South African Airways Flight 295

South African Airways Flight 295
	The crashed Boeing 747 "Helderberg" in 1986 at Faro Airport
Accident summary
Accident type	Loss of control or structural failure after fire
place	Indian Ocean
date	November 28, 1987
Fatalities	159
Survivors	0
Aircraft
Aircraft type	Boeing 747-200B Combi
operator	South African Airways
Mark	ZS-SAS
Surname	Helderberg
Departure airport	Chiang Kai-shek International Airport , Taipei
Stopover	Plaisance Airport , Mauritius
Destination airport	Jan Smuts Airport , Johannesburg
Passengers	140
crew	19th
	Lists of aviation accidents

On November 28, 1987, a Boeing 747 crashed on South African Airways Flight 295 as a result of a fire that broke out in the cargo area of the main deck, around 150 miles northeast of the island of Mauritius in the Indian Ocean . The South African Airways machine was on a scheduled flight from Taipei to Johannesburg . The crash occurred at 4:07 a.m. local time (00:07 UTC ), about 25 minutes before the scheduled stopover at Plaisance Airport in Mauritius. All 159 inmates were killed in the accident.

The Republic of South Africa entrusted Judge Cecil Margo to chair the commission of inquiry. According to his final report, the fire started from a cargo pallet that ignited during the flight. The actual cause of the fire remained unexplained.

The Association of South African Flight Engineers questioned the official accident report. In their opinion, the recorded cockpit conversations and the behavior of the pilots indicated that a first fire broke out on board earlier in the flight and the crew expected further problems. Independent experts have shown that the intensity of the fire was increased by an oxidizing agent . In 1998, the South African forensic scientist Dr. David Klatzow presented a chain of circumstantial evidence before the Truth and Reconciliation Commission (TRC) , according to which the machine had transported ammonium perchlorate on behalf of the state, circumventing the arms embargo that was in force at the time , and which had self-ignited in flight. The TRC Commission accepted this line of argument in part and came to the conclusion that the fire was unequivocally triggered by a cargo that was not recorded on the official shipping documents. A new state investigation of the incident has not yet taken place.

At the end of 2014, the South African investigative journalist Mark D. Young presented his own theory that a cable fire was the cause of the accident.

Combined passenger and cargo transport

In 1980, South African Airways had set up a regular service between Chiang Kai-shek International Airport in Taipei and Jan Smuts Airport in Johannesburg under flight numbers SA294 and SA295 . In order to be able to carry out the flights with sufficient payload , the route was routed in both directions from the mid-1980s via Plaisance Airport in Mauritius . There the machines were refueled before the onward flight. The flight time from Taipei to Mauritius was around 10 hours. Due to the high volume of freight, mainly Boeing 747-200 Combi aircraft were used on the route, of which the company operated two aircraft in 1987.

Plane and cargo

The Boeing 747-244B Combi ( registration number : ZS-SAS, c / n : 22171, s / n : 488) with the name "Helderberg" was delivered to the airline on November 12, 1980 and at the time of the accident had 26,743 flight hours completed. The aircraft enabled passengers and up to seven cargo pallets to be carried on the main deck at the same time. A partition wall with two doors separated the cargo area in the rear from the passenger cabin. Freight nets stretched behind the wall secured the cargo from slipping, but made it difficult for the crew to access this area.

The "Helderberg" was loaded in Taipei with six freight pallets, which were arranged in pairs on the main deck. The cargo consisted mainly of computer components, electrical appliances and sporting goods. Officially, there were no dangerous and / or flammable goods among them. A random check of the cargo carried out by Taiwanese customs before departure was without any objections.

Passengers and crew

nationality	Passengers	crew	Total
Australia Australia	2	0	2
Denmark Denmark	1	0	1
Germany BR BR Germany	1	0	1
Hong Kong 1959 Hong Kong	2	0	2
Japan Japan	47	0	47
Korea South South Korea	1	0	1
Mauritius Mauritius	2	0	2
Netherlands Netherlands	1	0	1
South Africa 1961 South Africa	52	19th	71
Taiwan Taiwan	30th	0	30th
United Kingdom United Kingdom	1	0	1
Total	140	19th	159

There were 140 passengers on board the machine. The passenger cabin was thus 68 percent full. Some other people, including the South African motor sportsman George Fouché , were booked on the flight, but did not make the trip because their feeder flights were canceled due to weather conditions or arrived late in Taipei.

