Skip to content

Positive Material Verification: Prevent Errors

Overview

On July 28, 2005, the BP Texas City refinery experienced a major fire in the Resid Hydrotreater Unit (RHU) after a maintenance contractor accidentally switched a carbon steel elbow with an alloy steel elbow during a scheduled heat exchanger overhaul. The CSB issued a Safety Bulletin to emphasize process equipment configuration control and positive material verification of critical alloy steel piping components.

Incident Snapshot

Field Value
Facility / Company BP Texas City refinery
Location Texas City, TX
Incident Date 07/28/2005
Investigation Status The CSB released its safety bulletin on October 15, 2006. On July 28, 2005, 4 months after a devastating incident in the Isomerization (Isom) Unit that killed 15 workers and injured 180, the BP Texas City refinery experienced a major fire in the Resid Hydrotreater Unit (RHU) that caused a reported $30 million in property damage. One employee sustained a minor injury during the emergency unit shutdown and there were no fatalities.
Accident Type Oil and Refining - Fire and Explosion Investigation Status: The CSB released its safety bulletin on October 15, 2006. On July 28, 2005, 4 months after a devastating
Final Report Release Date 10/15/2006

What Happened

The RHU heat exchangers were placed in service in 1984. Scheduled heat exchanger cleaning and inspection were performed in 1989 and 1991. Elbows 1, 2, and 3 were removed, set aside, and then reinstalled following the maintenance.

The next scheduled heat exchanger maintenance overhaul began in January 2005. The three elbows were removed, stored temporarily, and then reinstalled 39 days later. In February 2005, a carbon steel elbow was installed in the high temperature, high pressure hydrogen line instead of the required 1.25 percent chrome low alloy steel elbow.

On July 28, 2005, at about 6:00 pm, an RHU hydrogen gas heat exchanger process pipe ruptured. The venting hydrogen gas ignited and a fireball erupted in the unit. The fire burned for two hours.

X-ray fluorescence testing confirmed that an alloy steel elbow was installed in the carbon steel elbow position on the heat exchanger B inlet. The CSB concluded that carbon steel elbow 1 was inadvertently switched with alloy steel elbow 3 when the maintenance contractor reassembled the piping during the winter 2005 heat exchanger overhaul.

Facility and Process Context

  • BP Texas City refinery is the third-largest in the United States with a capacity in excess of 450,000 barrels per day of crude oil.
  • More than 1,600 BP employees and hundreds of contract personnel operate and maintain the facility.
  • Residual material from the crude oil processing unit is processed in the RHU to remove nitrogen, sulfur, and metals.
  • Hydrogen is pressurized to about 3000 psi and pre-heated in the RHU heat exchangers to about 600oF.
  • The RHU has three parallel operating systems.
  • Each system contains a heat exchanger assembly that consists of two series-connected heat exchangers to preheat the hydrogen.
  • The RHU heat exchangers were designed in the early 1980s.
  • The BP Texas City refinery has a material verification program, and PMI test equipment that quickly differentiates carbon steel from alloy steel piping components.

Consequences

  • Fatalities: 0
  • Injuries: One employee sustained a minor injury during the emergency unit shutdown.
  • Environmental Release: There were no offsite impacts.
  • Facility Damage: The RHU sustained major damage from the hydrogen-fed fire; reported $30 million in property damage.
  • Operational Impact: The unit burned for two hours and Texas City ordered a shelter-in-place for nearby residents until the fire was contained.

Key Findings

Immediate Causes

  • The 8-inch diameter carbon steel elbow inadvertently installed in a high-pressure, high-temperature hydrogen line ruptured after operating for only 3 months.
  • HTHA severely weakened the carbon steel elbow.
  • The carbon steel elbow was installed instead of the required 1.25 percent chrome low alloy steel elbow.

Contributing Factors

  • The maintenance contractor was unaware of the material differences in the elbows.
  • BP did not require the contractor to implement any special precautions to prevent inadvertently switching the elbows.
  • BP did not require any post-reassembly testing to confirm the alloy elbows were reinstalled in the correct locations.
  • Because carbon steel and low alloy steel are visually indistinguishable, special test equipment is needed to distinguish the two low alloy steel elbows from the carbon steel elbow.
  • The BP procedure requires alloy steel components to be verified when they are received in the warehouse and when alloy steel components are shipped from the warehouse for use in new construction, but it does not require PMI during maintenance.
  • Lacking post-installation PMI testing, or positive identification of the alloy steel components before and after installation, the maintenance crew’s reassembly error went undetected until the pipe failed.
  • The maintenance contractor did not adequately control the disassembly, storage, and reassembly of the piping components to ensure that all the piping components were returned to their original installed locations.

Organizational and Systemic Factors

  • Piping systems can be designed such that incompatible components cannot be interchanged.
  • All three elbows could have been made from the same low alloy steel material, even though this would have meant additional material expense.
  • Alternatively, elbow 1 could have been dimensionally different from elbow 2 and 3, although this would have meant additional construction costs.
  • BP did not alert the maintenance contractor that two of the three elbows were alloy steel piping components and must not be interchanged with the carbon steel elbow.

Failed Safeguards or Barrier Breakdowns

  • The BP procedure did not require PMI during maintenance.
  • No special precautions were required to prevent inadvertently switching the elbows.
  • No post-reassembly testing was required to confirm the alloy elbows were reinstalled in the correct locations.
  • The maintenance contractor did not adequately control the disassembly, storage, and reassembly of the piping components.
  • The crew did not clearly label or mark each elbow before removing them, then confirm that each had been reinstalled in the correct location.

Recommendations

  1. 2005-04-B-R1 | Recipient: BP Texas City Refinery | Status: Not specified | Revise the maintenance quality control program to require positive material identification testing or another suitable material verification process for all critical service alloy steel piping components removed and reinstalled during maintenance, and inform work crews of special material handling precautions.
  2. 2005-04-B-R2 | Recipient: JV Industrial Companies | Status: Not specified | Develop/update the written piping component installation quality control procedure to require positive material identification testing or other suitable verification or tracking process for all alloy steel piping components removed during maintenance.

Key Engineering Lessons

  • Positive material identification or another suitable verification process is needed for critical service alloy steel piping components removed and reinstalled during maintenance.
  • Piping components and their respective locations should be tagged or marked before removal, and the correct installed location should be verified after reinstallation.
  • Maintenance procedures should account for the fact that carbon steel and alloy steel can be visually indistinguishable.
  • Configuration control during maintenance is necessary because disassembly and reassembly can create unacceptable system modifications.

Source Notes

  • Consolidated from the provided structured extract for Safety Bulletin: Positive Material Verification (source_priority 4).
  • No higher-priority source documents were provided in the input.
  • All facts were taken directly from the supplied extract; no external information was added.

Similar Incidents

Incidents sharing the same equipment, root causes, or hazard types.

Same Equipment

Same Root Cause

Same Hazard

← View in Knowledge Graph