Friday April 25, 1980
Dan-Air Flight 1008
Flight 1008 was a fatal accident involving a Boeing 727-46 jet aircraft, operated by Dan Air Services Limited, on a non-scheduled international passenger service from Manchester to Tenerife in 1980. The accident killed 138 passengers and 8 crew members, because of a single misunderstanding.
Communication between pilots and Air Traffic Control is essential. When this process breaks down, disaster often follows. This accident marked the greatest loss of life in an air disaster involving a British-registered aircraft, as well as becoming Dan-Air’s second major accident in 10 years, and the worst incident killing fare-paying passengers in the airline’s entire history.
This was listed as one of the worst air disasters caused by miscommunication, which continued into the investigation, where blame was laid on the crew of the flight. While nothing about this flight went smoothly, one confusing and crucial line from an air traffic controller was skimmed over in the last report. This discrepancy was covered heavily in a documentary about the crash, available in parts on youtube channel DanAirCrash, also containing actual crash scene footage and audio from the cockpit. See still images above.
Dan-Air flight 1008 took off from Manchester Airport (MAN) on Friday, April 25, 1980 at 09:22 UTC, en route for Tenerife-Norte Los Rodeos Airport (TFN).
After an uneventful flight, the crew contacted Tenerife North Airport Approach Control at 1:14PM GMT, informing them that they were at FL110 and at 14 nautical miles from the TFN VOR/DME. Approach Control replied, “Dan Air one zero zero eight, cleared to the Foxtrot Papa beacon via Tango Foxtrot November, flight level one one zero expect runway one two, no delay.”
The Dan Air crew repeated the clearance and requested meteorological information, which was given as: “OK runway in use one two, the wind one two zero zero five, visibility six from seven kilometres clouds, two oktas at one two zero metres, plus four oktas at two five zero metres, plus two oktas at three five zero metres, November Hotel one zero three, temperature one six, dew point one, and drizzle.”
Approximately one minute later Approach Control told the aircraft to descend, and maintain FL60. Receipt of this message was acknowledged by the aircraft, whereupon the controller immediately requested it to indicate its distance from the TFN beacon. The crew replied that it was at 7 NM from TFN.
At 1:18:48 UTC the aircraft notified Approach Control that it had just passed TFN and that it was heading for the ‘FP’ beacon. The controller then informed them of an unpublished hold over Foxtrot Papa: “Roger, the standard holding over Foxtrot Papa is inbound heading one five zero, turn to the left, call you back shortly.”
Dan Air 1008 only replied “Roger” without repeating the information received, which was not compulsory under the ICAO regulations in force at the time of the accident.
Almost one minute later, the aircraft the crew reported: “Dan Air one zero eight, Foxtrot Papa level at six zero, taking up the hold” and Tenerife APP replied: “Roger”. Instead of passing overhead FP, the flight had passed this navaid at 1.59 NM to the South.
Instead of entering the 255 radial, the Boeing 727 continued its trajectory in the direction of 263 degrees for a duration of more than 20 seconds, entering an area with a minimum safety altitude (MSA) of 14500 ft.
The co-pilot at that point said: “Bloody strange hold, isn’t it?”
The Captain remarked, “Yes, doesn’t parallel with the runway or anything.”
The flight engineer then also made some remarks about the holding procedure. Approach control then cleared them down to 5,000 feet. The captain then remarked: “Hey did he say it was one five zero inbound?”
It appears that at this moment the information received on the holding flashed back to the Captain’s mind, making him realize that his manoeuvre was taking him to magnetic course 150 degrees outbound from ‘FP’, whereas the information received was “inbound” on the holding, heading 150 degrees towards ‘FP’.
The copilot responded: “Inbound yeh”.
“I don’t like that”, the captain said.
The GPWS alarm sounded. The captain interrupted his left hand turn and entered a right hand turn and ordered an overshoot. They overflew a valley, temporarily deactivating the GPWS warning.
The copilot suggested: “I suggest a heading of one two two actually and er take us through the overshoot, ah.” But the Captain continued with the turn to the right, because he was convinced that the turn he had been making to the left was taking him to the mountains.
