Bjorn’s Corner: Airbus’ A321neo has a pitch-up issue (now with a second update)

By Bjorn Fehrm

July 19, 2019, ©. Leeham News: The European Aviation Safety Agency EASA has issued an Air Worthiness Directive (AD) to instruct operators of the Airbus A321neo of a Pitch instability issue.

EASA writes “excessive pitch attitude can occur in certain conditions and during specific manoeuvres. This condition, if not corrected, could result in reduced control of the aeroplane.”

We analyze how this is similar or different to the Boeing 737 MAX pitch instability issues.

NOTE 2: We have got a further update from Airbus, see below.

A321neo pitch instability

The AD makes clear the issue is connected to a software version of the A321neo Elevator Aileron Computers (ELACs), the Fly-By-Wire computers used for the A320 series control of pitch and roll.

As the AD does not apply to the in-service A320neo, the issue is not connected to the pitch instability which comes as a natural consequence of mounting the larger neo engines on the A320 series. It’s restricted to how the present A321neo version of the ELACs handles the aircraft’s controls in an excessive pitch up condition in an isolated flight mode, the Flare mode when landing.

Note 2: Since writing the article and our first note Airbus has provided us further information, given verbatim here: 

The origin of the “excessive pitch” is associated to a reduction in efficiency of the angle of attack protection in a very remote combination of conditions, these being: The need for the crew to perform a dynamic manoeuver (such as a go around) AND the aircraft being below 100ft AND specific landing configuration AND a very aft CG.

  • The A321NEO flight controls commands were modified, in response to customer requests, for certain flight phases to improve the dynamism of the aircraft. The changes introduced, are not linked to the size, location or weight associated to the installation of the engines.
  • It is these changes, which, together with some existing control characteristics (again not related to the size, location or weight of the engines) mean that in the very remote combination of conditions, described above, they could lead to the reduction in efficiency of the angle of attack protection, and could lead to “excessive pitch”.
  • This was identified during extensive computer based simulator testing.  It was not identified neither during flight tests nor in response to in-service reports nor as part of studies being performed in relationship to any other events related to other aircraft types which are currently being widely discussed

Our comment to the new information: The Airbus information explains why the issue is limited to the A321neo. There were changes made to its FBW in the present release of the software to make the aircraft react livelier to commands in approach configuration Full flaps and when in FBW Flare mode. These changes can in a very remote case render the stall protection (Airbus FBW protects you from entering a stall) less effective. It wasn’t detected in test flights of the A321neo as it requires the Pilot to command a very strong pitch up when in Flare mode and keep this command when entering the pre-stall and stall region of the aircraft. If you pitch up an Airbus violently and don’t stop the pitch before coming close to stall, you’d expect the aircraft to always hinder you from entering the stall. In this remote case with a very strong-handed pilot and an aft CG, Airbus found this can be less effective. The AD is about avoiding this remote case, to guarantee the expected stall protection is at all times effective. The restriction is to avoid the most aft CG positions for the aircraft until the next release of the FBW software is available 3Q2020.

Pitch instability from larger engines

This part of the article is a general discussion about the physics involved in the pitch stability of aircraft. Read my series about this here for more background, it continues for nine more weeks on the Fridays. With the above explanation from Airbus, it’s not connected to the issued AD. The larger engines for the A320neo series are controlled by the aircraft’s inherent pitch stability which is constituted by the aircraft’s horizontal Stabilator and Elevator and its FBW which controls these aerodynamic surfaces.

Like the 737 MAX, the A319/320/321neos are affected by the mounting of larger engines with their larger nacelles ahead of the center of gravity, Figure 1. It follows the aerodynamic rule: any surface projecting an aerodynamic horizontal surface ahead of the Center of Gravity (CG) contributes to a Pitch-up moment and any surface projecting an aerodynamic horizontal surface behind the CG contributes to a Pitch-down moment. This is why aircraft have horizontal stabilizers behind the CG to balance the aircraft in pitch. The elevator can then control a balanced aircraft in pitch.

Figure 1. A321 shown on top of A321neo. The larger engine nacelles are marked with a violet color. Source: Airbus and Leeham Co.

As for the MAX, the larger nacelles for the A321neo are placed further up relative to the wing (Figure 2) but no further forward. The nacelles are larger than the 737 nacelles, however, so the pitch up moment from the engines in isolation can be similar. As the engines sit lower on the wing their combined effect with the rest of the aircraft and its wing is not the same as the 737 MAX. According to Airbus, it has not been forced to do anything special to control the changed pitch balance of the A321neo.

Figure 2. A321 below and A321neo above. The neo has larger engine nacelles, mounted higher relative to the wing. Source: Airbus and Leeham Co.

The AD and the explanation from Airbus singles out recently changed flight laws in the ELACs for the A321neo to be the root cause for the pitch-up described in the AD. It allows the Pilot to command fast dynamic pitch ups when in Flare mode (which is just before landing) and at a very aft CG. If the pilot keeps commanding the fast pitch-up into the stall region it can overwhelm the present anti-stall algorithm. The FBW software needs a change so this condition does not occur and the Airbus anti-stall function works as described in the Flight Manual. Until this change is fielded 3Q202o there are restrictions to the aircraft’s aft CG limit to avoid the described condition to appear.

