Technologically Advanced AircraftSafety and Training

AOPA AIR SAFETY FOUNDATION

Executive summary

T echnologically Advanced Aircraft (TAA) have been enteringthe general aviation (GA) fleet in large numbers since earlyin the decade. TAA are grouped into three categories: newly

designed aircraft, newly manufactured classic design aircraftequipped with new avionics, and retrofitted existing aircraft of vary-ing ages.

Our analysis, while preparing this report, shows TAA having propor-tionately fewer accidents compared to the overall GA fleet. TAAhave experienced reductions in the percentage of takeoff/climb, fuelmanagement, and maneuvering accidents, and increases in landing,go-around and weather crashes, as compared to the fleet.

Light GA pilots are now undergoing the transition that the airlinesand corporate pilots underwent in prior decades. The use of autopi-lots as an integral part of single-pilot IFR TAA operations shouldbe embraced. Training requirements center on differences in new-design TAA handling characteristics and the addition of capable butcomplex avionics packages.

Deliveries of new equipment have been accompanied by insurancecoverage requiring factory-approved training. CFIs and pilots areadapting along with the manufacturers and training organizations,gaining in experience and capability. More and better simulation isgradually becoming available to TAA pilots and ASF considers thisan essential part of learning to use the avionics.

Training to use nontraditional avionics using traditional methods isnot optimal. Use of CD/DVD and online simulation is a step for-ward, as is the development of relatively inexpensive simulators fornew TAA.

1

Technologically Advanced Aircraft Safety and Training

I. Introduction and overview . . . . . . . . . . . . . . . . . . . .2Questions this report will answer Technologically advanced aircraft (TAA) defined New and legacy cockpitsMore than hardwareHistory of TAA What’s next?

II. Safety implications . . . . . . . . . . . . . . . . . . . . . . . . .5The good newsThe challengeThe physical airplaneThe mental airplaneBeyond workload: over-reliance

III. TAA accident history . . . . . . . . . . . . . . . . . . . . . . .9Comparing glass-cockpit TAA to all GA aircraftCirrus accidentsType of operationComparing TAA accident pilots to non-TAA accident pilotsTAA and the parachute

IV. Training for the glass age . . . . . . . . . . . . . . . . . . .17Training requirements and sourcesA training sequenceTraining a new breed of pilots?Autopilot essentials Pilot performance and its effect on human factorsPilot performance and its effect on trainingThe automotive experienceTraining, liability and flight data recorders

V. TAA hardware and software . . . . . . . . . . . . . . . . . .26Integrated avionicsPrimary flight displaysWeather displaysTerrain awarenessAirspace displaysTraffic avoidanceEngine/systems monitoringTechnology abused?Some concernsAvionics maintenance and ownership

VI. Report conclusions . . . . . . . . . . . . . . . . . . . . . . .32

Table of Contents

Mooney Ovation 2 GX

2

T he first edition of this report was published in 2004as a preliminary review of TechnologicallyAdvanced Aircraft (TAA) accidents. Since then,

TAA have entered the general aviation fleet in significantnumbers, with more than 5,700 GA glass-cockpit aircrafthaving been delivered. This updated version ofTechnologically Advanced Aircraft: Safety and Trainingprovides a statistical analysis of TAA accidents, comparingtheir safety with that of conventional aircraft. This analysisis based on accident data contained in the AOPA AirSafety Foundation (ASF) Accident Database.

Questions this report will answerThis AOPA Air Safety Foundation Special Review ofTAA answers three questions:

1. What adaptations to the general aviation (GA) train-ing structure have been made as TAA have entered thefleet in significant numbers?

2. What GA accident trends have emerged involving TAA?

3. What changes to TAA or training might be considered?

Technologically Advanced Aircraft(TAA) defined Technologically advanced aircraft are equipped withnew-generation avionics that take full advantage ofcomputing power and modern navigational aids toimprove pilot situational awareness, system redundancyand dependence on equipment, and to improve in-cock-pit information about traffic, weather, airspace and ter-rain. By FAA pronouncement, a TAA is equipped withat least the following:• a moving-map display • an IFR-approved GPS navigator• an autopilot

Nearly all new aircraft go far beyond the basic defini-tion, sporting enough electronic displays to qualify ashaving a “glass cockpit.” ASF’s working definition of a“glass cockpit” includes a primary flight display (PFD)to replace the traditional “six-pack” or “steam gauges,”as round-dial mechanical instruments are known, and amultifunction display (MFD). The MFD, as the nameimplies, can show myriad items including a moving map,terrain, airspace, weather, traffic, on-board weatherradar, engine instrumentation, checklists, and more. Asthis went to press, more than 5,700 GA glass-cockpit air-craft had been delivered. According to a recent AOPAstudy more than 90 percent of new production aircraftare being delivered with glass, so it’s a safe bet that soon-er or later, most active pilots will be transitioning.

There is no current reliable estimate of how many exist-ing aircraft have been retrofitted to become TAA, but itwill be into the tens of thousands. New fleet sales toflight schools and university flight departments arealmost universally glass cockpit—even for basic trainers.Most leading aviation universities have adopted TAA toprepare pilots for the next generation of flight, be itGA, corporate, or air carrier.

New and legacy glass cockpitsSome TAA are completely new designs such as theCirrus, Columbia and Diamond, while others are updat-ed versions of legacy machines such as the Cessna,Piper, Beechcraft, and Mooney product lines.

Retrofitted, or retro, TAA are previously delivered legacyaircraft with instrument panels reworked to add TAAequipment. This report focuses on newly designed and updat-ed legacy aircraft with factory-installed glass cockpits.

Introduction and overview

New TAAinstrument panel

in a Diamond DA–40.

Legacy TAAInstrument panel

in a Mooney Ovation 2GX.

3

More than hardwareMany observers believe that the deeper importance ofthe TAA takeover goes beyond just equipment. Thelarger definition includes a new mindset for pilots,encompassing a revised view of what constitutes GA fly-ing, with airline-style procedures, regular use of autopi-lot, and greater dependence on avionics for multipletasks beyond pure navigation.

Although pilots flying classic high-performance aircraftunder IFR often use this approach, its application isessential in the successful operation of TAA. To processlarge amounts of information and not allow flight safetyto suffer, pilots must add “systems manager” to basicstick and rudder skills. This mental shift has proven to bea challenge for some conventionally trained pilots. Thereis a belief by some pilots, abetted by sales literature andaircraft sales personnel, that TAA has altered the funda-mentals of GA flying. Despite some significant differencesinvolving how the aircraft is operated, the core of pilot deci-sion making and many of the risk factors remain exactly asthey have been with non-TAA aircraft.

History of TAAFrom the beginning of powered flight, through the1970s and 1980s, traditional instruments and displaysdominated aviation. For much of that time, VOR,DME, and ADF were considered state of the art, butwere not a major concern in the aviation trainingprocess. Once pilots mastered the principles of avionicssystems management, transition to a new airplanerequired only cursory instruction on avionics because allequipment worked essentially the same way. The bulk ofpilot checkouts were spent learning the handling of air-plane characteristics and systems.

Then, in the late 1970s, the first GA area-navigation(RNAV) systems appeared. By the early 1980s, generalaviation began to embrace the technological revolutionas computers worked side by side with humans in thecockpit. The transition was visible first in military air-craft a decade or so before, but it wasn’t long before“glass” started invading the cockpits of business jets andlarge Airbus, Boeing, McDonnell Douglas andLockheed aircraft.

Technologically Advanced Aircraft Introduction and overview

Columbia 400

4

In the 1980s and early 1990s, the initial versions of com-puterized cockpits were relatively simple by today’s stan-dards: small glass TV screens (cathode ray tubes, orCRTs) capable of displaying graphics of traditional air-craft flight instruments. These electronic flight instrumentsystems (EFIS) came to be known as “glass” and aircraftsporting them as glass-cockpit aircraft. CRT displays weresuperseded in the mid-1990s by liquid crystal displays(LCDs) that delivered much larger pictures at a consider-able savings in weight and energy consumption.

Even the early CRTs, however, could graphically repre-sent multiple items of flight information in the samelocation on the screen, forever changing the basic six-instrument scan three generations of pilots had come toknow so well. For many pilots, the change to glass PFDswas straight-forward. The attitude indicators and flightdirectors looked pretty much the way they always hadand they were always in the center of the display.

Today, although the bulk of the existing 180,000-pluslight GA airplanes still use steam gauges, virtually allnew GA aircraft are delivered with glass cockpits. Whilesome manufacturers still offer the traditional six-packinstruments, few aircraft are delivered with this option,except those intended for pure recreation.

Many aircraft owners are retrofitting their classic aircraftto convert them to TAA with IFR-certified GPS naviga-tors, multifunction displays and upgraded autopilots.

What’s next?As technology continues to evolve, airliners and busi-ness jets are sometimes on the leading edge of evenmore sophisticated cockpit technologies, though GA air-craft are likely not far behind. The new Boeing 787,Airbus A380, and several business jets will work with

Microsoft Windows-like displays and trackballs to sim-plify data input. Knobs, in fact, will serve only a backupfunction as equipment tunes everything automatically.

The trickle-down of flight management systems (FMS)for light aircraft is already providing keyboards andother user interface enhancements, replacing multi-function controls that must first be configured beforedata can be entered. Keyboard and trackball data entrycan benefit the pilots of space- and cost-constrainedsmaller aircraft.

Cockpit space constraints were at least part of the ration-ale behind limited control interfaces, which experienceshows to be one of the more challenging aspects for pilotstransitioning to TAA. In the early 1990s there were atleast five manufacturers building IFR GPS navigators andall had different operating logic and displays. This con-tributed significantly to the training challenge for pilotswho flew multiple aircraft equipped with different units.At this writing, that number has dwindled to two or three.

Further down the road is the possible introduction ofhead up displays (HUD) and enhanced vision systems(EVS) in general aviation cockpits, although for thenear term these devices will likely go to high-end air-craft. Such systems allow an easier transition from fly-ing instruments to visual references during instrumentapproaches.

Light GA is leading the way over its larger and moreexpensive cousins with datalink and WAAS installations.In some cases these are on portable devices that are notofficially approved for IFR flight, but pilots use themfor supplemental guidance, thus gaining valuable experi-ence that can be applied if they upgrade to an approvedinstallation.

Left to right: Eclipsekeyboard, Garmindata entry pad in

Columbia, andDassault Falcon

EASy™ Flight Deckwith cursor controldevices (trackball

mouse)

CO

UR

TESY O

F EC

LIP

SE

AVIA

TIO

N

5

A s TAA were being introduced, both regulatorsand industry recognized that they were creating a new world of opportunity and challenges for

general aviation pilots. In 2003, ASF participated withthe FAA, academia, and other industry members to helpwrite General Aviation Technically Advanced Aircraft—FAA/Industry Safety Study.

The team findings were:

1. “The safety problems found in the accidents studiedby the team are typical of problems that occurred afterprevious introductions of new aircraft technology andall also reflect typical GA pilot judgment errors found inanalysis of non-TAA accidents.”

2. “Previous safety problems similar to those identifiedin this study have been remedied through a combinationof improved training and, in the case of new aircraftcapabilities, pilot screening (i.e., additional insurancecompany requirements of pilot experience).”

3. “The predominant TAA-system-specific finding isthat the steps required to call up information and pro-gram an approach in IFR-certified GPS navigators arenumerous, and during high workload situations they candistract from the primary pilot duty of flying the air-craft. MFDs in the accident aircraft did not appear topresent a complexity problem. The team also believes

that PFDs, while not installed in any of the accident air-craft and just now becoming available in TAAs, similarlyare not likely to present a complexity problem.”

4. “TAAs provide increased ‘available safety,’ i.e., apotential for increased safety. However, to actuallyobtain this available safety, pilots must receive additionaltraining in the specific TAA systems in their aircraft thatwill enable them to exploit the opportunities and operatewithin the limitations inherent in their TAA systems.”

5. “The template for securing this increased safety existsfrom the experiences with previous new technologyintroductions—the current aircraft model-specific train-ing and insurance requirements applicable to high-per-formance single- and multiengine small airplanes.However, the existing training infrastructure currently isnot able to provide the needed training in TAAs.”

6. “Effective and feasible interventions have been iden-tified, mostly recommending improvements in training,and effective implementation mechanisms for the rec-ommended interventions exist. Therefore, TAA safetyproblems can be addressed, and the additional availablesafety of TAAs to address traditional causes of GA acci-dents can be realized as well.”

We’ll explore these findings in greater detail while com-menting on the aircraft themselves.

Technologically Advanced Aircraft Safety implications

Safety Implications

A new Cessna 182equipped with aGarmin G-1000.

6

The good news The MFD provides an unprecedented view of the envi-ronment in which the TAA pilot operates. Moving mapsprovide pilots with significantly increased positionalawareness with pinpoint GPS navigational accuracy.Map overlays include data-linked weather information,terrain databases, obstructions, airspace, and trafficlocations. Additional information includes communica-tions and navigation frequencies, airport data, andengine and systems status. Some systems even providedepictions of the wind-corrected range based on theremaining fuel. Such tools have tremendous potential toincrease GA safety.