Suddenly the 727’s ground warning indicators began sounding, pointing out a mountain that cloud-obscured eyes couldn’t see. The captain contacted Approach Control at 13:21: “Er … Dan Air one zero zero eight, we’ve had a ground proximity Warning.”
About two seconds later the aircraft the plane ploughed into the side of the island’s Mount La Esperanza, at an altitude of approximately 5,450 ft (1,662 m), killing all 146 people on board. They were 11.5 km off course. Just before impact, the air-craft had pulled up and banked sharply to the right.
The Aircraft and Wreckage
As a result of the violence of the impact, as well as the nature of the terrain, the aircraft was totally destroyed. Damage was estimated at 100%. Damage took place on the accident area, consisting of the destruction of trees, mostly pines, various fires and the contamination of the area by human remains and spilled fuel.
The aircraft had refuelled with 49 800 lb, and the crew had carried out usual pre-flight operations. They had also requested and received appropriate meteorological information. The Boeing aircraft 727-100146, registered G-BDAN, was built in 1966 and certified for passenger transport. It was a low-winged monoplane, of metal structure and skin, propelled by three Pratt & Whitney JT8D-7 engines located in the tail of the aircraft. The landing gear was three-wheeled and retractable, with pressurized cabin.
The greater part of the three engines was identified. The damage to the revolving parts, and the way in which they had been broken, indicated that they had all been at full power at the moment of impact. The three devices for thrust reverse were in a position of forward thrust.
The ICAO Circular 178-AN/lll gives more information:
Impact and Wreckage Information
“The location of the initial impact was 28°23’53″N and 16°25’05″P, at 11.5 km on magnetic bearings of 222° from the approach end of runway 12 of Tenerife North, according to the military map of Spain 20/20.
This first impact took place at an altitude of 1,662 m (5,450 ft), approximately 38 m below the summit of a peak whose height is 1,690 m on a course of impact of 250° to 260° magnetic. The maximum height of the mountain is 1,752 m (5,748 Et) – Pico del Chiriguel.
The aircraft crashed obliquely on a slope located on a magnetic bearing of 033° and a 30° gradient relative to the horizontal. The actual gradient of the flight trajectory relative to the mountain was thus 20°.
The slope against which the impact took place forms the side of a valley whose floor is at approximately 1,310 m (4,300 ft) above sea level, running in a northwest to southeast direction.
The north-east side of the valley is formed by another peak of a height of approximately 1,464 m (4,800 ft).
The aircraft impact took place in a relatively large clearing, just a short distance before a densely forested area. A few small trees were broken off immediately before the impact against the ground, and the right wing tip grazed a rock. The marks made on the ground were covered up by a landslide caused by the impact.
The distribution of the wreckage indicated that most of the breakage suffered by the aircraft took place during the initial impact, although there were signs that a large section of the rear end of the fuselage, immediately behind the wing, and possibly with the empennage attached, remained relatively intact and was thrown several hundred metres away, starting a second trail of wreckage which included passenger seats and their occupants.
The main trail of debris was approximately 350 m long, stretching up the slope to about 50 m above the initial impact. The maximum width was about 200 m, although a large quantity of debris had rolled down to a considerable distance from the place of their first impact against the ground.
In general terms, the first part of the main trail included a large portion of the wing structure and the engines, the main left landing gear, the tips of the horizontal stabilizers and elevators, and some components of the crew cockpit. The trail continued with additional components of the crew cockpit, passenger luggage, the front landing gear, small pieces of the fuselage, wing, engines and a large portion of the central section of the plane and tail empennage, which were relatively intact except for the tips of the horizontal surface.The last part of the main trail was made up of a large number of pieces of the central and back section of the fuselage, as well as of electronic components.
The secondary trail continued for a further 250 m, going up 15 m above the cliff and 45 m down the other side, and crossing over the road.
A large quantity of small light fragments of material and burnt skin plate was recovered in the vicinity of “Las Lagunetas”, a village located at an altitude of 1,440 m (4,700 ft) above sea level and at 2 km, from the scene of the accident, according to a 390 magnetic bearing. A few similar fragments were recovered from the wood located between “Las Lagunetas” and the area of the impact.
