For years, transients generated on the utility grid have plagued industrial and commercial facilities with a high concentration of variable-speed drives (VSDs) and other sensitive loads. The common sources of transients include transfer switch operation, lightning strikes, and utility-level capacitor switching. These transients cause sensitive equipment to trip on momentary overvoltages and result in substantial losses in productivity. The most common source of transients is utility switching of medium- and high-voltage capacitors for voltage regulation and power factor correction.
The traditional method of protecting VSD loads from utility-side transients involves installing line reactors in series with the VSD. The purpose of this practice is to increase line impedance and limit the transient at the drive terminals. In many cases, however, the end user installs the line reactors and discovers the problem has been lessened — but not eliminated.
Traditional surge protection devices (SPDs) employing metal-oxide varistors (MOVs) protect against high per-unit transients, such as those created by lightning, but they will not protect VSDs from capacitor-switching transients. For example, SPDs will typically limit transients to 1.8 per unit to 2.0 per unit, but small drives will trip at 1.3 to 1.4 per-unit overvoltages. Typical silicon-controlled rectifiers (SCRs) used in industrial environments may have only a 1.75 per-unit withstand capability (i.e.,1200V peak inverse voltage (PIV) rating) before they suffer permanent damage.
Recently, a new approach addressing the problem of capacitor-switching transients has been developed. The device, called the zero threshold surge suppressor (ZTSS), employs a passive diode bridge and electrolytic capacitors to shunt transient energy away from sensitive equipment. Unlike traditional SPDs, which technicians install in layers (e.g., main switchboard, subpanel, motor control center, sensitive load), only one ZTSS unit is required to protect an entire low-voltage substation service. The ZTSS is not MOV-based, so it will not degrade over time as multiple transients are suppressed. It can typically limit capacitor-switching transients to 1.2 per unit or less, effectively protecting VSD loads downstream of the device.
Because MOV characteristics are unsuitable for protecting small drives, a suppressor with a lower voltage characteristic is necessary. The ZTSS is designed to reduce the voltage spike below the overvoltage trip level of the adjustable-speed motor drives. It is a capacitor-based, phase-to-phase surge suppressor, and the suppressed spike amplitude is dependent on the time constant of the ZTSS's R/C circuit.
The ZTSS consists of a 3-phase diode rectifier bridge and a DC capacitor bank (see Figure, on page 32). The circuit doesn't require a wye-connected secondary, and the diode bridge's peak output DC voltage is the peak line voltage of the supply. The capacitor bank consists of a number of resistors and capacitors that connect in parallel with each R/C leg protected by dual-element, time-delay fuses.
In practice, when the utility switches its power factor correction capacitors, the voltage on the line will fall first, followed by a sudden rise in voltage. This process repeats itself until the system settles down within one-half cycle.
The ZTSS absorbs the sudden change of the incoming high energy by charging and discharging the capacitors. The capacitors' rate of charge and discharge depends on the time constant of the R/C circuit.
The time constant of the R/C circuit is defined as the time required for the charge on the capacitor to attain 63.21% of its final value. Therefore, the critical design parameters of the ZTSS are the time constant of the R/C circuit and the total value of capacitance (in microFarads) contained in the ZTSS system.
The total capacitance of the ZTSS determines the energy dissipation limitations of the device and must be determined by the actual energy contained in the transient measured at the customer's site.
Compared to traditional transient voltage surge suppression (TVSS) systems, there are some distinct differences with the ZTSS solution, as summarized in Table 1.
The ZTSS's design attempts to provide sufficient suppression for most electrical networks. Designs are available for 208V, 400V, 480V, and 600V networks. However, some instances may require personnel to conduct site measurements and/or alternative transient program (ATP) simulations to ensure an appropriate ZTSS design for suppression of the available transient.
The economic validity of the ZTSS solution depends on a number of factors, including the number of drives requiring protection and their horsepower ratings, the cost and length of downtime as a result of drive trips, the cost of any scrap resulting from drive trips, and the installation complexity of alternative solutions, such as TVSS or line reactors.
Excluding installation charges, line reactors for 10 hp drives cost approximately $150, while TVSS systems cost between $3,000 and $5,000. TVSS systems must be installed in layers, with one TVSS at the service entrance and another at each motor control center.
Table 2, on page 36, compares the cost of the three types of solutions when trying to protect a 480V substation transformer with 30 VFDs (10 hp or less) connected to two subpanel motor control centers.
In some applications, the ZTSS is preferable to line reactors because the cost of scrap and downtime for plastic film extrusion is high, and the process requires substantial time to restart. A ZTSS has a higher probability of protecting the load, and the logistics of retrofitting line reactors into existing drives may be cumbersome.
On the other hand, if only a small number of drives need protecting, the series line reactor will likely be a more economically feasible solution. TVSS solutions, while crucial for protection from high-magnitude, high-energy transients, frequently won't protect small drives from relatively low-magnitude, capacitor-switching transients.
To provide an even wider range of transient protection for sensitive applications, it may be worthwhile to investigate a hybrid unit employing both ZTSS and TVSS technologies.
Although we reviewed solutions for protecting sensitive equipment from utility capacitor-switching transients, we did not discuss the methods for suppressing the transient at its source. These methods include pre-insertion resistors or inductors, tuning reactors with each capacitor stage, electronic “zero-closing” controls with vacuum switching devices, and solid-state switching of capacitor steps.