Some newly designed TAA themselves, with higher wingloading and sleek aerodynamics, are faster than tradition-al light GA aircraft with similar power. Better systemsredundancy reduces the probability of single-point failure.

The new look has an undeniable appeal for the light GAindustry, which has seen lackluster sales for more than20 years. With progress invariably comes responsibilityon the part of designers, regulators, CFIs, and, mostimportantly, pilots to make sure that all the features,performance, and extra information available with TAA

actually translate into safer flight.

Achieving the potential benefits will depend ontraining, and, ultimately, on a continuing evo-

lution in equipmentdesign. GPS navigatorshave evolved fornearly two decades,and the present gen-

eration is far superiorto early models. We have every reason to

believe that it is only going to get better.

The challengeThe AOPA Air Safety Foundation hasidentified three characteristics of TAA

that are likely to have the most impacton the GA safety record.

The first is the different physical handling characteris-tics of some new design TAA. This is obvious, straight-forward, and will be relatively easy to manage.

The second is the widespread adoption of new pilotingtechniques—different from the traditional role of theGA pilot. This may prove a bit more difficult.

The third challenge is finding instructors and flightschools that are knowledgeable and experienced on thenew aircraft, although this will improve as more TAAsenter the fleet and more flight schools becomeequipped with appropriate simulation devices to assistin avionics training. Again, we emphasize the impor-tance of an appropriate level of simulation early in thetraining process. Several manufacturers have embarkedon ambitious programs to educate CFIs, and they arecommended for their efforts. A related training issue isto bring the “planning ahead” skills of lower-time pilotsup to speed as they transition from slower training air-craft to faster, sleeker designs.

Any experienced CFI is well aware of the extra instruc-tion required for pilots to think further ahead in a fasterairplane. If the aircraft is descending at 180 knots intothe terminal area, the pilot had better be thinking at 220knots. With TAA, the CFI must guide the pilot alongthe additional learning curve of new avionics and devel-opment of the skills to manage their workload.

The advantages of TAA are many, but realizing theirbenefits will require pilots to shift from a typical GApiloting approach.

The physical airplaneIncreased speed and unique handling characteristics ofnewly designed TAA have, without proper training, ledless experienced pilots into difficulty in takeoffs andlandings and in managing arrivals into the terminal area.Some of these aircraft handle differently than conven-tional aircraft, with different “sight pictures” in the take-off and landing phases of flight. Using the “old” tech-niques with a new design may lead to pilot-induced oscil-lations, loss of directional control, or an inadvertent stall.

When the Boeing 727 was introduced to the airlinecommunity in the early 1960s, there were a number ofaccidents until pilots and instructors figured out thequirks of the new design. Different does not mean bad,but the training challenges for some new TAAs exceedthose for pilots moving between many other classic air-craft. High-wing loadings on some of the new aircraftproduce blazing speeds and give a smoother ride inturbulence, but they also develop a higher sink ratewithout power during approach and landing. They typi-cally increase the required landing distance as well, soshort field airports that may have been safe for legacyaircraft should be carefully evaluated for newlydesigned aircraft suitability.

The wing, fuselage, and

empennage area of a Columbia 350 is

superimposed on aBeechcraft BonanzaA36. Proper training

is necessary to overcome

different handling characteristicsbetween some

TAA and conventional

aircraft.

7

New aircraft designs are also prone to “teething prob-lems” in the first few years after joining the fleet.Examples of this include problems with both Diamondand Cirrus aircraft with doors opening or separating inflight. The Cirrus has also experienced several brakefires because of improper taxi techniques, and in-flightinstrument malfunctions because of water in the pitot-static system. As these new designs mature, such prob-lems are eliminated through changes to production air-craft and retrofits to the existing fleet.

The advanced avionics are also prone to growing pains.Reliable datalink connections, hardware reliability, anduser interface issues have all been encountered in thefirst few years of TAA service. One item of concernhas been the ability of the Garmin G1000 aural warn-ings to override ATC communications. This can betricky if the pilot needs to coordinate with ATC to dealwith the source of the warning tones coming throughthe headset.

Since Wilbur and Orville, pilots have defined “goodpiloting” primarily as a set of eye-hand or stick and rud-der skills that result in predictable outcomes:

• Maintaining VY precisely during a climb.

• Maintaining altitude within 50 feet.

• Tracking a VOR/GPS course within one dot of thecenterline.

• Landing with the desired speed and attitude, and therate of descent perfectly arrested at the exact instantthe tires brush the concrete.

As part of this mindset, alertness to the physical envi-ronment is valued (“keep your eyes outside the windowfor traffic”), as is an almost Zen-like unity with the air-plane (“can’t you feel that little buffeting? It’s tellingyou it’s ready to stall.”).

“Physical airplane” pilots, which is to say most GA pilotswho trained before 1980, often carry a do-it-yourself atti-tude that regards assistance as an affront. Popular writ-ings by author Ernest K. Gann capture this way of think-ing, telling of early airline co-pilots who were often toldby their captains to shut up, watch, and keep their feetoff the furniture. Autopilots were scorned as unnecessaryand were often only available on the top end of light air-craft so it was largely a moot point.

This view of the pilot has changed completely in air-line and corporate cockpits. The pros have recognizedthat the hardware is far more reliable than thehumans overriding it. This certainly doesn’t mean anabdication of pilot-in-command (PIC) responsibilitybut rather an acceptance that the autopilot does abetter job of mechanical flying.

The automation, however, is incapable of programmingitself and at times will significantly complicate a basicflying task. GA pilots are just beginning to face thistransition.

Technologically Advanced Aircraft Safety implications

Cirrus SR22 GTS

8

The mental airplaneIn TAA, piloting moves from the “physical airplane”—the stick and rudder skills—to a more mental approach.Pilots who successfully adapt will enjoy these aircraftwhile gaining situational awareness, and those who don’twill find challenge, complexity, and probably someunsafe situations when they are distracted from the pri-mary task.

The early corporate and airline operators who installedadvanced avionics employed primarily “physical air-plane” pilots, and the transition to glass cost consider-ably more time and money than expected. While mostpilots were eventually successful in the move to the glasscockpits of Boeing 757/767 and Airbus equipment, somewere not and retired. Some senior pilots admitted theyremained anxious about the complexities of glass rightup to their last day.

The transition to the “mental airplane” means copingwith distractions from the additional information andlearning unfamiliar displays. This is the root cause ofthe additional transition time.

Among the casualties: a good see-and-avoid lookout forother aircraft. In airline and corporate cockpits, much ofthis is negated by having two professional pilots, havingtraffic alert and collision avoidance systems (TCAS), andspending much of the flight in positive controlled air-space (Class A). Most operators have an inside/outsidepolicy where one pilot is clearing visually while the otherdeals with the internal systems. That they operate inlargely “sanitized” airspace of Class A, B, and C also con-tributes to a different approach to collision avoidance.It’s worth noting that with the advent of TCAS there hasnot been a single GA vs. airliner or airliner vs. airlinercollision in U.S. airspace. Traffic awareness systems foundon many TAA provide some of this protection. But forthe single pilot, the attention must be appropriately split.There have been numerous Aviation Safety ReportingSystem (ASRS) reports on crew confusion or distractionstemming from the use of TAA or equipment that is typi-cally installed in TAA. Reports included missing assignedroutes, mis-programming approaches, mode confusion,and altitude busts because of distraction with the equip-ment. It should be pointed out that pilots have alwaysbeen susceptible to distraction, and many of these sameproblems are manifested in classic aircraft. IdenticalASRS reports continue today, and for the same reasons.

In spite of manufacturer claims, the avionics in TAAonly provide the POTENTIAL for better situationalawareness. The tremendous flexibility and amount ofdata made available to the pilot of modern aircraft hasequal ability to inform or distract. Which result takesplace is largely dependent on how the pilot flies the mentalairplane and manages his use of that information.

In the case of corporate and airline operations, thelandmark TAA-related accident that graphically definedthe potential dangers of automation and FMS occurredin Cali, Colombia, in 1995, when an American AirlinesBoeing 757 struck terrain at night after the crew mis-programmed its FMS. After that tragedy, the airlineschanged their procedures in how crews interacted withcockpit automation. There are lessons to be learnedfrom Cali for GA pilots to write a safer history for TAA.

Beyond workload: over-relianceA related safety issue concerns pilots who apparentlydevelop an unwarranted over–reliance on their avionicsand the aircraft, believing that the equipment will com-pensate fully for pilot shortcomings. This is perhapsmore related to human nature than to TAA itself andwas raised more than a decade ago after several acci-dents that occurred shortly after the Piper Malibu wasintroduced. At that time, FAA instituted a SpecialCertification Review that ultimately exonerated the air-craft, finding that the Malibu problems were largelyself-inflicted by pilots unfamiliar with operations in highaltitude environments. Many of the fatal accidentsoccurred after encounters with convective weather whileen route.

Some pilots did not understand that FL250, the Malibu’shighest operational altitude, was one of the worst levelsto penetrate a thunderstorm. Clearly, these pilotsbelieved that the aircraft, a fine piece of engineering,was capable of more than reality allowed. The early mar-keting materials did nothing to dispel that belief by tout-ing that when flying a Malibu one could fly above theweather. To Piper’s credit, that approach was changed.

Related to the over-reliance on hardware is the role ofaeronautical decision making, which is probably themost significant factor in the GA accident record ofhigh performance aircraft used for cross-country flight.The fact that the aircraft involved was a TAA appears tobe coincidental.

9

A SF’s GA Accident Database contains NTSBdata on virtually every accident involving GAaircraft in the United States from 1983 to the

present (fixed-wing, weighing 12,500 pounds or less),accounting for more than 50,000 records. Unfortunately,government information-gathering on those accidentsgenerally contains no clear markers that define TAAfrom non-TAA. For the future, ASF has requested thataccident investigators note the on-board avionics inaccident aircraft. This will allow a more precise determi-nation of what avionics are involved in what type ofaccidents.

It is possible, however, to identify those aircraft that weredelivered by the factory with glass cockpits. Using aircraftserial numbers and delivery dates from NTSB and manu-facturer data, ASF has analyzed accidents involving glass-cockpit GA aircraft from 2003 to 2006 and compared themto the overall GA accident record. This analysis uses similarmethodology to that in ASF’s annual Joseph T. Nall Report.

Comparing glass-cockpit TAA to allGA accidents Between 2003 and 2006, glass-cockpit TAA accountedfor 57 of the 3,783 total GA accidents. Eighteen of the792 total fatal accidents were in such aircraft. It isencouraging to note that while 2.8 percent of the GAfleet were TAA, the advanced aircraft were involved inonly 1.5 percent of the accidents.

The distribution of these accidents also provides severalinteresting comparisons (Figures 1 and 2). For bothtotal and fatal accidents, TAA have had fewer than halfas many takeoff/climb accidents as the overall GA fleet.One contributing factor for this improvement may bethe ability to display critical V-speeds directly on the air-speed indicator. This gives the pilot an instant picture ofthe current airspeed relative to that desired.

Glass-cockpit TAA have had NO fatal accidents related tofuel management. This is an important victory over along-time cause of GA aircraft accidents. Many TAAMFDs include a “range ring” that superimposes the air-

craft’s range with availablefuel over the map display or adigital readout of fuel remain-ing and range, which is calcu-lated based on current fuelflow and groundspeed.

Technologically Advanced Aircraft Accident history

TAA accident history

0% 5% 10% 15% 20% 25% 30% 35% 40% 45% 50% 55%

Fleet

TAA

Other

Landing

Maneuvering

Go-Around

Descent/Approach

Other Cruise

Weather

Fuel Management

Takeoff/Climb

Preflight/Taxi3.5% (2)3.6% (137)

3.5% (2)16.1% (608)

0.0% (0)11.7% (442)

15.8% (9)4.7% (178)

5.3% (3)1.9% (9)

5.3% (3)5.9% (225)

10.5% (6)4.2% (158)

1.8% (1)9.2% (347)

52.6% (30)39.8% (1506)

1.8% (1)3.0% (112)

Pilot-Related Accident Categories, TAA vs. Fleet—Total

0% 5% 10% 15% 20% 25% 30% 35% 40% 45%

Fleet

TAA

Other

Landing

Maneuvering

Go-Around

Descent/Approach

Other Cruise

Weather

Fuel Management

Takeoff/Climb

Preflight/Taxi0.0% (0)0.8% (6)

5.6% (1)13.8% (109)

0.0% (0)7.6% (60)

44.4% (8)16.4% (130)

16.7% (3)5.4% (43)

16.7% (3)16.0% (127)

11.1% (2)4.0% (32)

5.6% (1)24.1% (191)

0.0% (0)2.9% (23)

0.0% (0)9.0% (71)

Pilot-Related Accident Categories, TAA vs. Fleet—Fatal

Fig. 2

Fig. 1

10

Maneuvering accidents, a leading cause of fatalities inGA overall, have also been greatly reduced in TAA.During the period studied, 9.2 percent of all GAaccidents, and a troubling 24.1 percent of GA fatalaccidents, occurred during maneuvering flight. This

compares with 1.8 and 5.6 percent respectively forTAA. While the data do not clearly point to the rea-son for this improvement, it is speculated that higherlevels of transportation use of these aircraft could bea factor—i.e., pilots are flying to some place ratherthan spending so much time in the practice/local areaor traffic pattern where maneuvering accidents areprone to occur.

