September 11th or 9/11 stands out in our minds for obvious reasons. However there was another 9/11, 11th September 1947, when a crankcase explosion on the Reina del Pacifico killed 28 men and injured 23 and led to the development of crankcase relief valves and oil mist detectors. Of course there had been crankcase explosions before this, but none which had such devastating consequences.
Between 1990 and 2001 143 crankcase explosions were reported to Lloyds Register which have about 20% of the worlds shipping in its class, so if we use that as a factor, we can estimate the total reported incidents were 715 in 11 years or about 65 a year. Don't forget that these are reportable incidents, i.e. those where the damage sustained has warranted a major repair or has resulted in injury. Minor explosions may have gone unreported, and it is possible that the actual number of incidents is more than double those reported. - maybe 3 a week!!
Of those incidents reported to Lloyds, 21 explosions happened in two stroke marine diesel engines and 122 in four stroke marine diesel engines. But this doesn't mean that four stroke engines are more likely to have an explosion; there are 7 times as many four stroke engines at risk than two stroke engines.
SEQUENCE OF EVENTS LEADING UP TO A CRANKCASE EXPLOSION
For an explosion to occur there must be a source of air (oxygen), fuel and ignition. Oxygen is present in the crankcase, but the lubricating oil splashing around in the crankcase is in too large droplets to start burning at the speed needed to cause an explosion, and the oil/air concentration is too weak.
If, however a mechanical fault develops with the consequent rubbing of moving parts, then a hot spot will occur. This could happen in the crankcase, chaincase, or camcase. When the temperature of the hot spot reaches 200°C the lubricating oil splashing on to this hot spot vapourises. The vapour then circulates to a cooler part of the crankcase where it condenses into a white oil mist. The oil droplets in this oil mist are very small - 5 to 10 microns in diameter. When the concentration of oil mist reaches 50mg/l (about 13% oil mist - air ratio), it is at its lower explosive limit. If this oil mist is now ignited by the hot spot - and tests have shown that it is necessary for a temperature of about 850°C to ignite oil mist in a crankcase under operating conditions - then an explosion will occur.
Although the most common cause of of a localised hotspot is due to friction, it is not the only cause of a crankcase explosion. A cracked piston crown, blowby or an external fire have caused crankcase explosions in the past.
PRIMARY AND SECONDARY CRANKCASE EXPLOSIONS
Severity of explosions vary between a puff which may lift a relief valve to a violent explosions which causes major damage and may injure personnel and cause a fire. Evidence indicates that the longer the combustion path, the more violent the explosion. This has become an area of concern with the large two strokes of today which may have a crankcase volume of 500m3 +.
When an explosion occurs a flame front travels down the crankcase with a pressure wave in front of it. The turbulence caused by moving engine components causing churning and mixing of vapours increase the speed of the flame front and its area, which contribute to the increase in pressure. Turbulence caused by venting of the pressure through relief valves can also influence the explosion.
Following the venting of the explosion through the relief valves, there is a drop in crankcase pressure to below atmospheric pressure. This can cause air to enter the crankcase resulting in another flammable mixture to be developed resulting in a secondary explosion to occur. The secondary explosion is more violent and can result in crankcase doors being blown off the engine, and fires starting in the engine room. If the relief valves do not reseal after lifting, or if they do not lift at all in the primary explosion ( due to lack of maintenance etc), then door(s) may be blown off in the primary explosion, giving a ready path for the ingress of air, which will make a secondary explosion more likely. Air can also be sucked in via the crankcase vent, although rules state that this must be as small as practicable and new installations must have a non return valve fitted.
If a primary explosion occurs, the pressure wave may send a large amount of oil mist out into the engine room. Although the flame arrestors on the relief valves should prevent ignition of this oil mist by the flame front, the mist will be sucked up towards the turbocharger where it may be ignited by an unlagged hot exhaust manifold. This ignition of oil mist can cause severe damage to plant and personnel.
CAUSES OF CRANKCASE EXPLOSIONS
The table below gives details of a number of accidents which have occurred since 1995 to large slow speed 2 stroke engines where the cause is known. In a number of cases death or serious injury to members of the crew occured.
This article and the following articles can be found in the members section under
Oil Mist Detection
The QMI Oil Mist Detector
The Shaller Visatron Oil Mist Detector
Crankcase Relief Valves
Crankcase Relief Valve Testing
Other Methods of Detecting a Potential Explosion Risk
Reducing the Risks of a Crankcase Explosion
Action to be Taken in the Event of a High Oil Mist Alarm