1.07 Operation in Turbulence

01. Introduction

Aircraft control during turbulence penetration has been the subject of considerable study in the USA. The study of gust and draft influence on the aircraft, as well as the reaction of the pilot as he tries to control the aircraft, has shown that his reactions are often incorrect, resulting in oscillations of the aircraft in the pitching plane. This phenomenon, known as Pilot Induced Oscillation (PIO) is a direct result of the following pilot inputs:

• Inputs, in the wrong sense (e.g. a Push when a Pull was required).
• Inputs, out of phase, with the aircraft’s turbulence-induced motion.
• Inputs, in an attempt to force an attitude change faster than the natural tendency of the aircraft permits.

A multiplicity of indications normally reflect a pitch change:

• The attitude indicator.
• The airspeed indicator.
• The altimeter.
• The vertical speed indicator.
• The experienced change in load factor.

Normally all show a change as a result of a pitch change, but only the attitude indicator can give true pitch indication during turbulence. Roll information is not a problem, as the attitude indicator is the only instrument for this purpose. False information of the aircraft’s motion can be given by the last four, the degree of falseness depending, among other things, on the gust direction.

02. Definitions

The Flight Path

The Flight Path is the path along which the aircraft travels. This flight path can be projected ahead of the aircraft.

A Relative Wind

A Relative Wind can be said to blow back along this flight path, at the speed that the aircraft travels along it.

The Angle of Attack

The Angle of Attack is the angle between the aircraft’s longitudinal axis and the direction of the relative wind.

ABC is the angle of attack. It can vary between 0° and +16° (ig stall). At a given weight, indicated airspeed and angle of attack are practically interchangeable terms. At a low airspeed, angle of attack is maximum. In high speed cruise, angle of attack is minimum, (probably slightly less then +1°).

Pitch Attitude

Pitch Attitude is the angle between the aircraft’s longitudinal axis and the horizon. It coincides with angle of attack only in level flight.

XYZ is the Pitch Attitude. An aircraft may climb out at a pitch attitude of +15°, but the angle of attack will be much less. About 1/2 of this value in a typical case. The difference between Angle of Attack and Pitch Attitude is important for understanding the following theory of control cues.

03. Effect of Relative Wind and Thrust

In the above, the angle of attack and the speed are the same. Note that the relative wind, opposite in vector of the flight path and speed, blows down in climb (a); level in cruise (b); up in descent (c). The aircraft by its motion creates this relative wind and then by its own weather-cocking action tends to face into this wind. The action of the adjustable stabilizer makes it possible for the aircraft to face into its own wind at any angle from 0° to +15°. Since angle of attack and indicated airspeed are in a sense the same, by setting a certain incidence to the horizontal stabilizer, the aircraft is being asked to maintain a particular speed. With no further control inputs, the aircraft will maintain the trimmed speed. If trimmed at 250 kt, in level flight, an increased thrust will cause a climb, and a reduced thrust a descent at a speed, for the purpose of this discussion, of 250 kt. Actually with unchanged stabilizer position the climb speed will be slower and descent faster, but that is associated with the pitching moment of a thrust change.

04. Effects of Gusts and Drafts

If a gust comes from the right, the vertical stabilizer is blown to the left, and the aircraft turns to the right. It turns into the “apparent” relative wind, which is a vector sum of the relative wind and the gust/draft velocity. Similarly if a gust comes from below, the aircraft will pitch down into the “apparent” relative wind. This is illustrated in the following figure:

With a sustained draft a false control cue is introduced by the altimeter which indicates a climb as the aircraft is lifted. This in itself could cause the pilot to apply forward elevator. The nose has already pitched down due to the influence of the draft. It is obvious that forward elevator to regain the old attitude is incorrect.

Table I. Airplane Response to Steady State Drafts in The Pitching Plane.

^a The response shown for the altimeter, vertical speed, and airspeed indicators assumes that pressure changes are consistent with steady state drafts, and ignores short period changes due to random pressure changes.

^b Decreases initially, then increases.

^c Increases initially, then decreases.

Note that drafts 1-4 cause the aircraft to pitch up, 5-8 cause it to pitch down. In 2-4 and 5-8, the aircraft tends to maintain a constant angle of attack. In 1 and 5, the lift change causes a pitch change.

