Transport Accident and Incident Investigataion BureauTransport Accident and Incident Investigataion Bureau
 
 
 
 

FINAL REPORT Nr.1/2005.

FINAL REPORT OF THE INVESTIGATION INTO THE CAUSE OF THE SERIOUS INCIDENT WITH ,,KESCOBERMUDA’’ LIMITED AIRCRAFT ,,RAYTHEON 390 PREMIER 1’’, VP-BAE, ON 25 JANUARY 2005 AT VENTSPILS INTERNATIONAL AIRPORT

.In accordance with Annex 13 of the Convention of Chicago as well as the Directive 94/56/EC of 21 November 1994 establishing the fundamental principles governing the investigation of civil aviation accidents and incidents of the Council of the European Union, the purpose of an investigation conducted under the responsibility of the Latvian Accident and incident investigation division is not to apportion blame or liability.

Head of Accident and incident

Investigation division

 Riga, November 2005

CONTENTS

 GENERAL INFORMATION OF THE SERIOUS INCIDENT

 THE INVESTIGATION

 SYNOPSIS

 1. FACTUAL INFORMATION

1.1 History of Flight

1.2 Injuries to Persons

1.3 Damage to Airplane

1.4 Other Damage

1.5 Personnel Information

1.6.Airplane Information

1.7 Meteorological Information

1.8 Aids to Navigation

1.9 Communications

1.10 Airport Information

1.10.1 Air Traffic Control Information

1.11 Flight Recoders

1.11.1 Cockpit Voice Recoder

1.12 Wreckage and Impact Information

1.13 Medical Information

1.14 Fire

1.15 Survival Aspects

1.16 Tests and Research

1.17 Organizational and Management Information

1.18 Additional Information

2. ANALYSIS

3. CONCLUSIONS

 3.1.FINDINGS

 3.2. PROBABLE CAUSE

 4. RECOMMENDATIONS

 6. APPENDICES

 GENERAL INFORMATION OF THE SERIOUS INCIDENT

 Unless stated otherwise all times in this report are UTC time.

 Registered Owner:                  ,,KESCO BERMUDA’’ Limited

 Operator:                                 ,,KESCO BERMUDA’’ Limited

 Aircraft type:                           ,,Raytheon 390 Premier 1’’

 Nationality:                                         Bermuda

 Registration:                                        VP-BAE

 Place of Serious Incident:        Runway 03 at Ventspils International Airport, Latvia

 Date and Time:                                    25 January 2005 at 09.48 hrs UTC (11.48 hrs LT)

THE INVESTIGATION

 The serious incident was notified to the Aircraft Accident and Incident Investigation Division of the Ministry of Transport of the Republic of Latvia on January 25th 2005. The Aircraft Accident and Incident Investigation team comprised Mr.Ivars Gaveika (Investigator-in- Charge), Mr.Ivars Zorgenfreijs – Investigator, Mr.Valentins Kaltenieks – Investigator and Mr.Stanislav Lobashovs – expert. Assistance in the investigation was rendered by National Transportation Safety Board USA.

 Synopsis

 On the 25 of January 2005 the airplane was cleared to VOR/DME approach to Runway 03 at the Ventspils International airport, Latvia. At the altitude approx 270 ft EGPWS (Enhanced Groung Proximity Warning System) ,,Too Low Flap’’ sounded. The pilot continued approach (single pilot flight) and after landing pilot applied brake pressure and continued to hold the permanent brake pedals pressure. He held pedal force the entire time after touchdown until start emergency braking before the end of the Runway 03. At the same time pilot reported very weak braking effectiveness.Getting closer to the end of runway pilot decided that it is not enough runway remaining to stop the aircraft and used emergency brake twice.

About 09.48 UTC (11.48 hrs LT) ,,Kesco Bermuda’’ Limited ,,Raytheon 390 Premier 1’’ aircraft VP-BAE after landing on the Rwy 03 at Vetspils International airport overrun 86 metres with Rwy 03 heading, bogged down and broken nose landing gear.

The incident was reported to the Aircraft Accident and Incident Division of Ministry of Transport of the Republic of Latvia on 25 January 2005 at 10.00 UTC from Ventspils International airport Manager and the investigation began on 25.January 2005 when additional details of the incident became known.

The incident occurred at Ventspils International Airport in weather conditions: wind 110 degrees 5 kts (ATC report 09.45.09 UTC before landing), meteorological visibility 8000 meters, NSC M03/M03 Q1024 03690340. Runway 03 was wet and covered with slush 3 mm.

Runway 03 was being used for approach and landing. Braking coefficient - 0,40.

Braking coefficient was to determined with special equipment ,,1155M’’(Deselerometer) Nr.437 which was placed in the VW Golf car. Runway was wet, partly (20%) covered with slush, braking actions was ,,Good”, braking coefficient 0,40/0,45/0,50.

 1. Factual Information

 1.1 History of the incident

 -Flight number – Charter Flight VP-BAE with 2 passengers on board. Total traffic load - 220 kg.