Despite the promising record for takeoff/climb, the acci-dents studied showed TAA have a higher percent of land-ing (52.6 percent vs. 39.8 percent) and go-around (10.5percent vs. 4.2 percent) accidents than the overall GA fleet.None of the glass cockpit landing accidents was fatal,however. With slick composite fuselages and wings,some new design TAAs can be difficult to slow down tothe desired approach speed, leading to porpoising dur-ing the flare or long landings. While trying to correctthe situation, or when initiating a go around, torquefrom the high-powered engine can lead to directionalcontrol problems and this has led to fatal accidents.

The area where TAA fared the worst was in weather relatedaccidents. These accounted for nearly half (44.4 percent)of glass-cockpit fatal accidents compared to 16.4 percentfor the GA fleet. There is still no way to determine howmany of these pilots had datalink weather available tothem. The news on weather accidents isn’t all bad, how-ever. Continued VFR flight into instrument meteorologi-cal conditions, while accounting for two-thirds (67.7 per-cent) of fatal GA fleet weather accidents, only accountfor a little over one-third (37.5 percent) of the fatal TAAweather crashes.

While the analysis of the NTSB accident reports does notprovide clear insights, there are several factors that couldcontribute to the high number of TAA weather crashes:

• As discussed above, TAAs are believed to have ahigher percentage of use in a transportation role,increasing their exposure to adverse weather com-pared to those whose primary use is for training.

• Unlike NEXRAD weather radar displays, METARsurface weather reports and most forecasts providedby datalink are typically presented on the MFD intext format. Lack of an easy-to-interpret graphicpresentation of nonradar weather data may negative-ly impact the pilot’s ability to get a clear mental pic-ture of overall weather conditions, and relate it to theroute being flown.

Accident 1 [ATL05FA034]December 9, 2004; Diamond DA40;Pelzer, South Carolina; Likely cause:Diverted attention to program newinstrument approach.

History of FlightNear the end of an IFR flight fromJacksonville, Florida, to Greenville,South Carolina, the CFI-rated pilot wasadvised by ATC that the weather wasbelow approach minimums and wasasked if he wanted to divert to his alternate airport. The pilot told thetower controller that he did not have an alternate filed. The tower con-troller advised the pilot that Donaldson Center Airport was nearby andasked the pilot if he would like to divert there. The pilot elected todivert to Donaldson and was given radar vectors for the final approachcourse for Runway 5. As the pilot maneuvered for the approach, theairplane descended below the minimum safe altitude (MSA) of 2,500feet, at which time the tower controller issued a low altitude warningwith no response from the pilot. Attempts to re-establish communica-tion with the pilot were unsuccessful.

Examination of the crash site revealed a damaged power line about 75feet above the ground and that the tops of four trees were also dam-aged. Airplane debris was scattered in an area 100 feet wide by 450feet long. No mechanical problems were reported by the pilot prior tothe accident, and post-accident examination of the wreckage failed todisclose a mechanical problem or component failure. Radar datashowed the airplane losing 600 feet of altitude in a period of 14 sec-onds before the airplane was lost on radar.

ASF CommentsThis accident appears to be a loss of altitude awareness leading todescent and striking of power lines and trees. TAA displays provideexcellent depictions of the flight path, desired course, and other dataon a map display. They are less helpful in providing a clear picture ofaircraft altitude compared to that desired. Altimeter “bugs” allow thepilot to set target altitudes, but not all pilots use them effectively. Inthis particular case, the pilot may have been reprogramming the navi-gation system for the newly assigned approach. Such a distractioncould result in loss of altitude awareness. Appropriate use of theautopilot is essential in these situations.

11

• Like traditional weather information sources, the pilotmust enable datalink weather displays. If they don’t“ask” for the weather, they don’t get it. Once a weath-er product is available in the cockpit, it is the pilot’sresponsibility to know how to interpret the informa-tion and integrate it with other weather information.

• A number of TAA accident pilots may have believedthat access to near real-time weather improved theirchances of dealing with adverse weather. ASF’sobservation is that reliance on the hardware, as previ-ously mentioned, must be accompanied by a muchstronger decision making regimen. When the decisionis made to go, that’s only the beginning of the ADMprocess and puts a significantly greater burden on thepilot to make the tough call to bail out or divert whenthe weather dictates.

There is one aspect that is impossible to measure thatmay mitigate this somewhat gloomy assessment. There isno way to know how many trips are successfully complet-ed in either TAA or classic aircraft. It is entirely possiblethat the trip completion ratio is higher with TAA thanwith classic aircraft but at this point that is speculative.We hope that a method will be devised to measure thisaspect of TAA to determine a better denominator formeasuring the actual weather accident rate.

Cirrus accidentsCirrus Design is the most successful manufacturer ofnew design TAA, as measured by delivered aircraft.They began deliveries of the SR20 in 1999 and nowhave several models, including a turbo-normalized ver-sion of the SR22. Through the end of 2006 they haddelivered more than 3,000 of the total 5,700 TAA. Tobetter understand TAA safety as it relates to the currentmarket leader, ASF analyzed glass-cockpit Cirrus acci-dents during the period from 2003 through 2006.

The Cirrus record shows improved safety versus the GAfleet for takeoff/climb, maneuvering, descent/approach,and fuel management. Like other TAA, fuel managementaccidents were entirely eliminated in glass-cockpit Cirrusduring the period studied. Fatal accidents followed trendssimilar to overall accidents (Figures 3 and 4).

Weather showed the largest negative difference whencomparing Cirrus accidents to the overall GA fleet, withnearly one-third (31 percent) of all Cirrus accidentsinvolving weather, compared to 4.7 percent for GA over-all. Weather proved to be uncommonly deadly in the

Technologically Advanced Aircraft Accident history

0% 5% 10% 15% 20% 25% 30% 35% 40% 45%

Fleet

Cirrus

Other

Landing

Maneuvering

Go-Around

Descent/Approach

Other Cruise

Weather

Fuel Management

Takeoff/Climb

Preflight/Taxi3.4% (1)3.6% (137)

3.4% (1)16.1% (608)

0.0% (0)11.7% (442)

31.0% (9)4.7% (178)

3.4% (1)1.9% (70)

3.4% (1)5.9% (225)

17.2% (5)4.2% (158)

0.0% (0)9.2% (347)

34.5% (10)39.8% (1506)

3.4% (1)3.0% (112)

Pilot-Related Accident Categories, Cirrus vs. Fleet—Total

0% 5% 10% 15% 20% 25% 30% 35% 40% 45% 50% 55% 60% 65%

Fleet

Cirrus

Other

Landing

Maneuvering

Go-Around

Descent/Approach

Other Cruise

Weather

Fuel Management

Takeoff/Climb

Preflight/Taxi0.0% (0)0.8% (6)

7.7% (1)13.8% (109)

0.0% (0)7.6% (60)

61.5% (8)16.4% (130)

7.7% (1)5.4% (43)

7.7% (1)16.0% (127)

15.4% (2)4.0% (32)

0.0% (0)24.1% (191)

0.0% (0)2.9% (23)

0.0% (0)9.0% (71)

Pilot-Related Accident Categories, Cirrus vs. Fleet—Fatal

Fig. 4

Fig. 3

12

Cirrus, accounting for nearly two-thirds (61.5 percent) offatal accidents. In the overall GA fleet, weather wasidentified as the cause in 16.4 percent of fatal accidents.

Go-arounds also proved troublesome in the Cirrus,accounting for 17.2 percent of all accidents and 15.4percent of fatals. This compares to 4.2 and 4.0 percentrespectively for the overall GA fleet. This problem maybe a result of higher wing loading combined with higherhorsepower engines.

Type of operationThe purpose of accident flights was also studied withsome interesting differences between GA and glass cock-pit accidents (Figure 7). While there were fewer (59.7 vs.67.5 percent) accidents when glass-cockpit aircraft wereflown for personal reasons, that difference was almostperfectly accounted for by the increase (13.4 vs. 3.5 per-cent) in business mishaps. Instructional flights alsoproved troublesome, accounting for 23.9 percent of theglass cockpit total, compared to only 15.1 percent of theoverall GA accidents. ASF’s experience in analyzing thesafety record of over a dozen different makes of aircraftis that the record largely reflects how the aircraft is usedrather than a fundamental flaw that was missed in the cer-tification process. In the case of Cirrus, this translates torelatively few takeoff accidents compared to the rest ofthe fleet and more cross-country accidents, often relatedto weather or terrain encounters. This is because the air-craft are used predominantly in transportation roles andnot in primary training where many takeoffs and land-ings are practiced. It is too soon to tell if Cirrus takeoffand landing accidents will increase on a percentage basisas they find their way into more primary training roles.

0%

5%

10%

15%

20%

25%

30%

35%

40%

45%

50%

55%

60%

Fleet

TAA

> 40

00

3501

-400

0

3001

-350

0

2501

-300

0

2001

-250

0

1501

-200

0

1001

-150

0

501-10

000-50

0

53

.7%

6.6

%5

.6%9

.3%

7.4

%

14

.8%

20

.4%

36

.6%

5.6

%2

0.9

%

3.4

%3

.7%

4.0

%3

.7%

2.6

%

0.0

% 1.9

%

0.0

%

Accident Rates by Hours of Experience, TAA vs. Fleet—Total

0%

5%

10%

15%

20%

25%

30%

35%

40%

45%

50%

Fleet

TAA

> 40

00

3501

-400

0

3001

-350

0

2501

-300

0

2001

-250

0

1501

-200

0

1001

-150

0

501-10

000-50

0

43

.8%

6.4

%0

.0%

10

.3%

12

.5%

17

.3%

31

.3%

29

.3%

0.0

%2

5.2

%

4.0

%1

2.5

%

3.1

%0

.0% 2.7

%

0.0

%1

.7%

0.0

%

Accident Rates by Hours of Experience, TAA vs. Fleet—Fatal

Type of OperationType of Operation % of % of % of

Flying TAA Fleet(2005) Accidents Accidents

Personal 49.4 59.7 67.5Instructional 18.4 23.9 15.1Aerial Application 5.1 0.0 5.5Business 15.1 13.4 3.5Executive/Corporate 4.3 1.5 0.4Positioning * 0.0 1.9Ferry * 0.0 0.5Other Work Use 0.5 0.0 1.3Aerial Observation 3.5 0.0 0.6Other/Unknown 3.7 1.5 3.7

* Included in Other/UnknownFig. 6

Fig. 5

Fig. 7

13

Comparing TAA accident pilots tonon-TAA accident pilotsPilot experience is another area of interest when exam-ining TAA safety (Figures 5 and 6, p. 12). When lookingat total time in all aircraft, pilots with 1,000 hours orfewer are more likely to experience a mishap in a glasscockpit aircraft than in a traditional GA aircraft. Fatalaccidents in TAA were more common for even moreexperienced pilots, with those logging 1,500 or fewerhours having over 85 percent of fatal TAA accidents,compared to 57 percent for the fleet.

Time in type was also problematic for the TAA pilot,with 300 hours in type or less accounting for more acci-dents in TAA than GA in general (Figures 9 and 10).This was even more exaggerated in fatal accidentswhere the TAA risk factor went up to 500 hours in type.

The proportion of accident pilots holding instrumentratings (Figure 8) was similar in overall TAA and GAaccidents, while a higher number of TAA fatal accidents(70.6 vs. 61.5 percent) involved instrument-rated pilots.This suggests that the transportation role of many TAAsmotivates a higher percentage of pilots to obtain aninstrument rating. It may also be related to the lowernumber of VFR into IMC accidents discussed above.

TAA and the parachute Some TAAs have added new features that did not existjust a few years ago. One such change is Cirrus Design’scomplete aircraft parachute. The chute should bedeployed when the pilot believes there is grave danger.

According to the Cirrus SR22 Pilot Operating Handbook,“The Cirrus Airframe Parachute System (CAPS) isdesigned to lower the aircraft and its passengers to theground in the event of a life threatening emergency.However, because CAPS deployment is expected toresult in damage to the airframe and, depending uponadverse external factors such as high deployment speed,low altitude, rough terrain or high wind conditions, mayresult in severe injury or death to the aircraft occupants,its use should not be taken lightly. Instead, possible

Technologically Advanced Aircraft Accident history

0%

10%

20%

30%

40%

50%

60%

70%

Fleet

TAA

> 10

00

901-10

00

801-90

0

701-80

0

601-70

0

501-60

0

401-50

0

301-40

0

201-30

0

101-20

0

0-10

0

63

.3%

4.1

%4

.9%

4.1

%8.3

%12

.2%

14

.7%

14

.3%

0.0

%1

.6%

2.2

%0

.0%

2.0

%1

2.4

%

44

.1%

2.2

%0

.0%3.1

%0

.0%4

.4%

0.0

% 2.1

%

Accident Rates by Time in Type TAA vs. Fleet—Total

0%

10%

20%

30%

40%

50%

60%

Fleet

TAA

> 10

00

901-10

00

801-90

0

701-80

0

601-70

0

501-60

0

401-50

0

301-40

0

201-30

0

101-20

0

0-10

0

45

.5%

9.1

%

3.5

%0

.0%

9.2

%1

8.2

%

16

.6%

27

.3%

0.0

%1

.4%

1.2

%0

.0%

0.0

%1

1.5

%

41

.8%

3.7

%0

.0%3

.5%

0.0

%5

.5%

0.0

% 2.1

%

Accident Rates by Time in Type, TAA vs. Fleet—Fatal% of Pilots with an Instrument RatingAircraft % of % of % of % of

Total Total Fatal FatalWith Without With Without

TAA 55.3 44.7 70.6 29.4Fleet 52.3 47.7 61.5 38.5

Fig. 9

Fig. 10

Fig. 8

14

CAPS activation scenarios should be well thought outand mentally practiced by every SR22 pilot.”

