CSB Publishes Sunoco Oil Terminal Investigation Report

Yesterday, the Chemical Safety Board published their report on the investigation on the 2016 flash fire and explosion at the Sunoco Nederland, Texas crude oil terminal. This was a ‘hot work’ incident where maintenance activities (welding) took place on a pipe that contained flammable liquids (crude oil). While there was an isolation device between the welding activity and the crude oil still in the pipe, the pipe volume where the crude oil was contained was not inerted (nitrogen purged for example).

Commentary

Interesting discussion about hot work practices, but the upshot is that the contractor did not follow well established hot work safety practices. Interesting point here, the post-accident OSHA inspection (see section 1.8 on pages 25-6) found the same issue and fined the company appropriately. While I have previously noted that post accident investigations by OSHA (and EPA, not applicable in this case) almost always find violations of safety regulations, they are seldom related specifically to the accident, that was certainly not the case in this incident.

On page 7 of the report, CSB notes that they conducted a “limited scope investigation”. What that means here is that there is no determination in this report what caused the ignition of the flammable gases behind the isolation device. While the isolation device ultimately failed, it should have certainly been adequate to prevent sparks and flame from coming in contact with the isolated crude oil.

It appears to me (with the standard caveat that I was not there, and only know what is contained in the CSB report) that probably happened is that the isolation device failed because of increased pressure in the isolated section of pipe due to conduction heating of the crude oil from the welding, not ignition of the material as described in the CSB report. If the pressure in the sealed pipe was sufficiently higher than the 25 psi used to seal the pipe (see the discussion about the CARBER isolation device on page 17) to cause that seal to fail then the flammable gas would have blown the device through the welding area providing an ignition source for the flash fire that resulted. This would have required some sort of failure of the pressure relief system CSB briefly described on diagram on  page 24.

I would have expected that a full investigation by the CSB would have determined how much pipeline pressure would have been required to cause the isolation device to move out of the pipe and then determine if the welding related heating of the pipe could have caused that pressure rise in the isolated section of pipe given the amount of crude oil present. That information would have been helpful because it would have also meant that inerting the atmosphere as required by OSHA regulations would not have prevented the incident. But the CSB only did a ‘limited scope inspection’ that uncovered nothing more than was discovered more quickly by OSHA. I am severely disappointed in this report.