A tail strike can occur during either takeoff or landing. Many air carrier aircraft have tail skids to absorb energy from a tailstrike. On some aircraft, the tail skid is a small bump on the aft underside of the airplane, while on others it is a retractable skid that extends and retracts with the landing gear.
Most tail strikes are the result of pilot error, and in general, landing tail strikes cause more damage than takeoff tail strikes.
In 1978, Japan Airlines flight 115 experienced a tail strike during landing that caused damage to the aft pressure bulkhead. The aircraft was repaired (although the repair was faulty) and returned to service. Seven years later, the aircraft, operating as Japan Airlines Flight 123, crashed as a result of the failure of the improperly-repaired pressure bulkhead.
This Boeing document is an excellent analysis of tailstrikes. A portion of the document is reproduced below:
Takeoff Risk Factors
Any one of these four takeoff risk factors may precede a tail strike:
A mistrimmed stabilizer occurring during takeoff is not common but is an experience shared at least once by almost every flight crew. It usually results from using erroneous data, the wrong weights, or an incorrect center of gravity (CG). Sometimes the information presented to the flight crew is accurate, but it is entered incorrectly either to the flight management system (FMS) or to the stabilizer itself. In any case, the stabilizer is set in the wrong position. The flight crew can become aware of the error and correct the condition by challenging the reasonableness of the load sheet numbers. A flight crew that has made a few takeoffs in a given weight range knows roughly where the CG usually resides and approximately where the trim should be set. Boeing suggests testing the load sheet numbers against past experience to be sure that the numbers are reasonable.
A stabilizer mistrimmed nosedown can present several problems, but tail strike usually is not one of them. However, a stabilizer mistrimmed noseup can place the tail at risk. This is because the yoke requires less pull force to initiate airplane rotation during takeoff, and the pilot flying (PF) may be surprised at how rapidly the nose comes up. With the Boeing-recommended rotation rate between 2.0 and 3.0 degrees per second (dps), depending on the model, and a normal liftoff attitude, liftoff usually occurs about four seconds after the nose starts to rise. (These figures are fairly standard for all commercial airplanes; exact values are contained in the operations and/or flight-crew training manuals for each model.) However, with the stabilizer mistrimmed noseup, the airplane can rotate 5 dps or more. With the nose rising very rapidly, the airplane does not have enough time to change its flight path before exceeding the critical attitude. Tail strike can then occur within two or three seconds of the time rotation is initiated.
If the stabilizer is substantially mistrimmed noseup, the airplane may even try to fly from the runway without control input from the PF. Before reaching Vr, and possibly as early as approaching V1, the nose begins to ride light on the runway. Two or three light bounces may occur before the nose suddenly goes into the air. A faster-than-normal rotation usually follows and, when the airplane passes through the normal liftoff attitude, it lacks sufficient speed to fly and so stays on the runway. Unless the PF actively intercedes, the nose keeps coming up until the tail strike occurs, either immediately before or after liftoff.
ROTATION AT IMPROPER SPEED
This situation can result in a tail strike and is usually caused by one of two reasons: rotation is begun early because of some unusual situation, or the airplane is rotated at a Vr that has been computed incorrectly and is too low for the weight and flap setting.
An example of an unusual situation discovered during the DPD examination was a twinjet going out at close to the maximum allowable weight. In order to make second segment climb, the crew had selected a lower-than-usual flap setting. The lower flap setting generates V speeds somewhat higher than normal and reduces tail clearance during rotation. In addition, the example situation was a runway length-limited takeoff. The PF began to lighten the nose as the airplane approached V1, which is an understandable impulse when ground speed is high and the end of the runway is near. The nose came off the runway at V1 and, with a rather aggressive rotation, the tail brushed the runway just after the airplane became airborne.
An error in Vr speed recently resulted in a trijet tail strike. The load sheet numbers were accurate, but somehow the takeoff weight was entered into the FMS 100,000 lb lower than it should have been. The resulting Vr was 12 knots indicated air speed (kias) slow. When the airplane passed through a nominal 8-deg liftoff attitude, a lack of sufficient speed prevented takeoff. Rotation was allowed to continue, with takeoff and tail strike occurring at about 11 deg. Verification that the load sheet numbers were correctly entered may have prevented this incident.
EXCESSIVE ROTATION RATE
Flight crews operating an airplane model that is new to them, especially when transitioning from unpowered flight controls to ones with hydraulic assistance, are most vulnerable to using excessive rotation rate. The amount of control input required to achieve the proper rotation rate varies from one model to another. When transitioning to a new model, flight crews may not consciously realize that it will not respond to pitch input in exactly the same way.
