Surge protective devices are standard on most specifications for industrial and commercial facilities.

A specification should focus on the essential performance, installation and safety requirements. But what criteria are important when specifying a surge protection device?

The following are considered essential performance, safety, and installation criteria for a specification:

Cutler-Hammer recommends 250 kA per phase for service entrance, 120 kA per phase for panelboards or other locations.

Performance should be specified based on the three standard IEEE test waveforms (IEEE C62.41 Category C3 and B3 combination waves, and B3 ringwave). Specify the required ratings for applicable nominal voltages on LG and L-N modes.

Based on the MIL-STD-220 insertion loss test, noise attenuation at 100 KHz should exceed 50 dB (L-N modes). Specify that test results (bode plots) are provided as submittals.

The SPD should be factory-installed as part of the distribution equipment. Ensure the installation minimizes lead length.

For safety and overcurrent protection, a 200 kAIC internal fusing system should be provided.

This should include foolproof status indication for each phase. A popular option is to include Form C contacts for remote monitoring.

To ensure a reliable construction and design, specify that all manufacturers submit results from an independent test lab, verifying the device can achieve the published surge current ratings (on a per mode and per phase basis).

To help you understand the importance of these criteria, let’s answer a few commonly asked specification criteria questions.

Surge Current Capacity is defined by NEMA LS-1 as the maximum 8/20 us surge current pulse the SPD device is capable of surviving on a single impulse basis, without suffering either performance or degradation of more than ten percent (10%) deviation of clamping voltage .

Industry standards publish surge current per-phase, by summing modes L-N and L-G in a Wye system, or L-L and L-G in a Delta system.

Surge current capacity is used to indicate the protective capability of a particular SPD design, and should be used on a per-phase and per-mode basis when specifying a SPD for a given application.

Surge current capacity is dependent on the application and the amount of required protection. The facility’s geographic location and exposure to transients should be considered. Also, consider how critical the equipment is to the facility in terms of downtime and repair costs.

Based on available research, the maximum amplitude of a lightning-related surge on a facility’s service entrance is a 20 kV, 10 kA combination wave (refer to IEEE C62.41). Above this amount, the voltage will exceed Basic Insulation Level (BIL) ratings, causing arcing in the conductors and/or the distribution system.

Cutler Hammer recommends 250 kA per phase for service entrance applications (large facilities in high exposure locations), and 120 kA per phase at branch panel locations.

If the maximum surge is 10 kA, why do many suppliers suggest installing a device that can handle up to 250 kA per phase? The answer is life expectancy .

A service entrance suppressor will experience thousands of surges of various magnitudes. Based on statistical data, a properly constructed suppressor with a 250 kA per phase surge current rating will have a life expectancy in excess of 25 years in a high-exposure location.

Some manufacturers recommend installing SPDs with surge current ratings of up to 600 or 1000 kA per phase. This level of capacity offers no benefits to the customers. A 400 kA/phase device would have approximately 500-year life expectancy for medium exposure location – well beyond reasonable design parameters.

Today's SPDs will not fail due to lightning surges. Based on two decades of experience, the failure rate of an SPD is extremely low; below 0.1%.

Should a suppressor fail, it is likely the result of excessive over voltage (swell), due to a fault on the utility power line (i.e., the nominal 120 VAC line exceeds 180 VAC for many cycles). A severe swell will damage surge protectors and other electronic loads, such as computers. Should this rare event occur, call the utility to investigate the problem.

Joule ratings are not an approved specification for surge protective devices. IEEE, IEC, and NEMA do not recommend using Joule ratings when specifying or comparing surge suppressors because they can provide misleading and conflicting information .

For example, on a 120 Volt system, a 150 Volt or 175 Volt MOV could be used. Even though the 175 Volt MOV has a higher Joule rating, the 150 Volt has a much lower let through voltage.

Joule ratings are a function of let through voltage, surge current, and surge duration. Each manufacturer may use a different standard surge wave when publishing Joule ratings. For this reason the power quality industry does not recommend the use of Joule ratings in performance specifications.

Manufacturers are not required to have their units independently tested to their published surge current capacity rating. Most published ratings are theoretical. They are calculated by summing the individual MOV capabilities.

