The recent, fatal crash of a vintage military trainer provides valuable lessons for pilots of twin and turbine airplanes as well. The NTSB reports:
…a North American SNJ-5 airplane impacted terrain following a loss of control during initial climb after takeoff from runway 13R at Kingsville Naval Air Station (NQI), Kingsville, Texas. The pilot and pilot rated passenger were fatally injured and the airplane was destroyed. Visual meteorological conditions prevailed. Witnesses reported that the airplane took off on runway 13R and had requested a right-hand teardrop turn for a departure toward the north. The witnesses reported seeing the airplane in a steep right bank with some reporting that the bank angle exceeded 90 degrees of bank. The airplane descended nose low and the right bank angle lessened before the airplane struck the ground.
Stalls and spins get the lion’s share of coverage in instruction and in article and videos concerning Loss of Control – Inflight (LOC-I). The record shows that LOC-I events are the most common fatal accident scenario, and most LOC-I events appear to be stalls that often develop into a spin before impact.
There is another LOC-I sequence that is neither a stall nor a spin. It is a natural outcome of aircraft stability, and a characteristic of all longitudinally (pitch) stable airplanes. Yet it is not mentioned by name, trained or evaluated in Practical Tests for pilot certificates or ratings. The sequence is a spiral dive, and it is what witnesses of the SNJ crash seem to describe.
Here’s how the FAA’s Airplane Flying Handbook (AFH) explains a spiral, with my emphasis added in bold font:
A spiral dive, a nose low upset, is a descending turn during which airspeed and G-load can increase rapidly and often results from a botched turn. In a spiral dive, the airplane is flying very tight circles, in a nearly vertical attitude and will be accelerating because it is no longer stalled. Pilots typically get into a spiral dive during an inadvertent IMC encounter, most often when the pilot relies on kinesthetic sensations rather than on the flight instruments. A pilot distracted by other sensations can easily enter a slightly nose low, wing low, descending turn and, at least initially, fail to recognize this error. Especially in IMC, it may be only the sound of increasing speed that makes the pilot aware of the rapidly developing situation. Upon recognizing the steep nose down attitude and steep bank, the startled pilot may react by pulling back rapidly on the yoke while simultaneously rolling to wings level. This response can create aerodynamic loads capable of causing airframe structural damage and /or failure.
The AFH recommends this spiral dive recovery technique:
- Reduce power to idle
- Apply forward elevator (“unload the wing,” i.e., reduce the G load)
- Roll wings level
- Gently raise the nose to level flight
- 5Increase power to climb power
This excerpt doesn’t explain why an airplane will naturally enter a spiral or how such spirals develop. This lack of emphasis in training syllabi and complete absence in Practical Test evaluation means many, perhaps most pilots may be unprepared to recognize and recover from a spiral. Let’s delve into why a spiral is a natural outcome of aircraft stability, how a pilot may enter a spiral (it’s not just an attempted visual flight into IMC phenomenon), and why knowing about spirals is important to VFR-only and instrument pilots alike.
Most airplanes exhibit some level of stability in at least two of the three axes. Almost all have built-in pitch stability. Disturbed upward or downward in pitch and then released, the airplane’s nose will oscillate up and down through two or three cycles before it returns to its original pitch attitude…not necessarily on its initial altitude, but at the same pitch attitude, angle of attack and indicated airspeed. Put another way, a pitch-stable airplane will seek the indicated airspeed (actually, angle of attack) for which it is trimmed. If it is disturbed in pitch, or if power or total drag (flap, landing gear position) changes, the airplane will pitch down or up as necessary to remain at its trimmed speed.
Most airplanes also have some level of stability in yaw. Kick a rudder pedal and release, or hit a wind shear that yaws the aircraft, and it will wallow back and forth a few oscillations before returning to straight-ahead flight.
Many aircraft are neutral in stability or even slightly unstable in roll. Enter a shallow bank and the airplane may remain banked or slowly return to approximately wings-level flight. But bank steeply enough and most aircraft will not level their own wings. In fact, in a steep turn most airplanes will continue to bank progressively more steeply. This is sometimes called the overbanking tendency, the reason it may take opposite aileron input to maintain bank once established in a steep turn.
