The internal air traffic of Japan had steadily grown, fueled by the growing economy from the early 1960s to the mid-1980s (The Japanese economy had grown on average 6 % per year during 1961-1980). Japan has several metropolitan areas, which are divided from each other by steep mountain chains, and to make logistics more difficult, the population centers are locatd on four major islands. This lead to rapidly growing demand of domestic airline travels, and travels soared from 18 million tickets in 1970 to 37 million in 1978. Many flights were sold out, but the airlines had little possibility to add more flights between popular destinations, as the runway capacity at airports had peaked. There was no room for extra flights, so the largest airline, Japan Air Lines, ordered a special domestic version of the Boeing’s brand new gargantuan, the Boeing 747, to keep up with the ever-increasing demand.

The Queen of the Skies, Boeing 747, had been introduced into service in the year of the dog, 1970. It was the first jumbojet, a widebody: it had two corridors along the main deck. Just to increase the mood, there were some seats in the upstairs, just behind the cockpit in the distinctive “hump” on the 747. This was due to the aircraft’s origins in the USAF Heavy Airlift program on the early 1960s: the nose of the aircraft had been designed to be an openable one, to enable loading the main deck with military cargo. As Lockheed won this bid with their C-5 Galaxy heavy lifter, Boeing decided to continue development and modify their model to meed civilian airliner needs, reaching a new market segment: large capacity airliners.

An early Boeing 747-100. Photo: Live from a Lounge

The 747 was designer for intercontinental traffic, but the domestic flights in Japan are relatively short as the crow flies. This meant that the aircraft would fly several sorties per day, which due to cabin pressurization meant rapid fatigue of the pressure hull. Boeing had designed the 747 to catty 24 600 flights in 20 years, and every flight meant a pressurization cycle in which the hull is pressurized above the ambient pressure of the cruising altitude of 10 000 meters. At this altitude, the athmosphere is at the pressure of 0,264 bar, and even sherpas cannot endure without pressurized auxiliary oxygen tank. This is why the airframe of an airliner is pressurized using either engine’s compressor stage bleed air or separate compressors. This creates a pressure difference between the airframe and ambient air, which in turn is a force stretcing the airframe outwards. This cycle repeats on every flight, and as the loads repeat, these repeating loads will eventually create fatigue cracks in metal. Thus, the airframes have a limited lifecycle, measured in pressurization cycles.

The Japanese customers, JAL and some years later ANA (All Nippon Airways), requested a specialized version of the 747: the 747-100SR (Short Range). Most of the stuctural components were strengthened, either using thicker metal or additional spars, and the designed lifespand was increased to 52 000 sorties on 20 years. Also the landing gear was made of heavier parts, as were the brakes. In return of this added weight, 20 % of the fuel capacity was removed, as it was not needed in this purpose anyway.

Boeing’s refecence seating arrangement fot the 747 was 366 seats divided into first, business and economy class, but FAA allowed 550 seats by the emergency exit capacity. In short Japanese domestic flights, there was little demand for first class service, but a high demand for seats. This made the 747-100SR the first airliner fitted regularly for more than 500 seats. This version was introduced to regular service in September of the year of the Ox 1973 (wearing the colours of JAL – the ANA would only procure the following model, the 747-100B SR ,in the yeat of the horse 1978).

JAL 123

It was a hot Monday, 12 August in the year of the ox 1985. A heavy roar engulfed the Tokyo Haneda Airport, as fout high-bypassing Pratt & Whitney JT9D’s spooled to takeoff power. A growling 747-100SR climbed towards the skies and continued to cruising altitude. One of the Japanese domestic mass transit flights, JAL 123, had begun.

The travel to Osaka Itami Airport (the todays’s Kansai Airport wasn’t finished until the year of the dog 1994) will take 54 minutes by the schedule, so the 12 attendants will have a busy time serving all the 509 passengers on board. The flight between the two largest cities of Japan is sold out, as today is the eve of the Buddhist celebration of Obon, and many Japanese are visiting their relatives.

JA8119, 17 August1984. Photo: ASN

Captain Masami Takahama, 49, is a Japan Air Lines instructor and a very experienced pilot of 12 424 hours of flight time. Today he is seated on the right side, at First Officer’s seat, as he is supervising Jutaka Sasaki, 39, who is about to finish his additional training and to be promoted to Captain. This means Sasaki, with 3963 flight hours on his books, will be the Captain of this flight, his supervising instructor being the Captain-in-charge. The third man on the filght deck is the Flight Engineer Hiroshi Fukuda, 46, who has 9831 flying hours on his records. This mass transit flight is commandeered by steel-eyed pilots; every one of them has more experience than most flight crews have combined. In total the 747 is carrying 524 sould on board.

