Aileron reversal led to horror ride

source: Aircraft Accidents Investigation and Prevention Department, Portugal

An aileron rigging error led to the extraordinary and frightening flight by an Embraer ERJ 190 over Portugal last November, Portugal’s Aviation Accidents Prevention and Investigation Department has said in a preliminary report. The single aisle 97-seat airliner encountered flight control difficulties soon after take-off at 1331, local time on 11 November 2018 and the crew declared a Mayday.

The aircraft belonged to Kazakhstan’s national airline Air Astana and was on a ferry flight to the country’s largest city, Almaty, after undergoing C-check maintenance in Portugal. There were three pilots and three engineers on board.

In the words of the Portuguese report, ‘Immediately after take-off, with adverse meteorological conditions, the crew felt that the aircraft was not responding adequately to the commands, developing oscillatory wing movements.

In the words of the interim report, ‘the performed trajectories caused the aircraft and the technicians on board to sustain intense G-forces, as well as the aircraft complete loss of control for some moments at multiple instances’.

‘All on board were physically and emotionally shaken,’ the report says, although there was only one minor physical injury—to an engineer’s leg.

Faced with flamboyantly unstable responses to control inputs but no warnings of any aircraft system failures, the crew reverted the fly-by-wire aircraft’s controls to direct law (selectable by switch on the pedestal of the ERJ-190), in which the flight control module is removed from the control system and control surfaces respond in proportion to human inputs on the yoke and pedals.

This improved the situation considerably without restoring normal operation and considerable difficulties persisted in roll control. The crew responded by minimising roll inputs wherever possible.

The crew was able to recover sufficient control of the aircraft and divert to Beja, 125 km south-east of Lisbon. The aircraft landed on its third approach about 15:36.

The preliminary investigation says a modification carried out in accordance with an Embraer Service Bulletin had changed a cable routing support near rib 21 of the aircraft. This change made it harder to understand the maintenance instructions and spot reversed aileron cables.

‘During maintenance, the engine-indicating and crew-alerting system (EICAS) displayed a caution message: “FLT CTRL NO DISPATCH”. This meant that one of the components of the flight control system had failed. Troubleshooting activities by the maintenance service provider, supported by the aircraft manufacturer, lasted 11 days. However, the ailerons’ cables reversal was not identified in this period,’ the reports says, noting that the incorrect rigging was not identified in the pre-take-off flight controls checks.


  1. Yes Many Years ago in Gisborne I watched a Completely Rebuild DHC-2 Beaver Start on it’s Test Flight only to Drop it’s wing into the Ground & Wrap itself into a Ball ! Caused by Crossed Control Ailerons !!

  2. GA trained 30 years ago/now RAA, I do a complete control surface response check (Flight Controls – Free & [importantly] Correct) before the first flight of the day, how come these x 3 pilots didn’t?
    Sounds to me they wiggled the controls okay, but no one actually looked out of the window (asked ground crew) to see if the aileron was going up when it should have been down.
    Same question of the maintenance/repair crew – I would expect a full check before sign off with a special emphasis on any systems worked on.
    Pretty basic failure -scary !- back to school.

    • Control surfaces still go up and down Walter however big the aircraft. Surely someone checks the controls are CORRECT as well as free?

        • You don ‘t have to fly jets to be passionate about aviation Walter. Instead of insulting us GA people how about trying to inform us? Does the ERJ have a Flt Ctrl page on the FMS? All Sean was saying above was that there must have been some pre-flight process for checking control movement, visual or electronic, and in the case reported this seemed not to have been done

          • The Embraer has a screen you can bring up which would have showed the flight controls being reversed. Checking it is part of one of the pre-takeoff checklists. The pilots apparently missed that the ailerons were cross-connected – probably they moved the yoke and verified that the ailerons moved, but not which direction. This would be a example of seeing what you expect to see during a preflight. I am *far* more puzzled how the people re-rigging the ailerons didn’t verify operation was in alignment with the controls – I’d think this would be something you would just assume could be wrong with cable style rigging.

    • Not sure they would have seen the ailerons from the cockpit. And I wouldn’t expect that an airline would have a ground crew person available to check, as they do in the military.

