Crashes of instructional flights: Analysis of cases and ...avhf.com/html/Library/Tech_Reports/cami/AM96-03.pdfLIST OFTARLFsS Gashes of Imauctional Flights.1989-1992 Table Page 1. Injury

DOT/FAA/AM-9613

Office of Aviation Medicine

Washington. D.C. 20591

19960305 031

Crashes of Instructional plights Ms: Analysis of Cases and Remedial Approaches

Susan P. Baker Margaret W. lamb Guam Li Robert S. Dodd

The Johns Hopkins School of Public Health Center for Injury Research and Policy Baltimore. Maryland 21205

February 1995

Final Report

This document is available to the public through the National Technics! Informatica Service. Springfield. Virginia 22161.

U.S. Department of Transportation

Federal Madan Adndaistredon

VMS QUA= 13SEMCTSD

PDF compression, OCR, web optimization using a watermarked evaluation copy of CVISION PDFCompressor

http://www.cvisiontech.com

NOTICE

This document is disseminated under the sponsorship of the U.S. Department of Transportation in the interest of

information exchange. The United Stares Govanrnent

assumes no liability for the contents or use thereof



Tandiedad lamed DOMIRMIlleSCO Nal L Ione Net

DOTJFAA/AM-96/3

1 ScierefrotAccenten 5b5 3. Reciter t Calabs

11199rel Malb

01131123 of lastmctional iPlighis: Analysis of Casa oad Reseal Approsches

Raoatt13519

Mowry 1996

memo SUsan P. Baker. Ainuet W. Lan* bibles Li. aod Robert S.Dcdd

Peen fftwe ationtokte Swat rect

9. Perm apan122112nNowe ales NIS The lobos Hopkins School of Public IinWa Cotter for lajury Rack and Policy

MD 21205 5georcalre Sewcy rata amIASSus

Office of Aviatim liedicine Federal Aviatioa Adminiseradon 800 Indeptodence Ave- S.W.

Webber°. DC 20591

It 1142ettra Na ORA5)

II. Coretal oreCre

91G-045 13. boo of Ispatorss Pored Comisct

Fad Report

Istramg /vac/ Code

iS nip sivamantan

This walk was performed at a grant from the Nueva Resoirces Research Division of the FAA's Gad Aeromethcal bedtime in Oidthocea City. Oklahoma, as part of task AM-A-94-111RR504. Dr. Tom Nashua served as FAA Technical Monitor. it saws honor -J ifeks empaiwee awe tha 300 awls wane awl we ameba is mon thett amums of a mak cabin fear ems tadaskse to irksaty as eireones et istemisool ashes sod dame* tams tamed to Wick shwa ad the aminmonot 11811140D% 3251141tot mks owe aka kr auks dwas 19110-1392 kvolved indent kles ors Sok lir losseatoosi passes PliZark During 19%4992. 1226 irosnames aglow mere knead it cakes. teselag is 250 deaths amt 128 sakes *dee hada coat= det ise those taw yews imbed 38 isonnieril sirgeetts (2.5%) sod aeconted kr 206 arm mama same yam muse atoms slew a sole inert a mar Say wee year bat ono expetioa mica m It arks en sob. Deli ad review of amiss dosing MP sod 1591 sevested ass less Waal es Ismiktelosatertad Z2 et the 638 auks, et 36%. Omuta mostaled

3316 eta assin Ile 94 stab owe MIS albs teats sod 46% ots% tad auks. Sake mks mere 51% at Me seam IS3 Sae 360 estritems ea sob tomokel 111 toss of amoral as !sodas lama Tawmad ptakyacasaeodacZSoralcalmsassmio. Arne Sett sada co cosavatatty sass. 26 (3111)esese dee to tasicmg ow et rity-foot(40%) erase tailless as aldose as aoss-emsrmy seas hat raffia ft* piss. Tosottikee Was kat wia sabot hos owe ayes etersive ese es Wit peseespes. k 13 Sae 25 amass doe a aaDa tesicia,riepi eta kips dories amine pose. brumes= wee mese to 30% at erwhes tom mob sod 32% oases km fat stet Simokted wonstoks ended 149 auks. Tie R115P2 it stified taws abed bo ere Snot se cossiboloty b oredisd of dm ant. Maw stalms moat a MRS eimened garbs viscamtemes lake so the auks. For essaple. lait a sberet, Mersaisaridairalai= atasreosiosiaipdedsecantPil otedbyasadeeasodsahoodasdteoartsskawig helicepos. mecum/a Grater eesplasis duds. tieWs tasks weds to he pima on acidic ma ad Skit allisiees. assers atemises. aid wassmodog the drab distaffeat) Sae eight dyareks win sob motitely ramiem bawl= eamaete to casks bedt dicer* dada tat anti= ad. leu Meaty. thatek ate to exonstally saw theis sakes ad any the an Ea% sod Fats flight. Ptabbas idwaBelli o this say tad comemiested Ilet imorecons ask ital inking as barsams. ditties preandoo kr their Mama alicestae. ard dom. sksenismicts dimwit ti ** barn 17.1thy Wads Plight Trak* Itankti08111 Rights Aircraft Cashes Ceetified Flight Instruction NUB Accident Database ASRS bead= Database teleoaavetat04eatefea 9 a Seca Oat (on* QOM

Unclassified Peon Darr 1700.7 (8-72)

Unclassified

1:2116514to Sialwriar*

Doan= is available to the public through the Maim& Technical Information Service *infield. Virginia 22161

21. Ka onacips 22 Pico

56 aepicasman at catmesoct pop cutasted



ACKNOWLEDGMENn

"We thank Robert E. Blanchard. Ph.D. and Thomas E. Nothus. Ph.D. of the Fedetal Aviation Administration for facititatingall aspects of the study and providing valuabk advice and feedback on our reports and Stanley Smith of the National Transportation Safety Board for providing the data for this investigation.



TABLE OF CONTENTS

A. Damiption of Coshes of Iastraaional Plights ?uncanniest Methods Results

Discussion Conclusion

1

1

1

2

17

18

B. Details of Individnal Crashes 18

Fatal Stalls 13

Simulated Emergencies 20

Loss of Control in Airport Surface 'Winds 22 Toothsand -Co's 22 Fed Starvation 23 Flight Instructor Performance 24

Student Pilot Carrying Passenger 25 Midair Collisions 25

C. Dissemination of Study Results 28 Highlights of Crash Repons 28 instructor Deficiencies 29 CFI Insights Into Student Behavior 29 Cases and Comments 30 Conveying the information to Right Instructors Recommendations

33 35

D. MRS Data 35

E. References 44



LIST OFTARLFsS

Gashes of Imauctional Flights. 1989-1992 Table Page 1. Injury Severity 2

2. State of Octant:ace 3

Crashes of bastructiond Meta 1989 and 1991 3. Phase of Hight by Type of Flight 5

4, Crashes of Aircraft Practicing Touch-and-Go Leans 6 5. Circumstances of Crashes by Type of Right 7

6. Circumstances and Outcome of Crashes Involving Stalls 8 7. Mechanical Failures 8 & Mishandled Emergencies 9 9. Crashes Resulting From Simukted Emergencies 9 10. Winds in Relation to Circumstances 10 11. Number of Crashes by Wind Speed and Angle with Runway Heading 11

12. Age of Students and Instructors 11

13. Principal Profession of Pilots in Crashes 12 14. Airplane Ratings of Pilots 13

15. Total Hight Time 13 16. Type of Soto Flight by License 13

17. Pilot Perforsnance Factors by Type of Flight 14 it Pilot Errors Leafing to Loss of Control on Takeoff or Landing 15

19. Flight Instructor Factors Cited by NTS13 15 20. Injury Severity by Circumstances of Crash 16 21. Factors Related to Pilot Survival 17 22. Examples of Hasty Actions by Student Pilots Hying with Instructors Resulting in Crashes 30

Aviation Safay Sporting System Dun 23. Ancona= in 164 ASRS Repent 36 24. Phase of Flight he 164 ASRS Reports 37 25. Hight Instructor Factors in 164 MRS Reports 37 26. Pilot Performance Factors in 164 MRS Report 38

LIST OF FIGURES Figure rage 1. Percent Distribution of Time of Crash. All Crashes vs Wait Collisions, 1989-1992 4 2. Proportion of Crashes in Eads Phase of Flight (Based on Phase of Flight. 1989 and 1991) 6

vi



ABBREVIATIONS

ac aircraft AC FAA advisory circular apt airport ASKS Aviation Safety Reporting System ATC air traffic control ATP airline transport pilot BM bicanial flight mire CFI flight instructor CIAF - common traffic advisory frequency eng FAA FRO

(Pm hood

engine Federal Aviation Administration

fixed base operator feet per minute

vision -restricting device for instrument training

lAP initial approach point IFR instrument flight rules as ------ instrument landing system IMC -instrument meteorological conditions IP instructor pilot lets knots LAX Its Angela airport LOC loss of control

LTT M/M MAC mil nul ndb

vii

light tampon (14301-30.000 lb) make and model midair collision

military altitude above mean sea level

non directional beacon NMAC near-midair collision NTSB - National Transportation Safety Board PIC pilot-in-command PTS nvy sins SMA SMT T&G tic TO TT curb Mt VFR zing

Performance Test Standards runway

simulated small aircraft (<5001 lb)

small transport (5001-14.500 lb) touch-and-go

traffic takeoff

total flight time turbulence

tower visual flight rules

crossing



EXECUTIVE SUMMARY

METHODS The National Transionation Safety Board (NTSB)

terms all aviation crashes (°accidente in NTSB terminology) that result in death within 30 days,

serious injury. or substantial damage to the aircraft. We we the terra `crash* because it isa more scientific teem-without the connotations of bad luck. ran- domness, and enpreventability.

NTSB computer tapes for instructional crashes of fixed-wing civilian airplanes during 1989-1992 were analyzed. Crashes of helicopters, ultralights, anti hamebuilt aircraft was excluded. For all midair col - lisions during the four years 1989-1992 end for all crashes during 1989 and 1991. two-page briefs from NTSB describing the crashes were read and mega rived as to the circumstances of the cashes and con- Shining futon. The years 1989 and 1991 were

representative of the four-year period. In addition to NUB data. a lay -woad search of

NASA's Aviation Safety Reporting System (ASKS) was made for reports related to instructional flights of general aviation airplanes during 1992 and 1993. These years were selected because they were the mon recent for which data were available; 164 reports that met study definitions were abstracted and categorized in a manner similar to the crash reports.

RESULTS

L NUB Daus 1909-1992 During the four years 1989-1992. 1226 instruc-

tional airplanes were involved in 1218 crashes included in the NTSB files. The enmities included 250 deaths, 128 serious injuries, and 270 minor injuries.

Mak cakes Thew midair collisions involving 38 instructional

airplanes occurred during the four years. They comprised 2.5% of all crashes of instructional flights and accounted for 20% of all deaths (508250). All °c- anted in VMC weather and during daytime hours, batmen 0800 and 1959 It. Sixteen of the trainees were

ix

piket-in-command on solo training flights compared with trainees on solo flights involved in cashes other than audits, these pikes were younger (with a median age of 24 vs 34) but more esperieneed - i.e.. they had more flight dine and were more likely to be

private pilots undergoing advanced training. Aircraft involved in the midair collisions included two helicopters and two bi-wiag acrobatic airplanes. Visibil- ity nary have been affected by the angle or gkze of the sun in four ease; and in six instances the student was

receiving instrument training, in which the trainee is

usually under * hood (a vision-Restricting device).

IL NISS Dams Details of Crashes in 1989 and 1991

During 1989 and 1991. 635 crashes occurred, involving638 instructional airplanes. The findings in this section relate to the detailed review of the crashes

of these 638 fiqOurs.

Fifty -one percent of the flights were solos of pilots with student enamors an addItional 5% were solos

of pilots with private licenses who west undergoing ins:mains for more advanced ratings. Thirty -mine

percent of crashes occurred in connection with dual instruction and 5% on checkout Bights or biennial Eight mien

Pine d light The most common phases of flight when the prob-

lem a determined by reading the two-page NUB briefs, were Unsling (26%) and touch-and-go's (19%). The 118 crashes on touch-and-go landings included 79 solos and 39 with an instructor. The touch-and- go's on solo constituted 22% of all crashes on solo flights and 19% of all crashes. Three out of four inadvertent gear retractions occurred when a dad flight (i.e., with an instructor aboard) was about to make a touch-and-go landing.

Crashes on go-arounds numbered 56, of which 27 (48%) involved stalk Five airplanes struck wires on go- ansunds. Fifteen go- around crashes followed simulated emergent



isansabet Saw"

Cinnamon Leas of control *a landing was the most common

type of crash, realties in 227 mashes (36%). Eighteen

of the crashes moaned while meaning a go-around following an unstacetafig stump to land. Twenty one occurred in connection with a simulated emergency. Crosswinds were involved in 109 cases orlon ofcontrol on landing (48%) and tailwinds in 21(9%).

Stall was the printery event in 94 cases (15%). An instructor was on board when half of the stalls occurred. Most stalls were taketadeparture (39 cases)

or approach to landing (28) nails. Of the 39 stalls on takeoffideparnue, 23 occurred on go- grounds, 17

involved crosswinds and 8, tailwinds. Thirteen simulated emergencies resulted in stalls. The 31 fatal stalls accounted for 46% of all fatal crashes.

Thad aunties resulted in 74 crashes (12%), 29 of which were on cross -country solos. An instructor was present in 24 cases, including 10 of the 14 frights on which the fuel selector was set on an empty tank. Nineteen crashes occurred when fuel was adequate, but the pilots misunderstood the fuel system or set the fuel selector for the empty tank. In 51 cases, the fad was exhausted because the pilot(s) misjudged the available fuel; Cessna aircraft, in which the high wings make visualization of the fuel more difficult, accounted for 861lbsithen cases, versus 60% *fall other czadats in the study.

Mechanical failure was the primary cause of 80 crashes (13%). The major problems were unexplained loss of power (31 cases) and failure ofthe landing gear (13). N ine instructon mishandled the resulting emergency, or even compounded it with an improper response.

Midair collisions numbered 15 (3%) and involved 18insetuctiondalphsatt.ofsshich lOweredualistsauction ached/am fight. the decries the 30 mishit coffering:

during the mire fstar-yess period.) Carburetor icing downed 25 flights (4%), 10 with

instructors on board. 3 in connection with simulated forced landings, and Goo student crose-country flights. Thirteen of the crashes were caused by carburetor icing during cruise, a phase off light when comet= icing is not usually anticipated.

S

Win unites occurred in 11 instances (2%), 5 of which involved a simulated emergency during dual instruction or a checkride.

Simulated emergencies led to 49 crashes (8%), predominandy loss of control on landing and stalls. Instructors sometimes killed the engine with the mixture at idle-cutoff or by shutting off the hail supply, then were unable to restart the engine.

Winds mac airport were an apparent factor in 232 crashes (36%). Crosswinds were involved in 28% of all crashes and tailwinds is 6%.

Piles The median total flight dine was 43 bouts for

trainees and 1552 hours for instructors. Eleven in- :Raton had less than 10 hoots time in type.

Eleven pilots with student licenses had more than 200 hours total time, suggesting that many of these "students" were not still flying under an instructor's supervision.

The primary circumstances of the 84 crashes on moss-country solos were fuel starvation (33%) and loss of control on leafing (31%). Eight of the students crashed after becoming lost

Twenty-three pilots flying with student licenses were illegally carrying one or more passengers.

Pilet performs/mat Poor crosswind conceded was apparent in 79 crashes,

a bounced landing in 48, adder misuse in 42, and poor has of =face wind or tubules= in 40.

Thirty -four of the 84 trainees who crashed on solo cross-country trips (40%) had not filed flight plum 6 of tee 34 were lots at the time of die crash.

Mitten students were geographically lost when they crashed; 8 had run out of fit&

lastrnmac parfitmemos Committals instructor-rated factors cited by

the NTS8 included ha/dogma supervision (77 cases), delayed remedial action (34), mishandled simulated emergency (20), and inadattent stalifspin (13).

Tuentraven percent of the solos reflected inadequate training, pdmatib it evaluation ofcrosswinds and weather (69 cases), touch-and-go landings (48),



and recovery from bounced landings (35). Pity -four instructors let the student get so far into an operational problem that the instructor could not recover

the aircraft. Nineteen crashes occurred because the CF1 did not anticipate a student's hasty action. Ten crashes occurred after the CFI: simulated forced landings in areas that did not afford safe landing places when the sirnulated emergencies turned into real ones.

Aisplane cbastattistica Twin-engine planes comprised only 6% of the

series, but 18% of the trashes following simulated emergencies and 13% of mechanical failures.

