Author -

Cubicle Explosion

Line crew was dispatched to change out a cubicle (U-2732). Upon arrival the crew learned that the cubicle to be changed, had blown up. The cubicle doors were found ~75' away, and the lid was reported by an eyewitness to have shot ~100' in the air before landing onto the cubicle's side panel. The explosion was of such magnitude that it caused this amount of damage. An employee from the claims department suggested that this could be a natural gas explosion. The crew thoroughly investigated the area and found no odor that would indicate to them that this was a natural gas explosion. The claims employee talked to a process coordinator and said that PG&E should come out and investigate the area for a gas leak. The process coordinator went to a system operator and asked if PG&E were on their way out to the scene. The system operator was speaking to a troubleshooter over the radio that said that the Claims employee was calling PG&E to come out. The system operator's answer to the process coordinator was yes. Note: The Claims employee and the troubleshooter were at the scene of the incident at the same time and communicating to two different people. Both thought that "the other guy" was calling PG&E. When in fact PG&E was never called. Process coordinator said to go ahead and change the cubicle out, thinking PG&E was on their way to investigate a potential gas leak. The foreman after investigating the scene and getting word to change the cubicle out reported on the circuit and the crew grounded the conductor and changed out the cubicle. There was no damage to the cable; it was a straight change out. After changing out the cubicle the foreman called dispatch to report clear of the circuit dispatch reported foreman clear and gave switching orders. Note: Dispatch thought that PG&E were called to the scene and that the foreman had communicated with PG&E and everything was resolved. When dispatch reported the foreman off the circuit the foreman assumed that it was okay to energize the cubicle. Note: This would be a normal assumption since the foreman received permission to change out the cubicle form the process coordinator. The foreman then performed the switching orders that hotted up the new cubicle. Part of the switching orders was to close the local fuses that would pick up load. The crew then closed a fused disconnect. As soon as the disconnect was closed the cubicle ignited and the cubicle well was on fire. The crew put out the flames by using a fire extinguisher. After the fire was out a crew member went up to the cubicle well to investigate what happened, when kneeling down to look inside the crew member felt heat from the edge of the cubicle well. Looking closely, the crewmember discovered a small flame that was burning along the outside edge of the cubicle well. The crew put dirt onto it and extinguished the flame, however they tested to see if this was a gas leak by using a torch to ignite the flame. They discovered they had a gas leak and reported the situation to dispatch, dispatch then called PG&E. When PG&E arrived they were able to determine it was a gas leak by using a "Sniffer" that detects natural gas leaks. A 2" gas line ended ~15' from the cubicle, the "End Cap" at the end of the gas line broke and the gas leaked. PG&E made permanent repairs to the gas line. Note: When talking with PG&E it was explained to us that when soil conditions are damp the odor that PG&E puts into the gas so it can be detected by smell becomes filtered out through the wet soil. This was verified with SMUD's Gas Control people. This is why the crew did not smell any gas while changing out the cubicle.

The seriousness of this near miss is a potential Fatality to several employees. The above chain of events shows that there was a lack of communication from the workers and control. This line of communication from control and crew, is in every case, vital to maintaining levels of safe operations to its highest potential.