13
$\begingroup$

The pilots of Turkish Airlines Flight 1951 had multiple issues with the auto-throttle setting itself to idle during approach: despite manually adding throttle, the plane stalled and crashed before reaching the runway, from a sequence of events initiated by the reaction of the auto-land system to one faulty transducer.

The fault was masked because:

  • The PF was the first officer, his radio altimeter was working correctly.
  • Autothrottle in the 737 only looked (looks?) at the captain's side.
  • The glide path was intercepted from above and throttle reduction was expected.

The F/O was an ex military pilot with 4000 flight hours receiving line training, when the stick shaker activated he immediately gave full throttle. The captain took over command, during which the autothrottle set back to idle again. The plane then stalled and re-applying full thrust was ineffective. Full investigation report pdf here.

The plane was a Boeing 737-800. Had it been an A320, would the autothrottle have reacted in the same way, and set the throttle to idle upon detecting zero radio altitude?

$\endgroup$
  • $\begingroup$ Yes the question is if the auto-throttle switches to flare mode (I assume this is the same as Retard in a B737) with only one radio altimeter detecting flare height. Indeed the DSB report mentions on page 66: The conclusion is that the captain should have made a go-around in accordance with Turkish Airlines’ standard operating procedures once the approach during the ILS approach had not been stabilised at 1000 feet. ... $\endgroup$ – Koyovis Jun 22 '17 at 7:25
  • $\begingroup$ ...It also mentions on page 66: The captain bears the final responsibility for a safe flight and compliance with the statutory regulations and standard operating procedures as long as these do not conflict with the safety of the flight. It is likely that, in light of the research above, the captain did not see the continuation of the approach below 1000 feet as a contravention of the safety of the flight. $\endgroup$ – Koyovis Jun 22 '17 at 7:25
  • $\begingroup$ what's the point (and source) of the image? $\endgroup$ – Federico Jun 22 '17 at 11:42
  • 1
    $\begingroup$ Own photo, it shows the throttles, cockpit and working environment. $\endgroup$ – Koyovis Jun 22 '17 at 11:43
  • $\begingroup$ I know what it shows. I still don't see what does it add to the question. $\endgroup$ – Federico Jun 22 '17 at 13:28
12
$\begingroup$

The Airbus thrust levers can only be moved manually. The levers do not move on their own to reflect the auto-thrust setting.

There are detents for the different modes of auto-thrust operation. Regarding the RETARD mode whereby the thrust auto-idles, Airbus doesn't use that (except in autoland).* Hence the callout, "RETARD, RETARD," for the pilots to retard the levers before touchdown.

In a normal uneventful flight, the third detent from the top (CL) is where the thrust levers stay throughout the whole flight from thrust reduction after takeoff, to just before touchdown.

enter image description here
(A320 manual)

* Airbus released an Operations Engineering Bulletin (no. 38) in Mar 2011 (also see the Airbus brochure on the matter). What happened to the Turkish 737 could happen to an A320 due to an "erroneous radio altimeter height indication".

Still, in the scenario given once the pilot flying had applied full power (TO/GA detent) it would have stayed there (unlike the Turkish 737).


Thanks to @user40476 for bringing up this autoland feature, which made me look up further details.

$\endgroup$
2
$\begingroup$

In AIRBUS, The «Flare mode» brings automatically the thrust to idle provided the autothrust is under the dependance of the FMGS, in this case irrespective of the pilot action moving back the levers or not, the thrust will go to idle. On the other hand if the approach is done in SPEED mode AND WITHOUT THE AUTOPILOT USED IN PARALLEL, it is mandatory to move manually the levers to idle, otherwise the autothrust will increase the power after landing just to maintain the approach selected speed, in this case it is important to move the levers to idle during flare.

Please refer to the following website:

https://forum.aerosoft.com/index.php?/topic/113217-need-help-with-manual-landing-a320/#comment-762234

You will read:

Remember that "RETARD" is a reminder, not an order: however, with the A/THR engaged in a manual landing if you do not retard the thrust levers to idle the system will add thrust to keep the the aircraft on target speed, which will result in you continuing to fly gently a couple of feet above the runway! In an autoland the system will automatically reduce the thrust to idle

Similarly, please refer to:

https://www.skybrary.aero/index.php/Autoland

You may read:

The pilots must program the FMS (or tune the appropriate radio aids), configure the aircraft for landing and engage the autopilot and autothrust systems in the normal fashion. The Autoland system then provide inputs to the aircraft flight controls and adjusts the engine power settings in order to maintain the required approach profile and land the aircraft safely without pilot intervention. Some systems require the pilot to reduce thrust to idle when performing autoland. The Airbus requires the pilot to move the thrust levers to the idle positon when the autocallout calls "RETARD" at 10' RA. HOWEVER, the autothrust has already reduced the thrust to idle before this point - the retard call is to remind the pilot to match the thrust levers to the demanded thrust requirement. In all cases, the pilot must select reverse thrust settings. The autopilots will be disengaged after landing to taxi clear of the runway

As a summary if on approach the AP/FD is in approach mode, the autothrust is in thrust mode and automatic idle occurs even if the pilot doesn’t move the lever to idle. On the other hand if the autopilot is not used, and the autothrust is in « speed mode », manual reduction to idle is necessary. In all cases, at landing the « retard » aural annunciation is done.

Important note: on Airbus, a single radio altimeter failure will change the autothrust redundancy, at worst case an automatic disconnection will not prevent to reengage the autothrust, this will not prevent correct functioning of the autothrust, that is the idle is automatic if the autothrust is coupled to the autopilot; the idle should be manually forced if the autothrust is in SPEED mode and is not coupled to the autopilot.

Anyway bringing back the levers to idle is mandatory if reversers will be used

Thank you for ymb1 about RA. RAs are not voted

$\endgroup$
  • $\begingroup$ Your answer appears to directly contradict the FCOM citation in the accepted answer. Can you justify your statement? $\endgroup$ – user71659 Jun 7 at 18:07
  • $\begingroup$ I added more details, thank you $\endgroup$ – user40476 Jun 7 at 19:02
  • $\begingroup$ The citation concerns the case « autothrust »is used without autopilot, when autothrust is used with autopilot the thrust reduction to idle is automatic irrespective of the levers position. $\endgroup$ – user40476 Jun 7 at 19:22
  • $\begingroup$ @user 71659, please read my updates. Kind Regards $\endgroup$ – user40476 Jun 7 at 19:26
  • $\begingroup$ You're right and thanks for bringing up this matter. However please note that the paragraph about the RA voting is incorrect, as Airbus released an OEB about it. $\endgroup$ – ymb1 Jun 7 at 21:13

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.