Put a dozen pilots in a room, and odds are youll hear a series of shocking stories of derring do as well as a parade of tales about other peoples mistakes. Put 200 in a room, and you might see something different.
The National Transportation Safety Board hosted a symposium in late September that brought together accident investigators, regulators, flight training experts, mechanics and manufacturers, among others, to discuss ways to make general aviation safer. The NTSB billed it as the first-ever General Aviation Accident Prevention Symposium. For several days, engineers, mechanics, human factors experts and flight training specialists examined a variety of common general aviation accidents. The goal: to pool resources and insights to try to figure out innovative ways of enhancing the safety of light planes. Although the group did not attempt to generate a specific list of initiatives, some potential solutions did emerge.
We were purposefully vague with our instructions to the presenters, because we wanted to try to get their ideas, says Dennis Jones, chief of regional operations and the general aviation division of the NTSB. We wanted to see what they thought was important, which was better than us trying to direct their efforts.
Perhaps one of the most prevalent views to come from the symposium was the necessity to shift the emphasis during flight training and flight reviews away from demonstrating physical skills and toward developing methods of making sound aeronautical judgments.
In generating the focus of the symposium, NTSB accident investigators selected accidents they considered both interesting and representative of the kinds of accidents common in general aviation. A panel of experts then examined in detail one accident of each type and came up with ways similar accidents might be avoided in the future. Most of the panelists used the accidents presented as a springboard to address the entire class of similar accidents.
The accident at hand involved a Cessna 205 used as a parachute jump plane. As one jumper was preparing to jump, the airplane stalled at 3,500 feet and spun to the ground. The jumper who was getting ready to go parachuted to safety. The other six occupants were killed.
The logbook of the 290-hour commercial pilot showed no entries reflecting spin training, although it did show he had ground instruction in spin recovery. The pilots roommate said he talked about performing a spin in an airplane one time.
In analyzing the accident, renowned spin instructor Rich Stowell pointed out that, when the Cessna 205 was certified, it took company test pilots using non-standard techniques to recover from the spin in one turn. Certification requirements under Part 23 stipulate that standard recovery techniques must work after one turn to lead to recovery within one additional turn.
Stowell suggested pilots pay more attention to low-speed phases of flight, such as airport traffic patterns. Estimates are that pilots spend about six percent of their flying time around airports, yet 57 percent of accidents happen there.
Spin training, he adds, does more than just demonstrate how to recover from an inadvertent spin, because traffic pattern spins generally happen too low for recovery anyway. More important, he says, is that pilots practice slow flight and stall recoveries more frequently to get people to understand the dynamics of handling airplanes at slow speeds.
Technological solutions were also on the agenda. FAA flight test engineer Lowell Foster presented research that showed that, of the 1,771 stall/spin accidents over a 20-year period, 93 percent happened at or below pattern altitude – too low for recovery. As a result, Lowell advocated applying technology to the problem.
Recent airfoil research shows that its possible to design wings that are more tolerant of stalls without suffering landing performance penalties. Such wings might incorporate leading edge cuffs, stall strips, vortex generators and stall fences. The new Australian-built Eagle 150, for example, includes all of these devices. The new Lancair and Cirrus airplanes also feature stall-resistant airfoils.
In NASA flight tests, a Grumman AA-1, a Beech Musketeer and Piper Cherokee were modified with an outboard cuff that made the outboard leading edge droop. The modified airplanes entered spins only 5 percent of the time, and then required prolonged aggravated control inputs or out-of-cg loading. The basic aircraft entered spins between 88 percent and 98 percent of the time.
Other potential modifications include separate stall warning devices for the clean and landing configurations and limiting elevator travel to moderate the angles of attack.
A combination of training and technology could drastically reduce the threat of stall/spin accidents. While no one advocated a return to mandatory spin training, most agreed that exploring the low-speed envelope with a qualified instructor would make such accidents less likely.
VFR into IMC
One of the biggest killers of general aviation pilots comes from flying VFR into IMC, even if the pilot is IFR qualified and the airplane is capable. The accident, outlined more completely in Accident Probe on page 19, involved a Piper Navajo Chieftain that flew into a 1,600-foot hill at 1,500 feet while scud-running on a 52-mile flight.