Tailwind airerah (61) were involved in only 9% of an crashes, but in 17% of postmen and 79% of ground loops.

Aircraft with retractable landing gar (86) constituted only 14% of all aircraft, but 26% of die crashes due to rnedtanical failure.

Crash outcome At least one fatality occulted in 11% of crashes.

The crashes that were most likely to be fatal were VFR into IMC (71% fatal), midair collisions (44%), and stalk (33%).

The pilot was mote likely to be killed if the weather was IMC. postcrash fin occuned, the aircraft was a twin-engine, or the pilot was nor restrained with a shoulder harness.

Nine percent of trainees and 16% of instructors known to have a shoulder restraint available were not wearing it.

HI. MRS DATA

The MRS data describe incidents that did not tea in crashes, but they offer insight into the cirannstances, as provided in the pion' own weeds. In get .al, they were consiseent with the findings front the study of coshes. Of petit:at inmost were the 34 nair-midair collision (NPAACs), which comptised 11% of the re- pa runner to the 30 midair collisions, they include' 3 aces in which a cadent was under a hood, and two cases involving helicopters. A description *Mc NMACs is provided in Section 13.

amain Searass,

Although ASKS is a voluntary reporting system

and. thenfore, not representative of all incidens, it is

noteworthy that many of the circumstances described in the ASKS reports are similar to those in the cashes. This suggests that it would be of potent! advantage to make greater use of this data base for identifying problem areas and their causes.

RECOMMENDATIONS

Special attention should be given to collision avoidance daring training eights in the tarn pattern, where midair collisions involving instructional flights typically occur (DeLacerda 1988). Avoidance of midair collisions in VFR flight primarily depends upon pilots to ̀ see and avoids other aircraft. a require- rant wida recognised limitations (NTSB 1985,1988). Pilot broadcasts of position and intention do not relieve the pilot of the obligation to thoroughly scan the area for other aircraft. Pilots working with ATC must be alert to the positions of other aircraft being controlled by Alt Helicopters, which do not fly a standard pattern, and bi-wing acrobatic aimknes, in which the pilot's visual field is limited. were each invoked in 2 of the .10 midair collisions. despite the fact that they representonly& very mall percentage of aircraft at airports where students undergo training. Pilots of these aircraft and students who share their traffic environment should rake organ= of the increased risk.

The abniay to cope 45 ataswilsis deserves an- Oats' since this appeared to be a factor in more than one-fourth of all crashes. Students may requite more dust inSEMCIii0EI under challenging crosswind conditions. Although the usual practice is to select * runway with ideal wind direction. it is often possible to choose a runway where the winds have a substantial crosswind component. Greater framing may occur if an instructor pidts runways with less favorable winds for landing peactice, until the student is proficient is crosswind landings and competent at assessing safe crosswind landing criteria.

Touchand-ges requisegnateentainy.This nee- tin. common to almost one-fourth tithe crashes, has me advantage of retaking the number of landings



frortribeSastaawy

that an be made in a time-limited lesson. It has the disadvantages, however, of 1) requiring a complicated series of operations in a few moments, often at high speed and 2) not affording the student the time needed to reflect over each ht and identify any deficiencies before initiating the neat. The specific

dynamics of flight canary/ usage (rudder, aileron, elevator, elevator trim, and flaps) and power usage

(throttle and carburetor beat) must be understood by the student in relation to landing rollout, transition to takeoff phase, and takeoff. Allowing touch-and-go's on the first few solos is inappropriate unless mastery ofthese elements has been achieved. The FAA should amend the Flight Instructor Practical Test Standards

to include instructions for teaching touch-and-go

landings and criteria for determining when students

are ready to practice touch-and-go's dtuizisolo flight. Adherent to guidelines far simulated emergen-

cies is crucial. Standards for practical tests of flight instructors sate that in the case of she- engine airplanes the examiner shall all the instructor appli- cant that such practices as "placing the fuel selector in the soff position or placing the mutate control in the Idle-cutoff position' to simulate a power failure 'are violation of FAA policy,' and that in the case of multi-engine aircraft this shall not be dom below 3,000 feet (FAA. 1991) Seven crashes, including 3 in twin-engine airplanes, resulted from shutting off an

engine by such means, rather than reducing power with the throtde. Furthermore, although continuing a simulated emergency approach below 500 feet s similarly (=bidden. it was dear that many crashes

occurred because recovety nes initiated too low. That 14 simulated antigenic: terminated in damaging hard

ail

landings or stalls dose to the ground semen that student pilots nerd additional training in power-off Wow.

Instructor mad to set a good maple, a fact that should be emphasised in instructor training. When an

instructor descends to within 150 feet of the ground while demonstrating a simulated forced landing, or initiates a stall at a low altitude, it sets a poor example for students. Evidence of some instructors' faun to set a good example was the fact that one-sixth of those

with available shoulder restraints were not wearing them. Moreover, pilots not wearing shoulder restraints have been shown to have more than three times as

great a ritk ofbeing killed when a crash occurs as those wearing shoulder restraints (Li and Baker, 1 993; Ed= and Lamb, 1989). For their own safety, as well as that of their students, instructors should stress the use of restraints by all aircraft occupants.

leanntor re- licensors an be red to convey that recomaaesdations, as well as the results of this research. Right instruction is the only category of Hying that requires re-licensute by the FAA; every two years,

a Right instructor must obtain a new centime from the FAA based upon training givers. or after tiding a

recertification muss t flight check ride. This contact provides a unique opportunity for information transfer between the two patties. la addition to recertification courses, tarts of the study an be incorporated into other modalities, inchrdinw pamphlets and newsletters distributed to all Can the FAA's Accident Prevention stria, which should be sent to all instructors and new students: modules used for initial training ofinstructorn and the FAA's Flight Instructor Practical Test Standards.



CRASHES OF INSTRUCTIONAL FLIGHTS ANALYSIS OF CASES AND REMEDIAL APPROACHES

A. DESCRIPTION OF CRASHES OF INSTRUCTIONAL FLIGHTS

INTRODUCTION

Crashes of instructional flights, averaging more than 300 each year, comprise 14% of all general aviation crashes. Their occupants account for 7% of all fatal and serious injuries that occur in general aviation. During 1987.1992, instructional flights were involved in 36% of all midair collisions (NTSB 1987-1993).

Despite the size of the problem, there has been no published research on the circumstances of crashes of instructional flights and the characteristics of the pilots involved. The content of flight training 'often is based upon tradition and upon insure:tors' judg- ments and unique experiences rather than upon detailed. systematic analyses of piloting task? (Cam, 1988).

Research was therefore undertaken to provide information that can be used in the development of relevant educational materials. The objectives were to determine the circumstances under which crashes of instructional flight occur, identify factors involved in suckcashes, and analyse the relationships between the circumstances of the cashes and the characteris- tic of the instructors and their trainees.

METHODS

The National Transportation Safety Board (NTSB) reports all crashes of non -government aircraft that result in death, serious injury, or substantial damage to the aircraft. NTSB computer tapes for instructional crashes °fazed-wing airplanes in 1989 through 1992 (the most recent years available) were analyzed. A four-year period was selected to obtain a large number of width collisions for analysis and to make the results generalizable. Selection of cases was based

' The exliag some is angst* fine the invenigetan upon roam

I

upon 'whether 1) the purpose of the flight was coded by the NUB as instructional (this included checkrides and biennial Right reviews (Nits) and/or 2) the pilot - in- command was i stud' ent, since pilots flying on nutlet :ionises are required to be under the supervision of an instructor. Crashes of helicopters and ulualight or bona:built aircraft were excluded.

Cases included not only student pilots undergoing their initial training but also pilots holding private, commercial, or even ATP licenses who were undergoing advanced minks checkouts, or air cattier quail- ficatinnipmficiency training. Inclusion of these cases

was based. in put. upon the high crash rates of piles of air saris and small commuter aircraft (Baker and Lamb, 1992; Baker et al.. 1993).

To obtain wore details than were in the coded data, the PITSIES two-page kids describing all midair collisions in 1989-1992, and all crashes in 1989 and 1991 were read. These cashes were categotited as to their drannstanoss and contributing factors, including pilot performance, errors contributing to loss of control, emergency handling. training deficiencies, and poor decisions.' These itsterpreationterne coded and combined with information on the data rapes for the purpose of condo cting modtiple-varaide analyses. The realm of the cues from this portion of the analysis, described under RESULTS: Part U, consti- tute the major part of the study findings.

Reading and coding the circumstances described in the briefs of ell of the crashes was an extremely time- consuming process; because of time and budget con- saaits, the briefs were studied and analyzed only for two years. Bated on analyses of the NTSB -oiled data for the years 19894992, the years 1989 and 1991 were representative of the four-year period with re- gard to circumstances dike crashes (e.g., wind speed), characteristics of pilots, and outcome. The NTSB investigations of 1992 cashes had act been completed at the initiation of this project; therefore, the 1992 cases were not chosen for review of the briefs.



awoke stagartnead' MO*

Denominator data as the number of instructional Rights was not available, making calculation of tams

impossible. Internal comparisons, however, permit- ted identification of associations between circuits- sooners of the crashes and characteristics of the pilots and aircraft.

Put II. NISI Data: 1989-1992 During the four years 1989-1992. 1226 instruc-

tional airplanes were involved in 1218 crashes. The crashes involved collisions with terrain, trees, foxes, etc., or with another aireah, plus one fatal propeller injury to OA-ground personal.

The casualties included 250 deaths, 128 serious injuries, and 270 minor injuries (Table I). Twenty- one of the fatalities were occupants of other aircraft involved in midair collisions with instructional airplanes. Of the 1226 pilots-in-command of the instructional airplanes, 126 00%) were Hied.

The states with the largest numbers of crashes were California (150), Fluids OWLTam (97).kabigan (55), Arizona (54), and Colorado (54) Fable 2). While the large nuntbets of crashes in C.alifornia and Texas probably correspond to the large: populations of those states, as well as the popularity of general aviation, the many cases in Florida and Arizona may reflect the presence in those sates of large flight

training, :thuds. Alaska. where 7% of all general

aviation crashes occur (NTSB 1993), had only 47 crashes of instructional flights, 4% of the total.

?akar collisions Thirty midair collisions involving 38 civilian in-

structional airplanes occulted during the four years.

They accounted for 2.5% of the crashes of instructional flights and for 20% of all deaths (501250). Eight of she nada* were collisions between two civilian instmaional airplanes. In a ninth case. a

milituy airplane (not one of the aircraft in the study series) smack a civilian trainer; in both aircraft, the trainees were undergoing dual ilLMUCtiOra.

Twenty-two of the 38 trainees involved in midair collisions were with an instranor and 16 were on solo flights. The 16 trainer on solo flights were generally more experienced than trainees not in midair collisions: 75% had over 50 boats total flight tints pared with 40% of pilots in other crashes on solo. Of the 16 solo pilots is midairs, 7 (44%) had private licenses, compared with 9% of solo trainees in other types of crashes. Although more experience4 they were younger The median age of these 16 pilots was 24 years. Tams 34 for other trainees.

Faint for four cans, the midair collisions occurred in or near an airport traffic pattern. One of the four exceptions involved two private pilots in training at the same flight school who were flying cross-

Tablet Sy Sew* Crashes of hauclansi ROO, 1289-1992

7 OCCUPANTS OP

INSTRISTIONAL AIRPLANES

TOTAL AlUISEtr OF PERSONS

X rF 220 12.1 250 122

Serious 123 (LS 125 Minor 201 13.7 270 13.7 Nora UM SU 113

1975 TOTAL 1009 100.0 100.0 I

°Was occupenes et honcoptes and nonpfratudional *Mama Invotaid In midair colthions.

2 PDF compression, OCR, web optimization using a watermarked evaluation copy of CVISION PDFCompressor


Osier arnonenkreal Alen

Table 2. Stabs at Ocatetence Crashes at Instnational Rights, 19al 992

STATE * STATE I STATE Alabama 17 Ladies 11 Ohio 35 Alaska 47 Maio: 6 Mahone 17 Mans 54 thoted 18 Oregon 21 Manses 9 gassechusetts 12 Puny halm 32 California 150 Michigan 55 Rhoda band 3 Colorado 54 Minnesota 30 South Carolina 5 Comacticui 7 Misainippi 13 SotAh Dakota 1

Delaware 2 thasouri 28 Tunas:: za Ponds 108 Montana 11 Tape 97 Gator& 15 Nettania 6 Utah 19 Nana 2 Nevada 12 Vermont 5 Idaho 4 HINT HarnaaNrs 3 Virginia 25 Ends 48 leaw army 17 Wannglon 43 BMW* 29 New Mato 16 Wed termini: 6 Iowa 8 New Yo* 26 Wasonain 20 Karam 18 North Caro** 17 Wyoming _2 lOadocky 4 Nest Dakota 12 TOTAL' 1219

lErciudar 7 caws n Ma t Ocont, amigo cony, Paid Non. or tonknova

country in formation. In addition to *is case, there were three other cases in which pilots at the same Eight ac cool corded with one another.

In three cases, one airplane descended on top of another - on final approach in one case, and in two instances, onto an airplane that had just landed.

Two tithe collisions involved a Pint (an acrobatic bi -'Ang airplane), one of which descended onto a Banta on final; the other Pitts collided over the takeoff runway with a helicopter. In another midair, a climbing helicopter collided with a Cessna practicing touch-and go's.

All midair: mama! in VATIC weather and during daytime hours. between 0800 and 1959 h. Figure I

shows that the midair* are more likely than other crashes to occur between 0800 and 1000 It, or in the

3

Actaeon between 1600 and 2000 h-- periods when the pilot's ability to an mother aircraft may be dimin- ished by the subs angle.

The bride indicated that one nude" °canted at sunset, sun in the pilot's eyes was a factor in a second. and is a dint the investigator noted that sortglare reduced risibility. In a /math midair, the sun was to the left of and in fiont *Moth Omsk which collided while Eying parallel westerly courses.

Six of the midair: occurred during immanent =into' which types ily is conducted with the trainee under a hood (two briefs specifically ramtioned that the traineewas under a hood). In 4 of these 6 instrument training amok other aitcraft was coming lava the right (instructor's) side of the wirplane. One student was under a hood practiang *sit week nearieuvere with as



Osage greastvettases- !Agit

instructor, when his Cessna in was hit in the right side by a military T38 going 330 knots during dual instruction.

In at least one midair collision, recommended radio procedures were not followed. In at least 6, the pilots of bothairtraft made radio calls, but did DOI see and avoid one awather's aircraft. In 5 other midair:, ATC (air trafficcontrol) did not provide separation to 2 aircraft that collided, although both were in contact with ATC; in I case, ATC was controlling 1 IM aircraft. but was apparently unaware eta Vat aircraft in the area before the two collide:J.

Each of the midair collisions is summarized in Section B.

25

20

15 C C o t. O I

10

5

0

Part IL WISE Dear Details Casskett la 1989 and 1991

During 1989 and 1991. 633 crashes occurred, involving 638 instructional airplanes. The findings in this section relate to the detailed review of the crashes of these 638 flights. Descriptive summaries' of many of the awes are to be hand in Section B.

Fifty-one percent of the flights were solos of pilots with student certify:sten an additional 5% were solos ofprivate pilots who were working toward a commercial license (Table 3). Thirty-nine peratnt of crashes occurred in connection with dual instruction, and 5% on checker:It flights or biennial Eight reviews.

Fes I Percent Distribute= of Time of Crash

MI Crashes vs Midair Cellisions, 1989 -1992

I:1 Ail Crashes II Midair Collisions

2 4 - 6 11- 10 12- 14. 18-

Local Time (00 it)



Owls sfb sosnisaa Highs

Tait L Phase et Flight by Type of Ned d Insblational Flights. 1989 and 1991

PHASE MOW

SOLO PRIVATE

PILOT SOLO

OVAL OFR OR CHECK.

RIDE TOTAL

I % Takata 21 1 21 1 44 6.9 anthout 12 0 35 1 48 7.5 Crain* 52 s 26 3 66 13.5 Demi 7 4 9 0 20 3.1 Dawnsind or bow Ing

7 1 4 1 13 2.0

Ftrisl 8PProth 8 2 17 5 32 5.0 Lending 99 7 50 9 185 25.9 Go-Mond 26 2 25 3 se aka

Taxi 4 1 3 0 a 12 Not in ention 0 1 6 0 7 1.1 Touehond-go 78 43 30 4 118 18.5 Pried:ins Odle 1 0 3 0 4 0.6 Turns 4 0 6 0 10 1.6 Odor 8 3 14 2 V 4.3 TOTAL 41 327 33 249 29 638 100.0

TOTAL % 51.3 5.2 39.0 4.5 100.0

Erzebles 7 toss to Mart Oaton. *sign cony. Pus Rico. or mbar.