The crew consisted of 19 people. Due to the long flight duration, three pilots and two flight engineers were assigned to the machine , who took turns on the way. The responsible captain on board was Dawid Jacobus Uys (49 years old, with 13,843 flight hours).

Fire protection concept on board

In contrast to the underfloor holds , the Boeing 747-200 Combi did not have any automatic extinguishing systems in the loading zone of the main deck. The fight against a freight fire should be carried out by the cabin crew using hand fire extinguishers and wearing respiratory protection masks . There were a total of eleven hand-held fire extinguishers on board, distributed over various cabin areas.

In order to protect passengers from smoke gases in the event of a fire , the cabin pressure in the cargo area was lowered slightly during the flight. The manufacturer Boeing had to prove the effectiveness of this concept when the Boeing 747 Combi was approved for traffic. The admission test was carried out with an inflamed ball of tobacco . The test fire could be extinguished quickly. In addition, during the fire simulation, no smoke gases flowed into the passenger zone due to the pressure difference.

After the South African plane crashed, the US Federal Aviation Administration (FAA) carried out another test under more realistic conditions, simulating a fire in a full hold with typical cargo items. The manufacturer's protection concept turned out to be ineffective. The higher temperature of the fire canceled the pressure difference between the cabin sections, so that the smoke gases spread throughout the aircraft. Furthermore, the cabin crew was barely able to fight the fire effectively because of the limited space, the smoke density and the heat. The FAA then set new fire protection regulations for all series of the Boeing 747 Combi in 1993. The retrofitting of the aircraft reduced their profitability, so that numerous companies renounced a combined operation and then used the aircraft in a pure passenger or cargo design.

Course of the flight and course of the accident according to the official presentation

Report of the Board of Inquiry into the Helderberg Air Disaster , viewable copy of the official accident report.

The Boeing 747 took off from Taipei at 10:23 p.m. local time (14:23 UTC ) due to weather conditions with a delay of over an hour. During the flight, the crew routinely reported to the various air traffic control points along the route. The radio messages gave no indication of an existing problem. About 90 minutes after take-off, the crew informed the technical operations center (ZUR) of South African Airways in Johannesburg about the planned arrival time in Mauritius. The stopover there should be at 4:30 a.m. local time (00:30 UTC).

The pilots first contacted air traffic control in Mauritius at 3:14 a.m. local time (23:14 UTC) and reported their current position (65 degrees East) and altitude (35,000 feet ). About 45 minutes before the stopover in Mauritius (the exact time could not be determined), a fire alarm sounded in the cockpit, which informed the crew of a fire in the cargo area of the main deck. The fire alarm was turned off very quickly. Four seconds later, a flight attendant also reported the fire via the on-board intercom. A further seven seconds later, a second smoke detector was triggered, which was located in a further back area of the cargo hold ( "It came on now afterwards." ). The flight engineer asked via the intercom connection whether the fire extinguisher from the cockpit was needed to fight the fire, which the flight attendant replied in the affirmative. The fire extinguisher was brought to the stern by the second flight engineer. 34 seconds after the alarm was triggered, the master ordered the cabin fire checklist to be worked through. At the same time, up to 80 electronic fuses failed because the fire destroyed the lines running in the stern.