In the area of the initial impact and its surroundings, the layer of earth immediately below the surface was strongly impregnated with fuel. Also, a large part of the vegetation in the surroundings showed signs of small fires caused by the fuel spilled during the impact.”
The coordinates of the impact area are 280 23’53” North and 16’ 25’05” West, with a bearing of 22Z0 magnetic from the threshold of runway 12 of Tenerife North, at an altitude of approximately 5 450 ft (1 662 m) and at 11.5 km off route.
138 passengers and 8 crew members perished in the disaster.
Approximately 50% of the passenger safety belts were found fastened, and others showed signs of having been fastened during the impact. It was determined that some seat belts had been unfastened by the rescue personnel.
Due to the total destruction of the aircraft, and the extensive area over which the wreckage was spread, it was very difficult to identify the casualties. Very few bodies showed signs of burns, but they were all seriously mutilated. The identification was achieved with the help of fingerprints and dental records.
Although it was not possible to identify the technical crew, their medical records gave no reason to suppose that they were not medically fit.
The Captain was born in the United Kingdom on October 11, 1929. He held a valid Airline Transport Pilot’s Licence, No. 51676 and obtained a Boeing 727 type rating in December 1976. All other relevant operating certificates were valid. His total flight time was 15,299 hours and he had 1,912 hours experience on the Boeing 727 type. During the last 30 days he had flown 37 hours and during the same period had recorded 86.49 hours duty time. He had flown into Tenerife North 58 times previously, the preceding flight being on January 29, 1980.
The Co-pilot was born in the United Kingdom on June 20, 1946. He held a valid Commercial Pilot’s Licence No. 95933 and obtained a type rating on the Boeing 727 in March 1979. All other operating certificates were valid. His total flight time was 3,492 hours of which 618 hours were in the Boeing 727. In the last 30 days he had flown 38.59 hours and during the same period had been on duty for 82.59 hours. He had flown into Tenerife North 9 times previously, the preceding flight being on December 22, 1979.
The Flight Engineer was born in the United Kingdom on July 11, 1946. He held a valid Flight Engineer’s Licence No. 2131 and obtained a type rating on the Boeing 727 in April 1978. He had flown a total of 3,341 hours of which 16.53 hours were flown during the last 30 days.
The Approach Controller was born in Santa Cruz de Tenerife (Spain) on February 22, 1946. He held Air Traffic Controller Licence No. AOITC 448 issued in October 1973. Ratings for the Tenerife North Airport and taxiing area were obtained in December 1978 and Tenerife North Approach in February 1979. A valid medical certificate was also held.
Once the alarm was given, after communications had been cut off between APP and the aircraft, a normal search procedure was carried out by the Air Rescue Service and the Civil Guard on the ground. The search lasted several hours.
Some people travelling along the road from La Laguna to the Teide found pieces of the aircraft which, at first, they did not recognize, nor did they pay them much attention, until they heard the news of the aircrafts disappearance over the radio, whereupon they reported them.
The aircraft was located at 2000 hours local time by members of the Civil Guard, and after an initial inspection the area was immediately cordoned off and some bodies recovered. This work was interrupted until the following day on account of nightfall and the nature of the place and the terrain. During the night, various groups for rescuing the bodies were organized and co-ordinated and they started work the following day, the 26th, at 7 a,m. They worked during the 26th and 27th, recovering human remains. The operations of search and rescue were undertaken by members of the Civil Guard, the National Police, the Army, the Spanish Red Cross, volunteers from mountaineering groups, airport personnel, etc., all 05 whom were co-ordinated and directed by the Delegate of the Civil Governor of Tenerife, They worked during the three days required for the recovery of human remains, while the Civil Guard continued to guard the aircraft: debris during the time it was being investigated.
To commemorate those who lost their lives in the tragedy, The Memorial Rose Garden was created at the front of the church grounds where, every year on the anniversary of the crash, prayers are said for the departed and their families.
A memorial in Southern Cemetery, Manchester commemorates the victims of the disaster, whose names are inscribed on a series of slate tablets within a small grassed enclosure.
Shortly after the crash near Tenerife’s Los Rodeos Airport, local Spanish officials voiced the opinion that the disaster was caused purely by pilot error.