Phase of Sight Phase of Bight was determined by reading the swo-

ps, briefs: it did not always coincide with the NTS& coded 'phases because it was based upon the period when the problem arose.

The ntost common phases wea. ;ending (26%) and

muck-and -do's (19%) (Table 3 and Figure 2). Al-

though the length of exposure to each phase of instructional Bight is not known, it is obvious that certain phases are ovenepresemed among the crashes in relation to exposure. For exam*, although there are as many takeoffs as landings, crashes were almost 4 times as amnion on loafing as on takeoff.

In touch-and-go landings, the airplane does not come to a complete stop before taking offagrim these 118 crashes included 84 solos and 34 with an instructor (Table 4). The touch-and/Ws on solo constituted

5

23% of all crashes on solo flights. Three of the four inadvertent gear retractions occurred when a dual flight (Le., with an instructor aboard) was about to make a touch-and-go landing.

Crashes on go-arounds numbered 56, of Alai 27 (48%) involved stalls. Five airplanes struck wires on go-grounds. Fifteen crashes on go-around occurred during simulated emeggencies.

Ciscamosaces On the basis of information included in the NTSB

two-page description of the crash, each case was as-

signed to 1 of 14 auguries (Table 5). buss of control so lama% was the most common

type of crash, resulting in 227 crashes, or 36% of the endue series. Although COM3011, they rarely tusked



4.0 C

s

tr. a w

n jcao nneog lea

N

at, em

s O

A D

I C

O

gni C

V r in

co 01

Von

Straconr.

tiorancla



Oates offeantentesed FRO*

Tabb 5. °mantras* of Cavan by Type of nip Instruction] RIO* 1989 and 1931

CIRCUSSTANCIES STUMM SOLO

PRIVATE PILOT

SOLO GOAL

OFR OR CHECK- MI

TOTAL I %

San 42 $ 44 3 94 14.7

Loss of control- Takeoff 35 2 15 3 55 8.0

Lana al control - Landing 148 10 59 12 VI 36.8 Rol alanallon a 7 ao 4 74 11.8

VFR No VAC 5 0 2 0 7 1.1

laddr colkian 6 2 9 1 141 2.8 Tesessrasuing 3 2 8 0 11 1.7

Matterieel kiln 21 0 SS 4 80 12.8

lieueleln train 2 0 2 0 4 MG

Gear lip bray 0 0 G 1 7 1.1

Carkseesor Icing 12 3 10 0 25 3.9 Wire doe a 0 4 1 11 1.7

Mier 4 2 15 0 21 3.4

Umnionvainati _2 -II 3 ..II .4 I&

TOTAL 327 33 209 29 8311 100.0

in sc.:sus injury and only 1 proved fatal. Within this category, the largest subgroups new ran off side of runway (69 cases), noseover (52), hard :awl* (37), and ran off end of runny (19).

Eighteen of the crashes occurred while executing a go-around following an unsuccessful attempt to land. Twenty-one occurred in connection with a simulated emergency. Crosswinds were involved in 109 user of loss of coma on landing (48%) and tailwinds Ia 21

(9%). Loss of central on takeoff resulted in 55 crashes,

9% of the series. Most commonly, pilots ran off the side of the runny (43) or off the end (8). As in the use of loss of control on landing, crosswinds were a

major factor. Stall was the primary event in 94 cases OM. An

instructor was on board when half of the stalls oc-

curred (Fable 6). Most stalls were takeoffideparnue

(39 cams) or approach to landing (28) stalls. Ofthe 39 stalls on takeoffIdeparnsre, 23 occurred on go-arounds, 17 involved crosswinds, and 8, tailwinds. Thirteen stalls ensued from simulated emergent

The 31 fatal stalls accounted for 46% of all fatal crashes and occurred under a variety ofcircumstances (Table 6). An instructor was present on 19 Rights that terminated in fatal stalls. Deans of the 31 fatal stall cases are resented its Section B.

Fuel ssatvatimaz malted in 74 crashes (12%), 29 of which were on clots-county solos. An instruans was present on 24 ofthe 74 flights, including3 instrument training flights and 10 of the 14 flights on which the fuel selector was set on an empty tank. Nineteen crashes occulted when fuel was adequate, but the pilots suisundanood the fuel system or set the fuel selector for the empty tank. Four crashes subsequent to fuel exhaustion terminated in fatal stalls. These we classified is this category, rather than as snail; because the lack offuel was the precipitating problem.

In 51 eases, the fuel was exhausted because the piloc(s) mitstiged the available fuel; Cessna abash, in which the high wings make visuafiation abbe fuel difficult, accounted for 86% of these can versus 60% ofall other avast in the study. Eighteen percent of fuel ia Texas. which Lad only 7% of all the crashes.

*hat soratias iededes bath fsel einanies. ia which dee rice is oar offset sad aria dasedoos eau* imulkiew fad to each she

west



CASs orflintwasiast

Table S. Cinnandances and Oedema at Cradtse Involving OkEs Inaftudianal Rifle, 1989 and 1991

CIROUUSTANCSS NONFATAL

SOLO DUAL* FATAL

SOLO war° TOTAL

SOLO DUAL

Prated% ea or spins

TaksoNdepsnore OA Approach to landing stall

IOther OS TOTAL

0

20

12

J 36

0

15

12

1 25

1

2

3

1 12

2

2

1

14

19

1

72

15

2 47

2

17

13

iti 47

lockein I bionaka mat ttherados laboolothee.

Tao 7. 1ie0hanical Fames Cashes of Inelendlorkal Rights, 1969 and 1991

TYPE OP mama Unsagasined power loes 31

Geer an 13

Stuck *dad vske S

Cartcretor Ttwolliehridtme owed asperalext $ Mocked by Reber sesta* door 2 Debris from ;Mom law bug 2 Fars 2 Hat control loon 1

Fuel Wdsolos 3 Corlakedon 3 Attain 1

CMS ars 3 Olielsevalion 2 CS conleastko 1

Crankshatainippiston 3 agree tea 3 Spa:lugs tarkevom 2 Nosewheal shirmv 2

Propalar anis di 2 Snag in cools* 2 tainieneous la TOTAL' 102

"*3 mass reme eibbtiod peenthey to mechanizei Sohn and in 22 Ole seas another taltin ais piney.



Mechanical failure was the primary rause of 80 crashes (13%). non of which ended in a forced landing with extensive damage to the airplane. In another 22 a there was a mechaniad problem, bur another factor was considered the printery muses The major problems were unexplained loss of power (31 curs). and failure ofthe landing gear (13) (Table 7). Twelve pilots (9 of whom were instructors) mishandled the resulting emergency or even compounded it with an improper response, such as failing to feather tbe propeller ou engine failure or reject a takeoff when the elevator control was binding (Table 8).

Midair collision during these 2 yaus numbered 15 and involved 18 instructional flights, of which 9 were dual instruction and 1 was a checkout Sr a Convair

Omits offsarrasfamel albs

pilot. These midairs are included in the four-year series of midair collisions described above in Part 1

and Section B.

Carburetor king downed 25 flights, 10 with instructors on board and 3 in connection with simulated forced land _age. Thirteen of the sashes (52%) were caused by carburetor icing during cruise. a phase of flight when carburetor icing is not usually anticipated. Three cases occurred on cfratbout, 3 on descent, and 2 during roach- and -go's. Carburetor icing occurred on 6 snidest cross-cosusay Sot In 12 of the 13 an of carburetor icing during cruise, the NTSB determined that the atmospheric conditions were conducive to carburetor icing CM the remaining

Table & Etairradsti Crashes ea Instructors' FUghts.1969 and 1991

anctairrencie flan no PRIVATE DUAL

11PR OR CHIECK-111011 TOTAL

to 14

1

3

2

13

1

1

2

at

a

UR et emerging retetheee

ireproper Room cam ounded ereergeney

TOTAL 15 3 n 3 a

Table 9. crambas Racidting Rom Illmuland Emergencies lasts tunnel Fields, 1989 Ed 1991

CMCMMIITANCES Sea

TakeriUdepanume 4 Approach to lending Other 3

(13) Loos et control - tending

tbatater 4 14wd lending Undershoot 3 Other 6

(21) thschalcel ken 3 Carburetor king 3 wire attire $ Other A TOTAL 49



Omsk. qfhastromel- Itqfket

case, this information was missingfrom the brief). Six

of the imports included the teraperaumdclewpoint relationship.

Wire soden canned ia 11 instances, 5 of which involved a simulated emergency or simulated forced landing during dual instruaion or a checkride. Of the six solo students who collided with wires, 3 were

cassYiall Pascars- Tering collisions were reported 11 times and

resulted in no injuries. The instructional airplane was

moving in 6 cases, and in 5, was struck by another aircraft while standing.

Other era eiremestances Simulated ensergencies. in which a flight instruc-

tor simulated a power loss, electrical failure, or gear system failure, led to 49 crashes- predominantly loss

*femme! on landing and stalls (Table 9). (Because a simulated emergency is usually considered a routine part of Right trod sing. it was not a separate categor, of crash circumstances.) Instructors sometimes killed the engine with the mixture at idle-cutoff, or by shuttingoff the fad supply, followed by difficulty in tannin the failed engine. la 3 instances. all Cessna 152s, the crew did not use carburetor hear. and the aircraft crashed. Three simulations were followed by actual mechanical problems, and the crew could not avoid a crash.

Winds played a role in 232 (36%) of the instructional crashes (Table 10). Crosswinds, is particular. were a problem, contributing to 177 crashes, or 2S% of the series. They appeared to have contributed to 36% of the crashes of student solos and 19% dottier crashes.

In the cases where CTOSSWimis were judged to have been a factor, wind direction usually ranged from 10

degrees to 90 degrees off runway heading, and wind speed from S to 24 knots (Table II). Esther than using a crosswind component than to calculate the crosswind component for each case, the wind speed and angle were evaluated relative to the expel dente of the student pilot, so that in a few cases, wind speeds less than 5 knots or angles greater than 90 degrees were considered to have been a factor in the crash. In most cases, the student appeared to have lacked the cape& ena to make a successful landing, as most crosswinds were within the skill range of a typical pilot.

Weather was typically fair, with DAC conditions in only 2% of cases, but advent demists other than winds contributed to 46 crashes. The predominant facton were clouds or fog in 17 cases, density altitude in 9, and smoke or haze in 6.

Nighttime crashes were rare except. in the case of air carrier pilots undergoing advanced training or checkouts. Of the 5 crashes of aircraft capable of au:tying more than 13 passengers. 4 ratted

between 2100 It and 0345 It. In one such crash at 020016 the instructor had conducted ground training all day and the trainee had been without rest for at least 30 hours.

Peat characteristics The miss reports routinely provide pilot age.

ratings. flight time, and other information specific to the pilot -in- command. who typically is the instructor, except when a trainee is soloing. In some cases of dual 'Instruction, it appeared that a trainee was re- garded by the NTSB as pilot-in-command if be had a

Tails 10. Melds lo Rollidon to domes Crashes of Instructionol Ratis.1999 and 1991

CINCURINTANCES CNOSIWAND TAILWIND ones OVID* TOTAL Statf Lam of carnal Loam of contaskireding Ohm TOTAL

22 So

15 177

10 3

21

sr

1

10 1 18

i

84 140 a 232

'Other *MS ecotone triads gusts, $ wig 4 dciendmits. 1 friatinit and 1 Ant dealt

10



private lkense, owned the aircraft in which he was

receiving instated's% or was undergoing a checkout

or biennial Bight review.

The marten at was 33 for instructors and 35 for pilots with student licenses. Eight percent of ants

lients were younger than 20; 5% of the :indents and

Osten if" 41414 its

9% of the instroassts were age SO orate (Table 12).

Ftfirseven percent of the student pilots and 58% of the Wanton were between 20 and 34 years old.

Fannin contented 15% of the trainees and 5% of the instmacos. The was no appestat relationship be-

nines plot me or tee and the ciscarnstanoes tithe cask

Table 11. Humber at Crashes by Speed and Wind Angie with Runway Heading

inetructional Rite*, 1989 and 1991

WIND ANNA WITH RUNWAY NEADS411 4

WIND SPEEDS Sines)

54 1014 15-10 20a TOTAL 10 - 0 8 4 2 9 12

20 - 2 11 4 1 2 20 30'- 0 10 11 1 1 23 40 - 0 10 0 1 1 18 S0`- 0 8 6 3 2 17 SO' - 0 8 II 4 1 19 70 - 1 9 5 2 0 17

SO - 2 7 5 0 0 14 90'- 3 4 8 2 0 15

100 - 0 8 2 1 0 11

110 - 0 2 0 1 1 4 120' a .9 .2 1 .9 .9 4

TOTAL 8 81 99 18 8 174

Table 12. Age of Students wad Inatruders Omaha at instructional Mrs. 1989 and 1991

AWE STUDENTS"' IF %

INSWN1CPORS 0 %

OTTER TOTAL II

40 28 Si 0 - 4 30 20- 53 13.7 52 21.3 15 120 25- 38 11.3 52 21.3 10 ea 30- 42 13.0 37 15.1 5 84 35- 51 15.8 23 9.4 4 78 40- 39 12.2 17 7.0 10 a 45- 30 9.4 18 7.4 3 51

50- 15 4.7 18 7.4 1 94 55- 12 3.8 5 Li 6 23

GO. II .122 St -I& -I .41 TOTAL 320 100.0 244 100.0 81 825

Thratudee 13 plots whose age was unknown. "Excludes halms with private icons included in 'other,



assist eiastrataadifisks

The principal profession of 88% ofthe instructors was -pike (Table 1*. The profession coded for most trainees was either `business" (40%) or 'student's (33%).

Of the 386 trainees for whom pilot information was provided. 34% shady bad private pilot licenses;

2% also had rauki-engine ratings, as had 84% of the instructors (Tabk 14).

The mean total flight time oldie instructors was 1552 hours (seen -3124 h) (Table 15). Sit MUM' WI* had fewer than 10 hours eight time in the 90 days prior to the crash. Eleven instructors had less than 10

hours experience in the type of aircraft flown at the time tithe and% ('time in types); 4ofthest 11 aircraft stalled.

The median number of flight hours as an instructor was 781 hours (mean.1384 W. Five instructors had less than 5 hours of previous instruaional time. Loss

of control on landing omitted dispmponionately among instructors with fewer than 300 hosts' experience as an instructor.

The median toad flight time of the trainees wet 43 hours (mean -108 h),suoi 19% had less than 25 hours' total flight time when they crashed (Table 15). Eleven pilots with student licenses had more than 200 hours' total time, sugrsting that many of these "students" were not still flying under an instructor's supervision.

Of the 327 crashes of trainees with student licenses, 39 (12%) were known to have occurred on the first, second, or third sob flight (Fable 16).17picallY due to loss of control on landing.

Almost one-fourth of the student solo crashes occurred on cross-country flights. Tice most common circumstances tithe 84 crashes on cross-country solos were fuel starvation (33%) and loss of control on landing (31%).

Pilots flying with student licenses may not legally carry passengers, yet 23 crashes in the series occurred when student pikes were carrying 1 or more passengers. Tea of these 23 clashes (43%) were foal (4 times the overall (stalky rate for the series) and 12 (52%) involved stalk Three of the 1! wire states in the series

involved students who vane flying with passengers.

Table 13. PrInelpal Pidsssloa of Pis la Crabs. Iralrudiana/ Fights, 1909 and 1991

MIMS Pik* 0. 5

Business 4$ 39 .8

Student 38 33.3

Doeinteattist 5.5

P01167b 4

Taschst 1 1.0

Engineer 3 2.8

Other -Li TOTAL 103 100.0

INSTRUCTORS

93

7

% 87.5

93

1 1.0

0

0

3 2.8

0 - A -2.2

112 100.0

Tato meetudas 410 Os tot titan protossien of kusrueter alma was unknown.



Cowen effestractisas lesits

Table 14. Airplane Ratings crashes of Inenzeraf ROW 1989 and 1991

DISTRUCTOSS 0 %

None 256 66.1 0 - Style amine land 123 31.9 40 1E3

ISW* and nit- area land -3 SR 2111 MI

TOTAL I eel 100.0 245 100.0

&cads 6 can rase es or flee awns a waravn.

RATING TRAINEES

Table It To Right Than Clashes of Instructional Fights, 1999 and 1901

NOM TRAINEES INSTRUCTORS

IP 11 i1 x 45 79 19.3 - - 25- 139 36.7 - - 50- 96 25.9 - - 100- 32 SA . -

200- 13 3.4 2 0.8 300- 14 3.7 24 9.9 500- 8 2.1 57 23.6

1000- 2 0.5 62 as 2000- 1 0.3 54 2E3 5000- 1 0.3 26 10.8 10000+ cult _33 II -.7.&

TOTAL 379 100.0 242 100.0

Eakin 17 as Ws Sal ilme or tante Mks Wawa rot *Watt

Table U. 'Type of Solo Right by lane Crashes of Instructional ROW 1909 and 1991

TYPE OF SOLO air LICENSE STUDENT PRIVATE

0 % # % TOTAL

8 % First solo 24 7.3 0 - 24 6.7 Second or laird solo 15 4.6 0 - 15 4.2 Solo oross-country 77 23.6 7 212 84 23.3 Me or on4Pflaled sob En. SE Ei aa EEL Mali

TOTAL 327 100.0 33 100.0 360 100.0



Osier eissersesitad

Mot petformance Specific pilot-related performance factors were in-

feted for 86% of crashes (Table 17). The most common pilot fa lost controV noted in half of the crashes, is detailed in Table 18. Poor crosswind correction was apparent in 79 crashes, a bounced landing in 48, redder misuse in 42, and poor handling of surface wind or turbulence in 40.

Thirteen :tridents crashed when they were lost: 8 who were lost had run out of fiat Six of the kat student had not filed flight plans.

Although it is standard and recommended practice to file a Right plan prior to a trip. 34 oftbe 84 trainees who crashed on solo cross-country tips (40%) bad not done so.

butratar performance In one-third of the cases, the NTS13 investipzor

cited instynceor-related factors as contributory to the crash. The most commonly cited factors were inad- equatesupervision (77 cases), delayed remedial action (34), mishandled simulated emergency (20), and inadvertent sad/spin (15) (fable 19).

Os the basis of review of the two-page NTSB briefs. it appeared that many of the solos reflected inadequate training, primarily is evaluation of cross- wkia and weather (69 cases), roach- and-go lanctings (48), and teen's* front bounced kstrusgs (35). In 54 cases, the instructor let the student get so far into an operational problem that the instructor could not recover, or even compounded the problem. Nineteen

Table 17. Plot Parforaustas fraeten by Typo of Right Crashes of Instactonal Rifle, 1989 and 1991

PLOT PERPOIIIIANIZ

8TiIO1itT SOLO

POPIATE MOT SOLO

OUAL AM 011 MC*

RIXE TOTAL 9 It

Poor penlight Ow 2 1 3 0 8 0.9 Impactor pritight 7 0 12 3 22 3.5 lekbidgecl bag 2 0 1 0 3 0.5 Used war land% tor Imp 0 0 2 0 2 0.3 Oil not bate pop 0 0 3 0 3 0.5 LIShinIscl Lips 4 0 3 0 7 1.1 kinused raft% 0 1 5 1 7 1.1 19.4udged fuel maths 23 4 6 0 33 8.0 Wofriffad lusi .0181n 0 0 1 0 1 0.2 Full wink" 821 enema 3 1 7 2 13 2.0 tigudged water 11 2 9 0 22 3.5 Poor hording d dinky altitude

1 0 1 0 2 0.3

1361 lad 11 1 0 0 12 1.9 Wed hone on =bolt 0 0 2 0 2 0.3 Alcohol 2 0 0 0 2 0.3 Fain to us Ind avoid 6 3 12 2 23 3.6 kaulblent thuds 3 0 5 0 3 1.3 PAsund bathos 5 0 1 0 8 1.0 Did not use arb hsat 11 2 8 0 21 3.3 Lost awe 200 15 92 14 321 50.2 02%. 9 2 15 1 21 42 Plot a factor 22 1 II I ..99 ILO

249 TOTAL 327 13 29 NB 100.0

Imo Tabie 13 for dna 14



Crash* efffsatestionsf Patio/

Talge it Rot Ens Leading to Loss at Control on Takeoff or Landna Crashes of InstructIonal Fights 1989 and 1991

STUDENT SOW

PANAY* PLOT SOLO

DUAL IIPR OR CHEM

R1011

Landed brg 4 0 2 0 Poeticised 11 0 0 a Bounced 40 1 7 1

Balloonsd 5 0 1 0 Appowsh eland to test Nan* speed to slow

e 5

0 0

2 1

0 0

Flared too high 12 0 7 0 Idisiudged landing surface 6 2 5 0 Rudder too RIR or misused 31 1 10 1

Pocecrowaselnd correction 64 4 10 4 lisjudged approach prone 17 2 16 5 Mishwried surface wecilturbularce 19 6 18 2 Igehandled wing 1p vortices 3 1 1 0 lateludged shodfnanow runway 4 0 3 0 liehandled tannic= runway conditions

7 2 5 0

Mused bakes 2 0 3 0 Unfamiliar with eitraft 0 0 5 2 Otter IR Si _I _3 TOTAL 246 19 99 11

IN 321 pilots web at tons snot, a second rot as moods* tot 63.

Table it Fight betrucbw Factors Cited by NTS8 Crashes rod Inetruclional Fights,1999 and 1981 FACTOR Iftetlequate saltation 77 353 Delayed remedial action 34 153 Mishandkal *nutted emergency 20 93 Inertattent stallApin 15 7.0 Direclional coned not maintained 9 4.2 A D D lanning 3 3.7 lane Mal twining 6 a Inadequale pre licit 5 2.3 Did not use cob heat 5 2.3 Airspeed not maintained 4 1.9 Used gear handle for Rap 2 OA Caw it JAR TOTAL 215 100.0

Table tocludes 423 easels Otwo4irds of d ems) ht stet no boscauctor Inn wan and.

15

TOTAL

6 11 49 6

10 6

19 13 43 82 40 45 5 7

14

5 7 a

384



Ossilet gfIscosssassmal" 'reit

crashes occurred because the CF/ did not anticipate a student's hasty action. Ten crashes occarted when the CPIs simulated forced landings in areas that did not afford safe landing places. when the simulated esna- gentles turned into real ones.

Abeam datarteektles Most ofthe aircraft (94%) were single-ensise uaia-

ing planes such as. Cessna 150s. 152s. and 172s and Piper Cherokees; twin-engine planes constituted 6% of the series. Twins were over-involved in crashes, due to mechanical Mum (13% of mechanical failures were in Milk compared with 5% of crashes from other muses). This is probably because Ion of control on takeoff or landing, which accounted for 44%of all crashes. and an even pester proportion of crashes involving inexpetienced students. was sands less common in twins, occurring in only 101439 cases (26%).

Nine of the 39 aashesoftwins (23%) resulted from simulated emergencies. in comnstto 7% tithe cashes of single-engine aircraft.

Tailwind aircraft (61) were involved in 9% of an crashes. 17% of nostovers. and 79% of ground loops. The 86 aircraft with retractable landing gear constituted only 14% of all aircraft but 26% of the crashes due to mechanical failure.

Crush ostemen The likelihood oft crash having fatal results varied

dramatically with the circumstances (Table 20). The kinds of crashes most lately to be fatal were VPR into 114C (crashes in instrument meteorological condi- dons when flying under visual fight rules) (71% fatal). midair collisions (44%) and stalk (33%). In connate. less than 1% of the crashes hoar loss of control on landing were fatal.

The total number of ewes in which 1 or more pew:insmete killed was 67, or 11% °fan crashes. The likelihood of being fiscal is related to the number of occupants of the aircraft; in order to avoid this bias, the Showing fatality rate comparisons are for the

Table 20. Injury have *y by astuentratn8 of Crab by Bond ROW nes arc 1991

ORCtiliSTANCES NONE INJURY

MINOR 919024119 FATAL TOTAL

S % FATAL

1 96 19 8 31 94 33.0 Loss of caravel - 48 6 0 1 55 1.8 TaluoN

Los col oonlral - 200 21 5 1 227 0.4 0

Rol starves on 45 19 6 4 74 5.4 VFR it INC 0 2 0 5 7 71.4 Radar Slim v 7 1 2 8 18 44.4 Taxlinpilandig

fare 9 11

44 0

19 0

13 0 4

11

80 5.0

Mourdien terrain 2 1 0 1 4 25.0 GM up lending 7 0 0 0 7 - Carburetor icing 20 4 1 0 25 - Wirt MN 4 4 1 2 11 182 Other 6 $ 4 6 at 2es Undorkud _2 II .2 .4 A 100.0 TOTAL 432 99 40 67 WS 10.5

I6

ADA VI/S190 PDF compression, OCR, web optimization using a watermarked evaluation copy of CVISION PDFCompressor


Cnsabar ifhistriaressomf. /Wee

TANG 21. Fetters Waist; Su Plat Survival Crashes el Insrucityral Fights, 1969 and 1991

FACTOR PLOTS

INVOLVE? PILOTS KILLED

CASE FATALITY RATE a PANS=

Wte lbw NC 16 4 40.0 <041 VIC 1128 54 8.8

Post= It Fie Yes 33 11 33.3 <0.01 No 605 49 8.1

Number al Engirt; Two 39 6 15.4 <0.20

599 54 9.0

Shoulder°1" Hamm Used No 160 24 15.0 <0.01 Yes 402 24 s.o

last rosy eyelet groups due to minima dot

piot-in-command. Sixty pilots-in-command (9.4%) were kill. Table 21 shows that the pilot was mote likely to be killed iE

the weather was !MC (40% Fatality vs 9% in VFK crashes. p<0.01)

*poem:ash fire occurred (33% vs. 8% if no fire. pc0.0i)

the aircraft was a twin (15% vs. 9% in single- engine plates, p<0.20) the pilot was not restrained with a shoulder harness (15% vs 6% of those restrained p<0.01).

Information on availability andior use of shoulder restraints was mining for two-thirds of cases. Nine. parent of trainees and 16% of instrutron known to have a shoulder restraint available were not wearing it at the time of the crash

One sash was precipitated when the instructor opened the door to retrieve his Capping sembelt, corm:luting to the manes loss of control on takeoff

Reflecting the fact that aircraft damage is 1 of die criteria for NTSB investigation of a crash. 83% ofdte aircraft sustained serious damage. and an additional

17

15% were destroyed. Thee wen 32 post-crash fires. almost half of which followed either stalls (3) or mechanical failures (7).

DISCUSSION

Until now, there has been little knowledge about the aim:mamas under which crashes of instructional flights oecur, except for intbsidual ease descriptions. This research provides infonnatioa ota fatten related to crashes of instructio' nal flights and shoalti be useful for improving; eying and decision- making skills, not onlyof sockets and instructors, but also of Berne pilots at all levek.

Student solos in this series of crashes illustrated several types of problems that gams* are not recog- nized, including carrying passengen (in violation of federal alit tegolofions), roisoodentstass the 4 namies of tonch-and io ion:Sags. inadequate planing and peetlight fuel check, and Faure m 6k tight plans

Even on a dual flight. most of the piloting error may have been made by the mime, who typically would have been at the controls; the insensate, however. is expected to closely monitor the student and to be able to awn a crash.



Crapies irhastreassed iflees

The involvement of flight instructors in certain types of training-flight crashes was surprising. Direct actions by fight instructors included retracting the landing gear prior to a touch-and-go landing, and initiating simulated emergencies at low altitude. Of- ten, there was evidence that the instructor had allowed the student to develop a problem past the point where recovery was possible, which occurred in many cases of loss of control on landing. Lack of instructor awareness of an impending problem was exemplified by the fuel starvation crashes due to failure to monitor the fuel supply while in flight, or failure to realize that the fuel selector was set to an empty tank

In the majority of flights that terminated in midair collisions or fatal stalls, an instructor was in the airplane. These 2 ategories of cashes comprised the majority of all fatal crashes. The involvement of instructors in crashes with such serious consequences (to themselves, their students, and people in other aircraft) underscores the need to ensure that instructors are trained as to the importance of their unique responsibilities. Flight instructors are called upon to split their attention between the teaching function - in the difficult raffia of a noisy cockpit - and the safety function. The latter demands high awareness, in terms of other air traffic, Arc communications, and airport surface winds, along with an awareness that the mince may respond in a surprising and hazardous way. Therefore, a greater focus on training instructors and improving the quality and results of their instruction may be an appropriate outcome of this research. Specific recommendations for disaerni-

flung the results ache research., especially to instructors, are set forth in Section C.

Solos 1-2149 8387

3-22-89 4131

3-2449 #902

CONCLUSION

For auto pilots, their early aviation knowledge and judgment skills are a product of both the flight training program and their instructors skills and teaching ability. The skills and knowledge that students take away from this training are what they build on as they continue to accrue flying experience. Thus, a pilot's basic flight training influences the safety of his or her entire flying career. Improvements in flight instnaction, therefore, should not only reduce crashes of training flights, but also enhance the safety of licensed pilots.

E. imams OF INDIVIDUAL CRASHES

The following summaries are based upon the narrative and other information in the 2-page NTSB briefs.

Topics were selected because of their apparent importance. Cases were chosen to be illustrative of the major problems encountered in instructional crashes.

The descriptions provide examples, rather than attempting to describe every case. with 2 exceptions: Because of their importsnce, all fatal stalk and all midair collisions are described.

Midair collisions from all 4 years, 1989-1992, are included, rather than being limited to the years 1989 and 1991. as in the case of the other categories. This was done to obtain an adequate number of collisions of this important type.

FATAL STALLS (31 cases from 1989 and 1991 instructional crashes)

33-hour student plot carrying 2 passenger: in C-172 fifted off1320-foot cow-panne strip below stall speed; 2 Idled and 1 seriously injured. 36-hour suntan pilot practicing stalls alone in C-154% made fanl descent into snow-covered terrain. S0 -hour student pilot carrying passenger in PA-38-112 flying low over river, pulled up sharply. Both occupants killed when right wing stalled and aircraft dived into river.



Fatal Stalls (csatlaned)

3-2849 #2082

4-15-89 #757

5-4-89 #1098

9.649 #2237

9-1949 #1770

10-14-89 #2040

5-5-91 #1856

5-26-91 #194

7-5-91 #1330

Dealt

4-649 *1303

4-1249 #2103

5-9-89 #1304

5-22-89 #402

6-25-89 #2369

7-2-89 #2101

8 -25-89 #2173

1-2-91 #141

2-2-91 #1957

2-25-91 #432

3-27-91 #981

5-18-91 *1735

Oasis. agfhwersetical /Was

46-hour student pilot solo in C-150 in pattern made ffight pith cask in altitude and Mara Math mud over sad tolled with train. Pilot killed. 33-hourstudent in C-I52 wed in approach-to-lantliogstal while attempting 5-nuns fee spacing khind another anctak Both 120 -lone student and pare in Atm= 11AC !died in takeollidepartute sal after low passes mat strip, 934our sxackat, 93 hours MIM, Lied in AA-56 matt out 0/hue-to-final turn on third attempt to had in 18-knot atuswind 34-hour student pilot with 16 WM doing touch-and-go's in es130 with passenger; stalled and cashed on takeoff phase. both killed. 30-hour student OR first supervised solo is C-152 left pattern lie second landing, staled into field after cirding house and flying low over golf mums 97-hour steam plot with 71 hours WM ((PA-38-112) lulled in sat NITSB &unmated stall as intentional. 30-hoursmient with 23 houssPA-23-140,btated his home, tolled inverted front steep bank fatal crash.

228-hour student pint, with 129 Louts in his Champion 7AC, carrying passenger, made inadvertent sail into imam

1500-hour of living duos in C-I50 to a CFrappficant Witnesses sitraitaafi pitch up and then one dawn. Both plots inflict. 1592-hour CFI, hours MIM unknown, giving seaplane mining in Lake LA 4200 to another pilot Hoch clew lad in fist spin. 4800-hour Ca 600 WM. giving an co:al:nation training in P444 -180 to two other piles; radar data Maned sallispin into sea. Part 135 Raining by 4000-bourxr? with 30MM. in DC-3. Two pilots and Oat passeapt bled is stallIspin. Private plot praaicing in C-152 stalled in nubtdeace and windsheu in the traffic panern; pilot and paw killed Eaperienced CFI doing tonch-and-go's with student in C-150. One killed and 1 sainuslyinjwcd in stalllspin in fist gust 1024-MtnCFI gavesimidatedengine failure in C-15Z audentseddenly pulled nose up into stet During recovery aircraft hit inountatatan tania. Student bled; CFI seam* lajotted. 2517-hour CR with 420 hours MM staffed attempting get-around in C-172 from snowy. 1500-foot runway. Time on-board; student killed. 130-hour private pilot with 13 hours Itift4 started takeoff phase of touch-and-go with 400 feet runway ntaminins 3 tailed in akeoledeparture sat C-172. 534-how CFI and student flying C-152 killed is apprtnde- to.Ldiag stet on final approach. After purchase can& 58P, owner-pilot hired CFI with 1000 bows WM to fly with him aril owner proficient Both Wed in spin into terrain.