The takeaways

There are several takeaways from the above (changed since the original article):

  • As we have written in several articles, pitch stabilities comes down to how several parts of an airliner work together.
  • Reduced pitch stability at certain regions of the Angle of Attack envelope when testing the aircraft in wind tunnel with fixed tailplane surfaces is present for many aircraft types.
  • Different aircraft implement different means to make the aircraft stable in pitch. It can be made with aerodynamic surfaces like the ventral strakes seen on the underside of the tail of the Learjet or with control of the stabilator and its elevator in the region of changed stability. It the latter case it’s important how the control is made to produce a safe aircraft.
  • In the case of the MAX, software-based control logic is used, controlling the movements of the horizontal stabilizer and elevator. It’s key how these controls are designed, tested and implemented.
  • The original MAX implementation was inacceptably badly done. It relied on a single sensor, commanded unnecessary repeated nose-down trim commands and didn’t have any global limitation on its authority.
  • The updated MAX software has an improved trigger where both Angle of Attack sensors values are compared (and it deactivates the function if these disagree), it only commands one nose-down trim sequence and it has a global limitation guaranteeing the Pilot can always take the aircraft out of a dive. In my judgment, this makes the 737 MAX safe to fly with MCAS. I’m on the record with this judgment, and I will not hesitate to fly on the updated MAX.

It also incentives me to continue my Stability and Control Corners where I will dwell deeper on the differences between a conventionally controlled aircraft and an FBW aircraft.

101 Comments on “Bjorn’s Corner: Airbus’ A321neo has a pitch-up issue (now with a second update)

  1. Interesting…
    Is the pitch up condition more severe with the CFM engines or is it similar for both PW and CFM engines. Also if it affects the A321neo, why does it not affects the A320neo and also the A319neo?

    • The a320neo and 319neo have shorter fuselage. Also they require Less thrust in response to average gross weight compared to the a321neo.

      Actually best way to reduce this tendency is to widen the wings to something as wide as 757. The 737max and a320neos series were bassically stretched fuselage with larger engines but little modification in wingspan width and breath.

    • @Vince

      I read that is the same issue for CFM as for PW.

      I supose that A321neo has a little bit different algorithm in ELAC then A320neo.

    • Exactly. Shouldn’t a smaller aircraft be more affected than a larger one?

    • The 321neo is a stretch so to avoid tail strike in landing lands at a different aoa than the 319/320. This is controlled by the landing flap configuration.

      That’s the starting point of this problem. Interestingly not the first time airbus flight laws have had issues around landing config.

  2. Presumably Airbus can just dial this deviant behaviour out?Or does it start getting complicated with recertification and different type certificates,etc.

    • FBW gives the OEM better tools and finer-grained authority to fix such issues. It should be a software update fix.

  3. Hi Bjorn,

    Why do you think the affect of the nacelles on the pitching moment balance is so non-linear with AoA, at least in the case of the MAX? I’m puzzling over this a bit.

    • Because the flow and by it lift characteristics over the AoA range for a nacelle is different to a transonic wing. It’s a blunt-nosed cylindrical lifting surface with rounded nacelle lips where the middle air cylinder gets sucked into the engine.
      Flow over such surfaces stays attached to a higher AoA than for a clean wing, dependent on the nacelle lip shapes.

    • Enter “lift curve AoA” into google. Then look at any graph. It may help. Lift with AoA is typically not linear.

      • In the AoA region before stall onset for wings and before separation for bodies of revolution it is typically, and remarkably, close to linear.

        • Yes, a typical graph does show a drop off in the pre-stall region of the envelope. Indeed many graphs show a significant drop off. So at some point it is linear.

          But, the pre-stall region is iffy because the airflow is on the verge of becoming unstable. All that is needed is a vertical gust to cause the airflow to become unstable causing a stall. Not a good idea to be there.

          It’s where the pre-stall region begins. The wing/engine mounting on the 737 MAX means the safe AoA before stall is smaller. The safe AoA before stall addresses gusts of winds and other external forces. So it’s below the pre-stall region. Going above it is not a good idea.

          The 737 MAX has a narrow safe envelope in pitch. Hence MCAS.

          Can pilots be trained to keep within the narrow envelope.
          Most are now saying training will be required. That may even apply to Boeing, although they have shown significant resistence to training.

          But we will see what outside regulators say!

  4. Airbus are smart to verify the neo for similar problems as the max, or was there incidents in service that promted EASA actions?

  5. I’m glad that A321 is “full FBW” and not “1/4 FBW” like MAX and I’m glad that Airbus + EASA came with that AD. So before ELAC update pilots shall be careful and avoid certain “far from normal” conditions and ELAC needs to be tweaked soon.

    “The updated MAX software makes it safe. I’m on the record with this judgment, and I will not hesitate to fly on the updated MAX.”

    Are you sure? Are you, or anybody, tested it to the end’s end? Are you sure MCAS is the only flaw of MAX is needed to be addressed? On which basis you plant such global assessment now? As far I can see for now they’re only yours bold suppositions. As far as I can see for example MAX’s chip/software is flawed, MAX’s manual trim system is flawed etc. etc.

    Wasn’t better to wait until the end of this MAX debacle and then make such asses? I’m against prematural, unproven “cutting corners” assessments.

    I know many pilots/ aviation specialists tend to “know” that this plane is 737 family which was doing great for 50 years, so it will be great again etc. But those persons miss that all was before MAX, and this completely new chapter, sometimes I wish to shout out: “stop with magic spells and back to reality”.

    For now, when I can see as Boeing is still not transparent and still cuts corners with testing (eg. FAA found the chip problem, not Boeing) I will really hesitate to fly MAX, as many I think.

    • I have reviewed the initial MCAS function and the fixed one (and I have a background in aircraft stability and control). Boeing is one of the top engineering companies in the business when the shareholder value-oriented board and top finance management keeps out of the loop. I’m convinced this is the case now.
      Boeing has lost an aircraft development budget from the debacle and it will lose another if the MAX doesn’t fly impeccably after this update. Any other issues are fixed in this revision, be sure. If not, the board and whole top management shall be replaced, as I’ve said in previous Corners.

      Re my faith; there is a limit how much you shall trust an OEM but also a limit how far the scaremongers shall be allowed to black paint the technical competency of one of our top OEMs without rebuttal. The engineers were handcuffed for the initial implementation, they cannot be for the fix. The transparency is just too high at the moment, any management doing it will have sealed its fate.

