Bjorn’s Corner: Analysing the Lion Air JT610 crash, Part 2.

November 8, 2019, ©. Leeham News: We started the series on analyzing the Lion Air JT610 crash based on the final crash report last week by looking at what went wrong with the aircraft’s Angle of Attack sensors.

Now we continue with looking at why an MCAS system is needed in an aircraft like the Boeing 737 MAX and why a correctly designed MCAS is not an irrational addition to the aircraft.

A note about the series

Before we go into the rationale for a system like MCAS, let me clarify how this series is intended. We are NOT rehashing the complete MCAS story in every article and how bad its initial implementation was. This has been done by us and other media over the last 11 months since the existence of the 737 MAX MCAS was revealed.

Rather we are methodically going through the themes around MCAS in more detail, now with the additional information provided by the final report. We cover one theme in depth in each Part of the series.

By it, we can correct and give context to a number of misconceptions around MCAS which has grown through the exhaustive but not too deep press coverage, where many times the interest has been to find more compromising data rather than correctly describe things as they are (the reporting around the Boeing pilots chat is a good example of this).

By covering the different aspects in detail and at the same time provide the state of the industry around these details, we try to give enough depth and context around the subjects so everyone can form more informed opinions.

Why do we need a Maneuvering Characteristics Augmentation System, MCAS?

The type of system that MCAS represents is no oddity for airliners. The need for such a system for the 737 MAX was triggered by mounting larger engines on the aircraft, but there was already similar augmentation systems in operation on the 737NG, the previous generation 737, and on other airliners.

The reason for MCAS is a shift forward of the aerodynamic center at high Angle of Attack (AoA), caused by the larger and more forward-mounted engines, Figure 1.

Figure 1. The larger and more forward-mounted engines (737NG yellow, MAX violet) affect the aerodynamics of the wing at high AoA, shifting the aerodynamic center forwards. Source: Leeham Co.

The larger nacelles, sitting further forward of the wing affect the flow of a clean wing (a wing with no slats or flaps deployed) at high Angle of Attack (AoA). The result is a moving forward of the wing lift, reducing the aircraft’s pitch stability margin at high AoA.

In aircraft engineering terms, it’s not a “pitch up” situation, as a further increase of AoA requires an increase in yoke force, but the magnitude is reduced. This is felt like a “pitch-up” by the pilot. It feels like the aircraft is pulling Gs by itself in the maneuver, which is not a nice feeling. An MCAS type system is then needed to help the pilot with a consistent feel when maneuvering in this extreme region.

There is another “Shift of aerodynamic center” augmentation system active on every airliner in the world, the Mach tuck system.

Here, the shift in the aerodynamic center is back at high Mach numbers. The system is similar in that it uses the powerful aircraft pitch trim system moving the whole horizontal stabilator to augment the behavior of the aircraft and compensate for Mach tuck (a tendency of the aircraft to dive at high Mach numbers). The difference is it’s triggered by Mach as it compensates for changed aerodynamics at very high Mach, close to the maximum Mach of the aircraft (MCAS compensates for changed aerodynamics close to the maximum Angle of Attack of the aircraft).

Speed Trim System for the 737

There is yet another augmentation system active on the 737, the Speed Trim System. This is triggered by a change in the speed of the aircraft, once again using the trim system for its augmentation of the aircraft’s behavior. It’s very similar to MCAS. In fact, MCAS is a variant of Speed Trim and it uses the Speed Trim System’s base functions for its augmentation functions.

It’s important to understand Speed Trim System because its existence on the 737 is the major reason the JT610 crew could not identify MCAS going active and what it tried to do. This has not been covered in the hundreds of articles around MCAS, as it’s complicated to understand and explain for most writers.

A classically controlled aircraft like the 737 has natural speed stability which is based on the Phugoid motion (see my Corners about aircraft stability for more on Phugoid). If a disturbance pushes the nose up, the speed of the aircraft decreases. The decreased speed creates a nose-down force, based on the center of gravity (CG) being ahead of the center of lift and the reduced speed decreasing the lift force. The nosing down of the aircraft increases the speed and the aircraft returns to its previous state. This is called the aircraft is pitch stable as the aircraft regulates the disturbance without pilot intervention.

When flying in aft center of gravity configurations, this tendency can be weak. In situations after take-off and before landing this can complicate the control of the aircraft for the pilot. Speed Trim then augments the stabilizing tendency of the aircraft, it makes the aircraft easier to fly for the pilot with different center of gravity positions.

Figure 2. The 737 cockpit trim wheel.

When speed decreases Speed Trim ads nose-down trim, when Speed increases Speed Trim ads nose-up trim. This means every pilot of the 737 is used to the aircraft constantly trimming in the background when flying manually (Speed Trim and MCAS are only active when flying manually).

It happens at every climb after takeoff (speed increases) and during every descent and approach for landing (speed decreases), which are the phases where a pilot flies an aircraft manually. At cruise, the autopilot is engaged and Speed Trim and MCAS are not active.

The trimming of the aircraft in the background is very visual in the 737, the large trim wheels with their white stripe (Figure 2) are centimeters from the pilot’s thighs. You see the wheels spinning in your peripheral vision when flying.

MCAS, a system which should never activate

It’s also important to understand MCAS is a system augmenting the behavior of the MAX way outside the normal flight envelope. In daily operation, a 737 MAX flies with an Angle of Attack below 7° to 8° and never passes 1.2 G in load factor (in fact it rarely passes 1.15 G, the load factor for a 30° bank turn).

MCAS is programmed to activate at almost double this AoA value and the pilot need to pull around 2G or more to get to where MCAS will be active. No person would stand up in the cabin if the pilot is flying like this. A mega-crisis situation is needed for an airliner pilot to need to pull 2Gs or more. I know no airliner pilot which has been even close to 2G during their flying of airliners in regular operation. Test pilots can reach these values to control the correct behavior of a new or repaired aircraft but even this is rare.

You only go to this corner of the flight envelope if the aircraft is in the utmost crisis. As MCAS is only needed when the wing is clean (no slats or flaps deployed), this means you are turning extremely hard to avoid a mid-air collision or maneuvering through extreme turbulence. If you hit clear air turbulence, you don’t maneuver as a pilot as this can worsen the effect of the turbulence. The only case where maneuvering might be needed is when passing strong wake turbulence of another aircraft (you can get a strong change in pitch and roll angle). But even then 2 G or more and more than 10° AoA is extreme maneuvering.

The consequence is, the likelihood an airliner pilot ever running into a situation where a correctly triggered MCAS would go active is highly improbable.

This is described to give context to why FAA accepted Boeing’s suggestion MCAS didn’t need to be described to the pilots. I don’t endorse this decision by writing the above. But as said, we need the correct context to be able to form a correct opinion. We can’t ask people to understand what is normal and what is abnormal in an airliner’s operation if we don’t spend time describing it.

And if we don’t understand the unlikely event of MCAS ever going active when correctly triggered, we don’t understand the system and its implementation. (For those who think I’m writing this to later defend MCAS in its initial implementation, I’m not. It was awful and we will come to that.)

Could MCAS have been implemented aerodynamically?

Yes. It would most likely be as large aerodynamic strakes at the lower rear part of the fuselage like seen on Learjet business jets, Figure 3.

Figure 3. A Learjet business jet with its ventral strakes adding stability at high AoA. Source: Wikipedia.

These strakes would add weight and drag for every flight during the aircraft’s life for a function that should never be used. The implementation of the MCAS function as software, re-using the functions of Speed Trim is, if done correctly, a rational decision.

Next step

In the next part of the series, we will look at the Boeing assumptions around how a rough MCAS would be recognized and contained and how the reality looked in JT610 based on detailed information in the final report and the things we discussed in this Corner.

398 Comments on “Bjorn’s Corner: Analysing the Lion Air JT610 crash, Part 2.

  1. Bjorn, a very clear description, but could you expand on “At cruise, the autopilot is engaged and Speed Trim and MCAS are not active.”

    How does the autopilot trim the aircraft ? Are there visual clues for the pilots e.g. moving trim wheels ?

    A very good description of Phugoid motion, demonstrated to me by a veteran pilot instructing me in a Cessna 152, trimmed the aircraft, then pulled back on the yoke until we were climbing at a high angle of attack, and let go of the yoke to allow the aircraft to regain it’s trim.

    For the video generation a nice demonstration at the co-pilot only uses the yoke to keep the wings level.

    • The autopilot has servos which work on the elevator for pitch changes/corrections. When those servos reach a certain constant force they ask for neutralizing trim via the autopilot software loops. This trim is using the same motor and systems as manual and augmentation trim so the trimming by the autopilot is visible, the wheels are spinning. The autopilot is not using a feedback force in the flight control system to judge how hard it’s commanding the control surfaces so it doesn’t need augmentation to help it with the control. It might use part of the Mach tuck logic to execute Mach change of lift center compensation but doesn’t need speed trim because of no need for control stick feel. An autopilot will never control the aircraft to the AoA or G region where MCAS goes active (it disconnects way before such flight envelope excursions and asks the pilot to take over), hence no MCAS involvement in autopilot flight.

      • Thank you Bjorn.

        Would I be correct in assuming that the E175/E190 flight controls are similar to the 737 ?

        I see American Eagle 4439 had some issues with runaway stabiliser, they managed to regain control by switching to FO controls.

        I wonder how similar 4439 was to Air Astana 1388 November last year ?

  2. ‘There is yet another argumentation system active on the 737, the Speed Trim System.’ Was this a Freudian slip Bjorn??

    • You mean “yet” as in “sigh there are too many” for the 737? No, I had described two and added a third. If you dissect the FBW systems (like the A320) you find up to 10 augmentation loops if not more (all the protection schemes are augmentations), so the 737 is not anything out of the ordinary with its three loops.

      • Now imagine there being 10 argumentation loops, with seven to go, instead of “10 augmentation loops”, some of which we will skim over…
        It was just a slightly naughty attempt at humour from peter, pointing out the spell error.

        • Thanks, nice nudge by Peter BTW, didn’t get it.

          Grammarly doesn’t understand aeronautics but it thought argumentation was better than augmentation :-).

          • Bjorn:

            Would “activate” be a better description for MCAS rather than kick in?

            Kick in tends to define as a sudden jolt vs a deliberate movement even if fairly aggressive movement .

          • Hi Transworld,

            agree, activate is better. The initial force build-up from MCAS is gradual but needs to be rather fast. It needs to fast provide incremental stick force in a dynamic nose up movement like in a brusk exit turn from a circling at say 3,000ft. When you circle at so-called green dot speed (the best L/D speed which is typically just above 200kts) you are the closest to a high AoA with a clean wing. As you go slower in the approach you do this with at least slats extended which takes away the need for MCAS (the center of lift no longer shifts forward at high AoA with slats extended).

        • And now imagine those 10 argumentation loops squabbling away in triplicate. That’d be the grumpy Airbus on a very, very bad day. And for the F117 it’d be in quadricate, though as that was designed to take some battle damage you could use a kinetic solution to quieten it down to triplicate with no reduction in safety and a modest reduction in yelling.

          BTW, anyone else notice GECAS buying Airbuses, A321 and A330? Yes, GE are buying non-GE. Talk about ringing endorsements… And Spirit have bought Bombardier aero structures, so they’re now operating in the Airbus camp.

  3. Why did you forget MCAS as antistall? That was added later as well for low speed high AoA, this change effectively turns out to be deadly. With this change they increased authority.

    • The description above is not restricted to the high-speed MCAS and Boeing is adamant MCAS is not an anti-stall system (like a stick pusher).

      It’s a system that shall provide a consistent stick force/AoA relationship in a region ahead of a stall. It does this by straightening the pitch moment curve (see previous Corners on pitch moment curves and MCAS). This is valid for high and low speed cases.

      • Hello Bjorn,

        In this regard, here is what the JTAR wrote in its report published last October (pages 13 and 14): ” MCAS used the stabilizer to change the column force feel, not trim the aircraft. This is a case of using the control surface in a new way that the regulations never accounted for and should have required an issue paper for further analysis by the FAA. If an issue paper had been required, the JATR team believes it likely would have identified the potential for the stabilizer to overpower the elevator.”

        The JATR adds on page 18: ” The JATR team considers that the STS/MCAS and EFS functions could be considered as stall identification systems or stall protection systems, depending on the natural (unaugmented) stall characteristics of the aircraft. From its data review, the JATR team was unable to completely rule out the possibility that these augmentation systems function as a stall protection system.”

        The MCAS does not seem to have been designed as an anti-stall system, but it can not be exluded that it nevertheless meets the characteristics of such a system. If that were to be the case, should not it be certified as such?

        Thank you.

        (English is not my language, I hope to make myself understood)

  4. “At cruise, the autopilot is engaged and Speed Trim and MCAS are not active.”

    How does the Autopilot work its control requirements?

    My guess it uses the the same method only a different hardware path that does not reflect into the “trim” part of the screw actuation?
    i.e. the tailplane spindle is turned with a different motor.

    • See the answer above, it uses the same trim system and motors, just with some tighter limitations on trim authority (implemented as stabilizer travel limiting switches which go active when in autopilot controlled mode).

          • so 737 AP mode is comparable to Airbus “autotrim” setup? i.e. have the trim system automagically trim to zero average ( low pass filtered) stick forces.
            A bunch of functions fighting over the trim system but with limiters of varying saneness. Onion layer approach to engineering at its (terrible) best.

    • When the BAC 1-11 prototype suffered a deep super stall crash McDonald Douglass looked at the issue for the DC9-10 and ended up adding wing vortilons at about 50% of span and wing fences at about 55% to ensure that the wing tips had a delayed stall compared to the inboard section and could thus recover by having a stronger forward pitch. The stabilizer was also enlarged. The Super VC10 received a dog tooth leading edge for a similar effect..
      I believe the Boeing test pilots that discovered the MAX pitch up issue were thinking of vortex generators perhaps behind the engines or ahead of the ailerons. Now there is a topic: B737-300 had Vortex generators at about 1;4 chord up to the aileron area where they move to 1/2 chord. The B737NG has vortex generators only to the aileron point, non thereafter. Ive read it that some NG don’t have any. Nevertheless I don’t think adding passive high lift devices like these would have worked. They help with a stall, they don’t counter a pitching up in the pre-stall area which is what MCAS does. So aerodynamically rear strakes, enlarged stabilizer or perhaps extended wing tips, perhaps folding like on the 777X.
      What I can’t understand is that both sensors weren’t used to activate MCAS. I can not see the reasoning behined that. That is surely the biggest mistake Boeing made and I ca see no reason for it. The had duel sensor activation, they apparently removed it.

      • Hi William,

        the vortex generators that are present on many Boeing wings, mostly ahead of the ailerons are there to attach the boundary layer at high Mach, by it preventing an instability at these speeds (i.e. they work in the low AoA region). They might help at high AoA as well but to my knowledge, it’s not their primary purpose.

        We don’t know the exact flow changes which are caused by the larger engines (only the center of lift moves forward according to Boeing), however, cylindrical surfaces have attached flow at higher AoA than a non-slatted wing. The nacelles will also change the airflow around the wing in their region of influence, including when the flow goes turbulent around the nacelles.

        My bet would be on attached nacelle flow concurrent with the rear of the wing aft of the nacelles having boundary layer separation (which moves the center of lift forward).

        • Thanks for your explanation. MCAS seems the most elegant solution and if implemented appropriately I don’t see why it shouldn’t be any less safe than speed trim or mach trim. I did also suspect that stall strips on the inboard wing might have induced an early stall there which could countered the “closed tubular wing” pitch up effect of the LEAP 1B nacelles as long as the inboard Krueger flaps cleared them during deployment.

          • The the question on elevator size and authority comes in. When you have a too small elevator you need “big brother help” from the horizontal trimmable stabilisers with their big actuator and its own logic. The actuator has its separate inputs from wheel, trim buttons and autopilot and is not failure proof.
            Like the MD80 lots of aircrat types have SB’s and AD’s to correct failure modes on the THSA, so on a commercial airliner trusting the THSA to augument a too small elevator can be dangerous and you have a bag of new failure modes you have to manage.

          • Are there any transport-sized swept-wing aircraft that do not use the moving tailplane? My understanding is elevators large enough to provide all stability and trim functions on such a design would be too large to be practical.

          • Like pylon mounted engines, a moving stab has long proven to be the best setup.

            Some fighters that is all they use (or used, ala later F-86)

          • Bubba, Rob, yes, I meant using a bigger lever arm, I wasn’t clear in describing what I was trying to lengthen. Actually, it reminds me of a story. A used plane was being looked at to purchase. I think it was a Cessna 210. Two CFI’s took a look at the log books and POH, and then took it up for a flight to see how well if flew. Everything was good, and they came back to land. They didn’t get three green lights for the gear, so they went for the alternate gear down checklist. It said to extend a handle between them and pump the gear down manually. Each one tried it but, the pump didn’t seem to have any resistance, and the gear wasn’t coming down. They radioed down to the ground and tried to diagnose the situation while planning for a gear up landing. The former owner finally found the problem after speaking with them. The handle they extended ok, but, that was all it was. An extension of the handle to gain more leverage. They were only extending and retracting the handle of the pump, not actually pumping the gear down (it was similar in design as an automotive grease gun). After they figured this out, they were able to pump the gear down. Needness to say, a picture is worth a 1000 words. And keep the Pilots Operating Handbook (POH) with you while flying.

      • William , regarding your VC10 comments here and previously I had commented but these didnt appear because my source was blocked ( not secure ?).
        Yes the dogstooth was for stall indentification, but the VC10 wing was less swept than the B707 not more , perhaps you meant the wing was the first specifically designed and supercritical foil shape.
        And regarding the T tail this sums it up best-
        “The VC10 is one of the few T-tailed airliners to have a natural tendency for pitching nose down at the stall, if not the only one. This was mainly due to the fact that the engine nacelles were comparatively wide and therefore started acting like a low-set tailplane in high angle of attack situations. A case was made for allowing the VC10 to fly without a stick pusher but in the end a decision was made to install one as an added safety feature to prevent the dreaded super-stall condition that caused the loss of the BAC 1-11 prototype in 1963.”
        A very good source but not https vc10*net (the wing design section)

        • Thanks for your info. A few books on VC10 development. Technical history is a great way to learn aerodynamics. In my study of the super stall issue on T tail aircraft I’ve found it is overstated. Aircraft manufacturers have, I think all of them, come up with aerodynamic technologies to recover T tail aircraft from stalls. There is no such thing as an irrecoverable super stall on a certified T tail airliner within CofG limits. The stick shakers, pushers and nudgers were added as an additional safety feature. The most common was devices such as wing fences, vortilons, slats & dog tooth leading edge extensions to ensure the outer portion of the swept wing recovers earlier from a stall than the inner to generate a pitch forward. Avoiding premature tip stall is critical on all swept wing aircraft, not just T tails since they would pitch up, but it’s more of an issue for T tails. The other was to increase horizontal stabliser area and sweep to get much more of the stabaliser out of the air stream shadow of the wing. The “stub wing effect” of the VC10’s 4 x duel prodded Conway turbofans was important but even two engined aircraft utilized this effect or used the engine pylons as stabilizer. Had the Super VC10 been given two RB211 I think it could still have been made to recover naturally by minor modifications. The other issue was to keep the engines going since the engines could stall at the same time as the aircraft due to interrupted airflow and powering out of a stall is one recovery technique. This involved profiling the wing airflow, adding fences to channel relatively clean air to the engines even during a stall. Initiation of the stick skater would also start the igniters “auto light” to restart the engines automatically. Early VC10 also had stall strips at the wing root to initiate a clean stall there. Straight winged T tail aircraft such as the Learjet May have ventral strakes to provide the required pitch up. One additional feature of the BAC 1-11 was the replacement of the spring tab elevator with a hydraulic powered unit as it was though that the interruption of airflow over the tabs had prevented recovery. The dawn of the transonic/supersonic era in the 1940s saw the problem of shock wave impingement interference on the elevator. The Americans initially looked at V tails (jakes Jeep), British developed all moving stabilators under Miles and the Germans used T tails to clear the shock waves. These German T tail aircraft are interesting in that they were configured as tailless aircraft using elevons for combined pitch and roll control with the horizontal tail only there only for trimming (not stability or pitch control)

  5. Is the authority of these columns compromised by the frequent, ignorant and annoying english language errors ?

    • @Leigh: English is Bjorn’s second or third language (he speaks four) and uses Grammarly to assist. So lighten up.

      • Speaking as someone labouring through the translation of a French aviation text, I can only hope I’ll be cut some slack too. Being largely monolingual is quite an impediment in the wider world, so I admire multi-linguists.