-Point of departure – Moscow Vnukovo airport (UUWW), Russia

-Time of departure – 08.17 UTC

-Point of landing – Ventspils International airport (EVVA), Latvia

-Time of arrival          – 09.48 UTC (11.48 hrs LT)

The whole flight did not show any abnormalities or failures. At descent to EVVA airport at approx 270 ft of altitude (according to the Honeywell EGPWS report of April 26,2005) the ,,Too Low Flaps’’ warning sounded. He selected flaps down after receiving the alert. Pilot continued the approach. The approach speed was maintained 110-108 kts less then Vref=112 kts  according to EGPWS report. After the disengagement of the autopilot he deviated below glide slope and touched down before aiming point markings.

After touchdown, pilot applied full braking pressure and extended lift dump, aircraft was continued the landing roll. At third part of landing roll distance before overrun the Runway 03 pilot was believed the main brakes had failed and started braking procedure with emergency brakes unit. He activated emergency brakes twice. Actually it was confirmed by controller , who have noticed that aircraft was kind of ,,nodding’’ couple of times before overrun the Runway 03. As the airplane passed in front of the tower and passed the taxiway exit the pilot transmitted something to the effect that his brakes did not function. Aircraft continued landing roll out of the Rwy 03 and bogged down 86 meters from end of the Rwy 03.

One of the Rwy 03 red end light (LGT) were broken during the landing roll out from the Rwy 03.

  1.2 Injuries to persons

 None.

 1.3 Damage to aircraft

 1.3.1. Broken the Nose landing gear - on the nose landing gear assembly the L/H fork was bent ninety degrees outboard and the shimmy dapper right hand attachment bracket was sheared apart as illustrated in Figures 1.3.1.1, 1.3.1.2, 1.3.1.3

1.3.2.The R/H wing tip has a crack propagating from the forward outboard corner of the lower access cutout as shown in Figure 1.3.1.4

 

1.3.3. Cracking of the paint was found on the L/H and R/H main landing gear outboard doors around the hinge attachment and the bracket attachment as shown in Figure 1.3.1.5

1.4 Other damage

 One of the Rwy 03 red end light (LGT) were broken during the landing roll out from the Rwy 03.

 1.5 Personnel information

 The Pilot, age 40, was hired by Kesco Bermuda Limited on 07 January 2004. He held a Commercial pilot licence (CPL) III P Nr: 005567 valid until 2005 April 17 and Russian medical certificate S/N AA Nr.:0034205 valid until February 17, 2005, with no limitations.

Pilot  flew  on AN-24, TU-134, IL-76, PC-12 and Raytheon 390 Premier 1 airplanes while on duty.

Pilot  completed a Premier Supplemental Performance Familiarization Fligjht on August 21, 2004. (RAC Demo pilot Mark Mohler endorsed the pilot’s logbook).

He had accumulated 5011 hours total flying time.

He had accumulated 2350 hours as Pilot-in-command.

He had 191 flight hours on type.

He had flight time in day of incident 01.31 hours. He had no other flights that day.  He had flown 06.49 hours in the 72 hours before the incident and 21.36 hours in the 7 days respectively before the incident.

The pilot last proficiency check was in USA Flight Safety International Training Center on 12 August 2004 and his pilot-in-command check occurred on 12 August 2004.

Summary of interviews with the Pilot and AFIS operator see at Appendix 8 and 9.

 1.6.Aircraft information

 The Airplane ,,Raytheon 390 Premier 1’’ VP-BAE, Serial number RB-66, manufacture by Raytheon Aircraft Company Raytheon 390 Premier 1 at 2003 and had a total of 397 flying hours and 312 landings.

Registration marks – VP-BAE

Owner – Kesco Bermuda Limited

Adress of Registered owner – CLARENDON HOUSE, 2 Church Street

Hamilton HM 11, Bermuda

Operator - Kesco Bermuda Limited

Validity Certificate of Airworthiness – until 16 September 2005

Engines – FJ-44-2 made by Williams International, a division of Rolls-Royce

Left engine – S/N – 105039; Right engine – S/N – 105040

Last maintenance inspection – 16 September 2004

After last maintenance inspection aircraft had 186 flight hours

Take off weight actual – 12.300 lbs (5.579 kg),

Max take off weight 12.500 lbs, (5.670 kg).

Center of Gravity – according operation limits and requirements – 30% MAC.

Landing weight actual – 10.000 lbs (4.536 kg)

Center of Gravity - according operation limits and requirements – 29,4% MAC.