      • Your suggestion thst nacelles on Airbus airplanes act as foreplanes is wrong. Airbus airplanes have pylons. So do Boeing airplanes, with the exception of the 737 MAX.

        To be clear, the pylon means that the boundary layer surrounding the nacelle flows under the wing with adequate separation from the wing. Without a pylon, the top part of the boundary layer surrounding the nacelle flows over the wing. This produces the lift.

        The 737 MAX is unique in it’s wing/engine mounting.

        • um. no. the 737 max engine is mounted on a pylon. it is not unique in any meaningful way.

          • I’ll get my microscope out. Look at it from the front. The nacelle is above the wing.

            As I said elsewhere the nacelle boundary layer must go under the wing without interfering with the wing.

            Sorry, not accepting the bracket between the wing and the engine as a pylon

          • @philip, google the coanda effect and learn just a little something about aerodynamics. I’m reasonably confident that Boeing’s aero department knows just a bit more than you….

          • People really need to listen to Philip. He is the emerging expert on all this.

          • Where has someone said Boeing disputes his comment.?
            Its an interesting point of view he making, that its more of a strut than a pylon

            The very first 737 never had a pylon ( or strut) supporting the engine to the wing ( maybe brackets internally)

          • If is faired into the wing per the original 737, then its not a pylon.

            Otherwise its a pylon. Sheese Louise

          • Boeings patents call the load bearing support structure between engine and wing ‘a strut’
            The Fairbanks Daily News -Miner may call it a pylon

          • @bilbo

            I don’t need to google it. Aren’t you making my point. You need to read more carefully, then you might learn something

          • Yea and Airbus calls Winglets by a stupid name as well.

            they are still Winglets.

      • @Bjorn

        I see that you have strong belief in Boeing’s engineering, based on the past, I understand it, but don’t share it. For me, as it is now, Boeing is troubled engineering company, was – let’s say in American way – a brilliant one, but this was in past. Doesn’t matter if it’s now because of management problem or engineers problem – one group without another wouldn’t screwed it up so colossally.

        For now your strong safety assessment of MAX that you made is based on beliefs but not on facts :/

        Why I state this? Because is still long roads for Boeing to make MAX fly again – MCAS still needs tweaking at least, chip/software problem needs to be resolved, trim wheel system etc. etc. I even suppose that Boeing is still hiding some flaws because wants to make it MAX fly again as soon as possible as low cost possible – is still not transparent and still challenging regulators eg. against simulator training for pilots.

        I share your convince that MAX will fly again, I think it’s probably but not 100% sure, maybe 2020 maybe 2021 but also maybe never – the case is developmental. But I don’t share your current belief it will be a safe plane, rather in best case scenario “deemed as safe” =with hidden & covered up flaws. Boeing is like a junkie that wants convince everybody that is not taking anymore, and he thinks is so smart in this, and everybody else is stupid (quite strong, sorry, but I think it makes a good parabola). Question is: shall I, will I, trust to a word of a junkie?

        Your current assessment, anyway, premature :/ is in fact a belief… but I’m starting to repeat myself… so time to end.

      • Björn, I think you meant “scrutiny” instead of “transparency”. Of the latter, there is still none, at all, at least to the public.

      • @Bjorn, well time will tell.

        I can’t see how one can be so sure that the engineers are presently not handcuffed, management out of the loop, free to fix it properly unless one were to be embedded in their engineering team and seeing what it’s like first hand.

        I feel the evidence is to the contrary. Boeing’s own public statements and actions suggests that the engineers are still restricted in what they can do; they’re still manufacturing aircraft by the hundreds that can’t fly, and have said very early on that it’s fixable in software.

        That pretty much precludes hardware fixes then. They’re not putting a bigger stabiliser on, they’re not increasing the elevator size, they’re probably not doing anything about the size of the trim wheels or replacing any of the flight control computer hardware, and they’re not putting in triplicate AoA sensors or a full FBW system. The engineers now are not free to pursue these avenues.

        Which is a pity because, on currently available evidence, the software fix failed to pass tests that the FAA wanted to carry out which, I’m guessing, Boeing didn’t anticipate being required. Arguably, all bets are off at this point in time.

        Plus I don’t see why a management which previously hasn’t listened to engineering sense, seemingly for a couple of decades, would decide to keep out of it now. Their own personal shareholdings, their employments, reputations and possibly their liberties are on the line (I’m sure that their lawyers have mentioned the VW cases to them…). It might not look it on the outside, but they really are fighting for their own personal survival. And we’re expecting them to take a cool step backwards, and tell the engineers “take however long you need, chaps, do whatever is necessary.”? Apologies, but I just can’t buy it.

        As you yourself has already pointed out, they’ve spent more money on this tragedy now than they would have developing a whole new aircraft, which many would say it the only “proper” fix. And they’re not doing that either.

        • Boeing probably have different teams working on different solutions. I can remeber PW4000 surge problems that initally should be solved by a software fix but finally got new hardware and software.

          • Well if they have anyone working on airframe changes, they’re wasting a lot of time building them in the current shape.

            If hardware changes are required, they’re going to have to be very tiny ones, otherwise the re-work cost on all those hundreds of airframes they’ve been building since March will be massively expensive.

      • The flight test vehicles (FTV) could be many, many planes and in every imaginable scenario possible. In fact maybe more planes than even for the first time the MAX was tested. Why? The two casualties of almost brand new hulls. They don’t just have to prove the MAX is air worthy; they have to prove it can never fall from the sky again under any circumstance associated with flight control and stability. After Boeing puts on many flight hours on the MAX 8, 9 and 10, SW, UA and AA, I think most likely will have to put each MAX through its paces. It’s hard to make comparisons, but this 737 MAX situation could outweigh the DC-10 cargo door latch by many folds when everything is considered.

  6. I think this acticle needs to concentrate on the pylon or lack of one. The pylon separates the airflow over the wing from the airflow around the nacelle. The consequence of that is that the engines on the A32X are not acting as foreplanes or lifting surfaces.