    • Far less so that by the lack of courtesy and utter inadequacy of authority shown by such as you.

      • LD:

        Nope, real technical people are used so sussing out what is meant as many a technical manual has poor phrasing and those are written by English speakers for the most part in my world (or was)

        I know one purportedly American who can’t tell a Sever from a Beer can. He speaks a variation of English called supideese.

        Bjorn is genius level by those standards and the best I have read by my own hard technical standards.

        I took a second language in high school. Incredible to do well in one language , do well in 3? Shoot, he is the Babe Ruth of tech discussion, purely out of the ball park.

    • I am very appreciative of the content Bjorn provides in these articles and completely satisfied with its presentation.

    • “Is the authority of these columns compromised by the frequent, ignorant and annoying english language errors ?”

      Leigh Dyer,

      If you knew anything about Bjorn – ANYTHING – then you would know that Bjorn is the authority on these boards – and to a lot of other people around the world. Bjorn makes some English errors because English is not his primary language – but that does not matter. Bjorn has always put forth the effort to make himself understood and, as a result, enlighten his audience.

      All of this experience Bjorn has he provides to us free – experience hard won through rigorous education and flying some very-dangerous fighter jets. Seriously, could you ask any more of the man? Could you be any less than grateful for what he does?

  6. I think it is settled that MCAS was firing out of its intended regime in both accidents. Your previous corner did its job explaining one of those reasons. Looking forward for more.


    When speed decreases Speed Trim ads nose-down trim, when Speed increases Speed Trim ads nose-up trim.(…)

    (…)at every climb after takeoff (speed increases) and during every descent and approach for landing (speed decreases)

    Therefore, STS for the phase of flight in question (climbing), would be working exactly in the inverse direction MCAS is always expected to be working: while STS trims Up, MCAS trims Down. This is interesting in terms of the logic that had to go into the FCC. Boeing made MCAS take precedence, as the LNI610 flight data demonstrates. It can be seen in the immediate STS auto trim activations just before the first MCAS activation, while we can also see STS and MCAS “agreeing” immediately after that as the flaps are back in play.

    We have two systems (MCAS/STS) driven by two different inputs (AoA/Airspeed) actuating the same surface (Stab). Why does the relationship between AoA and Airspeed, which may or may not have the same resulting sign, be under better control/understanding?

    • If the aircraft is in a very steep pitch, it would make sense for the protective MCAS to have priority over the STS, which was not designed for the same purpose. I think it’s been established that Boeing did not expect MCAS to be active in normal regimes of flight. This was an oversight in the fault tree and failure analysis. Boeing did not think this would happen so didn’t fully address it. Or if it did happen, they believed it could be dealt with as a runaway trim procedure.

      • “fault tree and failure analysis”, if so, then the issue has ceased the domain of physics and entered the domain of statistics. Somehow I am resisting that position.

        In the midst of a wind-up turn, the software would automatically swivel up the leading edge of the plane’s entire horizontal tail, known as the horizontal stabilizer, so that the air flow would push the tail up and correspondingly push the nose down.

        As the pilot pulled on the control column, this uncommanded movement in the background would counter the jet’s tendency to pitch up and smooth out the feel of the column throughout the maneuver.

        An engineer recalled Craig testing MCAS for the first time in the simulator.

        “Yeah! This is great,” Craig gushed after seeing how MCAS responded, according to the engineer. (Craig left Boeing before the operation of MCAS was revised.)

        (ST inside story of MCAS)

        It seams, at some point the trim rate of MCAS was found reasonable (was this rate, not its authority of 0.6, already set at 0.27 deg/secs by then?). While, so far, the STS have not presented many questions about its efficiency and “gentleness”.

        So the numbers the engineers at Boeing arrived at, regarding both MCAS/STS, must have been supported on some solid physics. Why, having increased their requirements on the side of MCAS, couldn’t they find a way to integrate them, specially since they are actuating on the same surface? ie. STS +0.2deg * 3sec & MCAS -0.27 * 6sec >> -0.22deg * 6sec (even if the integration is more complex then presented, even if MCAS is taking precedence, he is dull like a hammer not pondering its effort)

        • I don’t know how the control law is written, whether it is a simple sum or simple override or is a more complex integration. However they are combined, my guess is that MCAS would be expected to dominate in the regime it was meant to be active.

          The comments you referenced were of MCAS as developed for the original high-speed case. They are a confirmation that MCAS had provided the desired feel and responsiveness that Bjorn has outlined above.

          You seem to be suggesting that MCAS should have been a modification or extension of STS, rather than a separate system? With authority and reaction rate similar to STS? Or that it should have used some control feedback (closed-loop rathe than open-loop)?

          Bjorn has mentioned above that MCAS needed to act more quickly than STS to attain the desired effect and help the pilot arrest the pitch-up which is probably in progress at that time.

          I don’t know why open-loop was selected, or what the options would be for a closed-loop sensor, other than AoA which had already failed.

          • MCAS is a special mode of the Speed Trim System according to Boeing with a changed trigger signal. Speed Trim is triggered by speed change, MCAS by high AoA. Most of the time these two are connected as a high AoA creates a high induced drag, lowering the speed. Speed Trim would then trim nose down which is the command invoked by MCAS.

        • Speed Trim is replaced by MCAS above the MCAS activation AoA. Once there, the linear increase of stick force to get to even higher AoA is more important than an increase in speed stability. Also, at AoA above 10° it’s highly likely the speed is decreasing due to induced drag and then MCAS trimming nose down is fulfilling the speed stability role of Speed Trim.

        • I thank you both for your responses.

          I guess where I am getting at, is that the reliability of the STS input signal could be serving as a moderator of the MCAS input AoA signal. If the less reliable AoA signal somehow diverges, given a certain airspeed, that divergence would be accounted for. Gracefully dealt with if some integration was in place, or at least be bounded. But the end product observed is instead an all or nothing very powerdul MCAS, tied to AoA alone.

          Let me insist on the possibility of both systems being in contradiction. In the wind-up turn (turns which BTW, Boeing appears to be profusely testing these days on a MAX7), the airspeed increase is to be expected, STS is expected to be in ANU regime, reducing back airspeed, at the same time AoA is also being increased as a result. Now, where is the threshold? AoA takes precedence, and according to Bjorn, around 10º AoA. MCAS activates down against STS up. But there’s drag…

          No doubt this was expected, tests were devised to assess these limits. I am also sure these types of conflicting signals are common, so many are involved.

          • Regarding this integration, I answered myself downthread, while collecting some additional details, and it should have been obvious. The integration of the several input signals, mainly AoA and Mach, appear as a magnitude in time, not a trim rate.

            I still think this system is balancing a ball at the summit of round hill. Any AoA input, as unreliable as it is (airstreams, etc..), driving the final outcome is very likely to send the ball down the wrong way, which I find odd for the level of engineering which is implied.

  7. Looking at the picture of the ventral strakes on the Learjet, I assume they add pitch stability at high AOA by providing an upward force on the tail. Looking at the two stall tests, low speed and high speed, when they begin, the aircraft is in trim, at speed, and the horizontal stab is providing a downward force. By the time they reach the angle at which MCAS activates “almost double 7 degrees”, has this force reversed and the horizontal stab is now providing an upward force?
    It would be interesting to know the angle of attack on the horizontal stabilizer, at the point MCAS activates, in both tests. What is the difference in upward force through the last angles before stall, without MCAS, and with the added few degrees of MCAS, for both cases.

    • The horizontal stabilizer together with the elevator is exerting an even higher downward force at high AoA.

      The horizontal stabilator is working like an inverted wing when it generates a downforce, therefore, it’s the lower nose radius on the horizontal stabilator which is critical. When the elevator is commanding a higher downforce for higher aircraft AoA it’s increasing the stab+elevator combinations local AoA.

      MCAS is trimming the stabilator nose up, by it decreasing the local AoA.

      • I’ll assume (which may be wrong) in the low speed test the wing is at 7 deg and the inverted airfoil is at 7 deg. Then the aircraft rotates up 7 deg and the wing is at 14 and the stab is at 0, right?

        In the high speed test, I’ll assume they are both at 4 deg. The aircraft rotates up 10 deg, the wing is at 14, and the stab is at -6, the airfoil is flying upside down, providing upward force on the tail to counter the upward force of the engines?

        • First, the reduction in static stability in the region where MCAS is active is not changing a downforce horizontal tail to an up-force tail. It’s still a downforce tail and it needs additional force to push the plane to a higher AoA in this region. It’s just not as many N or lbf per AoA as before.

          Re the local AoA on a horizontal tail. To increase the downward force to fly the aircraft at a higher AoA you need to increase the tail’s local negative AoA. This is not a simple addition or subtraction of the change in wing AoA as the tail flies in wing downwash, which has the effect of increasing the local negative AoA.

          But to try and answer your underlying question, is there a risk of a tail stall and would MCAS increase such a risk? With local icing yes (it has happened) but without, the tails are sized and designed to not risk a tail stall even when flying in the wing stall AoA region at most forward CG. MCAS would reduce the risk for a tail stall with its stabilator nose up trimming (local negative AoA is reduced).

          • O.K. If the slipsteam on the wing is level(to the travel of the object), but the slipstream on the tail is angled down several degrees, I can see how that changes the situation.

          • Also, if the stab is always a downforce, then making it bigger for high AOA does nothing. The physical solution to provide less downforce would be to make it smaller.

            Or back to looking at figure 3 of the Learjet, isn’t the physical solution to counter increase aero area in front of the CG, to increase the areo area in back of the CG?
            Not that Boeing wants to do it, but would the Learjet ventral fins have been an alternative to MCAS?

        • Ted, I think your point is valid.

          This is the key issue. Is the elevator big enough to ensure the stabiliser/elevator combination does not cause an aerodynamic reversal OR insuffient aerodynamic action.

          Aerodynamic reversal means the direction of force is reversed to that expected. Aerodynamic reversal is an issue that has caused the death of many test pilots. But it is a severe situation.

          But full reversal isn’t actually necessary. It is sufficient that the aerodynamic action is insufficient. In this case, the forward lift produced by the nacelles will continue to increase AoA to stall because there isn’t sufficient aerodynamic action to control it.

          This we know. At some AoA, elevators alone are not providing sufficient aerodynamic action to control maneuverability. To correct this the stabiliser is being moved. What is the value of the AoA? At low speed, but with flaps up, I think it is below 10°. At medium speed, I think it’s below 8° and decreasing as the speed increases.

          This comes to the word irrational. This article uses the word irrational. Not appropriate for me, but I’ll leave it. I’ll use the word rational.

          The arguments that are put forward in this article are rational provided the aerodynamics of the stabiliser and elevators are rational. That’s an assumption for it assumes that the stabilisers and the elevators can provide the necessary aerodynamic action to control the pitch up caused by the forward lift produced by the nacelles.

          This comes to JATA. JATR says MCAS use of the stabiliser is novel. I agree it’s novel. But JATR also says that it is not compliant for the purpose of stall protection.

          My own view is that it can work provided that the stabiliser is upgraded to an all moving stabiliser with end to end fail safe redundancy that use high speed/high precision servo hydraulic actuators.

          But I’d still prefer a bigger stabiliser and elevators.

          Anyway, Boeing are going to have to prove that this tortoise of a stabiliser with an electric motor and a jack screw brought from Walmart and computers brought from Walmart can act as a primary control/manuevering system and can do it safely.

          RTS?. If ever? It’s March according to American and Southwest. They are US airlines. I wonder what it will be for airlines outside the US.

    • The Learjet has a T-tail so it is potentially vulnerable to a super-stall caused by the main wing blanking of airflow to the stabiliser in a stall. Because of the very low sweep angle the wing tips probably can not be used to generate the required forward pitch. The ventral tail strakes look to me like an add on to provide this forward pitch at high alpha as well as some additional longitudinal stability.

  8. Indeed, frequent errors like writing “English” with lower case “e”….

  9. There is one missing element prior to mcas input on the Max8, that of a pilot error, co-pilot error. If the flying pilot from the left sets in standard QNH for the FL and if the co-pilot on the right has his system on QFE and the pitot static is set on the right, the aircraft will hit the ground once the system is set to auto. The aircraft will only do what the prior crew set it to do. The flight simulator will not show you this problem as the thing is digital. Look at this possibility, Björn.

    • You have to explain this better. QNH and QFE are reference pressure levels to which the present static pressure for the plane is compared. Out of the difference you calculate altitude over sea level (QNH) and altitude over the level where QFE is measured which often is your airfield. What has it to do with an auto setting? What auto setting do you mean?

      • Yes. This is JT610 flight being analysed, Jakarta is at or only slightly above sea level.

  10. Each pass through removes the chaff (distills ) from this for me and clarifies the situation and what was going on.

    I think this falls into two levels.

    One is the general public and I think for the most part its been conveyed as well as possible. GP is very non tech informed (unless they work in a very discrete tech field but even then, outside their field no very good)

    Then there are people like Bjorn that are experts (Peter Leme) and down someplace in the deeply tech and interested group but not designers/engineer but more users, people like me.

    Sadly I spent the last 8 years of my career workign with management that believed in voodoo mechanics. That truly is scary. Sadly I have seen too much voodoo aeronautic espoused here, writing when they should read and listen.

    And yes, I will raise my hand with, I probably should write less and listen more.

  11. Is this page for a technical analysis or a grammar analysis? If the reader is a technician, an engineer or a pilot, a small mistake of the person who speaks four languages does not change the substance of the

    • Obvious hungry troll, I’m English,can’t speak any other language and my grammar is appalling. I just don’t care but I have to say that I found the very slight misspronunciations of the ABBA girls very appealing.

    • Grammar nazism is the preferred tool of the argument depleted.
      On occasion this is escalated into “native speakers have ‘being right’ precedence”. :-)))

    • I was born, raised and educated in Australia but I had two German parents and learned German/English at the same time. As German syntax is similar to but perhaps more convoluted than Swedish I find Bjorns Swedish style formalism and syntax superior and clearer than native English often.. Having worked in Sweden I find most Swedes speak better English than many native speakers in Aus and UK. Complex grammar helps you to construct complex logic in succinct way. Hard to explain but it’s sort of like thinking I could have written this code in python much better than C++. The big enemy of someone learning English is slang and metaphor.

        • Bugger is a very versatile word. Prettiest girl I knew at school, a Dutch immigrant girl, used it all the time. She just used it for emphasis never knew what it meant. We couldn’t figure out how to explain it.

          • The explanation of the word “bugger” is very straightforward and I believe homophobic.

  12. Anyone blaming third world pilots for bad ”airmanship” should checkout this approach.
    I would definetly fly with these guys rather then Southwest pilots peeking at passengers in the lavatory.

  13. Virtually pointless device kills hundreds of people. The most important question now is how did this happen? Boeing undoubtedly knew that they had gone down the wrong path at some point before entry into service and when they argued against grounding they were in posesion of more facts than any one else.
    Massive program, massive inertia,its more a question for mental health professionals than any one esle.

    • Two “small omissions” by Boeing. 1) Not to redo the Failure mode analysis and working with FAA while doing it. 2) Having one alfa probe tigger the system as it probably was in their own design rules to have at least 2 probes and a sythesised calculated alfa for comparison. These omissions should have been caught while they were working thru 1000’s of other issues before certification.
      Just compare how they did the 767-2C/KC-46A upgrade more or less at the same time where FAA/USAF did not back down and took the delays to meet USAF spec’s.

  14. No , Boeing didnt ‘know all this’ before grounding. They havent been totally forthcoming but this interview by NY Times with the CEO, Doug Parker, of American Airlines indicates Boeing told AA it was ‘poor maintenance and badly trained pilots’ as they were saying it wont effect ‘AA pilots and AA planes’. A case of Group think, which reinforced the theory’s used during flight testing not some mental illness.
    Try and keep it on the technical issues rather than heading off into unqualified diagnosis
    This would confirm why they havent been able to satisfy the FAA and other regulatory agencies much earlier, they kept sticking to initial conclusions rather than accepting they were wrong. Im thinking that all changed when the BCA CEO was fired, they finally realised he was a follower rather than a leader on how to fix the Max grounding .

    • “” Boeing told AA it was ‘poor maintenance and badly trained pilots’ “”

      Then why is Boeing changing MCAS for more than one year already?
      Making a safe plane even safer. Do you really believe this?


      “” No , Boeing didnt ‘know all this’ before grounding. “”

      Then Boeing didn’t know how to design safely. Of course they didn’t follow regulations too.

      It was a given to keep the NG certificate rating, no pilot training.
      Increasing the fan diameter more and more to reach better fuel efficiency.
      No taller gears or other hardware, achieving this with software only.
      All this with much pressure because Airbus was ahead.

      Congress already mentioned talks from 2015, when engineers questioned a single failure AoA. Boeing knew, but they didn’t care about safety, counting beans. Self certification and hiding, didn’t inform FAA (cheating) and after the LA610 crash didn’t ground the MAX (accepting to kill even more people).

      Now EASA is asking for a certain software standard, obviously Boeing didn’t follow regulations to present software in a certain form (Boeing said they did this always).

      Boeing killed !!!
      It’s cheap to pay some $$$ to victim families.
      So cheap that they don’t want to have court rulings in the US.

      Boeing will be taught to follow regulations.
      They still didn’t learn. Hard to change a culture of cheating.

      It’s their CULTURE
      and it’s not only the MAX, it might be EVERYTHING.

      Each day more and more comes into daylight
      and each day we learn that more needs to be checked,
      all regulators are watching this, not only JATR members, and it seems they all are supporting EASA.

      Next step will be checking the simulator software and I expect there will be mistakes found which will need to be fixed.
      Then simulator testings will be made and I expect something will be found or it will be asked if the MAX really behaves like this. There might be flight tests to fix the simulator.
      Then it might be April 2020 and certification flights are still not made but simulators can be produced so pilots can train. Maybe at that time Boeing learnt that they have to follow regulations and they start to redesign stabilators, elevators and triple AoA sensors. Triple AoA sensors are also needed for the 230 pax seating MAX-10 which might fly in 2021, but can’t beat 240 pax seating A321. Then Airbus announce an A220-900 and starts NSA with an A322, this is needed for decarbonisation per passenger.

      • Leon,

        I can’t agree more with what you are saying. However, I would like to add a few things.

        Number One: I am in awe of Boeing’s Stupidity in not electing to implement the Robust MCAS Solution that everyone already recognizes. This is going to cost Boeing a lot of time an money, but they don’t seem to care.

        Number Two: I believe the event which got us to the point where Boeing will not implement the obvious is the 787 Burning Battery Solution that the FAA and EASA and others let Boeing implement on the 787. Remember, instead of redesigning the Battery Cells so they didn’t overheat and burn, Boeing elected to encase the Battery in a steel sarcaphagus designed to contain the burning battery and vent all smoke off board. And this sarcaphagus was not just some interim solution – it was the permanant solution as far as Boeing was concerned. As a result, those batteries are still burning today.

        Number Three: Leon, like you I believe that the madness become part of Boeing’s Culture: especially the types of madness on display during the 737 and 787 incidents (oh…and the KC-46 incidents, too!). And this “Madness” – it took decades to ferment under such leaders as Shrontz, Condit, Stonecipher and McNerney, Muillenberg – and I fear that it’s going to be an awfully bitter vintage.

        Hell…can you imagine Bill Ford hiring any one of those gigolos listed above to save the Ford Motor Company? However, Bill hired Alan Mullaly to do the Job – and he did it. Remember Alan…the ex-Boeing guy that kept getting passed over for the Boeing CEO spot and whom Stonecipher threatened on multiple ocassions? Well…that’s what happened to the last tranche of Boeing Managers who could have cleaned up the company – and they are all gone now – and America ain’t making those types of potential CEOs any more!

        So here’s Boeing, it’s got a sick culture, and there’s no one left that can get the patient well: there’s no Louis Gallois, no Jean Pierson, no Thomas Enders, there’s no Alan Mullaly…NOBODY.

  15. Bjorn; I’d love to hear some thoughts (perhaps in an unrelated post) about the similar pitch force exerted by the A320neo’s similarly-upsized engines. Do they create a similar change in stick force near stall? Does their system handle it better (and how)? Or have they just been lucky?

    • Because the A320 has taller gear the Neo engines while larger did not have to be move up and forward to anywhere near the same degree. Consequently the A320neo will not see the same changes to flow over the wing at high AOA.

      That is not to say there will have not been any changes to aerodynamic characteristics, just that they will have been lesser and different. But because the A320 is a FBW system any these can be eliminated from the aircrafts normal law flight characteristics quite easily.