Take off fuel actual – 1.600 kg

Landing fuel actual – 557 kg

Refueling date - 25.01.2005

Type of fuel used – TC-1

 AFM AIRPLANE BRAKE SYSTEM DESCRIPTION

 The airplane has single wheel and tire assemblies mounted on each main landing gear and on the nose landing gear. Each main landing gear wheel assembly is equipped with a brake assembly that can be actuated from the cockpit rudder pedals or from the Parking/ Emergency Brake lever, which is located on the center cockpit pedestal. There are two modes of braking: power brake, and power brake with anti-skid. The power brake mode functions with the ANTI SKID switch in the NORM (the anti-skid control signal is available) or OFF (the anti-skid control signal is not available) position. Placing the ANTI SKID switch to NORM enables the anti-skid function. The anti-skid braking functions from about 10 knots to about 165 knots with anti-skid enabled. When a wheel-speed transducer signals a skid condition to the power brake/ anti-skid control unit (ACU), the power brake/ anti-skid control valve (PBCV) is signaled to release brake pressure to the brake assemblies. The automatic anti-skid function returns to the standby mode after the wheel skid condition stops or the pilot pedal pressure is below skid level. Brake control then returns to the power brake mode. When the ANTI SKID switch is in the OFF position, or with the anti-skid control signal not available, hydraulic pressure is applied to the wheel brakes in response to brake pedal application.

 Touchdown and locked-wheel protection is available. Touchdown protection is accomplished by inhibiting braking until the wheels spin-up (as detected by the wheel-speed transducers) to 60 knots or until 3 seconds after the detection of a weight-on-wheel (WOW) condition by either main landing gear wheel.

 When the parking brake lever is lifted, the parking brake control valve will pass either normal hydraulic system pressure to the brake assemblies, or if normal hydraulic system pressure is not available, the emergency/ parking brake accumulator will pass hydraulic pressure to the brake assemblies. Anti-skid brake protection is not available when the parking brake lever is activated.

 1.7 Meteorological information

 Weather information at Ventspils International airport are made an automated observing system AV-11.

Weather observations are transmitted in coordinated universal time (UTC).

The 250920Z METAR EVVA (11.20 local time) on the day of the incident the following: winds 110 degrees 02 KT, visibility 8000 m, NSC temperature M03, dew point M03 and QNH 1024  03690340

At 09.24:06, ATC reported the EVVA 250920 METAR weather and announced ”runway covered in slush, breaking coefficient 0,40.

At 09.45.09 UTC was the latest winds report by ATC before landing -110 degree 5 kts.

(See appendix 1 and 3)

 1.8 Aids to Navigation

 EVVA airport is equipped with VOR/DME – 113,6 VNT. No problems with any navigational aids were reported.

 1.9 Communications

 No communications problems were reported between the pilot and any of the air traffic controllers who handled the flight.

 The correct phraseology, to be used by air traffic controllers, is to be found in International Civil Aviation Organisation Doc.4444 Procedures for Air Navigation Services ,,Air Traffic management’’Chapter 7 – Aerodrome and Meteorological Information item 7.3.1.2.1 and 7.3.1.2.2.

 1.10 Aerodrome information

 Ventspils International airport (EVVA) is located 225º 2,7 NM southwest from the city of Ventspils and has an elevation of 19 feet (5,7m) MSL

Ventspils International Airport has one Runway 03/21.

AERODROME     57 21 28.08 N

CART – ICAO    021 32 38.99 E

ELEV 19’

Dimensions of Rwy – 1298x32 m

Slope of Rwy 03 – 0,24% downward slope for the first 877m and then an upward slope of 0,05% for the remaining 421m;

TORA – 1298m;

TODA – 1698m;

ASDA – 1298m;

LDA –    1298m.

Surface of Rwy – Asphalt-Concrete.

AUW 30T

Approach and Runway Lighting:

APCH lights: Rwy 03 – NIL; Rwy 21 – Simple ALS 300m LIL

RWY lights: White LIL; Green THR lights, Red End lights. (see appendix 2)

 Examination of the runway and tire tracks revealed that the airplane overran the departure end of Runway 03 on the right side of the runway center. The left main landing gear tire ran over a runway threshold light, separating the light assembly from its mounting pad. No corresponding damage to the left main landing gear tire/ wheel was observed. The airplane tire tracks continued in the direction as the runway heading for about 200 feet, and then the tracks made an abrupt turn to the left. The airplane came to a stop, pointing northwest with the nose tire/ wheel separated from the nose landing gear strut.

 The runway was 4,259 feet long and was constructed of interlocking concrete slabs covered with asphalt. The asphalt surface had multiple areas of repair. There were areas where the texture of the surface on the right side of the centerline differed from the texture of the runway surface on the left side of the center line. The runway was not grooved to aid in water dispersal. The runway had threshold markings, runway designator markings, runway centerline markings, touchdown-zone markings at 544 feet and 1,506 feet, and aim point markings at 1,077 feet. There was no glide slope presentation equipment displayed, nor was there any runway distance remaining information displayed. There were no taxiway(s) running parallel to the runway; there was a taxiway exit located on the left and 3,395 feet from the approach end of Runway 03 (864 feet remaining) that led to the airport terminal and parking apron. The terminal with the tower located on top was set 753 feet perpendicular from the runway. No tire marks (from any airplane) were observed on the runway surface.

