December 2005 Volume 48 / Number 12
Landmark Accidents: Down and Locked
A minor malfunction turns into a major accident — and the death of a U.S. senator
Pilots, like doctors, should take the Hippocratic oath: When faced with an emergency, don't make the situation worse — do no harm. It sometimes happens that we turn a minor mishap into a major accident. Breathless news reporting not withstanding, a landing-gear malfunction on a retractable-gear aircraft should be no more than a belly slide, some skin and prop damage, and possibly an engine teardown. Not to worry — local crews will have the runway cleared in a very short time. Heroic attempts to preserve the aircraft are sometimes successful, but also they can have horrendous results. Some accidents also capture the public's attention, as this one did with flaming wreckage falling into a schoolyard and the death of U.S. Sen. John Heinz of Pennsylvania.
On April 4, 1991, a Piper Aerostar departed the Williamsport-Lycoming County Airport, Williamsport, Pennsylvania, at 10:22 a.m. Eastern Standard Time for an IFR flight to Philadelphia International Airport, Philadelphia. The flight was operated as a charter with a captain, copilot, and one passenger (Heinz) on board. The weather was VFR and the flight was cleared for an ILS approach to Runway 17 just shortly after noon. While on the approach the captain reported that the nose landing-gear light had failed to illuminate. This could indicate anything from a burned-out bulb to a hydraulic leak or an electrical problem.
Help is on the way
Just as the Aerostar began its approach, a Bell 412 helicopter departed the company helipad at Philadelphia for a local VFR flight with two seasoned pilots aboard. The pilots heard the communications regarding the unsafe nose-gear indication. The Aerostar was told to maintain 1,500 feet to allow the helicopter to pass underneath. As the Bell passed under the Aerostar one of the Bell pilots said, "Looks like the gear is down."
The Aerostar pilot acknowledged the helicopter's transmission and replied, "I can tell it's down but I don't know if it's locked...." The pilot may have been referring to a reflection of the nose landing gear that could be seen from the cockpit on the propeller spinner. The tower acknowledged, advised that the helicopter was no longer a factor, and cleared the Aerostar to land on Runway 17.
The controller then declared an emergency and the tower supervisor alerted the airport's rescue and firefighting units. Runway 17 arrivals were terminated, and a clear communications channel was established. So far, everything was proceeding toward an uneventful outcome. At 12:03:35, the controller offered the Aerostar the option of making a low pass by the tower to observe the position of the nose gear, which the pilot accepted. The tower observed that the gear appeared to be down. Tower flybys do no harm but they don't help much either. This typical friendly gesture usually won't resolve anything because the gear is either down and locked or it isn't. In nearly every circumstance the tower will not be able to make a positive determination one way or the other. The pilot's options remain exactly the same as before — treat the nose gear as unlocked, land the aircraft with as much weight shifted as far aft as possible, and hold the nose off for as long as possible. If it's a main gear, the procedure is to hold the affected wing aloft as long as possible and anticipate that there will be a fairly sudden turn to the failed side.
The controller asked the Aerostar to enter a downwind leg for Runway 17 and advised that the Bell helicopter was in a position to take a further look at the nose gear, if the Aerostar pilot approved. (Previous comments about the value of this exercise apply, except that the risk level was about to go up significantly.) The controller provided vectors until the pilots saw each other and joined up in formation.
The aircraft were now on an extended downwind leg for Runway 17 at about 1,100 feet. The controller advised the flight of an antenna farm about six miles ahead and suggested a turn back toward the airport or a heading change. At 12:08:52, the Bell first officer told the Aerostar pilot, "We're going to come up behind you on your left side, so just hold your heading." The Aerostar pilot responded that the antennas were ahead and he might need to change heading by 15 degrees to the left.
At 12:09:30 the helicopter pilot said, "Aerostar, we're gonna pass around your right side now; take a look at everything as we go by." The Aerostar acknowledged. Around 12:10:00 the Aerostar pilot again stated that the gear did not appear to be down and locked. At 12:10:16, the Bell pilot stated, "Everything looks good." The Aerostar pilot replied, "OK, appreciate that. We'll start in." The last transmission was abruptly terminated by a loud noise. Shortly thereafter, the controller noticed a smoke plume to the north of the airport. Subsequent radio attempts to contact either aircraft were unsuccessful.