      Airbus is actually still making some of those changes. Recently some undesirable dynamic pitch characteristics were discovered if a landing is aborted just before flare with an after COG. Airbus will roll out a patch in 2020. Until the the aircraft COG range has been restricted.

      • – – – – But because the A320 is a FBW system any these can be eliminated from the aircrafts normal law flight characteristics quite easily. – – —

        Airbus did quietly announced that they are working on a software update for the NEO based on re-analysis after the MAX incidents, and that some configurations will have a COG restriction until the update is released. But clearly much less of a concern than with the MAX.

        That said, I’d like to make one note about the word ‘easily’: one of my classmates is working on a remediation program for a well-known FBW aircraft. It seems the sum total of algorithms added to the original design over the years to improve handling characteristics and especially efficiency have the control surfaces moving a lot more than in a classic mechanical design and in combinations that human pilots would never be able to generate. The end result has been cyclic loads on the structure that were never anticipated in the original design or seen in the fatigue frame, and there will have to be structural mods to the aircraft well before the design end-of-life.

        So – all is not always beer and skittles in the FBW design world either.

        • sPh
          You said: ” It seems the sum total of algorithms added to the original design over the years to improve handling characteristics and especially efficiency have the control surfaces moving a lot more than in a classic mechanical design and in combinations that human pilots would never be able to generate.”

          That is really interesting!

        • Well perhaps “easily” is not quite the right word, I’m not suggesting it is trivial. But the engineering process is straight forward for Airbus. The aircraft has the sensors, computing architecture and actuator systems in place to support flight characteristics augmentation via software. Moreover Airbus will have done similar augmentations many times in the past, both for the A320 family and for other families.

      • Specifically the excess pitch concern only effects the A321neo with the FLEX cabins. This is because an extra row of seats was fitted by moving equipment and the galley back. In addition the A321 has different trailing edge devices that make the wing 4% bigger. Specifically they use a double slotted flap instead of a single. The combination of further aft CofG and the different flaps means that excessive pitch might develop during a highly dynamic manoeuvre with flaps out . The solution is to alter the CofG limits temporarily. Lufthansa had simply blocked of the last row of seats. It’s importent to note this didn’t come out of in service reports, accidents or test flights. It came purely out of simulations. My readying is that in a dynamic manoeuvre that the tail can over shoot the maximum allowed pitch. Before the time this happens the elevator would already be limiting the pitch rate and restoring the correct pitch . The A321XLR gets a new optimized single slotted flaps derived from A350 technology which adjusts continuously. That flap may fund its way on to all A320 series aircraft. I don’t know why it’s taking so long to introduce the software upgrade.

    • Isn’t “stick force” what the pilot feels? Doesn’t Boeing hold that MCAS was designed to give the Max the same feel as the NG and therefore lessen the amount of training needed to go from one to another? Isn’t “stick force” irrelevant or moot in the A320, a fly-by-wire system that uses a joy stick or “side stick” like a drone or video game?

      • Yes “stick forces” forces are irrelevant in the Airbus FBW system on the A320. However Airbus provides other handling guarantees. Specifically a certain deflection on the side stick will result in a certain angle of attack. The issue discover is that an abrupt pull back on the side stick in landing configuration with an aft COF will result in an over shoot (e.g. in that configuration if 12 degrees is commanded the aircraft may exceed that for a brief period). This is not acceptable because it is unexpected. Until it is addressed the solution is to restrict the COG range.

          • In normal mode there is no chance of a stall (nor overspeed etc). The FBW systems implement “envelope” protection. The aircraft will simply not do some things no matter what the pilot asks of it.

            Bjorn has an interesting article on flying the A220 where he has the stick fully back with thrust at max and the aircraft just deals with it. Flying at 100knots 25 degree pitch angle.


          • In reading the link that Jbeeko posted about Bjorn flight testing the A220-300 (CSeries,) since I am not a pilot, I have to ask: Are you referring to the conditions in an airplane, when the MCAS software would “kick-in” on a b737MAX? TIA

          • sam, (if I can chime in and while not being a pilot)
            saying nothing about the values themselves (100kt ias and 25deg pitch), Yes.

            MCAS triggers under (undisclosed) low Mach and (undisclosed) high AoA. Notice pitch and AoA despite related are not the same thing, indicated airspeed and Mach likewise.

      • Stick force feedback is what prevents the pilot from over stressing the aircraft because as speed increases the effectiveness of control surfaces goes up by the square of speed. Look up dynamic pressure or Q. In an Airbus the FCS takes into account airspeed and air pressure and reduces the deflection of the control surface as speed goes up. That is called “alternate law”. Above that there is “normal law” which uses Rate Gyros that control say pitch, yaw and roll rate, accelerometers that limit stress on the airframe and alpha sensors that limit angle of attack. When you deflect a side stick you are demanding a certain pitch rate. It’s very useful. You can’t over rotate an Airbus and strike the tail since the angle is controlled. One way airbus have been able to increase MTOW is increase rotation rate to reduce takeoff distance. Read Bjorns “fly be steal or fly by wire”. it’s very good

    • Outside of the fact that the neo’s engines are not as far forward or up (the MAX’s are at leading edge height) the Airbus algorithm is more sophisticated. Firstly 3 sensors are involved in cross checking and intervention involves a 2oo3 (2 out of 3 sensors agreeing) rule. The other difference is that the Airbus algorithm uses the airspeed as well as alpha to calculate the maximum nose down attitude rather than just apply repeated unlimited stabiliser down trim. There have been cases of two alpha sensors freezing and alpha protection nosing the aircraft into a shallow but alarming dive. However the dive was shallow enough for the crew to be given enough time to trouble shoot the problem. The most famous is the incident that happened to and A320 flight LH1829 enroute from Bilbao to Munich on November 5th, 2014 with 109 souls on board. It lost only 4000ft of altitude.
      Airspeed and alpha sensors are not robust or diverse enough on any modern aircraft.

  16. Bjorn, Your comment “MCAS is a variant of Speed Trim and it uses the Speed Trim System’s base functions for its augmentation functions.” puzzles me a bit. The Speed Trim System functions off of speed. Its main function is to keep the stabilizer trimmed for the correct speed on takeoff. As the pilot changes speed and pitch settings, the Speed Trim System adapts and maintains the new speed setting. But, other than both STS and MCAS controlling the plane via the stabilizer, aren’t they separate systems in terns of functioning? MCAS is more of an Angle-of-Attack trim system, rather then Speed trim system. It’s function, in Boeing’s terms is to modify the column force, via the angle of attack not the speed of the aircraft? Does MCAS use the speed of the aircraft, as an input or is it only there to modify AOA using AOA as it’s input? In order to maintain AOA in a safe range.

    • To answer your last question. MCAS activates on AoA and Mach number. See Lion Air crash report.

      The report doesn’t provide details. That annoys me. I think we should have been told. But then Boeing may not be telling. They may not have told the regulators never mind the general public.

      I don’t know why it’s all a secret. It’s not IP.

    • To complement philip’s answer:

      From the Final Report and adding details from the NTSB report (as appended pp245-***)


      A schedule determines the desired incremental stab deviation from the last trimmed position as a function of airspeed and flap position

      When the flaps are down, the stabilizer rate is three times faster than when the flaps are up.


      Specific to the MCAS, the control law commands the stabilizer trim as a function of the following: Air/Ground, Flap position, Angle of attack, Pitch rate, True Airspeed and Mach.

      Further details:

      high speed maximum limit of 0.6°(…) low speed maximum limit of 2.5° of stabilizer movement

      physical maximum of 4.2° of nose-down movement

      The MCAS was revised such that depending on AOA, it would be capable of commanding incremental stabilizer to a maximum of 2.5 degrees at low Mach decreasing to a maximum of 0.65 degrees at high Mach

      The magnitude of the AND command was based on the AOA and the Mach.

      (!!!*)After the non-normal maneuver that resulted in the high AOA, and once the AOA fell below a reset threshold, MCAS would move the stabilizer ANU to the original position and reset the system.

      At any time, the stabilizer inputs could be stopped or reversed by the pilots using their yoke-mounted electric stabilizer trim switches, which also reset the system after a 5 second delay.

      with the magnitude based on the AOA and Mach sensed at that time

      MCAS would be active during manual flight only and would drive the stabilizer in flaps-up, high angle of attack conditions to improve pitch-up handling characteristics. The FCC software revisions include the following:
      *Logic to prioritize and command the stabilizer trim motor for MCAS operations using the active Speed Trim channel.
      *Output of MCAS Engage discretes from the FCC in command, to set the high stabilizer trim motor rate and inhibit the column cut-out function of the Column Switching Module in the aft direction.

      *MCAS may also trim up!

      I could not find the MCAS trim rate figure of 0.27deg/sec explicitly mentioned anywhere on the document. I did confirm it was consistent with values reported by the DFDR, if allowing for some rounding error (±0.5sec). This translates in a low rate figure of 0.09deg/sec when flaps are extended. Alike STS, flaps up, only verified by eye balling the slope in visual inspection of the trim graph on the report.

      I also noticed, manual electric trim was consistently behind MCAS rate, not sure if because an actual inferior rate output, more likely due to erratic switch pressure activation, hinted by visual inspection of the graph.

      Related with previous discussion here, it now looks more clear how integrating multiple signals is done, ie. not via trim rate but via the duration of the output signal.

      • Thanks Vasco,

        But I want the numbers. I’d like to know the AoA number for a Mach number at which MCAS engages together with the deflection of the stabiliser.

        • That I could not find, but these are big documents, I can’t be sure if they’re there.

          • They aren’t there. I’ve read the document front to back.

      • MCAS development and Certification overview (Boeing presentation to FAA – December 17) linked by Seattle Times, as obtained through House Transp. Committee.

        Details confirmed:
        Max command – 2.5deg (p7)*
        Hysterisis threshold – 0.5deg below the activation angle (p7)
        MCAS ANU – bellow hysterisis threshold (…) commands nose up (p7)
        High rate – 0.27deg/sec (p7 and p33)
        Low rate – 0.09deg/sec (p33)

        *I assume this implies a maximum magnitude (duration of output signal) of 9secs (10secs minus 1sec pilot reaction time = 9secs * 0.27deg/sec = 2.43deg)

        Still none of those numbers many are after, but:

        Maximum magnitude of stabilizer command is lower at high Mach number and greater at low Mach Number (for the same AoA above the activation threshold)

        Combined with “as a function of AoA and Mach” this means, different Mach only sets the magnitude of MCAS, AoA value above* activation threshold** is still the driving input.

        **(also uncertain)

    • Richard,

      As the pilot changes speed and pitch settings, the Speed Trim System adapts and maintains the new speed setting.

      As far as I understand it, this is not what speed trim does. Speed trim provides two augmentation functions.

      1) It changes the stabilizer setting so the pilot experiences more consistent handling over a wide range of speeds. MCAS essentially does the same thing except that it provides consistent handling at high AoA’s.

      2) It provides “active” damping for the slow Phugoid mode. It can only do this because the Phugoid mode is so slow.

      Peter Lemme talks about these functions.

      A stabilizer nose down trim increment as an airplane slows down will increase the stick force than otherwise, maintaining the stick force gradient.
      If the upset is a phugoid, the nose down increment will hasten the return to the target altitude and settle more quickly than otherwise.

      The LionAir final crash report states that the MCAS function is implemented as part of the STS (Speed Trim System) and both reside in the FCC. Peter Lemme’s description in the link I provided agrees as well.

      Finally, in my understanding the purpose of MCAS is not to maintain the AoA in a safe range or even to change the AoA, although it can. The purpose is to apply counter trim so that the handling near stall is consistent, reducing the likelihood of pilot over-control. An additional benefit is if the pilot does stall the aircraft, the AND trim provided by MCAS will aid in stall recovery.

      • Mike, When I watch this 4 minute video demoing the Speed Trim System on a 737, it makes sense to me that it keeps a constant speed. The pilot takes off and the STS system tries to maintain a target speed. Then, if the pilot re-trims the aircraft to a different speed, it targets to that speed.
        If a gust of wind pushes the nose of the aircraft up, it trims down to hold the target airspeed. If the pilot pushes the nose up and then lets go, the aircraft speeds slows, then the trim kicks in to bring the plane back to the original speed setting taking into account speed of pitch change etc so that it doesn’t overshoot the target airspeed. This all makes sense to me. Are we saying the same thing only with different viewpoints? One from a pilots perspective, and another from an aerodynamics perspective?

        • Richard,

          Are we saying the same thing only with different viewpoints? One from a pilots perspective, and another from an aerodynamics perspective?

          I think you are right about this. What you are describing sounds a lot like the damping of the Phugoid mode.

  17. Can I go back to basics. Elevators and stabilisers.

    If the stabiliser is in a neutral position then the elevators define up and down forces:

    Specifically if the trailing edge of the elevator goes down that creates an upward force -lift- so the tail lifts. If the trailing edge of the elevator goes up then that creates a downward force so the tail goes down.

    But what happens if the stabiliser is allowed to move – a trim stabliser. This means the stabiliser is not neutral. If a stabiliser moves – a trim stabiliser – then it can be used compliment the forces of an elevator OR oppose the forces of an elevator.

    That in turn means a trim stabiliser can be used to make the elevators inoperable. Boeing declared that the trim stabiliser can make the elevators inoperable.

    That’s the start, does anybody disagree. More to come.

    • What you said up to and including the sentence:

      That in turn means a trim stabiliser can be used to make the elevators inoperable.

      is true of any aircraft the employs a trim stabilizer. This includes the vast majority of jet transport aircraft in operation today, including all Airbus models as well as Boeing models.

      Then you say:

      Boeing declared that the trim stabiliser can make the elevators inoperable.

      It’s not at all clear to me what you mean by this.

      • I’m not aware any other OEM has declared the elevators become inoperable.

      • Your circular argument again. Everybody does it so there is nothing wrong with what Boeing does. Nobody but Boeing allows a trim stabiliser to be used in a fashion that makes the elevators inoperable.

        But I have to accept that it is possible. Why? The deflection of the stabiliser is independent of the deflection of the elevator. So the forces generated can be used to cancel each other out or partially cancel each other out, or even worse, reverse the intended force.

        If the tail forces cancel each other out then the airplane is tailless. This means the CoP is now forward of the AoA. The airplane is unstable.

        If the tail forces partially cancel each other out, then it still may be enough to move the CoP forward of the CoG. Again, unstable

        If there is force reversal. All hell breaks lose.

        The fact is Boeing are trying to use a stabiliser as an elevator. Does anybody else do that in the world of civil airlines? No, with one exception. So it can be done. The world of military airplanes prove that. The only civil airplane I know about is the Tristar.

        Trim stabilisers do not make the elevators inoperable unless they are being used for something they shouldn’t be used for.

        I told the truth for a year. It got me nowhere because of all the obscuration. At least I’ve cleared up one obscuration. But I had to lie to do it. It’s all about the CoP not about the CoG.

        You will need at some point to accept that what Boeing is doing is not typical, it is atypical. Indeed it is unique in aviation history. They have repurposed a trim stabiliser to be an elevator without doing the required upgrades. JATR used the word “novel”, I would say “stupid”. Not surprisingly they are getting it wrong. The same words applies to Bjorn as to you.

        RTS. Now March. So the continuous attempts at obscuration are not working. Thank God. Enough people have died.

        • Philip, In Peter Lemme’s article “737 FCC Pitch Axis Augmentation – Command Integrity Mandate for Dual Channel, Fail-Safe ” in the conclusions section he states:
          Of the Boeing types, only the 737 provides Mach trim, Speed trim, or MCAS functions for manual flight. As far as I am aware, no other Boeing models use the FCC to augment pitch axis while under manual flight control.
          I would like to know why only the 737 uses augmentation for pitch control? It does make you wonder about the harmonizing of the stabilizer and elevator on the 737. Most every 737 pilot has drilled into him, the quick response needed in a runaway trim situation. It’s a required memory drill. In the first Lion Air incident, the non-flying pilot (sort of acting as a flight engineer, in this case, was able to quickly turn off the stab trim motor). But, it wasn’t a true trim runaway. He was thinking only of the Speed Trim System, acting up, as MCAS was still a secret at that point. But, the column limit switches should have cut out the STS, unless you had a true runaway. MCAS wasn’t limited by those column limit switches as it had bypassed that limit. It was operating without any limits, except for the yoke trim switch itself, and a 5 second delay between firings. When the pilots wrote up the problem, they were still thinking about the speed trim system. Speed was unreliable. I”m still concerned about the possible mis-calabrated AOA sensors out in the field. If they are put on another aircraft other than the 737-MAX, then it will be a problem, but, probably not a crash.

          • Pilots are trained not to rely on the column limit switch for trim runaway, as it doesn’t correct the mistrim as the Boeing procedure recommends.

            Obviously it’s better to know and understand the cause of a runaway trim, and the Boeing documentation was flawed in that regard, for omitting MCAS.

            In the first crash, the captain did successfully interrupt and counter the MCAS inputs, despite not knowing about MCAS. In the second crash, the pilots did know about MCAS but still did not counter MCAS correctly. There are many factors outlined in the reports to explain why this happened, with recommendations made.

            New elevators and stabilizers? Not a word about that in any of the accident reports.

          • The 737 MAX is different. Your comments add to this.

            In these circumstances differential analysis is important.

            Keep going.

          • Richard,

            The 737 is not the only Boeing type to use augmented pitch control. Peter Lemme says “while under manual flight control”. The 767 for example, has mach trim and automatic stabilizer trim. It also has what Boeing calls the flight director. This flight director provides an automatic trim function in flight to keep the elevators in their “faired” position as much as possible during flight. I think the 767 was designed assuming the pilot would rely much more on automated systems during the climb and descent portions of flight compared to the 737.

        • Philip,
          great post. Is this affecting the 737 NG too?
          Then it might effect the MAX-10 too
          and new stabilizers and elevators are needed.

          • Actually its not, its some of if not the worst writing I have ever seen in any forum let alone a technical one.

          • Thanks Leon,

            I don’t know. But good question. One issue at a time given the current atmosphere.

            Best understanding:

            There is some correlation between the NG and the MAX but the NG did maintain a pylon to allow the airflow under the wing. The MAX put an end to that.

            That’s the best I have.

        • If the tail forces cancel each other out then the airplane is tailless. This means the CoP is now forward of the AoA. The airplane is unstable.

          This is fundamentally wrong. Static pitch stability is all about how the aerodynamic forces of each aircraft component change (location re. CG and magnitude) as AoA changes. It is not about the value and location of these aerodynamic forces at a given AoA. This determines trim not stability. So, in your example of tail forces canceling each other out, an aircraft originally in trim would experience an AND pitching moment because of the lack tail down force needed to maintain the trim AoA. This AND moment would reduce the AoA which then means that the tail now produces down force again. This down force then produces an ANU pitching moment that works toward increasing the AoA. This is statically stable behavior even though none of us would like the resulting trim of this aircraft under these tail settings.

          I told the truth for a year. It got me nowhere because of all the obscuration. At least I’ve cleared up one obscuration. But I had to lie to do it. It’s all about the CoP not about the CoG.

          I see you still can’t give up beating this horse. So it’s all about the CoP and not the about the CoG? Hmmm. So I guess this means that if a cargo aircraft loaded so the CoG is near the aft limit is climbing after takeoff and experiences a massive rearward cargo shift due to, I don’t know, say failed tie downs, that the pitch stability will not be affected because “it” is not about the CoG? No, I think it is pretty easy to see that the CoG location greatly affects stability.

          By the way, when you say the CoP must be ahead of the CoG for stability, what do you mean by CoP. Do you mean the CoP of the wing by itself, or the resultant CoP of the wing + engines, the wing + engines + fuselage, or the entire aircraft, wing + engines + fuselage + tail? It’s important to be clear here because it makes a big difference.

          • M. Bohnet,
            Your comments make me think of how MCAS might differ on the MAX and the tanker variant of the 767. For the tanker, at least, we know that MCAS is designed to account for a changing center of gravity as the mass of the plane changes during fueling.

          • By the way, when you say the CoP must be ahead of the CoG for stability, what do you mean by CoP.

            I meant to say:
            By the way, when you say the CoG must be forward of the CoP for stability, what do you mean by CoP?

          • RealSteve, They both use the name MCAS, but, the 767-tanker version uses 2 AOA sensors (and I believe a 3rd non AOA sensor, along with limit switches). The Air Force tanker uses their MCAS to help re-balance the aircraft as it dispenses fuel. The 737-MAX has a different goal in life. I doubt they use any common internal programming code, just the name MCAS.