Table II. Elevator Input “Command” By Cockpit Pitch Indicators.

^a Assumes that changes in load factor will call forth a control input that tends to restore normal 1 “g” conditions. There are reasons to believe that, particularly under negative “g” conditions, this may not always be so.

Control cues are given in this figures, which shows the influence of all pitch “indicators” as a PUSH or PULL pitch command. Control commands in the incorrect, or reverse sense are boxed, e.g. PULL. It will be seen that the command of the attitude indicator is always correct.

Every pilot has of necessity submitted himself to a long conditioning process by carrying out constant speed climbs, constant attitude cruise and constant speed descents, all with varying pitch attitudes. This causes the ASI and Altimeter to become a powerful influence on pitch control, with pitch control inputs being made as a direct result of attitude or airspeed changes. An airspeed that is dropping rapidly to a low value is almost certain to provoke, as a conditioned reflex, a strong PUSH input. Especially these cues must be consciously abandoned when flying in severe turbulence.

05. Pilot Induced and Natural Oscillations

A study of simulated turbulence penetration, carried out by the NASA in special simulators, showed that the simulated aircraft, without any control inputs, could fly through the most severe gusts without experiencing excessive “g” loads, or large airspeed and altitude variations. Large changes in pitch were involved but in each case the inherent stability of the simulated aircraft provided the forces restoring to the trim condition. With a pilot “flying” the simulator through a “storm”, the variations in load factor, airspeed and altitude depended on how closely the pilot tried to maintain pitch attitude. Some trials resulted in large oscillations, indicating that the pilot’s control inputs were out of phase with the gust/draft induced motions. Pilot induced oscillations or PIO’s are thus induced.

In a few cases oscillations increased until an upset occurred. Large oscillations of all parameters invariably resulted when pilots were instructed to ignore pitch attitude and concentrate on controlling the airspeed during simulated penetration.

A study of actual upsets had led the CAB to state that in all upsets, although many dissimilarities existed, there were a few common factors, i.e.

• Turbulence.
• Great variation in pitch attitude, airspeed and altitude.
• Large inputs of elevator and stabilizer by the pilots, and
• In some cases, a substantial attitude loss.

They conclude by stating that, although generalizing is generally unsound, yet it is difficult to avoid the conclusion, that the phasing relationship between the turbulence-induced aircraft motion and the control inputs is a factor.

Flight testing has shown that the initial response of large civil aircraft to an elevator deflection is not a good indication of the final response. These aircraft generally have slow response characteristics and as long as no attempt is made to force the aircraft, allowing it to respond well within its characteristics, no difficulty is encountered. If an attempt is made to force a more rapid response by using large elevator deflections, or backing up elevator deflection with stabilizer trim, as might be done in a gust disturbance, a low frequency PIO will result. This type of PIO is especially caused by overcontrolling the aircraft. It does not respond immediately, so the tendency is to put more elevator in, by this time the aircraft has started to respond, and suddenly it seems the response is more than desired. So the tendency is to reverse the process. In turbulence, if an attempt is made to maintain a tight control of attitude, it is most likely that a PIO will result.

Turbulence has known energies broad enough to excite aircraft natural frequencies between 0.2 and 4.0 cps. A pilot encountering this phenomenon described the effect as one might expect to encounter if sitting on the end of a huge tuning fork that had been struck violently. It resulted in all instruments appearing as white blurs, severe to violent buffeting, and complete loss of control, which was only regained on becoming visual.

06. Pilot Physiological Behaviour

In severe turbulence the pilot can suffer spatial disorientation due to loss of space references, and to false tactile stimuli due to accelerations. In this case, all sensory cues other than visual must be ignored, i.e. either from outside reference or by feel. Under negative “g” conditions it is instinctive to pull on the control column to establish positive “g”. This may be the wrong input. Do nothing under negative “g” until the cause is definitely established. Under vibrations of 3 – 4.0 cps the ability to read instruments and respond correctly is severely diminished. Tunnel Vision, head nodding, and eyeball bounce combine to compound the difficulties of only pilot control. The autopilot does not suffer from these problems. Use of the autopilot in the turbulence mode is recommended.