Good point. The party line has been talking about jet fuel as it's high octane aviation gasoline. Maybe some folks out there remember being on a piston-engined airliner when it first started up and there was a heavy smell of gasoline in the cabin. The "no smoking" signs were lit for a reason!
But jet fuel is kerosene based. And jet-A is also carried as a semi-gel which is designed to counteract the vapor problem.
Here's a good article on kerosene fuels:
From http://www.geocities.com/vialls2/ignition1.html:
Concorde Ignition
Copyright Joe Vialls - 15 August 2000 - All Rights Reserved
... Later on in the report, the Primus stove is used as a prime example of how difficult it is to ignite kerosene based fuels like the Jet A1 in Concorde's wing tanks. Even if Jet A1 is momentarily ignited, sustained combustion of this fuel is far more difficult.
Nearly a month after the crash of AF 4590, investigators are still uncertain what led to the catastrophe in France. Wild media speculation has largely abated, though rumours still hang in the air about a piece of metal on the runway "cutting" one of Concorde's tyres, which then allegedly exploded, hurling a large chunk of rubber up through the wing section and causing the devastating fire. This simplistic story fails to take two known facts into account. So far Goodyear has refused to confirm the small fragments of rubber later found on the runway came from a Concorde tyre, and all eight wheels and tyres were located at the crash site. Though all eight were badly damaged in the crash itself, no pieces of wheel or large chunks of tyre were missing. In other words, media speculation is not supported by known existing scientific evidence.
Aviation crash investigators work under enormous pressure, with governments and airline operators demanding instant answers to extremely complex problems. Sometimes the pressure is so intense that investigators take the "soft option", i.e. go along with government or airline wishful interpretations of events, without first taking all possible causes for the crash into account, and excluding those known to be scientifically impossible. This is of critical importance, because as Arthur Conan Doyle once wrote so accurately: "When you have ruled out the impossible, then whatever remains, no matter how improbable, is the truth." Though none of us yet knows the exact cause of the crash, we do have some of the earlier details, certainly enough to discriminate between known facts and wild media speculation.
Everything was fine on 25 July when Concorde's brakes were released and afterburners lit on the runway at Paris Charles de Gaulle. We know all engines and afterburners remained OK for at least eighty-percent of the take off roll, or Concorde could not have achieved V1 (its go/no-go speed). Remember here that Concorde is not the "rocket" many people believe the aircraft to be. With a full fuel and passenger load en-route for New York, AF 4590 needed every single pound of available thrust from all four engines and afterburners to accelerate to V1. If for any reason the aircraft was sluggish before reaching V1, or had used too much runway when approaching this crucial decision point, take off would have been instantly aborted by the Captain.
Fine so far, from the bald facts above we know that Concorde was past V1 when #2 engine first failed. This fact is corroborated by two radio messages sent back-to-back by the Captain: "failure in number two engine" and "too much thrust". Cryptic though his comments may sound, they tell the whole story at that point. "Failure in number two engine" was extremely serious at that stage and the Captain dutifully reported the fact to the tower. "Too much thrust" or "too much speed" literally meant that aircraft velocity was already higher than V1 when #2 engine failed, making it impossible to safely abort the take off. As previously stated, had #2 engine failed before reaching V1, take off would have been instantly aborted, as per Pilot's Notes. So we have now established scientifically from operating procedures, and from known aircraft performance and the Captain's own words, that the problem started at some point between V1 and VR - the latter being the speed at which the aircraft rotated off the runway.
