This incident occurred in an Oil Refinery, located at Feyzin, near Lyons in France. The date was 4 January, 1966, and being winter, temperatures were said to be around 0° C at the time of the occurrence (which had an important bearing on the incident). In the early hours, around 06.40 hrs, an operator, who was attempting to drain water from a Propane sphere could not do so, as the water had frozen. He opened both drain cocks fully, and suddenly, the liquid propane began gushing out. Somehow, the operator was not able to retrieve the cock key, which had dropped off, and he left the scene to get help. As the cock was now fully open, liquid propane was pouring out of the drain. There was almost no wind and the liquid began quickly vaporising, forming a large vapour cloud. The capacity of the propane sphere was 1200 M, and it was said to be around 3/4" full at the time prior to the incident.

Approximately 35 minutes after the leak began, around 7.15, the vapour cloud ignited, possibly due to a car which was travelling on a nearby service road. There was now a fire just below the sphere, and shortly the safety valve of the sphere blew, resulting in a vertical flame around 30 mtrs in height. About 1½ hours after the fire had started, the sphere ruptured (bleved refer section 6.3.1), resulting in a huge fireball believed to be around 300 mtrs in diameter. The extreme heat which evolved resulted in the instant death of 17 people, 11 of them fire service personnel. Around 80 people were injured, and approximately 40, severely burnt. In the resultant fire, four other spheres also ruptured, and a number of other crude and product tanks caught fire. The conflagration took over 48 hours to control.

Feyzin Refinery Disaster-effects and consequent safety upgradation

While some major industrial accidents had occurred earlier, the above disaster resulted in certain important findings and changes, as far as industrial fire safety was concerned. One of the major changes was the banning of the use motor vehicles within critical industrial areas, which is followed rigidly in most high hazard industries to this day. The other important finding was the lack of arrangements to draw the contents of the leaking sphere to other spheres, which are now standard in storage arrangements. It also made industry personnel aware of the need to provide proper bunding arrangments for storage tanks. Design of water spray cooling systems for such storage systems was deliberated upon and improved to provide effective cooling, especially at the bottom of the sphere, including support legs.

As far as fire fighting was concerned, it was quite clear that the affected people had little knowledge of the consequences of the fire. The analysis of the incident resulted in the practices of providing suitable heat insulation to spheres, providing cooling arrangement for the sphere surface/ supports and providing arrangements for remote venting of the sphere. The phenomenon of 'BLEVE' and 'Fireball' became important words in fire service jargon, and further studies were initiated on this subject.