The POH goes on to describe the types of situations inwhich CAPS use would be appropriate. These include:• Mid-air collision• Structural failure• Loss of control• Landing in inhospitable terrain• Pilot incapacitation

The parachute has stimulated strong debate within GAabout whether the presence of such a potentially life-sav-ing tool encourages pilots to intentionally fly into situa-tions they would not normally attempt in more conven-tionally equipped aircraft. Whole-airframe parachuteswill likely be offered on other manufacturers’ products inthe future. They are already available as retrofits onCessna products and a wide variety of ultralight aircraft.Perhaps the parachute’s effect on pilot decision making isas irrelevant as equipping an aircraft with shoulder har-nesses. One never intends to use them but they are there

in the event of need, regardless of whether the pilot cre-ated the problem or was a victim of circumstance.

To date, there have been more than 10 reportedinstances of use or attempted use of the CAPS system inCirrus aircraft. Some resulted in situations in which thepilot’s decision making placed the flight in jeopardy, butuse of CAPS likely prevented a disastrous outcome.Other CAPS deployments resulted from mechanical orother nonpilot factors. In at least one case, use of CAPSwas attempted in a high-speed dive after a severe icingencounter but the chute separated from the aircraft dueto the high deployment loads well in excess of the maxi-mum designed deployment speed. Following are sum-maries of several CAPS-related accidents:

CAPS Deployment 1 [FTW03LA005]October 03, 2002; Cirrus SR22; Lewisville, Texas;Likely Cause: The improper reinstallation of theleft aileron by maintenance personnel.

History of FlightDuring cruise flight the left aileron separated from anattach point, and the pilot executed a forced landing to afield. Prior to the accident flight, the airplane underwentmaintenance for two outstanding service bulletins. Duringcompliance with one of the service bulletins, the leftaileron was removed and reinstalled. The pilot confirmedwith the service center personnel that the maintenance onthe airplane was completed. After departure the airplanewas level at 2,000 feet msl for approximately one minute,the pilot noticed that the airplane began “pulling” to theleft, and the left aileron was separated at one hinge attachpoint. The pilot then flew toward an unpopulated area,shut down the engine, and deployed the aircraft’s para-chute system. Subsequently, the airplane descended to theground with the aid of the parachute canopy and came torest upright in a field of mesquite trees.

Examination of the left aileron and the airframe aileronhinges revealed that the outboard aileron hinge bolt wasmissing, with no evidence of safety wire noted.According to maintenance manual procedures, the boltand washer hardware were to be safety wired.

ASF CommentsHere is an excellent example of the safety factor intend-ed by Cirrus Design through use of CAPS. The aircraftwas being operated properly, and the pilot made anexcellent choice to deploy the parachute when a flightcontrol malfunctioned after routine maintenance.

15

CAPS Deployment 2 [NYC05LA110]June 30, 2005; Cirrus SR22; Haverstraw, NewYork; Likely cause: Pilot incapacitation.

History of FlightAccording to the pilot, the airplane was in cruise flightat 3,000 feet, when the pilot suffered a seizure and lostconsciousness. When the pilot awakened, the airplanewas in a high-speed descent. In addition, he felt disori-ented and felt numbness in his right leg. The pilotrecovered from the descent at an altitude of about1,700 feet and elected to deploy the CAPS parachutesystem. The airplane descended under the parachuteand impacted in a river. The airplane sustained sub-stantial damage to the underside of the composite fuse-lage. The pilot sustained a fractured vertebra and wasable to exit the airplane before it sank. Subsequentmedical examinations on the pilot revealed the pres-ence of a brain tumor.

ASF CommentsThis is another example of the parachute saving a pilotwho likely would not have been able to get back on theground safely. Each year there are a few accidentsattributed to pilot incapacitation. To date, there havebeen two cases where CAPS has been used to changethe outcome of an incapacitation accident. Ironically,both cases involved a water landing under the chute.While the water landing poses challenges of its own, theparachute at least gives the occupants the opportunityto increase their odds of survival.

CAPS Deployment 3 [ATL06LA035] January 13, 2006; Cirrus SR22; Childersburg,Alabama; Likely cause: Loss of control due to air-frame icing.

History of FlightThe experienced CFI departed Birmingham, Alabama,bound for Orlando, Florida. The airplane was equippedwith datalink weather. The airplane was identified byradar and cleared to climb to 7,000 feet. It entered theclouds at 5,000 feet on autopilot and climbing at 120knots. Upon reaching 7,000 feet the airplane encoun-tered icing conditions. The pilot informed the con-troller that he would like to climb to 9,000 feet, whichwas approved. As the airplane reached the cloud topsin visual flight conditions at 8,000 feet the airplanebegan to buffet. The pilot looked at his airspeed indica-tor and it indicated 80 knots. The airplane stalled andentered a spin back into instrument flight conditions.

The pilot deployed the ballistic parachute system andinformed the air traffic controller of his actions. Theairplane descended under the parachute canopy into anarea of trees.

The NTSB determined that the probable cause ofthis accident was the pilot’s inadequate preflightplanning, failure to obtain a current weather briefing,and his decision to operate the airplane into a knownarea of icing.

Technologically Advanced Aircraft Accident history

Accident 2 [DEN06FA131]September 15, 2006; Cirrus SR20;Maybell, Colorado; Likely cause:Inadequate preflight planning.

History of FlightThe private/instrument pilot and one pas-senger were enroute from Tooele, Utah,to Lincoln, Nebraska. The pilot contactedair traffic control and stated he needed alower altitude, as he was encounteringicing conditions. Several altitude changeswere assigned. Ultimately the pilot was assigned a block altitude from12,000 feet to 13,000 feet. The pilot reported serious icing conditionsand the controller cleared the pilot to an altitude of 11,000 feet. Shortlythereafter, voice and radar communications with the airplane were lost.

The wreckage was located scattered over a 1.5 mile area betweenColorado and Wyoming. Evidence was consistent with a ground impactdeployment of the Cirrus’s parachute recovery system, resulting in theairplane being dragged by high winds. Examination of the airplane’ssystems revealed no anomalies. Thunderstorm activity existed alongthe route of flight along with severe icing and turbulence. The pilothad not obtained a full weather briefing prior to the flight.

ASF CommentsInadequate flight planning has long been a contributing factor in weath-er-related accidents. It is possible that this pilot believed he could relyon the onboard datalink capabilities of his advanced glass cockpit toprovide the weather information needed to safely complete the flight.MFDs have the ability to display a variety of weather products. Sinceicing is one of the most difficult hazardous conditions to report and fore-cast, this pilot may not have recognized that he was entering an areawith conditions favorable to the formation of airframe icing until it wastoo late. Once the pilot lost control of the iced-up plane, the whole air-plane parachute system could have been used to make a safe descent.It was not. The chute deployed due to impact forces, and high surfacewinds dragged the aircraft on the ground for more than 1.5 miles.

16

ASF CommentsWhile the first two CAPS examples saved the day in acase where the pilot was not at fault, this one is a differ-ent matter. Here the pilot clearly entered dangerousflight conditions because of his own errors and oversight.The parachute was used to save the lives of those onboard, and without the chute this would likely have beenfatal. This was an expensive lesson but not a fatal one.

CAPS Deployment 4 [LAX05FA088] February 06, 2005; Cirrus SR22; Norden, California;Likely cause: Attempted deployment with excessive airspeed.

History of FlightThe private pilot was enroute from Lake Tahoe,Nevada, to Oakland, California, on an IFR flight plan.The pilot received a preflight weather briefing, whichadvised that there were no pilot weather reports (pirep)for the intended route of flight, and that the freezinglevel in the Reno area was 6,000 feet with no precipita-tion. There were no valid SIGMETs or AIRMETs foricing conditions along the pilot’s route. The pilot filedhis IFR flight plan for 12,000 feet, but indicated hemight request 14,000 feet once airborne. After takeoff,the pilot contacted Oakland Center and requested toclimb to 16,000 feet to try to get above the clouds. Uponreaching 16,000, the pilot reported that he was still inthe clouds and asked about going lower. Soon after, thepilot advised ARTCC that if he could go up another 200to 300 feet, he could get above the clouds. ARTCCrequested clarification if the pilot wanted to go up ordown. The pilot responded that he would like to go upfirst to build up some airspeed. The pilot was clearedfor a block altitude between 16,000 to 17,000 feet.About two minutes later, the pilot transmitted that hewas “coming down” and that he was “icing up.” Hedeparted from controlled flight, entered an uncontrolleddescent, and hit the ground.

Following the examination of the parachute system,investigators determined the system was deployed outsideof the operating envelope of the system, which is 133knots indicated airspeed maximum. The airplane was alsoequipped with an Ice Protection System that, when acti-vated, supplied deicing fluid to the wings, tail, and pro-peller. The aircraft was not certified for flight into knownicing and the Pilot Operating Handbook reads that,“Flight into known icing conditions is prohibited.”

ASF CommentsThis is a case where the parachute could have made adifference if it had been used in time. Unofficial reportsindicated the parachute was deployed at an airspeed wellin excess of the airplane’s red line speed. The loads onthe chute caused it to fail without any appreciable effecton the airplane’s descent. Pilots of parachute-equippedaircraft must have a clear understanding of when theyshould elect to descend under the canopy. This is a deci-sion that can be practiced effectively during training.

Accident 3 [LAX05FA032]November 10, 2004; Piper PA-32R;Santa Barbara, California; Likely cause:Controlled flight into terrain.

History of FlightThis VFR flight ended when it struckrising terrain during level controlledcruise flight on a night cross-countryfrom Bakersfield, California, to SantaBarbara. After departure the pilotclimbed from 4,900 to 5,200 feet andrequested information from ATC aboutthe elevation of the clouds. He admitted that he “seems to be in a littlebit of clouds...sort of in and out.” The pilot continued climbing into clear-er conditions. The flight continued and the airplane tracked near thecenterline of Victor Airway 183. The pilot was familiar with the round-trip route between his Santa Barbara home-base airport and Bakersfield,and he had previously flown over the route. During the last few minutesof the radar-recorded flight, the pilot was cruising at about 6,500 feet,as indicated by the mode C altitude reporting transponder. The pilot wasreceiving radar flight following service from a controller at the LosAngeles Air Route Traffic Control Center. The controller observed the air-plane and was aware that the minimum en route altitude (MEA) for air-planes on instrument clearances along the airway was 9,000 feet. Thecontroller and the pilot had sectional aeronautical charts available foruse that depicted a 6,840-foot msl mountain peak along the flight route.The pilot’s course did not vary as he approached and impacted themountain during the dark nighttime flight. The controller did not issue aterrain-related safety alert, as required by FAA procedures.

ASF CommentsThe pilot may have been lulled into a state of complacency. Flying avery well equipped airplane in smooth weather over a familiar routecould have led him to omit important planning and en route monitoringthat would have avoided this accident. The encounter with clouds dur-ing climb out by the VFR-only pilot suggests that preflight planningmay have been inadequate. Striking terrain in level flight is indicativeof a serious loss of situational awareness. This accident is also areminder that even when a pilot is in contact with ATC, full responsibil-ity for safety of the flight remains with the pilot.

17

Both aircraft manufacturers and traditional train-ing providers have jumped on the TAA trainingbandwagon. As mentioned earlier, FBOs and avi-

ation colleges are all rapidly adding TAAs to their fleets.Various commercial providers and equipment manufac-turers provide products and services to meet the needfor specific training on TAA avionics.

A wide variety of seminars, online training programs,videos, and computer-based simulators are now avail-able for all popular avionics systems used in TAA.Manufacturers of full motion flight simulators, formerlyreserved for airline and high-end corporate flightdepartments, are introducing models specifically for theCirrus SR20 and SR22 aircraft.

SimTrain, the first such company, provides full motionvisual simulators at locations near Atlanta, Georgia, andon both the east and west coasts in Cirrus TrainingCenters. The training programs include a parachuteactivation scenario for the Cirrus Airframe ParachuteSystem to emphasize the decision-making process lead-ing to CAPs deployment.