2401-hour CFI with 1 hour WM giving training to wife of owner of PA- 24150. Both Idled in accelerated sat

19



Credit eflassrsusissui agars

Fatal Stalls (asatkaued)

7-2841 #2054

8-7-91 #2085

8-20-91 #1505

9-24-91 #2130

11-16-91 #2010

12-21-91 #2079

12-28-91 #2121

1505-hourCM with 2 private pilots, skies mountain, flying training in PA-28-181,olnerved in climbing turn, sadden dements in mountain capon. All 3 Idled. 4502-hour CFI giving dual le PA-28-140, started spin ton fret 2 killed crashing into terrain. Sc hard clouds 2150 tan adore trade site may haw been a factor.

1575-hour CFI in C-150 tying to are 9-hour student of far of stalk both killed in spin into gonad. 2308 -boor at giving dual in C-150 could not recover from inadvertent stall/spin: both hand grips broken off left yoke and left grip broken off right yoke. Two deaths. 250 -hour private pilot with 167 MIM receiving dual in C-310 for multi-engine check ride..

Aircraft was observed below 3500 fee AGI.graffmg with gear and Baps enenderk power added. then redneck aircraft snapped into spin, recovery intposside. Two Earl. 540-hour CFI with 168 hours MIM giving intiodostoty Inns mean aging Imply st low afloat then diving into pound. Both occupants killed. al certified 7 months easier. Night commut er training. Beech 1900, instructor pilot disabled attitude imitator, then in addition, on procedure urn sinsulated engine faun. Captain trainee felt disoriented, asked IP to take over amuck IP refused; loss of control into ocean. Three fatal.

SIMULATED EMERGENCIES (Examples from 49 instroalortal crashes in 1989 attd 1991. listed by crash category)

Single engine aircraft Stalls: 1-16-91 #53

2-25-91 #259

7-6-91 #533

8-16-91 #1156

9-20-91 #870

12-28-91 #1770

Carburetor Ica 1-17-91 #59

312-hour at with 1I boon WM reduced power to idle on downwind: when attempting go-around PA-28-140 stalled into tees. Five-knot tailwind may have conflate& aircraft was ova poss weight 21-year oklissmactor, 3871T and 346 WM. simulated engine failure on dimbout racking in stall-spin in PA -28- 161. Sir-knot tailnind may have compounded situation. 6340-hour CFI with 30M1M. *pulled the engine" in BE-77 to simulate tangency; akaaftcodd not climb in high density altitude and nettled into vervain.

685-hour al with 96 hours WW1 stalled demonstrating nun back to runway in simulated emagemcy, Champ 7GCAA. Very apaienced CFI la 2-bout student in G112 r 'n simulated farad landing naivety 150 -200 feet AGL; student pulled up now and staled. 500-hour CFI retracted flaps on go-around from ponr-off vattach; C-172140 stalled into pound and ameateled.

On simtdated forced landing practice in 0152, age did not use card beaU cab ice prevented dimbont and aircraft overturned landing on uneven terrain.

Wire strika 5-1849 #369 C-152 kit wires on go-around following sinudated forced landing.

20



Chain elstnetised Pate

Simulated Exangeneks koadareere

4-13-91 01465 Ptacten'ag ground reference Mamma in C-177RG. 530-hour CFI with 55 bourn M/M gave

simulated fotcedlanding.WhenCR initiated gwerountL aintaft hitmrsres lines sad crashed into Sell.

Lon of control at Staab 9-11-91 #789 426-liner CFI in C-152 simulated engine faikre total power loss occurred and akaafi made

successful forced landiag. After engine stoned. CFI attempted takeoff ham unsuitable terrain and assbed.

Lon of osattra as badin: 4-9-89 *73 292-hour CR in C-152 shut off fuel to simulate engine kin could not Vegan. Abash landed

short of road. het fen 3-22-91 0988 CPI giving 8FR in CM puled minute to kik median approach. Ownetipaot tried to add

throttle to snake runway; that hit near runways and ad ova.

Maki engine aircraft Stalls:

114691 #2010 250-hour private paot with 167 WM receiving dual in 0-310 for taultienene deck tide. Akaaft was observed below 3500 feet AGL gliding with rat and Raps extends& power added. then reduce* airaaft snapped ino spin neeteryinepossark. The fresl.

12 -28- 9192121 Night commutes training, leech 1900. instrector pilot disabled attitude indicator. then in station as procedure a simulated engine fekre. Captain asinee Mt disotiented. asked IP to take ova contra; IP refuse& Ion of control into ocean. Three fatal.

Lon of amattol on Imam 2-4-89 *360 AM than field takeoff in FA-44-180, CFI idled left engine. Stein* retarded right throttle.

lowered Baps; CFI rook round. Bated too h#. collapsed landing gear on heed landing.

Fuel stanadont 104-91 #2167 9500 hour CR with 16 hours WM kit fuel seknot in C-337 on wrong at following simulated

engine-out lank* ran out of gas on takeoff. made Mad landing in cornfield near airport.

Maebenical Mkte 1-23-89 #158 785 hour CR with 46 heats Mild simulated power Ion tisk engine on approach pulled gear

circuit let Crew could not restate gen system aircraft would not paten single 0101* go-seound because right mop not feathered.

Otbon 3-2449 1531 CFI with 130 hours M/M shut down kfi engine in PA-34-200. told student to max single engin' e

landing on 2600-foot diet runway. Student high and fiat on find; on attempted go- around. aircraft would not climb en 1 engine.

21



tlatiss tekotroetkatel Atha

Crosswinds 03-24-89 9731

04-1949 *725

05-2659 #592

07-2949 *1806

02-26-91 #26

11-26-91 #1751

LOSS OP CONTROL IN AIRPORT SURFACE WINDS (Examples from 232 instructional crashes from 1989 and 1991)

1200 -bow CFI giving dual in a C-172, tried to tale control of sicced after =dent drifted off centerline crosswind. Snidest mould not relinquish controk *bath collided with a hater. 43 -hour student piodyingC-150, laying to hams runway 22, winds 270 at 10 Imam attempted go-around in crosswind, strata onto runway and hit a hangar. 20bour student in C-152, on third solo, wend off left tide tinning 32 and hit a sign. Winds 260 at 9 biota 30 -hour student with 17 hours MIM on short anal for runway 27 in C-In when he noticed a crosswind (190 at 13 knots). Trying to cope with crosswind. student let high sink rate develop; skald hit nonewhed ant and poquised. 27 -hour student dons touch-and-gots on fin_ solo, using runway27. Winds 230 m6 knots. On third takeolfround toll, aircraft yawed left. veered off runway, crossed a clearway and stopped in a plowed S&L 22 -hour student attempting takeoff in 0152 on runway 7. winds 030 at 8 knots. Student lost control, recanted throttle, misused rudder and brakes. ran off ninny into ditch and nosed over.

Tailwinds 07-0749 #497 37-hour student doing tondo -ad's on runway 34 in C-15% wind 180 at 5 knots and density

oldish 6500 feet S. Student sensed aircraft not dubbin properly not providing id power; nailed in torn avoiding obstacle.

08-27-89 #1361 39-hour student voids 2 hours in C-172 doing much vad-go's on nanny 17, winds 340 at 4 knots. Math bounced down ninny, porpoised and wand of

12-03-89 #1616 28-boot student landing on runny 22, winds 360 at 15 knots. Ancrafr bounced, drifted off tummy. Sew over cEsch, said and nosed over.

03. 15-91 #91 56-hour student with 27 hours MIM. landed C-152 on runway 2, winds 210 at 6 knots. Pilot landed aotmally, but puled back on yoke and shad ballooned, poipoited, and collapsed nose

Pat 08-21-91 #2143 34-hour student in C 152nsedtoofastonfenaimrunwayg ,winds300at5lmaa.Studenttaied

to extend flare to dissipate mimeo& aircraft bounced and landed had.

Solos 1 -22-89 #234 1-29-89 t537

2-10-89 *136

4-149 #530

TOUCH-AND GO'S (Examples from 118 inaruction crashes in 1989 and 1991)

16-hour student in FA-28-180 km control on takeoff phase, trying to attract the Saps. C-150 and C-172 both doing touch-and-goes at uncontrolled airport. C-150 landed on top of c4n. Dudack:An phase: left btakelacked for 41-hour student eying FA-38-112. Airadveeted off runway and nose and right gem collapsed. 40 -hour student in Beech C24R nude very bind landing, collapsed all three larding gear.



Towde-and-Gota fonettlanad)

4-22-89 #292 7-3-89 #1599 8-27-89 #1361

2-26-91 026

3-31-91 #1349 8-24-91 #1581

10-28-91 #1790

Dash 1-749 #226

1-2949 #98

2-8-89 #11

7-27-89 #2167

8-841 #774 12-23-91 #2119

Fatah 1-1449 #309

7-2349 12109

7-2449 *1043

9-12-91 #2199

Nestatals 5-12-89 *596

7-8-89 11604

Ossofess ebtainsebed AWN*

I8 -hour student in C-150 landed eon. boasted onto runway, noted over.

12-hour student practicing in C-I50 bounced, pound looped.

39-hour student with 2 hours is Gin bounced down the runny attempting landing in slight

tailwind. Met 3rd touchdown, pop hit runway and airaaft mend off 27-hour student on firstsob veered left offrunway 27 on groomed roll of third takeoff. 'Winds 230 as 06. 19-hour modem on second solo in C-172 stalled attempting go-arms& 15-hour student flyingC-150 lost control on landing, tried gemeund with fidl Alps and cubbeat at 17-hour smdent on first solo in C-152, attempting Jana. hammed. waned into bard landing.

CFI ring dual in C-172; slight tailwind and snowy runway. Aircraft drifted left off runway and nosed over in mow. Private Sot receiving dual in Commander 11210 banked steeply so land on runway centrum hit a wingtip and bent wing. 2215-bnur CFI with 165 bows WM giving rental duckpin to private pilot in PA-15-150. On takeoff phase, CFI could tot correct swerve to W. Following preflight bespeak& student roll at that aircraft needed fuck CR dim* not and aircraft ran out of gas doing touch -wad -goss.

2000-hour CH with 1 bow WM giving deal in PA-31-350. retracted gear instead of Sp. Student receiving dual in C-152 landed bag on touch-and-p, ran off runway end vying to stop.

FUEL STARVATION (Examples from 74 instructional crashes from 1989 and 1991)

268 -hour student =trying passenger in PA-24-180 took off with fuel selector on empty tank. Aircraft carded with ground on attempt to cum back to Open Not Sod. 103 -hour student in Beech 77 doing solo aces country radioed that engine losing power. Not sounded panic-stricken. Airaaft crashed inns pond; pant kilkd. Leh wing fuel carom unfastened and lying on wing in Salm fuel siphoned out. 30-hour student practicing touch-and-go's in C-152 entered steep spirting descent after takeoff and crashed. No usable fad left in tanks; plot Ia. 30 -bout student practicing touch-and-go's in PA-38-112 dove into ground attempting to turn back to runway after engine quit out of fiat. Not Wed.

502-hour CFI with 20 MIM giving private pilot checkout in C-1-21% after an hour of airwork engine pi, fuel selector on empty tank. Aircraft crashed in (steed landing. CFI did not notice that pilot did not twitch at 52-hour student carrying passenger in C-150 ran out of fuel. landed in barley field.

23

ADV104$390



Osaka ellegersaisi Mike

Feel Stantioa (coldnuod)

7-27-89 #2167

8-2549 42081

9-249 #1856

11-1449 #1327

1-6-91 #289

6-18-91 #1024

Following psth%bt inspection, student told CFI that aircraft needed fuel; CFI thought not and aircraft tan out of gas doing touch-and-go's. 20-hour student in PA-38-112 ran out of gas on final approach doing touch-and-go's landed ton. Engine had run 4.6 hours without refueling. 47-hour student with 29 hours MIM, flying C-150, got lost they returning to airport with rata problems and not tefueling before setting out again. Afar 4.1 bouts, engine quit due to fuel a/imitate. 30- hour student on solo cross country in C-150 got lost; wane* was guided by FSS bath to departure aispon, but tan out °fps and landed in an oil field pipe yard. 4900-hour CFI with 4300 Wan 0150, atteutpsed introductory Bight in C-150 with empty tanks. Aircraft crashed 114 mile from aid of takeoff runway. 44-boar student with 30 hosts M/M tan out *fps doing touch-and-go's in C-150; on clitabout ran out of fuel and entered accelemad stall on turnback to runway.

FLIGHT INSTRUCTOR PERFORMANCE (Examples from 278 dual instructional crashes from 1989 and 1991)

La pubis develop too fan 7-949 #593 9100 -hoer CFI with 10 WM gave :cadent simulated engine failure in PA-38-112. Student

performed emergency pea :duns but MC too high on final to land on nanny. CF/ then wok mead and staled the aka& attempting another 360-degree a

8-22-89 #1988 645 -bout CFI with 100 M/M ad as correct fin approach in Bella &ECM. Tminee saw power lists on ton final sad raised nom high sink rate developed and abash hit a parked truck short of the runway. CFI mid he tlid not apply power because of "tank pitch-up tendency.'

9-29-89 #1319 Two CFI:giving inansaion toes& other in short fieldWasp, in C-182. One CFI noticed that high detceet rate had developed on short Sat said other CFI to apply power. Flying pilot did not r e s p o n d and aircraft landed bard. PIC CFI said be could Ian prevented crash int Inti not delayed taking action

8-12-91 #929 1039 -bout C:FI with 374 hours M/M doing touchend-go's in PA-38-112 with nuclear in 10-knot left ansreistd. Os take's& aitaaft drifted Mk CFI told student to apply tight rudder, but left daft continued. CFI applied tight rudder but aircraft pivotal onto pass.

9-1-91 #2242 2062-hour a! with 1450 WM givingdeal in C-152. Student loaded hard and aircraft bounced about 20 feet into air. CFI could not recover from the bounce before the alt craft

9-20-91 .870 5175-howCFI widt 2000 WM gave 2-hour student sintuhred forced landing in C-I n. CFI bad student start seaway 150-200 feet AGL and student stalled the Sank CFI could not recover.

12-7-91 #913 1940-hour CFI with 6 M/M gave student in PA-15 anageacy landing on sod strip; 30-degree aoatvind at 14 knots. &fon CFI could recover, sloth veered off twiny and nosed over.

12-28-91 #2121 Might commuter waiting, Bach 1900, instractx pea Stabled attitude indicator, then in addition. on procedure a, simulated engine failure. Captain trainee felt &oriented, asked IP to take over amuck IP refuse* loss of control into an Three fatal.

Did met itaticipate improper Lan salsa by statient: 3.12-89 #263 518-hour private pdot with 57 hours M/M teceivieqg instruction in PA-34-220T. inadvettently

renamed par while lantlieq.

26



asks eirlannangeilfreidt

Ft leis bastraaor Porgrmance (amakstad)

4249 *120 591-hour CFI with 53 WM teaching in PA-30-112. 4-bour sans tuned takeoff tell with kfi aileron into 13 -knot left crosswind bat relaxed aileron prams. When CFI told student to Satan left thank studcin applied 'udder and brae; abash swayed left and CH could not recover.

41049 0436 5594tour CFI wadi 94 /4/1.4 nultinglowappoachwidt snidest in C-152. CFI insulated sudent to add powerto go around. Instead. thestmicartaised the airadinote ad the righavingdsopped. leaking in k atoll.

3-19-91 0429 315-hour CFI tidal; ?NM pve student in C-150 a simulated engine failure by palate dna& to idk. Student responded by polkas sinus to idle comfit which C1 did not notice until student applied thank to p atound. Forced landing.

STUDENT PILOT CARRYING PASSENGER (Examples from 23 instructional crashes from 1989 and 1991)

1-1449 0309 268-boor student in PA-24-180 took off with fed schwa on aapty tank Pays win* injure* piket bled.

2-12-89 0358 26-hour atidenttlyingC450with passenger (who dimmed tobeaskep) made forced lanitogafier wire mike. Pilot first said he bit a bitth kat said he hit a pradatist finally he said airceak hit "mg..

3-549 *400 23-hour student lying PA-28 with passage crashed into yea afar departing dam in fog, deb* at 0400 has Passenger Idle* pilot lad blood alcohol of .085%.