    I remember a previous article from LNA where it showed a graphic of the airflow around the engine of the A330neo. The engines on the A330neo are also forward. The graphic shows clear separation of the airflows, so interference in the lift of the wing is minimised.

    The A350 has it’s engines mounted forward. The A220 has it’s engines mounted forward. So forward mounted engines on Airbus airplanes is now standard.

    This comes to the 777X. Pictures strongly suggest that the engines are also mounted forward. But the 777X has a pylon to separate the airflow of the engine and the wing. We come to the 787: The same. I think the NMA will do the same, if built.

    There is actually nothing wrong with mounting engines forward. Indeed, I think it’s going to be the norm. Indeed, it is the norm. It does help reduce engine/wing interference, but there does need to be a pylon to ensure the airflow under the wing settles. It also helps with ensuring CofG is well forward of the CofL. But, the pylons do need to be cantilevers to transmit the load backwards to take away the torsional effects of having the engine forward – that means the pylons have to be stronger and stiffer.

    So the 737 MAX stands out because of a lack of a pylon. Indeed a frontal graphic of the 737 MAX shows engine nacelles above the wing. That does not apply to the 737 NG. The 737 NG does have a pylon, but pictures do suggest the existence of wing/engine interference because of insufficient separation between wing and engine.

    This then comes to the sharklets on the 737 MAX as opposed to winglets. The reason for the sharklets is lateral airflow above, and importantly, below the wings. This is a long conversation, but the bottom line is that wing aerodynamics are very messy on the 737 MAX because of engine interference.

    Boeing won’t develop any other airplane with sharkets when engines are mounted under the wings but then Boeing won’t develop any other airplane without a pylon when engines are mounted under wings.

    Bottom line, with regard to the 737 MAX, the engine/wing mounting is interfering in the lift aerodynamics. The interference is much more than Boeing will admit,

    With regard to the A321 issue. It’s calibration. They got it wrong, but not dangerously wrong. It is a software patch of the alieron/elevator computers.

    Calibration needs to be carefully done. But calibration needs detailed performance charts/graphs. That in turn means the CPUs need to be big enough to hold the detailed perfrormane charts/graphs and the software algorithms that control the movement of control surfaces based on the charts/graphs.

      • Thanks, it is. Airbus are very proud of the pylon. They subjected it to intense CFD to ensure there was no interference between the wing and the engine.

    • Philip, interesting point re airflow around the engine. when I look at an image of the MAX from the front it appears that the top of the nacelle is level with the top of the wing (not above it). Would be interesting to see modeling of airflow around that engine. I’m sure Boeing has done it.

      To your point, whether it is the MAX, 330neo or 320neo, they all have engines mounted higher than the original design. I suspect that at high angle of attack, all of them would see significant airflow that goes around the engine and then over the wing rather than below.

      • Not sure what front view graphic you are looking at.

        With regard to the rest of it … Don’t think so. If true moving the engines forward is a bad idea not a good idea. Moving the engines forward is a good idea.

        By the way, moving the engines forward allows a pylon with less depth because the airflow settles before it reaches the wing.

        With regard to your confidence that the 737 MAX is no different. It needs a pylon to be no different.

    • Philip,

      Here is a picture (hopefully it gets posted here) showing the engine nacelle mounting on the 737 MAX. Look at that space between the nacelle chevrons and the wing underside. Seems like a definite pylon to me. Also notice that the top of the nacelle ends right at the wing leading edge.

      At an altitude of 35,000 ft on a standard day, cruising at 0.79 M, and a nacelle top length of 12.5 ft, the turbulent boundary layer will be less than 1.9 inches thick (I assume worst case scenario where the boundary layer starts off turbulent). I don’t think the nacelle boundary layer is making it over the wing like you claimed, as long as the flow stays attached, which it will because the wing will stall before the nacelle flow separates.

      By the way, how did winglets and sharklets make it into this conversation? The wingtip devices are never going to significantly affect the flow near the engines and vice versa.

        • Thanks Richard.
          After I realized the picture didn’t post I tried to post a reply to my own comment containing a link to the picture, but it got attached to the comment above instead of mine.

      • Mike

        I can’t write a book. So without writing a book:

        I hope we both agree that the flow around the nacelle must stay around the nacelle and the flow over/below the wing must stay over/below the wing. Interference is bad.

        Going further means I have to start banging on about surface proximity and pressure contours produced by surface proximity. It’s a very, very long book.

        But to be clear as I can in short words.

        Moving the engine forward has advantages provided there is sufficient clearance between the wing and the nacelle. That is why it is now the norm.

        But Boeing have moved the engine up in a manner that makes it unique not only at Boeing but everywhere else in the world for no other OEM uses the 737 MAX wing/engine mounting.

        In my view, the 737 MAX winge/engine mounting is causing significant interference between flows. It causes a forward pressure build up under the nacelle and the forward wing. That’s moving the CoL forward reducing pitch stability. It also looks as though it is causing lateral flows under the wing. Hence the one off sharklets.

        Can I be sure? Not without seeing the pressure contours caused by the proximity of surfaces.

        Boeing won’t do the wing/engine mounting again. They won’t do the sharlets again. Both are a one off. Will they do forward mounted engines with sufficient clearence between wing and nacelle. Always. The latest is the 777X!

        Anything else requires a book. But it is all about how surfaces compete when they are proximate

        • I might add, the airflow at the top of the nacelle is speeding up because of the upper surface of the wing. So reduced pressure above and increased pressure below. Equals lift.

          Can I end by making clear that I have repeatedly said Boeing engineers did not volunteer the solution. I’m of the view they were given no choice.

          Example: Not repeated with the 777X. Aerodynamically, the 777X is back to best practice. So, won’t be able to argue the aerodynamics or the stability. Bit heavy though.