Training requirements andsourcesWith the introduction of new design TAA, there wasconcern about pilots’ ability to handle aircraft thathave both state-of-the-art aerodynamics and avionics.The manufacturers of glass-cockpit TAA respondedto these concerns by offering factory-approved train-ing for both pilots and instructors. This solution tothe pilot qualification problem has been expensivebecause of the limited number of CFIs who haveacquired or maintained the rigorous qualificationsrequired by some manufacturers’ programs. The lackof affordable, widely available part task trainers foravionics is also problematic.

Early in the life of the glass-cockpit TAA, insurancecompanies expressed the unknown level of risk in theform of higher premiums and additional training andflight experience requirements. As loss experience withthese aircraft increases, coverage rates are beginning todecrease and permitted sources of training are becom-ing more numerous. This results in a reduction in thecost of owning or operating a TAA.

Technologically Advanced Aircraft Training for the glass age

Training for the glass age

Instrument trainingin a Cirrus (note theback-up instrumentsin front of the pilot,under the PFD).

18

A training sequenceIn ASF’s opinion, the best way to train pilots, eitherfrom the beginning (ab initio) or for transition to TAA,is to start learning the aircraft on the ground. That’snothing new.

1. System training and basic avionics should be donewith CD/DVD, part-task trainer, or online.According to our surveys, most pilots do not findprint media particularly helpful for advanced avion-ics systems. Too much interactivity is required tolearn effectively by just passively reading. Quick-tipcards with shortcuts, after the pilot has a basicgrasp, are appropriate. Much training can andshould take place long before the pilot shows up atthe training center or before starting with a CFI,especially as a transitioning pilot. Jeppesen hasteamed with Diamond and Cirrus to provide anonline learning program. Pilots can use the programeither prior to flight training or afterwards to rein-force the concepts.

2. The next level might be a part-task trainer that simu-lates the GPS navigator or PFD/MFD cockpit.Having the actual knob/switch configuration of themost complex part of the instrumentation and properreaction to all pilot inputs will go a long way topreparing the pilot for flight. Here is an area whereboth avionics manufacturers and training providershave typically fallen short in offering an inexpensiveway to actually practice with the equipment outsideof an aircraft. This is gradually changing, as trainingproviders understand what is needed to effectivelytrain pilots in the new environment. Some of theolder GPS units came with ground power suppliesand simulation software so pilots could practice byactually removing the unit from the aircraft and set-ting up at home or at the school. With glass cockpitsand large moving map displays this is clearly not fea-sible. Short of having a dedicated ground trainer, thenext best alternative is to plug the aircraft into aground power unit. The disadvantage is that both theaircraft and power must be available.

3. Ideally, the next step is a cockpit simulator or flight-training device. This may or may not have a visualsystem or motion but it duplicates all other aspects ofthe aircraft. Simulation has been proven very effec-tive in larger aircraft. With the advent of relativelylow cost visual systems and computers, the new sys-tems now typically cost less than half than the aircraftthey replicate and can be so effective in preparingpilots that we wonder why anyone would train fromthe beginning in the aircraft itself. Professional pilotscertainly don’t.

4. Finally, it’s time to go to the airplane. This doesn’tpreclude experiencing some basic physical airplanehandling and local flights before sim training is com-plete, but the full-fledged cross country VFR andIFR departures and arrivals should wait until thepilot has a solid grasp of the glass or MFD/GPSequipment. Too much training is currently done in theactual airplane, resulting in great inefficiencies andhigher risk situations because of pilot and instructor dis-tractions. These include midair collision risk, airspaceblunders, blown ATC clearances, possible loss of con-trol, and extended training time required in the aircraft.It may be entertaining for the CFI but is far fromoptimal for the pilot who is attempting to grasp thebasics of the avionics. As soon as the pilot has mas-tered the most basic handling and after havingdemonstrated proficiency with the avionics on the

Screen shot of

(bottom).Cirrus full motion

flight simulator by Fidelity FlightSimulation, Inc.

(far bottom).

Cirrus TransitionTraining lesson

19

ground, we recommend as much actual short, highworkload cross-country experience as possible.

Droning around the pattern practicing touch andgoes at slow speeds in aircraft with wide-rangingspeed operating envelopes does not prepare pilotsfor the critical transition phases of flight. Few pilotshave difficulty leveling off at pattern altitude, throt-tling back to pattern speed and performing thebefore-landing check while staying in the pattern.En route, at altitude, the workload and risk is alsolow. It is the airspeed/altitude transition that causesthe problem. Unless the pilot is very light on cross-country experience and dealing with weather, thetraining time is better spent in the high-workloadareas such as the departure/arrival phases whereproblems invariably arise with altitude, speed, andconfiguration changes. Heavy use of autopilot andappropriate division of attention is critical.

New pilots who have limited cross-country experi-ence— and by this we’d arbitrarily say several hun-dred hours on cross-country trips of more than 200miles—should fly with a mentor in actual weather.This seasoning process should not be rushed as thenew pilot develops an appropriate level of respect andknowledge that cross-country flying requires, regard-less of onboard hardware and software. It can take theform of the mentor not necessarily being on board,especially in the latter stages. The mentor is there toprovide guidance in the planning and decision to goor not go, just prior to departure.

How long should all this take? As always, it will dependon the pilot’s experience and the tools available. A newpilot could take five days or longer and for very lowtime pilots, particularly those who are transitioning tofaster TAA, a reasonable mentoring period is suggestedthat might extend over several months. Pilots should begradually introduced to the broad range of conditionsthat the aircraft will ultimately encounter.

An experienced and instrument-competent pilot withconsiderable high performance time—and a good graspof the avionics—might transition successfully in two orthree days. If they haven’t mastered the GPS navigator,expect to easily double the time to IFR proficiency.

One size certainly does not fit all, as convenient as thatmay be for the training schools, CFIs, or manufacturers.Each pilot will bring different strengths and weaknesses

that need to be addressed, and flight instructors shouldperform an assessment to specifically identify thoseweaknesses, and tailor the training accordingly. Aftertraining it is essential for all pilots to get out and practicewhat they’ve learned. Wait longer than one week to get backinto the aircraft or into a simulator and much of the reten-tion is gone without additional instruction. Considerablepractice is the only way that pilots will develop and retain ahigh skill level. This is more critical now than it has everbeen with the new complexity and capabilities that theseaircraft introduce. This can be done in conjunction withsupervised operating experience (mentoring), to work onoperational proficiency (for example, dense traffic areas).

A final point—the traditional method of spending a fewhours in ground school on aircraft systems and a cursoryreview of the avionics before hopping in the aircraft for afew hours of familiarization is now long outdated. Anytraining institution or CFI that attempts to do in-the-airtraining on advanced IFR GPS navigators, FMSs, or glasscockpit aircraft before having a thorough introduction

Technologically Advanced Aircraft Training for the glass age

Training in a 182.

20

and practice on the ground via simulator, ground pow-ered aircraft, or at the very least with computer basedinstruction, is just not performing in the best interests ofthe client.

Training a new breed of pilots?Some market analysts have theorized that a new breedof pilots may be emerging, one that represents a signifi-cant change in the pilot population. Many are thoughtto be successful business people who want aircraft strict-ly for personal and business transportation and are notnecessarily aviation enthusiasts. They view an airplane,like a car or a computer, as a business tool.

These people typically do not hang around airports forlong periods to pick up an hour or two of flight time.They are busy professionals who will not be satisfiedwith a VFR private pilot certificate and want to beunrestricted by weather. Consequently, they need toearn a private pilot certificate with an instrument ratingquickly and efficiently.

The traditional training approach needs modificationfor this customer. These people are focused on results,not the process to get there. This group may also placeunwarranted trust in technology to compensate fordeveloping skills and their inexperience. They may alsobe persistent and decisive in running a successful busi-ness. These are not traits that serve new pilots well.

There is little evidence to prove or disprove that newpilots are more focused on transportation flight asopposed to local recreation flight. It is logical, however,to think that pilots who buy aircraft capable of flight atmore than 150 knots might be interested in going some-where. There have always been the “fast burners” wholearned to fly in basic aircraft and within a year or twoupgraded to high-performance cross-country machines.

The traditional sequence is still followed by many pilots:Start in a basic trainer, upgrade to a slightly larger four-place aircraft, and spend several years getting cross-coun-try and instrument experience before making the jump toa high-performance aircraft. This allows seasoning andjudgment to take place in addition to formal training, afactor that some think is lacking with the fast burners.

We believe a split still exists, often dictated by personaleconomics. Those who have a need to travel and the finan-cial wherewithal will buy a high-performance aircraft. Andthose who previously followed a traditional approach toaircraft upgrading may now become fast burners becauseof some TAA system simplicity (fixed -gear, full authoritydigital engine controls, etc.) and attractive pricing.

There may also be a new group of pilots who enter thesystem through the sport pilot certificate. They will havelearned basic flight skills, but there will be a significanttransition into a full-fledged TAA and longer trips.Because the sport pilot certificate is so new, it is toosoon to tell how this will play out: A pilot tries out flyingand as he or she becomes financially able and desirousof more capable aircraft, moves from a very basic physi-cal airplane into a mostly mental one—the TAA. This isa big step but not insurmountable with the right trainingapproach and appropriate mentoring.

Autopilot essentialsFor single-pilot IFR operations in TAA, we believe thatautopilots are essential. All single-pilot jets require anautopilot and pilots are trained to rely on it right fromthe beginning.

While TAAs are simpler and slower than jets, the work-load is nearly the same. Since pilots operating TAAs arerequired to function more as programmers and man-agers, it only makes sense to delegate much of the phys-ical aircraft handling to a reliable piece of hardware.GA pilots need to view the autopilot as their second-in-command, and use it appropriately.

This is not how light-GA pilots have traditionally beentrained. The autopilot was considered ancillary rather thanessential. The airlines and corporate world left that conceptbehind decades ago, recognizing that a properly managedautopilot can reduce workload tremendously. First, the useof the autopilot must be considered as core to the opera-tion of TAA and pilots should be trained in its routineusage. The FARs require single-pilot IFR flights under Part135 to have a fully functional three-axis autopilot.

CheltonFlightSystem’s

autopilot.

21

Departures, en route operations, arrivals, and approach-es should be flown such that the pilot is comfortableand completely proficient. Some hand-flying training isnecessary in the event of an autopilot failure, but inmany cases hand flying is indicative of pilots who do nothave the requisite autopilot skills to properly managehigh workloads in single-pilot TAA.

Proper programming is critical. Mismanage themachine and the workload is increased well beyondnormal. Pilots must learn all the modes and their limi-tations. Confirm that the aircraft is doing what theyasked it to do—trust but verify—and how to react whenthe autopilot is, inevitably, misprogrammed. Learnfrom those mistakes to reduce their frequency in criti-cal situations.

Some potential problem areas include fighting theautopilot by holding onto the control yoke or side stick.Runaway trim is one example. The autopilot willmethodically trim against the pilot and will either winthe fight or disconnect with the aircraft badly out oftrim and very difficult to control. Pilots need to diag-nose an autopilot problem quickly and know how to dis-able both electric trim and autopilot quickly.

Some autopilots have a vertical speed mode selection.In our opinion, this capability is a potential trap, espe-cially in piston aircraft. In a few documented cases, ver-tical speed mode was selected—for example, at 700fpm—and as the aircraft climbed, the engine perform-ance declined with altitude. As the airspeed declined,the autopilot attempted to maintain the selected rateand caused the aircraft to stall. A better mode selectionwould be to use airspeed but that usually requires anair-data computer, which increases the cost and com-plexity of the system.

Malfunctions are rare, far less than with human pilots,and these must be handled appropriately. Malfunctionsare best practiced in a simulator where pilots can actual-ly experience the sensations and learn the properresponses. In actual IMC this will include advising ATCthat the flight has an abnormal situation. The concept ofan abnormal situation may be new to GA pilots, butsimple to understand. It is in between normal opera-tions and a full-blown emergency. The situation may notyet require drastic action, but if not handled properly, areal emergency could be imminent. When in an abnor-mal situation, ask for help. This might be nothing morethan insisting upon radar vectors to the final approach

course and no changes in routing. It may also be pru-dent to divert to an area of better weather, lower trafficdensity, or an easier instrument approach. It is not thetime to show just how good you might be. Studies haveshown that pilots persistently believe their skills to behigher than they actually are.

The FAA has recognized the realities of autopilot use inTAA and made appropriate modifications to theInstrument Practical Test Standards requiring demon-stration of autopilot skills as part of the InstrumentAirplane flight test.

Pilot performance and its effecton human factorsTAA accidents examined for this ASF report werelargely indistinguishable from accidents with non-TAAequipment. Would a more direct approach to humanfactors in GA accidents make sense? Some will refer tothis as the big brother approach to safety, since itinvolves using monitoring devices permanently installedin the aircraft to record flight operations.

Technologically Advanced Aircraft Training for the glass age

Accident 4 [SEA06CA187]September 22, 2006; Cessna 172;Naples, Florida; Likely cause: Studentlanding accident.