JAL 123 climbs with a slight nudge to the right, and at 18:16:55 they request ATC for a right turn towards Kushimoto VORTAC beacon sooner than expected, before passing Mihara Sagara Mountain of Oshima island. ATC approves at 18:18:33, and the 747 makes a right turn.

From left to right: Captain Takahama, First Officer Sasaki, Flight Engineer Fukuda. Kuva: Namuwiki

At 18:24 a flight attendant calls the flight deck, informing that a passenger has a severe need for the restroom, but the seatbelt light are still on. Fukuda answers the call and allows it, under the condition of being careful. JAL 123 has almost reached cruising altitude of FL 240 (24 000 feet or 7300 meters).

There is a loud bang at 18:24:35. A vast gust of wind screeches towards the rear of the aircraft. The inner roof in the rear of the aircraft falls down. The main deck is filled with a white fog. The 747 shambles, and enters a irrate motion.

The flight crew type SQUAWK code 7700 to the transponder, the emergency code, which visible on Air Traffic Controller’s rada display. 18:25:21 Captain Takahama declares emergecy via radio, and requestes flight level 220 (6700 meters) and a radar vector to Oshima. He plans to return to Haneda Airport. Something has happened to the 747, but he doesn’t know what. Tokyo Area ATC officer clears JAL 123 for a right tunr to heading 090, and Sasaki makes the instructred input to flight controls. The 747 banks rapidly and steep, 40 ° to the right, and everyone on the flight deck are surprised. Sasaki immediately turns his control yoke to te left to reduce the bak angle, but the 747 does not respond. Takahama orders Sasaki to pull up, but this iput is also in vain. Flight Engineer Fukuda notices that the hydraulic pressure is falling rapidly. Takahama and Sasaki look at him in disbelief, and request him to chech the hydraulic system. This does not alter the result. All four hydraulic systems bleed to zero pressure. Despite of any auxiliary or emergency power, any hydraulic system will not function without a fluid. The steering system of the 747 has bled to death.

Tokyo ATC confirms at 18:27:02 that JAL 123 has declared emergecy, but when the controller asks a reason for it, he gets no answer.

He repeats his question, and only then Takahama answers with slow speak, that the aircraft is out of control. Nobody knows why, and neither that JAL 123 has lost pressurization. At 7300 meters of altitude, the air presssure is 0,31 bar and the partiual pressure of oxygen is only 0,065 bar. The flight crew and everyone alse on board JAL 123 are haunted by hypoxia, their lungs cannot gather enough oxygen from the thin air.

Fukuda understands what is happening, and makes a question: “How about the air pressure? Are the oxygen masks deployed?” At the same time, Tokyo ATC requests the flight to descend, and Takahama responds that JAL 123 is descending. The intercom rings, and Fukuda answers the call. A flight attendant informs that the oxygen mask have been deployed aoutomatically, and that the roof around the door R5 has collapsed. Fukuda suggests that also the flight crew would wear oxygens masks, and Takahama replies that it would be wise. Still, none of the three men ever wear their masks. No-one knows why, but it is probably related to hypoxia, which makes their minds numb. They are succumbing to the big sleep.

The last picture taken on board of JAL 123. The oxygen masks have been deployed. Photo: Namuwiki

The 747 descends, but only for a short while. Suddenly the nose turns upwards, and the aircraft starts to climb. Nobody knows why.

JAL 123 has enterd a phugoid motion. AS the flight crew cannot give any inputs, the aircarf’ts motion is controlled purely by aerodynamic forces. As the 747 descends, it gathers speed, which increases the lift genrerated by the wing. This lifting force in turn makes the nose to rise, turning the 747 into a climb. AS the aircraft climbs, it loses airspeed, which decreases lift, and the cycle will repeat itself. The wavelength of this cycle is about 90 seconds. In addition to the phugoid motion, the 747 has entered a Dutch roll: it is making steep banks to left and right, and the nose of the aircraft is making a constant lift-right yaw. The banking cycle stabilizes to a 50 ° arch, and the amplitude of the motion settles at 12 seconds.

