Showering Arc Testing of Electrical Products

Jerry Ramie

Products headed for utilities and other industrial settings need to withstand the electrical noise produced in these harsh environments.

When mechanical switches open to interrupt the currents flowing into inductive loads, arcing across the switch contacts occurs. Repeated ignition and quenching of the arc's plasma is typical in this instance. The showering arc test was developed in the United States and Europe to simulate this environment's effects on power and input/output (I/O) control and monitoring lines. These environments include those found in utility power switching, industrial installations, or other high-reliability applications (transportation, military, aerospace, etc.).

Industrial automation and control products used in power generation and substation environments in the United States are addressed by the National Electrical Manufacturers Association (NEMA) in its Industrial Control and Systems (ICS) series of standards. These standards encourage the production and testing of reliable, safe equipment for the power industry. The ICS series of standards consists of the nine standards shown in Table I.

The ICS 1-2000 (General Requirements) standard was developed by the Industrial Automation Control Products and Systems section of NEMA. It addresses many aspects of the manufacturing and testing of industrial and power automation products. Section 8 provides performance requirements and tests. Section 8.10.2 states that "manufacturers of industrial control equipment establish performance criteria for each of their products and conduct the checking they consider necessary to provide assurance that products shipped meet these performance criteria."1

Table I. NEMA Industrial Control and Systems (ICS) standards.

To provide assurance that solid-state devices will survive and work as intended in these high-reliability but harsh industrial environments, Annex E of ICS 1-2000 suggests that electrical noise tests be performed on the solid-state logic, I/O modules, and power supplies used on these types of equipment. These tests "are intended to detect whether noise signals injected into the power supply and input and output wiring have penetrated the isolating means afforded by these devices to a degree that would cause malfunction of logic gates." With ever-increasing demand for electrical energy in the United States, many companies that provide equipment into this market will be required by their customers to perform the showering arc testing described in Annex E.

Dolan Labs at American Electric Power (AEP; Columbus, OH) states on its Web site:

Figure 1. Oscillating-polarity waveform.

The showering arc test is essentially an electrical noise susceptibility test. A NEMA standard noise generator is used to perform the test. The test set generates broadband electrical noise via an arcing spark gap, and couples the noise onto individual conductors within a multiconductor cable. Conductors are then used as input/output paths for the device under test. The test is designed to test logic input and output circuits, excluding low-level logic such as TTL, and is appropriate for devices with solid-state control input and output circuits such as [programmable logic controllers].2

A more precise description of electrical noise is given in NEMA ICS 1-2000 and in IEEE C37.90.1-2002, "Surge Withstand Capability Test for Relays." Both utility equipment standards describe two threat phenomena: fast-rise repetitive waveforms and oscillatory waveforms.3 Typical fast-rise repetitive (single-polarity) and oscillatory (oscillating polarity) waveforms are shown in Figures 1 and 2.

Figure 2. Single-polarity waveform.

These threats are delivered by two separate showering arc generators, which are shown in Annex E of the NEMA standard. The schematic for the single-polarity generator is shown in Figure 3, and the schematic for the oscillating-polarity generator is shown in Figure 4. Both showering arc generators use 3-kV luminous-sign transformers to generate high peak voltages with 10 mA of current.

The oscillating-polarity showering arc generator applies the transformer's secondary 60-Hz high-voltage sine wave across a spark gap, and the single-polarity showering arc generator rectifies the transformer's secondary output and applies the resulting dc voltage across the gap. In either case, the physical dimension of the spark gap is controlled with a lever-reduced micrometer to allow adjustment of the ionization potential (arc-overvoltage or distance) across the gap. This mechanical adjustment sets the amplitude of the test voltage delivered from the generator to the multiconductor cable coupler.

Figure 3. Single-polarity schematic.

Before testing can begin, a verification test of the delivered voltage and induced current into nearby loaded conductors must be run according to Clause E5, "calibration procedure for noise generator and coupling cable assembly." This procedure requires that the generator be sourced into opposite conductors within the cable. The loading resistors and current probe are to be connected as shown in Figure 5.