          Same with the 787. And I think the same will apply to the NMA.

          • Philip, and others. As a non-Aerodynamic trained, small plane pilot, I”ve been trying to understand the pitch up tendency at high AOA as a lot of others have here.
            I see that pressures change at high AOA settings
            It’s the non linear, sudden change that puzzles me the most. It must have something to do with the boundary layer separation near stall? That seems to be what worries Boeing and why MCAS is there, as other planes don’t seem to have this particular problem, or they must have it solved in some other manner? Of course, there may be a lot of factors influencing the plane at that point. It’s probably the combination of a few of them. The size and position of the engines certainly is the focus. Other Boeing planes have the larger size engines and don’t seem to have the sudden pitch problem? How does one find the answer, if they aren’t inside of Boeing? The wind tunnel testing, doesn’t have the engines running? Is there enough aerodynamic formula’s to be simulated via a computer? For example MATLAB
   or does it take more complex formula’s that are beyond that system?

          • Point 1:

            If the CoL moves forward the flow is likely to separate at a lower AoA. So the pre-stall regime begins at a lower AoA. This in turn reduces the safe AoA. The
            If the CoL moves forward the static margin is reduced. This will cause an increase in the pitching moment. Increases in the pitching moment will move the nose up. That in turn will move the CoL further forward. That again reduces the static margin. That inturn increases the pitching moment. And so on.

            In simple terms the CoL needs to be kept well behind the CoG. That’s the job of the elevators and the stabiliser.

            Look up static margin. It’s the difference between the CoG and the CoL. It must be kept positive, the more positive the more stable (within limits).

            The problem is that the nacelle is producing lift. This is moving the CoL forward. The greater the AoA the more lift it produces. This moves the CoL further forward. Reducing the static margin.

            To add complexity, the fact that the nacelle produces lift is likely to cause the wing to stall at a lower AoA. But I use the word ‘likely’. There is complexity.

  7. Bjorn do we know if the issue with the A321 NEO pitch up was discovered by Airbus, and reported to EASA, or if EASA discovered it ?

    It’s reassuring to know that the airframers / regulators are paying more attention than before. We should never take anything for granted.

    • Issue discovered by Airbus. Concerns particular manoeuvres where protections do not cope with pilots hard inputs. No issue with A/C aerodynamics.

      • It sounds to me like heavy input by the pilot in certain landing configurations ie go around can generate pitch up rates that could allow the airframe to overshoot the critical stall angle even though stall protection has activated. It seem a sort of damping issue.

    • I agree with John.

      But to go further, Airbus continuously monitor how pilots are flying their airplanes. They clearly came to the conclusion that the calibration of the aileron/elevator computers was an issue and reported it to EASA.

      Nothing wrong with the airplanes natural stability. It will just be a software patch. But it does need to be done carefully. So no quick turnround.

  8. “The updated MAX software makes it safe. I’m on the record with this judgment, and I will not hesitate to fly on the updated MAX.”

    Totally unnecessary sentence. Come on, is this article sponsored by Boeing? Which updated MAX do you mean? As far as we know there is no update available yet. The Max are still grounded and more issues were found.

    • @Chad

      I agree. Bjorn went to far by presenting his belief (based on Boeing past, and I’m completely fine with that as long as belief is presented as belief) as a fact/assertion (based on nothing in this moment of MAX history).

    • Don’t do that. Argue why LNA are wrong. This site allows you to do it.

    • @Michel BRUN

      If you don’t agree – make a constructive critic.

      I did.

  9. Bjorn does a great job , bringing out the technical aspects of the software changes at Boeing and Airbus (re Max and 321 neo). A non technical person like me can conclude that Airbus did a much better and safer job re the pitch control on its 321 Neo than Boeing which failed to do the right quality ; instead it ducked and tried to find a way out of the mess. For me, the integrity of its safety processes is in question. Boeing went against its famous philosophy of giving the ultimate control of the plane to its pilots with the MCAS system .Not sure how long it will take for Boeing to restore its safety brand.

  10. Nah,that was just marketing, they didn’t have FBW at the time,and it was a credible position in the early days of FBW.

  11. Many weeks before you stated that you would not hesitate to fly on the updated MAX. Very brave.
    Ok, you are a very experienced combat pilot, but I would suggest that you leave your loved ones at home and take the jump seat.
    Are you sure that only a softwarefix will solve the problem? The (aerodynamic) hardwarebugs are the problem. Will they complete the lonely AoA-sensor with three of them? Will they provide a second electric trim motor (hopeless to handle the unredundant trim wheel at higher speeds)?

    • I see no relevance to Bjorn and being a Combat Pilot (which I believe he is not, he flew fighters but not combat)

      He is an aeronautical engineer.

      That means his words have a vast amount of weight, more leverage than MCAS 1.0 in other words.

      I to will fly the MAX with no problems.

      Its the unknowns that get you.

  12. The thing that concerns me most is how happy all the professionals were to allow the MAX to keep flying after the first crash.
    This whole saga reminds me of the original A class Mercedes. Not actually a bad car,pretty good for old people to get in and out of. The trouble was it didn’t handle very well when pushed.The company and all the professional journalists knew this but said nothing, a bit like the Emperor’s New Clothes.It was only when a Swedish car magazine tested it properly in an extreme manoeuvre and it ended up on its roof that the truth was confronted.
    As Bjorn says, there is no aviation company on earth with more expertise, there is no way someone within the company didn’t flag this up either before or after the Lion air crash. I still have to hear a convincing explanation and I am pretty certain of more revelations.

    • A-Class “Moose test”:
      you can do the same with any US market softsprung SUV.

      Roll Over in the US appears to be much more of a thing than elsewhere.

  13. I wonder why Airbus were poking around in the corners of the flight envelope after finishing development, not that curiosity is anything other than a good thing.Were they checking to see if the MAX issues applied to them as well?