  18. I’m not sure the figure provided helps me better understand the discussion. Does the line across it really represent the center of gravity or lift? My mind had the drawing pivoting from left to right, not up and down. All of the journalistic representations that I’ve seen of angle of attack have been side views, not top views. Is there a yellow and purple, side view figure of the Max/Ng? Isn’t the issue not just the forward positioning of the engines, but their lifting?

    • Like all aerdynamics it looks to be a complex relatioship that in the end is really is so what?

      Bjorn describe it if I am folloionw as more movement with the same force.

      The big qreustion seesm to be is this a flying problme or a smoothing problem.

      I take it as a smoothing problem that any pilot could deal with fine if he by some bizarre circumstances wound up in that high AOA regions.

      Is the MAX going to go nuts control wise?


      All aerodynamics are various compromises. This is nothing more than one aspect having a relatively narrow affect on another at an unusual attitude, so unusual most pilots would fly 30 years and never come close to it.

      What was a very miner issue got turned into a lethal problem by horribly written software. You can have good software, you can have bad software, you can have buggy software. This was as bad as software gets (killing people).

      Vastly more relevant is the manual trim problem. That still needs to be addressed from what I see so far.

      Is the two computer integration and the MCAS a finished product? No? There are still confirmation question on it and rightfully so. Don’t make another lethal issue out of this. Sort it and test it until it passes full muster.

      The issue at its heart is about the MAX and is it a safety aerodynamic aircraft and the answer to that is a simple yes.


      • TW,
        Your comments about the software fix for aerodynamic issues are interesting — and they have been interesting. It seems that you have deviated somewhat from your earlier questions about the computer architecture of the Max, its capacity and its ability to handle this type of solution. The Airbus fly-by-wire system is designed for this. But I gathered from your earlier comments on other articles that the two computers on the Max has not been designed to make these types of judgments. Have you changed your thinking about this?

        • Real Steve:

          My thinking was changed for me by people that know and understand it far better than I do.

          As it stood originally, it was a solid situation that with two computers you could not have a vote (and I still wonder what happens if none of the computers agrees with the other two on a FBW? )

          As it stood the system was not capable of cross linked for a mechanical setup and that was how its been certified since the Classic (two computers in a mechanical system aircraft) .

          Suddenly, they decide to integrate the two computers somehow in a cross check but not a vote out? .

          Now I know it can be done, what I did not know was you could do it and maintain the certification as it was not intended to be (or they chose not to originally maybe a better way to put it)

          All statements said not and then Peter Leme said it would be better if they did and they did do it.

          So I don’t think its a vote so much as a cross check and I i don’t know what the end result is as far it what it does if the disagree.

          Its why I think this has been a lot more major change than just saying, ok, you guys talk. Like all this is cascaded into some serious complexity.

          They have to talk, not muck each other up (and yes I have seen corruption spread from one to another) as well as what they do and what the control situation continue to be.

          Its a case where I have to accept its the right move without understand the underlining logic or approach (at least yet)

          And that of course begs a question that if it was the best way to go why was it not done original? That is rhetorical as we have the FAA and Boeing messing with things that far back (trail rudder issue)

          So I am watching to see the details (and if I can grasp them)

    • The Graphic says the line is ‘aerodynamic centre’. Yes the internet is split over whether is it blue or gold the centre of gravity or lift.
      The definition of “aerodynamic centre” is
      The Aerodynamic center is the point at which the pitching moment coefficient for the airfoil does not vary with lift coefficient (i.e. angle of attack)- making analysis simpler.

    • RealSteve,

      The top view is best for the purposes of this discussion because it not only shows differences in the engine location, but also possible differences in the wing and tail locations as well. All of these affect the aircraft static pitch stability. The wing, tail, and fuselage are the most important followed by the engines.

    • Based also on Boeing ACAP, but vectorial version of the B737-800 (green) and MAX8 (red) overlay. It is a heavy 1.1Mb svg, allows indefinite zooming, accuracy is stated around 6inch, before overlay editions. Includes aproximate CG per Leeham original drawing, and my vertical gross estimate. (Don’t know how reliable is this svg host, download for yourself just in case, full free use)

      • Wow, Vasco da Gama, that works! So the black line is an estimate of the center of gravity?

        • I placed the black line according to visual cues on the original Leeham image (yellow magenta). Vertically I had nothing to go by but some general info which is based on the theoretical wing CG not the plane itself which should imply an offset. So take the vertical placement with a much bigger grain of salt.

          • Why would the placement of the center of gravity and the center of lift for the Max and the NG be identical? Isn’t this the entire question?

          • RealSteve,

            The two diagrams are overlaid so the rearmost CG locations are lined up for the two models. The diagrams say nothing about the CoL.

          • RealSteve,

            I think when the diagram was first published was to aid our intuition in understanding what was at stake, not so much stating precisely where these CoGs would be in reality (they are also dynamic depending on pax, cargo, and fuel). But should in any case give us some relevant hints regarding MAX relative to NG: CoG should be offset to the left in the top and side views; CoL could also be offset (effect of the nacelles); Thrust/Power is also offset to the left and up* (higher power translating in the pitch up tendency).

            *It is also noticeable that only the intake is raised, the exhaust has the same center. In the side view this means there is an increase in slope of the air current passing through, not a mere displacement of the force center. In addition, on the top view, MAX engines are displaced slightly towards the inside (I am sure along with its own implications).

          • I think scientific illustration — particularly in journalism — is fascinating. It is meant to make issues more graspable for common people. Unfortunately, it often misrepresents the science. A good example is the way the tree of evolution has been presented historically. This whole discussion is about how the new engines affected the aerodynamics of the Max. The original drawing — presenting center of gravity (when center of lift might be the actual point of contention) — makes the difference between the NG and Max seem negligible. Your drawing and discussion adds more information. But probably what would have been most useful, is something that establishes as a given a particular, speed, altitude and angle of attack and plots on the drawing how center of lift might differ between the Max and the NG. Or maybe it’s the same?

          • RealSteve, I understand your relative disappointment. The first time I saw Leeham’s NG/MAX overlay, already having heard of the engine placement, I was also expecting a pronounced difference. It was not the case. This made me pay additional attention to lateral issues, ie the nacelles, power increase… Only to make me suspect the issue is much more subtle, there are a symphony of factors in the balance. The balance is definitely not the same for these birds. I could even accept that MAX achieved its own balance, while pushing its design to include these new engines. Somehow on this trail the required balance was neglected.

            We could, if we had access to the details, plot all those points of interest, but in a way it is beyond the main concern. Can further design improvements be found allowing for an admissible, certifiable MAX to go back in the skies?

            It is certainly proving difficult, seeing the recurrent return to service postponements. Fortunately, the onus remains with Boeing, after despicable early attempts at laying it in full at the feet of the pilots. I am sure they are working they’re asses off right now.

    • After 5 years as a software engineer followed another job of 16 years as a factory automation controls engineer, maintenance manager and quality control and 11 years in drawing office management combined with software writing combined with I can tell you it was incompetent engineering management. The kind of mix ups and mistakes Boeing made are well understood. Changes were being made with effected parties not being informed and not assessing the effects on their part. Boeing are claiming they did everything right but didn’t asses the pilots reactions to MCAS upset quite right. Everything was muddled. Grandfathering the A/C made it so much worse.

      • William, I quite agree. I’m hoping Boeing goes back to their huge design review meetings. At some points in the design of a new aircraft or a large redesign, they call everyone in to a large room and each area goes over their changes in some detail. Most of the audience is bored, until some part of the discussion turns to their piece of the pie, but, with lots of eyeballs watching and ears hearing how the whole design is put together, sometimes a small change in one area, effects another, that hadn’t been considered before. Maybe, with today’s work environment, they’d be doing this via a teleconference or video conferencing. It seems like a lot of wasted time for a lot of busy people, but, it were to uncover just one fatal MCAS type flaw, it should be worth the effort.
        I don’t think the new safety reporting committees and groups that Boeing has set up would accomplishes the same level of communication between far ranging designers and engineering groups.

        • In many industries they have a “squad check” where all the disciplines and outside parties meet at a large table go through and review approve various drawings over a few days. You can avoid the expense of these but I doubt drawings being ‘approved’ locally or by email/mail are getting as much attention as people think. I’m sceptical of parties attempting to do squad checks via software. squad check should get most of your work done very rapidly. Obviously documents and drawings need to be well developed before you get to the squad check.

        • You don’t have to have huge design review, just an ongoing constant one.

          You do need to have all parties review it and assess it as well as have the assessment or issue acknowledged and addressed yea or nay – and what is the only way to do that is constant communications and having all the affected parties closely located, hopefully in the same building.

  19. Mike Bohnet quoted the following from the JATR report in another posting:

    “Finding F2.7-A: The FAA was not completely unaware of MCAS; however, because the information and discussions about MCAS were so fragmented and were delivered to disconnected groups within the process, it was difficult to recognize the impacts and implications of this system. If the FAA technical staff had been fully aware of the details of the MCAS function, the JATR team believes the agency likely would have required an issue paper for using the stabilizer in a way that it had not previously been used. MCAS used the stabilizer to change the column force feel, not trim the aircraft. This is a case of using the control surface in a new way that the regulations never accounted for and should have required an issue paper for further analysis by the FAA. If an issue paper had been required, the JATR team believes it likely would have identified the potential for the stabilizer to overpower the elevator.”

  20. I don’t think anyone would dispute that the stabilizer can overpower the elevator. That is obvious from the area ratio alone, and is a consequence of the all-moving stabilizer design, which is not exclusive to the 737.

    With such a design, there is always a possibility of stabilizer runaway with overpowering control forces being generated. So there is a memory-item procedure for that.

    I don’t think anyone would dispute that for the stabilizer to move as it did in the two crashes, was a runaway situation. The deflection was too large and the only safeguard was for the pilots to interrupt & counteract the runaway with electric trim, then turn off the trim motor if the runaway continued, and use manual trim. That should not have been the only safeguard within MCAS. But runaway trim can occur for other reasons besides MCAS.

    I get the JATR point that the MCAS concept should have received more scrutiny, again no one would dispute that point. Or that a clearer conveyance of that information would have resulted in a more thorough review (especially the extension of MCAS to the low-speed case which greatly expanded its authority, and in my mind was a major failure in MCAS design and review process).

    I also get the point that using the stabilizer in this way was novel from the perspective of stall prevention, which more traditionally would be done with elevators via a stick pusher. It would not be novel from the perspective of trimming out behavior of the aircraft, which is all MCAS was ever meant to do. A similar idea had been used before on other aircraft. A similar idea is written into the control laws of Airbus fly-by-wire.

    So a lot of this comes down to perspective & semantics. Is MCAS stall prevention or is it aircraft trim at high AoA? I hope there is a review of the regulations that will clarify this issue. If the stabilizer is permitted to be used in this way, then there should be safety requirements established for it, just there are for stick pusher systems. I think this is the point that JATR was making.

    • Rob:

      I am not sure why this is relevant to MCAS (desing of aircraft and the regs maybe, I don’t know what they say about the relationship)

      What can do what is ?

      You have to know the moment arm, the size of stab and elevator and how much each one can move (speed may be a factors as well – iffy on that but then I am not an aerodynamics engineer)

      The stab has a larger size and a limited amount of movement.

      The Elevator is about 33% the size of the stab and has a lot more movement.

      If the elevator moves enough or is large enough it can overcome the stab.

      Unless is a basic violation of the regulation in affect then its a non issue. Boeing built the stab and elevator and was approved a long time ago.

      Should the elevator be able to overcome the stab? In a mechanical derived system maybe it should but that is not what is at issue here.

      Getting diverted into a irrelevant channel does what to deal with the issue at hand?

      Sure it makes buzz but whats the point?

      • TW, I did not raise the issue of elevator vs stabilizer, I was only responding to it as raised by others here. It’s clear that in these two crashes, the stabilizer did overpower the elevators, otherwise the pilots could have recovered with control column alone. People are identifying that as a design flaw. I was pointing out, as you also have, that the design is long-standing and long-approved, and the same principles have been used in other aircraft as well. It has also not been raised thus far in any of the accident reports.

        As you said in a previous post, the argument circles around whether the proper view of MCAS is stall prevention or aircraft trim. JATR pointed out that if MCAS is stall prevention, then the regulations were not written for that case, and that may have contributed to the lack of adequate review.

        • No, the discussion is simply what did MCAS 1.0 do to the aircraft and to the pilots trying to fly it.

      • It was approved.

        Should it have been? Then go back 50 years and stop it? Ok, that gets us where?

        Per current regs Boeing has build 10,000 legally –

        So, remove 7,000 or so legally build aircraft from the air?

        Or what is the po0int here?

        Oh, take those pickups with the in cab fuel tanks off the road to, even though they were legally build that way.

        • TW, The point is that 50 years after the 707 and 727 had accidents caused by severe stabilizer trim, we’re looking at the same issues on the 737-MAX. They issued two AD’s 50 years ago, to limit stabilizer trim.

          Should they look at the same fix for the 737-MAX? 50 years ago, the yoke mounted motor couldn’t work the stab trim at extreme limits, the same issue the MAX was having in the Etheopean accident, with the stuck trim wheel.

          A lot of the longitudinal pitch force FAR’s were written from the accidents in the 1950’s and 60’s that MCAS was put in place to adjust for.

          Having the stablizer over powering the elevator isn’t a new issue with the 707/727/737. It’s been happening for on this over 50 year old design. But, it’s having the same results in accidents. Runaway trim is something every 737 pilot is trained on in the simulator. But, MCAS inserts a new wrinkle, as it didn’t operate consistantly. It pulsed the trim in 10 second ON / 5 second OFF sequences. Not a typical runaway trim senario. If lower limits of travel were on the 737 stabilizer, the co-pilot would have been able to turn the trim wheel manually. The only question being at what new limit?

          • It’s possible that the output of MCAS was unbounded, as the input also was. We’d have to know what values of stabilizer deflection were expected or needed for MCAS to fulfill its purpose.

            The green zone on the stabilizer position display above the trim wheels may indicate what values were expected in flight. From photos, this appears to be 2 to 8 units of deflection with normal values between 5 and 6 units. So if the display is deviating beyond those values, there is a problem with the trim systems.

            We know that MCAS did drive the stabilizer beyond those limits after repeated uncontested activations. The captain of Lion Air did successfully interrupt and countermand MCAS to keep the deflection near the center range.

          • Following up on this, the conversion for stabilizer unit display to degree deflection is 1:1, however the units are offset and reversed to make them always positive and to reflect the effect on the nose, rather than the deflection direction of the stabilizer.

            The actual stabilizer deflection is roughly +4 to -11 degrees (15 degrees total travel)

            So roughly speaking, a unit display of 4 would correspond to a neutral stabilizer position. Values of 5 or 6 might correspond to normal tail downforce in trimmed flight (-1 to -2 degrees of deflection).

            Also the green zone on the trim position indicator is for takeoff, it shows the acceptable range to be sure the stabilizer is in takeoff position.

            if the indicated value was driven to near zero, as it was for the two accidents, that corresponds to +4 degrees deflection, so tail upforce is being added to the existing nose-down moment created by the offset of CoG being ahead of CoL.

            I couldn’t find a reference for elevator vs stabilizer forces, but my guess based on the crash evidence is that the exiting CoG-CoL moment combined with stabilizer upforce at full +4 deflection, can overpower the elevator.

            This would seem to be true for all aircraft that use the all-moving stabilizer, so may not be appropriate to think of it as design flaw, in and of itself. But lack of safety precautions to preclude it are an obvious concern.

          • RD:

            As this is a MCAS issue the rest is irrelevant.

            Should the MAX be allowed back in the air with the fact that the second trim motor was removed with the NG? (and the NG?)

            That is a relevant question as it changes the very basis of mechanical system.

            Bottom line this was simple in its execution (literally execution).

            MCAS 1.0 was done horribly wrong and it had nothing to do with the basic tail system and all to do with a bit of software that was lethal in the way it was implemented.

            Next this morphs into the landing on the moon never occurred.

            You can conspire into the MCAS issue anyt9ing you desire, that does not make any less absolutely bogus.

          • TW, the issue was raised by others, so I looked up the information and presented it, it’s correct to the best of my knowledge. If I said something that was incorrect, please feel free to present that.

            If it was known hat the stabilizer could swamp the elevator, then MCAS operating the stabilizer should have had precautions to prevent it from driving the stabilizer to full deflection, unless that was somehow necessary for its successful operation.

          • Rob & Co. Aviation Week back in April 11, 2019 was reporting on the changes that MCAS was undergoing then (it’s been changed since, but, I”m assuming this change is still there.
            (MCAS 2.0 is being explained)
            That MCAS cannot trim the stabilizer so that it overpowers elevator pitch control authority. The MCAS nose-down stab trim is limited so that the elevator always can provide at least 1.2g of nose-up pitch authority to enable the flight crew to recover from a nose-low attitude
            There is no FAR that I’m aware of that specifies that the Stabilizer can’t overpower the Elevator. The closest I can find is Part 23.655(b). Maybe there’s some verbiage not mentioning Stabilizer or Elevator, but, describing the aerodynamic functions that place some limits or loss of control elements to the Elevator?
            (b) If an adjustable stabilizer is used, it must have stops that will limit its range of travel to that allowing safe flight and landing.
            I”m wondering what new FAR’s may be in the works from the two MAX accidents.

          • Thank you Richard. I believe that MCAS 2.0 will address the runaway trim issues that occurred with MCAS 1.0.

            As far as whether the stabilizer can overwhelm the elevator in general, I believe it can for most all-moving stabilizers, at least for some flight conditions.

            The question is whether that much stabilizer movement is needed. I don’t have the answer but I would think this would be looked at closely in aircraft design, and the range of motion would be limited by reason and purpose. Doing otherwise would create liability without benefit.

        • “” Per current regs Boeing has build 10,000 legally –
          So, remove 7,000 or so legally build aircraft from the air? “”

          Now it’s in question what was legally built.
          MCAS was approved by the FAA, but it was an earlier version which was approved. Then Boeing made it more powerful and hid it. That was not approved, that was cheating. There was no MAX legally flying.

          The ways Boeing used to get exemptions with political pressure on the FAA and cover it from public knowledge is another thing.
          I believe there is no 737 NG legally flying either.

          Now other regulators know and don’t trust FAA anymore. Boeing’s cheatings will be uncovered and Boeing will have to pay to uncover it, EASA won’t do this for free.

          • Leon, this is your opinion and is not a finding in any of the reports. Boeing has made serious mistakes for sure, the FAA too, probably Congress as well. but no authoritative body is suggesting the 737 aircraft are not legally flying, were not legally flying.

          • Rob, only the documents which were stamped during certification are legal, changes built into the planes are not legal. If important information were kept secret and not described in the documents, those certifications are not valid too.

            EASA’s MAX certification was based on the FAA certification in the believe that all regulations were followed. If exceptions in the FAA certification process were made and kept secret from EASA’s knowledge, EASA’s certifications are not valid too.

            Since the MAX is based on the NG, I’m sure it will effect the 737 NG too. This will be a lifetime job and fortune for lawyers. EASA just started to discover facts. After over one year of JT610 we are just at the beginning.

            Boeing used the shelter of the FAA for cheatings and now Boeing is hiding and therefore will face complete investigations and rulings. To think otherwise is foolish, many people died. The world will uncover the cheatings.

          • Leon:

            Like Philip you don’t understand the system and make comments (Leeham does not edit those generally)

            The reality is that the MAX is a long evolution of the 707/727 and the earlier 737s.

            It has also been approved well before there was an EASA by UK, French and German authorities primarily at the time.

            In the case of the EASA they allowed two iffy engine to be put on the same aircraft when in fact from a true safety standpoint, you should never even have one iffy engine on an aircraft (the FAA quietly nixed that and as a lot of those birds fly US that in affect makes them adhere to it) though they still allow an iffy engine.

            So what you are arguing is that none of those prior certification’s are any good because FAA was corrupted by Boeing on the MAX.

            The FAA also allowed Boeing to continued to fly the 737 with a known severe rudder issue some yeas ago.

            What you don’t understand is the aviation system is not and never will be flawless.

            EASA makes mistakes, FAA may be corrupted.

            EASA is not going to get money from the FAA. Its an agency whose job it is (or should be) to certify European products. That is just silly. Its a fixed cost to society as the FAA is here.

            Clearly 99.9999% of the assessments and applications are valid and not corrupted or in question.

            The FAA also allowed Boeing to self certify the battery on the 787, that was taken away and now is not allowed.

            The only people more appalled than I am on the MAX are the family members who had people killed in the two crashes.