The biggest questions raised were why a 4,300-hour ATP in a lightly loaded Part 135-capable twin would drive into the side of a mountain. Aviation consultant Rudy Kapustin said that, while VFR into IMC accidents have been declining fairly steadily over the past two decades, its uncertain why. Improved training, better weather information and the increased situational awareness brought about by such things as moving map GPS units all might get some of the credit.
Even so, pilots who get into VFR-into-IMC trouble repeatedly follow the same pattern.
They start the accident chain by preparing poorly for the upcoming flight. In many cases they skip weather briefings. Sometimes they stray into unforecast bad weather. Real or perceived pressures from family, business associates, bosses or passengers induce some to continue when they should turn around. Finally, the pilot, intent on showing he has the proverbial right stuff, attempts to press on even as the situation deteriorates around him or her.
Kapustin and John King of King Schools reiterated that judgment is a subject given short shrift in flight training.
King said a big problem he sees is that pilots are misled into thinking flying is safer than it actually is. Flight training and biennial flight reviews are focused on the skill with which the pilot can maneuver the airplane rather than the judgment required to plan and execute flights safely.
King proposed adding risk management to the bag of tricks pilots carry with them every time they buckle into the cockpit. It was a refrain echoed throughout the two-day conference.
We are in denial, King said. We tell people that the most dangerous part of their trip is the drive to the airport, and they believe it. Instead, we have to tell them that there is risk in general aviation and that their most important job is risk management.
King outlined the competition between minimizing risk and completing the flight, saying that the pilot who cannot properly assess risk will be more likely to press on to the destination.
Aviation Safety editor Ken Ibold added that the fear of an FAA violation can lead some pilots into trouble – particularly commercial pilots who hope to move into an airline position. Regulators need to make it clear that they will not issue violations to pilots for correcting their mistakes unless the offense is particularly egregious.
While judgment may be hard to teach, the panelists agreed that its often the reason experienced pilots come to grief. In fact, about 80 to 85 percent of general aviation accidents stem from pilot error – with errors in judgment leading the way.
No matter how carefully pilots conduct their preflights or how involved in maintenance they get, unless they do all their own work they have to trust their mechanics to some extent.
Driving the discussion was a 1999 accident in which the propeller came off of Sen. James Inhofes Grumman Tiger as he cruised at 2,500 feet. Inhofe glided to an airfield, landed off the runway and the airplane nosed over. Investigators determined that a propeller installed under an STC had been installed incorrectly, leading to the propeller bolts fracturing.
The propeller shop that owned the STC said as many as 40 percent of the Grumman Tigers that went through his shop had the propellers installed incorrectly because of what may be a design flaw in the bushings that align the propeller assembly.
Although the design of the engine and the propeller assembly made installation more difficult, its clear that other factors came into play, said Douglas Macnair, vice president of government and technical affairs for the Professional Aviation Maintenance Association.
Although the maintenance shop knew of many deficiencies, it made no effort to spread the word through the FAAs Service Difficulty Reporting System or to encourage the FAA to issue an Airworthiness Directive. Several other examples of the same installation error were found by investigators. All illustrated the difficulty in installing the propeller correctly, especially if working alone.
The difficulty was compounded by the installation instructions, which did not make it clear just how sensitive this apparently straightforward installation was. In addition, the same economic problem that is decimating the ranks of flight instructors is also working on aviation mechanics, which is diluting the pool of knowledge of how to do unusual repairs.
Panelists also pointed out that many STCs contain installation instructions that do not follow the airplane from owner to owner. Even if they do, some owners keep them with the airplane, some with the engine and airframe logbooks – which doesnt help much if you have a problem away from home and your logs are at your mechanics shop.
In addition, mechanics have a tendency to ignore maintenance instructions for many tasks, contending the proper way is obvious.
As the airlines scramble for more mechanics, the ranks of general aviation mechanics will continue to be thinned. This, the panel concluded, represents a real threat to the continued airworthiness of aging general aviation airplanes.
Loss of Control
Losing control while in flight obviously isnt something pilots intend to do, yet it happens to experienced general aviation pilots as well as inexperienced ones.
