Handling a nuisance trip
The main 4,000 amp service breaker associated with a research and development facility has experienced a number of unexplained trips.
View the full story , including all images and figures, in our monthly digital edition .
The main 4,000 amp service breaker associated with a research and development facility has experienced a number of unexplained trips. A power monitor was installed on the service to determine if the cause of the circuit breaker's operation was the result of a fault condition in the building's electrical distribution, a utility supply issue, or a problem with the electronic control circuitry (trip unit). Because the building was undergoing a significant remodel, the general consensus was that there was a wiring problem introduced during the construction activity.
A Dranetz-BMI PowerGude 4400 (PG4400) power monitor was connected to the load side of the 4,000 amp circuit breaker. Voltage connections for phases A and C were connected to the utility supply side of the breaker, and phase B was connected to the load side; this configuration ensures that the event capture anytime the breaker trips. LEMFlex RR3035 current probes were installed on the respective phases and neutral to document current levels at the time of event capture. The PG4400 was configured for cross-channel triggering, i.e., any event trigger will result in all voltage and current changes being recorded.
Three significant events were recorded during the monitoring period. A graphical summary of the events is shown in Figure 1. Two of the events were associated with a trip of the problem breaker. The other event was a utility sourced sag, most likely the result of fault clearing operations somewhere in the utility distribution system. The following is a summary of the significant events:
Event #104 on Tuesday at 8:13 a.m. (see Figure 2)
Event #687 on Saturday at 6:53 a.m., first breaker trip (see Figure 3)
Event #808 on Sunday at 2:12 a.m., second breaker trip (see Figure 4).
In addition to the three significant events, the monitor data revealed other sag-related events, one utility sourced and, most likely, two load turn-on events. These events were not sufficient to cause an event trigger of the PG4400. The following is a summary of these lesser events:
Utility sag between 1:30 and 1:40 p.m. on Wednesday
Load sag between 12:30 and 12:40 p.m. on Thursday
Load sag between 4:30 and 4:40 p.m. on Thursday.
Utility sourced voltage sag
The voltage sag shown in Figure 2 is typical of a utility fault clearing operation. The duration is five to six cycles, and both the voltage and current decrease indicating a utility sourced fault. Phase C was the most impacted and decreased to approximately 241 Vac during the sag. Most rotating equipment will ride through these types of events, but some electronic loads (tools) may respond the short duration voltage sag.
Typical solutions for equipment impacted by these types of events would be some type of voltage regulation or voltage support (uninterruptible power supply or some other type of energy storage device).
Circuit breaker operation
Figures 3 and 4 show the circuit conditions at the time of the two breaker operations (trips).
In both cases, the pre-trigger waveform data show no abnormal current surges that would account for the operation of the breaker either from a phase-to-phase or ground fault condition.
Zero sum calculations of the monitoring data do not show any significant current that could account for the breaker's unplanned operation. It is as though it (or someone) just turned the breaker off.
Based on the data from the two recorded breakers operations, the most likely cause is the breaker's electronic trip circuitry.
The suspect 4,000 amp circuit breaker is equipped with additional control/diagnostic electronics that can provide state and fault conditions on breaker trips. The additional electronics requires an optional power supply to provide the additional functions. The circuit breaker will operate without the additional power supply, but will not provide any state or fault conditions information in the event of a breaker operation. In this case, the 4,000 amp circuit breaker was not equipped with the optional power supply.
The circuit breaker and associated trip circuitry were tested and found to be in satisfactory operating condition. Discussions with the manufacturer's applications support group revealed that using the circuit breaker, without the optional power module, under low load conditions with harmonics present can result in nuisance tripping.
The breaker in question was carrying 300 to 400 amps of current at the time of the breaker trips. Also, from the current waveform, it is apparent that harmonics are present.
Figure 5 shows the current time plot versus the third, fifth, and seventh harmonics. Both of the trips coincided with lower current levels and slightly elevated harmonic levels. The fifth and seventh harmonics are actually increasing as a percentage of the fundamental, which is a function of the types of loads remaining in operation. Apparently, the unpowered control circuitry responds under these somewhat unusual conditions.
The breaker had been in service for quite some time without an incident, and the low current (load) levels were the result of the remodeling activity. This appears to be an example of “Murphy's Law”: if anything can go wrong, it will.
Operating a 4,000 amp circuit breaker at only a few hundred amps is an unusual situation brought about by the remodeling/construction activity, coupled with some load generated harmonics—not all that unusual—and the decision not to implement the breaker's electronic features on the original installation. These factors created the “perfect storm” for this situation to develop.
Once the problem was identified, then the most cost-effective solution would be to install the power supply for the circuit breaker's electronics, which should resolve the problem.
Lonie is president and a co-founder of PowerCET Corp., Santa Clara, Calif. Previously, he was vice president of ONEAC Corp. and product support manager for the telecommunications division of ROLM Corp. His experience with GTE included operations, customer support, and engineering management positions. A recognized expert in the evaluation of power and grounding problems for the data and telecommunications industries, Lonie has written two books and regularly publishes articles on power quality and the electromagnetic environment. He is co-publisher of PQToday and serves on the editorial advisory board for Power Quality Assurance Magazine. He is a graduate of the University of Redlands, a member of IEEE, and has more than 30 years of experience in computer and telecommunications service and equipment protection.