            But I have seen this happen before in Aviation as well as other areas.

            It goes in cycles. It will happen again.

            For the most part the system is better, but it will always be pushed to compromise and people will die.

            Its a fact of reality.

      • Richard,

        It is a long term issue, but:

        All other Boeing airplanes have been naturally stable. By that I mean, if the control surfaces are at neutral the airplane will still fly. Typically that means it will settle with a small positive AoA. Not always, as somebody else said, the B52 settles with a small negative AoA.

        The MAX is different. It wants to go nose up. This is bringing to the front the issue you have clearly researched extremely well. Thanks for that. It’s an education.

        In the simplest terms the trim stabiliser is being used in positions that are giving the elevators problems. So they repurposed the stabiliser to act as elevators. Fine if the required upgrades to the stabiliser are performed as per FAA/EASA regulations.

        In civil aviation, the only airplane I know that uses the stabiliser to manueve in pitch is the Tristar. But it complied with FAA/EASA regulations and it was naturally stable.

        • The MAX is not naturally unstable. It has similar stability characteristics to other 737’s and other commercial aircraft. There are no reports from pilots, in the hundred thousand hours of accumulated flight time, of instability or inherent flight design flaws.

          It does have an accelerated pitch tendency at high AoA, outside the normal flight regime, due to the larger engine fans that are required for greater fuel economy. At high AoA, they shift the CoL forward and decrease the normal pitch-down moment that aides stability. This effect is not exclusive to the 737, all aircraft that use those engines are affected. Airbus found a similar effect when they checked, currently their plane has tail seating restrictions to adjust CoG forward and reduce the accelerated pitch tendency. They are working on a software solution, just as Being is.

          MCAS use of the stabilizer for pitch control is not unprecedented. There is a difference in perspective as to whether this use constitutes stall prevention (traditionally done with elevator) or trim modification to enhance handling and feel (traditionally done with stabilizer). Regulations don’t currently address this issue, but I hope they will clarify it in the future.

          • Rubbish. The MAX was in service for just a few months before the first crash. The second crash was five months later. So rubbish.

            So why the use of the trim stabiliser? The use of the trim stabiliser is unique. It is unprecedented. Try your obscuration on others.

            Bit by bit it will be exposed. After 9 months the regulators are not convinced. Will they ever be convinced? We will see.

            I thank you for your certainty. But I don’t agree with your certainty.

          • To add to my previous comments. It will take a brave regulator to say what you have said. There is no precedence.

            My own view is that if the regulators do declare the MAX to be safe it will go to the courts. The courts will demand a public explanation.

            There are no precedents. If there are, provide the proof.

          • Philip, you are expressing your opinion as fact, and thus are only interested in those facts and other opinions that support your opinion.

            That is the same confirmation bias that prevented Boeing from identifying problems with MCAS. You cannot only develop evidence for one side, you have to develop and reconcile evidence from all sides, then draw your conclusions.

            If you start with your conclusion and then try to justify it, you will only learn or see your own side. That leads to comments like “rubbish”. You may not like facts that don’t support you, but they exist nonetheless. It would be better to try to integrate them into your view, or refute them with reason if you believe them to be incorrect.

          • First 737 max into service with Malindo Air (a Lion Air subsidiary) on May 22 2017. Lion Air crash was Oct 29 2018.
            Almost 18 months .

          • Rob

            Other way round. The use of the trim stabiliser in this fashion is unprecedented. Not been done before. That’s why the regulators are taking their time. If there were precedents the regulators would rubber stamp the MAX. There are no precedents. So your comments apply to you not me.


            Thanks for the clarification. How many were in service at the time of the Lion Air crash, remembering Boeing ramped up to 52/month? Not many.

          • Rob,

            To add.

            Richard is doing differential analysis. The 737 MAX is different to other Boeing airplanes and all other airplanes. Richard is doing great work on this. Read his posts.

            I’m also doing differential analysis. I have no evidence that a trim stabiliser has been used in this manner before. It is unprecedented.

            As I said it will be a brave regulator that declares the MAX to be safe for the use of the trim stabiliser is unique.

            You have said it happens all the time. I’m listening. But not to posts like that.

          • MCAS is not new with the 737, it was used previously on the KC-46 tanker.

            The purpose of the stabilizer is pitch control. The question raised by JATR is whether MCAS extended the purpose to stall prevention. If the purpose is stall prevention, that is a novel use of stabilizer. However, JATR has said this also depends on the characteristics of the aircraft.

            Boeing’s assertion is that MCAS is not a stall prevention system, it’s a pilot and handling characteristics aide, and was designed within that scope.

            JATR has said that if the purpose is stall prevention, then the existing regulations were not written for that case, and Boeing could have written an issue paper so it could be considered within the framework of regulations. Had that been done, it might have revealed the problems with MCAS.

            I believe that should happen now, and if changes are needed they can be undertaken. Both positions can be reviewed and a decision made in a fair and unbiased manner.

            It’s best to consider and evaluate both positions. You can’t assume one is completely correct and the other completely devoid of value.

          • Rob, it’s not clear to me that MCAS has the same function for the KC-46 tanker as it has for the 737 Max. From what I’ve gleaned from this discussion, the center of gravity in a tanker changes because the mass changes during a refueling operation. Furthermore, this is a slow and steady process. It seems MCAS might behave differently and be used for different purposes in the 737 Max, which does not undergo changes in mass during flight.

          • RealSteve, I agree that the cause of the initial pitch-up motion is different for the tanker, and possibly the rate as well, although free surface effects can be quite sudden and unexpected (they can capsize ships at sea).

            My guess is that tankers internally distribute fuel during refueling so as to avoid imbalances, and MCAS is there more for the unexpected.

            The essence and effect of MCAS is the same, though. The stabilizer is used to obtain the proper stick force and control dynamics during pitch-up. I’m sure that it does consider CoG shifts as well as CoL. CoG changes obviously would be a lesser factor for the MAX.

          • Amazing that you keep going back to the standard use of the stab for trim.

            If you put a 737MAX in neutral you will get a phugoid.

            It only has any added pitch up in an stall. And even then it does not appear to be that bad.

            The only innovation was to use it at high angles of attack as a smoothing method so the controls stayed uniform.

            Now you can jump up and down and bark at the moon as to what that amounts to as stall protection or pilot feel, it makes zero difference. It was put there to meet a specification – period.

            That the programing and the logic of the concept of how it triggered and by what and all the rest totally sucked is what matters.

            In fact, the Stablizer is used for speed trim, more authority at low speed and less at high speeds.

            You clearly don’t want to get over it.

          • TW, I think JATR’s point was that it might be better to seek permission than forgiveness.

            If Boeing had said, “hey, we’re thinking about adding MCAS at high AoA, how does that fit into the regs?”, in answering that question MCAS would have been examined more fully, which likely would have identified the flaws.

            Or thinking about it another way, the FAA might have said “we agree it’s a handling characteristics system, but it has failure modes that are similar to stall prevention failures, so we should conduct that level of failure analysis.”

            It wasn’t about Boeing doing something wrong, JATR agreed that MCAS was a reasonable solution in concept. But for Boeing to ask the question, might have avoided the whole tragedy.

            I thought that was a fair point, it shows JATR was looking at the big picture, while also trying to be reasonable.

          • Rob,
            “” I thought that was a fair point, it shows JATR was looking at the big picture, while also trying to be reasonable. “”

            No, every point in the JATR report must not be the opinion of the majotity of members, it could be a single opinion.
            But I wonder how many points in the JATR report were in agreement with the FAA staff too.

          • Rob:

            I think its more complicated than that.

            Boeing deliberate did not inform the FAA the had increased the authority of the system.

            No one did a real world failure analysis of AOA

            MCAS was approved for the purpose intended.

            MCAS was deemed not needed and could be turned off if required.

  21. The boeing slides presented to the Congressional hearing-
    Interesting – and some so called data around pages 15 16 17 as to what boeing ‘ believed’ or postulated re MCAS

    Note although it shows at the bottom being proprIetary – it was sent to the hearing and is now a matter or congressional record

    The link was posted in the seattle times- who simply asked for it.

    Depending on your system- it can be made into pdf format or simple OCR’d text format.

    at least in the chrome browser

    • The really useful or legally problematic stuff has been blacked out. And why include the AoA disagree cockpit display when that was an extra cost option not on all Maxes

      • Just a guess- By that date when submitted to FAA- Someone in Boeing knew that incorporating that the AOA ‘ clock’ display option had disabled the AOA disagree display note on all MAX displays even though the disagree light had been on all NG models.

        Covering posteriors

        • Bubba:

          If you don’t know about MCAS 1.0, what goes and relevance does the AOA disagree or the display do you?

          Its not till you pull up the flaps that things happen and then it hits the fan big time.

          Another indicator on the panel when all hell has broken loose is simply going to be ignored.

          What is effective is if there is an AOA disagree, MCAS gets turned OFF.

          For 346 people its too late, but that is what a disagree light does for you and who cares about the display?

          • As I understand it- when Boeing-collins- provided for the ‘ AOA clock ‘ option in the software, it made no difference whether or not the ‘ option’ was ordered at a ridiculus price, the software change involved disabeled the disagree light- message in either case. Thus those who did NOT order the ‘clock display ‘ option also wound up with the disagree message-light disabled.

            Whereas in the NG, the disagree light worked and was standard.

            Boeing became aware of this gotcha, but delayed doing anything about it- as it was not considered a ‘ safety’ issue.

            after all, doing it correctly would be a cost to Boeing and eat into the profits . ..

            Whether or not it ( disagree light ) would have made a difference is arguable, since the only way to turn of HAL (MCAS) was to shut off all trim and leave one with the Hamster wheel . . try pedalling your non electric bike with only one pedal while going uphill to avoid the falling rocks.. or avalance or flood waters impending.

          • Reality is that non safety of flight issues are often left to the next software upgrade.

            I don’t agree with a lot of it but that is how its done by all the AHJs.

            So while I have no issue beating Boeing up where its due and it certainly is, but to pile on everything diminishes the real issue.

            EASA dropped the need to not have known bad engines on an aircraft, FAA put it back in after the Rome 787 Trent debacle.

            Me? I don’t think any engine with a problem should be allowed.

            You no longer have N+1, you have N – 1.

            There is a lot of ugly lazy faire in the AHJs.

    • Most repeated 6-gram in that Boeing slide presentation to FAA:

      this, that, yada, yada
      No process violation or non-compliance

      There is also a 7-gram more often repeated, but does not count:

      Copyright © 2018 Boeing. All rights reserved.

      I think this is exactly what Chief Technical Pilot Mark Forkner meant regarding jedi-mind tricking regulators. ie. “You will accept republican credits (waves hand in up-right/down-left motion)”

  22. Can I ask everybody a question.

    The JATR report said the use of the trim stabiliser is “novel”. I took that as sarcasm but some appear to think it was a complement. The reason I took it as sarcasm was that the JATR report was beyond damning but the Lion Air crash report was also beyond damning.

    I think the JATR report and the Lion Air crash report are being ignored by some on this web-site. So how damning does something need to be before note is taken?

    So is the word “novel” sarcasm or a complement?

    • EASA is very polite, but they know that they will catch every Boeing/FAA mistake and won’t hide it.

      Recently FAA and EASA were checking simulator software, but the software wasn’t presented in a certain form, so they gave it back to Boeing to change the form. Boeing said that they always used this form.

      No Excuses, regulations need to be followed. There is a new sheriff in town.

      • Well if they did not tell Boeing that the form changed then?

        EASA is in its own way as bad as the FAA.

        Trent 1000 should be grounded until its fixed as much as the MAX, but is allowed to fly and try to kill people

        Two bad Trent 1000 are allowed on the same aircraft when none should be. The FAA at leas stepped in and amended that to you can only have one bad engine on an airplane (ungh)

        • TransWorld,

          I’m not into Trent 1000. Did the FAA change the ETOPS too?
          If there is really a big problem I would agree with you, pilots can’t find a Hudson River everywhere. I don’t care if it’s Boeing or Airbus, safety should be the priority, no excuses.

          I’m confused, Scott Hamilton wrote about RTS in 2021.
          Were the software form about the simulator software Fake News?
          Now it seems the FAA approved the simulator software.
          Will the FAA go alone?
          The primary report of the TAB seems to be politics involved.
          Is this a Gong Show or will EASA follow?
          I can’t believe after the JATR report that the MAX can be fixed so fast.
          If the Jackscrew is involved it might crash again.
          I only know I don’t trust Boeing or FAA or politics, they all didn’t learn something.

          • Leon:

            ETOPs has been amended (shortened) on the Trent 1000 based on what the current assessment of the engines are.

            EASA erased the you can’t fly with two known issue (I simply call them bad) engines on the 787.

            As an aside, they were trying to do that in India with the A320NEO PW and the Indian AHJ told them to knock it off which is the right thing to do.

            The Trent 1000 has now had 4-6 in air failures on their known issue. RR has no handle on fix (pushed to 2021 now I believe)

            Their current fix is to replace the old blades with new blades of the same issue.

            That you don’t trust Boeing or the FAA is beyond fair.

            But this is not new. Boeing has used denial years in the past and will do so again. The FAA has failed to manage Boeing and airlines in the past and will do so again.

            I would suggest learning before posting (you have the right to post per Leeham policy) but so far I don’t see anything of any value and a complete lack of misunderstanding. Following Phillip is not a recipe for being informed.

            I am truly serious about getting a MS Flight Simulator on your machine.

            Phillip has posted total nonsense about how you can’t fly with no moon in a storm when in fact that is standard. Its normal to do so. Its even normal to do so with some equipment not working (some requires landing as soon as possible if it fails)

            Aviation is an extremely complicated field overall, piloting is a sub set of that and is a skill and gets into areas the normal population has now knowledge of.

            FS programs give you an idea of whats involved in the flying end.

            The regulatory end is one that there is no Sim for but should be.

            Well as jackscrews have been the default for a long time, simply a jack scfrew is not an issue.

            AK had a non 737 Jackscrew take down an MD-80 (or one of hte MD series they flew at the time)

            Combo of worn out gears and the wrong grease (aproved by Boeing but not allowed for the metals in that MD jackscrew) .

    • philip

      ” So is the word “novel” sarcasm or a complement?”

      suggest your question can be answered if you replace the word “novel ” with “unusual ” or ” different ” or ” rare ” or ” not normal “

    • I think they weren’t editorializing, but, just claiming that is was a ‘new approach’ and that it needs more investigation since there isn’t a lot of previous similar designs to compare it with.

    • It’s neither, the word “novel” was used in the sense of not having been previously considered in the regulations, for the case that MCAS was being used as stall prevention. Hence the finding that Boeing could have submitted an issue paper to have it considered and clarified.

      Both JATR and Lion Air crash reports have been referenced extensively in this discussion. “Damning” is a conclusion, not a finding. You won’t find that word in either report. A full reading of the reports shows the reality to be more nuanced than that.

      Your agenda is to keep the MAX on the ground, you believe it is patently unsafe. So, you are on the hunt for evidence to show that to be true, beyond that which has been presented in the reports.

      Others (including many experts I have read) have read the reports and believe the MAX can be safely returned to flight, with corrections of the problems that have been found. That is my belief as well.

      None of the accident reports have referenced or recommended a permanent grounding, as they surely would if they believed the MAX was patently unsafe. They have identified issues with the expectation that those will be addressed. There should be follow-up on those things to make sure they are addressed, before a return to flight.

      • Rob,

        I should add that none of the accident reports or the JATR report have referenced or recommended increasing the MAX stabilizer and/or elevator size. I think you already mentioned this in one of your previous comments but I felt it bears repeating here.

        The JATR report also clearly states:

        MCAS used the stabilizer to change the column force feel, not trim the aircraft.

        This is the same thing Bjorn has been saying all along. The JATR body clearly understood what Boeing was trying to accomplish with MCAS. But like you’ve said, they didn’t necessarily agree with Boeing that MCAS was not part of stall identification.

        • Thank you Mike, I totally agree. I can also see JATR’s view that it might have been better to raise the issue in a paper submission, for clarification. MCAS was getting into a grey area where it could be viewed either way. This is 20/20 hindsight but it has merit, and JATR was right to point it out.

        • “” none of the accident reports or the JATR report have referenced or recommended increasing the MAX stabilizer and/or elevator size “”

          Reports can only hint to regulations which were not followed. There are different ways to reach a regulation. Reports are not meant to redesign the plane, it’s Boeing’s job to choose a design to follow regulations.

          • Yup. So I guess the JATR report or any of the accident reports can’t really be used to vindicate any silliness like thinking the MAX is statically pitch unstable, or thinking that the stabilizer and/or elevator will need to be made larger because of this, or thinking that the 737 is somehow the only aircraft ever to use the stabilizer for pitch augmentation. Go figure.

          • Because Boeing was cheating and is still hiding, I can think of many things and I think EASA and other regulators are thinking too that there’s something wrong. That’s why flight tests with and without MCAS is one of the plans. I’m sure EASA will find the mistakes, it will only take longer.

            Thinking about that the MAX won’t fly next year gives this a new dimension. After 8 months grounded Boeing lost $10 billions with strange accounting. In 2021 it will be grounded over 21 months. Soon Boeing will stop MAX production.

          • Boeing is actually starting to delver 737MAX to some Airlines so they are pre-positioned to start ops when its cleared to fly.

            Now we will see if that works, but Airlines are staring to say that they believe the fixes are done, FAA has to confirm it of course and they will resume operation when the ok is given and the pilots have the training the FAA has agreed to or the Airlines feel they need.

            Airlines can exceed FAA, they just are not supposed to do less.

    • It is neither sarcasm or compliment. In patenting, novel means that the idea is new and was not used before. For a certification agency, it means it is a new approach that has never been certified according the current rules and regulations.

    • Novel, meaning unusual, new, or unprecedented, I think, is a just statement of fact in this case, isn’t it?

    • The ST reported that the pitching moment at high AoA and high speed were first seen in wind tunnel testing.

      • MIke, Thanks for the tip about the ST article. I took a look at it,
        and it seems that wind tunnel testing showed the high speed problem, when Ray Craig was a test pilot around the
        2012 timeframe, in a simulator. That is what I refer to as MCAS 0.5. But, the later change to MCAS 1.0, which is the one that caused all of the problems (taking out the g-force sensor and column limit switches), after test pilot Ed Wilson found some problem with a low speed stall situation, that wasn’treally tested for or found during wind tunnel testing?

        • Richard,

          Not sure if the low speed problem was at all detected in the wind tunnel. I’ve read that the high speed problem was driven primarily by compressibility (high subsonic Mach) effects at the higher AoA’s. The low speed problem, on the other hand, is probably caused by the flow remaining attached around the nacelles at high AoA’s while at the same time the turbulent boundary layer near the wing trailing edge separates. At least this is what Bjorn is saying and his guess is by far the best guess I’ve read so far in relation to this whole issue. Compressibility effects are easier to account for when going from sub scale wind tunnel results to the full scale aircraft. Boundary layer transition and flow separation can be more tricky to account for when changing scales.

          I think that the column limit switches were already taken out in what you are calling MCAS 0.5. Otherwise they would interfere with MCAS activating when the pilot pulled back to execute high speed wind-up turns.

          • Mike, I think you’re correct, they took out the limit switches probably on the original MCAS 0.6 version.
            That really brings up an interesting point. Boeing was now putting a computer in control, overriding the pilot. Not a stick shaker, not a buzzer or light, not a stick force stiffener, but, a complete takeover of a primary flight control of the aircraft. I think that’s one reason Boeing doesn’t come out and mention the exact MCAS activation is XX.456 degrees, and then resets when the AOA comes down to YY.678 degrees. That would show that MCAS is more of a stall prevention system than a pitch augmentation system, which seems to me to be semantics, but, probably has more legal or regulatory consequences.

          • Richard, the column limit switch could not be functional for MCAS, because pulling back on the column would lead to the situation MCAS was intended to help avoid. I don’t think it was ever intended to remove control from the pilot, it was just a functional necessity. I believe this still is true for MCAS 2.0.

            I think it’s important to understand that MCAS actions in the two accidents were the result of errors in input and errors in design & implementation. There wasn’t an overt intention to take control from the pilot.

  23. I wonder,
    Why B737 MAX test pilots did not discover MAX problem during testing, and how come such of this problem did not discovered by FAA during test certification cycle.

    • Test pilots were replaced. There was pressure to keep the mouth shut. Gathered information was kept in small groups.
      FAA was pressured too, Boeing did that through politics.
      I even question that FAA made own test flights. Certification flights might have flown with Boeing pilots flying and FAA pilots only monitoring.
      This was all about cheating for profit.