The crew then initiated an emergency descent. At 3:49 pm local time (23:49 UTC), the pilots informed the approach control in Mauritius that they are due to a "smoke problem on the plane" in the descent (about 4,200 meters) were on 14,000 feet. The pilots only officially declared an emergency when asked. The captain could not tell the air traffic controller the distance to the destination airport because the electronic on-board systems had failed ( "Now we've lost a lot of electrics, we haven't got anything on the aircraft now." ). After the machine had reached an altitude of 14,000 feet, the crew initiated a ventilation of the cabin, which was filled with smoke and combustion gases, although the procedure actually requires the fire to be extinguished beforehand. Using the smoke evacuation checklist , the airspeed was reduced to below 200 knots and two outer doors were opened. At 4:01 am local time (00:01 UTC), the air traffic control captain reported that the doors had been opened ( “Eh, Plaisance, Springbok Two Nine Five, we've opened the door (s) to see if we (can ) ... we should be okay " ). The crew received clearance to continue the descent to an altitude of 5,000 feet (1,500 meters). While the commander answered the radio messages, the co-pilot gave a distance specification ( “We're now sixty five miles” ). The air traffic controller related this information to the distance to the airport. In fact, the copilot meant the remaining distance to the next waypoint ( waypoint XAGAL ). The distance to the airport was about 270 kilometers (145 NM ). The air traffic controller then transmitted the weather and landing information for Mauritius. These were confirmed by the pilots at 4:04 a.m. local time (00:04 UTC). At 4:08 a.m. local time (00:08 UTC), air traffic control tried again to contact the aircraft. However, the crew no longer answered.

Search for the crash site and recovery of wreckage

Around two and a half hours after the last radio contact, planes and ships from several nations began searching for the machine. Due to the misleading distance information, this initially started in the wrong sea area. In the late afternoon of November 28, an aircraft sighted objects about 250 kilometers (136 NM) northeast of Mauritius. On November 30, ships reached the rubble field and began to recover victims, luggage and lighter fragments of the machine. The search continued through December 10, 1987. Some of the objects and debris recovered showed significant fire damage or were sooty . A total of 15 bodies were recovered.

Three ships of the South African Navy reached the sea area on December 11, 1987 and tried to locate the radio signals from the flight recorders . The promotion was discontinued on January 2nd without success. In parallel, South Africa commissioned a civilian company to search the seabed with sonar equipment from January 25th . The research vessel deployed discovered two main debris fields in 4,400 meters of water on January 28, 1988.

The northern main field had an extension of 900 meters. The field consisted of the shattered rear fuselage (from the fourth outer door), the tail unit , the loading door , numerous cargo parts and various pieces of the assemblies and systems located in the rear. The center of the second main field was about 600 meters further south-west. The southern area contained, among other things, the debris of the front fuselage, the wing structure , the engines and the landing gear . The degree of destruction of the finds in the southern area was significantly higher than in the northern area. There was a 200 meter wide zone between the sections that was almost free of wreckage. At a distance of 1.5 miles, a third field of debris was located made up of lighter objects that had drifted there as they sank from the surface of the sea. The position of individual pieces was recorded with cameras, including the position of the cockpit voice recorder (CVR). Due to the depth of the water, it was initially not possible to recover the objects.

On January 6, 1989, the rescue work began with the help of a modified diving robot . The cockpit voice recorder could be lifted on the first day of the operation. The second recording device on board, the flight data recorder , however, has disappeared. In an action lasting several weeks, the wreckage of the rear part of the fuselage and other debris were recovered. The pieces were examined in South Africa.

The cockpit voice recorder, damaged by water ingress, was brought to the USA and opened and evaluated there by employees of the National Transportation Safety Board (NTSB) in the presence of Rennie Van Zyl, the chief accident investigator and director of the South African aviation authority.

Cause of accident

The fire started from the front right freight pallet, which according to the loading list contained computers and electronic components. The fuselage sections next to and above this pallet showed the greatest fire damage. The floor below showed no signs of fire. This suggested that the fire broke out at the top of the cargo pallet. It could not be determined what caused the fire.

According to official reports , a smoldering fire initially developed that spread to the packaging material. The hot smoke gases collected under the cabin ceiling and caused a flash over . The fire subsequently destroyed the electrical lines, so that numerous on-board systems failed, including the Cockpit Voice Recorder (CVR), whose recordings ended just 81 seconds after the alarm sounded.

The cargo fire reached a high temperature due to the packaging material. Numerous recovered objects, including a tennis racket made of Kevlar ( melting point : 600 degrees Celsius), showed significant traces of fire. The temperature on the outside of the fuselage rose to around 300 degrees Celsius in this area, which led to the formation of bubbles in the paintwork. In addition, the packaging material caused a lot of smoke. The smoke and the fire gases flowed into the passenger zone. The autopsy of the 15 recovered victims revealed that at least two of them had suffocated from the fire gases before the impact, including a passenger whose seat was in the front part of the cabin in business class . The airways of all corpses contained soot particles.