History of FlightThis student pilot, on his second soloflight at Naples, Florida, reported thathe had completed a practice landing onRunway 14 and was applying power inpreparation for another takeoff when theaircraft encountered a “wind gust” fromthe right. The pilot applied correctiverudder and aileron, but the airplane veered off the runway and struck aditch. The weather observation at Naples indicated that the wind wasfrom 140 degrees at eight knots.

ASF CommentsTAAs are entering the training fleet in increasing numbers, with theresult that more new pilots are learning to fly using the latest technol-ogy. This is an example of an accident that would have occurredregardless of the type of avionics installed. An inexperienced pilotencountered a situation that he couldn’t handle and lost control of theairplane. The difference between aircraft used primarily for transporta-tion and those used for training will have to be considered as TAAsafety is analyzed in the future.

22

The airlines have employed this technology, calledFlight Operations Quality Assurance (FOQA) for years.It allows airlines to periodically download data from theaircraft and to look for major anomalies from normalflight operations. This might include unstabilizedapproaches, improper use of flaps, poor speed and alti-tude control, etc.

British Airways has employed this approach for morethan a decade and claims that it has allowed them tocatch pilot performance problems and correct thembefore accidents or incidents occur. It is too early in the

transition to see how this approach might be applied toPart 91 operations or if it is cost-effective.

Tracking pilot performance and itseffect on trainingAs we transition into the glass age, it’s still essential tostudy accidents and mishaps to understand how theyoccurred and what can be done to prevent them. Thishas ramifications for aircraft design and, perhaps mostimportantly, for training. If we could reasonably andinexpensively capture what the aircraft and the pilotwere doing just prior to impact it would help distinguishbetween aircraft malfunctions, pilot judgment, and skillissues. That would help to improve training curricula,identify where a piece of equipment did not performproperly, or where poor pilot judgment was the culprit.

Highly sophisticated flight data recorders (FDRs) havebeen used in large corporate aircraft and airliners fordecades to track dozens of parameters regarding flightcontrol input, switch positions, aircraft configuration,attitude, altitude, engine parameters, and speed. TheFDR and companion cockpit voice recorders (CVR)have become essential in identifying the probable causeof heavy aircraft accidents. Their use in light aircraft hasbeen impractical due to very high cost, complexity, andweight constraints.

However, the digital data used for PFDs, MFDs, andnavigation in new and in newly-built classic TAA isstored and can be downloaded for analysis. In somecases, pilots can access such information to review theirown performance and that of their aircraft. The NTSBoccasionally uses such data during its accident investiga-tions, although in many cases the equipment isdestroyed due to fire, impact, or water intrusion.

Those concerned with privacy or “big brother” willobject to this approach to safety, since it involves usingmonitoring devices permanently installed in the aircraftto record flight operations.

Microprocessors in new aircraft engines and in enginemonitoring equipment have the ability to track how theengine is being flown. Engine monitoring has been success-fully and inexpensively retrofitted to many airplanes aftermanufacture. It guides both pilots and manufacturers inrunning engines more efficiently, is used in troubleshoot-ing, and is widely available for existing aircraft, althoughnot without some expense. Engine management has beengreatly simplified and improved with this equipment.

Accident 5 [NYC06FA072]February 22, 2006; Columbia 400;Stafford, Virginia; Cause: Descent belowminimums during instrument approach.

History of FlightThe private pilot was conducting an IFRflight between Winston-Salem, NorthCarolina, and Fredericksburg, Virginia.The pilot attempted a night GPS instru-ment approach, but executed a missedapproach. He subsequently requestedand flew an ILS approach to theStafford, Virginia, airport. Radar and transponder returns confirmedthe airplane flew the localizer course down to about 200 feet aboveground level (agl). Weather about the time of the accident includedcalm winds, 1.25 statute miles visibility, light drizzle, and an overcastceiling of 500 feet. The airplane’s wreckage was located in a woodedarea, about 300 yards left of the runway and three quarters of the waydown its 5,000-foot length. Tree cuts were consistent with the air-plane having been in a 30-degree left turn. The missed approach pro-cedure was to climb to 600 feet msl (400 feet agl), then make aclimbing left turn to 2,000 feet, direct to a VORTAC, and hold. Therewas no evidence of mechanical malfunction.

ASF CommentsThe evidence in this case is consistent with the pilot failing to estab-lish a positive climb while following the missed approach procedure.The Columbia 400 is representative of the new generation of slick,high-powered TAAs. When executing a missed approach, the applica-tion of power and subsequent need to trim for a climb could lead thepilot into a difficult situation if priorities are not firmly set. The oldmaxim of “aviate, navigate, communicate” is as valid for the TAA as itis in traditional aircraft. Training to maintain proficiency in challengingmaneuvers such as missed approaches in night instrument weatherconditions is also important.

23

Technologically Advanced Aircraft Training for the glass age

The automotive experienceThere is no doubt that human behavior changes whenparticipants know they are being watched and usually itimproves. When police use radar, laser, and cameradevices to monitor speed on the highways, drivers slowdown. To see how FDRs might affect GA, it’s instructiveto look at how event data recorders (EDRs) have affectedthe automobile industry. Automotive fleet studies haveshown that the installation of EDRs can reduce collisionsby 20 to 30 percent. Since 1990, General Motors hasequipped more than six million vehicles with the monitor-ing capability. Events commonly recorded by automotiveblack boxes include vehicle speed; brake and acceleratorpedal application forces; position of the transmissionselection lever; seatbelt usage; driver seat position; andairbag deployment data—very similar to FDRs. The datacollected belongs to owners except when requested bypolice or court order. Auto manufacturers also will use itas a company defense in a product liability lawsuit.

Some automakers are reluctant to use EDRs for fear ofhow the information will be used in court. GM, howev-er, believes that the potential for improvements in autosafety far outweigh any possible increase in litigationand in most cases, driver mishandling has caused theaccident, not the vehicle—exactly the same circum-stance as with aircraft. Here are some examples:

• Data from a black box caused jurors to question theprosecution’s argument that the driver was speedingrecklessly before a fatal head-on crash with anothervehicle. The driver was found not guilty after histruck’s black box showed 60 mph at impact—notabove 90 mph, as a witness had claimed.

• A police officer won a major settlement for severeinjuries he suffered when a hearse struck his squad car.The hearse driver claimed a medical condition causedhim to black out before he hit the police car. But thehearse’s black box showed the driver accelerated to63 mph—about 20 miles more than the posted limit—seconds before he approached the intersection, thenslammed his brakes one second before impact. The black-box information was an unbiased witness to the crash.

• After a high-profile crash that killed a former profootball player, the family filed a $30 million civil suitthat claimed the vehicle’s air bag deployed after thecar hit a pothole and that caused him to hit a tree.Data from the black box showed the air bag deployedon impact as designed, and the survivors lost the case.

Training, liability, and flight datarecordersSome large U.S. flight training institutions using TAAshave installed small digital cameras and flight datarecorders (FDRs) that allow fast, comprehensivereviews of training sessions on what actually occurred inthe cockpit or simulator. The electronics revolution ofthe last decade—which itself has helped make TAA pos-sible—offers small and relatively inexpensive digitaldevices ideally suited for this purpose. The fact thatthese are usually installed at the time of manufactureversus an expensive retrofit have made them an inex-pensive benefit in training. There’s nothing like seeingvideo or a flight path of a training scenario to guideinstructors and students. Olympic athletes, skiers,golfers, and swimmers all use monitoring to improveperformance.

One leading GA aircraft manufacturer has seen its air-frame liability insurance premiums triple in the past fewyears because of consumer legal action claiming defec-tive equipment. It is rumored to be considering someform of FDR in its new production models to reduce itsliability from speculative lawsuits and to improve theaircraft. For the builders of very light jets, several com-panies have mentioned that FDRs and CVRs might bea part of the package.

After many accidents, when lawsuits against manufactur-ers ask for millions in compensation, it is to everyone’sbenefit to see that the facts are presented unemotionallyand correctly. From the manufacturers’ standpoint,claims for maintenance and warranty service can often bemore fairly adjudicated with data from the devices.Historically, about 90 percent of the accidents investigat-ed by the NTSB show no design or manufacturing defect.

FDRs can also support the legitimate claimsof pilots, and in those cases where an aircraft or piece of equipment isshown to be defective or improperlymaintained, the manufacturer ormaintenance provider should settlethe claim fairly, and then quicklyresolve the technical or proceduralproblem for the rest of the fleet.The advent of new productionmodel TAA equipped with FDRsmay improve safety where productliability and tort reform advocateshave been unsuccessful.

Appareo GAU 1000flight recorder.

24

Autopilots have been around for nearlythe entire history of powered flight,starting with a primitive model inventedby Elmer Sperry just before World War I.Today, autopilots are essential on turbineaircraft, commonplace on GA cross-coun-try aircraft, and even showing up ontrainers. How should pilots use them andhow much should we depend upon them?

There are those who live by AP:wheels up, AP on; wheels down, AP off,or, if the ceiling is low, the pilot-engage-ment point becomes “runway in sight.” Ifthe gear is fixed, 500 feet above groundsubstitutes as the on-off point. Thosewho routinely fly this way are abdicatingtheir basic flying skills to the hardware.

The other extreme is the pilot whohand flies a four-hour trip even thoughautomation is available. He or she maybe highly proficient, but the wear andtear on the pilot is likely to be signifi-cant, depending on the physical condi-tion, age, and experience of themarathoner. Such a pilot’s flight pathmay be somewhat erratic, especially ifthe avionics require much attention, andflight management may occasionallybecome secondary to aircraft control.

When I started flying, back in thePleistocene era, autopilots were a rarityin light aircraft. New instrument pilotshad to keep all the balls in the air manu-ally, and while it may not have always

been graceful, it worked—mostly. Weweren’t trained in the use of automationand even when it was available, many ofus considered it a crutch.

After checking out in bigger, faster air-craft with more sophisticated avionics, Ilearned that flight management wasmore than just a term; it was a usefulconcept for negotiating high-density air-space and dealing with inevitablechanges, allowing the aircraft to performthe basic flight tasks, with supervision.The autopilot became an essentialcrewmember in the single-pilot cockpit.

As a new IFR pilot, I checked out in aPiper Arrow. The avionics packages inthose days were mostly basic and stan-dardized, and the instructor provided a cur-sory briefing on the autopilot. It was a sin-gle-axis device that held heading or track,but did not have an altitude-hold function.

On one of my first Arrow trips in theclouds, I was hand flying, minding myown business, and marveling at my abili-ty to survive. New York Center decidedthat things were going entirely toosmoothly, and gave me the obligatoryreroute. I copied the clearance, whichincluded intersection names known onlyto FAA planners, while attempting tohold straight and level, retune the VOR,reset the OBS, and fumble with thetransponder. On most tests, getting fourout of five elements isn’t bad and 80

percent is considered a passing grademost everywhere. This overlooks my Fgrade for altitude control. At some pointduring the breakdance the controllerpolitely reminded me to check thealtimeter setting, probably knowing fullwell what was going on. The autopilot,even without altitude hold, would havebeen a fabulous asset if I had used it,and the whole deal would have beenaccomplished far more elegantly.

We are required on the instrumentpractical test to handle the aircraft,reroute, reprogram, and stay withincheckride tolerances. In the real world,while dealing with turbulence, fatigue,passenger distractions, and myriad otheritems, the reality is that those toler-ances are sometimes stretched into“pink-slip” territory. It shouldn’t happen,but then IFR life isn’t exactly as por-trayed in training. Ask me sometime whythat’s nearly impossible to do effective-ly. Also understand that I’m not advocat-ing loosening the standards.

When Cessna introduced the Citationline of jets, the autopilot was integral tothe FAA’s single-pilot approval. Pilotswere taught that the autopilot flew theaircraft and it was to be used in all nor-mal circumstances. If the autopilot brokebefore takeoff, the flight was canceledand if it failed en route that was an abnor-mal procedure. The pilot was expected tobe able to handle the aircraft, but it wasappropriate and expected to ask for ATCassistance, if needed, and divert to thenearest suitable airport. The extra timeand mental processing power was used tomanage the avionics and stay at least 10miles ahead of the aircraft. Requiredautopilot use was a major attitude shiftfor the light-aircraft crowd, most of whoused autopilots periodically and didn’treally trust them. With the arrival of verylight jets, you can be certain that theautopilot will be a required piece of equip-ment and that pilots will be expected touse it religiously.

Safety PilotSafety PilotSafety PilotSafety PilotAutopilot supermenBy Bruce Landsberg

Reprinted from the October 2006 issue of AOPA Pilot.

A s we move into increasingly sophisticated aircraft with moredata, more glass, and more speed, what role should an autopilot(AP) play? You wouldn’t think that a great laborsaving device

like the autopilot could create such a diversity of opinions.

25

Technologically Advanced Aircraft Training for the glass age

Let me clarify something for the“what-if-it-fails” crowd. In searching theAOPA Air Safety Foundation’s accidentdatabase back to 1983, we were unableto find a single accident where NTSBconsidered autopilot failure as the proba-ble cause. That doesn’t mean that itcan’t happen or that you should give uppracticing hand-flying skills. But a shiftin attitude is appropriate as we transi-tion to complex avionics packages thatdeliver a far better flight-path productand situational awareness, but needmuch more programming and demandmore attention than the old ones.