5-6-89 0204 61 -how student with 29 ?MK lying with another sedate in C-150. stalled ea ea final. 9-19-89 *1770 34-how student pilot with 16 SUPA thing soudtand-ges is C-150 with pastengar; staled and

andsd on takeoff phase. both Med. 9-29-91 01610 137-hour student pilot in Amen 7W stall-mush while WO., passenger spot alheators. 7-31-91 *2253 SnadmavidtpausagerinC-150lasdedformeal apneas abstrip. Restanastdoank indention

canal on Seal in Sght aosswind and clothed.

MIDAIR COLLISIONS 00 cases; 38 instructional aircraft;1989-1992

Fatah 8-2949 01846 Private pilot in C-152 receiving Stases dual in uncontrolled airspace as =met diabad out

at 700-8004unon pathconnteisgwidtBE135aktisinglevel az 1600 isanslathertangcrtrady into an Fatal to four.

9-349 02230 38-hour student in pattern doing touch-and-Ps in C-152 hit an base kg by istkopter crabby oat for posiabwagat aessbylhowitaL Asylums= had dosed 46 sainawatastrec meow airman was bend giving talk ashiscaits. Three kSai.

9-2449 #1661 27267-boas at giving dow Blebs Awl in C472. at 800 feet sad Radar plan medal another akar& attest cmaineC-172 flight path. Rewiring spar dein Sad in &Pt. The markfiont otharairash found on C-172 wing tap. Two am; killed in C-172 crack Other airakft. a C-152 operated by a 24-hour student. stunted to open. Stades said he could not real hitting anything



Oman citiastnesiseeindar

S Collisions (cemdamod)

11-5-89 .1997 1600-hour CFI with 200 MIM doing touch- Lid -go's with student in C-172 at eniconuolled

airport. runway It Win& 190105 lees. A Beech C90 flown by 2 ATPs was deeming runway 6 on an IFIthight plan. Both crews made ratio calls but aircraft collided over runways aida 4 people

killed. 2-3-90 .895 147 -hour pirate pilot with 49 hours MIM conchsding cross country in PA-28-161, made left

be While PA-23 on final. AA-1A an inside on doer & bate, anise final slightly. and prop and nosewheel hit S ofPA-28. Both &bereft broadcasting positions ma pilot killed, and AA-

IA pilot could not real deals of ocantence. 2.6-90 .1629 4300-hour al giving anal in Beech 95-55, doing touch-and-go's with right traffic for noway

31R. Beech twin made 30-degree constant tom with no rollout dishing our from ansewind and tuning downwind 0182, which had repotted NW imitwfiy. end been cleared to enter left downwind, repented ads side of sigma and was also dared fee right downwind; the 2 aircraft collided 2 milts north ofakpom 3 kilted C-182 did not enter downwind abeam mawaymidpoint and C-182 pilot tiered positive for marijuana.

3-90-90 ATCwoekkag trarscon nordrand south rturnaysoal slifferentratliolitesprencim 1135-howCFI giving dual in 0152, planned touch-and-go's on south runway, where 2 other Camas were alremlyiapateent Abannenowitegairaali radioed= nerds freguencynad wasclearod to descend themes proem for banner drop. Alt informed banner aircraft of Ceuta: on downwind. and the latter, of the banner Scree When the C-152 tuned downwind, it and the banner towing air raft collided 3 fatal.

7-1-90 .982 Two nevdylicassed private pilots with 118 hoursWM were flyers PA-38-1 Ireton& same cross country route. After an intmeneclime stopover, the 2 flew fommtiem the tram aircraft bit the boding airerafr while saaneuvaing. Both pilots lulled.

7-23-90 .1826 CFI pang nutrarnene dual in ma was circling out on cense of 222 degrees, at 80 Lana. A PA-60 was asking at 2100 .`at and at 165 !mots, cosine 258 depots. The PA-60 converged on the PA-28 from the right man the PA-28 canard an the PA-60 from is loan left forward an= The abash collided at 2100 feet sad all 3 pilots were killed. Neither Sash had obtained Ara radar assistance.

8-19-90 42190 271-hour CFI with 114 hours MM giving dual Is C-172. which was flying a course parallel va a PA-28RT-201, which was costfipired for sky* Spat with par and flaps down. Cessna lying faster and hit Piper with left wing, seeming outhand portion of Piper right wing. Both thrash flying toward sum Csalon killed 4. (No ATC information)

94-90 #1902 Taylotanft Gown by 100-hour mom pilot collided head-on with another aircraft. Both pilots inMclembyNarianal %karts m500knAGILchechingon Meats athuntingcampc Bockplan

2-3-91 094 Cessna 182 climbing out with 4 skydivers converged lateraily with approaching instructional PA-28-140 abort 1.5 miles from airport Aircraft tweed sad crashed 7 people killed.

2-13-91 9905 17,300- IsoarCF3givingd ug in Pius 5-2Acollided witb Waiver which had lifted offa pad near runway and turned to depart over same rummy. Fitts new Hutt

12-7-91 92111 847-boor CPC with 561 MIM giving instrument dual in C-172; weather VMC in controlled ainpoce. Alt gave C-172 missedapproach clearance to altitude 42000 feet When 5 roles east of depots. 0172 involved in midair at 2,200 fret with VFR aircraft cruisingon VFlit Bight plan and not in contact with AM. Two in cm Wei

26



Adak Cedisions (coattiaawl)

6-1-92 4148

7-7-92 92170

8-27-92 9679

Neaseah 1-10.89 01640

1-29-89 9537

24-89 0337

9-21-89 02328

7-22-90 02012

8-7 -90 92236

10-20-90 92038

4-36-91 #323

4-23-91 0255

11-20-91 0481

*Mee fhansemaiigin

1430-hom CFI eyeing dual in 0172 returoing to uncontrolled airport at 2500 Meet risk a C-182 was also inbound to airport at 1500 feet. CFI in C-172 odd eau he would rude and fast C- HM 0182 pilot saw C-177t later looked out tight window and saw C-172 converging in deuending tight tun. Left wing of C-172 bit dght side of 0182. Two kited in C-172. 560-hour CFI with 300 uht doing such- sakes with seass s C-172. Another C-172 reported inbound from mandireaion; ATC did not idendlyes tadar and controller.in briefing replacement.didnetpoiatoutinbound C-172. which meserreportedendownwindesimancred. but clod wpm turning base. Second C-172 was tracking Mode light path of dual C-172, and aircraft converged and crashed. Four peeps lolled.

2300-hoer CFI with 1500 MIM doing touch-and-go's with student in C-150. broadcasting intentions on 1229, the mom MR C-182 made fight traffic for tame tunny sad tamed a high final. C-150 cropped 0450 as C-172 turned lets bate to fiat C-182 pilot kaki

20.700-hour CFI giving dual in Banana 7GMC broadcast position on fiat Picas pilot announced potation on downwind. Pim overtook and landed on Bellows on show find. 27-hour studentfiyingC-1501andedon tap eflarefingC-172 at unconoolled airport. Bode pdots said they broadcast intention% neither Ewan ofother's pukka. Both sitar& woe doing touch- and-gen ATC handfuls 3 ma high wing Camas at busy airport rabitimesed tit C-152 flown by a 53-Itottratadentpdot and gme lea doom= to land, inteeding tkohnsoce foranother COMM The C-152 and a C-150 ceded about 20-30 feet AG!. on find. C-172 in which 440-hour CFI giving bwmantent wank; collided in mite while "mainetwer- ine with a training 0152. 0152 Noma to home airport 0472 made fenced knew* C-150 being &MU by 27-hors madames on fund at tit when it was hit on tipper cockpit and right wing root area by kinks aitaaft that had turned to ftnaL Neither pilot son other. (No information on radio peocedrares)

3232-hots CFI doing airwork in C-172 with Ss under hood; sigh side demise hit by tastary T-311 withal stabiliser. T-38 was being vectored for Kay AFB U S, going about 330 bats. C-172 did not lave Mock C. Two C-152s flown by a 213-hour 'commerce pilot and a 67-hour private plot corset in a flight school traffic prawn at an uncontrolled Sport on a dew day. There were 7 aircraft in the paw= ad ratio fieguencywas mutated One aircraft is daunted as in the pattern. Smother on a go- mound. Hots be visual comma with arch other. Two Bike school PA-28-1614 ofrenued by rivets pica with 83 and 75 been respeaiveiy, collided while approaching aspen under positive control. Propeller of 1 bit bottom% of other's steam Both aircraft did ms's following incident. 50-hour scuba plet in PA-28451 had just touched down when a 69-hour student pilot in a 0150 with 1 Imes head landed on top. (Apparently the airport was unceatestied) 7200 -boar CFI with 5100 hours NUM tin practice instrument approaches so comwenciri Soc.under aimed. in lefi SgaofConveir 600.A developmental contsellureas controlling uaffic under supervision and dewed alike& 19 to cross thearea stockists-2400feet Abash converged and collided. after 019 pilotviedto avoid Connie. ATC bad net notified eidter aircraft clear?: position.

27



Crash. ghettoised Fels

Midair Cages. (continted)

11-22-91 #1893 1657-ban CFI giving dual in C-150. made close-in bare* C-150 We wing hit the leaded stab- Amer of a C-172 that had attend the paean on downwind and was melting a long final

3-31-92 #1920 61 -hour student pilot lying 0I72 planning kenos at torencontrelled *Warn cleared to land from a right downwind. 16004m CFI with 700 NW giving dual touch-and-go's in another C- 172 to wodenn on lit downwind. CFI and his student misidentified ninth ATC said to follow, and on final "posit, descended onto student's shalt. Student landed safely on runway; CFI node foxed landing in sough tenant.

11-11-92 19762 3,500-hour CFI with 1500 ?AIM giving dual in C-182; made improper traffic pattern entry and collided with C-150 gowns by 390-hour CFI with 43 hams LUMI, giving Sett check. C-150 was tolling out of am from climb to downwind; left seat pant sew the Cr182 approaching from behind and to the left, pulled nose up shasply. C-182 rear seat prameger alerted 182 crew, which made imp left turn. (No information in report on cmmuaicasionn airport apparently uncontrolled)

Notes Neat of the Stir otilihisee occoreeti it might

C. DISSEMINATION OF STUDY RESULTS

Results of the investigation of crashes cf instructional (fishes are described in detail in other sections of this report to the Federal Aviation Adsninbaation. This section presents more details on some of the specific problems identified and focuses on how the results out be disseminated by the FAA to the aviation community, especially tight instructors.

HIGHLIGHT'S OF THE CRASH REPORTS

Our analysis of 638 crashes in 1989 and 1991 involving instructional Bights has brought to light a number of fags that probably are not wen known, even to instruaors. Among the surprising findings were the large numbers of crashes that occurred in 2 years under the following circumstances

aosswinds at the airport (120 crashes on solo, 57 dual),

touch-and-1p landings (84 on solo, 34 dual). fuel starvation, due to tidier inadequate planning/ peal& or misunderstanding the fuel system (SO on sob, 24 dual),

28

genarounds (28 on solo. 28 dual), and sinalated emergencies (sometimes at a low altitude or after ilLttnlaftt shut down engine) (1 on solo. 48 dual).

The above problems occurred in the course ofdual instruction, as well as on solo flights.

Mentionsl problems or deficiencies noted in connection with crashes of student solos include&

students carrying passengers (23 crashes). getting lost (13), and

failure to file flight plans on cross-country solos (34) - al hough not then. of she crash, these cases suggest poor oversight by the instructor.

Less common problems, of special interest because they an not widely reconnired, included:

carburetor icing while in cruise phase (13) and midair collisions with helicopters or acrobatic airplanes -2 types of sin:raft in which the flight patterns may be unusual -or the pilots' visual Reid may be limited (4 cases among the 30 midair collisions in the 4-year series).



That the most common event inuaediandy pieced- ins the crash was loss of control on lending (227 cases)

may surprise no one. The 94 stalls, however (half of them with an instructor oa board), are noteworthy, not only because oftheir numbers, but also because of their severity; Stalk snotnatied fen dart Sa f all final erasion

INSMUCIOR DEFICIENCIES

Solo andent Eights A student on solo is the pilot in command (NC)

and must be held responsible for anything that hap- pens on the fight. The identification ofsome crashes as being suggestive of inadequate training is often subjeaive, as is the choice of categories, fuel stan- dee, for example, was not mimed ea one ofthe above categories because training in this area is relatively smightforward and it is tenable to assume that the asentials have been taught by the tinter a student solos. The skills required fie the shone 4 categories sequin much more interaction between student and insane- tor before mastery is achieved. Even with the best trainiag, some student pilots may be Hedy to be involved in a crash because oftheir permed approach to decision-making and risk-taking.

Nevertheless, loss of control of the aircraft by the solo student. as well as many other problems, may reflect directly on the certified flight instructor (CFI) because of either inadequate training or CFI misjudg- ment of the student's lapel:Rides.

On the basis of renew of the 2-page NTSB briefs, it appeared that many of the 360 crashes on solo were suggestive of inedequute training. primarily in;

*evaluation oferomminds and weather (69 crashes), touch-and-go landinp (411),

recovery from bounced landings (35), and navigation (13).

In addition, 8 crashes occurred when instructors allowed students to solo irs hazardous conditions (poor visibility, snowy runny, etc.).

29

Oatiot eflannartlead Might

Dad flights Based on review of the 2-page briefs describing the

278 crashes of flights with an instructor on board, the most common instructor deficiencies were

letting the student get too far into an operational problem, so that the instructor was unable to recover control of the *Bait -or even compounded the problem (54 crashes); not anticipating a student's hasty action (19) (Table 22): simulating a forced landing in an area that did not afford a safe Isoefing place when the simulated emergency turned into a real one (10); and inadequate training in touch-and-go landings (8).

FMB appeals* In one-third of all crashes in the series, the NTSB

investigator cited instructor-related factors as con- taken so the aash. The factors most commonly as were

inadequate supervision (77 crashes), *delaying remedial action (34), emishandling of a simulated emergency (20), and inadvertent seallhmin (15).

Unandolpssod student anion Flight instructor training often does not include

preparing the CFI for the student's surprising, inappropriate reactions. As an be seen from the exemples in Table 22, students do unpredictable things. Flight instructors must anticipate and guard spina these maims

CFI INSIGHTS ncro srumorr BEHAVIOR

Looking at situations in which students experience problems, it is fairly simple to say that student crashes are due to student lack ofexperience, lack ofpractia, or insufficient undentanding of aerodynamics and the function of Eight controls, or poor preflight inspecdon or planing. But no matter what the direst ease,iostractioaalt echniques hover in the backpouncL



Crain fehessrebeal Piet

In addition to improved teaching, the instructor must constarady be perceptive about the student's behavior, and must also have insight into student reactions during each kwon.

The CFI must be particularly *kw to any student tendency to respond automatically in a hazardous way, such as pulling the nose up if too low on final, rather than adjusting pitch *Suede with power. Each time the instructor solos a student, the CFI mast balance objective factors. such as airport surface winds, traffic, etc., with the student's apparent capabilities on that particular day.

An instructor must have insight into &cols that might buoy a student (such as the ruder having 6 prior *bath on the fueling list, so the student departs with partially-filled tanks). It is the CFI's job to perceive things that may be stresses on the student pilot. Flight instruction cannot be hasty; the student

needs time, even during lessons, to absorb examples and techniques. Especially during takeoffs and landings, the student pilot needs time between each landing and takeoff to review the events of the previous circuit around the pattern.

CASES AND COMMENTS

The following situations typify many of the crashes involving instructional Bights- Section fi includes scores of cases this could be used to illustrate these problems, in addition to the ones presented below. (Although the individual case reports in Section B contain less detail than the following, the case number and date make it possible to obtain the teleran two- page brief from the NTS11.) Each of the cases that follow could be toed in educational material for student pilots and instructors.

Table St Examples of Katy Actions by Student Plots flying with Instructors Reseuling in Crashes st Instruedonel Riga 1911 and 1391

CIRCUKSTANCESI Student puled up nose for goated laad of adding power. On high ads take* Wen stall waning bind. CFI oiled for gear reaction; private plot trainee thought that wait wrong and retracted S When amphibious airplane bounced into air on pow boars S student reduced power. Cet pow reducdon to irte (simulated entrgency). at to nose heal eta Mar a boa on leas vitiate plat trainee Waded gear before aircraft WSW bath on runway.

Following rollout on tainted checkout. private plot trainee added par and swerved Ian oil nava. riot using rudder.