The crew carried out a fire fight, but probably gave up because of the heat and / or smoke. A still full hand-held fire extinguisher was found to which remnants of the melted cargo net were stuck. The fire extinguisher came from the front area of the passenger cabin (position: second right outer door).

Due to the intensity of the fire, the cable systems of the rudder and elevator warped , which presumably made it increasingly difficult for the crew to control the aircraft.

The smoke evacuation checklist was used to ventilate the cabin filled with smoke gases , which actually requires the fire to be extinguished beforehand. The crew probably - as provided in the checklist - put the entire ventilation back into operation, so that the fire gases were more widely distributed in the aircraft despite the outer doors being open. At 4:02 a.m., flight captain Uys ordered the cockpit door to be closed because smoke apparently entered the upper deck. The investigators did not rule out that the pilots were passed out by the gases from the fire or that the incoming outside air intensified the fire.

The aircraft hit the sea at 4:07 a.m. local time (00:07 UTC). Two analog wristwatches that were found together in the same piece of luggage had stopped at this point.

Due to the two fields of debris on the sea floor, the investigators assumed that the machine probably broke before the impact at low altitude. Loss of control or structural failure of the rear torso was considered to be the cause.

Representation of the Association of South African Flight Engineers

The Cockpit Voice Recorder (CVR) had a recording time of 30 minutes. The first 28.5 minutes of the CVR recording (all cockpit conversations until the fire alarm sounded) were not published by the Margo Commission and are also missing in the official final report. According to the Commission, this should protect the privacy of the pilots. In addition, in the opinion of the Commission, the discussions did not contain any content that was relevant to the investigation or to the public.

The Association of South African Flight Engineers ( Suid Afrikaanse Lugdiens Boordingenieursvereiniging ) received a complete recording of the conversation and then questioned the official description of the course of the flight. In the opinion of the association, the CVR records were not complete, but came from two different flight phases. The first segment of the conversation was 28:31 minutes and was probably recorded about two hours after the start. The device then temporarily failed. The second part of the recording starts when the fire alarm is triggered and ends 81 seconds later. The association justified its view as follows:

In the first phase of the recording, there were two pilots and a flight engineer in the cockpit. After the fire alarm was triggered, however, all three pilots and both flight engineers were present. The entry of the two remaining crew members was not recorded by the CVR.
In the unpublished conversations, the three crew members discussed, among other things, the dinner that was to be served shortly. One pilot stated that he was hungry and was waiting for dinner. The flight captain, who was allergic to certain foods, told his colleagues that he had to be particularly careful when choosing the dishes on offer. The men used only the word “dinner” in their conversation . Dinner on this route is served around two hours after take-off in Taiwan, but never in the early morning, one hour before landing in Mauritius.
Neither the first radio contact with Mauritius at 3:14 a.m. nor the emergency call sent at 3:49 a.m. were recorded by the CVR. According to the Margo Commission , the CVR recording began shortly after the first radio message (around 3:15 a.m.). The alarm sounded around 28 minutes later (at 3:43 a.m. at the earliest). The aircraft remained at cruising altitude until the end of the CVR recording (at least until 3:45 a.m.). Up to this point in time, the crew had neither started to initiate the emergency descent nor to work through the cabin fire checklists. According to the flight engineers, processing the emergency checklists took at least seven minutes. However, the crew made the emergency call just four minutes later (at 3:49 a.m.). The checklists were not read during the radio contact with the air traffic controller in Mauritius.

The flight engineers' association assumed that a first fire on board led to an interruption of the CVR recording. In their opinion, the behavior and communication of the pilots as well as the procedures on board indicated that the crew expected further difficulties after the first fire. The association justified its assumption as follows, among other things:

The pilots reacted to the fire alarm extremely quickly. The alarm bell was turned off immediately without the crew having to find out where the fire was.
Only four seconds after the fire alarm had sounded, the cabin crew reported the fire via the on-board intercom. This indicated that at least one crew member was positioned at the door to the cargo area and was on guard there.
The flight engineer was apparently immediately aware that the number of fire extinguishers on the main deck was not (no longer) sufficient to extinguish the fire, so he offered the flight attendant to provide the fire extinguisher from the cockpit.
Initially, the pilots did not seem to be surprised by the triggering of the fire alarm. Only after the second smoke detector had triggered the alarm did the crew become increasingly nervous.

