…testing for product longevity is not as important as testing for and ensuring system and user safety.
Surge protection devices (SPDs), or transient voltage surge suppressors (TVSSs), have become an important part of residential, commercial, and industrial power quality applications. The best SPDs provide enough safety that, in the unlikely event of SPD failure, they won't damage nearby equipment, shut down a process by tripping the upstream circuit breaker or fuse, or cause physical harm to personnel as a result of smoke, fire, or flying debris. Fortunately, market demand and manufacturer innovations have led to improved SPD design and features, based on new fusing techniques and advancements in metal-oxide varistors (MOVs). This article discusses the causes of SPD failure, proper SPD testing, and the recent design improvements.
The Cause of SPD Failures
SPDs with high surge ratings (kA) usually have multiple MOVs connected in parallel to each other so they can share the surge current. However, laboratory tests show that if MOVs are connected in parallel circuits, the MOVs will fail one at a time (not all at once). This occurs because the remaining MOVs have a much higher impedance and will not conduct a significant amount of current after the first MOV in line fails. Therefore, the current will flow through the failed, low-impedance MOV, creating upstream and/or downstream interruption and unsafe conditions on the system.
Although SPD failures rarely occur, they may do so if used incorrectly (e.g., a technician installed a wye-configured SPD into a delta system) or if the MOV experiences a sustained temporary overvoltage (TOV). A TOV up to 200% of normal voltage can result from utility faults; loss of the neutral on 3-phase, 4-wire systems; or an improperly wired device. A TOV, which is not a surge event, accounts for approximately 99% of all SPD failures. These occurrences degrade the MOV gradually, changing its resistance from megohms to milliohms.
Fig. 1, on page 14, depicts how an MOV reacts to overvoltage conditions. The typical maximum continuous operating voltage (MCOV) of a 120/208V 3-phase, 4-wire system is 150V. The MOV could likely handle a 50% increase to the nominal system voltage (1.5 × 120 = 180V) for a period of days. However, prolonged or frequent overvoltage occurrences will reduce the reliable life expectancy of the MOV.
SPDs that promote fuses with excessive surge-current ratings don't provide the proper system coordination. They sacrifice low-level fault protection and fail to disconnect during low-current faults. This can result in catastrophic failure (fire) and eventual tripping of the upstream breaker or fuse.
Proper SPD Testing
SPD manufacturers must adhere to numerous tests, which should undergo independent verification. The limited current test and the available interrupt current (AIC) rating test are two tests generally accepted by the industry. (The AIC rating is sometimes referred to as interrupt rating (IR) or maximum fault current rating.) An SPD must pass both tests in order to receive the UL 1449 listing. To pass the tests, the device must not produce a flame and the internal parts should not become exposed due to the tests.
The limited current test is conducted by applying an abnormal overvoltage (e.g., a line-to-line voltage) to the SPD via a conductor pair, which is connected to two of the following SPD terminals: phase-to-neutral, phase-to ground, and neutral-to-ground. The current is limited to 0.125A, 0.5A, 2.5A, and 5A by means of the power- source impedance.
The AIC rating test is conducted in the same manner, except the impedance must allow for a current of at least 5000A. The end user must note the difference when comparing the SPD AIC rating with the AIC ratings of other system components, such as circuit breakers, contactors, and relays. The SPD AIC rating is tested at abnormal overvoltage, and other components are usually tested at nominal system voltage. For example, on a 120/208V SPD, the test will apply a line-to-line voltage (208VAC) on a phase-to-neutral SPD terminal (120VAC).
These are important tests because they relate directly to equipment and personnel safety. However, there are other tests that are currently in use within the industry.
The NEMA LS-1 specification, published in 1992, designed test parameters around the SPD's commonly promoted surge rating (at the time) of 100kA per phase. The specification required “no more than 10% clamping voltage degradation after ultimate surge.” The test is relatively simple and fast to perform. This allowed the ultimate surge test to gain support and popularity because it was an attainable standard to reach.
However, the clamping voltage, or let-through voltage, during ultimate surge is up to 10 times higher than the nominal system voltage. Therefore, even if it were possible to receive a 200kA surge at the SPD, the clamping voltage would damage or destroy the system's electronic components downstream. Though a number of manufacturers promote higher surge- current ratings and rely on the ultimate surge test, the purpose and relevance of this test becomes questionable for two reasons: (1) the end user's equipment is not protected during such surges, and (2) test labs are not able to generate peak surge currents greater than 240kA.
