to a load with a delay on the order of seconds. In contrast, a static voltage regulator (SVR) provides voltage boost during voltage sags by using thyristor switches that rapidly change taps on three single-phase transformers. Generally, this type of device is limited in design to provide up to 50% boost. The rating of an SVR needs to be the same as the full rating of the load that it will protect, because the SVR will carry the entire load during sag or swell events. An SVR is not effective during voltage interruptions because there is no voltage to transform.

An SVR can be configured to have a 1:1 transformer winding ratio to act solely as a load-protection device. However, it also can be configured to operate as a step-down transformer.

During voltage sags in which the SVR switches to full 50% boost, the current drawn by the unit can be twice as high as normal. Therefore, upstream protection devices need to be coordinated with the SVR so that they do not operate in response to the higher current levels. The SVR does not compensate for the change in the voltage waveshape that occurs during voltage sags. If the load is sensitive to changes in waveshape, such as phase-angle jump, then it may still malfunction during the event, even though the magnitude of the voltage is within design requirements. Since an SVR is designed to operate with individual phase control, it can correct for unbalanced voltages during steady-state operation.

Backup Energy-Supply Devices

Backup energy-supply devices are products that have some type of energy storage such as batteries, capacitors, superconducting magnets or flywheels to supply seconds or even minutes of backup power. This enables protected loads to maintain operation during complete (although usually brief) power interruptions. When configured with longer-term backup generation, these devices can mitigate short-term PQ variations and improve reliability.

All of the devices available on the market can be considered to be a variation of the UPS, with the major differences being the type of storage used, the duration of load support and the mode of connection (standby, line interactive or online).

Online Connected Devices

In a traditional online energy-storage system, the protection unit feeds the load. The incoming AC power is rectified into DC power, which is used to charge a bank of batteries or to energize some of the other energy storage mediums. This DC power is then inverted back into AC power to feed the load. If the incoming AC power fails, the inverter is fed from the batteries and continues to supply the load. Generally, UPS systems are designed to provide 5 to 15 minutes of backup. In addition to providing ride-through for power outages, an online UPS system provides high isolation of the critical load from all power-line disturbances. Online UPSs are available in a variety of sizes.

Custom Power Device Application Matrix
Static VAR Compensation Static Shunt Compensation Source Transfer Static Series Compensator Switch Static Voltage Regulator Backup Energy Supply Device
Voltage Sag 1 4 1✓6
Voltage Swell
Momentary Interruption
Capacitor Switching Transient 2
Voltage Regulation
Harmonics 3 5
Reactive Power Compensation
Cost Range (US$) 50 to 200 kVAR 120 to 175 kVAR 500 to 1000 A 150 to 250 kVA 80 to 125 kVA 750 to 1500 kVA
1 Generally corrects up to 100% of nominal for a “load-induced” sag to 65% to 70% of nominal. Corrects up to 90% of nominal for a “load-induced” sag to 55% to 60%. This technology is not a typical solution for power-system “fault-induced” voltage sags.
2 When installed to replace traditional switched capacitor banks.
3 One model corrects 5th and 7th. Other two models have de-tuning filters provided for harmonics determined to be present.
4 Generally corrects up to between 90% and 100% for a sag to 20% of nominal.
5 One model corrects to the 25th harmonic, one model corrects to the 17th.
6 One device corrects for voltage sags and momentary interruptions. Others typically correct for voltage sags down to 50%.

Offfline Connected Devices

An offline energy-storage system allows the utility to power the protected loads until a disturbance is detected and a static or a mechanical switch transfers the load to the battery or other energy-storage-backed inverter. Since there is a short-duration interruption during the time it takes to detect a mains failure, start the inverter and transfer the load to battery power. A load with some inherent ride-through capability is required for the interruption to go unnoticed. A standby UPS using a static switch rather than a mechanical switch can provide nearly seamless transfer from utility power to battery power during utility events.

Line Interactive Devices

The line interactive energy-storage system is a variation of the offline system in which the design incorporates some means of load-voltage regulation. This can be accomplished by integrating some of the technologies previously described, such as a static regulator or an injection transformer.

Currently, nearly all designs for medium-voltage energy-storage devices are standby using a static switch. These units can achieve greater efficiencies than the online and line-interactive units, and the power devices do not have to be rated to carry the full load all the time. There are tradeoffs with this approach as it has no voltage regulation and results in a slight break in supply during the transfer to the backup energy supply; however, cost-wise, this is the most economical approach.