    • Its an ongoing process.
      This covers other historical accidents, some where the plane was lost, others it was a violent maneuver
      “Shortly after an Eva Air A330’s departure in 2012 it suffered an angle-of-attack sensor jam, at 5°, as it climbed through 11,000ft. Although the angle was shallow, angle-of-attack margins become narrower, increasing the risk of stall, as an aircraft climbs and its Mach number increases.
      When the A330 reached an altitude at which this false angle-of-attack data exceeded a critical threshold, the aircraft’s stall-protection mechanism responded by automatically commanding nose-down.
      Investigation of the incident revealed that not only could the flight-control laws command a nose-down pitch, but pilots might not be able to counter the attitude – even if they pulled fully back on the sidestick.”

      “Simultaneous jamming of two angle-of-attack sensors, and the rejection of a valid third, had previously led to the fatal crash of an A320 during a check flight at Perpignan in November 2008.
      “Water ingested by the sensors, left unprotected during routine washing, froze as the aircraft cruised at 32,000ft. The sensors jammed at low angle-of-attack settings – respectively 4.2° and 3.8° – and maintained these readings as the crew conducted the descent.
      As a result the sensors were rendered inoperative and failed to detect the A320’s increasing angle-of-attack when, as part of the check flight, the crew deliberately reduced airspeed at low altitude to test the stall-protection system. The aircraft slowed and the horizontal stabiliser trimmed nose-up but the protection system did not activate.
      “The crew waited for the triggering of these protections while allowing the speed to fall to that of a stall,” the inquiry by French investigation authority BEA found”

      • A big part of the problem is the use of alpha sensors of the same type, based on the same principles and of the exact same model number by the same manufacturer. Because of this they are prone to fail in the same way at the same time defeating redundancies ability to detect the fault by comparison. Why use only vane type sensors when a mixture of vane and pressure nulling types would ensure failure by different modes. The pressure nulling type uses two slightly offset ports facing the airflow whose pressure differences are driven to a pressure null by a servo motors. Often used on fighters but seldom seen due to its compact nature. This type could also be self testing and calibrating because it can be rotated in a self test. Airbus seems to have a record of sensors freezing indicating too much reliance on redundancy rather than primary robustness ie adequate heating. It’s also surprising that the reis no “alternate air data system” that infers air speed, TAS, IAS, angle of attack from GPS or inertial data, thrust setting and flap position. In general I’m not Impressed with aviations use of automation.

  14. Lets see, A321NEO software fix, good (well that all they can do isn’t it?)

    What happens when the computers crash (Oh My Got we have an UNSTABLE aircraft!!!!!!!!!!!!!!!!!!!!!!!! – ayeeee, the sky is falling (see Chicken Little)

    737 MAX software fix (done right) bad. We should revert to small people in the back with cranks following signals (see ship engine room on telegraph system, avast you me harries, give me 1 degree down stabilizer (redundant of course) – don’t forget, up Periscope , damn MCAS, full speed ahead.

  15. Allow me a dumb question. Max designers had to place the engines in an “not optimal” position due to a low ground clearance, right? I read somewhere that the Max-10 landing gears Will be longuer than previous models (either NG and Max-8 & 9). They are testing a telescopical gear in order to increase ground clearance (and keeping the current gear’s dock sizes). If it is true, the telescopical gear wouldn’t allow Boeing to place the engines back to the “correct, optimal” position ?

  16. Duke:

    Thanks. I didn’t know that Boeing referred to the 737 MAX wing/engine mounting as a ‘strut’.

    A ‘strut’ transmits a load between two bodies. A ‘pylon’ transmits a load between two bodies but also ensures that the airflow over the two bodies remain separate.

    So Boeing accept that the 737 MAX wing/engine mounting isn’t a pylon. That means they accept that the airflow over the wing and the nacelle do interact.

    If the engine is back, 737 classic, the consequence is drag. If the engine is forward, 737 NG (partly) and 737 MAX (fully), the consequence is drag but also lift.

    It’s no wonder Boeing are not refuting what’s being said. They know.

    Airbus airplanes have pylons. Boeing airplanes have pylons, with the exception of the 737.

    • Yes. I like the differentiation you make.
      Trouble is the patents are either functional or what it looks like.
      The aerodynamic effects and will probably be kept private as intellectual property.

    • “Airbus airplanes have pylons. Boeing airplanes have pylons, with the exception of the 737.”

      Mincing words to reduce compareability. 🙂

      As long as there is structure between wing and engine that has to transfer moment arm (enginewing) it is a pylon.
      Independent on how Boeing tries to redefine it.

      A single strut would only transfer linear forces. A bunch of struts could make pylon though. ( see any power line support )

  17. FlightGlobal have reported that Airbus will issue a software patch in the 3rd quaarter of next year for the A321neo pitch up condition as set out in this [corrected] article. In other words, 12-15 months time.

    The timeline is typical for a non-critical/non-catastrophic problem. Why does Boeing take the view that much smaller timelines are possible for a critical/catastrophic problem? Remember the latest issue was found just a month ago but we are told the patch will be there by September; back to service in October/November.

    • That’s a dammed good question. Maybe, the answer is more to do with how complex the patch is rather than how critical.You need to be more constructive, as Loren Thompson advocates!It’s also hard to see how the aircraft can be ungrounded without the final accident reports.

      • Yes, I need to go on a Loren Thompson course on positive thinking.

        But your right, complexity is the biggest issue. But in my time, a year was the minimum. So more complex issues took longer, a lot longer.

        • Also, while the MAX wasn’t known to have a dangerous flaw,Boeing had been already been working on a solution to that flaw for a significant but not fully determined amount of time.This is a very tricky one for Boeing’s lawyers.Its amazing how little information has leaked so far.