Cecil Margo, chairman of the official commission of inquiry, called on the flight engineers' association to retract its account of the events in order to avert claims for damages from South African Airways . At a meeting with representatives of the association, he explicitly mentioned the sum insured of 400 million rand that the company might have to repay.

Hearing before the TRC Commission

On June 1, 1998, the crash of the "Helderberg" was the subject of a hearing by the Truth and Reconciliation Commission (TRC). Among others, the South African forensic scientist Dr. David Klatzow, who carried out his own investigations after the crash and collected numerous statements from employees of South African Airways .

Former pilots had reported to Klatzow that aircraft from South African Airways , which at the time was still state-owned, were regularly used by the ARMSCOR company to import and export armaments and that the crews would face reprisals if they refused to board. The state armaments company ARMSCOR was subordinate to the South African Ministry of Defense. During these transports, which violated the UN embargo and the IATA regulations, there were several incidents.

Klatzow assumed that rocket fuel containing ammonium perchlorate or ammonium perchlorate itself was on board the "Helderberg" . South Africa needed this oxidizer to make solid propellant for anti-aircraft missiles in the war against Angola . Ammonium perchlorate, which can cause spontaneous combustion in contact with flammable materials , was subject to the UN embargo and was therefore not listed in the official shipping documents. According to Klatzow's testimony, witnesses reported that flight captain Uys had informed the company's technical operations center in Johannesburg (ZUR) about a fire on board over the South China Sea , a few hours after take-off from Taiwan. Uys asked to be allowed to take the plane to Changi Airport in Singapore . This request was forwarded from the operations center to government circles, which declined the request. Klatzow suspected that the crew managed to put out the first fire, but there were injuries and the machine was damaged. An emergency landing in Singapore would have led to an investigation by the authorities there, which they wanted to avoid. The pilots had to continue the flight to Mauritius. About 45 minutes before the stopover there, the material ignited again. The crew, who had already emptied most of the fire extinguishers during the first fire, no longer had sufficient resources to effectively fight the second fire. Laboratory tests of the wreckage showed that the fire reached a temperature of over 1000 degrees Celsius. According to fire experts, the high temperature could only be explained by the presence of an oxidizing agent. The burning substance supplied itself with oxygen , similar to a sparkler .

The radio messages from all South African Airways aircraft were stored on magnetic tapes by the operations center in Johannesburg (ZUR) . Internal company regulations stipulated a general archiving period of 30 days for all magnetic tapes. The tape with the magnetic recordings from the day of the accident was requested by the investigators shortly after the incident, but could not be found. The Margo Commission did not investigate the whereabouts of the tape. According to a statement by Klatzow, employees at the operations center had reported that the ZUR station manager handed the tape, which was initially locked up, to Gert van der Veer, the then director of South Airican Airways , a few days after the incident .

Gert van der Veer was invited to the TRC hearing himself and in a subsequent cross-examination denied having received the magnetic tape. However, he thought it possible that the tape in question had been accidentally reused and re-recorded.

Doubts about Klatzow's theory

Critics of Klatzow's theory pointed out that South Africa had been producing ammonium perchlorate (AP) in Somerset West near Cape Town since 1970 and has been self- sufficient ever since. South unneeded AP only as a solid propellant for anti-aircraft missiles , but also for the development of long-range missiles to nuclear weapons to carry. Klatzow said he had found indications that the then ARMSCOR subsidiary Somchem could not keep up with production and urgently needed supplies from abroad. Although there had been brisk shipping traffic between Israel and South Africa in the 1980s , Klatzow claimed that solid propellants containing ammonium perchlorate were also flown in from Israel by cargo planes, among other things. Under normal conditions and in its pure form, e.g. B. the US Air Force AP also as air freight. Another argument against pure ammonium perchlorate as the primary fire source is that it tends to explode at temperatures of over 200 ° C. Clearly no explosion had taken place on board the “Helderberg” .

