Failure of a strainer allowed sand from brine filters to get into mercury cells in a chlorine plant. It appears that the sand obstructed flow and caused excessive hydrogen generation. An explosion occurred in the chlorine collection system.

A 7 inch thick reactor of a wide range hydrocracking unit being operated by computer at 2500 psi failed explosively due to localised heating. The blast caused widespread damage over a 300 yard radius including an adjoining catalytic cracker and crude pipe still where the roof of the control building collapsed. Other units were safely shut down from a blast resistant control building which sustained minor damage.

A vertical cylindrical hydrodesulfurisation reactor charge heater was being started up with hydrogen flow only rather than hydrogen plus oil as in normal operation. As a result of the low heat load only a couple of the burners were being fired. The uneven firing resulted in one pass of the furnace being overfired and a tube ruptured. This problem was solved by installing a single centre burner for start-up. In addition multiple skin thermocouples were installed on each tube pass at the hottest points.

A vertical cylindrical fired heater was used on the charge to a hydrodesulfurisation reactor. The flow was two phase at the heater inlet (hydrogen and hydrocarbon). The distribution of flow to the multiple tube passes depends on symmetrical pipework and the pressure drop through the passes. There were no individual pass flow meters.During start-up when the flow was low a radiant tube ruptured in one pass. This was though to be due to static head in the U tubes of that pass with liquid in the up legs and vapour in the down. Thus there was no flow in that pass and it overheated.

A short circuit developed in the wiring of a fire eye system on one of two 600 psig boilers serving a refinery. This triggered a rapid sequence of events that resulted in a one week shutdown of much of the refinery. Fortunately no one was injured and equipment damage was negligible.When the hydrocracker make-up compressor started slowing down, the control system at the reformer released the excess hydrogen to the flare. However, no increase flame was observed at the flare and the crude unit pressure went over 50 psig. The trouble was quickly diagnosed as a blockage in the flare system.

The fire ignited roofing materials which had fallen down as the result of the explosion.The operator, who was outside the plant when the explosion occurred, pulled the emergency switch which shut down all the compressors. Although in a dazed condition he then checked in the motor room that all the motors had stopped. He was subsequently taken to hospital where he was found to have no physical injuries but was moderately shocked. The fire alarm was raised by the Gatekeeper who had heard the explosion and seen flames and smoke in the vicinity of the hydrogen plant.

The fire started as a result of an explosion in the process section of a refinery. An open structure of steelwork protected with reinforced concrete housed the usual furnaces, reactors, heat exchangers, storage vessels, overhead pipe racking, pumps and ancillary equipment associated with the refining of crude oils.Also included in the area were three 46m (150 ft.) high stripping and drying columns built on a concrete base approximately 91m x 21m (300 ft.

The event occurred at an ethylene production plant. The plant was shutdown at the time as the supply of compressed air to the gauges had inadvertently been switched off. This failure of the air supply was not immediately recognised because of the low pressure alarm was installed before the isolation valve. Shortly after the plant was restarted an unusually high temperature was observed in an acetylene hydrogenation column.

A weld on a 1 inch diameter instrument tapping from a Octafiner reactor overhead line fractured, resulting in a small but intense fire, lasting about 50 minutes until the unit was nitrogen purged.The fire was always under control and the fire fighting effort was directed solely to cooling the surrounding metal with fire steam lances. Plant Inspection reported that the failure was due to hydrogen embrittlement of the heat affected zone of the tapping and probably resulted from improper or no stress relieving of the connection during construction.

The vent occurred at the hydrogen manufacturing unit: a pipe of this unit burst, followed by an explosion. The fire was extinguished in about 15 minutes. One employee nearby had a consequence from a burn in the face for two weeks. According to the police, the explosion occurred while checking the temperature in the hydrogen manufacturing device after pouring liquefied butane gas as fuel in the device.