The video the CSB just released shows in such great detail what happened and how maintenance confusion and training caused the massive fire. This actually fits into a valve and a flange issues all in one. Not realizing that the “cap” actually is a pressure boundary to the top of the valve is a major oversight and specific training would of assisted in avoiding this happening in the future.
One issue that is not addressed in the video is the root cause of this being the failure of the actuator. It is hard to tell if the issue was the gear box or the packing friction could of been too large to actuate with the hand wheel. When the final report is released there will be more details on what actually happened.
Been reviewing this fire at a Refinery in Corpus Christi, Texas a few years ago where the Chemical Safety Board investigated. One disturbing point they found was a flange had been having a slight leak for 6 months and had not been address barely at all! Quote from CSB report “In late January of this year, maintenance was performed on the flange, tightening the existing bolts, but the leak persisted. Further maintenance was performed on February 10 – over three weeks prior to the actual incident. At that time workers replaced the flange bolts and a work order was submitted to order a clamp to enclose the leak.”
What I find interesting is they did not mention the gasket anywhere! It seems to me they first did a “re-tightening” effort that might of been with a torque wrench but as you can see in the photo there were NO FLAT WASHERS! Also hard to see any anti-seize on the bolts but hard to say. Either way, the torque would have some errors. Second it seems they might of replaced bolts but kept the gasket in place since they did not bring the system down.
In summary, a flange leak in one of the most dangerous parts of the plant was left to leak and eventually caused massive damage to the plant. leaks are not something to take lightly in a refinery!
The CBS Letter here.
Report on CSB news conference here.
I listened to a fantastic podcast yesterday about the oil industry and changing the macho culture. the story focuses on Shell oil was building a world class off shore platformed (the Ursa, pictured above) and was worried about how technical complex and dangerous it was. They wanted to make sure that compuciation between everyone was a peak levels to lower injuries and death.
They turned to the obscure physiological technique called EST where you force people who work together to share their life experience and break down barriers to help them communicate better. The experiment was well documented and showed massive drops in safety of 84%.
I have been to many similar places and have seen first hand this inward non cooperative behavior and could really see how these techniques could help. Also I could see how this is a generational thing that the millennials focusing more on coopartive work would get more out of this.
I thought this was such a good quote because I have seen it so many times in my 27 year experience working with the Nuclear industry. People often overlook technology like valves and heat exchangers and taking them for granted – this non focus can sometimes end badly an even death. Also I like how Rickover calls out the supply chain of such components and how manufacturing errors in workmanship and materials can cause the entire system to fail. Procurement can be looked down upon by engineering sometimes but their role is also needed for true safety in the nuclear industry.
I was digging a little deeper and found an article from the Chicago Tribune on this speech from Rickover.
Another video that shows how people neglect procedures in place because they seem to be burdensome but end up causing massive damage, and in this case death. The story also highlights how “tribal” knowledge can end up not being captured. I feel when the changes were made no one thought they would need to explain how the system now works – and they would never recommend opening the heat exhanger without having the vent engaged, but, like many times, that information got lost in translation to paperwork.
I think something like having a lockout tag on the 2nd valve would of been noticed if someone did a thorough MOC but, as stated in the video, some questions were left blank.
I have been thinking of this accident in Canada. Not only was the issue with putting the hand brakes on with the air brakes on already as a root cause, but also the leaks in the air line that must have lowered the air pressure in the system. Air leakage is accepted as normal and no one really spends much time fixing but this accident should reflect on how something so simple as an air leak can cause a disaster.
Fittings around air piping are a typical air leak problem. Simple checks and using higher grade tape could of assisted. I am not the only person to wonder http://about this – here is a great article with more depth of the issue. One good quote from this article:
“Hand Brakes should have been redesigned many years ago, to make them easier to use and more likely to work properly. The problem is, no one is pushing for it. Costly air leaks should have been addressed and improved, but nobody is pushing for that either.
It’s difficult to explain in this format just how much leakage the railway industry is OK with. Locomotives are equipped with an air flow sensor that shows the engineer how much air is flowing/leaking. Get the picture? Leaks are OK, we just have to manage them. Our Lac Magentic engineer was driving the train, so he knew what the leakage was on that train”