Today’s autopilots are much more reli-able than the humans programmingthem. I concede that a few units built onFriday afternoons before a holiday maybe less trustworthy than a politician at a

PAC reception. But as a group, humansare far more likely than our electro-mechanical helpers to deviate from aheading, miss an altitude, blow througha final approach course, or wobble downthe localizer.

As the equipment has changed, so tooshould the testing and training for lightaircraft, mirroring the single-pilot jet thatour cockpits are now emulating. Howabout treating autopilot failure in actualinstrument conditions in the same wayas an engine malfunction on a multi-engine aircraft? When an engine acts upor fails, the mode of operation changes.We set priorities very carefully, adviseATC that we have a problem that maydevelop into an emergency, or declarethe emergency outright. A diversionshould be made to the nearest suitableairport, not necessarily the nearest air-

port, which may have a complex and dif-ficult approach. Vectors-to-final is asmart way to handle this.

I have played and inflicted “what-if”games on students, including scenarioswhen a controller’s radar is down, or aposition 300 miles from nowhere, and abogeyman that jumps out if you don’tkeep it all going perfectly. The tradition-al approach to training for an autopilotfailure is often to continue the trip as ifnothing had happened. Just suck it up,son, and, by the way, I’ve got a reroutefor you with a hold and a course reversalon a back-course approach with a doglegat the final approach fix. It’s a par four,and mind the sand trap. The properanswer in the real world is, “Unable—we’ve got an equipment failure and I’llneed vectors to the ILS at downtown

municipal.” If you’resinking, ask ATC forthe localizer frequen-cy, inbound course,and altitudes insequence so the work-load remains manage-able.

If you can reasonably handle a bitmore, that’s good, but in the real world,the idea is to manage risk and workload. In the minds of some instructors,pilots should just meekly accept all thestuff that is shoveled into the cockpitby the CFI or ATC, instead of acting aspilot in command. Here’s what’s badabout this 400-pound bench-pressapproach to training: In an actual situa-tion, pilots tend to react as they havebeen trained. If the autopilot dies, theymay revert to the Superman mode they

learned in training, even though most ofus aren’t Superman or Wonder Womansix months after training. In really nastyweather after a series of long workdaysand perhaps not flying quite as muchas we would have liked, it’s not goingto be the same as the environment cre-ated by the tough-as-nails coachomnipotently sitting in the right seat.To continue the sports metaphor, thisisn’t the time for the Hail-Mary pass—just get first downs until you get to therunway. If you can only bench-press100 pounds, train to where you mightget to 120.

So how do you balance being a goodenough hands-on pilot in command withintelligent use of an autopilot? I like tohand fly departures until about 5,000feet, let the machinery do the mindless

en route part, and, on at least everyother trip, hand fly the approach. Thiskeeps me conditioned, but I also go forweightlifting sessions and coachingevery six months.

Some flight schools are now buyingfull-glass-cockpit aircraft without autopi-lots to save money and perhaps to trainpseudo supermen and women. It’s afalse economy and premise. Either gowith the full package and learn how touse automation intelligently or stick tosteam gauges and basic avionics. Let’strain for the real world!

So how do you balance being a good

enough hands-on pilot in command with

intelligent use of the autopilot?

M odern integrated avionics systems use largeliquid crystal display (LCD) screens to displaydata to the pilot. The primary flight display

(PFD), as its name implies, provides the most importantinformation the pilot needs to operate the aircraft. Instreamlined format, the PFD shows: • Attitude• Airspeed• Altitude• Primary navigation data• Supporting data

Multifunction displays (MFD) come in a variety offorms and accept input from aircraft and datalinksources. MFD data can include:• Engine and systems status• Moving maps with airports, navigation aids, and way-

points• Approach, taxi, and navigations charts• Terrain and obstructions• Traffic avoidance• Datalinked weather including NEXRAD precipita-

tion, TAFs, and METARs• Airspace

Integrated avionicsAvidyne, Garmin, and Chelton are currently the leadingsuppliers of GA integrated avionics systems. In somecases they provide equipment and components forretrofit into legacy aircraft. While each manufacturertakes its own approach, the pilot interface is similar.

Integration means that most information about the airplane and its environment can be controlled and

displayed through a single system. The two main dis-plays can be configured to meet the pilot’s needs andpreferences. Useful information is brought up as it isneeded while less important material remains hidden—but available.

Common hardware components in integrated systemsallow the displays to be switched back and forth in theevent of equipment failures. Such reversionary capabili-ties greatly reduce the risk resulting from critical instru-ment failures. It also puts an increased burden on themanufacturers to ensure that single point or cascadingfailures do not catastrophically degrade safety. Utilitycan be adversely impacted where a component in anintegrated system results in a unable-to-fly conditionwhereas a noncritical instrument or system failure in alegacy aircraft is a minor inconvenience but not flight-canceling.

Primary flight display In general, the PFD replaces all six of the traditionalflight instruments, plus some. The “directional gyro”mimics the more sophisticated HSI (horizontal situationindicator) combined with a radio magnetic indicator(RMI). Recent advances also provide a capability rarelyavailable to light GA pilots—the flight director. Theflight director provides computed attitude commandsthat allow the pilot to hand fly the aircraft with the sameprecision as the autopilot, provided that the pilot reactsin a timely fashion to the flight director’s directions.

Weather displaysUntil TAA, anything approaching real-time display ofconvective weather in the cockpit was limited to aircraftwith onboard radar. Radar is the gold standard for tacti-cal avoidance of thunderstorms but is expensive, some-what fragile, and heavy. Smaller GA aircraft usuallymade do with lightning detection devices such as aStormscope or Strikefinder to mark the location of sus-pected turbulence, but they provided a display thatrequired considerable interpretation. It should be notedthat one doesn’t need glass to get datalinked weather, asit is available through the use of some excellent portabledevices than can be used aboard any aircraft.

In TAA, however, suppliers of datalinked weather aremaking major inroads and such displays may greatlyimprove utility for light GA. Weather graphics datalinkcan simplify in-flight decision making. Depending on air-

TAA hardware and software

Datalinked weatheris displayed on

a Garmin GMX200.

26

27

craft and pilot capability, the decision can be made todivert, delay, continue, or land ASAP. Likewise, the avail-ability of the latest TAFs, METARs, winds aloft, andother products allow both VFR and IFR pilots to monitorweather ahead and around them. There will be very fewexcuses for being surprised, however, pilots are still capa-ble of getting themselves into trouble, either by failing tounderstand the limitations of the product, or not knowinghow to correctly interpret the information provided.

Terrain awarenessIntegral to most new GPS navigator units these days isterrain and obstruction awareness, usually displayed onan MFD in a format using different colors to indicatedifferent elevations. Symbols show obstructions such astowers and buildings and their relative height. In somecases, the terrain shown near the aircraft will changecolor, based on the GPS-derived separation between theaircraft and the ground.

TAWS (terrain awareness warning system)While GPS mapping modules with integrated verticaldimensions (elevation data) displayed via different col-ors are becoming an expected part of new TAA displays,an extra feature designed to prevent perfectly good air-planes from smacking the ground while under control isbecoming popular. Terrain awareness warning system(TAWS) became mandatory on March 29, 2005, for allturboprop or jet aircraft with six or more passengerseats, including those operated under FAR Part 91.TAWS has emerged as a common component in theTAA cockpit as well.

TAWS evolved from radar altimeters, devices thatemitted a warning when terrain directly below the air-craft became closer than a preset value. The originaldevice, called a ground proximity warning system, orGPWS, used ground return radar to measure the alti-tude from the airplane to points directly below. Thedevices worked fairly well, and the rate of controlledflight into terrain (CFIT) accidents in the late 1960sand early 1970s was significantly reduced. But theradar altimeter GPWS units had a major shortcoming:altitude measurements and thus the warnings of poten-tial CFIT were unable to prevent fast-moving aircraftfrom striking rapidly rising terrain if the aircraft had ahigh rate of descent. The integration of GPS naviga-tion and terrain database technology allowed thedesign of equipment that computed aircraft position,groundspeed, altitude, and flight path to calculate a

dangerous closure rate or collision threat with terrainor obstacles, and provided a predictive warning. This isthe technology behind TAWS.

The five functions provided by TAWS units most com-monly installed in high-end general aviation TAAincludes the appropriate audio alert for:

• Reduced required terrain clearance or imminent ter-rain impact. This is the forward-looking terrain-alertfunction. This warning is generated when an aircraft isabove the altitude of upcoming terrain along the pro-jected flight path, but the projected terrain clearance isless than the required terrain clearance. The warningsdepend on the phase of flight, and whether the aircraftis in level or descending flight. There are sixty-secondand thirty-second warnings. Sixty-second aural warn-ing: “Caution, terrain; caution, terrain” (or “Terrainahead; terrain ahead”) and “Caution, obstacle; cau-tion, obstacle.” Thirty-second aural warning: “Whoop,whoop. Terrain, terrain; pull up, pull up!” or “Whoop,whoop. Terrain ahead, pull up; terrain ahead, pull up.”The “whoop, whoop” sweep tones are optional.

• Premature descent alert. This alerts the pilot ifthere’s a descent well below the normal approachglidepath on the final approach segment of an instru-ment approach procedure. Aural warning: “Too low,terrain!”

• Excessive descent rate. This is a carryover fromGPWS, and alerts you if the rate of descent is dan-gerously high compared to the aircraft’s height aboveterrain—and, for example, if flying level over risingterrain. Caution alert: “Sink rate!” Warning alert:“Whoop, whoop! Pull up!”

Technologically Advanced Aircraft Hardware and software

Terrain is displayedon a GarminGMX200.

28

• Negative climb rate or altitude loss after takeoff.Another GPWS function, this is to assure a positiveclimb rate after takeoff or a missed approach.Caution alert: “Don’t sink!” or “Too low, terrain!”

• The 500-foot “wake-up call.” This occurs wheneverterrain rises to within 500 feet of the aircraft, or whenthe aircraft descends within 500 feet of the nearestrunway threshold elevation during an approach tolanding. It’s intended as an aid to situational aware-ness, and doesn’t constitute a caution or warning.Call-out: “Five hundred.”

Airspace displaysMost current generation GPS navigators include air-space information in their databases. The pilot cansuperimpose graphic depictions of complex airspacesuch as Class B on the MFD maps and access relevantaltitude and communications information. Usingdatalink sources, temporary flight restrictions (TFR)can also be displayed.

Traffic avoidanceToday, many TAAs have the ability to display symbolsrepresenting other transponder-equipped aircraft ontheir MFD. This information allows the pilot to haveanother set of eyes to spot and avoid traffic. While thissystem is useful, there are future developments that willenhance its function.

AOPA has assisted the FAA in the testing and selectionof a system that promises not only weather datalink butalso collision avoidance, even in nonradar areas. As it isimplemented across the country in the coming years, itwill represent a dramatic departure from the traditional

full-time separation provided by ground-based air trafficcontrollers. It may also help push TAA more quicklyinto the realm of “free flight,” a new model for air traf-fic control now under FAA consideration as one possi-ble answer to over-saturation in the existing radar-basedATC system.

A three-year program called Capstone was designed toevaluate various avionics systems that could become animportant part of air traffic control within the NationalAirspace System. Most of the testing was conducted ina remote corner of Alaska, with GA aircraft serving asthe test vehicles. Why test in a remote corner ofAlaska, rather than a high-density area in the Lower48? The answer is that when Free Flight is fully imple-mented all participating aircraft are expected to be fullyequipped with appropriate avionics. Therefore, anyevaluation of Free Flight concepts becomes more real-istic as the percentage of equipped aircraft flying in thetest airspace increases.

In Bethel, Alaska, the FAA was aiming for nearly 100percent participation. Excluding the high-altitude air-line traffic and a few daily commuter flights, it’s esti-mated that there are fewer than 200 aircraft operatingwithin 100 miles of Bethel. Mainly, these are single-engine air-taxi “workhorses” such as Beavers,Caravans, and a host of smaller machines, down toCessna 180s, plus a handful of helicopters. These werethe Capstone participants. The FAA equipped 195 ofthese aircraft for the project, outfitting each with aGPS receiver, a color multifunction display, and anautomatic dependent surveillance-broadcast (ADS-B)transmitter/receiver.

The ADS-B equipment allows aircraft to broadcasttheir positions to each other—and to air traffic con-trollers on the ground—via special transceivers andground stations. By the same token, air traffic paintedon ground radar can be datalinked to aircraft dis-plays. So can Doppler and other weather radarimagery, as well as text messages such as ATC clear-ances and weather reports. Even e-mail messaging ispossible.

In the ideal world of the future, pilots and controllerswould see the same targets and the same informationon a single display. Pilots could see potentially con-flicting targets as far away as 100 nautical miles, andalter their courses and altitudes to avoid midair colli-sions. For more immediate traffic threats in heavily

Traffic is displayedon a Garmin

GMX200.

29

traveled airspace, ADS-B could work equally well,although ATC would issue traffic advisories, orTCAS-equipped airplanes could follow any traffic orresolution advisories issued by their own on-boardequipment.

