by Pat Veillette
In a world dominated by checklists and standard operating procedures, it can be unsettling to find that sometimes deviating from the norm is the only way to make everything work out right.
A recent incident rammed this home on a startlingly personal level. The evening news was interrupted by a special report that an aircraft was currently circling the airport, waiting to make an emergency landing. I glued myself to the television. The jet happened to be the same make and model of jet that I fly at work, so I watched the video with even more interest. The news camera focused on the aircraft as it circled high over the airport and then subsequently made a flawless approach.
The aircraft rescue and fire fighting units rolled down the runway after the jet as it passed by their location. It rolled to a stop straight ahead on the runway and the rescue vehicles immediately pulled up to the jet. After a few uneventful minutes, the aircraft was towed off the runway.
I thought the paint job on the jet looked familiar, though it was hard to tell from the distant view through the news cameras telephoto lens. Two weeks later, I was back at work and, when I checked the aircraft logbook, I found the incident airplane was the very airplane I was about to fly and discovered that the captain on that day was a friend of mine.
Steve, the captain, had just obtained his copy of the videotape from the evening news and we found a VCR where we could replay the tape. It was interesting to listen to Steves Monday morning re-evaluation of his incident. The lessons it contained held some surprises.
It was a 98-degree summer day at the high altitude airport, so the density altitude was considerable. The captain reviewed the takeoff numbers and determined the aircraft needed nearly 8,000 feet of runway to meet its takeoff requirements, including the accelerate/stop distance. The runway was nearly 10,000 feet long, so there was plenty of excess runway length available.
The flight was being conducted under an IFR flight plan, which normally requires a 3.3 percent climb gradient, but the nearby mountainous terrain required a steeper-than-normal climb gradient for the IFR departure.
Instead, the captain chose to conduct the climb out using a VFR climb on course request, which means he would take responsibility for terrain clearance during the climb out. This was clearly legal, and it meant the airplane only had to meet the VFR climb gradient requirement of 2.4 percent.
The takeoff was done with the normal takeoff setting of 15 degrees of flaps. Taxiing took several minutes because the jet had to taxi nearly two miles to get to the end of the runway.
During the takeoff roll, the aircraft accelerated slowly due to the high density altitude. It took several thousand feet of runway just to get an indication of airspeed, and the final couple thousand feet of runway quickly rolled by with the airspeed increasing slowly. As the aircraft neared rotation speed, the pilot noticed that the aircraft wanted to divert toward the side of the runway.
It took all of the rudder deflection he could muster to keep the aircraft on the runway. It also seemed like the runway had quickly turned real rough. The acceleration seemed to slow, but he kept control of the aircraft and rotated at the rotation speed.
Normally a pilot will retract the gear as soon as the aircraft shows a positive rate of climb and the remaining runway is insufficient to abort the takeoff. This time, the captain chose to leave the aircraft in its present configuration. He knew he had a flyable aircraft and that was enough. In the VFR conditions, he was able to see and avoid the terrain and climb to pattern altitude.
He told the tower that he might have a problem and needed to orbit while he worked out the problem. Steve then picked up the air phone and called the company, who then put the manufacturers engineers on the line.
Finding a Culprit
The most likely reason for the slight rumble on takeoff and diversion toward the side of the runway was a blown tire. As the flight crew and the engineers tried to figure out what was working on the aircraft and what wasnt, they also tried to rebuild the failure process. The aircraft had to burn off some fuel to make its landing weight anyway, so they had plenty of time to troubleshoot and make some decisions.
The groundspeed at the time of the tire failure was approximately 165 knots, on a heat soaked runway. The long taxi probably heated up the tires. The takeoff distance was considerable, nearly three times the normal distance that would be required on a cooler day at sea level, and the resultant ground speed at the high density altitude was very high.
Since the centrifugal acceleration on a rotating body is proportional to the square of the speed, the heat and acceleration stresses on the tire were very high, increasing the chances of tire failure.
No one knew if the tires had failed catastrophically, but they all assumed the worst. The only option that held out a glimmer of hope to determine the condition of the tires was to fly by the tower. Steve coordinated with the tower for a fly-by, and tried to go as slow as he could, knowing that it is very hard for tower personnel to get a good look at the landing gear, even if they have binoculars. The tower reported that all three gear appeared to be down, but they couldnt confirm the condition of the tires.
The drag of landing on blown tires is much stronger than you might imagine. Anyone who has ever had a blowout while driving a car is acquainted with the sudden compromise of directional control. In an airplane, where ground manners are already compromised, the problem is exacerbated because nosewheel steering alone is probably not adequate to overcome the drag of the blown tire.
On this particular aircraft, there are three hydraulic lines next to the landing gear. One hydraulic line is the normal hydraulic pressure to the anti-skid braking system. The second line is the return pressure to the reservoir. The third line goes to the emergency side of the brakes.
The pilots didnt know if any of the hydraulic lines had been punctured, but they had to assume the worst. They checked the hydraulic pressure and noted that they had lost all of their hydraulic system pressure. This meant normal operation of the landing gear, brakes, nosewheel steering and thrust reversers was out of the question.
Fortunately the landing gear was already down and locked, thanks to the captain quick decision not to raise it after takeoff. But losing the primary braking system meant they were without the anti-skid braking system, which would seriously erode their braking ability on the runway. To make matters worse, they would have no nosewheel steering, which would be especially important when landing because of the problems with directional control caused by the blown tires.
