I'm sure many of you are familiar with "rudder hardover", where an aeroplane's rudder deflects to an extreme position - full left or full right. Many of you will be aware that such an event, caused by a faulty Power Control Unit, was implicated in the loss of United Airlines Flight 585 and USAir Flight 427. A rudder hardover was also involved in an incident on Eastwind Airlines Flight 517 and the investigation into that incident unravelled the mysteries around the loss of the two aforementioned flights.

My question is about the effects of a rudder hardover on flight - and how it led to the two crashes. I'm interested in what happens, aerodynamically, after the rudder has moved to its extreme position and the sequence of events which follows. It's obvious to me that a rudder hardover must be an extremely destabilising event. But I'd like a more detailed explanation of precisely what happens. What effects does it cause? What happens to the affected aeroplane and why might it crash?


With a fully deflected rudder the airplane will fly with substantial sideslip, and by dropping the windward wing it can be held on a straight course. This is usually done intentionally with gliders and light aircraft when they want to lose altitude quickly, because the high sideslip angle will increase drag substantially. If $\zeta$ is your rudder deflection angle, the slip will normally have a sideslip angle $\beta$ of between -$0.5\cdot\zeta$ and -$\zeta$, depending on the details of the aircraft.

If the hardover happens at high speed, the resulting aerodynamic forces will certainly overstress the fuselage and tail structure, and the aircraft will break apart and crash. No amount of pilot skill can prevent this!

If the aircraft survives the hardover incident, the best the pilot can hope for is a controlled crash: The sideslip angle will cause the landing gear to be torn off, and the dropped windward wing will provoke a ground loop - if the pilot tries to level the wings during the flare, the resulting yawing motion will ground loop the plane in the other direction.

I once had the chance to fly the prototype of the Swift S-1, an excellent Polish aerobatic glider. During one maneuver my left pedal got stuck, and I saw myself already sideslipping the plane into the ground. In the remaining minutes I tried to get the pedals unstuck and got lucky, ending the flight with an uneventful landing. Afterwards, I found out that the pedal mechanism had caught the wire from the battery, and my maneuvering had freed it up again. This was the closest I ever came to land with a rudder hardover, and I am not eager to ever try this out for real.

  • 8
    $\begingroup$ I'm glad you survived the close call. You're much too valuable to this coming. $\endgroup$ – RoboKaren May 31 '15 at 19:42
  • $\begingroup$ Ps, your answer is based around unpowered flight. How would thrust steering affect the navigability equation? $\endgroup$ – RoboKaren May 31 '15 at 19:44
  • $\begingroup$ Agreed, mostly -- it takes quite a bit of work to land a jet safely with severe crosscontrol (the EGF4649 crew was able to do it with lots of aileron, and heavy tiller through the rollout). However, re: AA587, that required alternating rudder deflection to happen -- a simple full-deflection command is structurally fine at any speed below V-sub-a. $\endgroup$ – UnrecognizedFallingObject Jun 1 '15 at 4:09
  • 3
    $\begingroup$ @RoboKaren: Thrust would help a little in level flight by the yawing moment of asymmetric thrust, and again during landing by reducing the sink rate. The yawing moment due to thrust depends on engine location, but is generally small - remember, the tail is sized to compensate for failed engines on one side. $\endgroup$ – Peter Kämpf Jun 1 '15 at 8:31
  • 1
    $\begingroup$ @Ksery: It was the repeated deflection into the wind which tore the vertical off American Airlines 587. The vertical is sized for full deflection at zero sideslip and v$_A$, but not the added load of a sideslipping aircraft. And a commanded deflection is not a hardover in my book. $\endgroup$ – Peter Kämpf May 21 '18 at 20:34

Hardover of a control surface basically means that it has become jammed against a mechanical stop; think of having a poltergeist in your plane, applying full rudder (or aileron, or elevator) one way or another.

In the case of the rudder, it indeed will cause a severe yaw and roll with the resulting sideslip. Dealing with this requires a hefty helping of opposing aileron control (and spoileron, if your plane has that); this creates a forward slip, which keeps the plane on course at the cost of increased sink, which requires more pitch and power to overcome.

For most jets, this is hard on the crew, but doable: American Eagle 4649's crew shot an ILS into Bangor in blowing snow with their Embraer 135's rudder jammed full right, landing safely. However, in the 737's case, the rudder has so much low-speed authority that below a certain airspeed, there isn't enough aileron/spoileron roll authority to overcome the adverse roll from a hardover rudder. Worse yet, the 737 PCU problem not only sent the rudder hard over, it caused a reversal to occur -- the pilots stepped on the pedal they thought would make things better, and the rudder problem got worse, confusing the absolute snot out of them. You might be able to get out of it with differential power, provided it was applied soon enough to counter the roll before you went over on a wingtip and down...

  • $\begingroup$ Why would a rudder hard over cause the aircraft to roll? $\endgroup$ – Au101 Jun 1 '15 at 13:13
  • 3
    $\begingroup$ @Au101 -- because rudder input causes a plane to roll as well as yaw (this coupling is related to the infamously annoying Dutch roll). $\endgroup$ – UnrecognizedFallingObject Jun 1 '15 at 22:44
  • 1
    $\begingroup$ @Au101: Because rudder input causes the plane to yaw to one side, which causes the wing on the outside of the yaw to move faster than thee inside wing, which causes the outside wing to produce more lift than the inside wing, which lifts the outside wing relative to the inside wing, which causes the plane to roll in the direction of the rudder hardover. $\endgroup$ – Sean May 26 '18 at 23:40
  • $\begingroup$ Yes, diferential thrust saved this 747. fss.aero/accident-reports/look.php?report_key=1012 $\endgroup$ – Camille Goudeseune Jul 31 '18 at 16:52
  • 2
    $\begingroup$ @CamilleGoudeseune: Differential thrust plus having two rudders instead of just one. $\endgroup$ – Sean Jan 21 '19 at 4:49

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service, privacy policy and cookie policy

Not the answer you're looking for? Browse other questions tagged or ask your own question.