By Simon Hradecky, created Tuesday, Dec 27th 2011 17:01Z, last updated Tuesday, Dec 27th 2011 17:01Z A Jeju Air de Havilland Dash 8-400, registration HL5256 performing flight 7C-502 from Jeju to Busan (South Korea) with 74 passengers and 5 crew, was climbing through 9500 feet out of Jeju when the crew received RUD CNTL and #2 RUD HYD caution lights. After working the relevant checklists the crew continued the flight to Busan. After completing an ILS let down to runway 36L while joining the downwind for the circling approach to runway 18R the crew received further caution indications amongst them elevator feel, rudder controls and pitch trim. Upon landing on Busan's runway 18R, winds from 130 degrees at 13 knots gusting 18 knots, the crew lost directional control and the aircraft veered left off the runway and impacted a concrete drainage ditch between runway 18L and 18R. 4 passengers received minor injuries, the aircraft received substantial damage.



South Korea's Aviation and Railway Accident Investigation Board (ARAIB) released their final report concluding the probable cause of the accident was:



The Aviation and Railway Accident Investigation Board determines that the cause of the runway excursion of the Flight 502 was that the rudder failure was not recognized by the pilots during flight and as well as during landing roll.



Contributing to this accident was that

- the rudder was failing to respond to the pilots' rudder pedal input and

- after departing from the runway, no appropriate alternative measure was taken to control the aircraft direction.



The ARAIB stated, that the first officer (30, CPL, 1,213 hours total) was pilot flying for the sector, the captain (50, ATPL, 8,655 hours total) was pilot monitoring.



After receiving the caution indications while climbing through 9500 feet the crew worked the relevant checklists after reaching cruise level 150 and consulted with a mechanics at Busan Airport.



When a number of additional caution indications, the crew could not completely recognize and remember, appeared while joining the downwind for runway 18R the captain regarded those messages as nuisance, no particular action was taken. The captain permitted the first officer to continue the approach and landing. The aircraft had been configured for landing at flaps 15, gear down.



During the flare, about 2 feet AGL, the rudder pedals were gradually pushed to the right, right aileron was applied and the aircraft touched down on the center line of the runway however began to immediately veer left after smooth main gear touch down at about 118-120 KIAS and a heading of 174 degrees magnetic, while the right rudder pedal was further pushed. Instead of crossing the controls (editorial note: normal cross wind handling, apply aileron opposite to rudder input) the control wheel was further turned to the right instead of forcing the left wing down (by applying left aileron).



According to the cockpit voice recorder the captain kicked the right hand rudder and uttered an expression of surprise, the first officer responded "it doesn't respond". Two seconds later the nose gear touched down (4 seconds after main gear touch down), the right rudder pedal pressure increased until the rudder pedal hit the mechanical stop, brakes were applied symmetrically and the propellers selected into reverse wth engines accelerating symmetrically, another two seconds later the aircraft departed the left edge of the runway about 3500 feet down the runway at a heading of 171 degrees magnetic followed by groaning sounds of one of the pilots another four seconds later.



13 seconds after nose gear touchdown the captain exclaimed "Oh No! Gosh!", 15 seconds after nose gear touch down the sound of impact with the concrete ditch along with a scream by the captain was recorded by the cockpit voice recorder.



The aircraft came to a stop about 340 feet left of the runway center line and 4,600 feet (half way) down the runway at a heading of 158 degrees magnetic. The engines were shut down and the aircraft was evacuated.



The captain said in post flight interviews that immediately after main gear touch down he recognized the aircraft drifted to the left and thinking the first officer was applying rudder kicked the right hand rudder, however, the aircraft did not respond.



The aircraft received substantial damage: the left wing was bent upwards and leaked fuel although the left wing tip showed little damage from impact, the left main gear collapsed, the nose landing gear was bent backwards and its structure sustained severe impact damage, its hydraulic lines were leaking, the left hand propeller received severe impact damage.



The ARAIB quoted Jeju maintenance personnel stating they often had fault and caution indications on for a while then off again on that particular aircraft. As a result the aircraft's cautions and fault indications were generally regarded as nuisance, maintenance took the simple way to reset the systems to get rid of the indications.



The ARAIB stated, that the RUD CNTL caution illuminates when the flight control electronic control unit (FCECU) determines that both #1 and #2 hydraulic systems are not able to pressure control the rudder or when the airspeed becomes invalid.



The #1 RUD HYD caution light illuminates when the #1 hydraulic pressure drops below 350 psi (+/- 50 psi), the FCECU shuts hydraulic pressure down on the left channel or the crew pushed the #1 RUD off switch. A compensation signal is sent in this case to permit the rudder to operate normally on the remaining channel.



Based on the FDR the ARAIB determined that the left hand spoilers had opened at 12 degrees prior to the takeoff roll and remained in that position until just before landing, without indication. Upon main gear touchdown at 119 KIAS the rudder position was 7 degrees to the left despite right rudder input.



During examination and tests of the FCECU it was found that the FCECU did not pass any test. A small crack was found on the #1 circuit board as well as the #2 circuit board, measurements however showed similiar values to those of a non-damaged FCECU. Engineers by the manufacturer, under supervision of ARAIB, opened the resistance bundle of the channel 2 card and cleaned between the connector pins 6 and 7. The FCECU was subsequently reassembled and passed all tests. The unit also passed two more tests under different environmental conditions.



The investigation however determined that the cleaning of the space between connector pins 6 and 7 could not explain the behaviour and loss of rudder control. No scenario could be identified that would be able to explain the accident flight.



The aircraft manufacturer rated the loss of rudder control a major issue, however not a dangerous or catastrophic issue. Simulator tests showed that despite the rudder position as experienced on the accident flight directional control of the aircraft could be maintained up to 25 knots from 130 degrees by applying asymmetric brakes.



Maintenance records indicated that the aircraft had been exposed to lightnings twice, in February and July 2007. These lightning strikes would require a report to the manufacturer and relevant maintenance tasks, however, there was no documentation that such steps were taken. Tests and examinations of the FCECU also ruled out the malfunction of the FCECU may have been the result of a lightning strike damage.



The ARAIB analysed that according to FDR the rudder position was a 7 degrees to the right for most of the flight, however, the abnormal rudder position was not recognized by the crew and consequently was not dealt with.



It would not be impossible to see the left spoiler deflected to 12 degrees with the naked eye. The ARAIB analysed: "Therefore, if abnormal indications of the spoiler and the rudder had not been verified by the pilots during flight, it would mean that their monitoring was not cautiously carried out, or perhaps the pilots might have ignored the abnormal indications even after verifying them."



The checklists for the RUD CNTL and RUD HYD caution messages did not contain any particular action except for maintaining speed below 200 KIAS. The pilots were not concerned regarding a possible rudder malfunction.



When after touchdown the aircraft did not respond to right rudder inputs the crew did not use asymmetric braking, probably as result of surprise.



Three safety recommendations were released to the aircraft manufacturer, 5 to Jeju Air and one to Busan Airport as result of the investigation. Busan Airport completed the recommended work to cover the ditch by Oct 31st 2009.



HL5256 in its final position:



