f-pm980928a
REPORT - TRANSLATION on the incident on 28 September 1998
at Saint Etienne (42) to the Beechcraft 1900 D
registered F-GRPM
operated by Proteus Airlines

F O R E W O R D

This report presents the technical conclusions reached by the BEA on the circumstances and causes of this incident. In accordance with Annex 13 of the Convention on International Civil Aviation, with EC directive 94/56 and with Law N° 99-243 of 30 March 1999, the analysis is intended neither to apportion blame, nor to assess individual or collective responsibility. The sole objective is to draw lessons from this occurrence which may help to prevent future accidents or incidents. Consequently, the use of this report for any purpose other than for the prevention of future accidents could lead to erroneous interpretations.

SPECIAL FOREWORD TO ENGLISH EDITION

This report has been translated and published by the BEA to make its reading easier for English-speaking people. As accurate as the translation may be, please refer to the original text in French.

Glossary

DGAC

Direction Générale de l'Aviation Civile

(General Civil Aviation Directorate)

FAR

Federal Aviation Regulations

FL

Flight Level

ft

Feet

JAR

Joint Airworthiness Regulation

kt

Knots

lb

Pound(s)

QNH

Altimeter setting to obtain aerodrome elevation when on the ground

UTC

Universal Time Coordinated

SYNOPSIS

Date of incident

Aircraft

Monday 28 September 1998 at 16 h 00 UTC[1]

Raytheon Beechcraft 1900 D registered F-GRPM

   

Place of incident

Owner

Marseille FIR, 40 NM west of Saint Etienne

Kansas Beech Leasing Inc.

10511 East Central

   

Type of flight

Operator

Scheduled public transport flight PRB 405 from Saint Etienne to Bordeaux

Proteus Airlines

 

Persons on board

 

2 crew, 8 passengers.

Summary

About eleven minutes after leveling off at FL 190, with the engines set at 2 950 lb.ft torque, the crew noticed two torque surges on the right engine up to about 4 500 lb.ft, followed by a clear drop, then by stabilization at about 2 000 lb.ft. They decided to shut down the engine and return to the departure aerodrome. The return, approach and landing were performed on one engine with no further problems.

1 - FACTUAL INFORMATION

1.1 Flight Crew

On the day of the incident, the copilot was Pilot Flying up until the event, when the Captain took over the controls.

Captain

Male, 39 years old, employed by the company since 1982, Airline Transport Pilot'sLicense since 1982, type rating in 1997. License valid until 31 December 1999.

Flying experience :

  • total flying hours : 8 612 of which 1 000 on type,
  • in the previous 90 days : 172 of which 159 on type,
  • in the previous 30 days : 72 of which 69 on type
  • Copilot

    Female, 33 years old, employed by the company since 1998, Airline Transport Pilot'sLicense since 1991, type rating in 1998. License valid until 31 December 2000.

    Flying experience :

  • total flying hours : 1 130 of which 300 on type.
  • in the previous 90 days : 107 all on type.
  • in the previous 30 days : 37 all on type.
  • 1.2 Aircraft Information

    1.2.1 Aircraft

  • Manufacturer : Raytheon Aircraft Company, USA.
  • Type : Beechcraft 1900 D.
  • Serial number : UE-300.
  • Service entry date : 6 February 1998.
  • Certificate of airworthiness : situation V, category N/TPP, issued on 5 February 1998, expiry date 3 February 2001.
  • Flying hours : 1 070.
  • Number of cycles : 1 304.
  • 1.2.2 Engines

    Number of engines : 2.

    Manufacturer : Pratt et Whitney, Canada.

    Type : PT6A-67D.

    Left engine :

  • Serial number : PCE-PS 0181.
  • Total hours : 1 070.
  • Right engine :

  • Serial number : PCE-PS 0180.
  • Total hours : 928.
  • Hours since overhaul following foreign object damage : 109.
  • Both engines were installed on the aircraft when it was new.

    1.3 Meteorological Information

    Observation by the Saint Etienne Bouthéon station at 16 h 00 :

    Wind 080° / 4 kt, visibility over 10 kms, clouds 1/8 of St at 300 m, 3/8 of Cu at 1 200 m, 7/8 of Sc at 1 500 m, temperature 16 °C, QNH 1009 hPa.

    Estimated conditions at the place of the incident : sky very cloudy to overcast with 1 to 2/8 of Stratus towards 1 000 ft, 3 to 4/8 of cumulus towards 3 600 ft, 5 to 6/8 of stratocumulus towards 4 700 ft, then doubtless 5 to 6/8 of altocumulus, taking into account the air mass. Humidification at lower levels at the end of the day with frequent rainy periods. Visibility very variable, from good to very good to mediocre in the rain. Weak ground wind turning to north-east, 5 to 8 kt. Wind at altitude from west-north-west : at 500 m 8 to 12 kt, at 1 000 m 15 to 25 kt.

    1.4 Tests and Research

    1.4.1 History of the Failure

    Following a bird strike, the right engine was removed, checked and reinstalled on the aircraft on 7 September 1998. Following this procedure, a ground run revealed no anomalies. On 8 September 1998, a 40-minute check flight was performed on the Lyon Satolas-Saint Etienne route, which generated no particular remarks on the part of the crew.

    On 10 September 1998, after 12 hours and 42 minutes of flight performed on 14 legs, a problem with the propeller synchrophaser was noted in the Technical Log. On 16 September, during the 5D and 5C checks performed on the aircraft, the connections  were cleaned, and a satisfactory ground test was carried out.

    N.B. The format of these checks does not include any operations on the propellor synchrophaser and the autofeather system.