The official cause of the disaster was a misinterpretation, blamed mostly on the Boeing 727’s flight crew. The plane entered a holding pattern, awaiting its turn to land, informed to take an unpublished, not officially approved, and potentially dangerous holding pattern above Los Rodeos Airport by traffic control. The pilot seems to have mistaken the word “inbound” for “outbound” in the instructions he received, flying in the opposite direction to which he was supposed to.
This turn in the wrong direction took the plane through an area of exceptionally high ground, and due to the airport’s lack of ground radar, the air traffic controllers were unable to tell the flight crew that the plane was off course.
Heavy clouds obscured the crew’s vision, likely preventing them from seeing the looming threat of the mountain. The first sign they had of any impending danger was when the plane’s ground proximity warning device was triggered. The crew attempted a steep climb, but the aircraft slammed into the mountainside, killing everyone on board instantly.
Flight International on June 14, 1980 discussed further cause:
Much of the current concern about Tenerife’s ATC capability centres on the fact that neither of the island’s two airports (Los Rodeos and Queen Sofia) has surveillance radar. Aircraft flying in and out are directed to follow one of several published procedures while ATC monitors the whereabouts of aircraft by listening to position reports the pilots transmit as they pass each beacon or designated reporting point. The procedures at Tenerife are, in themselves, simple and safe, keeping air-craft well clear of high ground with a reasonable margin for error. So although surveillance radar would be desirable and would definitely raise safety margins, it is not actually essential. Radar could have prevented this accident because such a gross deviation from the ordered flight path could hardly fail to be noticed. From all the unofficial detail known to date, three factors seem to have combined to cause this accident: the short notice given the crew to carry out an unexpected procedural manoeuvre; the non-availability of a relevant chart; and the failure of the captain to query the ATC instruction if he was confused by it.
The report laid full blame on the crew:
The captain, without taking into account the altitude at which he was flying, took the aircraft into an area of very high ground, and for this reason he did not maintain the correct safety distance above the ground, as was his obligation.
Contributing factors were:
a) the performance of a manoeuvre without having clearly defined it;
b) imprecise navigation on the part of the captain, showing his loss of bearings;
c) lack of teamwork between captain and co-pilot;
d) the short space of time between the information given and the arrival at ‘FP’;
e) the fact that the holding was not published” (Spanish report)
UK authorities agreed in general with the report, but added some comments to give the report ‘a proper balance’:
1. The information concerning the holding pattern at FP, which was transmitted by ATC, was ambiguous and contributed directly to the disorientation of the crew.
2. No minimum safe altitude computed for holding pattern.
3. Track for holding pattern at ‘FP’ is unrealistic.
Due to the overwhelming amount of information, this entire transcript is copy and pasted from the official report on the incident:
Since there is no evidence of any functional abnormality in the aircraft itself, this analysis will concentrate mainly on the communications with Tenerife Norte approach control (APP) from the time it was first contacted by the aircraft, and in particular the communications and conversations taken from the CVR for the last five minutes before the accident, i.e. from the point when the aircraft was approximately 7 n.m. from the ‘TFN’ VOR-DME.
Between 1245.39hrs. (35:39) and 1246.52hrs. (34:26) control of DA-1008 was transferred from the Las Palmas Control Centre to Tenerife Norte APP.
From the first communication by DA-1008 at 1314.28hrs. (06:50) until 1317.39hrs. (03:39), APP was in contact only with this aircraft and IB-711; thereafter until the time of the accident, apart from these two aircraft, it was also in contact with Hapag-Lloyd 542 which took off from runway 12 at the airport.
At 1314.28hrs. (06:50) DA-1008 contacted APP:
`Tenerife, good morning, Dan Air one zero zero eight’. At 1314.33hrs (06:45) APP replied: `Dan Air one zero Tenerife, go ahead’.
At 1314.36hrs. (06:42) DA-1008 reported:
`Good morning sir, levelled at one one zero, fourteen miles to Tango Fox November’.
At that moment the aircraft was between 14 and 16 miles from the `TFN’ VOR-DME and was approaching FL 110 with an indicated airspeed (IAS) of approximately 280 knots.