You’ve probably seen diagrams that show the relationship between bank angle and stalling speed (see photo). What’s not often well-explained is that this relationship is only valid in level, coordinated flight. If the pilot does not resist the airplane’s tendencies and its nose drops to seek the trimmed airspeed, the G load does not increase. In fact, the load factor will increase only if the pilot, an autopilot, or a runaway electric trim system resists the airplane’s natural tendency to change pitch if it gets off its trimmed speed. An airplane will not stall on its own. The pilot (or an automated pilot) has to actively pull against the airplane’s stability to make it stall.
What happens then if the airplane enters a steep turn and the pilot provides more or less resistance than is necessary to maintain level flight? We’ll use the 60-degree bank example:
- If the pilot adds more than 2G of resistance, the airplane will climb. The nose will rise above the horizon and, if there is sufficient power, the airplane will enter a sustained climb. With insufficient power the wing will quickly enter an accelerated stall.
- If the pilot applies exactly 2G of resistance, the airplane remains level. Airspeed will decrease from the drag of high angle of attack flight, so the pilot will have to add power to maintain airspeed. If airspeed increases, the airplane will climb or the pilot may reduce back pressure – more power means the same G load is sustained at a lower angle of attack. If airspeed decreases, the airplane will descend and its nose will drop below the horizon seeking to attain and maintain the trimmed airspeed.
- If the pilot does not apply at least 2G of resistance with elevator, power or both, the airplane will descend. Its nose will drop below the horizon, seeking to attain and maintain the trimmed airspeed.
- Further complicating this situation is the overbanking tendency. Unless the pilot corrects for it, once in a steep turn, the wing will continue to bank further. This means the nose will drop even more. The airplane, now sensing more airspeed than that for which it is trimmed, will naturally pitch upward to return to the slower, trimmed speed. Except this pitch change is “up” relative to the airframe, not relative to the horizon. In a steep (and getting steeper) turn, this just tightens the downward spiral, increasing airspeed even more. Airspeed and vertical speed increase incredibly fast. As bank angle and speed increase, G load increases to (and eventually beyond) the airplane’s structural limit.
Put simply, a spiral is a steep turn that the pilot allows to go bad. Once in a spiral, one of five outcomes results:
- The pilot recovers from the spiral using the recovery technique described earlier.
- The airplane spirals rapidly into terrain.
- The airplane is high enough at the entry into the spiral that it has time to accelerate beyond VNE before it impacts terrain. Exceeding structural load limits causes the airplane to break up in flight.
- The pilot does not recognize the spiral for what it is, or does not know the proper recovery technique, or panics. She/he pulls back on the controls, perhaps instinctively. The G load builds and overstresses the airframe; the airplane breaks up in flight.
- The pilot attempts a recovery but does not apply forward control pressure to unload the wing. The airplane exceeds structural limits in the pullout and breaks up in flight.
Those sequences may sound familiar. The outcome of attempted visual flight in Instrument Meteorological Conditions (“VFR into IMC”) follows one of these patterns. The same goes for a thunderstorm or other strong turbulence encounter, even for an instrument pilot. An airplane’s natural spiral tendency helps explain the hazards of the visual portion of an IFR circle-to-land approach, and to landing at a “dark hole” airport at night. In both of these situations, visibility is reduced; the pilot is usually focused on the runway trying to keep it in sight and unusual visual cues tempt the pilot into flying steep banks close to the ground. Frankly, I think more airplanes impact the ground out of spirals entered from uncorrected steep banks in the traffic pattern than they do from stalls.
Training from this Trainer
Let’s go back to our example. The SNJ pilot clearly intended to make a right-turning departure; he asked for permission to do so. I interpret a “right hand teardrop turn” from Runway 13 for a departure to the north to mean a tight turn to overfly the airfield, as contrasting from a more conventional left turn on course after departure. An airshow-type, Navy-historical airplane departing from an air show at a Naval Air Station in this “look at me” departure path at least suggests the pilot intended to make a fairly steep turn shortly after lifting off, a turn that could quickly degrade into a spiral. It’s at least consistent with what I see at air shows all the time.
We don’t know yet if there was an engine issue, or the pilot pulled into an accelerated stall, or if there was some sort of control issue, or whether there were medical or other issues that led to the flight path that witnesses described. But whether a spiral was a factor in the SNJ crash or not, hopefully you now know a little more about spirals and how to avoid them.