A failed fire sprinkler system

We were asked to investigate the cause of a stand pipe connection failure in the fire protection and sprinkler system of a major high rise building, located in Los Angeles. The failure occurred at the penthouse level when a section of horizontal 6 inch diameter pipe pulled out of a “Gruvlok” coupling.   As you would imagine, an open 6 inch pipe puts out a lot of water.   Damage was extensive-was it caused by a faulty installation, a manufacturing defect, or some other cause?

The sections of the pipe on both sides of the failed coupling and the “Gruvlok” coupling were examined.   Corrosion was not an issue.   The dimensions and location of the rolled grooves which retained the coupling in place were within the manufacturer’s specifications.   The groove diameter was 0.013 to 0.043 inches smaller than required (making the groove depth greater), and the groove width was .005 to 0.015 inches wider than required.   These discrepancies did not affect the integrity of the joint.   The plastic, rubber-like gasket that was installed inside the “Gruvlok” coupling did not show any “pinching” or other distortion that would indicate that it had been installed incorrectly.

The pipe line had been spray painted after the installation.   The marks of the hanger that were not painted were used in identifying the correct orientation of the pipes and couplings.

The presence of paint in the bottom of the grooves was not an indication of faulty installation.   The coupling was in two halves that were bolted together around the plastic, rubber-like gasket and then tightened.   There was an intentional gap between the metal coupling and the bottom of the groove on each end of the pipe.   This gap was to allow for movement and misalignment of the two sections of the pipe being joined by the coupler.

If a coupling is incorrectly assembled, the mistake is usually discovered the first time the system is filled and pressurized.

When the fire protection piping system was installed, it was inspected and hydrostatically tested for a minimum of two hours at 335 psi.   This test was passed without leaks and a certificate of occupancy was issued .   The system did not leak for over five years.

The term “water hammer” is used to define destructive forces, pounding noises and vibration which develop in a piping system when a column of noncompressible liquid flowing through a pipe line at a given pressure and velocity is stopped abruptly.   Tremendous forces generated at the point of impact or stoppage can be compared, in effect, to that of an explosion.   To quote Wikipedia, “Moving water in a pipe has kinetic energy proportional to the mass of the water in a given volume times the square of the velocity of the water.”

If a pipe is suddenly closed, the water is still moving and builds up a high pressure shock wave.   Quoting Wikipedia again, “In domestic  plumbing this is experienced as a loud bang resembling a hammering noise.   Water hammer can cause pipelines to break or even explode if the pressure is high enough…In the home water hammer often occurs when a dishwasher, washing machine, or toilet shuts off water flow, resulting in a loud bang or banging sound.”

When water hammer occurs, a high intensity pressure wave travels back through the piping system until it reaches a point of relief, such as a large diameter riser or piping main.   The shock wave will then surge back and forth between the point of relief and the point of impact until the destructive energy is dissipated in the piping system.

In commercial applications the common cause of water hammer is the quick closing of electrical, pneumatic, spring-loaded valves, as well as quick hand closure of valves.   The speed of the valve closure time, especially during the last 15% of the valve closure, is directly related to the intensity of the surge pressure.   The resultant water hammer shock wave travels back and forth in the piping system at a rate of 4,000 to 4,500 feet per second.

Although noise is generally associated with the occurrence of water hammer, water hammer can occur without audible sound or noise.  Quick closure always creates some degree of shock, with or without noise.   Therefore, the absence of noise does not indicate that water hammer or shock is nonexistent in a water distribution system.

To return to the story, three days before the coupling failure, the fire protection and sprinkler systems were recertified by a new vendor.   This required testing of the fire pump and the flow of water through the various pressure-reducing valves.   If you are in a hurry, it is difficult to slowly close valves.

Our conclusion was that during the testing of the fire protection system, overly aggressive valve closures created a water hammer condition that stressed the standpipe to an early failure.   The failure was not caused by an incorrect installation or defective materials.