JAL 123 bounces restlessly between flight levels 200 and 250 (6100 to 7600 meters) during the following 18 minutes. Hypoxia is an immediate danger at these altitudes, and as the control systems are lost, the situation is confusing. Japan Air Lines’ gound serviced makea SELCAL call to Flight Engineer Fukuda, and they learn Fukuda suspects that the door R5 has failed, and that JAL 123 is attempting a emergency descent. Takahama and Sasaki work hard to get the 747’s nose downwards, but the Jumbo Jet won’t budge.

The passengers wrote farewell messages, isho, to any piece of paper they fould find. This was written by Shirai Mariko, 26: I'm scared, I'm scared, I'm scared. Help me.I feel sick. I don't want to die. -Mariko.

Fukuda suggests that they would deploy the landing gear. It might work like a drag anchor, stabilizing the aircraft’s motion. Takahama nods, so Fukuda begins his attempt. With no hydraulics, the actuators have no chance of working, but Joe Sutter had made a series of fail-safes when he designed the 747: there is an electric backup system for the landing gear. All four engines are running and in addition to thrust, they are providing electric power via connected generators. Thus the gear is slowly deploying.

The landing gear creates a stabilizing force, and the increased drag also slows the aircraft down. The phugoid cycle winds down, and ceases almost entirely. Takahama and Sasaki manage to control the altitude by using engine thrust levers. To climb they get more power from the engines, and vice versa to descend. They also find that they may control the yaw of the aircraft by lincreasing thrust on left or right side engines at a a time.

But the deployed landig gear makes this yaw control much more difficult. 18:47:19 Takahama again informs the ATC that JAL 123 is out of control. The flight crew is now more responsive and speak to each other much more often, as the flight has descended over 2000 meters and the ambient pressure is 0,61 bar. The hypoxia is releasing it’s claws. The flight crew succeeds in turning the 747 to west, towards Tokyo.

The flight path of JAL 123. The red dot marks the location of first sign of emergency. Photo: Ministry of Transport, Japan.

But the fate had only given them a respite, not mercy. The aircraft is now not reponding to thrust levers. The 747 antaers a left turn, which cannot be corrected by engaging the #1 engine – the leftmost – to takeoff power. Also the altitude control is lost, and the 747 descends to 2100 meters.

JAL 123 has drifted above the mountains of the largest of Japanese islands, Honshu. The phugoid cycle begins again at this altitude. Takahama ordesrs a full power at 18:48, an idle power at 18:49. The mountains are frightingly close. The engines roar again to full power, and the 747 climbs at a heartstopping 40 °. The airspeed decreses to 108 knots – dangerously low. The stick shakers are activated. The 747 is very close to a stall, but the pilots have no other opion than engage full emergency power.

Takahama is tying to get the flaps deployed, to inrease wing lift. These actuatros are hydraulic and thus cannot operate without fluid, but Joe Sutter’s design again has redundacy: the flaps have an electric backup mechanism. It is very slow thoigh. and it takes 3 minutes and 10 seconds to get the flaps eve to a 5 ° position. The 747 is climbing again, to 4000 meters. Japan Air Lines ground services try to contact JAL 123, but no one is answering the SELCAL call.

18:53:21 Takahama contacts Tokyo ATC and informs that JAL 123 is still out of control. The flight crew continues to deploy the flaps, and they reach 20° position. Takahama requests Tokyo ATC to give him his own position – no one had any chance to track it in this turmoil. Tokyo ATC tells him they are 55 nautical miles northwest to Haneda. The airport fire department has been prepared for a crash landing. But the 747 now makes a bank to the right.


JAL 123, 12 August1985. Photo: Namuwiki

Engines #1 and #2 at the left wing are running at higher power than the riggt wing engines, steepening the bank, for reasons unknown. The bank steepens to 60 °, when Takahama orders to retact the flaps and give more power, but for some reason, the left engines are still running at higher power. The bank approaches 80 °, which means the 747 is practically on it’s side. The flight crew pust the trhust levers to emergency power, which reduces the bank to 70 °, but airspeed has now increased to 340 knots (640 km/h). Ground Proximity Warning System activates. The flight computer sounds a warning PULL UP, and keeps it on.

This is the final sound captured by the Cockpit Voice Recorder. JAL 123 strikes the Mikuni Mountain’s lower cliffs at 1610 meters by her right wing, which is torn away. The 747 is immediately bent to her back by he dissimilar lift, and crashes into the next cliff at 1565 metes. The Tokyo University seismometers detect one wave at 18:56:27, followed by another, a louder one, at 18:56:32. The turmoil has become to an end.