 1.11 Flight recorders

 The serious incident airplane ,,Raytheon 390 Premier 1’’ was not equipped with Flight Data Recorder.

The airplane was equipped with CVR S/N 000187264 and data dump was performed.

The CVR was sent to the USA NTSB audio laboratory in Washington, D.C., for readout and evaluation. The CVR data started at 09.20 and continued uninterrupted until 09.51.03. The recording consisted of four channels of audio information. A transcript was prepared of the entire 31-minute 03 -second recording (see appendix 3)

 The CVR provided a wealth of information. The CVR information allowed an incident sequence of events to be constructed; the determination of time between specific events and the collaboration of other information.

 -After the pilot received the airport surface weather information, the controller asked the pilot what runway he wanted to land on.

-The pilot requested Runway 03.

-Two minutes later, the pilot contacted the controller and requested Runway 21.

-Two minutes after that, the pilot again contacted the controller requesting Runway 03.

-At CVR time 21:16, which corresponded to ATC transcript time 0938:55 UTC, the pilot

reported “D8.8 leaving 1,500,” which is a point on the VOR/ DME Runway 03 approach plate.

-At CVR time 27:27, the pilot reported, “wheels down, passing 5.7 DME inbound,” which is a point on the VOR/DME Runway 03 approach plate.

-At CRV time 28:03, (09.45:48 UTC) the avionics artificial voice stated “1,000 feet.”

-At CVR time 28:54, (09.46:39UTC) two clicks are heard, which evaluated as the pilot unlocking the lift dump handle.

-At CVR time 29:00, (09.46:45 UTC) the avionics artificial voice stated, “500 feet.”

-At CVR time 29:25, (09.47:10 UTC) the autopilot disconnect chime was heard.

-At CVR times 29:31, 29:36, and 29:39, ( 09.47:14, 09.47:19 and 09.47:22 UTC) the avionics artificial voice stated “TOO LOW FLAPS.” Coincident with the third warning, two clicks were heard, which this investigator

evaluated as the pilot moving the flap handle to the flap DOWN position.

-At CVR time 29:55, which corresponded to 09.47:37 UTC, touchdown was heard.

-At CVR time 30:10, (09.47:52 UTC) and again at CVR time 30:13, (09.47:55 UTC) a noise was heard, which evaluated as possible tire squeals from the emergency brake lever being lifted.

-At CVR time 30:15, (09.47:57 UTC) the pilot made a radio transmission concerning his brakes.

-At CVR time 30:23,(09.48:05 UTC) a click was heard, which this investigator evaluated as the pilot moving the flap handle to the flaps UP position.

-At CVR time 30:25, (09.48:07 UTC) the airplane left the runway.

-At CVR time 30:31, (09.48:13 UTC) background noise ends, and the engines were heard unwinding and switches were heard being moved.

-At CVR time 31:03.5, (09.48:45 UTC) the CVR recording ended.

 The investigators did not hear the audio associated with the engine igniters being energized, during the approach. The Premier Airplane Flight Manual (AFM) approach checklist item #8 is (engine) “Igniters-ON.” The investigators did not hear any change in engine noise before the touchdown noise that would correspond to the throttles being retarded to IDLE before landing.

AFM approach checklist item #1 is “Thrust – IDLE” at 50 feet agl. The exact engine power setting during the approach was not determined.

 Honeywell Principal Engineer – EGPWS Yasuo Ishihara downloaded the accident airplane’s EGPWS flight history. The downloaded data provided information to establish the airplane’s flight path for 30 seconds during the time period about 45 seconds prior to landing to about 15 seconds prior to landing as a result of a Mode 4 FLAP warning (TOO LOW FLAPS) being triggered about 25 seconds prior to landing (landing gear down, lower than 245 feet agl, airspeed less than 159 knots, and flaps not down). The airplane location when it was 50 feet agl was also recorded. Ishihara plotted the recorded information. Those plots showed the airplane stable and slightly above a 3-degree glide slope, tracking inbound to the airport along an extended center line to Runway 03, and at a consistent 106 knots calibrated airspeed. The flaps being selected to DOWN was recorded about 15 seconds prior to touchdown. Ishihara stated that the GPS latitude/ longitude position listed for each EGPWS time was accurate to within 2 to 3 seconds of the airplane’s actual position; i.e., at the airplane approach speed (106 knots) each airplane position was accurate to within 350 to 525 feet of the airplane’s actual position.

 1.12 Wreckage and impact information

 A quantity of 1,200 pounds of fuel was removed from the airplane.

