"Kicking out the crab"

"Kicking out the crab"

Steve Seibel
www.aeroexperiments.org

This page was last modified on September 15, 2006

Throughout this article, unless explicitly stated otherwise, when we speak of the "flight path" we mean the aircraft's flight path through the airmass, not the aircraft's ground track.

In "Sideslips and forward slips", we noted that in the presence of a crosswind, to make an aircraft's ground track coincide with a runway or any other line on the ground, the pilot can simply aim the aircraft's nose, and the flight path through the airmass, in the "upwind" direction and fly the aircraft in a normal, "coordinated" manner with no slip or skid. We noted that this maneuver is called "crabbing", which is a somewhat misleading term in that it seems to imply that the aircraft is moving sideways through the air, when the reality is that aircraft is actually moving through the air in the same direction that the nose is pointing. We also noted that this "crabbing" maneuver will not allow the aircraft's fuselage (and landing gear) to be aligned with aircraft's ground track (and the runway). We described how when landing in a crosswind, the pilot of a "conventional" 3-axis aircraft often uses a sustained, non-turning slip (typically called a "sideslip" in this situation) to yaw the nose of the aircraft in the "downwind" direction in relation to the flight path through the airmass, so that the fuselage (and landing gear) are aligned with the ground track (and runway). We noted that to make the aircraft fly in a straight line when the nose was yawed to one side in relation to the flight path, the wing had to be banked in the opposite direction, so that the horizontal component of the lift from the banked wing cancelled out the horizontal component of the "aerodynamic sideforce" created by the impact of the airflow or relative wind against the "upwind" side of the fuselage, so that there was no net horizontal force trying to make the flight path curve either toward the left or toward the right.

Sometimes pilots of "conventional" 3-axis aircraft use a technique called "kicking out the crab" to deal with a crosswind during landing. With this technique, the pilot simply flies the aircraft in a coordinated, non-slipping manner, with the flight path (and the nose of the aircraft) aimed in the "upwind" direction in relation to the runway centerline, so that the ground track coincides with the runway centerline, until the aircraft is about to touch down. Then the pilot applies firm pressure to the "downwind" rudder pedal to swing the nose (and the landing gear) into alignment with the ground track (and runway). If the aircraft touches down immediately after the pilot "kicks" the downwind rudder, the aircraft will touch down with no sideways "drift".

It is enlightening to consider what happens if the touchdown is delayed for several seconds after the pilot "kicks out the crab"! As soon as the pilot applies rudder in the downwind direction and yaws the nose to point in a different direction than the aircraft is actually moving through the airmass, the relative wind strikes the upwind side of the fuselage, generating an "aerodynamic sideforce". This force creates an acceleration: the flight path begins to curve in the downwind direction. This curvature is a gradual process. If the pilot were to release the pressure on the rudder one second after he "kicked out the crab", the aircraft's natural yaw stability or "weathervane effect" would bring the nose back into alignment with the flight path through the airmass at that instant. If we look at the flight path one second after the pilot has "kicked out the crab" or applied pressure to the "downwind" rudder pedal, we see that the flight path has only curved a small amount in the downwind direction. Therefore if the pilot releases the rudder one second after he "kicks out the crab", the nose will swing back almost to its original heading. Also, the curvature in the ground track will match the curvature in the flight path through the airmass: since the pilot was only applying the rudder briefly, the ground track would have had time to curve a small amount in the downwind direction. From the pilot's point of view, this will mean that the aircraft will have only have accumulated a small (but still non-zero!) amount of crosswind "drift" in relation to the runway.

This brings us to an important, but often overlooked, point about the "kick out the crab" method of correcting for crosswinds: the pilot does not simply "kick" out the crab. Instead, to keep the nose of the aircraft aligned with the runway heading, the pilot applies the rudder in the "downwind" direction, and then he must keep the rudder deflected in the "downwind" direction until the aircraft touches down, or until the flight path has curved so much that the relative wind is no longer striking the side of the fuselage. (We're assuming here that the pilot will not try to compensate for any sideways "drift" or downwind curvature of the flight path and ground track that may develop if the touchdown is delayed for several seconds after the pilot "kicks out the crab". We'll assume that the pilot will focus entirely on making whatever rudder inputs are needed to keep the aircraft's heading the same as the runway heading.)