      Then Karma stroke, now Boeing will need to follow regulations 100% to get certifications and will likely be not competitive.

    • To answer factually, Boeing did not test for the AoA vane failure scenario that led to the crashes. They classified the MCAS system in a safety category that did not require testing to that degree of rigor. The FAA concurred with that assessment. at the time.

      At least initially, that classification may have been justified, although that is still open to question. However as MCAS development continued, it acquired more authority, and it’s widely agreed now that this should have triggered a new safety assessment and additional testing. The FAA knew of this change, but it was not presented in a manner that generated sufficient attention.

      Without adequate testing based on a new assessment, the fault mechanisms were not fully explored or identified, which led to MCAS being approved by the FAA under the original assessment.

      Additionally Boeing’s assumption was that an MCAS fault could be handled as a runaway trim scenario by pilots. Thus they did not detail MCAS operation in the aircraft documentation.

      However it was found that pilots had a mixed response, some were able to correct the excessive trim, some were not. So this assumption was not valid, nor was the decision to omit MCAS from the documentation.

      Investigative bodies have extensively detailed these contributing factors, among others. The result has been extensive scrutiny of MCAS, the MAX, and the certification process. The regulating authorities have said the MAX will not be certified for flight again until they are satisfied the identified issues are resolved.

      MCAS now has been redesigned and tested. Boeing is also working on reconfiguring the two flight control computers to function as dual masters. so each will monitor the other to flag errors. This is a major task and has taken considerable time, but should enhance safety.

      Regulatory issues also need to be addressed. Funding cuts at FAA have resulted in extensive delegation of certification assessments to Boeing, under FAA supervision. Overall that system has functioned well, but clearly failed in the case of MCAS. So it needs to be reviewed, improved and possibly restructured to avoid a recurrence.

      • Rob,

        Very good synopsis of the whole affair.

        Some people seem to ignore or forget that Boeing is completely changing the software architecture of the two FCC’s. They are creating true duplex redundancy in response to FAA demanded testing that was unprecedented according to reporting in the ST. This takes time. Instead these same people choose to view the fact that Boeing is not done yet as evidence that Boeing is trying to hide the terrible MAX from regulators. Like that could ever work.

        • “” Boeing is completely changing the software architecture of the two FCC’s. They are creating true duplex redundancy “”

          Boeing should have done that decades ago. But as we know now, safety was not a priority. I wonder how many lives could have been saved. By time former accidents will be reviewed, especially when Boeing blamed pilots.

        • Full duplex requires failure isolation. As far as I’m aware, Boeing are simply accepting sensor disagree. From memory, the AoA sensor disagree is to be a delta of 5°. But no attempt will be made to determine which of the two sensors is wrong as per the definition of failure isolation.

          If sensor disagree occurs, MCAS disengages. The pilots then are on their own. At night with no moon in a storm, therefore no external reference point, and an AoA disagree of 5° or more, not knowing which sensor is wrong!

          Here is an interesting crash.


          The point I’m making is they are changing the problem, not addressing the problem. It will be interesting if it’s accepted for it is not true duplex. It’s dummed down duplex.

          • According to the reporting, it will be dual master, which is a shift from master-backup as used previously.

            Dual master systems have some ability to do fault detection, diagnosis & response. It’s not the same as 3 computers with election as used on Airbus FBW. 3 is better than 2, just as 4 is better than 3 for military applications.

            Still it represents an enhancement of safety. It’s not an attempt to implement FBW, it’s an attempt to eliminate potential points of failure.

            The test that precipitated that change, was a series of bit-flips that were focused in pitch control circuits, to see if they could cause a similar sequence to the accidents. The odds of that happening are extremely remote.

            The results were that the MAX under these conditions was recoverable, but the safety margin was not great enough. So Boeing had to increase their tolerance for pilot response, which means the computers have to handle the issue either entirely, or more quickly to allow more time.

            With dual master architecture, the presence of a bit flip in one computer should be detected by the other, so it would be alarmed immediately rather than waiting for the effects to be noticed.

          • Active/active or dual master is a minium. It’s not relevant.

            I’m addressing the sensors not the computer. There was no failure checks of the sensors. This was proven with the Ethiopian Airlines crash. It accepted an AoA of 75°. That is never valid and should have caused the FCC to ignore it. That is what I mean and everybody else means by failure isolation.

            What you are addressing is something else, physical failure of part of the CPU or memory circuitry. Modern CPUs identify failed circuitry and reroute. The same applies to memory. So failure isolation is performed by the CPU or memory . But the CPUs and memory used by the FCCs are 80286. They are 40 years old and don’t isolate circuitry failures.

            So no.

          • Philip,

            You are remembering info on the original software fix that Boeing had ready in June. You don’t really know what possible failure isolation Boeing is going to implement in the their latest dual master architecture because none of us do. Boeing hasn’t shared it publicly. However I’m sure the FAA knows since Boeing already submitted the documentation that the FAA ultimately deemed non compliant.

            This architecture improvement makes it much easier to implement potential failure isolation checks using e.g. an artificial AoA derived from the ADIRU’s. But either way, at minimum MCAS will fail safe i.e. do nothing if there is AoA disagree. As the several flights prior to LNI043 proved; no MCAS no problem in the normal flight envelope.

            So, Boeing is addressing the problem that needs to be addressed.

          • Yes, Mike is right, addressing the need for redundant sensors has been accomplished in MCAS 2.0.

            The dual master architecture is a separate issue raised by the FAA regarding response times. It’s inline with the JATR and Lion Air report recommendations that expectations for pilot response times should be re-evaluated and lengthened.

            I suspect this is also the FAA response to Boeing’s assertion that recovery is possible, they are being more stringent and sending the message that recovery should be not just possible, but likely or overwhelmingly likely.

            It’s all to the good since it promotes and reinforces safety. It’s not that the earlier architecture was unsafe, but the new architecture will be safer. Boeing has got with the program and is willingly doing the work, despite the extensive delay and cost.

          • I don’t think philip’s contention is whether or not Boeing satisfies a proper MCAS certification alone, as it should have done from the get go. Philip insists on this, I think, reflecting on the original reason that lead Boeing to create an MCAS in the first place. That reason, and all that goes with it, must necessarily be certified, MCAS alone may not be a sufficient measure from this point of view to respond to certification requirements.

            Raising MCAS up to the task, is not the same as raising MAX8 to the task.

          • Adding:

            Specially if MCAS effect (due to its novel compliance with the certification) is diminished in this 2.0 version. Will we pass the threshold about which the the aircraft becomes uncertifiable again?

          • Vasco, I think the re-certification for flight, is ongoing right now. EASA has said it’s a month or so behind the FAA and wants to run some additional tests in December.

            All results are not yet in, but it appears that re-certification will occur. Once it does, there is still a long delay to bring the MAX out of mothballs and apply the fixes.

            I think what Philip wants now, is to either scrap the MAX, or begin the certification over from scratch, as a new airframe. There doesn’t appear to be much momentum for that. I think the regulatory authorities will be satisfied with the changes they are recommending. Those seem reasonable to me, based on the facts that we know.

            I could be wrong about this, I guess time will tell. The authorities have been given unbounded time to consider the MAX, and I think that process has been fair. I don’t think Boeing has any complaint about that, or the delay, or the costs.

          • Phillip:

            “At night with no moon in a storm, therefore no external reference point, and an AoA disagree of 5° or more, not knowing which sensor is wrong!”

            You don’t get it. No visibility, on instruments in a a storm is what a pilot lives and breaths for! The rest is just boring as hell.

            You also do not get that failures occur and we train for those failures.

            Get a Flight Sim for your computer and run a basic pilot instrument course.

            The only thing I could not recover from is if the aircraft went upside down when I was on instruments and only because the AH we had would not handle that upside down area (right ones can).

            We not only train on instruments , we train for instrument failures.

            The only thing as fun was a 90 degree 25 knot cross wind landing (yea I nailed it though the instructor handled tower comms) .

            If you can’t handle that stuff , then you are not a pilot.

            The issue with MAX is not that its bad, its the MCAS put so much crap into the system with other twists that they never got focused on the problem.

            In the end if they had kept counter trimming it would have been a bit rough but easy return to airport (or got it neutral and then turned off the Stabs but they did not know that in Indonesia)

          • Rob, I tend to agree with most of those points. I think the parties involved know what is at stake, after two accidents, allowing a third would only check mate them. Philip’s expression was “brave” the one to leave this credit in the wrong hands, at any level of responsability.

            I will keep following as many other.

          • Rob,

            My original response disappeared. But it might come back so I will summarise and then add.

            I wasn’t addressing CPU bit corruption. Modern CPUs do that. The 80286s as used my the MAX aren’t modern so don’t address bit corruption. I was addressing the failure of the FCC to identify invalid sensor inputs.

            To add.

            In the simplistic terms, 346 people died because the FCC didn’t recognise that an AoA of 20° is invalid and an AoA of 75° is invalid. Such identification is easy.

            So the FCC doesn’t identify easy failure states with regard to sensors . If it doesn’t do the easy it won’t do the hard.

          • TransWorld

            The nose wants to go up and the flight deck is useless!

          • Philip, you are correct in your criticism of the failures to provide bounds on inputs (and possibly outputs) of MCAS. That was a software error though, the CPUs were not responsible. It should be resolved now in MCAS 2.0.

            Boeing is adjusting the architecture for another reason. They had to pass a more stringent test for bit corruption caused by cosmic rays, in the wake of the accidents. The number of bad bits was expanded to 5, and additionally they were focused in the stabilizer control circuits. The intent was to cause an un-alarmed stabilizer runaway (similar to MCAS failure) and see if 3 pilots could recover. They did, but with not much margin.

            So then they added a delay in pilot response (similar to the film “Sully”) and then only 2 of the pilots recovered. The FAA deemed that unacceptable.

            I think they were sending Boeing a message that recovery should not just be possible, but likely. Unrealistic estimates of response times have been a focus of the accident reports. Also, the FAA showed it was willing to hold Boeing’s feet to the fire, even though they argued that the odds of 5 simultaneous bit flips in a few circuits were nearly infinitesimal.

            Given that the source of the error was bit flips, and Boeing could not control pilot response times, the solution was to monitor for bit corruption continuously. That is a significant shift and it required the flight computers to monitor each other.

            Boeing has now had to certify those changes, which accounts for the nearly 1000 hours of flight testing. But it should make the MAX safer, and it drilled home the points that safety is critical and pilots need time to recover.

          • Richard, the Seattle Times reporting on this issue from August, clarifies that the first early reporting from June was incorrect. It was not an issue of the older microprocessor being slow, as was first reported. It was pilot response to the bit flip issue, and the lack of any internal detection of that issue, such that pilots awareness began with a misbehavior of the aircraft. In this case, an aggressive dive since one of the bits chosen was the stabilizer drive.

            Boeing … “emphasized how extremely improbable it is that five single bits on the microprocessor would flip at once and that the random bits would make these specific critical changes to the aircraft’s systems. While it’s a theoretical failure mode that has never been known to occur, we cannot prove it can’t happen, so we have to account for it in the design.”

            Boeing … “added that early published accounts of the fault suggesting that the microprocessor had been overwhelmed and its data-processing speed slowed, causing the pilot-control column thumb switches that move the stabilizer to respond slowly, were inaccurate.”


          • Rob, Thanks for clearing that up. I stand corrected.
            I”m amazed that this is the reason they are going to dual channel. I wonder if they will be doing any hardware changes to add protection also? Collins makes the FCC for both the NG and MAX, while Honeywell made the FCC for the Classic series I wonder if they will update the FCC software and/or hardware for the NG also? That’s an awful lot of programming changes and testing to do in a few months. A new FCC usually requires years to get approved. I’m glad the FAA is not rushing this certification process.

          • Richard, I agree, the delay and large number of flight test hours since June has mainly been about recertifying FCC’s, as well as waiting for the various accident reports.

            Collins is feeling the heat too, since they produced the faulty software. It was also them that did the outsourcing to an Indian company. However, it appears that they produced the software to spec, so it comes back to Boeing.

            The FCC win from Honeywell was a huge boost for Collins, and they did it mainly based on cost. So this has really shaken them. But as yet, there is not evidence that cost or outsourcing was a contributing factor.

            Although one wonders, if an experienced software engineer wouldn’t have asked about basic things like bounding inputs or outputs on MCAS. But then again, most of us have been asked to be quiet and do the work at some point in our careers.

          • Philip:

            Well if you are not a pilot I can see the flight deck confusing you.

            Well that and the whole thing about aerodynamics and stability.

            But as a pilot, when I had no alarms and just the basic instruments, you could not shake me, and I was just an average pilot.

            My instructor go bored with no challenge on instruments with me because the recovery was so simple (even with instruments masked out)

            He was fairly new at it and he asked me if I wanted more of a challenge. You bet.

            He never would tell me what he did (aerobatic were against school policy) but I believe we did loops and barrel rolls (though we came out in less than 90 deg horizontal and less than 90 deg vertical before I got it back)

            I was under the hood and could not see what he was doing.

            Never missed one, never.

            My inner ear was toast, vertigo rained supreme.

            And as a new pilot that was just average over all he cold not shake me. I had less than 60 hours and was not even close to an instrument rating at the time.

            You really have to be a pilot to understand this stuff.

            Frankly I am willing to bet with what I know about MCAS I could recover a 737 with minimal fuss from MCAS 1.0.

            You just have to trust, know and understand the instruments , run your scan and cross checks. Well you can’t panic like AF447 did.

            The Lion and Ethiopian pilots did not panic. They did not do it all right but they did not panic.

            Been there, done that, have boxes of T shirts (successful survival in all sorts of crisis conditions).

          • Rob, Internally, at Boeing and Collins, I”m sure there are some discussions going on about the MCAS software and how it was specified, tested and verified.
            It had to be a very small group that initially implemented the fatal changes. How do you make sure something like this doesn’t happen to future flight control system software changes?

    • Bouroux,

      It would be nice if there were an English version of this video ( so others who don’t speak French could understand it. Thank you!

      If this video reflects the thinking of the rest of the world about the Max, it is devastating! In short, it deviates from much of what has been said here. Like Philip (though for different reasons), it locates a central problem in the operation of the stabilizer, but it blames the antiquated jackscrew system. They plainly say that the system hasn’t changed since the original 737, and that the forces of the new design, MCAS, and quasi fly-by-wire systems ask the jackscrew to perform way beyond its capacities. The jackscrew system operating the stabilizer, say two engineers, would not be certified today. They also criticized the manual control wheels, and the idea of using human muscle power as a backup system. This would not be acceptable, or certified today as well. I think serious engineers, pilots, flight attendants, regular passengers and others interested in flight can see the obvious — though a lot of smoke tries to obscure it (to hide it in plain sight!). Boeing tried to soup up a 1967 airplane to rival the 21st century technology of its competitors. Brilliant and hard-working engineers did the best they could by making “rational” design decisions that would make this greedy and short-sighted corporate decision fly. And the 737 Max might, or probably will fly again. But to do so ethically and with an abundance of caution, and following some of the arguments laid down in this film, I think Boeing would have to bite the bullet, spend the money and follow the recommendations of Philip, though it may be for different reasons. The engineers in the film would probably agree that the Max at the very least needs to replace the grandfathered and archaic jackscrew and “upgraded to an all moving stabiliser with end to end fail safe redundancy that use[s] high speed/high precision servo hydraulic actuators.” This would be the responsible thing to do for the passengers who get on these planes. As the Brits say: full stop.

      • For the record, I have addressed the atequated electric motor and jack screw. It’s needs to be undated to servo hydraulic actuators. I’ve said it repeated. It doesn’t need FBW.

        To help you servo hydraulic actuators are at least 50 faster.

        • I believe you said the motor and jackscrew were from Walmart … for the record.

          • philip:

            You provide no link so its an unknown and as the jackscrew speed has been fine, its highly likely that an activation is going to be slowed down by software to avoid a pitch up.

            You should remember what being allowed to slam the rudder from stop to stop on the A300 (vertical tail snapped off)

            Classic case of PID statements to deal with too rapid a response and an out of control circuit.

  24. “” The Technical Advisory Board (TAB) was created shortly after the second crash, as a panel of government flight-safety experts for independently reviewing Boeing’s redesign of the MAX. It includes experts from the USAF, the Volpe National Transportation Systems Center, NASA and FAA. “The TAB is charged with evaluating Boeing and FAA efforts related to Boeing’s software update and its integration into the 737 Max flight control system. The TAB will identify issues where further investigation is required prior to FAA approval of the design change”, said the FAA. The TAB reviewed Boeing’s MCAS software update and system safety assessment. On November 8, the TAB presented its preliminary report to the FAA, finding that the MCAS design changes are compliant with the regulations and safe. “”

    The Madness Continues

  25. Those theories have been circulating for awhile. I’ve read before that the cause of the crashes was the jackscrew motor being overstressed and overheating, and no longer being able to move the jackscrew.

    I thought at the time that if those things were true, they would show up in the crash evidence and analysis. I have not seen that happen thus far.

    Also the claims about older technology may be true, but we haven’t seen a strong objection to the technology and operation from the pilots. I have read many comparisons of A320 to 737, pilots mention that the A320 is a newer design and nicer in many ways, but never in the context of safety.

    The manual trim wheels are something that the regulators are examining to see if they can provide enough control for the pilots, as a backup system. Again we have not heard complaints from the pilots, but possibly the wheels are very seldom used. It’s definitely worth examining.

    • Maybe related, but on the Découverte program (Radio Canada) recently linked previously (minute 18), two aero engineers comment on a trim slipping (glissement) occuring twice on the ET302 preliminary flight data with no auto or pilot input associated. They relate this with increased airspeed and pitch towards the end of the flight (pilot commanded elevator counter force could be aiding in this phenomenon, increasing stress). To be fair, the resolution of the data on the first claimed instance is inconclusive, the second instance is evident.

    • RE Trim wheel use- back in the dasy ( 1982 ) pilots were told to use the roller coaster or yo yo technique to reduce loads enoiugh to use the trim wheel – great if you had XY thousand AGL.

      But about 2 decades ago, that method was no longer mentioned in the manuals- and I doubt if since that time any pilots were ‘ trained’ in that method.

      Perhaps someone can correct me ??

      • Bubba:

        As I understand it that is correct.

        Add in the Simulator also went stupid on the fidelity of the manual trim and it was easy to turn ion all modes when it should be difficult or impossible

        1. High speed (you can unload if you have the altitude)

        2. With a seized motor (at which point the manual trim has to break loose the clutch and no answer for that one yet)

        • 5th Apr 2019, 14:44
          In pprune on tech log thread ( locked ) one can find this post on 5th april 2019 re trim wheel
          Boeing advice on “aerodynamically relieving airloads” using manual stabilizer trim


          Originally Posted by Takwis 
          SOMEWHERE on this great big site, buried within one of several threads now applicable to the two MAX crashes, SOMEONE made the statement that the trim wheel was made smaller, because of a “chokepoint” (or “choke point”) between the trim wheel mechanism and the new instrument panel/screens.
          Remembered that post because it was a reply to one of mine, the phrase used was “pinch point” – until we have HAL to do searches looking for choke point won’t find it 

          —I find they don’t always work that well – gets you near the post not on it – so the relevant text is (and it’s unattributed – copied from another forum):

          I agree it’s a flawed design. And I used to work there. I’m glad I don’t now.

          Regarding the trim wheels: When the NG was being introduced, I happened to be the Lead Engineer in charge of them and a whole lot of other stuff. There were some issues. The new display system created a pinch point between the dash and the wheel. We had to make the wheel smaller. And the new trim motor resulted in the wheel, which is directly connected to the stabilizer by a long cable, springing back when electric trim was used. It was an undamped mass on the end of a spring. We had to add a damper.
          Result: Depending on the flight conditions, the force to manually trim can be extremely high. We set up a test rig and a very fit female pilot could barely move it.
          As I said, I’m glad I’m no longer there. ”


          so what has changed ??

      • I believe that’s correct, the roller-coaster method was given to help unload the stabilizer when heavily loaded. I think in addition, the trim wheel diameter has been decreased slightly since then, which would further increase the turning force required. This is why they are looking at the trim wheels.

        There was also a case of the trim motor seizing, for which pilots were instructed to use high force on the wheels to break the motor clutch grip.

        • RE trim wheel – seems to be mentioned in Euro agancies but no mention in U.S. The slight change in diameter is IMHO not a real issue resulting in a slight increase maybe 10 percent in ‘ required ‘ force. Lets fram the problem a bit different.