At the time of the accident, ceiling and visibility were essentially unlimited. The wind was reported from several stations at slightly different times as anywhere from 8 to 10 knots with an occasional gust to 15. There were no gusts reported at Philadelphia at the time of the collision.
Formation flight at any time is challenging and more so when the aircraft are not similar. The helicopter was generally below and behind the Aerostar, so the fixed-wing pilot needed to maintain heading and altitude while the chopper maintained adequate distance. Witnesses differed on exactly what happened — some saw the helicopter climb while others said the airplane veered right. It was agreed that the rotor blade hit a wing panel, rendering both aircraft uncontrollable. There were three fatalities on the Aerostar and two on the Bell, while on the ground there were two fatalities, one severe-burn injury, and four minor injuries with more than $6 million in property damages alone, not counting lawsuits. Seems like a terrible price to pay for what should have been a minor incident at worst.
The captain on the Aerostar held an airline transport pilot (ATP) certificate with a multiengine rating and a flight instructor certificate with airplane and instrument ratings. He had just more than 1,500 hours of single-engine flight time and slightly more than 400 hours in twins. He had taken a Part 135 pilot-in-command check in the Aerostar on March 26, 1991, administered by the FAA principal operations inspector for the charter company. The inspector recalled they had covered emergency gear-extension procedures in the oral part of the check. At the time of the accident the pilot had about 72 hours as captain and 42 hours as first officer in the Aerostar.
The accident flight was his second revenue flight as captain of the PA-60. On April 1,1991, his first revenue flight in the PA-60 had not gone well, either, from a mechanical perspective. The flight was aborted shortly after reaching cruise altitude because of a surging engine. The charter passenger, an experienced pilot himself in PA-60 operations, had accumulated more than 300 hours in type. He reported that the captain had difficulty starting the engines and that he needed instruction on proper starting techniques. The takeoff roll was described as "pretty erratic" because the captain was overcontrolling the electric/hydraulic nosewheel steering. At altitude, the right engine began to "surge about 200 rpm," which the passenger believed to be a problem with the fuel controller. The captain did not appear to respond to the problem and the passenger had to persuade the captain to return to the airport. Subsequent inspection found that the fuel controller was defective.
The Aerostar first officer was not a required crewmember per the regulations, although Heinz did personally require two crewmembers when he flew. The first officer had about 1,300 hours of single-engine experience and about 200 hours in multiengine equipment. The first officer had been on duty the night before from 9 p.m. until 6 a.m. and reportedly slept from about 6:30 a.m. to 9 a.m. While the first officer's flight and duty time did not exceed the limitations for unscheduled one- and two-pilot crews, the NTSB believed it was ill advised for the first officer to take the flight, having been on duty the entire previous night.
Both helicopter pilots held ATPs for multiengine rotorcraft and both had in excess of 8,000 hours with several thousand in company helicopters. Ironically, the first officer had logged prior time in the Aerostar. None of the pilots involved indicated he had ever received any instruction on formation flight. The chief pilot for the flight department that owned the helicopters noted that company policy for "in-flight" inspection of another aircraft was not to get closer than 300 feet, the length of a football field. That's good advice.
By the book
Oddly enough, the FAA-approved flight manual for the Piper PA-60 does not contain emergency gear-extension procedures in section 3, the emergency procedures section, which is where you would expect to find them. However, in the systems description, there is information about the hydraulic pump, which explains what to do if hydraulic pressure is lost. Essentially, the landing gear will free-fall to the down and locked position using gravity and springs when the gear handle is placed in the down position. In addition, the landing-gear warning horn will sound if the throttles are retarded to idle and the nose gear is not locked. This is the preferred method to check a gear anomaly, and most aircraft with retractable gear employ a similar method for checking gear position.
The training/check pilot for the charter said he did not instruct the Aerostar captain on the operation of the landing-gear warning horn but had explained the push-to-test function of the gear indicator lights — not exactly a thorough checkout.
Without a cockpit voice recorder, it is impossible to determine whether the captain took any action to isolate the problem to the indicator light or verify that the nose gear was locked in the down position. He did not mention any such tests during communications with the tower. The NTSB believed some discussion would have been likely if the checks had been made.
The NTSB examined another Aerostar and concluded that while it was possible to see if the gear was down from a safe distance, it was impossible to verify by visual inspection at any distance whether it was locked. This is true of most aircraft whether viewed by the tower or another aircraft, as previously mentioned.