The airplane traveled about 285 feet over loose/ muddy ground before coming to a stop. The airplane remained upright and sustained minor damage to the nose landing gear assembly. The nose wheel was laying ninety degrees to the left of the nose wheel fork and was attached to the nose wheel fork’s left fork. The shimmy damper’s right hand attachment bracket was separated. The right wing tip had plowed into the ground during the aircraft’s turn to the left. Dirt was found on top of the wingtip assembly, and there were multiple mud streaks on the upper surface and leading edge of the right wing and under the spoiler panels. There was mud and debris around the inside of the number two engine intake and mud on the front face of the fan. Except as noted previously, the airplane had remained intact with no visible damage. On April 27, 2005, RAC Mechanical Engineer II Joseph L. Vondracek and RAC QA

 Manager/NDT Level III Bryce Bee conducted a damage survey on the accident airplane at EVVA. Their report of the damage survey appears at Appendix 10.

 The flight controls were operated from the cockpit. The flaps were in the retracted position and the flap handle was in the UP position. The lift dump handle was in the retracted position, and the LOCKED/ UNLOCKED switch was in the LOCKED position.

 The landing gear was extended, and the landing gear handle was in the DOWN position. The brake ANTI SKID switch was selected to the NORM position. The main tire/ wheel assemblies remained intact. The brake assemblies remained intact. Break wear pin lengths were well extended. The airplane Parking Brake/ Emergency Brake system was activated. The brake assemblies were heard to clamp down on the brake discs. There was no adverse or uneven wear observed on the landing gear tires.

The brake cables from the cockpit to the Crane power brake control valve moved freely when the brake pedals were depressed. The Crane power brake control valve lever arms moved their full range of travel when the brake pedals were depressed.

The airplane was configured as a 3-tube instrument panel. There was no damage to the instrument panel. Airplane electrical systems were not energized. The airplane Hobbs meter read 0397.1 hours.

 1.13 Medical and pathological information

 There were no medical aspects relevant to the incident.

 1.14 Fire

 Not applicable.

 1.15 Survival aspects

 Not applicable.

 1.16 Tests and research

 All reports from USA laboratory. See appendices for more information.

 Anti-skid brake components (two wheel transducers, brake control unit, and power brake control valve) – to be examined at Crane Hydro-Aire in Burbank, California;

 Honeywell Mark V Enhanced Ground Proximity Warning System (EGPWS) computer – to be examined by Honeywell in Redman, Oregon;

 Two Rockwell-Collins Flight Management Computers (FMC-3000) and Maintenance Data Computer (MDC-3000) – to be examined by Rockwell-Collins in Cedar Rapids, Iowa.

 The Crane brake components removed from the airplane and examined by Crane were forwarded to RAC at the completion of the Crane tests. The brake components were installed on RAC Flight Test Model 390 (S/N RB-1), during planned anti-skid brake operation braking tests.

 1.17 Organizational and management information

 None

 1.18 Additional information

 The AFM from the airplane did not contain Landing distance information for Wet or Contaminated runways (Manufacturer Approved Airplane Flight Manual Supplement for Airplanes Operating on Wet and Contaminated Runways, Issued: May, 2004 P/N 390-590001-0077).

According the AFM supplement, performing calculations for the Ventspils (EVVA) airfield Runway 03 length 1298 m (4259 feet), for runway with slush, roughly required landing distance should be 1279 m (4194 feet). (Manufacturer Approved Airplane Flight Manual Supplement for Airplanes Operating on Wet and Contaminated Runways Page 53 0f 80 Revised: May, 2004).

 RAC Service Bulletin (SB) 27-3608 (Lift Dump Actuation Kit P/N 390-3607-0001 and Main landing gear switch replacement Kit P/N 390-8103-0001) had been complied with.

 On December 24, 2004, the airplane’s right main landing gear tire ruptured during a landing at Ufa, Russia. The right tire/ wheel and brake assembly was replaced by PremiAir Engineer Whitman. Whitman reported the brake assembly was leaking hydraulic fluid which caused the brake assembly to seize. The brake cables from the cockpit to the Crane power brake control valve moved freely when the brake pedals were depressed. The Crane power brake control valve lever arms moved their full range of travel when the brake pedals were depressed. The airplane’s brake system operated normally after the right brake assembly was replaced. The airplane had flown for 28 hours since the brake assembly replacement, and the incident landing at EVVA was the 14th landing since the brake replacement.

 Witness Information – According airport manager and ATC controller (Tower) interview the aircraft touchdown point was in normal touchdown area (See Appendix 9).

 1.19 New investigation techniques

 None

 2 Analysis

 Before flight to EVVA airport there was not performed a proper calculation of the actual landing distance based on the AFM supplement for the Wet or Contaminated runways operation, due to absence of this document. The AFM from the airplane did not contain ,,Landing Distance, Wet Runway’’ Page53 of 112 and ,,Landing Distance, Slush’’ Page 85 of 112 Issued: January, 2003 and Pages 23 of 80 , 53 of 80 Revised: May, 2004.

 Current AFM (Airplane Flight Manual), which was actually used on that aircraft, was not amended properly. Manufacturer Approved Airplane Flight Manual Supplement for airplanes operating on Wet and Contaminated runways (Issued by Raytheon on May, 2004) was not incorporated before the Serious Incident happened.