At 1314.45hrs (06:33) APP called:
`Dan-Air one zero zero eight, cleared to the Fox Trot Papa beacon via Tango Fox Trot November, flight level one one zero, expect runway one two, no delay’.
This message, with its closing words ‘no delay’ probably fixed in the commander’s mind the idea that they would be making the shortest possible approach, judging by the following reply from the aircraft:
`Roger, cleared to the Fox Papa via Tango Fox November, runway one two. Can we copy the weather?’.
APP immediately asked:
`Confirm, did you copy the weather?’.
`I would like to copy the weather and the pressure, please’.
APP passed on the following information:
`O.K. Runway in use one two, the wind one two zero, zero five visibility six from seven kilometres, cloud two oktas at one two zero metres, plus four oktas at two five zero metres, plus two oktas at three five zero metres, November Hotel one zero one three, temperature one six, dew point one one and drizzle’.
The aircraft acknowledged the message and requested the QFE for runway 12.
`Roger, can we have the Fox Echo, please, for runway one two’.
At this moment APP asked IB-711:
`Iberia seven one one, level?’.
IB 711 replied:
`Iberia seven one one we are past the VOR and have left six, we are now five thousand, five thousand, maintaining’.
This allowed APP to clear DAN-1008 to descend, and the following communications took place between APP and the aircraft.
`Received. Break. Dan Air one zero zero eight descend and maintain flight level six zero’.
`Roger, leaving one one zero for six zero’.
`Report your DME reading, please’.
The aircraft replied:
`Er, we’re reading seven DME Tango Fox November and requesting the QFE, please’.
According to the reconstruction of the path followed by the aircraft, taken in conjunction with the information from the flight data recorder and the cockpit voice recorder, at the time of this message the aircraft was between 7.5 and 9 nm. from the `TFN’ VOR-DME and was descending from FL 110 to 60.
APP supplied the information on the QFE which had been requested by the aircraft:
`Nine four three’.
`One zero zero eight, nine four three many thanks’.
Later APP corrected the QFE.
`One zero zero eight, for your information QFE for runway one two is nine four one’.
`Roger, nine four one for one two, thanks’.
In the cockpit, the commander asked to co-pilot to tune in the ILS on his side, to use the back beam for guidance, at 1317.18hrs (04:00).
`You can put the ILS on your side we might get it on a back beam for a lead in’.
There were no further messages between DAN-1008 and APP until the aircraft reported that it had just passed the `TFN’ VOR-DME, but according to the CVR, the crew noted passing the radio aid approximately 33 seconds before reporting the fact, and this delay in reporting might have influenced subsequent events. The cause of the delay cannot be determined, since no other conversation took place in the cockpit and communications during this period were scanty, there being only a 10-second exchange between the Hapag-Lloyd flight and APP.
From the time it reported its position as 7 n.m. on the DME, approximately 4:50 before impact, the aircraft’s speed continued to increase.
Data from the CVR:
P 1 || Commander
P 2 || Co-pilot
FE || Flight Engineer
APP || Approach Control
DAN || Dan-Air 1008
HP || Hapag-Lloyd
IB || IB-711
FE ‘That’s what I reckoned it should be by calculation’.
P1 ‘I won’t go out the full procedure if you know because it takes you way out to sea on this’.
APP ‘IB-711 report on completing procedure turn’.
IB ‘Roger IB-711’.
Morse Code ITF ILS audio ident.
P2 ‘ITF three oh five is in my box’.
P2 ‘Three oh two I’m sorry excuse me’.
P1 ‘Just about to go overhead going for’.
P2 ‘Two five five out of here’.
HP ‘Tenerife good afternoon — Hapag-Lloyd five four two’.
APP ‘Five four two, good afternoon report ready’.
DAN `Dan-Air one zero zero eight has just passed the Tango Fox November heading to the er, Fox Papa’.
While the above exchange was going on, the aircraft commenced a wide turn to the right, taking it on to 263°magnetic which placed it to the south of the 255° radial of the ITN’ VOR-DME, which was the radial it ought to have followed in order to pass over the FP beacon, according to the clearance it had at the time.