This scenario started well enough. Advising the tower of a potential problem is always a good idea because it allows other resources to be used where appropriate. Emergency equipment can be standing by and there may be some technical help available, by telephone or radio, to help troubleshoot. As always, flying the aircraft and balancing risk against a minor accident should be kept in perspective.
The NTSB believed that the Aerostar captain's inexperience as pilot in command in revenue operations was a significant factor. During both his revenue flights there had been mechanical malfunctions, and in neither case had he exercised the care that one would expect. In the first, an experienced passenger had to take control of the situation and in the second, the captain failed to recognize the risk posed by the formation flight.
A more experienced pilot probably would have accomplished the emergency procedures and landed the airplane, accepting the possibility of a nose-gear collapse during the landing roll. Some pilots might shut the engines down after main-gear touchdown to minimize the potential propeller and engine damage. Incidents like this happen with some regularity and usually do not result in a major accident.
The Aerostar captain should have rejected the offer for the in-flight inspection since there was nothing to be gained. A less experienced pilot frequently will abdicate responsibility to someone experienced and well intentioned who only jeopardizes the safety of all concerned. Would you let a complete stranger take responsibility for the safety of your aircraft in such a critical maneuver? Remember, none of the pilots had any experience in formation flight.
While we're engaged in Monday-morning quarterbacking, let's look at some other risk-reduction strategies. Having accepted the offer from the captain of the Bell to approach his aircraft, the Aerostar captain should have insisted that the in-flight inspection be accomplished with minimal hazard. The pilots should have agreed on minimum separation and the maneuver should have been conducted so that each aircraft was in sight of the other at all times. A communications strategy and a bailout option should have been discussed to minimize the chance of collision, if not eliminate it. The NTSB also faulted the pilots for not conducting their hazardous inspection over open terrain rather than the densely populated suburbs of Philadelphia. After the collision occurred, there were two fatalities on the ground and one severe injury because of burns, not to mention significant damage to buildings and property.
A review of the literature on the aerodynamic interaction between fixed- and rotary-wing aircraft in close proximity noted that there were two distinct and potentially hazardous aerodynamic concerns: (1) turbulence-induced blade stall and settling experienced by rotary-wing aircraft while flying in the turbulent area behind and below a fixed-wing aircraft, and (2) opposing pitch changes experienced by both aircraft when one flies close behind and below another.
The textbook Aerodynamics for Naval Aviators specifically refers to the case of one aircraft inspecting the landing gear of another. The lower aircraft may experience a nose-up pitching moment and the higher aircraft, a nose-down pitching moment. The author states that the opposing pitch-moment changes can be large and must be anticipated or a collision may result. Engineers at Bell Helicopter confirmed that the Bell 412 would experience such a nose-up pitch change.
The NTSB found the probable cause to be the poor judgment by the captain of the Aerostar to permit the in-flight inspection after he had determined to the best of his ability that the nose landing gear was fully extended, the poor judgment of the captain of the helicopter to conduct the inspection, and the failure of the helicopter crew to maintain safe separation. Contributing to the accident were the incomplete training and checkout that the captain received from the charter operator and insufficient oversight by the principal operations inspector for the FAA flight standards district office.
If the gear malfunctions, get away from the ground, out of the traffic pattern, and up to a safe altitude, and methodically run the emergency gear-extension procedure. Flybys by the tower and formation passes by other aircraft will likely not answer the "locked down" question. Insurance companies are much happier to write a relatively small check for mechanical mishaps than policy limits for loss of life.
Gear-Up Saves — A High Risk!
Sometimes the story involves an impromptu race down the runway with a pickup truck precariously positioned under a ruptured Cessna retractable with a lame main gear. The story ends happily when the gear is miraculously pulled into place and the aircraft lands successfully. This has happened several times in recent memory and Lady Luck has smiled on the gladiators, but do not be deceived by the very real risk that such an act entails. The conservative approach to landing-gear malfunctions will always lead to the occupants' being uninjured and relatively minor damage to the aircraft. Unpracticed formations, as we've seen, can have unintended negative consequences. It takes only a small gust of wind and an unanticipated response from either driver, pilot, or gear extension "specialist" to result in decapitation, fire, destruction of both machines, loss of all participants — you get the picture. — BL