The airplane was owned and operated by RAC from 2003 until delivery to Kesco Bermuda Limited on September 17, 2004. At the time of sale , Kesko Bermuda Limited registered RB-66 with RAC to receive Technical Publication`s Safety of Flight Information (SOFI). But AFM Supplement for airplanes operating on Wet and Contaminated runways was issued on May 2004, and therefore on the moment of delivery should be already incorporated.

 The first AFM supplement dated 31 March 2005 that included landing distance information in tabular form concerning Wet/contaminated runway awareness was received by Operator at April 2005.

 RAC Beechcraft Model 390 Premier I FAA Approved Airplane Flight Manual (AFM) (P/N 390-590001-0003B8), Section V, VA, and VB Performance, OPERATIONAL FACTORS, states the following:

Performance data has not been determined for wet or icy runway conditions.

 The AFM, Section V, VA, and VB Performance, states that all landing performance in the manual is based on flight test data and the following conditions:

 1. Thrust was set to establish a 3 degrees approach with the airspeed stabilized at VRef.

 2. Both thrust levers were moved to idle position as the airplane passed 50 ft. above the runway surface.

3. At touchdown, maximum braking was immediately initiated.

4. Lift dump spoilers were selected immediately after nose gear touchdown.

5. Maximum breaking was maintained to a full stop.

 The AFM, Section V, VA, and VB Performance, LANDING DISTANCE graph has the following associated conditions listed on the graph:

RUNWAY …………………………………………………... PAVED, DRY SURFACE

APPROACH SPEED ………………………………………. Vref

FLAPS …………………………………………………….…. DN

ANTI-SKID ………………………………………………….. NORMAL

 The Model 390 AFM has a Manufacturer Approved Supplement (P/N 390-590001-0077) issued May, 2004. The supplement provided contaminated runway landing distance graphs. A LANDING DISTANCE WET RUNWAY graph and a LANDING DISTANCE SLUSH RUNWAY graph along with other graphs are provided. Wet and contaminated conditions are defined as follows:

A runway is considered wet when there is sufficient moisture on the surface to cause it to

appear reflective, but without significant areas of standing water.

 

A runway is considered to be contaminated when more than 25% of the runway surface area (whether in isolated areas or not) within the required length and width being used, is covered by surface water more than 3mm (0.125 inch) deep…

LANDING DISTANCE CALCULATIONS REVIEW

 No accident pilot landing distance calculations were presented.

 The reported accident landing conditions were as follows: airplane weight 10,000 lbs, Vref 112 knots, temperature -3 degrees Celsius, wind nil, runway gradient -0.2%. Pilot reported the brakes were applied after touchdown, not immediately applied maximum, rather with ever increasing pressure. The pilot reported the LIFT DUMP was extended after brake application.

 Using the LANDING DISTANCE graph and the accident information known, the landing distance and the ground distance required for a landing on a dry runway at Vref was calculated.

 Using the LANDING DISTANCE WET RUNWAY graph and the accident information known, the landing distance and the ground distance required for a landing at Vref was calculated.

 Using the LANDING DISTANCE SLUSH RUNWAY graph and the accident information known, the landing distance and the ground distance required for a landing at Vref was calculated.

 

LANDING DISTANCE @ Vref

TOTAL DISTANCE

GROUND DISTANCE

 

Dry

2,873 feet

1,373 feet

 

Wet

3,613 feet

2,113 feet

 

Slush

4,194 feet

2,694 feet

 

  The LANDING DISTANCE (TOTAL) value is composed of two parts; the air distance and ground distance. The air distance is that distance the airplane travels horizontally across the ground from the position directly below 50 feet above ground position to the position at touch down. Geometrically the distance covered from 50 feet on a 3. glide slope would be 954 feet. The AFM landing performance graphs allow for a 3 second duration flare to reduce the descent rate before touchdown. That 3 second allowance will extend the touchdown point from 954 feet to about 1,500 feet. Subtracting the 1,500 feet air distances from the above total distances will produce the GROUND DISTANCE values. To achieve the ground distance values the AFM landing performance graphs assume the pilot applies a maximum continuous brake effort by touchdown plus 1 second, and the pilot has the lift dump extended by touchdown plus 2 seconds.

 Referring to Wreckage Diagram Section 2.1, the runway had 4,259 feet of useable length.

Babenkov stated the touchdown point was before the runway aim point marks, which were located 1,077 feet from the runway threshold. The amount of useable runway after the aim point marks would be 3,182 feet.

 STOPPING DISTANCE AWARENESS

 Raytheon Aircraft Safety Communiqué No. 246 Landing Performance Awareness was released in June 2004. The communiqué defined the landing distance obtained from the AFM graphs as being composed of an air component and a ground component. The communiqué defined the value of the air component used in the AFM Landing Distance graph as being 1,500 feet, as described previously. The communiqué then quantified the percent Vref speed reduction that occurred during a normal landing in the first, second, and third part of the calculated ground distance. The expected percent reductions in Vref speed were as follows: 8% total in the first third of the calculated ground distance, 31% total in the second third of the calculated ground distance, and the remaining 69% in the last third of the calculated landing distance.

