EXCELLENT Presentation from the CSB on
the West Fertilizer Exploison
Friday morning (9/23/16) the United States Court of Appeals, For the DISTRICT OF COLUMBIA CIRCUIT announced that OSHA could not redefine "retail" using a Letter of Interpretation (LOI) or Memorandum, OSHA would have to go through the Notice of Public Rule Making (NPRM) in order to redefine the term. Ironically, it was a 1995 LOI that actually defined "retail" and yet no lawsuit was brought forth on that LOI and it stood for 20+ years. The court even acknowledges that the "retail" exemption was to be applied to "small containers" and true retail businesses; but in the end, the court sided with the complainants (AGRICULTURAL RETAILERS ASSOCIATION and the FERTILIZER INSTITUTE) and said the Memorandum was essentially a standard. Here is the cleaned up version of the decision, the actual decision can be downloaded...
This is an interesting outcome, of which I have no knowledge as to how this deal was struck. But there was no monetary penalty and the facility has 60 days to make some seriously big changes/improvements. The biggest thing I noticed is that EPA is requiring the facility to make changes to comply with the latest version of CGA 2.1 - 2014 rather than the RAGAGEP the process was most likely built to: ANSI K61. Here are the facts as we know them...
In the last 10 years, RMP data show that there have been more than 1,500 reportable accidents, about 500 of which had off-site impacts. These accidents are responsible for nearly 60 deaths, some 17,000 people were injured or who sought medical treatment, almost 500,000 people evacuated or sheltered-in-place and more than $2 billion in property damages. Source
UPDATED on 9/26/16 - as more information trickles out from the investigation I will continue to post. I still have not been able to get the actual report, but this morning we learned...
Among the safety equipment that was not functioning, was the refrigeration compressor and condenser for the tank had been dysfunctional for three (3) years. Two (2) pressure gauges on the tank that are used to calculate the pressure inside the tank had been dysfunctional for "long". In addition, the two (2) pressure vents of the tank - which are used to release gas in times of excessive pressure inside the tank - were dysfunctional during the explosion. We have also learned that the fire water loop for the fire protection system at the plant was not operational at the time of the incident.
We are also hearing that the 10-year old tank had some MI issues that were identified 3-years ago and not addressed. For some reason, the tank walls experienced severe thinning over its VERY short 10-year life! The wall thickness was suppose to be 16-18 mm (0.6") and was found to be 8 mm (0.31"). The sister tank has said to be found within spec. Personnally this wall thickness is not what caused this accident and may have been a "saving grace" as the tank failed at a lower pressure! The Plant Manager has admitted the plant never went down for annual mantenance since 2005.
Absolutely it can, and allow me to explain how. We start our Risk Management Plan (RMP) program(s) level assessment with Program 1, then if our process(s) does not qualify for Program 1 we ask ourselves if our process(s) would be a Program 3, and then if the process does NOT qualify for Program 1 or 3, then our process(s) will fall into Program 2 by default. So we look at Program 1 first, then Program 3 and if neither fit then we are Program 2 by "default". But somewhere a nasty rumor got started and many facilities began to believe that if the process was already covered under OSHA's Process Safety Management standard there was no need to perform this Program Level Assessment and that the process would automatically be a RMP Program 3 process - this just is NOT how it works.