Under the Free Flight proposal, aircraft would befree to fly more direct routes using GPS; pilots couldsee virtually all of the traffic around them, and domore to safely separate themselves; and ATC couldbe freed of much of their en route controlling work-load, letting controllers focus more on the efficientmanagement of the entire airspace system, and toconcentrate their energies on sequencing and separa-tion in terminal areas.

Engine/systems monitoringAnother area where the MFD excels is in helpingpilots to manage their engines. Some of the new instal-lations have FADEC (full authority digital engine con-trol), which allows the pilot to move only one powerlever, much like a turbine. There is no need to adjustpropeller or fuel mixture—it is all done automaticallycorrecting for ambient temperature and altitude. Goneare the concerns of detonation, temperature control,and fuel flow.

If a parameter moves into the “yellow” for whateverreason, unlike gauges of old where the pilot mustconstantly monitor a needle for a 1/8-inch move-ment, the MFD automatically advises the pilot thatsomething is out of tolerance before it becomes crit-ical. The equipment also monitors the engine’s over-all performance and is routinely downloaded duringmaintenance to allow technicians a quick look at theengine’s history. This holds great promise toincrease reliability. Even routine engine parameters,such as cylinder head temperatures, EGTs, carbure-tor temperatures, and duty cycles are now monitoredas an accepted part of TAA instrumentation. TAAinstrumentation often provides more data than mostpilots know what to do with so there is another needfor training.

Technology abused?All tools have the potential to be misused and newtools have the greatest risk because users have to learnthe limitations of those tools and the pitfalls that canoccur if those limitations are ignored. Much of the newtechnology aboard TAA falls into this category. A few,including some regulators, have suggested that because

something can be misused, it should be severelyrestricted or not developed at all. That logic wouldhave forestalled the development of aviation itself andthe installation of airborne weather radar or deicingsystems. Current statistics do not indicate any wide-spread systematic trends toward the misuse of theadvances of TAA. There have been, and will always be,some individual failures.

Some concerns• Weather datalink—There is some potential danger

for TAA pilots who mistakenly believe theirdatalinked radar images constitute true real-timeweather, such as the case with an onboard radar. Thetime lag between capture of the radar image and thedatalink display may be anywhere from five minutesto 20 minutes. In a very active thunderstorm situa-tion, a pilot attempting to navigate around cells usingold data could be in serious jeopardy. This hasalready happened on several occasions. Similar dan-gers exist with radar-equipped aircraft when a pilotgets too close to a cell. This has happened infre-quently in both airline and corporate flight. No onewould suggest that on-board radar be removedbecause it is occasionally misused. Rather, we identifythe incident or accident as an anomaly, publicize itfor educational purposes, and move forward.

• Terrain—As with weather graphics, there is potentialto misuse the terrain databases for scud running oran attempt to operate VFR in areas of IMC. A CirrusPOH Supplement warning states: “Do not use theTerrain Awareness Display for navigation of the air-craft. The TAWS is intended to serve as a situationalawareness tool only and may not provide the accuracyfidelity on which to solely base terrain or obstacleavoidance maneuvering decisions.” There was oneaccident in the Capstone project in Alaska where this

Technologically Advanced Aircraft Hardware and software

Engine monitoringdisplay on an AvidyneEntegra system.

30

happened. On balance, however, the value of know-ing that obstacles lie ahead dramatically lowered thenumber of Alaska accidents.

VFR into instrument conditions is a leading causefor weather accidents in all aircraft, TAA or legacy.A classic accident occurred in 2005 when a CirrusSR22 piloted by a 1,100-hour flight instructor andthe plane’s owner struck a mountain while scud run-ning up the Columbia River gorge at night. Friendsnoted that the pilot had done this sort of thing anumber of times before in the Cirrus. Even with thelatest avionics, including terrain awareness systemson a large MFD, this activity is as deadly as it hasalways been.

• Traffic avoidance—As mentioned earlier, pilots gen-erally can acquire targets visually faster with on-board avoidance systems. Airline and corporate sys-tems have worked very well to date. To be sure, thereare two pilots and they tend to operate in highly con-trolled environments. In the more open areas andsmaller nontowered airports there will be moretransponder-less traffic so pilots will have to continueto scan outside.

As the Cirrus POH supplement points out, “SkyWatchcan only detect aircraft that are equipped with operat-ing transponders. Traffic information…is provided asan aid in visually acquiring traffic. Pilots must maneu-ver the aircraft based only upon ATC guidance or pos-itive visual acquisition of conflicting traffic.”

• Engine/systems monitoring—The only negative thatwe can see is if the system fails. Cessna’s experiencewith fuel monitoring has been so positive that evenan occasional malfunction will not override the bene-fits derived from spotting problems sooner.

• Parachutes—A minor downside to aircraft para-chutes is that pilots may come to rely on them whenbetter decision making would have prevented themfrom getting into a bad situation in the first place.Several fatal accidents have occurred when pilotsmay have rationalized that the chute would savethem if problems got out of hand and then failed todeploy when needed with fatal results. The technicalsolution is to have an “auto-deploy” system whenthe aircraft senses itself in grave danger. That levelof machine intelligence is probably still a number ofyears off.

Piper Saratoga II TC

31

Technologically Advanced Aircraft Hardware and software

There is another downside to use of the parachute.If deployed over an area with surface high winds, itis possible that the parachute can drag the aircraftalong the ground after touchdown. This happenedafter a fatal accident near Maybell, Colorado, in2006. Evidence at the scene suggested groundimpact caused deployment of the parachute recov-ery system, resulting in fragmentation of the air-plane over a 1.5-mile area as it was dragged by highsurface winds.

In the final analysis, the benefits of whole airplaneparachutes—as described earlier in this report—faroutweigh the downsides.

• Integrated Systems—Modern integrated avionicssystems offer a high level of flexibility and allow thepilot to set up preferences that suit personal oper-ating style. In a rental environment, this could leadto pilots not knowing just what data is going to bedisplayed without a comprehensive inspection ofthe many setup pages on the MFD. Work by theavionics manufacturers to allow portable prefer-ences or to allow EASY access to a default page toreset to a basic simple configuration would over-come this issue.

Avionics maintenance and ownershipThe owners and operators of TAAs are finding thatmodern avionics change several maintenance aspects ofthese aircraft. First, not every avionics shop is trained orequipped to work on such systems, and even if they arethey often troubleshoot down to the line replaceableunit (LRU) level only, exchanging the malfunctioningunit for a functioning one. LRUs often can only beopened and repaired by the manufacturer. It should benoted that FAR 91.187 requires the pilot on an IFRflight plan to report loss of any navigation, approach, orcommunication equipment as soon as practical to ATC.It’s also a good idea to have the avionics technician fillout a Service Difficulty Report, or SDR, on any signifi-cant problem.

Software updates are another maintenance considera-tion. Pilot using GPS navigators are likely familiar withthe need to update the navigation database on a regularbasis. Like other computers, however, TAAs’ sophisti-cated computers and software are updated regularly toadd new features and correct errors. Occasionally, theseupdates also require hardware updates. Almost all new

technology goes through growing pains and it is no dif-ferent with TAA. Several MFDs have had multiple soft-ware updates and reconfigurations to address slowupdate rates, mislabeling, or outright failures. As withall computer equipment, upgrades and updates areprone to potential failures and it is critical for manufac-turers to advise pilots of problems and address themimmediately.

Accident 6 [IAD05FA032]January 15, 2005; Cirrus SR22; CoconutCreek, FL; Likely cause: Loss of controlbecause of avionics failure.

History of FlightThe commercial pilot departed from FortLauderdale, Florida, on a flight toNaples, Florida, to gain experience inIMC. Shortly thereafter, he misinterpret-ed a series of air traffic control instruc-tions to be for his airplane when theywere for another airplane. Callouts and responses by the pilot indicat-ed confusion, to the point where he stated, “I gotta get my act togeth-er here.” Less than one minute later, the pilot reported “avionics prob-lems,” and about 40 seconds after that, during his last transmission,he stated that he was “losin’ it.” The airplane subsequently descendednose-down, out of the clouds, and impacted a house and terrain. Theairplane was equipped with a primary flight display (PFD), as well asseparate backup instruments in case the display failed. The airplanehad approximately 98 hours of operation since being manufactured,and had a history of reported PFD problems. The pilot had previouslypracticed partial panel (no PFD) flight. The airplane was also equippedwith a parachute system, which was not deployed, nor was the autopi-lot engaged, despite over two minutes of significant altitude and head-ing deviations.

ASF CommentsThis accident suggests that the pilot was struggling with a flightinstrument problem and became increasingly disoriented and con-fused. Failure of a TAA glass cockpit display should not be a fatalproblem. These aircraft are equipped with traditional round dial flightinstruments that are available as backups. As has been the case fordecades, there is an alternate static source that can be selected inthe event of water or other obstructions in the system. As a lastresort, the Cirrus offers the pilot the use of the CAPS parachute sys-tem in the event of pilot incapacitation or impending loss of control.Thorough initial and recurrent training programs address complexemergency situations. In particular, flight simulators tailored to TAAallow the pilot to practice dealing with such emergencies effectively.

32

Technologically Advanced Aircraft Report conclusions

W hile TAA are moving GA forward, they stillshare many characteristics with older aircraft,at least at this point in the transition. The

penalties for poor judgment, misinterpretation, mispro-gramming, or clumsy flight-control handling remain thesame as they always have. Learning to fly TAA willchange the flight-training world, and it should paynoticeable dividends to all segments of the industry.

While current accident figures are generally comparableto classic single-engine aircraft, there are some causalfactors such as weather where TAA pilot decision mak-ing may create a higher risk factor than traditional air-craft. This is troubling in aircraft that provide unprece-

dented access to weather information in the cockpit. Inmultiple cases, parachute-equipped aircraft have cer-tainly saved lives, but in other cases, although available,pilots did not use them at all, or in time. While the trackrecord of that technology is still being written, there isevidence to show that even though a pilot may havemade a bad in-flight decision, the negative outcome wasmeasured in insurance dollars rather than lives.

In the end, these discussions are not so much about air-planes but about the people who operate them.Although the on-board technology and performance ofTAA are rapidly evolving, and despite the fact that thepilot-training industry is making a strong attempt to bet-ter integrate pilots with their aircraft, pilots, for themost part, have not changed. A VFR-rated TAA pilotwho departs into an area of deteriorating weather maywell have attempted the same trip had he been flying aclassic aircraft, or he may have been enticed by the

machine’s capabilities. Poor judgment will always bepoor judgment. Did the new TAA cause the ensuingaccident? Certainly not! As long as pilots are humanthey will continue to make mistakes.

It’s also about the environment in which they operate.Automobiles are not affected much by low ceilings orvisibilities, strong winds, or thunderstorms. They arelargely weather-tolerant machines. Light aircraft areaffected to a much greater degree by all of these phe-nomena and while changing the avionics may helpsomewhat by giving the pilot more information, it doesnot change the fundamental environment. A small TAAor a small legacy aircraft all share the same weaknesses.

Until we address those shortcomings, the advances willbe smaller than some marketers would have us believe.

New generations of autopilots might allow for fullauto-land capabilities in small GA aircraft. This mayallow a low-time IFR—or in an emergency, a VFRpilot—the opportunity to fly an approach to mini-mums. On-board systems may eventually function asthe equivalent of a senior instructor, able to offeradvice based upon the inputs of all aircraft system sen-sors combined with up-linked information from theground to form a forward-looking picture of what theaircraft is about to encounter.

TAA offer increased safety with added situationalawareness. But for pilots to avail themselves of theseimprovements, the key ingredient will remain a balancebetween training tied to experience and ever-improving,smarter technology and retention of basic piloting skills.

Report conclusions

“Get rid at the outset of the idea that the airplane is only an air-going sort of automobile. It isn’t.

It may sound like one and smell like one and it may have been interior-decorated to look like one;

but the difference is—it goes on wings.”

—Wolfgang LangewiescheFrom Stick and Rudder, originally published in 1944

Publisher

Bruce Landsberg

Executive Director

Writer

Neil C. Krey

Editor, Statistician

Kristen Hummel

Database Manager

Editors

Bruce Landsberg

Executive Director

David Wright

V.P., Operations

Steve Harris

Chief Pilot

J. J. Greenway

Chief Flight Instructor

Design and Production

Michael Wescott

Graphics Manager

Angie Ebersole

Associate Art Director

Becky Richter

Production Coordinator

Mike Fizer

Senior Photographer

AOPA Air Safety Foundation

421 Aviation Way

Frederick, MD 21701

800/638-3101

asf@aopa.org

© Copyright 2007 AOPA Air Safety Foundation

Technologically Advanced AircraftSafety and Training

AOPA Air Safety Foundation wishes to express its deepest gratitude tothe Trustees of the Emil Buehler Trust for their support of the AOPA ASFAccident Database, GA's most authoritative leader in data analysis.

AOPA AIR SAFETY FOUNDATION421 AVIATION WAY

FREDERICK, MARYLAND 21701WWW.ASF.ORG