At 1318.54hrs. (02:24) APP, in view of the information on the aircraft’s position, notified the aircraft that the holding pattern at ‘FP’ was as follows:
`Roger, the standard holding over Fox Papa is inbound heading one five zero, turn to the left, call you back shortly’.
This transmission ended at 1319.00hrs. (2:18).
The aircraft immediately acknowledged the information.
`Roger DAN-AIR one zero zero eight’.
For the next 56 seconds there were no transmissions between the aircraft and APP, or between APP and any other traffic, but there ensued a series of comments amongst the crew on the information they had received and these need to be analysed because they reveal doubts, due to the phrase ‘no delay’ in their first exchange with APP, the lack of clarity in the information received, which began with ‘standard holding’ and ended with ‘turn to the left’, and the idea which had formed in the commander’s mind one minute and twelve seconds earlier, and these may enable us to formulate a hypothesis on the motives which led the crew to direct the aircraft to the accident site.
Information from the CVR:
Nothing during this period
P1 ‘Inbound one five zero to your left’
P2 ‘One five zero left yeh’
P1 ‘That’s an odd sort of one the runway…’
P1 ‘One to go’
P2 ‘One to go’
FE ‘One to go’
P2 ‘No I’m not er suppose it’s alright’
P1 ‘I’ll just turn straight round left on to one five zero when I go overhead then’
P1 ‘The only thing is we’re hmm we’re just about to miss it ah ah it’s too close’
P2 ‘Would you like the other one on the Fox Papa as well for this?’
P1 ‘If you put them both on as we’re going to hold yeh’
P1 ‘That’s er that’s the Foxtrot Papa now’
After the information about the hold had been acknowledged, given the expansive and extroverted nature of the co-pilot it is possible that he made some gesture of surprise; the commander confirmed ‘Inbound one five zero to your left’, which created a mental image of an inbound heading of 150° towards some unknown point which we have been unable to determine, given that the aircraft was on a heading of 263° magnetic at the time. A further difficulty is that the commander apparently understood that the 150° inbound heading was a course to be taken up at ‘FP’.
The co-pilot accepted what the commander said, but the latter still appeared to have doubts when he said ‘That’s an odd sort of one the runway…’
The crew then repeated the commander’s words ‘one to go’. Subsequently the co-pilot still appeared to have doubts about the information they had received and the action which the commander had apparently decided to take.
The commander’s response was: ‘I’ll just turn straight round left on to one five zero when I go overhead then’.
This appears to imply that he would make a turn to the left immediately on passing over ‘FP; on a heading of one five zero, which again reinforces the idea which he had already formed when he said ‘Inbound one five zero to your left’.
Once again the co-pilot, in spite of his uncertainty, confirmed what the commander had said, rather than voice his doubts which might have led to the crew calling APP.
Nevertheless the commander, who was still maintaining his aircraft at a high speed considering his proximity to’FP: remarked: ‘The only thing is we’re just about to miss it ah ah it’s too close’.
This appears to refer to the ‘FP’ beacon and possibly the difficulty in performing a manoeuvre to get from his heading of 263° on to 150°, which appears to be the picture he had in his mind.
The co-pilot suggested:
`Would you like the other one on the Fox Papa as well for this?’
We cannot say whether this suggestion was an attempt to draw the commander’s attention to the fact that the information they had received referred to ‘FP’.
The commander replied:
`If you put them both on as we’re going to hold, yeh’.
He appears to have been convinced that they had to hold, and believed he knew how the hold should be performed.
A moment later the commander remarked:
`That’s er that’s the Foxtrot Papa now’.
The co-pilot answered: Yep’.
It is notable that at a time when instrument monitoring was most needed, the co-pilot’s attention was mainly focussed on selecting and identifying the `FP’ beacon on his ADF.
The crew were probably engaged in checking their charts for the hold, therefore they did not anticipate their arrival at ‘FP’ at that moment.
The co-pilot immediately reported to APP:
`Dan Air one zero zero eight is the Foxtrot Papa level at six zero taking up the hold’.
APP replied: ‘Roger’.