At this stage it is also critical to note that the Captain did not report an engine fire, which he certainly would have done if an engine fire existed and he knew about it. Nor did the fire sensors in #1 and #2 engine bays report an engine fire at any time. Fire warning Klaxons or bells on the flight deck have considerably more volume than a human voice, meaning they would be heard clearly in the background on the cockpit voice recorder, as they have been on several occasions in the past. Not on Air France Flight 4590. The Concorde CVR contains no sound of fire warning klaxons or bells at all. So we either believe that all of the fire sensors in engine bays #1 and #2 suffered a spontaneous failure, or we take the scientific view that the sensors failed to activate simply because there was no fire inside engine bays #1 and #2. In the absence of any hard evidence to the contrary, investigators are obliged to take the scientific view. Media claims that the tower reported a fire to the Captain "on the runway" are rubbish, or an unfortunate misinterpretation of events. The official sequence shows that the tower advised the Captain about the fire "fifty-six seconds after take off commenced", by which time Concorde was airborne and vanishing rapidly in the distance.
For all practical scientific purposes it is also possible to rule out a tyre failure before V1, again based on known operating procedures, aircraft performance, and an earlier incident with an Air France Concorde at Washington in mid 1979. When a tyre fails during the take off roll, the first manifestation is drag, which in the case of AF 4590 using only four high pressure tyres per main bogie, would have been massive. The failed tyre or tyres would have dragged the aircraft to the left, and more importantly reduced acceleration to the point where take off was aborted. In the Washington incident tyre failure occurred beyond V1, but was certainly noticed by both passengers and crew. The Captain immediately notified the tower, while a passenger later told the media on camera: "The vibration was incredible, it felt like we were driving over speed bumps at fifty miles an hour!" The Captain of AF 4560 at Paris Charles de Gaulle reported no such vibration at any time, though in isolation this does not prove a tyre or tyres did not fail beyond V1. Perhaps the Captain was simply preoccupied with #2 and then #1 engine failures. Perhaps he was, but "perhaps" is not a scientific term likely to impress any member of the investigation team.
For the discerning investigator, the Washington incident serves to clarify this latter point. When the Air France Concorde burst two tyres after V1 at Dulles In 1979, all four engines and afterburners maintained 100% thrust, allowing the Captain to just get the aircraft off the ground, dump fuel and then return for a landing. But according to the reports it was a very close call. Though not operating at Maximum All Up Weight (MAUW), the Washington Concorde slewed violently to the left, and was almost out of runway when it finally became airborne. Now superimpose these known data on AF 4590 at Paris Charles de Gaulle, which was operating right on Maximum All Up Weight and lost one and a half engines and afterburners between V1 and VR. The harsh scientific reality is that the massive drag of a burst tyre or tyres at that point, added to a known 25 to 37.5 percent reduction in engine thrust, would have prevented the aircraft from achieving VR and getting airborne. Instead, Concorde would have remained on the ground, eventually ramming the scenery at the end of the runway. End of story.
So to summarise at this mid point in the report, investigators know from existing data that Concorde AF 4590 lost #2 and then #1 engines between V1 and VR, but these engine failures were not caused by a known or reported fire in either engine bay. Because the aircraft managed to get airborne at MAUW with a 25 to 37.5 percent reduction in overall engine thrust, investigators also know that Concorde did not burst a tyre or tyres before V1, or between V1 and VR. As stated in reports posted on this web site immediately after the crash, the media managed to get everything back to front. The only realistic scientific possibility for the engine failures, is that the raging fire in Concorde's inboard port wing cut the fuel supply to #2 and then #1, causing both to fail because of simple fuel starvation. The two main questions facing investigators are therefore (1) how did the fire start, and (2) how did it manage to stay alight?
There are many who claim that the fuel was simply lit by the intense heat of the afterburners on #1 and #2 engines, but this fails to take into account the scientific reality that the fire in the inner port wing section is well forward of the turbine and afterburner engine stages, as shown clearly on the Japanese amateur photo. Fire cannot and does not run upwind, in this case head-on into the 200+ mph slipstream blowing from forward to aft over Concorde's wing section. Proof of this fact abounds on thousands of feet of video taken at various car race tracks. Out of interest I played a video of an Australian "V8 Supercar" race in which two of the cars collided, then spun apart again, finally coming to rest 110 yards from each other with a substantial raw fuel trail between them. The first car burst into flames, in turn igniting the highly volatile gasoline trail leading to the second car. According to my stopwatch, the flame front advanced from car one to car two at 57 mph, proving that any such scenario with Concorde was impossible. There are other examples of course, and older readers will recall how during World War Two, pilots frequently dived to increase airspeed and "blow the fire out".