    • It sounds like Boeing’s management has driven a stake in the ground for the Engineering time line. And Engineers didn’t have a say in the decision. I’ve been there, done that too many times. You will either get a botched up solution, or the time line will be pushed back. Boeing hasn’t delivered any solution to the FAA yet, for the FAA, and others to begin verifying and certification. So, setting a time line six months away seems rather aggressive. I assume they have to say something when asked the question, and have to be optimistic. But, Airlines are looking at this from a month by month basis. Longer term, they must be planning for a worst case situation. It’s common sense to plan for the worst and hope for the best.

  18. Also, while the MAX wasn’t known to have a dangerous flaw,Boeing had been already been working on a solution to that flaw for a significant but not fully determined amount of time.This is a very tricky one for Boeing’s lawyers.Its amazing how little information has leaked so far.

  19. If you’re the Engineer assigned the task of fixing the 737-MAX .. You have to fix the obvious. Use two AOA senors. Put in limits to the Stab control of MCAS. Get the AOA disagree light working. Get the trim yoke switch to respond faster (computer issue?). Rewire the Stab Trim cutout switches back to their original design. Then you have to figure out how to get a speed trim system or another solution to the sudden change in pitch at high AOA, while keeping the control feedback to the yoke within FAA limits. That to me is going to be the fun part of the challenge. I assume they already know what the squiggly diagram of the speed vs AOA vs pitch forces comes out to. But, do you have only the high speed / low speed setting of the stabilizer acting for “X” number of seconds to work with? Or keep running the Stab till the AOA gets back to “Y’ degrees? Does the rest of the speed trim system / AP get overridden during this? And then you stop and wait for a weight on wheels to reset and rearm MCAS or some other parameters? The non linear portion of the pre-stall sounds abrupt. That’s where I think the solution gets interesting. You’re really bordering on a FBW system at this point. Feeding pulses of AutomaticNoseDown till you get to a safe AOA seems too much of a fudge to me. Could spoilers be used?

    • I believe Boeing already has the MCAS issues worked out, those were under extensive testing and there have been no reports that those problems are not resolved. The new issue is a data overrun condition that results under very specific circumstances of partial flight control failure. It’s a valid issue and needs to be fixed. That will take additional time, but there is not currently a reason to believe it won’t be possible.

      The Airbus issues just point to the complexity of pitch control and the systems that control them. As Bjorn pointed out, it’s not a simple problem. The Airbus issue is not a result of bad design & implementation, whereas the Boing MCAS situation was, and that is an important point. But I suspect most avionics could be shown to have flaws that may surface under the right conditions.

      That’s actually true of any computer code. Bugs are always there, the thing that varies is whether the situation that triggers them arises or not.

  20. That is rather a non issue but it smells a little bit like MAX and there are several big issues.

  21. Duke:

    Thanks, I’m sure your research is spot on as usual. But, I maintain the distinction. Pylons have structural properties and aerodynamic properties. The aerodynamic properties have the purpose of guiding the airflow when two surfaces are close together.

    The bottom line is this. If the airflow over the nacelle is enticed to flow over the wing then the nacelle is part of the wing lifting surface. In other words, LNA’s ‘foreplane’ effect. To turn it round. If the airflow over the nacelle is enticed to flow under the wing at the same speed or nearly the same speed as the airflow under the wing then the nacelle has no lifting effect.

    The “same speed or nearly the same speed” is key. That’s why Airbus put so much effort into the pylon for the A330neo.

    I hope everybody notes that the pictures provided by Airbus and reproduced by LNA – link provided by JakDak – is about the flow under the wing not over the wing.

    I keep looking at pictures of the 737 MAX. Doesn’t work for me. The airflow over the nacelle is being enticed over the wing.

    • With both aircraft, the engine is mounted forward enough that at high AoA, the nacelle rises above the wing and obstructs the mount, whether pylon or strut. In that case the nacelle area replaces the cross-section of the wing as a lifting surface, and that tends to move the center of lift somewhat forward, which adds to the pitch-up tendency.

      The pylon design may have advantages in level flight or low AoA, as it better controls the airflow around both nacelle and wing. But at some point as AoA increases, separation will occur to direct nacelle airflow over the top of the wing, and the nacelle will become the leading surface. This will happen sooner for the strut than for the pylon, so the effect may be delayed somewhat for the pylon.

    • From my looking at the patents , its seems that technically the strut is the part that carries the load and a fairing is added for obvious reasons mainly in front of the strut but could be sides and rear as well.
      An elegant approach would be to call the strut and ( aerodynamic) fairing together a ‘pylon’. It would fit your reasoning and keep everyone happy

  22. Rob has highlighted a element of the narrative, helpful in understanding risk when AOA is high, regardless of what type of airliner we are considering.
    “With both aircraft, the engine is mounted forward enough that at high AoA, the nacelle rises above the wing and obstructs the mount, whether pylon or strut. In that case the nacelle area replaces the cross-section of the wing as a lifting surface, and that tends to move the center of lift somewhat forward, which adds to the pitch-up tendency.”
    Exceptionally, below 100 feet, in the case of the Airbus FBW family, aircraft revert to Flare (Direct) Law. Therefore response to sidestick inputs is materially different below 100 feet, especially if high AOA is sensed; the specific risk in relation to the observed A321neo behavior being unanticipated pitch-up during a mishandled go-around below 100 feet.

  23. LNA

    Pleased the record has been put right.


    I will call it a bracket not a pylon or strut. All sorts of crap is going on around the wing/engine mounting of the 737 MAX. The crap doesn’t apply elsewhere

  24. According to the excerpts below from the 8-1-19 FlightGlobal article at the link after after the excerpts, a pitch up problem has been discovered with the A320neo, that is “outwardly similar” but “different”, than the one found with the A321neo, and a manual revision AD with a 30 day deadline had been issued. As best as I can remember, I haven’t seen any mention of this on this blog.

    “Airbus has determined that the A320neo is potentially vulnerable to an angle-of-attack protection weakness which could result in excessive pitch attitude under certain circumstances.