USAF Airbase Yokota had been listening to the radio communications, and offerd their runway for emergency landing should JAL 123 head this way. At 19:15 Yokota Approach informed Tokyo rescue center that an approaching C-130 had detected a fire on ground at bearing 305, 35 nautical miles from Yokota TACAN beacon.

Tokyo rescue center alerted the Japanese Self-Defence Forces, whose Air Force wing mobilized it’s forces at 20:33 and Ground Force at 21:30. A Defence Force helicopter overflew the crash site at 20:42. The Sun had set at 18:34, so the pilot had no more means of observations than a searchlight. He detected only burning debris scattered over a large area, and as the site was on a steep hillside, he could not try to land. There was no any sign of survivors. The crash of JAL 123 was (and still today is) the worst disaster involving only one aircraft.

Based on the helicopter pilots’ report, the Japanese Self-Defence Force concluded that there were no survivors, and it would be more beneficial to crate a large, well-equipped camp at the base of the mountain. There were very few roads at the mountain and the risk of a landslide was imminent, especially in the dark. They could amass a larger force during the night and reach the crash site at daylight with enough manpower.

The Japanese society enjoys a reputation of efficiency (during the evening of 12 August, more than 2500 police officers and 1000 soldiers had been dispatched and transported to the area), but the authority ranges were also sectioned. In addition to the Self-Defence Forces, the area was searched also by the local police, the Marine rescue center, the Japanese Red Cross, Uoeno municipality Fire Department and the Ueno Hunter’s Association (responding to a reequest by the Gunma Prefecture Police). Communication and access to information between all these parties cannot be graded as perfect, and thus a Nagano Police helicopter re-discovered the crash site at 05:37 (even the military could not confirm the exact location until 04:39).

The firemen had been marching from the early hours of the night, and reached the crash site after a long and exhausting climb soon after 09:00 on 13 August. At 10:45 on of them was surprised: he saw movement in a ravine nearby.

Yumi Ohciai. Photo: Namuwiki

The movement was caused by Yumi Ochiai, 26, a JAL flight attendant, who was off-duty during the flight and was seated as a passenger at seat 56C. Her hand an pelvis were broken, but to everyone’s astonishment, she was alive. The rescue crew startted a wide search with newfound enthusiasm and hope. Soon they found Kawakami Keiko, 12, entangled in the trees covering the hill. Tho much more astonishment, she only had bruises and muscle tear. Soon they also found Yoshizaki Hiroko, 34, and her daughter Mikiko, 8. But they were the only ones found alive.

The rescuers are frustrated. As they had found survivors 15 hours after the crash, It was obvious that at least some more would have been found, had the site been reached sooner. This was later confirmed by a doctor: severl of the deceased had injuries, which had not been immediately lethal.

The survivors had been gathered at 11:40, and they had been triaged and stabilized by 13:29, when JSDF helicopters arrived for them. The childer were lifted to helicopters in a harness with surface rescuer, and the adults in stretchers. They were all airlifted to Ueno camp, where they were examined by doctors and then airlifted to the sports ground at Fujiko elementary school, where amulances waited for final transport to a hospital at 14:17.

Keiko Kawakami is being prepared for airlift. Photo: Namuwiki

And so all the rush had faded away. What was ahead was the long and heavy work of aviatiopn forensics: why had the 747 ran out of control? The culprit was not detachmnent of the door R5 as the flight crew had first suspected: it is found within the wreckage.

Soon a series of photographs is handed to the investigators. An eyewitness had taken the photos, as they had seen the aircraft behaving oddly. And the photographs reveal why: the vertical stabilizer is missing. The Japanese Navy finds it floating at 18:55 floating at Sagami Bay, approsimately the position at which JAL 123 had first declared emergency. This is a important clue. The loss of the vertical stabilizer and rudder makes controlled flight of an aircraft extremeley difficult. At the crasch site, the hydraulic pumps are found. They show clear sighs of having been running dry, which is evidence that the 747 had lost all hydaulic fluid, which in turn had made all other control surfaces inoperable.

Fuji TV has enhanced this photograph of JAL 123

Photo: Namuwiki

...and added what is missing. Photo: Namuwiki

The massive tail structure won’t just detach by it’s own. There had to be a reason. But there is no sign of an extenal strike, manufacturing fault, fatigue or corrosion. The vertical stabilator had been struck loose from within. The investigation looks into JA8119’s history.

JAL 115

JA8119. Photo: ASN

It was a hot Friday, 2 June in the year of the oHorse 1978. A heavy roar engulfed the Tokyo Haneda Airport, as fout high-bypassing Pratt & Whitney JT9D’s spooled to takeoff power. A growling 747-100SR clibed towards the skies and continued to cruising altitude. One of the Japnese domestic mass transit flights, JAL 115, had begun.