Because flight and ground engineers are not experts in the complexities of fuel ignition and sustained combustion, I approached the research labs of two major Australian oil refineries for assistance. Most puzzling of all was that the Jet A1 Concorde was carrying in its wing tanks is probably the most stable fuel known to man. Basically kerosene with minor non-flammable additive packages, Jet A1 has a flash point of 38 - 40 deg C, but this has very little to do with a raging fire inside a closed wing section devoid of oxygen. Flash point is measured artificially under pressure in a chamber, and officially defined as: "The temperature at which the vapour above a volatile liquid forms a combustible mixture with air. At the flash point the application of a naked flame gives a momentary flash rather than sustained combustion, for which the temperature is too low."
So inside the wing, provided there was room in the compartment for a vapour cloud to form, and provided oxygen was present, ambient temperature might have been high enough to momentarily "flash" the vapour, but that temperature was far too low for the sustained combustion visible on the amateur photo, and on the later video of Concorde in flight. With this in mind I asked the lab scientists for their expert opinions on the minimum temperature required for sustained combustion. All were unanimously of the view that the 100% boil off temperature of Jet A1 (250 deg C) would be essential, and even then only if oxygen was present in sufficient quantity. For purists reading this report, the initial boiling point of Jet A1 is 150 deg C, 10% boil off is 170 deg C, 50% boil off is 200 deg C, and 100% boil off is 250 deg C. One scientist was clearly worried about the process of combustion, and kept asking me "But where did the oxygen come from?" Unfortunately I was unable to provide an answer.
In an attempt to simplify his explanation of the difficulties involved in lighting and sustaining combustion with kerosene, a lab technician at the second refinery reminded me of the old-fashioned Primus stove, which is why there is a picture of one at the top of this report. It is an excellent analogy, because the original fuel specification for the Primus stove was almost identical to Jet A1. So much so, that an enterprising camping equipment store owner in Vancouver nowadays buys Jet A1 in bulk, then repackages it and sells the resulting "camping stove fuel" at substantial profit.
To ignite the main kerosine fuel supply in a Primus stove, it is first necessary to fill the circular tray beneath the burner with methylated spirits, which is then ignited in order to pre-heat the fuel supply pipe running up through the middle from the kerosene tank at the bottom, to the outlet jet at the burner on top. Critically, the meths must heat the centre fuel supply pipe to a temperature which ensures the kerosine flowing through it is heated above its 100% boil off temperature of 250 deg C. When this is done, you simpy light the kerosine vapour at the jet, and oxygen is drawn in from atmosphere to provide sustained combustion at the burner. If you try to light the kerosene burner without first burning enough meths in the tray, the result will be failure. Hopefully most readers will recognise the importance of this analogy, and its direct relevance to the fire in Concorde's wing. In order to determine the root cause of the Concorde crash after take off from Paris Charles de Gaulle, investigators must work out which bays and components were involved, and identify the sources of fuel, flame ignition, oxygen, and 250 deg C temperatures. Not an easy task.
Joe Vialls, former member Society of Licenced Aeronautical Engineers & Technologists .
------------------------------------------------------------------------
At 4:44 P.M. Paris time on 25 July 2000, with pilot Christian Marty and co-pilot Jean Marcot at the controls accelerated down the runway and subsequently blew a tire on its left main gear. As the crew fought to keep the airplane under control, it began drifting left on Runway 26R with a long sheet of flame trailing from its left wing. The alarm gong sounded in the cockpit, and engines one and two (of four) lost power. "Watch the airspeed, the airspeed, the airspeed!" Marcot yelled to Marty as the airplane continued heading left, in the general direction of a taxiing Air France 747 bringing President Jacques Chirac back from a Tokyo summit.