The accident that started the discussion was the Linden, N.J. crash of a Bonanza piloted by an experienced pilot who taught partial panel in Bonanzas for the American Bonanza Society. (See Accident Probe, October)Author and aviation writer Dick Collins of Flying magazine pointed to the need for currency and proficiency, and noted that even though the accident pilot appeared to be both, the pilots use of a medication may have been the performance-inhibiting factor that led to the accident.
The panelists used the specifics of the case to examine flying while under medication, both because of side effects the medication might have as well as physical problems associated with the underlying medical condition.
Over the counter antihistamines and antidepressants were the two drugs most commonly linked to airplane crashes, reported Dennis Canfield, manager of the Toxicology and Accident Research Lab of the Civil Aeromedical Institute. More than 100 accidents have been linked to antihistamines in the past five years and more than 67 accidents have been linked to antidepressants. Marijuana is the most common illegal drug, with 43 cases reported between 1994 and 1998.
Canfield went on to say, however, that the real danger in flying while using unapproved drugs comes when something goes wrong. Although reaction times are slowed, coordination suffers and mental processes break down in all cases, the trouble only becomes critical if unexpected trouble arises.
Dealing with the unexpected means using all of the meager resources available in a general aviation cockpit, said Martha King of King Schools.
When flying partial panel, for example, pilots are discouraged from using the autopilot (which may still be available, depending on the installation and the particular failure) and the GPS, King said. Both can help ease the workload that makes partial panel flying so dangerous.
A GPS ground track, for example, can be used instead of the magnetic compass if the directional gyro is inoperative. If the GPS has a moving map that shows desired track, heading information becomes almost irrelevant – as long as the pilot knows how to use the equipment and takes the time to set it up correctly.
King said her company has taken the controversial position in its videos that, if the airplane is equipped with an approach-capable GPS, the pilot should shoot a GPS approach if flying partial panel in IFR conditions. Such a plan allows the pilot to program and brief the approach at altitude, then fly a much simpler approach than the typical VOR approach generally taught.
Pilots also need to know how the autopilot works so theyll know if its available if the vacuum system fails or one of the flight instruments goes belly up. King suggested making operation of the autopilot mandatory on general aviation check rides, as it is for airline operations, to decrease the workload on the pilot during a critical time.
Although the sky is sometimes a lonely place to be, other airplanes can appear suddenly and without warning. Somewhere in the back of most pilots heads is a fear of mid-air collision – and the place where thats the most likely is near an airport.
A Cessna 172 and a Beech A-36 Bonanza collided about three miles south of Meigs Airport in Chicago in 1997. The Cessna was on a sightseeing flight and the Bonanza was inbound for Meigs from an airport in Indiana. All seven aboard the two airplanes were killed.
After some maneuvering, the Cessna was heading south. The Bonanza was heading north. Analysis showed the pilots would have had about 25 seconds to see each other, and it takes about 12.5 seconds to evaluate a threat and take evasive action.
An analysis of Aviation Safety Reporting System data by ASRS program manager Vince Mellone revealed 96 reported near mid-air collisions involving two general aviation aircraft near an airport with a control tower. The vast majority, 96 percent, occurred in VMC and during the day.
The most frequent resolution was that the pilot detected the near collision after it was too late to react or the threat had passed. The next most-frequent resolution was that controllers saved the day, followed by the crew seeing the threat and avoiding it.
In the accident under discussion, it was clear that a miscommunication between the Cessna pilot and the controller played a role in the collision. The pilot missed reporting its position abeam the airport as it passed south and the controller didnt follow up.
The investigation also pointed out some of the deficiencies of the contract control tower in place at the time. Training was brief, the DBRITE radar equipment for the facility had never been installed and the controller generally worked alone. In fact, after the collision the controller had to stay on duty for another four hours until she could contact someone to relieve her.
Other sessions stemming from a helicopter accident and a runway overrun by a business jet reinforced the notion that determining how to instill better judgment and decision-making skills would go a long way toward reducing the risk of flying general aviation airplanes. Technology also has its place, but the cost-sensitive market and the long product lives mean technological developments made now will take years to work their way into the industry in meaningful quantities.
In the meantime, pilots were urged to try harder, true, but also to demand training that reflects the real world of flying instead of just the mechanics of operating an airplane. The real test will come in developing new strategies and then finding ways to put them to work.
Also With This Article
Click here to view “Presenting Their Views.”