For example, a manufacturer may claim a rating of 100 kA, but due to the poor construction integrity, the unit may be unable to share current equally to each MOV. Without equal current sharing, the expected life expectancy cannot be met.

Specifiers should request that manufacturers submit independent test reports from lightning labs confirming the published surge ratings.

Let-through voltage (or Clamping Voltage) is the amount of voltage that is not suppressed by the SPD and passes through to the load.

Let-through voltage is a performance measurement of a surge suppressor's ability to attenuate a defined surge. IEEE C62.41 has specified test waveforms for service entrance and branch locations. A surge manufacturer should be able to provide let through voltage tests under the key waveforms.

Installation is the most important factor in determining the effectiveness of a particular SPD.

Installation lead length (wiring) reduces the performance of any surge suppressor. As a rule of thumb, each inch of installation lead length adds between 15 to 25 Volts to the let-through voltage. Because surges occur at high frequencies (approximately 100 kHz), the leads from the bus bar to the suppression element creates impedance in the surge path .

Published let-through voltage ratings cover the device or module only. These ratings do not include installation lead length, which is dependent on the electrician installing the unit.

Therefore, the actual let through voltage for the system is measured at the bus bar and is based on two factors:

the device rating (quality of the suppressor)

the quality of the installation work

For example, consider an SPD with a 500 Volt rating. This is the true rating only if the SPD is integrated into the panelboard it is protecting. If it is connected to a panelboard with 14 inches of #14 wire, it allows approximately 300 Volts to be added to the let through voltage. The true let-through voltage at the bus bar is 800 Volts.

Figure 18. Installation Plays a Large Role in Determining Let-Through Voltage
(Based on IEEE 6KV 3KA Combination Wave)

Filtering eliminates electrical line noise and ringing transients by adding capacitors to the suppression device.

Filtering is often referred to as “sine wave tracking” or “active tracking.” These are marketing terms, and are not relevant to filter performance. Also, not all SPDs provide filtering. Many SPDs claim to possess sine wave tracking, sine wave contour, or EMI/RFI noise attenuation, but may not employ a quality filter.

With all the confusion on the subject, make sure the manufacturer can provide proof that the SPD meets these key filtering specifications:

MIL-STD-220A Insertion Loss Test
Attenuation at 100 KHz, measured in dB. A dB rating above 40 dB (at 100 kHz) reflects better performance. The higher the dB rating, the better the filtering.

Let Through Voltage - IEEE C62.41 Category B3 Ringwave
On a 120-volt system, L-N should be less than 200 volts.

Maintenance is not a requirement for a quality SPD. A quality SPD should last over 25 years without any preventive maintenance program.

The SPD should come with a diagnostic system that will provide continuous monitoring of:

fusing system and protection circuits (including neutral to ground)

open circuit failures

overheating (in all modes) due to thermal runaway

SPDs protects against surges, one of the most common types of electrical disturbances. Some SPDs also contain filtering to remove high Frequency noise (50 KHz to 250 KHz).

An SPD does not reduce harmonic distortion (3rd through 50th harmonic equals 180 to 3000Hz).

An SPD cannot prevent damage caused by a direct lightning strike. No device can. A direct lightning strike is a very rare occurrence. In most cases, lightning causes induced surges on the power line, which can be reduced by the SPD.

An SPD cannot stop or limit problems due to excessive swells (overvoltage). A swell is a rare disturbance caused by a severe fault in the utility power, or a problem with the ground (poor or non-existent N-G bond). A swell occurs when the AC voltage exceeds the nominal voltage (120 volts) for a short duration (millisecond to a few minutes). If the voltage exceeds 25% of the nominal system voltage, the SPD and other loads may become damaged.

An SPD does not provide back-up power during a power outage. An Uninterruptable Power Supply (UPS) is required to provide battery back-up power.

Most people think that the main cause of SPD failure is a direct lightning strike. This is not true. The number one cause of failure for SPDs is exposure to severe swells and overvoltage disturbances .

Under normal operation, the internal components of an SPD are designed to conduct a short-duration (microsecond or millisecond) surge to ground.

A swell (increased RMS voltage lasting a half cycle to a few seconds) or an overvoltage condition (swell lasting more that a few seconds) causes the SPD to conduct beyond specifications. The result is a reduction of life expectancy, or in severe cases, SPD failure.