 The Vref speed calculated by Babenkov was stated to be 112 knots. Babenkov stated his approach speed was less than the calculated Vref speed, and Babenkov stated he allowed his glide path to dip below the 3 degree glide path. The EGPWS recorded the approach speed flown until 15 seconds before touchdown was a consistent 106 knots. Even though there was no audible indications on the CVR to indicate the throttles were retarded to idle before touchdown, the information known would be consistent with the accident airplane landing at the expected touchdown speed. This investigator considered the touchdown occurred at the aim point markings, which left 3,182 feet of useable runway to stop the airplane. Applying the percent speed reductions stated in the communiqué to the ground distances calculated for a landing with lift dump on runways that are dry, or wet, or with slush, the following are the expected speed, ground distance traveled from touchdown (TD), and useable runway remaining values:

 

 

Dry @ Vref

First third

Second third

Final

 

Vref speed change

112 to 103 knots

103 to 77 knots

77 to 0 knots

 

Distance traveled from TD

458 feet

915 feet

1,373 feet

 

Runway useable remaining

2,724 feet

2,266 feet

1,807 feet

 

 

 

 

Wet @ Vref

First third

Second third

Final

 

Vref speed change

112 to 103 knots

103 to 77 knots

77 to 0 knots

Distance traveled from TD

704 feet

1,408 feet

2,112 feet

 

Runway useable remaining

2,478 feet

1,774 feet

1,070 feet

 

 

 

 

Slush @ Vref

First third

Second third

Final

 

Vref speed change

112 to 103 knots

103 to 77 knots

77 to 0 knots

 

Distance traveled from TD

898 feet

1,796 feet

2,694 feet

 

Runway useable remaining

2,284 feet

1,386 feet

488 feet

 

 

 The runway distance from the aim point marks to the taxiway intersection was 2,318 feet. The runway distance remaining at the taxiway intersection was 864 feet. The CVR captured the touchdown at CVR time 29:55. The tower operator reported the airplane was passing through the taxiway intersection when the pilot made a transmission concerning the airplane brakes. The CVR captured that transmission at CVR time 30:15. At CVR time 30:25 the airplane left the runway. This investigator calculated that the average speed of the airplane from the taxiway intersection to the end of the runway was 51 kts. If the rolling friction of the airplane is considered to be 1 knot/sec, then there would be a 10 knot speed change as the airplane traversed the last 864 feet of runway. This would correspond to about a speed of 56 kts at the taxiway intersection and would correspond to about a 46 knot speed when the airplane reached the end of the runway.

Using the following factual information:

-The airplane touched down at about 106 knots airspeed by the aim point marks.

-The pilot applied brakes and extended the lift dump after touchdown.

-At some point prior to the taxiway intersection the airplane brakes were released and

two quick emergency brake applications were made.

-The airplane had been on the ground for 20 seconds and had traveled about 2,318 feet

when it reached the taxiway.

-A transmission was heard from the airplane as it was observed at the taxiway.

-Ten seconds after the transmission at the taxiway, the airplane reached the end of the

runway surface, which was 864 feet from the taxiway.

-The airplane brakes were not applied as the airplane traveled from some point before

the taxiway intersection until the end of the runway.

The following information can be obtained:

-From the touchdown to the taxiway the airplane speed changed from about 106 knots to

about 56 knots as it traveled 2,318 feet of ground distance in 20 seconds

-The calculated average deceleration from touchdown to the taxiway was about 4.2 ft/sec2.

-The airplane left the runway at a speed of about 46 kts

Based on the EGPWS system data analysis (airspeed, altitude, alt rate, track, pitch and roll angles) the aircraft was established on stabilized approach on three-degree glide path. The landing weight was 10 000 pounds (4536 kg). Based on the AFM supplement for the Wet or Contaminated runways the Vref should be 112 knots. According to EGPWS report the speed was 110-108 kts well below Vref. The vertical speed was corresponding to a typical three-degree glide path and the pitch angle was quite stable.

According to ATC (Tower) controller interview the aircraft touchdown point was in normal touchdown area.

,,Too Low Flap’’ warning indicating that Mode 4B of EGPWS system was activated due to landing gear were down and aircraft penetrated the boundaries getting lower then 245 feet with speed less then 159 knots with flaps not in landing configuration. Pilot reported that the flaps were extended. The aircraft was stabilized on three-degree flight path and touches down before aiming point markings. So delayed flaps extension does not really affects the landing.

According Raytheon Aircraft ,,Safety Communique No.246 at ½ nautical mile from the threshold, a standard three-degree glide path would place the airplane at 160 ft above the landing field elevation. At this point, the airplane should be in the landing configuration, trimmed at Vref with a normal power setting for the approach.