SVC and Harmonic Compensation

When PQ problems arise from nonlinear customer loads, such as arc furnaces, welding operations and others, voltage flicker and harmonic problems can affect the entire distribution feeder. Several devices have been designed to minimize or reduce the impact of these variations. The primary concept is to provide dynamic capacitance and reactance to stabilize the power system. This is typically accomplished by using static switching devices to control the capacitance and reactance, or by using an injection transformer to supply the reactive power to the system.

There are three basic configurations of static var compensators (SVC):

  • Fixed Capacitor/Thyristor Controlled Reactor (FC/TCR) behaves like an infinitely variable reactor. The unit consists of one reactor in each phase, controlled by a thyristor switch. The reactive power is changed by controlling the current through the reactor by controlling the duration of the conducting interval in each half cycle by issuing gating pulses to the thyristors. A fixed harmonic filter provides the capacitive VARs necessary for voltage regulation under the worst design conditions. With the filter supplying VARs, the TCR controls the amount of reactive power supplied.

  • Thyristor Switched Capacitor (TSC) provides capacitor-based voltage regulation for the distribution system. It consists of several sets of TSC steps. The major components include capacitors, thyristor switches, fuses and possibly a soft-start resistor system. When the TSC is started, a resistor in series with the capacitors can ensure that they are charged slowly, avoiding high inrush currents and corresponding system disturbances that can trip customer-power electronic loads. After the capacitors are initially charged, a contactor can automatically bypass the resistor.

  • Thyristor Switched Capacitor/Thyristor Controlled Reactor (TSC/TCR). A combination of TSC and TCR, is, in the majority of cases, the optimum solution. With a combined TSC/TCR compensator, continuously variable reactive power is obtained, as well as full control of both the inductive and the capacitive parts of the compensator. This feature permits optimum performance during large disturbances in the power system.

Static Shunt Compensation

STATCOM (STATic COMpensator) typically describes an SVC used in both transmission and distribution applications. However, DSTATCOM (Distribution STATCOM) specifically applies to equipment used for PQ improvement in distribution applications.

The DSTATCOM is a shunt-connected, solid-state switching power converter that exchanges reactive current with the distribution system. It uses three-phase inverters to transfer leading and lagging reactive current with the distribution system via a coupling transformer. The DSTATCOM supplies reactive power by synthesizing its output for insertion into the AC power system via high-frequency power-electronic switching. More specifically, the DSTATCOM employs a pulse-width modulation (PWM) scheme to generate higher-than-fundamental-frequency currents for injection into the distribution system. This injection of high-frequency current allows the DSTATCOM to provide harmonic load current compensation.


With a large array of custom power devices available, along with a variety of PQ problems that can lead to misoperation of electronic equipment controlled process, it is useful to have a means of cross-referencing the hardware against the disturbances that this hardware can mitigate.

The table on page 53 provides an overview of events that can be mitigated by the custom power devices mentioned in this article. Each of these technologies has found its way into several applications, either as demonstration projects or as cost-effective and viable solutions to distinct types of PQ problems.

Perhaps the day will come when the electric power system have the built-in capabilities to perform some of the energy-storage and voltage-compensation functions of the described technologies. Until then, the good news is that distribution-level solutions are available to provide the equipment and power system compatibility that enables mission critical customers to keep their process equipment operational when they need it.

Charles Perry is a senior engineer with EPRI PEAC Corp. (Knoxville, Tennessee, U.S.) He has more than 14 years of experience in medium-voltage engineering and PQ investigations. He was the principal investigator for EPRI's Guidebook on Custom Power Devices. He has a BSEE degree from West Virginia University and an MSE degree from West Virginia Graduate College. He is a registered professional engineer in West Virginia and a senior member of IEEE.

Doug Dorr is the director of business development for EPRI PEAC Corp. He has been involved in a variety of power-quality and research projects, including power-conditioning device testing, electric vehicle battery chargers, variable speed drives, fuel cells, ultracapacitors and distributed generation. Prior to joining EPRI PEAC, Dorr directed a five-year North American power-quality monitoring study for National Power Labs. He also chairs the revision effort for the IEEE Emerald Book, Recommended Practices for Powering and Grounding of Electronic Equipment, and chairs an IEEE Low-Voltage Surge Protective Devices Subcommittee. Dorr has a BSEE degree from Indiana Institute of Technology in Fort Wayne, Indiana.