This left the pilots with the emergency brake accumulator, which stores up a supply of hydraulic pressure for a one shot chance with the emergency brakes.
The flight crew and engineers feared they might have an additional problem caused by the high-speed failure of the tires. At 165 knots, it was almost certain that portions of the shredded tire would strike other nearby structures on the aircraft, particularly the flaps.
The pilot not flying went back to look at the flaps through the cabin windows, but couldnt see any apparent damage to the upper side of the flaps. It was the underside of the flaps and the flap actuating mechanism that worried them more.
The pilot didnt know for certain, but he had to assume the worst – that tire debris had damaged the flaps. He chose to leave the flaps in the takeoff position, rather than risk moving them and getting them stuck in an asymmetrical position.
The crew planned to land the aircraft on the side of the runway with the good tires, hoping that it would give them some extra margin in the event the blown tires tried to drag them off the side of the runway. The approach was flown at the proper partial-flap speed, which required touching down about 10 knots faster than normal.
The crew requested and was given the longest and widest runway at the airport for the landing, and the airport rescue and fire fighting teams were pre-positioned along the runway. Steve flew a flawless approach and landed in the touchdown zone on the good side of the runway, then applied continuous pressure to the emergency brakes, trying to use differential braking to control the direction of the aircraft.
He managed to bring the aircraft to a stop straight ahead on the runway. The airport rescue and fire fighting forces immediately surrounded the aircraft. The fire crew checked the areas of the brakes, since hot brakes and leaking hydraulic fluid are not a good combination. The crews pinned the landing gear and the aircraft was towed off the runway without further incident. Airport management subsequently swept the runway to make certain that no debris was left over on the runway.
Good Problem Solving
There are some important lessons to learn from Steves experience.
For starters, Steve did not mindlessly apply the automatic procedure of raising the gear and flaps on takeoff. Pilots have been trained for years to raise the gear at a positive climb rate and to retract the flaps at the proper speed and altitude. In jets, especially, both of those happen very rapidly after liftoff.
Instead, Steve chose to leave the gear extended and the flaps in the takeoff position. This was an outstanding reaction because retracting the landing gear with blown tires has a good chance of causing the gear to jam in the wheel well, making the problem even worse.
He accurately noted that something was wrong with the aircraft, kept control of the aircraft, and kept the aircraft in a safe condition while he flew to safe maneuvering airspace to figure out the problem.
Second, he divided the workload between himself and his copilot so that one pilot maintained control of the aircraft at all times, while the other pilot handled communications and the abnormal procedures.
Third, he sought the advice of other resources to help him determine the problem and formulate a safe recovery plan. This included communicating with others such as ATC and manufacturers engineers who would have critical pieces of information to resolve the problem.
Fourth, Steve knew enough about the associated systems that he was able to investigate the possible failures induced into those sub-systems, and he was able to apply corrective procedures for those failures. In this case, the tire failure may have induced damage in many other sub-systems, including the anti-skid brakes, landing gear, nosewheel steering, and flaps. This was a long list of things to consider.
Rethinking the Scenario
Another point that is interesting about this incident is how it is a clear exception to the accepted wisdom regarding aborting a takeoff. Standard procedure, reflected in virtually all training programs, is that you should abort a takeoff in its earliest stages for any reason. While that is an accurate statement, its far from exception-free.
Aborting a takeoff at relatively slow speeds is a low-risk maneuver because there is plenty of runway remaining and the aircraft doesnt have the deceleration and control problems associated with high-speed roll-outs.
High speed aborts are another problem. High-speed aborts require maximum braking. In the process, you risk losing directional control or blowing tires. Even in the best situations, you need plenty of runway remaining.
As such, the recommendation for high-speed aborts has usually been reserved for severe or dire circumstances, such as fire, engine failure, loss of directional control or any serious master warning caution.
This begs the question of whether Steve should have aborted the takeoff. For what its worth, I think he took the best action possible.
A high-speed abort in this scenario could have serious consequences. The accelerate-stop takeoff performance charts are all predicated upon having good tires and normal braking. In Steves case, he had neither.
The tires were blown on the one side, meaning that he had essentially zero braking ability, and the loss of the hydraulic system meant that his normal anti-skid braking system may have been rendered inoperative, depending on how fast the hydraulic system was losing pressure.
A high speed abort with only a couple thousand feet of runway remaining and with inoperative braking systems and the directional control problems associated with blown tires is a condition no pilot ever hopes to experience. It makes more sense to try instead to touch down at the beginning of the runway, with plenty of runway in front of me.
I dont know the cause of the tire failure. There are many possible causes, including debris on the runway, undetected weaknesses in the tire or under-inflation. In any case, Steve made a good split-second decision, and then avoided the temptation of automatically doing the normal ingrained procedures.
This incident supplies a lot of food for thought. Tire failures cause more aborted takeoffs than engine failures, so the probability of tire failure is significant enough that you need to give it some serious forethought. If you have a nice long and wide runway for the takeoff, and if you can still maintain directional control, then perhaps your safest action is aborting.
If each main gear has two tires, there is a chance that only one of the tires took the damage, in which case the previous scenario applies well to you. This is a good example of how the failure of one component can cause a cascading series of mechanical failures, each of which will require methodical trouble-shooting and proper corrective actions to recover the aircraft safely.
It goes beyond tire failure. It goes down to the core of developing sound judgment as pilot in command.
-Pat Veillette is an aviation safety researcher who flies transport aircraft for a living.