However, according to the reconstruction of the flight path, the aircraft continued on heading 263° for approximately 20 seconds, during which time it covered about 2 n.m. (this action is difficult to comprehend since the commander had already stated previously that he would commence a turn immediately after passing `FP’). At this time the aircraft was flying in a sector with a minimum safe altitude of 14,500 ft.
Following a comment by the flight engineer ‘that’s the fuel’, which appeared to relate to his own specific duties, the co-pilot apparently continued expressing vague doubts to the commander, when he said: ‘Bloody strange hold isn’t it’.
The commander replied:
`Yes, doesn’t parallel with the runway or anything’.
But he continued the manoeuvre he had initiated.
A few seconds later the co-pilot expressed further doubts and the flight engineer joined in:
P2 ‘It’s that way isn’t it?’.
FE ‘That is a three isn’t it?’.
FE ‘That is a three isn’t it?’.
P2 ‘Yes, well the hold’s going to be here isn’t it?’.
The co-pilot’s comments are an indication of the uncertainty he felt about the manoeuvre which was being performed, but the meaning of the flight engineer’s comments and precisely what he was referring to, is not clear.
While this conversation was going on in the cockpit, APP received the following transmission from IB-711:
`Free five thousand now, we are in the procedure turn’.
APP immediately contacted DAN-1008:
`Dan-Air one zero zero eight recleared to five thousand on the Quebec Foxtrot Echo and the Quebec November Hotel’.
This clearance should not have been given in this form, which could lead to an error if the crew did not choose the correct datum for interpreting altitude. However, the crew acknowledged the message, read back the information and used the QNH as the datum.
This information was given by APP at a time when, as far as APP knew, the aircraft was entering the hold, according to the transmission received 35 seconds earlier.
The crew, bearing in mind the zone in which they were flying (assuming they had determined where they were), should have notified APP of their position and of the risks involved in descending to the assigned altitude.
A few seconds after reading back the information, the commander, who was quiet and not particularly vocal by nature and who appears to have been thinking about the manoeuvre, exclaimed:
P1 ‘Hey, did he say it was one five zero inbound?’.
It would appear that at this moment the commander’s mind went back over the information he had received on the hold, and he realised that his manoeuvre was taking him on a heading of 150° magnetic away from ‘FP’, whereas the information he had been given referred to an inbound heading of 150° to ‘FP’ for the hold.
This question led to the following dialogue:
P2 ‘Inbound yeh’.
P1 ’-I don’t like that’
P2 ‘They want us to keep going more round don’t they?’.
The conversation broke off at that moment because the Ground Proximity Warning System (GPWS) sounded.
The commander broke off the turn to the left, ordered an overshoot and remarked:
P1 ‘He’s taking us round to the high ground’.
He initiated a turn to the right and because this happened to take the aircraft over a valley the GPWS was deactivated.
This manoeuvre alerted the co-pilot, who said to the commander:
`I suggest a heading of one two two actually and er take us through the overshoot, ah’.
But the commander continued the turn to the right, being convinced that the left hand turn he had been making was taking the aircraft towards the mountains.
`He’s taking us round to high ground’.
The co-pilot did not repeat his previous suggestion, and simply acknowledged the comment.
The commander made a call to APP approximately 4 seconds before impact:
`Er, Dan Air One zero zero eight we’ve had ground proximity warning’.
At the same time as the commander was completing his message, the flight engineer exclaimed:
The impact then followed.
Company procedures require the flight engineer to alert the pilots whenever the bank angle exceeds 30°, and although the aircraft had been exceeding that angle for fifteen seconds, the reason he did not give a warning was probably that he was adjusting the engine settings, in accordance with the commander’s instructions.
APP, in view of the traffic being handled, should have kept the Dan :Air flight in the hold at `TFN’ but, possibly not wishing to delay the flight and being convinced he knew the position of the two aircraft, he kept them at a separation of 1,000 feet.
The controller did not determine correctly the position of the Dan -Air aircraft. This is evident from his failure to appreciate that it was overhauling IB-711. Thus, when DA-1008 reported passing ‘TEN’, perhaps earlier than he had calculated, he found himself in an unexpected situation and therefore passed information about a hold at ‘FP’, revising his calculations and estimating that at the theoretical speed for the sector, the aircraft would take nearly two minutes to reach `FP’; but it reported passing ‘FP’ and taking up the hold just 63 seconds later.