Newer theories

In his theory, presented in 2014, the South African investigative journalist Mark D. Young suspects that there may have been an electrical short circuit in the on-board electronics. So-called " wet arc tracking " occurs when the insulation of a live line is damaged, for example, and consists of a leakage current to another damaged line with a corresponding potential difference or a structure leading to ground potential . The resulting arc flashover can reach temperatures of up to 5000 ° C. This temperature is sufficient to ignite the thermoacoustic insulation mats that were used until the late 1990s. Such a short circuit was also the cause of a fire on board Swissair Flight 111 , which caused the plane to crash in 1998.

Web links

Individual evidence

↑ ^a ^b ^c ^d ^e ^f ^g ^h ⁱ ^j ^k ^l ^m ⁿ ^o ^p ^q ^r ^s ^t Republic of South Africa, Report of the Board of Inquiry into the Helderberg Air Disaster ( Memento of the original from February 17, 2015 in the Internet Archive ) Info: The archive link was inserted automatically and has not yet been checked. Please check the original and archive link according to the instructions and then remove this notice. (in English), accessed February 17, 2015 @1@ 2

↑ ^a ^b ^c ^d ^e ^f ^g ^h ⁱ ^j ^k ^l Truth and Reconciliation Commission, Special Hearing, Helderberg Flight ( Memento from October 30, 2004 in the Internet Archive ) (in English)

^ Homepage of MD Young (in English)

↑ politicsweb (in English), accessed February 20, 2015

↑ JP airline-fleets international, Edition 87

↑ Aircraft accident data and report from South African Airways Flight 295 on the Aviation Safety Network , accessed on January 4, 2015.

^ Federal Aviation Administration, Airworthiness Directive December 18, 1989

^ Federal Aviation Administration, Airworthiness Directive July 15, 1993

^ South African History Online, Jan. 6, 1989

↑ Ammonium perchlorate, safety data sheet in accordance with 1907/2006 / EC, Article 31, page: 2/5 ( Memento of the original from March 5, 2016 in the Internet Archive ) Info: The archive link was inserted automatically and has not yet been checked. Please check the original and archive link according to the instructions and then remove this notice. @1@ 2
^ US Air Force air shipments , accessed February 19, 2015
↑ ChemPage: Ammonium Perchlorate , accessed on February 19, 2015

-19.168055555556 59.615833333333Coordinates: 19 ° S , 60 ° E

[Margo-1] ↑ ^a ^b ^c ^d ^e ^f ^g ^h ⁱ ^j ^k ^l ^m ⁿ ^o ^p ^q ^r ^s ^t Republic of South Africa, Report of the Board of Inquiry into the Helderberg Air Disaster ( Memento of the original from February 17, 2015 in the Internet Archive ) Info: The archive link was inserted automatically and has not yet been checked. Please check the original and archive link according to the instructions and then remove this notice. (in English), accessed February 17, 2015 @1@ 2

[TRC-2] ↑ ^a ^b ^c ^d ^e ^f ^g ^h ⁱ ^j ^k ^l Truth and Reconciliation Commission, Special Hearing, Helderberg Flight ( Memento from October 30, 2004 in the Internet Archive ) (in English)

[Young-3] Homepage of MD Young (in English)

[politicsweb-4] ticsweb (in English), accessed February 20, 2015

[5] JP airline-fleets international, Edition 87

[aviation-6] Aircraft accident data and report from South African Airways Flight 295 on the Aviation Safety Network , accessed on January 4, 2015.

[7] Federal Aviation Administration, Airworthiness Directive December 18, 1989

[8] Federal Aviation Administration, Airworthiness Directive July 15, 1993

[9] South African History Online, Jan. 6, 1989

[10] Ammonium perchlorate, safety data sheet in accordance with 1907/2006 / EC, Article 31, page: 2/5 ( Memento of the original from March 5, 2016 in the Internet Archive ) Info: The archive link was inserted automatically and has not yet been checked. Please check the original and archive link according to the instructions and then remove this notice. @1@ 2

[11] US Air Force air shipments , accessed February 19, 2015

[12] ChemPage: Ammonium Perchlorate , accessed on February 19, 2015