If the touchdown is delayed for several seconds and the pilot continues to maintain whatever pressure is required on the "downwind" rudder to keep the nose of the aircraft aligned with the runway heading -- i.e. to keep the aircraft's heading the same as the runway heading -- the deflected rudder continues to make the aircraft point in a different direction than it is actually moving through the air, so the airflow or relative wind continues to strikes the "upwind" side of the fuselage, causing the flight path and ground track to continue to curve in the downwind direction. As the flight path and ground track continue to curve in the downwind direction, the difference between the direction of the flight path through the airmass and the runway heading steadily decreases. This means that the amount of "downwind" rudder needed to keep the aircraft's heading the same as the runway heading also steadily decreases. Therefore the pilot can steadily decrease his pressure on the "downwind" rudder pedal. This creates a steady decrease in the rate of downwind curvature of the flight path and ground track. In other words there is a steady decrease in the rate at which the sideways "drift" of the aircraft in relation to the runway is increasing. The cycle ends at the moment that the flight path through the airmass has curved so much that it coincides with the direction that the nose of the aircraft is pointing, which in this case happens to be the same as the runway heading. At this point the rudder will need to be centered to keep the aircraft's heading constant. In other words at this point the pilot is no longer forcing the aircraft to fly sideways through the air. And at this point, the aircraft's ground track is angled off quite significantly toward the "downwind" side of the runway. In fact, at this point the aircraft's ground track has attained a sideways "drift" component, in relation to the runway, that is exactly equal to the crosswind component in the external, meteorological wind.

At the instant the pilot "kicks out the crab", the aircraft experiences a sideways component in the relative wind or airflow that happens to be equal to the crosswind component in the external, meteorological wind. Thus we can say that the external, meteorological wind is "pushing" the aircraft in the downwind direction and creating a curvature in the ground track. From a more fundamental point of view, this is nonsensical: an aircraft does not "feel" a steady wind. Instead, the curvature of the flight path and ground track is entirely caused by the fact that the pilot uses the rudder in a way that makes the airflow or relative wind strike the side of the fuselage. In other words, the "kick out the crab" maneuver is absolutely no different than any other wings-level skidding turn performed with the rudder. At the instant the pilot first "kicks out the crab", the only reason that the sideways component in the relative wind happens to be equal to the crosswind component in the external, meteorological wind, is that the pilot constrains himself to applying exactly as much rudder as is needed to yaw the aircraft heading into the alignment with the runway heading. If the touch down is delayed, and the pilot keeps the aircraft heading aligned with the runway heading and keeps the wings level, the only reason that the aircraft ends up "accumulating" a cross-runway velocity component in the ground track that is exactly equal to the crosswind component in the external, meteorological wind is that the pilot continues to apply exactly as much rudder input is needed at any given moment to keep the aircraft heading identical to the runway heading. Since the rudder yaws the fuselage in relation to the relative wind and this creates a sideforce that makes the flight path curve, the end result of this can only be that the flight path curves until it is aimed in the same direction as the aircraft heading, which happens to be the same as the runway heading. At this point simple geometry dictates that the sideways (cross-runway) "drift" in the ground track must be equal to the crosswind velocity in the external, meteorological wind , even though the external, meteorological wind was not the real cause of the sideways "drift" in the ground track in relation to the runway.