He authorised a descent to 5,000 feet, in the belief that the aircraft would be on the entry segment to the hold, i.e. on a heading of 330° outbound from ‘FP’ in accordance with the procedure for entering the holding pattern. APP began the information on the hold with the words ‘The standard holding …’, which according to ICAO Doc. 8168-0PS, Part II, Chapter 1, Section 18.104.22.168, indicates turns to the right unless otherwise specified. Although what APP meant to say was `… . turns to the left …’, what was actually heard was `… turn to the left …’, and this, if the information was not read back, could give rise to incorrect interpretation.
It proved possible to determine, by reconstruction of the flight path, that after passing ‘TEN’ the aircraft flew at an average speed of approximately 250 knots, higher than that advisable for this part of the approach. At the same time, the navigation was very imprecise, and more specifically the aircraft passed to the east of the ‘TEN’ VOR-DME at a distance of approximately 0.79 n.m. and at no time did it intercept the 255 radial of ‘TEN’, which would have brought it to `FP’; it therefore passed to the south of ‘FP’ at a distance of 1.59 n.m. The crew reported passing over ‘TEN’ some 33 seconds after actually passing it; APP immediately issued instructions for a hold at ‘FP’, which was accepted without read-back or any request for clarification.
The crew were never certain what action to take on reaching ‘FP’. The following hypotheses can be put forward:
If they had only interpreted the first part `the standard holding over Fox Papa is inbound heading 150 …’, they should have turned to the right on passing `FP’ in order to take up the ‘standard’ hold at ‘FP’ with an inbound heading of 150.
If they had only taken the second part of the message `… turn to the left’ as mandatory, one would assume that they should have turned left as soon as they received the message, or commenced the turn over `FP’; instead of continuing on 263° for a further 20 seconds before commencing their turn to the left.
It appears that the commander’s intention as soon as he received the information (and in spite of his having said ‘Inbound 150 …’) was to make a left turn to take up the ‘standard’ hold at `FP’,.approaching the radio aid on a heading of 330°, or simply to follow a magnetic bearing of 150°, because he thought that was the information he had been given.
With all the accumulated navigational errors, the activation of the GPWS led the commander to alter his flight path because he did not know his position. The deactivation of the GPWS led him to believe that his avoiding action had been correct.
Standard operating procedure for an avoidance manoeuvre requires the aircraft attitude to be adjusted to achieve the best rate of climb and it must be said that the commander made a mistake in not levelling the wings, with the result that instead of starting to climb, the aircraft simply reduced its rate of descent.
The combination of events led to the accident.
The confusion in the information and its acceptance by the crew lead us to examine the existing ICAO rules on holding instructions; it is apparent that they are not sufficiently clear and can give rise to errors of interpretation, since some documents such as Doc. 8168, Part II, Chapter 1 referred to earlier and Annex 4, Chapter 8, section 8.2.1 are ambiguous and contradictory and require clarification.
Today’s History Lesson
Aviation Safety Database
The Memorial Rose Garden
BBC Active Video for Learning
Flight Global Archives June 14, 1980
Report on the Accident # AAIB AAR 8/1981
The Gazette – Google Archives – May 10, 1980
10 Deadliest Air Disasters Caused by Miscommunication
The Spirit of Dan-Air, Simons, G.M., GMS Enterprises, Peterborough, 1993
Kompass – Winter 1974/75 Edition, Dan Air Services Ltd, West Berlin, 1974 (German)
Berlin Airport Company — Report on Dan-Air’s Berlin operation, July 1975 Monthly Timetable Booklet for Berlin Tempelhof and Berlin Tegel Airports, Berlin Airport Company, West Berlin, 1975 (German)
In Flight – Silver Jubilee Anniversary Edition, Dan Air Services Ltd, London, 1978
Airliner Classics (Dan-Air – Popular British Charter operator: The Boeing 727), Key Publishing, Stamford, UK, November 2011
Follow mayday365 for complete aviation disaster history every day!