The travel to Osaka Itami Airport (the todays’s Kansai Airport wasn’t finished until the year of the dog 1994) will take about an hour by the schedule, so the 13 attendants will have a busy time serveing all the 379 passengers on board. As the flight crew of this first-generation Jumbo Jet consists of three aviators, the 747 is carrying 524 souls in total.

The flight is uneventful, and passes through some beautiful scenery, as the Mount Fuji, Nagoya Metropolitan area and the historical Kyoto. Then it’s time to call out the service and prepare for descent.

JAL 115 is approaching Osaka Itami Airport at 15 knots wind. It is a regular automated ILS aprroach to runway 32 Left, and the pilot is preparing for a flare, to pull up slightly to decrease the descent just prior to touchdown (usually at 30 to 50 feet or 15 to 10 m above the runway).

At 15:01 the main gear wheels bite the pavement. The pilot’s flare is a steep one and possibly too late, so the 747 bounces off the runway, back into air. The pilot continued to flare too steep, and the 747 comes into another touchdown, now in an angle too extreme. The massive Jumbo jet hits the runway tail first, and the aluminium fuselage screams like a sheep reversing into blazing cactae. The tailstrike smashes the nose rapidly down, and 25 people suffer injures, two of them serious.

JA8119, 2 June1978. Photo: National Geographic

The 747 comes to a halt, and the injured are evacuated, but Japan Air Lines has yet another patient: Juliet Alpha 8119, the tailstruck 747, whose farts could no longer be trusted. The tailstrike and the following slide against the runway pavement had peeled the fuselage open, and also the rear bulkhead had been cracked by the strike. This sort of damage repair was beyond JAL’s maintenance expertise, so the repair was ordred from the manufacturer Boeing. The company accpeted the order, and sent a detachment of engineers to Japan. They replaced over 16 meters of the fuselage skin, but the rear bulkhead was a more complicated case: it it a structure resembling an umbrell, consisting of 36 triangular panels held together by radial stiffeners. The bulkhead had craked at sector L18 at bays 2 and 3, and they had to be replaced.


Bulkhead sector L18. Photo: Ministry of Transport, Japan

The same bulkhead at the crash site. Photo: National Geographic

Welding aluminium is like turning a hex screw with a cauliflower. It’s reputation as a “non-rusting” metal is in fact caued by it’s easy oxidation, which crates a solid layer of aluminum oxide Al₂O₃ to it’s surface, preventing oxygen from reaching the metal beneath. This passivation makes aluminum in many ways more resistant to corrosion than steel, but as a trafe-off welding of aliminum is nearly impossible: the aluminum oxide has a high melting point of 2072 °C, whereas the base metal melts at 660 °C, and as a great conductor of heat, the aluminum has melted from a large area much before the wleding arch has even penetrated the film-thin oxide layer. As a result, the metal acts like a vat of metal oatmeal. Thus, aluminum structures are assembeld using other tehcnologies than welding (which requires well controlled conditions and careful planning), such as riveting and adhesives.

Bulkhead panels cracked at the riveting. Photo : National Geographic

Thus the JA8119’s damage had to pe rapaired by pulling off old rivets, removing the cracked metal plates, and replacing them with new ones, and joint plates called splice plates.

As the splice plates had to be riveted in place, this created an issue. Riveted parts have to overlap each other, so that a row of rivets always attaches at least two plates together to crate a continuous structure. Rvets carry longitudal tension only by their cross-section, which is minimal comapred to the entire structure, and a single row of rivets also crates a bending line for the bending forces.

This is why a single row of rivets was not anough when inserting a splice plate in between of two parts of the bulkhead. According to Boeing’s repair manual, a minimum of three rows of rivets was required: One row between the outer bulkhead panel and the splice plate, another between the inner bulkhead panel and the splice plate, and a third row whih panetrated both inner and outer bulkhead panels and the splice plate. This, both longitudal and shearing tension are all carried by all of these three rows of rivets and all three panels.

The priciple of a correct repair and how the repair was carried out to JA8119. Photo: Wikipedia

For an unkown reason, Boeing engineers used two slice plates. The did shoot three rows of rivets, but not through a single structure as was required. The third row of rivets attached onlu the outer bulkhead panel to a single splice plate which was not attached to any other structure, so that the outer bulkhead panel was attached to another splice plate only by the central row of rivets, which were shot correctly through all three panels. This structure carried only about 70 % of the streght of a correct one.