  3. Where humans and machines are merged together to be operated such events will happen, fact of life!

  4. Every time an aircraft leaves maintenance, a very thorough check is crucial. Crossed controls won’t suddenly happen during normal operations – the risk is biggest following being in the shop. The same counts for checking no tools have been forgotten anywhere in the aircraft – but hopefully aircraft engineers following similar procedures to hospital operating theatres, ensuring tools are back in their spot and accounted for before signing off an aircraft.

  5. Many years ago I was ferrying a C172 from Goulburn to Moruya after it had just undergone a 100 hourly inspection. I was about half way across the Great Dividing Range on my way back to Moruya when all of a sudden the engine RPM dropped about 150 so I immediately carried out a magneto check which indicated the left hand Mag was inoperative, but I continued on at reduced power (what else could I do) and safely landed at Moruya. We lifted the engine cowling and immediately saw the left hand magneto completely detached from the engine and hanging by the wires. It turned out a mechanic had removed the magneto from the engine and finger tightened the long bolts that hold it onto the engine and went off to morning tea with everyone else with the intention of properly tightening the holding bolts when he returned to work, trouble is he forgot about it and apparently the engine vibration caused the loose bolts to slowly back out which they eventually did.
    Needless to say that after that worrying episode, everytime I flew an A/C back from Goulburn to Moruya after maintenance which I frequently did, I very carefully checked everything I could before departing because about the last place on earth you want to have an engine failure is when you’re halfway across the Great Dividing Range as it is almost impossible to find a suitable forced landing area, as can be appreciated!

  6. Well said Ian.
    I have a background in large jet maintenance & GA flying. Both sizes of aeroplanes have flight controls that need to be checked for “full and free movement in the correct sense” & there is never a valid excuse for not checking this.

  7. Funny thing Walter, while reading all your arrogantly belligerent comments where you seem to have a sneering contempt for anyone who isn’t an airline pilot and is in your view a vastly inferior general aviation pilot. It occured to me that your average G/A pilot actually does more hands on flying of their A/C in one day than what your average airline pilot does in a whole week, or maybe in a month even. I once sat in the check captains seat on a 747 on the way to NZ from Sydney a few years ago, the Captain being a long time friend of mine and noted the actual hands on flying of that A/C over the entire flight would have been no more than 10 minutes or so at most. The flight crew spend most of their time monitoring the automated flight control and navigation equipment and radio communications etc than actually flying the A/C and in fact the airplane does most of the flying all by itself, so have a guess who I reckon is a better and more experienced hands on flying pilot out of the two Walter, a G/A pilot or an airline pilot operator who sits in a plane that most of the time flies itself. I also noticed there are two other pilots and a flight engineer sharing the workload whereas we have to do the whole lot on our own. With the greatest of respect, I invite you to have a bit of a think about that the next time you sneeringly look down your nose at one of us thank you!

    • Haha people like you amuse me! Do you really think Airline drivers don’t hand fly? Plenty of Airline drivers own/operate light A/C! Amazing what some out there believe to be true, it’s a fantasy land for some! -:)
      I enjoyed yr little story re the mag falling off the engine, I just shook my head ! -:)
      No need to thank me either -:)

      • Trouble is Walter the Mag actually did fall off the engine because I actually saw it with my own eyes same as the person that was there with me and my front seat flight on the Qantas B747 actually happened too no matter how much you ‘think’ it didn’t happen. I also take great offence at anyone insinuating I’m a liar, which I am most certainly not. As for the B747 flight to NZ, the Captain manually flew the A/C off the runway and engaged the autopilot as we were climbing out over Botany Bay and his hands only went back onto the controls on final approach to Auckland and just like I said, the aircraft flew itself throughout the entire flight baring the takeoff and landing at its destination no matter how much you declare it didn’t happen!

      • Forget him Paul. He isn’t worth the angst. In fact I am beginning t think he is a fraud. Nobody could be in the industry for 40 years as he claims with that attitude.

  8. All true and one thing I do know from way back is the fact that one of the biggest killers of pilots is arrogant overconfidence and bearing that in mind it’s a wonder our friend has lasted as long as he reckons he has, if he is fair dinkum that is, which as you say is doubtful and from now on I will studiously ignore the man!

  9. I was flying wide body jets and most of them had Control surface indicators which displayed actual deflection. Those indicators were not connected through FCS. Malfunction was obvious when idicators failed to follow surface movements or vice versa. In either case, we considered airplane undafe for the task.

Comments are closed.