Marty tugged on the controls and tried for liftoff at 188 knots (11 knots below normal rotation velocity). With trust coming only from two engines the aircraft barely reached 200 feet in the air, before suddenly nosing up, rolling over and crashing into a hotel.
Air France President Jean-Cyril Spinetta watched horrified as he viewed Concorde F-BTSC trail flames and crash outside his picture window overlooking Charles De Gaulles runways. "For all of those who were eyewitnesses to this catastophe, and I'm one of them, the cause of the crash was an engine fire on takeoff." He immediately grounded the Concorde fleet until further notice and British Airways quickly followed suit.
On 16 Aug 2000, Frances Bureau Enquites Accidents declared that a tire blowout caused the crash. It also recommended the suspension of Concordes certificate of airworthiness. France's Direction Generale de l'Aviation (equivalent of the U.S. FAA), pulled the planes certificate. "What is uniquely differant in this case is that tire debris alone is thought to have led to this catastrophic accident." Two official French inquiries were set into motion. The BEA searched for causes, while a judicial investigation headed by three magistrates tried to determine legal responsibilites.
The first puzzle to solve was a piece of the number 5 fuel tank found on the runway that had somehow been torn loose from the inside out. Researchers from the European Aeronautic Defense and Space Company, successor to the original Concord builder, Sud-Aviation, ran a computer simulation that showed when a large piece of tire weighing about nine pounds struck the underside of the left wing, the impact generated a shock wave, propagated through the full fuel tank, that first moved up, then down, exploding outward. The resulting one foot square hole created a massive kerosene leak - on the order of 20 gallons per second - that somehow ignited.
Why was there such a large chunk of tire, bigger than usual, after a blowout. Investigators found a 17" titanium thrust reverser wear strip on the runway from an airliner that had just taken off minutes before Flight 4590. Striking the metal strip at high speed, they theorized scalped the number two tire of a 5' length of tread, which was whipped up at the wing by tremendous centrifugal force.
The most controversial finding however, was that a spacer that holds two lateral rings in position on the oleo/bogie coupling of the main left gear and is vital to wheel alignment, had not been reinstalled after routine maintenance work performed four days before the crash, because of an Air France maintenance error. Air France being traumatized by the accident and being sued for $100 million by families of the crash victim, declined repeated interview requests by Air & Space Smithsonian. However, BEA ruled out the missing spacer as a cause of the crash.
"The truth is because of the missing spacer, the left main gear was slightly skewed on the takeoff roll. Skidding heated and wore down the tire, caused the plane to drift to the left side of the runway, and kept it from accelerating normally." charges Jean-Marie Chauve, a 37-year Air France vetran and retired Concorde pilot who has done his own calculations - and has had them verified by independent experts - based on published information from the flight data and cockpit voice recorders. His version is seconded by Michel Suaud, a longtime Concorde flight engineer also retired. They spent several months preparing a detailed report on the crash, which they presented to the investigating magistrate of the judicial inquiry.
"Our figures show that the plane was moving to the left at the start of the takeoff roll, not just after the blowout and loss of engines one and two. The tire burst at around 174 knots and only after the blowout did it strike the metal strip. If acceleration had been normal, the plane would have been airborn about 50 yds before reaching the metal strip. The BEA says the leftward yaw was caused by loss of thrust from the left engines, not by the skewed bogie. But they've never shown us where our figures are wrong."
The bottom line is that Jet A1 fuel does indeed burn and quite fiercly too. It did so after two planes hit the WTC towers, it did so when a plane crashed into the pentagon, it did so when the F4 crashed at Magoo Airshow a few weeks back, and when Flight 255 crashed at Detroit Metropolitian Airport 16 Aug 1987 thirty miles or so from where I live.