When CFI told student to Wean kit ahem. seat applied lei rudder and Wake. During mullegine instruction, student overcoirected; pia left. then right. When CFI loid student to deity rotating on takeoff, student rejected takeoff, abed balm. Student handed king on touchzand-go; CPI thought student would atop hut taint anted bisect

Student was a kit on lam added full power, aggravating dr& CFI pultad Cessna 150 throde to ids to simulate engine Isla; shatierta then puled mixture I to Idle- cutoff, which CFI did not notice.



Loss of coatrad on takeof- Crosswind Case 592, 5-26-89. A 19-year old student with 20

hours flight time was nying to Ste offin a Cessna 15? on runway 32, oa his third solo. The winch were 260 at 9 knots. After rotation and during initial climb, the aircraft veered over the left side of the ninny and struck a runway locator sign. Upon impact. the nosegear collapsed, tight wheel broke off, and right wing hit the ground.

Comments; In this cue, student inexperience seems to be a major factor. The case raises the question of how to improve the tesponaible flight instructor's teaching. This sequence of events appears to be a

typical case of not properly using the flight controls (initial left aileron, right rudder) in response to the crosswind and in response to the developing left- turning tendency following rotation and liftoff. The student needed to apply even more tight rudder with rotation and liftoff.

The crux of the problem with crosswind landings and takeoffs may be intructional emphasis on flight control coordination. Obviously, it is important to cruise and turn in a coordinated fashion, but there may be so much emphasis upon coordinated flight that the need for independent use of controls in handling crosswinds is ignored, or perhaps disci- plined away. For some students, warnings against the dreaded cross- contra) stall on the hose/final nun may have precluded using rudder and aileron =depot:len:1y in situations where that is an obvious =quietism=

In the crashes resulting from loss of control, over and over again. it is evident that pilots are not using enough redden Flight instructors must emphasize use of the rudder throughout the entire count *fudging. Students must be taught that rudder and aileron can be used in harmony, as when initiating a turn, but that it is perfectly natural to use these controls in opposi- tion: That upon takeoff, one can hold the upwind wing down by holding aileron up into the wind on that side, and still lamp the aircraft moving straight ahead by using tight rudder.

Basic to all of the student solo crosswind loss-ofcontrol crashes is the CFI's initial decision to permit solo flight under than crosswind condition e. In this and other cues, an instructor allowed an inapt/se eased student to solo with a substantial crosswind.

31

Oaths ofinbeet Higbee

While the crosswind Ices-of-control crashes in this series generally did not cause serious injury or death, the large number of crosswind aches indicants that remedial measures are media The FAA might con- sider recommending that Right instructors often se-

lea crosswind runways for dud training.

Loss of rand our landing - Totack-and-p Case 23`.1- 22-89. A 16-hour student pilot flying

a Cherokee 180 at Show Low, AZ (devation 6412 ft.. nun, had completed her second touch-and-go land- ins and was beginning her third takeoff on runway 24; winds 180 at 8 knots. The student remembered that the flaps were still full down, added power to continue the takeoff. and lost directional control when teeth- ing dawn to retract the flaps. The aircraft wad left and collided with a int bank. The student said she may have stepped on left radar as she tried so :enact the flap.

Commenran This is an example of a situation re- quizing mead teaching of the dynamics of touch- and-go's. Then was a 60-degree left crosswind at 8 Inoue the student was using a 75-foot wide mnway. to addition to the chubs*/ control promo required on landing and rollout, and then on adding power to take off, the student must be warned of the potential trouble resulting from teaching foe the flap handle (especially in this aircraft. which has a atedsanically- linked Rap ban& on che &or between theseats). The instructor might diagram the flight control and power kinematics for stages final approach, flares touchdown, rollout, flap resetting. takeoff, dianbout, etc. (Control and =neer usage vs. condiitions.e.g,aecwind left or right. tailwind left oe tight, and density skint).

Interestingly. available FAA training materials do not contain intonation on how to perform or teach touch-and-pis. AC 60-14, Aviation Instructor's Handbook (1977), asestans doing sand er's in dual lesson plans 7 and 8; recommends 3 takeoffs and 3 full-stop landings on fits solo; and in lesson 10, suggests for the second supervised solo, 3 takeoffs, 2 touch-and-go's. and 1 full-stop landing. The Ffight Instructor Practical Test Standards (PTS) apparendy contain no description of, or instructions for, teaching touch-and-go's. Because of the high incidence of crashes on touch-and-go's, the FAA should amend the YTS to include touch -and-go criteria.



Oat elfessernsbam4 Flea

Fuel starvation Case 1327,11- 14-89. At 6:00 p.m.. with darkness

approaching, a 30-hour student pilot became lost on a solo cross country in a Cessna 150. After trying for some time to locate hkuself, he called FSS for help; FSS was guiding the student against headwinds back to the departure area when the aircraft ran out of gas; the pilot landed 1 mile north lithe akport in an oil field pipe yard. The student listed 1 hour of instrument time.

Case 581, 6-11-89. A 47-hour student, sent on a

solo cross country in a Cessna 152, was told that the aircraft had been flown 1.5 hours since being topped off with fuel. The student became lorddisoriented; said he could not find his destination airport due to area flooding and a faulty radio, and was trying to return to departure airportwhen the aircraft exhausted fuel after 1 hr 50 gain. Student was able to land on a road. Aircraft actually had been flown 2.5 hours prior to student departure. Student listed 1 ban instrument time.

Gonwnents: Of the 51 fuel exhaustion cases, 26 were student ado cross country flights, and 8 of these students were lost Flight instructors should give more thorough in:unction on how to handle gettinglost. In addition to basic radio orientation procedures, the student should be aware that panic is a common effect of getting lost. Flight Training Handbook, AC 61- 21A Rev. 1980, under the topic losing track of posi- don, says. The greatest hazard to a pilot failing to arrive at a given checkpoint at a particular time, is panic' (p. 172). In addition, being lost is extremely distract taw student pilots should be taught not only to the aitplanes but ako to nsalatain awareness ofthe fuel supply and plan a landing prior to fuel exhaustion.

A number of cases in the crash series hatred insufficient preflight fuel check Pilots either did not look in the fuel tanks, relied upon someone else's estimate of fuel on board, or, for other reasons, misjudged the fuel available. The FAA should require, as pan of evny preflight aircraft inspection, that the pilot actually measure the fuel in the tanks. At least well-known pilot shop advertises fuel sticks calibrated for high wing Cessna tanks. A currently manufactured European trainer, the Czech Zlin 240, has a calibrated fuel dipstick in each gas cap.

32

Simulated emergency raseldag in stall Case 2173, 8-1549. A 22-year old, 1024-hour CFI

was introducing a 5-hour primary student to simulated engine failure. The CFI selected an area of mountainous terrain and reduced power to idle in the Cessna 152. The student rapidly raised the nose beyond stall attitude. The instructor pushed the nose over, but the Casa: collided with terrain during recovery as it was achieving a climb attitude. The steam was killed. The weather was noted as being 3.0 miles visibitny, with ceiling partially obscured in fog.

Comment= This simulation should have been started at a much higher altitude. Instructors, moreover. should initiate simulated emergencies in an area where there is a place available to land. This case exemplifies the fact that students can be counted upon to do something surprising. la this case, the surprise

resulted in an unanticipated stall at an altitude too low for recovery by the CFI. The NTSB record indicates CFI flight time of 267 hours in the last 90 days, and 92 hours in the last 30 days. This young flight instructor say have beat pushing hard to build hours - so hard that he was doing maneuvers with a primary student in near-IMC. In addition, the sky conditions may have influenced the choice of practice area and the student's reaction.

Siandatod essurgoacy Wire strike Case 1492, 44-91. A 23-year old, 311 -hour CFI

was giving a checkout to a private pilot in a PA-28- 140. near Carefree, AZ After doing maneuvers, the CFI retarded the throttle and told the pilot to do a simulated forced landing. On final to the landing area, the private pilot noticed power lines and began to add power to go around. The CFI told the private pilot to fly below the wires, took the controls and nosed the aircraft down. The Cheroker struck the power lines and crashed.

Comment= The Flight Instructor PTS for emergency approach and landing says not to continue a simulated emergency approach below 500 feet AGL, unless over an area where a safe landing can be accomplished, in compliance with FAR 91.79. The above report does not list the height AGL of the wires. In 5 cases in the series, aircraft hit wires on simulated forced landings. One preventive measure is to comply



with PTS criteria. Another recommendation is that the CFI find a good location for skated emergency practice and work there solo, in order to make certain that the area is a safe place for dual practice.

Snadeat pNot carrying passengers Case 1803, 7-2449. A 361-hour student pilot

(holding a valid medial certificate) was aiming 2 passengers in a PA-28-140. departing a 1940-foot gravel strip at Vashon, WA, destination, Port Townsend, WA. The pilot said later that during takeoff, he rotated about midpoint down the runway; after the Cherokee climbed about 30 feet, the stall wanting horn sounded. The PIG realizing the aircraft would not dear trees at the end of the strip, tried to set the aircraft down in an open field to the kft; the landing gear collapsed in soft terrain.

Comments: Unfavotatde conditions -3 people in a 150-hp aircraft in July; a short, gravel strip, and unspecified ̀ unfavorable winds.- make amended takeoff unlikely. Without carrying the 2 passengers prohibited by FAIts. takeoff might have been possible.

Many of the 23 students who were illegallycarrying passengen exhibitedexttentelypoor judgment in other tartest. Two ashes in the series were known to involve alcohol. and both were student Rights with pamengers.

Crashes of students carrying passengers were 4 times as likely to be fatal as other crashes in the series. They also were more likely to involve plots with more than 100 hours total time flying en student licenses.

Eleven student pilots had more than 200 bouts eight time. The FAA may wish to ascertain the cir- mamma= under which high-time student pilots are flying on student licenses - for example, whether they are pilots who have been unable to pass flight examinations, whether they are flying under the supervision of an instructor, and whether instructor endorsement; for unlimited solo cross-country trips between specified airports are appropriate.

Go-ersmad Case 2000, 8-13-91. A 1700-hour CFI with 300

hours in make and model was giving dual in a PA -23- 253. To simulate engine failure during the initial climbs, the CFI turned off the fuel supply to the

33

Gnaws efrassezesteati Item

right engine. The student performed the emergency checklist. which included feathering the right propeller (aircraft was not equipped with propeller unfeathaing system). The CR told the student to re- enter right traffic pattern; he clad se and overshot final. The left main gear down light was not on. The CFI took the controls in an attempt to do a single-engine go-around. The aircraft would not climb; it landed in the sweet and hit a sign and a pickup truck.

In 3 of the 9 cases of simulated engine failure in twin-engine aircraft, the CFI initiated the simulated emergent; by actually shutting down an engine, and ensuing single-engine go-stottnds were unsuccessful. Twin-engine simulated emergencies also involved unrelated mechanical failures, fuel starvation of the working engine, wire strikes, and controlled descents into terrain or structures.

While the FAA Flight Training Handbook (AC 61- 21A, 1980), approves shutting down an engine at a safe altitude (rainbows 3000 ft. above terrain), the text also specifies that such a shutdown be within landing distance ofa suitable airport. At Iowa altitudes, the aoudad= is accomplished by power reduction.

CONVEYING l'HE INPOI0dATION TO FLIGHT INS111UCTODS

A major past of the FAA's mission is to promote aviation safety; training Right instructor nose ofthe most important ways in which the FAA accomplishes its statutory objectives. This research provides a basis for developing Advisory Circulars (AG) and other materials likely to be used by students. In addition, it is important for the FAA to ensure that Right insane- tots benefit from the results of this investigation.

The FAA series on Accident Prevention (FAA-P- 8740-1 to 53. Rev. 1987) contains a gnat deal of useful information for beginning and advanced pi- Ion. Ideally, the series &odd be actively distributed to each student when the FAA is first informed of issuance of a student license. since some instructors may act make use ofthe material in than. The present study suggests, however, that more attention should be even to instructon, themselves, so that they will



Choke ofisserwaveasel-

know what the problem areas are. and will gain some insight into what they can do to prevent crashes, such as the ones described in this report.

Rlgia lastrecter Ladd eatifkatios Information gleaned from this research should be

incorporated is materials that the instructor is required to master in only to be a CFI. This could be accomplished through several avenues, including a) aarricelar materials used at schools that train instructors, b) questions on written tests for CPIs. and c)

discussion during CFI flight tests by FAA flight exam- Men.

Flight instramor recertification Flight instructors must renew their cettificates; ev-

ery 2 years. This provides a potentially valuable opportunity for interaction between the FAA and CFI:.

Over the past 20 years, the FAA has moved from directly conducting the flight instructor refresher counts, to complete privatisation of CFI renewal courses, the curricula of which an reviewed by the FAA and approved. The course content is developed by die contractor, for example the Aircraft Owner's and Pilots Association orjeppesen Sanderson. Course content does not necessarily reflect FAA knowledge of instructional crashes, but could be modified to do so, for example through d'emlopment of a module based on the findings of this study.

The second method of recertification is based upon a flight instructor's record, i.e., having successfully trained a certain number of students.

Neither of the above methods requires an in-depth encounter with knowledgeable and experienced FAA personnel. Although it is possible for a CFI to renew by taking a check ride with an FAA inspector, this method may be undemtilited. During the time when the FAA conducted the CFI renewal course at the Training Academy in Oklahoma City. flight instnsc- ton often expressed a perception that the personal contact and communication with FAA personnel were beneficial. We fed that some results of the present study suggest that it may be advantageous for the FAA. once again, to have more direct personal contact with CM during recertification.

34

Dina commemieation with flight inatructors - flight instructor pamphlets

Currently, the FAA sends flight instructors copies of its flight examiner newsletter. A separate, detailed publication specifically targeted to instructors would provide a means of emphasizing potential problems and solutions faced by students and their instructors. The FAA could mail to each instructor holding a current metal certificate a quarterly Flight Instruc- tor Pamphlet, perhaps designed to be 4 pages long, describing a significant instructional problems, suggesting ways in which the instructor could anticipate and prevent this problem.

Fortin*, our tardy identified 13 cases ofcarbn- maim icing in cruise - serious enough that the aircraft crashed. The FAA could develop a pamphlet describing these cases and °Sittig symptoms of carburetor icing, such at a drop in engine RPM: (fixed pitch prop), decrease in manifold pressure (constant speed prop), engine rougluans, or unexplained high fed consumption.

Pilots are interested is crash reports. If, in each pamphlet. the FAA discussed several instructional crashes that illustrate a significant instructional problem, and outlined preventive measures. this information should be well received.

It would be helpful for the FAA to list the particular unanticipated hasty actions identified in this study, and to develop a flight instructor pamphlet describing the circumstances, and how to handle that

For example, a common failure is retracting the gear instead offlaps. The CFI should always neck the motions of the student's Std. from throttle to flap bandit Training students always to say 'flaps identified; flaps ups and not retract flaps until dear of the runway, may be actin. If doing touch-and-go's in a retractable gear aircraft, the CFI must monitor pilot hand motions, or even guard the gear handle.

The FAA has not revised its instructional mining bock since publication of AC 61-21A. Flight Train- ing Handbook. in 1980. The latest and most specific trainee' g information is contained in the Flight In- structor Practical Test Standards, which could be incorporated as reference material in the suggested FM flight instructor pamphlet&



RECOMMENDATIONS

Comavanitattion of study results We recommend that the findings from this study

be communicated in some way to every CFI and to each entity producing CFI renewal court. This may be accomplished through a variety of approaches.

including but not limited to: 1. Development of a series of flight instructor pamphlets could be made available during CFI renewal; this would provide a way for the FAA to consuunkate the information directly to instructors. Instsuctional training manias from the Prac-

tical Test Standards could also be incorporated in the FAA Flight Instructor Pampldets. 2. Flunk' bine a copy of the entire Final Report to sponsors of flight instructor refresher courses, and

anyone else known to be developing instructional materials. Copies could be provided to F1RC span-

:on upon their renewal.

3. Development of modules for use in flight instructor courses based on the findings otitis study. illustrating what can be learned from crashes involving students and instructoes. 4. Development of Trigger Tapes that emphasize

the problems identified by this research.

5. Publication of this study's results in FAA Avia- tion News and the development of 'popularized* version of these findings in articles for instructors in Image:inn midi as Flight Trani% Ariasines Praftektety et Careers Afsgaziar (Publisher: Mel- ina Murphy. Editor: Scott Spangler).

For the initial certification of inattuctors, findings from this research could be incorporated in materials that the instructor is required to master, including

1. Curricular materials used at schools that train instractors. 2. Questions on written tests for ale 3. Discussion during CFI Right tests by FAA flight examiners.

35

Crean eflanniskadayits

liettonnendadona specific to asscilt-anata's The FAA should *stead the Flight Instructor !Prac-

tical Test Standards to include instruction for teach-

ing touch-and-go twangs and criteria for determining when students are ready to practice touch-and-go's during sole flight.