          Assume that for whatever unplanned super remote low probability reason the new system – (or the systems on NG) go fubar and you are left with a few degrees AND trim.

          Ok- now lean down, unfold the handle, and with say 30 lbs force applied uniformly ( tangent to a fixed spot – handle ) on the wheel for say 10 turns per degree. How long by yourself per regs ( one average pilot ) will it take at say 2ookts ?

          For practice – ride your recumbent bike uphill using only one ‘ pedal’ crank – while having your knees about half way up your torso and being tightly strapped in ( since you are probably at maybe .5 or more negative G. )

          Are they going to incorporate that drill and related forces into your simulator session and then readout your net altitude loss until level flight ? You fail if over say 2000 ft loss?

          signed ‘ curious SLF ‘

          • I think the trim wheels are setup so both pilots can crank together if needed. Mentour Pilot has some good YouTube videos of manual trim training in a 737 simulator (although not the MAX).

            He also talks about the increased forces required with aerodynamic load. His videos give you a rough idea of what manual trim would be like. He also goes through the runaway trim memory items.

            One thing that struck me is how obvious and loud the wheel motion is. It would seem like un-commanded trim would be readily apparent. The wheels move in short bursts with STS and MT, but would spin 40 or more revolutions in a runaway.

            But as mentioned in the reports, if you have stick shaker and other alarms going off, you might easily be distracted, or view the wheel motion as a response rather than a cause. Human factors play a big role.

          • Rob, The trim wheel can also be ignored easily, as it is a ‘normal’ background sound in the cockpit, being heard on every flight. Pilots are used to the sound.
            Bubba, Maybe Boeing can incorporate a larger spring loaded handle? If the Frisbee disc is only 10 inches now, have a extended handle that pops out to maybe 15 inches in size to give the pilots more leverage when needed?

          • Rob and davenport- believe you both are missing two items-issues.

            1) Somewhere in the regs are the words ( paraphrased ) that ONE pilotmust be able to fly the plane undder all conditions

            2) A longer handle adds nothing to the ‘leverage’ needed. Consider your bicycle pedal- it ismounted at right angles to the chained wheel- a longer pedal adds nothing to the leverage needed except maybe for a bigger foot width.

            As to the leverage needed- try riding the bike with only one pedal- no inertia or coasting motion such as going uo hill. Try when the pedal is at the bottom applying say 30 pounds force with your foot directly backwards and then changing upwards( assume foot is strapped to pedal.

            Of course the ‘ system ‘ is now supposed to NOT move the stab more than ?? .6 degrees in AND, etc.

            Thus the argument that ‘ manual trim ‘ at any speed is ‘ easy ‘ when all else is fubar ?

            If so why are Euro types concerned about trim wheel force ??

          • Bubba, you have a valid point, no one is saying that manual trim is easy. Pilots train on it so they will be familiar with the feel and work required. One of the issues that surfaced was that the simulators may have underrepresented the trim wheel forces, which means the training was not accurate.

            My assertion of two pilots came from the training videos, where it was said that one pilot could help the other if needed. Possibly that is a violation of the design regulations requiring one pilot operation. However if you were struggling in flight, I don’t think you would hesitate to ask for help or use both pilots.

            Richard is right about the lengthening, it would be equivalent to extending the pedal crank length in your bicycle analogy. A longer crank means more torque at the hub which effectively increases the mechanical advantage (or gearing ratio). A question might be if there is clearance in the cockpit for rotation of a longer handle. I think Richard was throwing some ideas out there for consideration.

            When I saw it, I thought the trim wheel was an awkward motion too, the pilot has to lean forward and to the side, and then crank vigorously while also controlling the column. That may be another reason to use the second pilot, one can control while the other cranks.

          • I know you guys are not mehanics (I was a mechanc, thencia, engineer with letters)

            What you are missing is there is a slip clutch betweent he motor and the stabalizer.

            If he motor seizes (bearings) then you have to break the clutch loose to move the jack screw as the drive mechanism goes through the motor shaft)

            So that is an issue, its where the muscle thing kicks in.

            Also missed was that there used to be two motors and that went away in the NG. Not sure what the ramifications for the whole system are and if it would help in this case or not.

            It might move the manual trim to N3 not N2.

          • re rob …”Richard is right about the lengthening, it would be equivalent to extending the pedal crank length in your bicycle analogy”

            I read it as making the handle longer. Since the handle when used is at right angles to the ‘wheel ‘ there would be no change in mechanical leverage even IF there was room ( unless you could remove your leg 😉 ) I agree that in most – all cases- the other pilot would be asked to help. But even so, consider that majority of persons are right handed, and thus majority can apply more force in odd directions than their left hand. Again refer to cranking your bike up hill with one pedal. nearly impossible- that is why most- all bicycle cranks are different on both sides by 180 degrees. Since you sit a bit to the crank, you can push at the top a bit closer to the tangent of the wheel, or even stand on it.

          • Hey guys, there are reported cases where both pilots cranking could not move it.

            Throw in a less upper body strength mixed crew be it female /male – male/ female – or female/female and most people left arm is not as strong as the right (right handed)

            It needs to be sorted out and a solution implemented.

            If its the final backup it should take no more force than to crank it.

          • Interesting background on human factors dating back to WW2 – from a publication called wired and a article review. Should be titled- how much has boeing forgot re MCAS…

            ” How the Dumb Design of a WWII Plane Led to the Macintosh

            At first, pilots took the blame for crashes. The true cause, however, lay with the design. That lesson led us into our user-friendly age—but there’s peril to come.
            Illustration: Elena Lacey; Getty Images

            The B-17 Flying Fortress rolled off the drawing board and onto the runway in a mere 12 months, just in time to become the fearsome workhorse of the US Air Force during World War II. Its astounding toughness made pilots adore it: The B-17 could roar through angry squalls of shrapnel and bullets, emerging pockmarked but still airworthy. It was a symbol of American ingenuity, held aloft by four engines, bristling with a dozen machine guns.

            Imagine being a pilot of that mighty plane. You know your primary enemy—the Germans and Japanese in your gunsights. But you have another enemy that you can’t see, and it strikes at the most baffling times. Say you’re easing in for another routine landing. You reach down to deploy your landing gear. Suddenly, you hear the scream of metal tearing into the tarmac. You’re rag-dolling around the cockpit while your plane skitters across the runway. A thought flickers across your mind about the gunners below and the other crew: “Whatever has happened to them now, it’s my fault.” When your plane finally lurches to a halt, you wonder to yourself: “How on earth did my plane just crash when everything was going fine? What have I done?”

            For all the triumph of America’s new planes and tanks during World War II, a silent reaper stalked the battlefield: accidental deaths and mysterious crashes that no amount of training ever seemed to fix. And it wasn’t until the end of the war that the Air Force finally resolved to figure out what had happened.

            To do that, the Air Force called upon a young psychologist at the Aero Medical Laboratory at Wright-Patterson Air Force Base near Dayton, Ohio. Paul Fitts was a handsome man with a soft Tennessee drawl, analytically minded but with a shiny wave of Brylcreemed hair, Elvis-like, which projected a certain suave nonconformity. Decades later, he’d become known as one of the Air Force’s great minds, the person tasked with hardest, weirdest problems—such as figuring out why people saw UFOs.

            For now though, he was still trying to make his name with a newly minted PhD in experimental psychology. Having an advanced degree in psychology was still a novelty; with that novelty came a certain authority. Fitts was supposed to know how people think. But his true talent is to realize that he doesn’t.
            Cover of the book User Friendly

            Adapted from User Friendly: How the Hidden Rules of Design Are Changing the Way We Live, Work, and Play. Buy on Amazon. Courtesy of MCD

            When the thousands of reports about plane crashes landed on Fitts’s desk, he could have easily looked at them and concluded that they were all the pilot’s fault—that these fools should have never been flying at all. That conclusion would have been in keeping with the times. The original incident reports themselves would typically say “pilot error,” and for decades no more explanation was needed. This was, in fact, the cutting edge of psychology at the time. Because so many new draftees were flooding into the armed forces, psychologists had begun to devise aptitude tests that would find the perfect job for every soldier. If a plane crashed, the prevailing assumption was: That person should not have been flying the plane. Or perhaps they should have simply been better trained. It was their fault.

            But as Fitts pored over the Air Force’s crash data, he realized that if “accident prone” pilots really were the cause, there would be randomness in what went wrong in the cockpit. These kinds of people would get hung on anything they operated. It was in their nature to take risks, to let their minds wander while landing a plane. But Fitts didn’t see noise; he saw a pattern. And when he went to talk to the people involved about what actually happened, they told of how confused and terrified they’d been, how little they understood in the seconds when death seemed certain.

            The examples slid back and forth on a scale of tragedy to tragicomic: pilots who slammed their planes into the ground after misreading a dial; pilots who fell from the sky never knowing which direction was up; the pilots of B-17s who came in for smooth landings and yet somehow never deployed their landing gear. And others still, who got trapped in a maze of absurdity, like the one who, having jumped into a brand-new plane during a bombing raid by the Japanese, found the instruments completely rearranged. Sweaty with stress, unable to think of anything else to do, he simply ran the plane up and down the runway until the attack ended.

            Fitts’ data showed that during one 22-month period of the war, the Air Force reported an astounding 457 crashes just like the one in which our imaginary pilot hit the runway thinking everything was fine. But the culprit was maddeningly obvious for anyone with the patience to look. Fitts’ colleague Alfonse Chapanis did the looking. When he started investigating the airplanes themselves, talking to people about them, sitting in the cockpits, he also didn’t see evidence of poor training. He saw, instead, the impossibility of flying these planes at all. Instead of “pilot error,” he saw what he called, for the first time, “designer error.”

            The reason why all those pilots were crashing when their B-17s were easing into a landing was that the flaps and landing gear controls looked exactly the same. The pilots were simply reaching for the landing gear, thinking they were ready to land. And instead, they were pulling the wing flaps, slowing their descent, and driving their planes into the ground with the landing gear still tucked in. Chapanis came up with an ingenious solution: He created a system of distinctively shaped knobs and levers that made it easy to distinguish all the controls of the plane merely by feel, so that there’s no chance of confusion even if you’re flying in the dark.

            By law, that ingenious bit of design—known as shape coding—still governs landing gear and wing flaps in every airplane today. And the underlying idea is all around you: It’s why the buttons on your videogame controller are differently shaped, with subtle texture differences so you can tell which is which. It’s why the dials and knobs in your car are all slightly different, depending on what they do. And it’s the reason your virtual buttons on your smartphone adhere to a pattern language.

            But Chapanis and Fitts were proposing something deeper than a solution for airplane crashes. Faced with the prospect of soldiers losing their lives to poorly designed machinery, they invented a new paradigm for viewing human behavior. That paradigm lies behind the user-friendly world that we live in every day. They realized that it was absurd to train people to operate a machine and assume they would act perfectly under perfect conditions.

            Instead, designing better machines meant figuring how people acted without thinking, in the fog of everyday life, which might never be perfect. You couldn’t assume humans to be perfectly rational sponges for training. You had to take them as they were: distracted, confused, irrational under duress. Only by imagining them at their most limited could you design machines that wouldn’t fail them.

            This new paradigm took root slowly at first. But by 1984—four decades after Chapanis and Fitts conducted their first studies—Apple was touting a computer for the rest of us in one of its first print ads for the Macintosh: “On a particularly bright day in Cupertino, California, some particularly bright engineers had a particularly bright idea: Since computers are so smart, wouldn’t it make sense to teach computers about people, instead of teaching people about computers? So it was that those very engineers worked long days and nights and a few legal holidays, teaching silicon chips all about people. How they make mistakes and change their minds. How they refer to file folders and save old phone numbers. How they labor for their livelihoods, and doodle in their spare time.” (Emphasis mine.) And that easy-to-digest language molded the smartphones and seamless technology we live with today ” —–

  26. “” The Technical Advisory Board, a panel of government flight-safety experts, will identify issues where further investigation is required prior to FAA approval of the design change. The TAB reviewed Boeing’s MCAS software update and system safety assessment. On November 8, the TAB presented its preliminary report to the FAA, finding that the MCAS design changes are compliant with the regulations and safe. “”

    There must be really different forces and politics on top.

    Obviously MCAS is needed. How can it be safe if MCAS is off.
    I thought regulations were not meant to activate stabilizers for anti-stall purposes.
    Smells like undue pressure again.
    At least the FBI can watch the TAB live, or will they close their eyes.

    Is it still save for EASA experts in the US?

    • Leon:

      It would help if you acualy learned basic aerdynaims.

      MCAS acgivae (or should) only in the stall region.

      LCA (large commeial aircfat to you) do not operate in that region, ever.

      A pilot would only see that in a simlar (and only rcnely, most simlar were not programed to do a stall less han 2 years ago)

      From a pilots view, if you get a stick shaker you just lower the nose, end of stall.

      MCAS is a kind of cutesy thing that is not needed but was mandated by a reg concerning how much pitch up is allowed (and in this case the engine cause a bit more than allowed at stall).

      You did note, MCAS failure (TURN OFF THE STABILIZER MOTOR) ??????

      By definition if you can turn it off then its not needed.

      Like Philip you should by the new Microsoft Flight Simulator and go through the basic pilot training.

      Trust me, there are a lot more challenge things to deal with than a simple stall.

      Shame I can’g get you both up and do a spin. Now that is something that will suck the hair off your head.

      • TransWorld

        I didn’t kill anyone and I’m sure everyone with a right mind would have done a better job too.
        Interesting how this unfolds, it’s politics and culture.
        Nobody will accept Boeing’s certifications anymore. If Boeing has to follow the rules they are not competitive, they were not competitive even with cheating.
        Many people will avoid Boeing. How many are needed to put Boeing out of business.

        • Leon:

          That is getting hysterical.

          Boeing does not certify the aircraft, the FAA does.

          No argument there has been corruption of the process, that is part of what should be sorted out.

          No reality that people are afraid to fly Boeing’s, including the 737NG that still has the manual trim issue.

  27. Two Canadian engineers are suggesting that the jackscrew brake might have been slipping on Lion Air due to the increased horizontal stabilizer loads on the MAX.

    Here is a full report on the MAX grounding with a discussion of the jackscrew slippage in the second part.

    1- The first link is for the article associated with this video:

    2- The second link is for the video itself:

    3- The third link is a preview of the video on Twitter:

    It’s all in French but worth watching even if you don’t understand the language, for the article has some nice illustrations and in the video the animations are quite remarkable.

    • I ran this article through Google Translate. Here is a brief summary. Translate is not perfect but basically:

      The authors assert that they found a slip of about 0.2 degrees between stabilizer and jackscrew positions during the Lion Air flight. They acknowledge that this was not enough to contribute to the accident, but they feel there should be no slip at all, so they are concerned that the older technology used in the 737 does not meet current standards, as this was a basically new aircraft with little wear.

      They point to newer & improved technologies used on fly-by-wire aircraft that would monitor and maintain exact stabilizer position and prevent slip. They are concerned that the MAX is a lengthened aircraft using components originally designed for smaller versions dating back to 1967. They would like to know why Boeing has not updated their actuation technology. They don’t believe the grandfathered technology used in the 737 MAX is truly safe, if it can allow slip.

      They also are concerned that slip of this kind may have occurred and been worse on the Ethiopian flight, as it experienced much greater aerodynamic loads.

      Finally they said that Boeing did not answer their questions, but provided a statement that changes to stabilizer controls have been made so that similar accidents could not occur again.

      • will Boeing order inspections of the jackscrew on all 737s — NG and Max?

        • The French video should be a serious clue as to what international regulators are thinking, and what people who have inspected the wreckage of the flights think, and what serious academic, aeronautical engineers all over the world (who have puzzled, and puzzled through the design of the Max and tried to make sense of the information that Boeing has not provided) are thinking. The American century is over. The second decade of the 21st century is about to end. And I think for future generations, the poor 737 Max will be illustrative AND symbolic of something very significant that will be taught in engineering, business, ethics, computer science, and even creative writing programs for generations to come.

        • My understanding is that this is strictly a MAX problem and the NG should not be affected by this because the loads are not the same.

          • At what point should the clutch slip?

            We don’t know why its there nor what its settings are.

            My best guess is exactly for his so you don’t wipe out the motor.

            Its a basic part of he certification b and he loads are no different so we have to questions its use on NG, Classic as well.

            Does he FBW aircraft have the jack screw and moor clutch setup?

            More at issue is he manual trim have to break free the clutch to move the stab.

            Though if it goes wonky then in turn the manual trim should be easier with a motor issue.

    • Here’s a link to translate into English for those of us multi-language challenged.
      The article does question the jack-screw brake, that hasn’t been updated since the original 737 classic. How far can you push a type certification? If you double the thrust on the engines, should it require a new type certificate? The same airframe, flight control surfaces and systems, now having to deal with double the thrust? They don’t need to be looked at because of a grandfathering clause?

    • I thought I posted something about the Canadian video yesterday, but it disappeared. It’s shocking and it seems obvious to everyone but some Americans that Boeing executives made a deadly decision to try to compete with cutting-edge 21st century rivals by forcing its engineers to make “rational” software upgrades to an aging 20th century aircraft. It’s a technological embarrassment to Boeing and to the US that is hidden in plain sight. The quasi fly-by-wire fixes don’t work well on the antiquated system. The two engineers in the video draw attention to the jackscrew controlling the stabilizer, which they say never would have been certified today — it was grandfathered. And it may seem like a joke to some, but I feel crushing sadness imagining those Ethiopian pilots actually trying to turn those small, antiquated, frozen trim wheels desperately attempting to do their jobs and save 157 people with all of the force of their bodies. Again, according to the two engineers, this system would not be certified today — it was grandfathered. What modern aircraft includes human brawn in its backup safety system? I think those engineers would agree with Philip but for different reasons. The only abundantly cautious and safe engineering fix to the Max is not more software to drive an antiquated system beyond its capacity to deliver, but might be altering the control of the stabilizer so it can better respond to the various aerodynamic forces and demands placed on it by the Max with its larger, more powerful and differently placed engines. The 1960s vintage system doesn’t cut it. So Boeing should probably bite the bullet, take a painful hit to its profits, and do what’s right and what Philip recommends: “upgraded to an all moving stabilizer with end to end fail safe redundancy that use[s] high speed/high precision servo hydraulic actuators. ” And then, end the Max program and spend its resources, energy, brilliance and know-how on developing the most amazing airliner the world has ever seen — after, of course, they get rid of the current executives and their board and find someone who has real vision.

      • It helps to look at real-world evidence. The statistical rate of pitch trim incidents in the 737 is not excessive, over millions of hours of flight time. Pitch trim incidents occur for other aircraft as well. They have happened for the actuating systems being recommended here. Those are not immune to problems because they are newer.

        Grandfathered doesn’t mean there is no regulation or review, it just means the system is held to the standards in place when it was certified, plus any safety-related bulletins that have been issued since then. Like all other manufacturers, Boeing willingly incorporates those into newly manufactured planes, as well as those in service. Any problems formally identified for the stabilizer systems, have been addressed. It’s not like no one has looked at this for 50 years.

        I think that’s why this issue doesn’t get much traction. Saying that a newer technology is safer, is not the same as saying the older technology is unsafe. Changing things also involves some risk that new problems will be introduced.

        Also there is no evidence from Lion Air that the stabilizer actuation was a contributing factor. We have yet to see the Ethiopian report, that aircraft was well beyond maximum rated speed so it’s possible that a newer actuator or brake would have stood up better, we’ll have to see if that was a factor.

        Would it be better for Boeing to design & build a new plane from scratch? I think we’d all agree it would be. But it’s not a trivial request. New aircraft investments are immense. The 747 development nearly broke Boeing, as did the 787. Both were meant to be state-of-the-art aircraft that we all want to see.

        The truth is that the market readily accepted an update of the 737, and that is what ultimately drives those decisions. Now the cost of the MCAS failure is extremely high, and that will alter the market in the future, so Boeing will have to respond.

        A interesting lesson here is that, in the information age, the general population will not accept even the appearance of repetitive failure. That is a new thing, in the past multiple crashes due to similar problems were not uncommon, until they were understood, but it was left to the regulators. That is no longer true, everyone is feeling the heat, and the reputational and financial costs of that are very real.

  28. Returning to the word novel.

    The MAX has a novel [unique] use of the trim stabiliser. The MAX has a novel [unique] wing/engine mounting. The MAX will have a novel [unique] definition of duplex FCCs channels for fail-safe redundancy.

    I’m very interested in how or whether the regulators address these firsts.

    • The system on all 737s is not novel.

