Though not the most common power quality disturbance in most locations, the transient, or rather, the source of the transient, can be one of the loudest. Transients—also called spikes, surges or impulses—can be generated when lightning strikes electric power or telecommunications lines, a building, or an object close enough to either of those. It’s especially bad when it couples into the electrical transmission or distribution system at the utility level or the distribution system within a building. Whereas the original Big Bang is said, by some, to be the start of the Universe’s creation, this one can be the end of your business or home. The unleashing of one of nature’s most powerful phenomena into electronic and electrical loads and circuits can damage and/or destroy thousands of dollars worth of equipment faster than the blink of an eye, unless proper protective measures are in place, properly installed and properly working.
The National Electrical Code (NEC) has definitions in Article 100 for surge arrestors and surge protective devices that are similar and as follows. Color and bold characters are for emphasis and are not part of the Code:
SURGE ARRESTOR. A protective device for limiting surge voltages by discharging or bypassing surge current. It also prevents continued flow of follow current while remaining capable of repeating these functions.
SURGE PROTECTIVE DEVICE (SPD). A protective device for limiting transient voltages by diverting or limiting surge current. It also prevents continued flow of follow current while remaining capable of repeating these functions and designated as follows:
TYPE 1—Permanently connected SPDs intended for installation between the secondary of the service transformer and the line side of the service disconnect overcurrent device
TYPE 2—Permanently connected SPDs intended for installation on the load side of the service disconnect overcurrent device, including SPDs located at the branch panel
TYPE 3—Point of utilization SPDs
TYPE 4-—Component SPDs, including discrete components, as well as assemblies
FPN: For further information on Type 1, Type 2, Type 3, and Type 4 SPDs, see UL 1449, Standard for Surge Protective Devices.
[National Electrical Code, NFPA 70, 2008 Edition.]
The UL 1449 third edition defines the electrical environment as “Surge Protective Devices (SPDs) designed for repeated limiting of transient voltage surges as specified in the standard on 50 or 60 Hz power circuits not exceeding 1,000 V and designated as follows: … ,” where it goes into a little more detail on explaining the differences in the types and locations at which they are to be used, such as for Type 1, “including watt-hour meter socket enclosures and intended to be installed without an external overcurrent protective device.” Articles 280 and 285 cover the installation for surge arrestors and surge protection devices, respectively. The NEC states “280.1 Scope. This article covers general requirements, installation requirements, and connection requirements for surge arrestors installed on premises wiring systems over 1 kV,” and “285.1 Scope. This article covers general requirements, installation requirements, and connection requirements for SPDs permanently installed on premises wiring systems 1kV or less.”
Then there is the working group in IEEE on Project PC62.34 working on a document titled Standard for Test Methods and Performance of Low-Voltage (1,000 Vrms or less, 48–62 Hz) SPDs (Secondary Arrestors) with a scope, stating the following:
“This standard applies to surge protective devices designed for application on the low-voltage supply mains (1,000 Vrms and less, frequency between 48 and 62 Hz) and intended to be connected at locations between, and including, the secondary terminals of the distribution transformer and the line side of the service entrance panel. Such surge protective devices are also known as secondary arresters. This is coordinated with C62.44 (the application guide), NEC Article 580, and UL 1449 3rd Edition.”
Returning to the nasty transients, they come in a variety of types, so we will define and limit our discussion of those as well, since only some of them are effectively limited by the aforementioned SPDs. Transients are normally subcycle phenomena relative to the power line frequency, meaning that their duration is measured in microseconds to milliseconds. The amplitude can be several times the normal peak voltage on the system. From IEEE 1159 Recommended Practice for Electric Power Quality Monitoring, we have two classifications of transients: impulsive and oscillatory.
Impulsive transients can be further defined by the direction that the transient is relative to the point on the voltage sine wave: negative if it is toward the zero axis (subtracts energy from the waveform), positive if it is away from the axis (adds energy to the waveform), and if it only goes one direction—unipolar—or goes two directions—bipolar. An oscillatory transient can start out in a negative or positive direction, such as a power factor capacitor switching transient that goes negative first, but then oscillates above and below the normal sine wave curve as its amplitude is decreased to zero, usually within a quarter cycle or so (4 milliseconds). Examples of such can be seen in the figures below.
Of these types, SPDs can be effective in clamping the peak of the positive kick-back portion of the PF cap switching transient and also can clamp voltage swells (which are the opposite of the more common voltage sags). But the SPDs usually are put in to protect against the positive unipolar impulsive transients, which is what lightning usually causes.
When the electrical current of a lightning strike is coupled into the wiring, it is, by nature, adding a very significant amount of electrical energy into the system—the big bang that destroys things. SPDs work as the definition said. The clamping voltage level, referred to by UL as the “suppressed voltage rating,” can vary and ranges from 330V to 4,000V, with a lower rating being better, as long as it is not below the normal range of peak voltage. Peak voltage usually is 1.4 multiplied by Vrms of the system so that it would be clamping every cycle.
Depending whether it is the common mode-clamping device or the differential clamping device inside the SPD, it will either divert the current to ground or absorb it, respectively. This is a key point to remember with such devices. If they are to divert the common mode developed transient voltage, then there better be a low impedance connection (at 100 KHz to 1 MHz) to earth ground. It is essential that surge protectors only be used with properly grounded outlets and not with 3-to-2 prong adapters or on the end of long extension cords. The voltage can rise kilovolts per foot even in No. 6 AWG fire from the lightning current levels.
The differential or normal mode transients are “eaten” by the device, which can only eat as much energy as its joule rating. Each time it does this, its stomach shrinks a bit. At some point, the influx of energy will be more than the device can handle, and it will fail, normally in a short-circuit mode. SPDs should have a current-limiting protective device, such as a fuse, so this potential short circuit does not result in a fire.
Can you sleep well with just one SPD at the service entrance or an outlet strip with such protection where your computer is plugged in? I guess it depends on how lucky you feel. The odds are against you. Critical equipment should have protection right at them, and telecommunication wires that leave the premises also should have such protection. To really be protected, any external outlet box or fixture should have such protection, as that is a potential gateway for lighting to enter the facility. The properly rated devices for the application should be installed per the manufacturer’s instructions and the NEC and other applicable local codes.
By this nature of diverting or absorbing energy, it should be clear that such devices cannot “reduce energy consumption” as some ads would like you to believe. Nor can they have any effect on negative transients, since they aren’t rising in amplitude above the normal peaks of the sine wave to where the clamping voltage begins to take effect. Likewise, they cannot stop transients that cause zero crossing errors. And lastly, they cannot do anything to help ride through sags or interruptions, since they only absorb energy, not create it to fill in the missing energy from those types of events. When properly installed and maintained, they can, however, protect your equipment and facility against the “big bang.”
BINGHAM, a contributing editor for power quality, can be reached at 732.287.3680.