None of this is nearly so complicated as it may sound. With any "conventionally" shaped aircraft, where yawing the nose to point toward the right (for example) of the actual direction of the flight path and relative wind creates a net aerodynamic sideforce to the right due to the impact of the relative wind against the left side of the fuselage and other surfaces of the aircraft, the following is true: if the pilot is flying on a heading of 360 degrees, and applies right rudder to yaw the nose of the aircraft 10 degrees to the right to a heading of 010 degrees, and then applies whatever rudder pressure is needed to hold that heading, the flight path will curve to right due to the impact of the airflow against the left side of the fuselage and other surfaces of the aircraft. As the flight path curves to the right, the difference between the aircraft's heading and the actual direction of the flight path will steadily decrease, which means that the amount of rudder pressure needed to keep the nose on a heading of 010 will steadily decrease. As the flight path curves to the right, the aerodynamic sideforce created by the impact of the relative wind against the left side of the fuselage will also steadily decrease, which means that the rate of curvature in the flight path will also steadily decrease. At some point the flight path will end up travelling in the direction of 010 rather than 360. At this point the pilot will no longer be applying any pressure on the right rudder pedal, and the airflow will no longer be impacting the left side of the fuselage, and the aircraft will be travelling in a straight line. This is one way to perform a wings-level skidding turn. (And naturally, the direction of the external, meteorological wind is of no consequence whatsoever in this exercise.) Of course, another way to perform a wings-level skidding turn to the right is to gently apply and hold a small but steady amount of right rudder pressure, and release the rudder pressure when the flight path has curved as desired. Note also that with either method, if the aircraft has dihedral, some left aileron will also be needed to counteract the rightwards "downwind" roll torque created by the interaction between the sideways relative wind and the dihedral, so that the wings remain level. On an aircraft with dihedral, this same left aileron input is also needed to keep the wings level when a pilot applies right rudder to "kick out the crab" during a crosswind landing. (We explored how dihedral creates a roll torque in the "downwind" direction whenever there is a sideways component in the relative wind or airflow over an aircraft in "Oblique views and side views of aircraft with dihedral"). Of course, in situations other than landings, in many cases when a pilot makes gentle steering corrections with the rudder he doesn't mind if this also causes a roll torque in the intended direction of turn--in this case the rudder is being used to produce a turn that is partly due to skidding the nose and creating an aerodynamic sideforce from the impact of the airflow against the side of the fuselage, but also partly due to banking. Many aircraft can easily be controlled in this manner, allowing the pilot to take both hands of the control yoke to attend to other tasks in the cockpit.

At any rate, the "kick out the crab" method of crosswind landing is simply a special case of the first type of skidding turn that we described in the above paragraph, where the pilot quickly yaws the nose to the desired heading, and then maintains whatever rudder pressure is required to hold that heading until the direction of the flight path through the airmass fully coincides with that heading. Unfortunately, in the case of "kicking out the crab", this ensures that if the moment of touchdown is delayed too long, the flight path will eventually coincide with the runway heading, which ensures that the ground track will be angled off in the downwind direction. This ensures that the aircraft cannot stay over the runway centerline. This is why the pilot should wait till the instant before touchdown to "kick out the crab"!

If the pilot "kicks out the crab" and the touchdown is delayed and the aircraft begins "drifting" in the downwind direction, the pilot can always salvage the situation by lowering the upwind wing to stop the downwind drift, while continuing apply whatever downwind rudder pressure is needed to keep the aircraft heading aligned with the runway heading. This would be a normal "sideslip" type of crosswind correction: the "kick out the crab" maneuver is really nothing more than an "incomplete" version of a normal "sideslip" crosswind correction.

Note that the "kick out the crab" techniques and the "sideslip" technique can be mixed. A pilot could fly the aircraft in a "coordinated", non-slipping, "crabbed" manner until just before touchdown, and he could then apply downwind rudder to yaw the nose into alignment with the runway heading, while he also applied upwind bank to prevent the flight path from curving, i.e. to prevent sideways "drift" from "accumulating". This is just the normal "sideslip" technique, delayed till shortly before the touchdown. In a strong crosswind, if the normal "sideslip" technique requires too much bank to allow for safe ground clearance of the low wing tip, the pilot could confine himself to a lesser bank angle, along with a lesser yaw angle that did not align the nose fully with the runway heading. Then shortly before touchdown, the pilot could apply extra rudder to yaw the nose into alignment with the runway heading. As with the normal "kick out the crab" technique, the goal would be to touch down immediately after the "extra" rudder was applied, before much sideways "drift" could "accumulate", i.e. before the flight path and ground track could curve very far in the downwind direction.

