Catalogue of Errors Causes Crankcase Explosion
It was lucky that no one was injured in this
crankcase explosion which lifted and buckled the engine room floor plating, blew the
spare gear store door (which was 2 decks above the crankcase) through the frame , blew
the oil mist detector off the engine, and caused extensive damage to lighting,
vent trunking, paintwork and of course the crankcase relief valves and doors.
The workshop door was buckled and de-laminated The control
room door wire reinforced glass was blown through into the
control room where shards of glass became embedded in
The explosion happened at about
1800 hrs, not long after the
staff had left engine room for the day.
On investigation the cause of the explosion was found to have been a hole in
number 1 cylinder piston crown. Hot combustion gases had then been able to pass
down the piston oil cooling return to the crankcase, forming and then igniting
the oil mist.
The photograph shows
the oil puddle on top of the piston, with a pen lying next
to the crack for scaling.
Burn away was
excessive on the piston crowns, with numbers 1 and 6 out of
tolerance. (number 6 piston crown was also cracked)
The hole in number one piston crown had been caused by erosion due to poor
fuel injection. All pistons were examined, and erosion damage and cracking found
on all of them. The "burning" of the piston crowns giving them the telltale
"elephant skin" surface. The underside of the piston crowns had 3mm build up of
carbon deposits which would have compromised the cooling of the piston.
The fuel valves were examined: Blockages were found, the nozzles were not
genuine spare parts and the push rod profiles were incorrect. The injectors were
lifting at 30bar below the recommended pressure, and the nozzle holes
had worn so that a jet of fuel was being injected rather than an atomised mist.
Bearing damage had resulted from overloading due to poor combustion.
The crankcase relief valves were found to have been incorrectly assembled
with defective flame arrestors.
This caused them to jam open once they had lifted, and this may have contributed
to a violent secondary explosion.
The oil mist detector
had alarmed in the five second period between the first abnormality indication -
ME governor minor fault and the ME cyl no.1 outlet high temp. Due to the short
time-span between first indication and explosion, it would be doubtful if the
monitor could have slowed the engine quickly enough to prevent an explosion.
Information of this event
was taken from the
Isle of Man Accident Investigation
website website. The full report can be downloaded
here in pdf