As simulators reproduce airplane responses with remarkable fidelity, simulator training can help flight crews learn the appropriate response. A concentrated period of takeoff practice allows students to develop a sure sense of how the new airplane feels and responds to pitch inputs. On some models, this is particularly important when the CG is loaded toward its aft limits, because an airplane in this condition is more sensitive in pitch, especially during takeoff. A normal amount of noseup elevator in an aft CG condition is likely to cause the nose to lift off the runway more rapidly and put the tail at risk.
IMPROPER USE OF THE FLIGHT DIRECTOR
As shown in figure 1, the flight director (FD) is designed to provide accurate pitch guidance only after the airplane is airborne, nominally passing through 35 ft (10.7 m). With the proper rotation rate, the airplane reaches 35 ft with the desired pitch attitude of about 15 deg and a speed of V2 + 10 (V2 + 15 on some models). However, an aggressive rotation into the pitch bar at takeoff is not appropriate and may rotate the tail onto the ground.
Landing Risk Factors
Any one of these four landing risk factors may precede a tail strike:
A tail strike on landing tends to cause more serious damage than the same event during takeoff and is more expensive and time consuming to repair. In the worst case, the tail can strike the runway before the landing gear touches down, thus absorbing large amounts of energy for which it is not designed. The aft pressure bulkhead is often damaged as a result.
An unstabilized approach appears in one form or another in virtually every landing tail strike event. When an airplane turns on to final approach with excessive airspeed, excessive altitude, or both, the situation may not be under the control of the flight crew. The most common cause of this scenario is the sequencing of traffic in the terminal area as determined by air traffic control.
Digital flight recorder data show that flight crews who continue through an unstabilized condition below 500 ft will likely never get the approach stabilized. When the airplane arrives in the flare, it invariably has either excessive or insufficient airspeed, and quite often is also long on the runway. The result is a tendency toward large power and pitch corrections in the flare, often culminating in a vigorous noseup pull at touchdown and tail strike shortly thereafter. If the nose is coming up rapidly when touchdown occurs and the ground spoilers deploy, the spoilers themselves add an additional noseup pitching force. Also, if the airplane is slow, pulling up the nose in the flare does not materially reduce the sink rate and in fact may increase it. A firm touchdown on the main gear is often preferable to a soft touchdown with the nose rising rapidly.
HOLDING OFF IN THE FLARE
The second most common cause of a landing tail strike is a long flare to a drop-in touchdown, a condition often precipitated by a desire to achieve an extremely smooth landing. A very soft touchdown is not essential, nor even desired, particularly if the runway is wet.
Trimming the stabilizer in the flare may contribute to a tail strike. The PF may easily lose the feel of the elevator while the trim is running; too much trim can raise the nose, even when this reaction is not desired. The pitchup can cause a balloon, followed either by dropping in or pitching over and landing flat. Flight crews should trim the airplane in the approach, but not in the flare itself, and avoid "squeakers," as they waste runway and may predispose the airplane to a tail strike.
MISHANDLING OF CROSSWINDS
A crosswind approach and landing contains many elements that may increase the risk of tail strike, particularly in the presence of gusty conditions. Wind directions near 90 deg to the runway heading are often strong at pattern altitude, and with little headwind component, the airplane flies the final approach with a rapid rate of closure on the runway. To stay on the glidepath at that high groundspeed, descent rates of 700 to 900 ft (214 to 274 m) per minute may be required. Engine power is likely to be well back, approaching idle in some cases, to avoid accelerating the airplane. If the airplane is placed in a forward slip attitude to compensate for the wind effects, this cross-control maneuver reduces lift, increases drag, and may increase the rate of descent. If the airplane then descends into a turbulent surface layer, particularly if the wind is shifting toward the tail, the stage is set for tail strike.
The combined effects of high closure rate, shifting winds with the potential for a quartering tail wind, the sudden drop in wind velocity commonly found below 100 ft (31 m), and turbulence can make the timing of the flare very difficult. The PF can best handle the situation by exercising active control of the sink rate and making sure that additional thrust is available if needed. Flight crews should clearly understand the criteria for initiating a go-around and plan to use this time-honored avoidance maneuver when needed.
OVER-ROTATION DURING GO-AROUND
Go-arounds initiated very late in the approach, such as during flare or after a bounce, are a common cause of tail strike. When the go-around mode is initiated, the FD immediately commands a go-around pitch attitude. If the PF abruptly rotates into the command bars, tail strike can occur before a change to the flight path is possible. Both pitch attitude and thrust are required for go-around, so if the engines are just spooling up when the PF vigorously pulls the nose up, the thrust may not yet be adequate to support the effort. The nose comes up, and the tail goes down. A contributing factor may be a strong desire of the flight crew to avoid wheel contact after initiating a late go-around, when the airplane is still over the runway. In general, the concern is not warranted because a brief contact with the tires during a late go-around does not produce adverse consequences. Airframe manufacturers have executed literally hundreds of late go-arounds during autoland certification programs with dozens of runway contacts, and no problem has ever resulted. The airplane simply flies away from the touchdown.