This structure was prone to metal fatigue, both due to lesser stregth and a fatigue line which had unintentially formed. It held for 12 318 pressurization cycles, until at 18:24:35 on 12 August 1985 gave way, at pressure difference of 0,597 bar. About 2 to 3 square meters of bulkhead separated simultaneously, which let the cabin pressure inside the tail cone of the 747. The impluse struck the APU generator’s bulkhead, which gave way in 0,04 seconds. In 0,29 seconds the vertical stabilizer’s attachment bolts began to give way and the surface plating cracked open to rushing air, creating a domino effect if increasing pulling force and breaking bolts.

At 1,21 seconds, the decreasing air pressure within the cabin increased relative humidity to 100 %, creating a thick fog. The pressure alarm was activated at 1,56 seconds and the oxygen masks were released at 2,40 seconds. The Boeing 747 has two ruddersstacked vertically, with hydraulic lines 1 and 3 operating the upper and lines 2 and 4 operating the lower. As the entire vertical stabilizer was torn away, all the hydraulic lines inside were cut, which bled all the hydraulic fluid out.

From this point on, JAL 123 was doomed. Test pilots try to solve the situation in a simulator. Not a single one manages to regain control of the 747. Takahama, Sasaki and Fukuda managed to continue flight for 32 minutes after the failure, which even some of the experinced test pilots cannot reach.

Epilogue

In a way, Boeing saved face, but in another way it lost it. There was no inherent design error in the 747, which had or could have led to the disaster. Contrary, the 747 had several redundant systems to give even some tools to solve the hopeless situation. But the repair was failed by Boeing, against Boeing’s own manual. They had used splice plates of a wrong size, which were riveted in place incorrectly.

The world is not a void. Dennis Fitch, an United Airlines DC-10 instructor, had reserched the JAL 123 disaster, and was curious about the flight crew’s solution to use differential engine thrust to control an aircraft with no hydraulics, and worked it in a simulator. He later became one of the irreplaeable heroes of United 232 Heavy, whose expertise and experince held on to 184 souls in the year of the Snake 1989.

In aftermath of United 323 Heavy, McDonnel Douglas hastily installed hydraulic fuses on all DC-10s under the tail engine in service to prevent a total loss of hydraulic fluid. Would hydraulic fuses might have saved JAL 123, remains an open question. The fuses would have spared the fluid and granted control of aircraft roll using ailerons, and possibly pitch using elevators or horizontal stabilizer trim, but no amount of hydraulic fluid would have brought back the separated rudder and yaw control, nether the stabilizing effect of the vertical stabilizer. The question was not asked, as the 747 had not been doomed by a design flaw, but an external factor – which was Boeing.

The location of the rear bulkhead. Photo: Namuwiki

But this was not enough in the eyes of the world. Japan Air Lines’ maintenace manager Hiroo Tominaga committed seppuku, a honor suicide with a knife, 22 Spetember in the year of the Ox 1985. Tajima Susumu, the engineer who had ispected Juliet Alpha 8119 after the tailstrike repairs, followed suit, taking poison in the year of the Tiger 1986. They had committd suicide, but it was not their cause of death. They had died of guilt.

An thus was 522 names written in the stone book of souls. It was all due to a piece of aluminum sheet, and a few rivets more.

References:

Shun Takada et al. Aircract Accident Investigation Report. Minsitry of Transport, Japan, 1987. Available:

https://jtsb.mlit.go.jp/eng-air_report/JA8119.pdf

https://weatherspark.com/h/d/149383/1978/6/2/Historical-Weather-on-Friday-June-2-1978-at-Osaka-International-Airport-Japan#Figures-Temperature

https://asn.flightsafety.org/asndb/328849

https://asn.flightsafety.org/asndb/327151

https://www.jal.com/en/jal-int-70th/history/

https://www.researchgate.net/figure/Number-of-passengers-on-scheduled-domestic-and-international-flights-offered-by-Japanese_fig1_240167842

https://en.namu.wiki/w/%EB%B3%B4%EC%9E%89%20747-100

https://en.namu.wiki/w/%EC%9D%BC%EB%B3%B8%ED%95%AD%EA%B3%B5%20123%ED%8E%B8%20%EC%B6%94%EB%9D%BD%20%EC%82%AC%EA%B3%A0#s-3

https://hoodcp.wordpress.com/2016/01/15/the-jl123-isho/