    The condition is “different” from the excessive-pitch anomaly recently discovered during analysis of the larger A321neo, says the European Union Aviation Safety Agency, although it appears outwardly similar.”

    “EASA has opted for quick implementation of an airworthiness directive, requiring the changes within 30 days of 14 August, postponing the commentary period until after publication.”

  25. Although the stab trim cutoff switches have never been advertised as capable to work separately, on the NG they can: one turns off autopilot/STS, another turns off all electric power including the control wheel switches. On the Max both switches cut off all electric power, so you can have everything (autopilot/sts/mcas/manual electric trim) or nothing (only manual trim wheel, which is sorely inadequate).

    Here is a question: should not a proper fix include re-wiring stab trim cutoff switches to the pre-Max configuration? I think it should, and I would never fly the Max if these switches are not rolled back to the NG configuration.

    • Michael T. Yes, I’ve been wanting an OFF switch for MCAS for quite a while. It probably would have saved ET302, if they could have switched OFF MCAS and kept manual electric via the yoke mounted switch to the stabilizer. As a matter of fact, I recently found out that the reason there are two switches was that in the original 737 Classic, there were physically two stab trim motors. One for manual trim via the yoke and a separate one for the autopilot. Again, redundant systems. Now, if the one stab trim motor fails, you are left with the truly manual cable system, hoping the smaller trim wheel is enough leverage to hand crank the system. It would be nice to have a hand pump or electric hydraulic system backup?

      • Have Boeing explained why exactly they re-wired the switches for the Max?

        • Michael T, Boeing has said nothing. I can only assume the lawyers are in charge. Peter Lemme, a very experienced former Boeing avionics engineer has been doing some very detailed explanations from what he knows. He has some excellent insight as to why the column stab cutout switch was taken out of the loop MCAS in his most recent posting.
          (I’m only 1/2 way done reading it)

          As for the stab cutout switches on the center console, Peter has been puzzled. The best explanation that I’ve found is the following discussion from


          Mick Gilbert says:
          April 7, 2019 at 1:36 am

          @Don Thompson


          Re: ‘The MAX switches appear to imply left (PRI): STAB TRIM driven by FCC-A or control wheel STABILIZER TRIM switches; or right (B/U) STAB TRIM driven by FCC-B or control wheel STABILIZER TRIM switches.‘

          I see what you are getting at with that explanation but I think that the operation may be different. The following is from an NG FCOM:

          ‘Stabilizer Trim

          ‘Stabilizer trim switches on each control wheel actuate the electric trim motor through the main electric stabilizer trim circuit when the airplane is flown manually. With the autopilot engaged, stabilizer trim is accomplished through the autopilot stabilizer trim circuit. The main electric and autopilot stabilizer trim have two speed modes: high speed with flaps extended and low speed with flaps retracted. If the autopilot is engaged, actuating either pair of stabilizer trim switches automatically disengages the autopilot. The stabilizer trim wheels rotate whenever electric stabilizer trim is actuated.

          ‘The STAB TRIM MAIN ELECT cutout switch and the STAB TRIM AUTOPILOT cutout switch, located on the control stand, are provided to allow the autopilot or main electric trim inputs to be disconnected from the stabilizer trim motor.‘

          For the NG, at least, that suggests both FCCs operate the stab trim motor through one circuit when autopilot is engaged (the autopilot stabilizer trim circuit) and the other when in manual flight (main electric stabilizer trim circuit). That suggests that the Speed Trim System (STS) commands the stab trim motor through the main electric circuit.

          On the basis that using the control wheel stabilizer trim switches will disengage the autopilot if it is engaged, it’s not 100 per cent clear (at least not to me at this point) whether the control wheel stabilizer trim switches use only the main electric stabilizer trim circuit or both circuits.

          Under normal circumstances if Boeing changed the basic structure of the main electric and autopilot stabilizer trim circuits you would expect that to be covered in the Differences Training. With the MAX, however, that rule may not apply. What the MAX differences training does include is a slide noting the nomenclature change for the cutout switches but suggesting that the circuits controlled have not changed;

          viz NG ‘MAIN ELEC’ = MAX ‘PRI’ and NG ‘AUTO PILOT’ = MAX ‘B/U’.

          The circuit that MCAS uses is almost certainly the ‘old’ main electric stabilizer trim circuit (the same as the STS), referred to as the PRI(mary) circuit on the MAX.

          • Richard, thanks a lot for referring to the Peter Lemme’s article. I’ve read most of his words and about half of what he quotes. Then I searched for “cutout” and the switch he mentions most is the column aft cutout switch. The center column cutout are mentioned in the runaway procedure, where his (or the commissions?) take is to define the usage of electric trim switches before turning off STS/MCAS as a “must” not “can”. But he does not touch the history or the reasons of the center column re-wiring. OTOH, the info on the introduction of STS was illuminating.

        • Michael T. Peter Lemme’s most recent posting thinks that the reason Boeing took out the column cutout switches was because of the windup turn/stall test producing different effects than the 737-NG. So they took out the safety valve (column cutout switch). I just found out that the 737-NG has a mini MCAS like feature recently (1999) added to it for pre stall feelability enhancement (or whatever term folks want to call it). But, it keeps the column cutout switch function intact.


  26. Boeing should have stuck with the 757 which did not need gimmicks to stretch the design beyond normal aerodynamic boundaries. The 757 was designed for that segment of the market that Boeing attempted to stretch the 737 to. An updated 757 with composite wings, engines, lighter DIW would have been a better path and a cost efficient offering. Now the 321 NEO is walking away with the prize market share.

    • Boeing couldn’t do that because SW Airlines only flies 737’s, and their pilots would have to be retrained, along with their mechanics to work on a different type certificate. (I say this half in jest and half in truth). BTW, someone else agrees with you about the 757.

Leave a Reply

Your email address will not be published. Required fields are marked *