    A failure in this system recurred on September 17. The subsequent trouble-shooting did not solve the problem. The propeller synchrophaser, which is included in category C in the Minimum Equipment List, was declared inoperative and noted down in the Deferred Work List (cf. § 1.4.3).

    The work carried out following the 28 September 1998 incident allowed us to :

    Note : the report on the ground run carried out on 7 September 1998 includes some parameters associated with autofeather and propeller synchrophaser tests. In the report on the check flight performed on 8 September 1998, the check box concerning the result of the autofeather test has not been filled in. Between the 7 and the 28 September 1998, no comments at all on the operation of the autofeather system were made in the Technical Log.

    1.4.2 Technical Factors

    This inversion of the plugs does in fact lead to the phenomena noted. Firstly, it makes propeller autofeather impossible, and secondly it modifies the operating performance of the propeller synchrophaser, causing apparently random failures.

    The investigation showed that the inversion of the two plugs was possible. Both connectors are in fact identical and belong to the same wiring bundle. This is also the case on the Beechcraft 1900 C. According to how the bundle is tied, a sufficient length of cable can be freed to allow inversion in the connection. This assembly error was probably committed during operations on the engine on 7 September 1998.

    FAR and JAR 23 certification regulations do not, for this type of equipment, include any obligation in design or distinctive marking which would minimize the possibility of mis-assembly. The Beechcraft 1900 D maintenance manual gives information, in the sections on "propeller autofeathering" and "propeller synchrophaser", on troubleshooting the causes of failures, describes probable malfunctions and action to be taken to remedy these. It gives details of operations to be performed and the precautions to be taken, but does not explicitly mention the possibility of such an inversion.

    Note : During the investigation, it was noted that this risk of inversion of the cables was known outside of the airline by some technicians who work on this type of equipment. These technicians stated that they passed on this knowledge when they were called upon to train colleagues. It appears, however, that the manufacturer of the aircraft was not aware of it. In any event, Raytheon never warned operators of this danger, either in the maintenance manual (see below) or in a technical service note.

    1.4.3 Airline Operations Manual

    Proteus Airlines' Operations Manual, section 9, specifies the implementation procedures for limitations and their presentation in the minimum equipment list. The propeller synchrophaser[3] is presented as a category C item with a notation (*) indicating that the time limit for repair is 10 calendar days, starting on the day after notification of the problem at zero hours. The (*) notation indicates that the inoperative equipment must be flagged by positioning an "Inop" placard in the cockpit.

    The same document specifies, in section 6 on normal procedures, a test of the autofeather system at the holding position before each flight. A new procedure being introduced (which came into force on 22 March 1999) had already been mentioned during type rating. The document had not yet been published at the time of the incident. This procedure, which corresponds to that recommended by the manufacturer, specifies that the autofeather test may be omitted by the pilot after the first flight of the day.

    None of the members of the incident flight crew had used the aircraft between the 7th and the 28th September 1998.

    2 - ANALYSIS

    We saw in paragraph 1.4.2 that the design of the aircraft allowed the two plugs to be inverted during a maintenance operation. This is clearly what happened during the reassembly of the engine on 7 September 1998. The error was not identified during the checks which preceded the aircraft's entry into service.

    Two malfunctions could have brought it to light subsequently, these being the non-functioning of the autofeather system and the propeller synchrophaser. failures.

    No problems related to the autofeather system were reported. However, even though it is likely that the annunciators associated with the entry into service of this system were not affected, the test, which should be performed at the holding position under normal company procedures, should have established an anomaly on every flight or, following the new procedure, on the first flight of each day. It is therefore highly likely that this test was not performed systematically at that time, contrary to procedures set out in the Operations Manual.

    A functional anomaly in the propeller synchrophaser was noted twice in the Technical log, on the 10th and 17th of September 1998. This anomaly had still not been corrected on the day of the event, that is to say 12 days after its listing on the deferred work list, even though the time limit for repair of this item is ten days.

    Thus, the incident occurred after failures at four levels of verification :

  • checks during reassembly of the engine,
  • the ground run,
  • the check flight,
  • airline operations.
  • 3 - CONCLUSION

    The incident resulted from errors in maintenance, made possible by the design of the system, and the failure by the airline to apply checking and operational procedures strictly.

    4 - RECOMMENDATION

    In March 1999, the airline indicated that it had taken the necessary steps to correct internal procedures. In this context, the BEA will not make any recommendation to reinforce strict application of checks and operations within Proteus Airlines.

    The inversion of the plugs, which is the cause of this incident, would have prevented autofeathering if it had been necessary. Consequently, the BEA recommends :

  • that the DGAC inform French operators of the Beechcraft 1900 C and D of the risk of inversion of the propeller synchrophaser and the autofeather solenoid plugs during maintenance operations, and request that the FAA, as primary certificator of the aircraft, ensure that a study of a modification be undertaken to guarantee that this error can no longer occur.
  • APPENDIX 1 -Ground run report

    APPENDIX 2 - Check flight report

    APPENDIX 3 - Operations Manual procedures

     

    APPENDIX 4 - Deferred work list


    [1] All times in this report are UTC, except where otherwise specified. Two hours should be added to express local official time in metropolitan France on the day of the incident.

    [2] The crew indicated that they did not operate the propeller synchophaser before the event.

    [3] In the Proteus FCOM, the propeller synchrophaser is called, in French, the "synchronisateur" in section 11 on system description, "synchrophase" in the MEL, doubtless as a result of transfer from the English term "synchrophaser" and "prop synchro" in the Technical log. Similarly the French term "mise en drapeau automatique" also appears as "autofeather".