After the touchdown the full brake pressure were applied and lift dump was deployed. After incident inspection of the aircraft revealed that there was some grass remaining and dirt under the lift dump surface, which indicates that the surfaces were deployed. The pilot reported usual feeling of additional aerodynamic braking when used lift dump handle.

When the aircraft was getting closer to the runway end the pilot had appreciation that it will not enough remaining distance to stop the aircraft.

Despite of the awareness of the anti-skid system performance on Raytheon Aircraft 390 (according to Raytheon Aircraft Safety Communique No.246 – Landing Performance Awareness of June 2004), the pilot did not follow precisely the recommendations of fully applied brakes pressure till the stop of the aircraft. Pilot released brake pedals pressure before overrun the Runway 03 and twice activated emergency brake.

As the pilot reported he tried to use emergency braking very carefully in order to avoid the problems, which happened with the aircraft earlier on December 2004 event, when the R/H wheel were badly damaged.

Therefore he activated the emergency brake twice to a limited extent. This technique could be confirmed by the report of Tower controller, who noticed couple of ,,nodding’’ movement of the aircraft in the proximity to the runway end.

Based on results of braking system components testing investigation team can not conclude that the power braking/anti-skid system was operating normally.

( See attached P/N 38-811-1 S/N 204, Antiskid Power Brake Unittesting results – appendix 4)

3. Conclusions

3.1. Findings

 The main findings of the performed investigation is following:

 1. The pilot was qualified to perform the flight.

 2. The valid AFM used on the aircraft was not revised properly. The issued in May 2004 Manufacturer approved AFM supplement for airplanes operating on Wet and Contaminated runways was not introduced into the AFM before the Serious Incident happened.

Therefore calculations for landing distance were made incorrectly.

 3. The pilot did not follow procedures established by the AFM and in the Raytheon Aircraft Safety Communique No.246.

 4. According to CVR readout there was no indication that throttles were retarded to idle before touchdown, as prescribed in AFM procedures.

 3.2. Causes

3.2.1. Primary cause

- The pilot exhibited poor judgement in landing at Ventspils airport which has a runway length that must be considered to be at best marginal for a landing under the prevailing conditions. With a 1 298 meters(4259 feet) runway and an air distance of as much as 291 meters (954 feet) at a 3 degree glideslope, the runway available for braking would be

1 007 metres (3305 feet). For slush runway the calculated total landing distance will be 1279 meters (4194 feet) and ground distance 821 meters(2694 feet). According AFM landing performance graphs for 3 seconds duration flare the touchdown point will be extended from 954 feet to about 1500 feet (457 meters). In this case ground distance will be about 841 meters (2758 feet) Runway length remaining is 20 meters (66 feet).

- Not appropriate using of the power braking/anti-skid system by pilot on short field landing. Based on perception, the brake pressure was released, which cause the aircraft going out of the runway.

- The throttles were not retarded to idle at 50 feet AGL as prescribed in AFM procedures.

 3.2.2. Contributing causes

 - AFM was not revised appropriately with AFM supplement for aircraft operating on Wet and Contaminated runways. If landing distance calculations would be made using revised AFM, then possibly the flight to EVVA airport could be either cancelled or delayed until getting more acceptable runway conditions.

 - The pilot plays an important role in minimizing stopping distance by using a touchdown technique that rapidly develops maximum vertical load on the landing gear and the consistent application of brakes. This is important to ensure rapid wheel spin-up and to allow the development of the maximum braking force. Further, the pilot must maintain a high and consistent brake pedal force to allow the braking system to develop maximum braking force.

 4. Safety Recommendations

 The following recommendations are made:

  Safety Recommendation 2005 - 1.

 Revision service should be provided by manufacturer in more proper manner, to avoid airplane operation without introduced revisions. In the AFM on board of the aircraft, on the moment of incident, there was no evidence that issued in May 2004 AFM Supplement (before delivery to operator in September 2004) was inserted into the manual.

If some revisions in aircraft documentation to be made, especially which are related to the safety, the all responsible people from owner/operator side should be informed urgently with the request of receipt conformation.

 Safety Recommendation 2005 – 2.

 Operator should establish and maintain RAC Technical Publications Safety of Flight Information (SOFI) monitoring process by regular check and search for revisions on applicable Web site: www.racpubs.com.

 APPENDICES:

 Meteorological information

Aerodrome information

  1. Radio communication records and CVR transcript
  2. Braking system testing results

5.   Spoiler Control Unit (SCU) Test Pt. No 233700-106 S/N 012

6.   EGPWS Test Honeywell International Inc.

7.   Raytheon Aircraft Safety Communique

      8.  Summary of interviews with the Pilot.

      9.   Summary of interview with the AFIS operator.

10. Damage Survey following incident at Vetspils Airport 25of the January 2005.

 

16.12.2005.

Investigator-in -charge

Head of Division Aircraft Accident and Incident

Investigation Ministry of Transport of the

Republic of Latvia                                                                             Ivars Alfreds Gaveika