At this point the reader might be wondering why on earth anyone would intentionally use the a technique that can be described as "nothing more than an 'incomplete' version of a normal 'sideslip' crosswind correction", or "a wings-level skidding turn performed with the rudder, that happens to be interrupted by the wheels touching the ground before the ground track can curve very far away from the runway heading". This doesn't seem like an ideal way to land! In my own flying in "conventional" 3-axis light planes and sailplanes I always use the "sideslip" method of crosswind correction rather than the "kick out the crab" method. But note that the more massive an aircraft is, the less fuselage side-area it typically has in relation to its mass, so for a given yaw (slip or skid) angle, aerodynamic sideforces become somewhat less important in a heavy aircraft than they would be in a lighter aircraft. The actual shape of the fuselage plays a role too--we'll revisit this point below. Also, unlike tailwheel aircraft, tricycle-geared aircraft are not prone to ground-looping. Therefore with a tricycle-geared aircraft, the strength of the landing gear becomes the main limiting factor in the aircraft's ability to withstand touching down in a "sideways" attitude with respect to the ground track, regardless of whether the "sideways" attitude respect to the ground track is caused by the pilot flying the aircraft onto the runway in a "crabbed" attitude, or is caused by the pilot "kicking out the crab" too soon so that sideways "drift" accumulates before touchdown. However, if the runway is narrow, the former of the above two options is much better than the latter! If at the moment of touchdown the landing gear withstands a sideways loading and abruptly yaws the aircraft's heading into alignment with the aircraft's ground track, this will be little comfort if the ground track is aimed off the edge of the runway and the pilot doesn't have enough room to make the ground track curve back into alignment with the runway heading before the edge of the runway is reached! If an aircraft touches down with a strong sideways drift on a narrow runway, this is exactly what may happen. Therefore it may be better not to "kick out the crab" at all, than to "kick out the crab" much too early, unless the pilot is prepared to cancel out any downwind drift that develops before touchdown by lowering the upwind wing if necessary. Of course, if the pilot's timing is just right and the aircraft touches down immediately after the pilot has "kicked out the crab", then the sideways drift will be low regardless of the shape and size of the aircraft.

Aircraft with rudders that cannot be controlled independently of the ailerons (e.g. Ercoupes), or with rudders that serve as the primary roll control and thus cannot be used very well as an independent yaw control (e.g. the old MX ultralights), or with no rudders at all (e.g. weight-shift-controlled "trikes"), are simply flown onto the runway in a "crabbed", non-slipping manner. The main wheels must withstand a sideways loading at the instant of touchdown without collapsing, and the main wheels also must apply a yaw torque at the instant of touchdown to swing the nose of the aircraft into alignment with the direction of the ground track. Such aircraft always have "tricycle" landing gear. An aircraft with a tailwheel landing gear configuration would be prone to ground-looping when the main wheels, located ahead of the aircraft's CG, contacted the runway with the nose of the aircraft pointing in the "upwind" direction in relation to the ground track.

A few aircraft have specialized "crosswind" landing gear that are freely castoring (e.g. Cessna 190/195/LC-126) or can be manually be set by the pilot to match the expected "crab" angle at the instant of touchdown (e.g. Boeing B-52). These aircraft can simply be flown onto the runway in a "crabbed" non-slipping manner without imposing any sideways load on the landing gear. Another class of aircraft with specialized "crosswind" landing gear are hang gliders and paragliders: at the instant of touchdown, the pilot's feet are free to run in the direction of the ground track, even if this is quite different from the direction that the nose of the aircraft is actually pointing!

When the pilot uses the rudder to create some given angle between the aircraft heading and the actual direction of the flight path, the rate at which the flight path and ground track curve will be highly dependent on the shape of the aircraft. In other words, it is much easier to do a flat "skidding" turn in an aircraft with a slab-sided fuselage (and with a low mass in relation to the fuselage side area) than in an aircraft with a slender streamlined fuselage (and with a high mass in relation to the fuselage side area). In other words, when the pilot applies whatever rudder inputs are needed to keep the aircraft heading the same as the runway heading, this will create a higher rate of curvature of the flight path (and a faster accumulation of cross-runway "drift" in the ground track) in an aircraft with a flat slab-sided fuselage than in an aircraft with a slender streamlined fuselage. This difference arises not because the two aircraft "feel" the external meteorological wind in a different manner, but rather because the two aircraft "feel" the sideways relative wind created by the pilot's rudder input in a different manner. We'll explore the relationship between the shape of an aircraft and the aerodynamic sideforce generated by a sideways component in the relative wind in more detail in "The aerodynamic sideforce generated by a slip depends on the shape of the aircraft".

Could an aircraft's aerodynamic characteristics be such that it begins drifting in the upwind direction, not the downwind direction, when the pilot applies "downwind" rudder to "kick out the crab"? This would certainly drive home the point that the external, meteorological wind is not the true cause of the "sideways drift" that begins to accumulate after the pilot "kicks out the crab"! We'll explore this question in more detail in a later section of this tutorial.

 

For more practical reading on crosswind landing techniques, see Harvey S. Plourde's superb book "The Compleat Taildragger Pilot" (1991).

 

Advance to "The aerodynamic sideforce generated by a slip depends on the shape of the aircraft"

Up to the Aerophysics Exploration Pages index

Up to the Aeroexperiments site map

Copyright © 2004 aeroexperiments.org