The ultimate surge test may support claims of longer product life; however, testing for product longevity is not as important as testing for and ensuring system and user safety.
Thermal Protection
MOV degradation happens gradually. The increase in leakage current (caused by TOVs) through the failing MOV progresses at the same rate as the MOV's degradation. This heats up the MOV and adjacent thermal disconnector, causing the disconnector to trip. When an SPD fails, the MOV short-circuits and must be disconnected from the system immediately.
SPD manufacturers recommend using a circuit breaker or an internal or external overcurrent fuse in front of an SPD. To disconnect properly designed SPDs, individual MOV fusing is required. In general, the characteristics of the materials that make up overcurrent fuses don't allow for a simple design. The same overcurrent fuse cannot conduct high-surge current and perform low-fault current disconnection. Because MOVs and fuses share the current well during a surge event, you can improve an SPD's surge-current rating by having one individual fuse per MOV.
On the other hand, you must use overcurrent fuses in combination with a thermal disconnector for low-fault current disconnection. The thermal disconnector will open for sustained low-fault currents but will remain closed for fast high-fault current conditions. Therefore, the combination of thermal disconnectors with over-current fuses on individual MOVs is the best solution.
An advanced SPD on the market today contains a new surge protection platform called Thermo-Dynamic Fusing™. It consists of multiple MOVs, and each MOV is accompanied by a thermal fuse spring (TFS), a fuse trace (FT), or a fuse trace with soldered hole (FTWSH) connected in series. The TFS, FT, and FTWSH should withstand a surge current equal to the surge-current rating of the MOV with which they are associated. These enhancements are based on empirical test data, which prove that the combination of thermal disconnection and overcurrent protection provides the highest level of component disconnection safety relating to fault occurrences.
An overheated MOV produces sufficient heat to prompt a thermal disconnection. For small fault currents, or if the occurrence lasts a longer period of time, the TFS will disconnect first. Some SPDs use an FT and an MOV in series with a 30A or 60A circuit breaker. Those SPDs may pass standard safety tests, but can fail in real world applications.
For example, if the fault current is less than 30A, the SPD might catch fire, but the circuit breaker will not trip. SPDs designed with a TFS will allow disconnection of the shorted MOV at the overheating stage.
However, in incorrect installations or highly abnormal overvoltage conditions, an FT will help in the disconnecting process. At very high fault- current levels, the FT will open faster than the TFS. In these instances, the FT improves the SPD's AIC rating.
Silver FTs have been used for years, but they only protect up to a certain level of surge current unless the cross-sectional area is large. The new design of FTs use copper in their circuitry, which provides better surge (kA) ratings but also less cross-sectional area, allowing for disconnection on low-fault currents.
In very rare situations, such as a high-fault current above 1000A, the TFS might not disconnect the MOV quickly enough. In that situation, an FTWSH provides additional help. The solder-filled hole improves the disconnection time during a fault condition. Tests have shown that if an FT and FTWSH have the same surge current rating, the FTWSH will disconnect at a 50% lower fault current than the FT. These two designs used in tandem provide disconnection capability for both high- and low-fault currents.
Conclusion
In order for an SPD to protect the electronic components of a system from damage or degradation, it must function with a high degree of safety and reliability. Typical well-designed SPDs are maintenance-free, dependable, and last the lifetime of the facility.
At the manufacturing stage, proper testing helps ensure the reliability of SPDs. In a facility application, adding integral thermal protection in series with individually fused MOVs provides protection against low-fault currents (high-impedance faults) by preserving high-surge current protection. A cascaded approach for coordinated, facility-wide protection, as recommended by the IEEE, offers the best solution for a safe and reliable surge-suppression system.
Alan Chiste is product manager for Eaton's Cutler-Hammer business in Calgary, where he oversees product development, applications, and technical training for surge suppression and power conditioning products and solutions. Chiste received his electrical engineering degree from the Southern Alberta Institute of Technology.
Dalibor Kladar works in research and development at Eaton's Cutler-Hammer surge protection and power conditioning division in Calgary. He received his bachelor's and master's degrees in electrical engineering from the University of Sarajevo.