When AC 60-14 is revised. consideration should

be given to deleting the suggestion that the second

supervised solo include touch-and-go's. Guidance should be provided to instructors as to how to teach

touch-and-go landings and evaluate student readiness

for solo touch-and-gess

D. INFORMATION FROM MRS REPORTS

Information on lassos:done, flights was obtained from the Aviation Safety Reporting System (ASKS).

ASKS is a eonfidentiol. voluntary reporting system for safety-telated incidents involving aircraft or associ-

ated facilities. The ASRS system coatains over 178.000 reports. These reports are submitted by pilots, air traffic contralto, and others concerned about aviation safety.

For this study, a request was made for a sample of 200 reports involving general aviation instructional fights canning in 1992 and 1993, These reports were selected from the ASKS database, using key word searches on the report narrative. since flight instruction is not one of the formatted fields in the &sabots. The repeat used for this analysis seise limited to general aviation airplane instructional Higbee Heli- copter and air taxi instructional flights were excluded. One hunted sixty-four of the 200 reports were determined to be valid flight instruction reports.

Each report was reviewed and coded by an experienced aviation safety analyst. Descriptive information was collected including the month and year of the incident, weather was, state of oanurence, aircraft type, and reporter function (student, instructor. etc.). Information was also coded on phase of flight when incident occurred, the type ofinstruction being performed, type of event (NMAC, etc.), pilot



Canso qthattetasteell- Mate

factors associated with the event, flight instructor factors associmed with the event, specific characteristics of the event and a short narrative synopsis.

Tables detailing these ASRS reports are followed by summaries of the near-midair collisions.

On average, there were 1.9 anomalies mentioned in each ASKS report. Reflecting the kinds of events that are most likely to be self-mponed by pilots, the most common anomalies were such conflicts as near-midair

collisions, enechanical problems. and violations of the Federal Air Regulations (PARS) (Table 23).

The most common phases of Right when the yob- km occurred were landing and final approach (Table 24).

As in the case of the crashes of instructional flights. the most common flight instructor performance fac-

ts were inadequate supervision and delayed sane- dial action (Table 25).

Table 23. Ananias Si 124 ASRS lispolls ANOMALY Conflict

ConlitUNMAC Canitataborne less SWIM Conlikaround less are Conga/ground cake Lees Men legal separation

Weather !!FR in IMC Wight encouraedweetter Matt encountediehef

Mechanical Airadt equipment problemicdtital Akan* equipment geobletelses severe

Deflations AllitudeheadIng S. deviation Speed deviation laced upon talk pattern deviation Erroneous penetration or St of *apace Track or heating deviation

Control Loss of mats Controlled fight lovird terrain

Tranagressiona Rummy or Rainy mansion Rummy Ira nspesawoulhonzed landing Rummy traragrealaher

Akb.de Deviates Alt deaumeshoot on Snit or descent Alt cadescuroin from aligned

FAR Mebane Non adherence legal requimmeteclearance Non adherence legal requiremenUpub

procedure Non adherence legal requiremenbFAR Non Wheat legal requitement/ether

TOTAL

25 9

14 0 2

4 12

1

45 4

0 1

7 9 9

23 12

13

10

5 9

21 35

310



S 9:11.rooftram

anostn:V9f

01111111,1111

110$1200$141411



ors C9 IN 2 ye Pi. CD gm CV CD C9

CD CD CD v. on. 9- v- CD f% ::

"VX

IV.

ograntler.voi

1

I i

!I 1141

i

I 'wog, i

; 11111/Iihi g



The pilot performance factors presented in Table 26 provide insight iato some of the actions that led to the problems. The most commonlycited weredistrac- don, stress, poor pm :flight planning, and poor corn- ratoicseion between instructor and student.

Additional details on the 34 near- midair and other conflicts, are presaged in summary form at the end of this section. It is interesting that, as in the case of the 30 midair collisions described in Pan I of the study results, 3 of the student pilots were under a hood at the time of diecoaRkt. Also as in the midair collisions, 2 conflicts invoiced helicopters. One flight instructor reported a near- midair collision with a nooconummicating helicopter, at an airport where helicopter traffic and aircraft from a large training facility nearby offer substantial potential for midairs.

cradies offerowaistsai Sten

Another near-midair collision in the ASRS data base for 1992, although not included in the sampled cases, provides insight into the problem of conflicts with acrobatic planes. In addition to the possible restriction to vision because of its hi -wing ©Agin*. don, the case involved an acrobatic plane (with no radio) flying a shorter pattern than aircraft from a nearby flight school, which commonly flew a longer. wider pattern.

In conclusion. the events described in the ASRS reports, although they rarely involve mashes, provide insight into the citannuances surrounding "near- misses" and, therefore, may be valuable in the preven- don of actual crashes.

Oats and Nearlilidide Collisions Repotted to MRS lostnntiosel Flights Daring 1992-1993

ASRSSYNQ COMMENT1: COMMENTI

ID: ASRSSYNO:

COMMENT!: COMMEND

ID: ASRSSYNO: COMMENT!:

COMMENT1

ID: ASRSSYNO: COMMENT):

ASRSSYNO: COMMENT':

COMMENT2:

ANOMALIES LIMITED TO CONFLICT/1044C 237955 INSTRUCTOR war STDNT HAD NMAC WITH SMA. STUDENT UNDER HOOD, 500 FT CLIMB, MISSED OPPOSITE SMA TRAFFIC NO EVASIVE ACTION TAKEN.

238636 NMAC WHEN TWR ISSUED RIGHT TURN OUT OF PATTERN INSTRUCTION TO AC THE FLIGHTWAS IN THE PA'TTERN.CONFLICTOCCURRED.ATC ISSUED RIGHT TURN OUT INSTRUCTION. NMAC ALMOST OCCURRED.

204024 NMAC CFR WITH STDNT UNDER HOOD ON APPRCH, HAD NMAC, NEVER SAW OTHER AC, WAS INFORMED BY PUT OF OTHER AC

205119 MULTIPLE RWY OP, INTRSCTNG RWYS, NMAC, SYS ERROR (CONTROLLER) CH ALMOST HITS SMALL TRANSPORT LANDING ON XDIG MN. CONTROLLER ERROR.

206577 NMAC CFI ON DOWNWIND NOTICE HELI ENTERING DW IN ERRATIC MANNER. HELL TURN TOWARD CR WHO TOOK EVASIVE ACTION. SOLO STUDENT IN HELL



Oman eflatransawl' diegks

1111

ASRSSYNO:

COMMENT I: COMMENT2:

Pik ASRSSYNO: COMMENT': COMMENTI

ASRSSYNO:

COMMENT': COMMENT 2:

ID: ASRSSYNO: COMMENT':

COMMENT1:

3m ASRSSYNO: COMMENT':

COMMENT12:

ASRSSYNO: COAOAENT1: COMIAENT2:

ID: ASRSSYNO: COMMENT1:

COMA4ENTI

206773 PLT EXECUTES 180 ON FINAL AND ENCOUNTERS ANOTHER AC ON APPROACH,

NMAC NEW PILOT (2 DAYS) ON 1.5 MILE FINAL THOUGHT HE WAS APPROACHING WRONG AIRPORT. DID 180 TO GO TO OTHER APT, ALMOST MT FOUJOWING TRAFFIC

207176 NMAC WITH OAT DEPARTING FROM SAME AIRPORT CFI AND STDNT ALMOST COLLIDE WITH MT DEPARTING SAME AIRPORT DURING CUMBOUT.

208170 NEW CFI GIVING BM TO OTHER CFI HAD NMAC IN APT TIC AREA. BOTH AC WAS PATTERN WORK ON 25L ASKED FOR FULL STOP ON 25R. ATC CLRED AC FULL STOP ON 25L PLT FLYING SET UP FOR 25R. CROSS IN FRONT OF AC FINAL ON 25

208377 NMACAFTERATCCALLEDIVIUMPLE TARGETS,ACUNDERRADARADV/SORIES TFC CALLED BYAPPRCH, STDNT RESPONDED "am TRAFFIC IN SIGHT, WRONG TRFF cH HAD TO TAKE CONTROL TO AVOID AIR CARRIER.

209199 IN PATTERN, CFI HAD NMAC. Ric WAS NO HELP, DID NOT CALL TRAFFIC CR SAW OTHER AC ABOUT 100 FEET AWAY MANEUVERING TO LAND FOR OTHER RWY CFI TOOK EVASIVE ACTION. ATC DID NOT CALL OUT OTHER TRAFFIC

20901 NMAC AVOIDED AS CF1 HAS TFC SIGHTED AND DIVES AS AC JET CLIMBS CFI HEARS HIS AC CALLED AS TRAFFIC TO AC JET. STARTS LOOKING AND SEES

DIVES TO AVOID NMAC, SEE AND AVOID WORKED BUT ATC SHOULD DO BETTER JOB

210731 2 VFR AC HAD NMAC WHILE GIVING INSTRUCTION IN EXCEL LENT VISIBILITY ON APPRCH TO NDB, SMA ALMOST HITS ANOTHER SMA ON APPRCH TO SAME NDB, LEFT AREA AND RETURNED AND ALMOST HIT SAME AC AGAIN.



/ lit .7#1

cP 4se

9+1

.4,

a 49 I /



Creabanthwarankneiftfies

LEk

ASRSSYNO: CO).Q4ENTI: COMMENT/

alq ASRSSYNO: COINAENTT:

COMMENT2:

ID: ASPSSYNO:

COIAMENTI: COMPAENTI

ID: ASRSSYNO: COMAENTI:

COMMENT/

ifs ASRSSYNO:

COMMENT1:

COMMENT2:

Qk ASRSSYNO:

COMMENT':

COMMfl4Th

222164 3 AC IN A NIGHT TOUCH AND GO PATTERN. 2 HAD NMAC. MITT DID NOT SPACE WELL IN PATTERN, OVERTOOK AC IN FRONT. CFI HAD TO TAKE CONTROL

222463 2AC HAD NMAC IN MARGINAL CONDITIONS DOING IFR TRAINING. TWO TRAINING AC ALMOST COLLIDE IN MARGINAL Vat CONDITIONS CON- DUCTING IFR TRAIN BEFORE OBTAINING Alt CONTROL

223611 ON FINAL APPROACH TAKES EVASIVE ACTION TO AVOID SECOND AC ON FINAL TWIN ALMOST RUNS SINGLE OVER ON TURN TO FINAL AT CONTROLLED AP.

224045 2 SMAS HAD NMAC IN PATTERN NMAC DUE ID CFI LETTING STDNT HANDLE AIRCRAFT TOO LONG. NEVER ANY RISK OF COLLISION.

224747 CFI WITH STUDENT HAS NMAC WITH 11,7/14 WHICH THEN MAKES A SECOND PASS. AGGRESSIVE TWIN PLT GETS MAD AFTER NMAC AND MAKES SECOND PASS AFTER SLOW ACROBATIC HIGH WING.

225674 INMOST FLIGHT HAD NMAC wry ANOTHERACTICOFF IN OPPOSITE DIREC- TION. ON PRACTICE INSTRUMENT APPROACH TO LAND, AC HAD NMAC WITH M- artin AC CLEARED BYATC TO TAKEOFF OPPOSITE DIRECTION.



ID: ASRSSYNO:

CX3MMENT1:

00M/ANTI

ASRSSYNO:

COMMENT':

COMMENT2:

ID: ASRSSYNO: COMMENT1: COMMENT/

ED:

ASRSSYNO: OOMMEVIT: COMMENT2:

ID: ASRSSYNO: COMMENT':

COMIANT2:

ID: ASRSSYNO: COMMENTI: COMMENT/

ID: ASRSSYNO: COMMENT!: COMMENT/

*can of hantesissal Higbee

ANOMALIES 'LIMITED TO COATPL/C77LESS SEMIS

204124 INSTIAT TRAINING FLT TAKES EVASIVE ACTION TO AVOID CONFLICT ON APPROACH ON CIRCLE TO LAND Wm SHORT FINAL AC ALMOST HIT ANOTHER AIR- CRAFT ON APPROACH. STDNT UNDER HOOD UNTIL SHORT FINAL CFI SCAN NOT GOOD.

209433 SMA HAS ENCOUNTER WTTII AIR CAR WAKE TURBULENCE DURING INITIAL CLIMB CFI AND STDNT SHOOTING T+G. COMMUTER DOING PRACTICE NEB AP- PROACHES TO OTHER RUNWAY. ON TWEE, THE SMA ENCOUNTERED THE WAKE TURD, 45 DEG ROLL + FITCH UP

209522 HDG TRACK DEV ON ',MILS APPROACH ON LONG FINAL ON ES (10 MILES) CFI TURNED OFF LOCALTEEA DUE TO FAST TIC BEHIND. CFI TURNED BACK ONTO COURSE WHEN HE HEARD CERN FOR OTHER AC

216413 ALTDEV BY MULTI - ENGINE TRAINEE STUNT WAS CLIMBING TO 6500 FT, TOLD TO LEVEL FOR TIC AT 3500. MISUNDERSTOOD AND CONTINUED TO 02M11 INTO OPPOSITE TIC PATH.

217116 LOSS OF RADIO RESULTS IN FOSS NMAC, SEVERAL UNMAN LANDINGS. RADIO VOLUME TURNED DOWN WHILE IN PATTERN AT BUSY T'WR APT. CFI AND STDNT DID NOT NOTICE UNTIL LATER.

218246 CFI INMATES SIM ENG FAILURE IN ON BASE LEG AT NONTWR AFT. 01 INMATES 8114 ENG FAIL IN PATTERN TURNING LEFT BASE WHILE OTHER TIC FLYING RIGHT BASE. cur OTHERS OFF IN PATTERN (AND AWARE OF TI).

222373 CFI AND STDNTIN TIC PAT ACCUSED OF (MOSE PROX BY FAA OFFCL. IN TWIN CFI HASSELED BY FAA EXAMINER WHO CLAIMED CH AND STUDENT GOT TOO CLOSE IN PATTERN. CATS FBI AND FAA HAD BAD BLOOD OVER PAST EVENT.



Oasis eltestentbasl Saber

REFERENCES

Baker, 52., Lamb, M.W.. Li. G., Dodd, R.S. Human Futon in crashes of commuter airplanes. At Spar Fain Ake( 1993M: 63-68.

Baker. S.F., and Lamb. M.W. Hannon haws ie Crakes sf Garanwa aid Air rani A Report to the Federal Aviation Administration. Contract *DTFA01-90-C-00046. May 11,1992.

Baker. Si?., and Lamb. M.W. Hazards of Mountain Flying Crashes in the Colorado Rockies Asiat Spa Enka Meet 60:6:531- 536, 1989.

Caro, P.W. Flight Training and &minion. In E.L. Wiener and D.C. Nagel (eds.). Haan Foam in Austin. Sabot San Diego, California: Academic Peen, 1988.

DeLacetda. F.G. Sze and Amid Stillwater OK Delta Aviation, 1988.

Federal Aviation Administration. Fret Its Para- dad Tea Suardark)brAtalase .Simegit begirt and Aida Engine. AFS-632. 4-9-1991.

Federal Aviation Administration. DearisyAltitada FAA- P-8740-2; AF0-800. (No date.)

Li. G., and Baker, S.P. Crashes of Commuter thank and Air Tot What Determines Not Sutvivan J Ckarsiabsafei 35:1244 -1249,1993.

National Transportation Safety Board. ellarrfiAreidese Rent Am* Cain e Magi Was dirks° inch C-99 WOWED eadAestketr4 Inc, Also

mosamoder 11270 12SM) Mar San Luis Sap CA. Aqua 24 1.9 Report No. NUB AAR-85-07. Washing' goo. D.C. 1985.

National Transportation Safety Board. AirerafiAcra kat Rear Midair Carstaue of Area Airbus:. Semi* est Mara II. MOS,: asd Misty Ma HMO& flat Usab. jam" iS 1987. Report No. NTSRMAR-88103. Wadtingwe.D.C, 1988.

National Transportation Safety Board. U.S. General Aviation Calendar Year 1990 (also 1987. 198& 1989, 1991. 1992). Report No NTSBOARG-91/ 01. Washington, DC: National Transportation Safety Board. 1993.

44 twatounnanlaintilanalnMIS-laralltlear PDF compression, OCR, web optimization using a watermarked evaluation copy of CVISION PDFCompressor


Documents

Crashes of instructional flights: Analysis of cases and ...avhf.com/html/Library/Tech_Reports/cami/AM96-03.pdfLIST OFTARLFsS Gashes of Imauctional Flights.1989-1992 Table Page 1. Injury