By nature, power failure recovery testing will subject the systems under test to the risks associated with an improper or unintended response to the loss or reapplication of power. The most common places for damaging deficiencies to be encountered are:
n Localized outages where the outage affects a component or subsystem but not the entire system.
For example, a power outage to the interlock circuit controlling the outdoor air dampers on a 100% outdoor air system that does not impact the fans could subject the intake plenum to extreme negative pressures unless proper provisions had been made in the design to protect it via permissive interlocks, limit switches, pressure relief doors, etc. If the outage affects the fans and the damper interlock circuit, the potential for trouble is significantly mitigated.
n Outages where portions of the building and its systems are recovered by emergency power while others remain off line until the return of normal power.
If the fans in the preceding example were served by emergency power, but the damper interlock wasn’t (either intentionally or accidentally), problems could ensue.
n Recovery from a power outage.
Problems at this point are usually the result of an uncoordinated reapplication of power either because the power is restored to different areas at different times or because control software and programming have not been properly coordinated to ensure a smooth recovery.
Even in a well designed and fabricated system, power failure recovery testing will probably uncover deficiencies. In retrospect, the deficiencies will seem obvious and sometimes they are. But in many instances, they are subtle and the words “Darn, I never would have imagined that could happen” are frequently uttered in the field as the impact of the deficiency is assessed and repaired. In most cases there will be less risk associated with a properly managed and coordinated testing process in contrast to a “test” triggered by natural phenomenon without benefit of previous manually triggered and monitored functional testing and tuning. Manual testing will be conducted by knowledgeable individuals who have anticipated that things might go wrong; i.e. they will be anticipating specific problems and also anticipating that there could be some things they might not have anticipated. Since they know the systems and are familiar with the weak points, they can be prepared to act before things get out of hand, even if the specific failure was not foreseen. For example, a team testing a large, 100% outdoor air handling system served by normal and emergency power sources would most likely be wary of problems that could collapse the intake plenum. They would have ideas about specific things that could go wrong (dampers not on emergency power, sequencing problems, limit switch adjustments, etc.) But they would also know that something they hadn’t though of could happen. As a result, they would react to any sign of impending disaster and stop the test even if the reason for the problem was not immediately apparent. Once the equipment was in a safe state, they would figure out why it happened. Generally, nature would just continue the test despite signs of trouble.
Additional discussion on this topic can be found in Chapter 2, Section 2.12. General Precautions and Preparations.