479487 Dangers of Hot Work: Lessons Learned from Fatal Welding Operations on a Gasoline Storage Tank

Tom DeBold, GexCon US, Bethesda, MD, Bryant Hendrickson, Gexcon US, Bethesda, MD and Scott Davis, GexCon US Inc., Bethesda, MD

An explosion occurred on May 12, 2009 at a mid-stream storage terminal facility that fatally injured three welding contractors. The welding contractors were hired to install a gage pole inside a floating roof storage tank, which involved cutting into the floating roof with an oxy-acetylene torch. At approximately 2:20 PM, the explosion occurred, blowing the geodesic dome roof off of the tank where it fractured into multiple sections, splitting the top plate of the floating roof from the bottom and propelling it up and out of the tank, deforming the bottom section of the floating roof towards the floor and northern section of the tank, deforming/collapsing the sidewalls of the tank, and killing three of the welding employees. These men were on the floating roof at the time of the explosion and were blown out of the tank.

The storage tank had been last used to store unleaded fuel, and, according to operations staff, the tank was drained of fuel a week or two before the incident occurred. The work plan for the gauge pole installation included several steps: (1) isolate the tank from operations; (2) clean the tank such that it was ready for hot work; and (3) install the gauge pole. The tank was isolated from operations almost 10 days prior to the incident. This procedure entailed stripping the tank and the lines feeding the tank, installing skillets, and removing manway covers. In addition, a fan was installed in the north manway to aid in removal of vapors and drying the tank. This fan was operated from May 1 to May 5. The second step of the procedure, cleaning the tank, was completed on May 8-10, 2009. The third step of the installation procedure, the actual installation of the gauge pole, was contracted to welding contractors.

On the morning of the incident, oxygen and lower explosive limit (LEL) readings were taken inside the tank. The tank was inspected earlier that morning at 7:00 AM and found that the oxygen level was 20.8% and that the LEL was 0. A confined space entry permit and a hot work permit were issued stating that the atmosphere in the tank had been tested. Despite the necessary permits, the explosion occurred when the contractors began cutting into the floating roof. This paper will discuss the investigation findings and lessons learned from this devastating, yet preventable incident.


Extended Abstract: File Not Uploaded
See more of this Session: Case Histories and Lessons Learned - GCPS Joint Session
See more of this Group/Topical: Global Congress on Process Safety