Images from the Board Meeting on USAir Flight 427, March 23, 1999

A schematic of the hydraulic fluid flow path in the main rudder dual concentric servo valve after a jam of the servo valve's secondary slide (gold object) to the valve body or housing. This view shows the primary slide (silver object) porting fluid to both left and right rudder directions effectively canceling the rudder command from the jammed secondary slide. This condition is called "cross flow" by the designers.
 

A schematic of the hydraulic fluid flow path in the main rudder dual concentric servo valve after a jam of the servo valve's secondary slide (gold object) to the valve body or housing. This view shows fluid porting for reverse rudder operation. With the secondary slide jammed to the body and an input command to move the primary slide in a direction opposite of the jam, the primary slide moves beyond its intended design point and ports fluid for a left rudder movement with a right rudder pedal command.
 

Simulations of 737 Rudder Hardover Events
These movies illustrate the aircraft control and response concepts involved in a rudder hardover event, and depict NTSB simulations of three 737 rudder hardover events.  These simulations were developed during the investigation of two 737 accidents and one 737 in flight control incident.  All animations require the free RealPlayer from Real Networks; file size is 1M or less unless otherwise indicated.

Illustration of Airplane Motion and Control Surfaces

This movie illustrates how the airplane’s flight control surfaces are used to control motion about the airplane’s vertical, longitudinal, and lateral axes.  The control surface movement and resulting airplane motion is shown independently for each axis.  The rudder controls yaw (rotation about the vertical axis); the elevators control pitch (rotation about the lateral axis); and the ailerons and spoilers control roll (rotation about the longitudinal axis).

 

Depiction of Rudder Blowdown

The control surfaces move in response to pilot control inputs through the action of hydraulic actuators.  The actuators push the control surface away from its faired, or neutral position, into the oncoming airstream.  The force of the airstream tends to oppose the surface deflection and to return the surface to its neutral position.  This opposing force increases with airspeed.  At lower airspeeds, the opposing force is not as great and the actuator can move the surface further into the airstream, resulting in a larger deflection; at higher airspeeds, the actuator can not move the surface as far, since the airstream is “blowing” the surface back “down” to its neutral position.  For the rudder surface, this phenomenon is called “Rudder Blowdown.”

The following aviation event reconstructions exhibit selected data from the Flight Data Recorder, simulations, and edited comments from the Cockpit Voice Recorder transcripts.  These simulations do not depict the weather or visibility conditions at the time of each event.  The cockpit simulations are modeled on a Boeing 737-300 cockpit interior. 

Airplane Simulation Chase View:
  USAir Flight 427
  Aliquippa, PA
  September 8, 1994
  Boeing 737-300


 

USAir 427 Airplane Simulation: Cockpit View

 

Airplane Simulation Chase View:
  United Airlines Flight 585
  Colorado Springs, CO
  March 3, 1991
  Boeing 737-200

 

United Airlines 585 Airplane Simulation: Cockpit View (1.6M)
 

Airplane Simulation Chase View:
  Eastwind Airlines Flight 517
  Richmond, VA
  June 9, 1996
  Boeing 737-200

 

(1.6M) Eastwind Airlines 517 Airplane Simulation: Cockpit View
 

The following simulations represent the rudder pedal positions and leg orientations of the corresponding event. These simulations were developed as an educational aid although, whenever possible, the scaling and motions of the manikin and cockpit control were modeled after those of the Boeing 737 event being studied. For all the simulations, the color of the manikin's leg indicates the force output. Blue indicates that no force is being applied to the rudder pedal. Yellow indicates a normal force application while red indicates a large force application, larger than that needed for normal cockpit operations.

 (1.3M) Manikin USAir 427: The first portion of this simulation shows a typical rudder check performed by the pilots on the ground. The pilot pushes the right pedal slowly to its full extent to check that the use is easy and free of obstructions. He then relaxes the right leg, then pushes the left pedal to its full extent.

The next portion of this simulation shows the first officer's legs during a general rudder reversal event. The pilot attempts to enter a small amount of right rudder, but something goes wrong. Almost immediately, the right pedal begins to come up rather than down. The pilot is surprised, and pushes harder to make the pedal go down. Although he pushes as hard as he can, the pedal continues to rise until it reaches the full upper position. Notice that as long as the manikin pushes on the pedal the reversal continues but if the manikin removes the force on the pedal, the rudder returns to a neutral position.

The last portion of the simulation shows the rudder pedal motion and the corresponding leg motion for the Pittsburgh USAir 427 accident, as reconstructed by Safety Board staff. This model accounts for the aerodynamic forces acting on the rudder and for factors such as cable stretch. By pushing hard, the pilot is able to move the pedals closer to neutral even though the reversal is making the rudder move to a full left position.

Manikin United Airlines 585: This simulation shows the rudder pedal motion and the corresponding leg motion of the Colorado Springs United 585 accident. This airplane was flying at a slower airspeed than the 427 flight, and there is a greater motion of the rudder and the rudder pedal due to the lower aerodynamic forces opposing rudder motion.
 

Manikin Eastwinds Airlines 517: This simulation shows the rudder pedal motion and the corresponding leg motion of the Richmond Eastwinds 517 incident. This airplane was flying at a faster airspeed than both the 427 and 585 flights, and there is less motion of the rudder and the rudder pedal due to the higher aerodynamic forces opposing rudder motion. The pilot also reported 'standing' on the rudder pedal in an effort to generate greater force on the pedal, as seen on the simulation.


 

The following video clips regard the NTSB's Wake Vortex Flight Test, conducted at the FAA's Technical Center in Atlantic City, New Jersey in September, 1995.

(4.8M) This video includes the airplanes that participated in the tests as well as representative samples of flight test conditions. The video includes shots of each of the participating airplanes - the vortex generating B727, the vortex-encountering B737, and the T-33 chase plane. The video also documents wing-mounted smoke generators on the B727, and shows several test conditions, in which the B737 flies into the smoke-identified cores of the B727's wake vortices. The encounters are shown from the chase plane, the B737's cockpit camera, and the B737's tail-mounted camera.
 

(1.5M) This video is a short segment extracted from the previous video, and shows only the test conditions, in which the B737 flies into the smoke-identified cores of the B727's wake vortices.