While these methods provide some reduction of the transient, it is impractical to assume utilities can be convinced to alter their method of capacitor-switching control.
The ZTSS is a capacitor-based design that acts just like a damper or a shock absorber. The ZTSS starts acting as soon as the spike exceeds the nominal voltage level. If properly designed for the application, it should protect sensitive downstream equipment from utility- or customer-generated, medium-voltage capacitor-switching transients.
The ZTSS is suitable for facilities with a high concentration of sensitive drive loads and where the cost of downtime and scrap is high.
Other advantages include its single point of installation and global protection scheme. Use of a ZTSS may eliminate the need to replace standard customer-owned, medium- or low-voltage capacitor banks to detuned or filter systems, as it can safely deal with the transient magnification that these banks often create.
Allan Ludbrook, “The Zero Threshold Surge Suppressor: PWM Drives and Capacitor Switching,” Electricity Today, Vol. 5 No.4, April 1993.
Stephen D. Boutiller, “Mitigation of Magnified-Capacitor-Switching Transients,” Canadian Electrical Association spring meeting, April 1992, Vancouver.
Van E. Wagner, “Utility Capacitor Switching and Adjustable-Speed Drives,”
IEEE Transactions on Industry Applications, Vol. 27 No. 4, July/Aug 1991.
Gary W. Chang, “A Review on Solutions for Mitigating Utility Capacitor Switching Transients,” Power Quality 1999 Proceedings, November 1999.
Roger C. Dugan, Mark F. McGranaghan, and H. Wayne Beaty, “Electrical Power Systems Quality,” McGraw-Hill, 1995.
Traditional Transient Solutions
Transient voltage surge suppression (TVSS) — also known as surge protection devices (SPDs) or surge suppressors — is the most common solution designed to protect a facility against transients. Although similar, SPDs are not the same as lightning arresters or surge diverters. Lightning arresters are used to protect the utility's system and equipment rather than customers' sensitive electronic equipment. A lightning arrester consists of a resistance with series spark-gap devices connected between line and ground. This gap is usually set to break down at approximately a 50% overvoltage, and the lightning arrester does not begin to absorb the energy of the surge until the gap has actually flashed over.
TVSS systems reduce or eliminate harmful transients, surges, and electrical line noise (Figure), thus preventing damage to sensitive electrical equipment. Many TVSS systems feature multiple parallel metal-oxide varistors (MOVs). As the voltage reaches the MOV's rated voltage level, the impedance of the MOV changes state, providing a low-impedance path for the transient to follow. This allows the excess energy to be diverted away from the protected load.
However, when the MOV sustains an overvoltage or a large transient exceeding its capacity, it can go into a “thermal avalanche” or “thermal runaway” condition. This means the zinc oxide material of the MOV breaks down and initiates a short-circuit condition. The clamping characteristic of the MOV is too high to protect small pulse-width modulation (PWM), adjustable-speed drives from the 200%- and-higher voltage transients generated by utility-switched capacitor banks.
A well-designed installation includes a lightning arrester, service entrance SPD, and downstream SPDs. This is the cascade protection scheme detailed in IEEE C62.41. By UL1449 definition, an SPD (or TVSS system) is a surge protection device intended for electrical connection on the load side of the main overcurrent protection, in circuits not exceeding 600V rms.
A bearing manufacturing plant located in London, Ontario, Canada, experienced periodic equipment failures that coincided with the times the utility switched capacitor banks. Transient voltage waveforms showed the presence of a magnified capacitor-switching transient between 7 a.m. and 9:30 a.m. (Fig. 1). The 480V bus captured the waveform, which showed a spike of 1200V peak (1.77 per unit) when the utility energized the capacitor bank.
The utility's 13.8kV distribution station supplies power to the plant, which owns a 1500kVAR-rated capacitor bank that was installed on this incoming 13.8kV bus. The plant has four main substations that step voltage down to the utilization level of 480V. Three of the four main substations have installed fixed-type power factor correction banks.
Discussions with the customer revealed that overvoltage trips were occurring on most of the 1 hp drives from two to four times per day. Fuses in the lighting system and power transistors in the pulse-width modulated (PWM) drives also exhibited a high failure rate.
To investigate potential solutions, network models were made using an alternative transient program (ATP). An ATP is the royalty-free version of the electromagnetic transients program (EMTP) and a widely used transient simulation program. The full 3-phase model included a minimum of 50% inductive load on the 480V bus, a ZTSS unit connected between phases at the 480V level, and a utility capacitor switch. Models of various capacitor sizes — switched at both 13.8kV and 115kV — were made to recreate the measured transient and determine a worst-case design scenario.
The customer installed a ZTSS in the Grind Building seven years ago, and it proved to be effective at suppressing the incoming transient for that service. Based on the ATP simulations and site measurements, the customer installed a newly designed ZTSS in another substation (the screw machine building), and the transient waveforms were captured before and after the new unit's installation (Fig. 1 and Fig. 2).
The transient on the screw machine building substation is now no greater than 1.30 per unit (the average is 1.14 per unit), compared to 1.77 per unit before the installation of the ZTSS, and drive trip problems have been eliminated. ATP simulations show that the new ZTSS unit (patent pending) is directly protecting the transient-sensitive equipment on the same bus and improving the transient magnitude on the line side of the substation by 10% to 30%. That means each ZTSS will assist in protecting the other substations next to the ZTSS-protected substation.