The voltage (spark-gap dimension) is set to 1500 V, and the current probe reading must indicate that 6 A of minimum current were induced into a nearby pair of conductors to be used for the remainder of the tests. Figure 6 shows the showering arc generators, calibration and spark-gap assemblies, and the coupling-cable assembly performing this calibration.

Figure 4. Oscillating-polarity schematic.

A loom of 15-conductor cable is wound onto a form to make the coupling-cable assembly, which is essentially a transformer used for coupling the generator's transients onto the equipment under test (EUT) input, power, and output lines. After the generator's voltage and minimum induced current are validated, the coupling-cable assembly is used as the path for routing input circuits, output circuits, and power supply circuits to the EUT.

For each type of showering arc generator, the test is run for 1 minute each on the EUT's power, input, and output lines. This procedure provides a total of six tests, using three connections and two arc generators. The EUT must not change state (relay drops out or chatters) in order for the product to have passed the test.

Figure 5. Connection diagram for coupling-cable assembly standardization.

Although the acceptance criteria above may sound somewhat arbitrary, it has worked for many years for testing utility equipment used in generation and substation plants. The generators' delivered energy, voltage ring, or overshoot may not be fully specified, but users report correlation between good performance on the test and reliability of the equipment in the field. However, some questions arise. The standard's minimum current waveform is typically exceeded by 25% during the test, but shouldn't reliable utility products be overtested? A spark gap is subject to wear, and the amplitude stability of the generator's voltage changes with the gap's dimension. Ionization sputtering will increase the gap dimension, hence raising the test voltage. Is that going to cause the EUT to fail? It shouldn't be that close to failure, should it? The coupling- cable assembly uses #22 wire, which limits its current capability and exposes inattentive operators to dangerous voltages.

Conclusion

Despite these shortcomings, showering arc testing is necessary. The nature of arcing switches on inductive loads in high-power utility settings is always somewhat unpredictable. A real-world power-switching setting is very noisy and random. It is recognized as a mixture of fast-rise-time repetitive and oscillatory events. This could certainly describe the threats delivered by the showering arc generators and coupler shown in Annex E.

Figure 6. Calibration setup.

Power and industrial control customers will continue to use the NEMA ICS 1-2000 standard as a benchmark or figure of merit for the equipment they purchase. Manufacturers want to sell equipment into the growing power market, and showering arc testing to NEMA ICS 1-2000 will be required to succeed. The availability of commercially built test equipment for this standard has improved, and a number of laboratories provide showering arc testing.

In the United States, many products sold to the power industry are currently tested for showering arc immunity to repetitive and oscillatory threats. In Europe, tests to address these threats are specified in the EMC product standards IEC 60255-22-4 (burst or single polarity) and IEC 60255-22-1 (1 MHz ring-wave or oscillatory). The standards specify measuring relays and protection equipment. It is clear that utility products bound for either market will require testing for immunity to these threats.4,5

References

1. NEMA ICS 1-2000, "Industrial Control and Systems, General Requirements," National Electrical Manufacturers Association (NEMA), Rosslyn, VA. Available on Internet: http://www.nema.org.

2. "American Electric Power, Electromagnetic Compatibility & Qualification Testing," Dolan Laboratory, Columbus, OH. Available on Internet: http://www.aeptechcentral.com/emcqt.htm.

3. IEEE Std C37.90.1-2002, "IEEE Standard for Surge Withstand Capability (SWC) Tests for Relays and Relay Systems Associated with Electric Power Apparatus," IEEE Power Engineering Society, IEEE, New York City.

4. IEC 60255-22-4 (2002-04), "Electrical Relays-Part 22, Electrical Disturbance Tests for Measuring Relays and Protection Equipment- Electrical Fast Transient/Burst Immunity Test," International Electrotechnical Commission (IEC), Brussels, Belgium.

5. IEC 60255-22-1 (2002), "Electrical Disturbance Tests for Measuring Relays and Protection Equipment-Section 1: 1 MHz Burst Disturbance Tests," IEC, Brussels, Belgium.

Jerry Ramie is president of ARC Technical Resources, a provider of EMC test equipment in San Jose, CA. He can be reached at 408-263-6486 or at http://www.arctechnical.com.