      Using the Stab as a tweak is not novel either, its done in speed trim that goes on continuously in flight.

      Pylons go back to the B-47.

      Engine mount locations vary all over the place due to designs and needs (in fact the 737 use to have quasi form in the wing engine mounts – almost British in the full in he wing) the Classic that put pylon on it was a vastly more huge change than the tweak on he MAX.

      Seizing on a single application within a whole framework of operation and then jumping up yelling he sky is falling just shows ignorance.

  29. Does anyone know if the AOA disagree light triggers when the MCAS deactivates because of of the 5.5 degree difference in the new ‘MCAS 2.0? The reason I ask is you could set up an Eastern Flight 401 type accident situation on 737-MAXs (I assume they’ve put in an FAR since then to synchronize the warning lights with the actual control cut out?) The only change I’ve found from the Flight 401, is that flashlights are now required near flight attendants.
    For example, if you know that AOA’s in the field can vary by 3 percent. You set the AOA disagree light to 5 degrees. You set the AOA disagree cutout of MCAS to 1 degree. You tune up your test aircraft so that the AOA sensors are calibrated to 0.5 degrees and fly the certification with a passing grade. MCAS passes certification, but, in real life is constantly disengaged, but, never shows on the warning light. Are warning lights synchronized to the actual system cutout signal of the system by regulation? Does the FAA ever retest planes after 5 years or 10 years to see if they can pass a flight certification test again?

  30. @RealSteve: So Boeing should probably bite the bullet, take a painful hit to its profits, and do what’s right and what Philip recommends: “upgraded to an all moving stabilizer with end to end fail safe redundancy that use[s] high speed/high precision servo hydraulic actuators. ”

    It’s obvious to me, and many others as well, that this would be the right thing to do. However, it wouldn’t be practical and the rules allow Boeing to get away with this for that very reason.

    Therefore the dilemma for the FAA and other agencies is to determine if the airplane is safe (as in safe enough) even though it does not meet modern safety requirements.

    What is made clear in the Radio-Canada video is that the MAX would not be certifiable today if it was a new model.

    • Fly by wire introduces a new type of fatigue in modern aircraft. What kind of toll does a quasi-fly-by-wire system like MCAS, coupled with the speed trim system, and all of the new and unexpected forces acting on the stabilizer take on the jackscrew?

      • It’s doesn’t require FBW. Mechanical wires work with what I’ve said. The Tristar worked, it was’nt FBW.

        It’s part of the obscuration. Specifically, we are being told, to upgrade the stabiliser requires FBW. No. The elevators already have servo hydraulic actuators and fail-safe redundancy. The same applies to the ailerons. The elevator and ailerons are not FBW. No.

        Obscuration. Rubbish.

        Still prefer a bigger stabiliser and elevators

  31. Boeing got away with it, but nearly 350 people in almost back to back accidents of newly built aircraft didn’t. The implications of this are beyond disturbing. It’s wrong. Again, the image of the Ethiopian Air pilots trying to solve a “runaway stabilizer” situation by trying to muscle those too small and ineffective trim wheels deeply troubles me. Well, as America’s most important philosopher has said: “There’s a sucker born every minute.”

  32. Rob,

    Whilst I understand your support for Boeing, your words are not realistic.

    To provide dual channel redundancy on the flight deck is at least 2 years, usual far more.

    It’s seat of your pants stuff. It’s winging it. Sorry, I don’t believe Boeing are that good.

    Let’s see what the regulators say.

    • I agree the regulators have the final word. But this change was instigated by the regulators, and they are participating in it. Boeing would not undertake a change of this magnitude unless they had indications from the regulators that it was an acceptable solution. It’s no longer Boeing working and overseeing themselves, as it was during original MCAS development. Regulators have participated in all of the flight testing.

      My support is for a good process rather than for Boeing. I’ve been critical of them when appropriate, but you can’t completely dismiss their considerable knowledge and expertise. It’s foolish to do that, it’s best to consider all sides to gain the benefits of an expansive and inclusive view.

    • Agreed, lets see.

      What happens when they agree its fine?

      Do we then get a no, no, no, its not fair they did not listen to me?

  33. Boeing have not been told by the regulators that they will accept their solutioon. That’s not allowed. Boeing have said they can do it with a software fix. The regulators have said let’s see it. The regulators cannot collude. Any suggestion of collusion means the courts.

    As of today, the regulators have not seen it regardless of Boeing’s public statements

    Any suggestion that Boeing are offering dual channel will cause the regulators to fall over laughing.

    • Wow. Cooperation with a regulator, and having an open process, is not collusion. Collusion would be if the regulator and manufacturer conspired to defraud the public. I realize you have some extreme views, Philip, but this one is really out there!

      First you say that Boeing’s process was closed and they hid things from the regulator. Now you say if Boeing shares and is open with the regulator, that is collusion.

      I think the truth is that Boeing and the FAA both have a strong incentive to arrive at a good solution. That does not mean the regulator will not thoroughly vet the system when done, or give it a rubber stamp. It just means that there is intermediate oversight along the way, together with the final testing and approval. If the final testing fails, the solution will be rejected, as it should be.

      That cooperation & oversight can only be a good thing, in my view. If that had happened with the original MCAS, we would have avoided a tragedy.

      • Agreed.

        Nothing wrong with working hand in hand.

        Its when one side pulls the wool over the others eyes.

        Years back we had an issue in a stairwell that the AHJ decided it had to be fire wall inviolate so they made them fire wall over my access hatches to the damper actuators that were part of the generator system (which in turn was part of life safety)

        All the AJH said was no, we needed to inspect and ensure the damper linkages and actuator themselves worked.

        Finally our Architect (he was great to work with) got a bolt out of the blue thought and he asked the AHJ , what if we put in fire prated hatches and ALARM them?

        That would be just fine said the AHJ. What we needed was help. I think it took us 3 years (fortunately I got in early when the original hatches were there and got the actuators checked out , adjusted and tightened the hardware up and none had failed)

        Just saying no is not a help to anyone.

        If the Generator could not run as they had no outlet, then other parts of the system for life safety were not going to work including the stairwell smoke flush system.

        • Let’s not confuse “the sky is falling” with “planes are falling out of the sky.” I don’t find radical but creative solutions pessimistic. Instead, I find fatalism, much more pessimistic. Saying it would be impractical for a company to build and make airplanes safer because of the cost is worse than pessimistic — it’s downright cynical. I find Philip’s observations, incredibly optimistic. They are a fix for what seemed like either an impossible problem or one that could not be repaired in a such a way that it would make me want to fly airlines other than than Jet Blue and Delta. He is saying: Do it right and spend the money as if doing a job right were more important than making money. And like most of the people on this site (with the exception of two or three well-paid trolls), he seemed to honestly think through the problem to the best of his ability.

          • You can’t attribute thinking to the process Philip is engaged in.

            Did you miss his statement about the Cockpit is useless?


            He throws a lot of verbiage out that is technically and proven bogus.

            He insists you cant blind fly an aircraft (or at least a 737MAX may be).

            Being uninformed is one thing but when proven wrong and not saying you were? That is political not technical .

            I think if you boil it down, nothing but a FBW aircraft is safe (I don’t know if its just Airbus or includes Boeing FBW)

            Massive numbers of non FBW aircraft are flown each day just fine as long as the pilot does their job.

            Rarely does a non FBW aircraft fail.

            A fair number of FBW aircraft have crashed so its not the panacea he makes it out to be, you still need a capable pilot.

      • IMHO the frustration is that Boeing is not sharing the information with the aviation enthusiast subset of the public. Which they are not going to do, so the frustration level will continue to grow amongst some commenters but will not be alleviated.

        • The bigger question is if they are now sharing the information within their own company, the FAA, other regulators, the Airlines, Pilots, etc.

        • The public will not fly with the MAX thinking Boeing is still cheating.
          Making a safe plane even safer. People know how that turned out.

    • Sorry for intruding, but I would hate to lose both of Robs and philip’s knowledgeable input due to some misunderstanding of either side. I have learned so much from both of you. Please take a moment to actually read what each of you are saying each other. Anyone can see the misunderstanding being raised in these last words. Please tone down the coffee, I will gladly offer you beer.

      • Words of wisdom, thank you Vasco. There’s room here for all opinions, and we want to be inclusive even when we disagree. Exclusion or dismissal serves no one.

        The collusion thing really hit me because it criticizes progress within the process we are all striving to improve, so I reacted. Will try to reign it in next time.

      • I noticed how the word collusion appeared and noticed how it could trigger a reaction. I have to agree that this is not the time for that word. It is much more serious and belongs to the courts. In a previous thread for a different reason I made a reference towards that care that must be taken. Also the technical side is what fascinates me, and I have to thank our hosts for facilitating these discussions.

  34. @TrasnWorld: “At what point should the clutch slip?”

    It is not allowed to slip within the normal flight envelope. In the Radio-Canada video this particular point is mentioned at 19:10 when they refer to FAA regulation 395A.

    The two engineers, both horizontal stabilizer specialists, have noticed that on Lion Air 610 the slippage was increasing with speed. In other words the faster the airplane was flying the more slippage there was. Which means the loads were exceeding the jackscrew braking power.

    • I read a description of the mechanism that said there is a clutch positioned for manual trim, such that the two jackscrew brakes can be overridden by motion of the trim wheels. The brakes are engaged/disengaged automatically for electric trim.

      The clutch system priority is:

      1. Manual trim
      2. Electric trim (pilot trim switches)
      3. Electric trim (autopilot and auto-trim systems)

      Thus manual trim can always override electric trim and pilot controls can always override automatic.

      So if the issue is un-commanded motion of the jackscrew and stabilizer together, then manual trim wheel motion might account for it. Or there might be some slip in that clutch when first engaged or disengaged.

      If the un-commanded motion was stabilizer but not jackscrew, then that is an obvious slip of the mechanism. Since it is a screw drive with double redundant threads, that seems unlikely.

      Not sure whether this accounts for the observed slip, I don’t know nearly enough about it. Just pointing it out as a possible explanation.

      • Boeing uses a recirculating ball ” nut” on the threaded jackscrew. As such it **can** be relatively easy to back drive that is for the nut under load to ” unwind ‘ . BUT proper design of the threads will prevent that. It is unclear if Boeing also uses a ‘ brake” arrangement to prevent such motion under load.

        for a description figure of the jackscrew – ” ball” nut arrangement.
        And search on ball screws will show details of effects of thread pitch, etc
        for example

        • Thank you Bubba, those were good links. You’re right about the ball screw/nut as opposed to ACME jackscrew, and the lower friction.

          I could not find an internal diagram of the gearbox, but various text descriptions indicate it has two rachet-type brakes, one for each direction of rotation, on the electric actuator shaft. The corresponding brake is disengaged to permit one-directional rotation when either ANU or AND clutches are commanded.

          The manual trim wheels are on a separate higher-geared shaft that bypasses the braked electric actuator shaft. The shafts are geared together in a differential or planetary arrangement so that the trim wheels are always engaged. Or at least, that was the best I could determine from what I could find. There may be more accurate sources.

          Philip is also right about the superiority of hydraulic actuators and servos. The 737 has a long-standing mechanical design that has worked well over the years, so has remained. It’s had some FAA actions over the years, so the MAX must comply with those.

          • Thank you Richard, that diagram is for an earlier version of the drive, but it does show the basic components and is very useful.

            It shows the differential gearing, and the clutch for manual trim being inside the drum itself. I had wondered how the manual clutch fit into the scheme of things, I couldn’t find it described anywhere

            I believe the current design has separate ANU/AND clutches for the trim motor and also the brakes are now racheting rather than friction or V-brakes. But this diagram gives the general idea of how it works.

  35. @Rob: “So if the issue is un-commanded motion of the jackscrew and stabilizer together, then manual trim wheel motion might account for it. Or there might be some slip in that clutch when first engaged or disengaged.”

    They were talking of uncommanded motion of the jackscrew itself. In the case of Lion Air the amplitude of the motion was increasing with speed and was occurring at speeds far in excess of a possible trim wheel activation.

    See the Radio-Canada video at 18:45 for a complete explanation:

    • Thanks Normand. I don’t speak French so may have missed some things from the video. I enlarged a screenshot of the chart that was presented. I saw it was for the Ethiopian flight, not Lion Air as I previously thought.

      The chart shows an offset developing between the expected and actual stabilizer positions, at 4 occasions during the flight. The offset seems to max out consistently at 0.2 degrees, except for the 3rd instance, when there was a 0.3 degree offset.

      I could not tell from the chart if they are saying the errors were accumulating, so as to create a larger total error. It did not appear so.

      Also I noticed that the direction of the offset was the same for 3 of the instances, the expected position was a larger deflection than the actual position. In the fourth instance this reversed, and that was at the highest speed before the crash.

      There is no trace for manual trim wheel inputs, but maybe that can be deduced from the voice recorder. I don’t know how to interpret this data any better, so will accept their conclusions. It’s clear that some form of slip was occurring, on the order of 5% to 7% of the 4.2 degree AND stabilizer deflection, or 2% of the total 17 degree stabilizer range, with the tendency to lessen the deflection.

  36. Rob,

    Other way round. As I have made clear there is no historical precedent for these “novel” solutions, as follows:

    1) A wing/engine mounting with no pylon, causing

    2) The use of a trim stabiliser as an elevator, causing

    3) A non duplex FCC to be redefined as a duplex FCC when it isn’t.

    We can all redefine the meaning or purpose of something to pass through regulations. That’s why we have the courts.

    With regard to collusion, by law the regulators cannot say something works until they have seen it work. Equally they cannot say to an OEM that will work. The regulator must have an open mind.

    So the regulators must allow Boeing to do it their way, but not agree their way is right until they have seen it’s right. Anything else is collusion. Again, that’s why we have the courts.

    I do think it will go to the courts for both reasons stated in this post: 1) meaning/purpose is being redefined and 2) the FAA (but not other regulators) do have a prior agreement with Boeing.

    The only person who is being extreme is you. You do need to offer precedent for all of this. You can’t. So you revert to abuse. I’ll leave it at that.

    If you don’t know the meaning of precedent. It means established practice or an established solution. That will be the first question of a court. As this been done before and repeatedly done before?


    • When they say “novel”, I wonder if this means this is the first time use on the stab has been enlisted this way on a commercial aircraft? That during the two stall tests, the stab moves during the sequence of the test of pitching the airplane up from a steady angle to stall AOA, instead of just the elevator.

      Is it against the rules? No
      Is it in the spirit of the intent of the test? A gray area, hence the use of the word “novel”?

      Did they ever perform these two stall tests with MCAS 1.0? What did the time versus AOA, time versus pitch, time versus elevator deflection, and time versus stick force look like on that or MCAS 2.0?

      It may pass the test of requiring a decent amount of elevator and stick force in the final angles, but I bet it is hardly linear with time. It’s probably an ugly fix, in terms of having a smooth behavior, but meets the letter of the law.

      • Ted,

        I find your spirit of the law (or regulations) verses the letter of the law argument compelling and appropriate. In some ways this seems to define the whole Max project and the issue of grandfathering. It treats regulations like unreasonably strict parents, not reasonable safety protections. This is not a good attitude to have when updating an airplane.

  37. Philip, you have referred to my posts, and others you have disagreed with, as rubbish. Garbage, refuse, having no value. To me, that is abuse. Yet I have overlooked this and tried to further explain my positions in terms of evidence and facts.

    I referred to your assertion, that cooperation between the FAA and Boeing represented collusion, as an extreme view. I don’t believe that was abuse, I was truly taken aback by your comments. They seemed to undermine the process improvements we’ve been hoping to see. However if you felt that it was abuse, I apologize. It’s not my intention to abuse anyone or be dismissive of any other point of view.

    To Vasco, I posted a response but it vanished. Thank you for your words of wisdom. You are right, there is room here for all viewpoints, and we want to be inclusive. I should have taken a breath before diving in.. Sometimes the old Scotsman gets the better of me, a family failing I’m afraid. I light up when I should calm down, especially when there’s an issue of truth or fairness. I’ll try to reign things in for the future.

    • You are disregarding the evidence in the JATR report and the Lion Air crash report.

      To give an example, both reports say the MAX has a pitch up tendency. That means it isn’t naturally stable. To simply just tell me I’m wrong needs explanation as opposed to just telling me I’m wrong. If it’s naturally stable provide the evidence.

      Another example, both you and Mike keep telling me the use of the trim stabiliser is typical. JATR says it’s novel. If it’s typical provide the evidence.

      Otherwise I do have the right to say rubbish. The evidence is in the reports.

      Address the evidence, don’t ignore it.

      • Philip,

        To give an example, both reports say the MAX has a pitch up tendency. That means it isn’t naturally stable.


        From the KNKT report pg. 246.

        During the preliminary design stage of the Boeing 737-8 (MAX), Boeing tests and analysis revealed that the addition of the LEAP-1B engine and associated nacelle changes produced an aircraft nose-up pitching moment when the aircraft was operating at high angles of attack (AOA) and mid Mach numbers. This nose-up pitching moment was deemed likely to affect the stick force per g (FS/g) characteristics required by FAR 25.255 and the controllability and maneuverability requirements of FAR 25.143(f).

        A nose-up pitching moment is not the same as a “pitch up tendency”. The nose-up pitching moment is just one of several pitching moments acting around the aircraft CoG simultaneously at any given time during flight. It can exist without the aircraft pitching up because the moments can be balanced.

        A “pitch up tendency” implies that the aircraft pitch tends toward ANU without pilot. The KNKT report plainly does not say this. I’ll recheck the JTAR report as well, but I’m positive it will concur with KNKT.

        • To clarify, when I say nose-up pitching moment, I mean the nose-up pitching moment contribution of the engines.

        • Bad hunch because both JATR and the KNKT reports do mention “pitch-up tendency” as philip puts it, but the reports use the expressions interchangeably:

          KNKT p246 (same paragraph of your extract)

          Boeing implemented aerodynamic changes as well as a stability augmentation function called the Maneuvering Characteristics Augmentation System (MCAS), as an extension of the existing Speed Trim System (STS), to improve aircraft handling characteristics and decrease pitch-up tendency at elevated angles of attack.


          to address potentially unacceptable nose-up pitching moment at high angles of attack at high airspeeds

          JATR p2 (before index)

          (…) and then modifying the software for a different scenario – in this case reducing the pitch-up tendency at higher angles of attack at low speeds.


          Observation O3.4-B: (…)The possibility of a pitch-up tendency during approach to stall was identified for the flaps-up configuration prior to the implementation of MCAS.

          The distinction you make – depending on other concurrent pitch moments balancing out – is expressed in the report in terms of “possibility”, “potentially”, “likely”, and do originate in Boeing’s own attempts at compliance with certification. It has been said, we don’t have access to Boeing’s data, only the latter may quantify objectively how likely; how possible; how potential.

          Boeing would not exhaust resources unnecessarily developing MCAS, so even without data it is a totally valid assumption for the outside observer that it represents an actual and likely possibility, that pitch moments are not always balanced, and when not, it is nose-up.

          This unrelenteless scrutiny is warranted given the import of the matter, and if Boeing wants to keep its cards close to their chest, given the circumstances of the lethal accidents, the public/consumer is not only prone but also entitled to assume the worse, at least, until the mistake can be considered honest by allowing an objective peer review, which so far they are not.

          I am convinced, not even JATR members had access to the relevant data, having reserved their final assessment pending future tests, or somehow agreed to keep it from the report, in which case they would be failing the respective public they each represent too.

          • Vasco,

            I guess I was wrong about the expression “pitch-up tendency” not being used in the KNKT and JATR reports. Thanks for pointing that out. Truly.

            I really don’t like it when technical reports use imprecise language. It causes misconceptions. I expect this from news reports, but not technical reports.

            However as you pointed out, the KNKT report uses the expressions interchangeably. Everyone has a tendency to apply their own bias when they read something. I’m no different. So when I read the KNKT report I read it with my bias which is based on my knowledge and experience in this subject matter. Again, everyone does this, including Philip, you, and of course, me.

            My reading of the reports is that while they use the expression “pitch-up tendency”, I don’t think they mean that the MAX will suddenly pitch up, uncommanded by the pilot, at some high AoA like Philip claims. Based on my own knowledge and experience, my reasons for reading it this way follow:

            1) Before either report came out people were long claiming that the MAX is not naturally stable. I asked myself, if this is true then how would this manifest itself? A naturally unstable aircraft without any artificial stability from a control system would exhibit a distinctly awful handling characteristic; control column reversal in both the ANU and AND directions. LNA commenters have never addressed the spontaneous pitch-down tendencies that have to be present in a naturally unstable aircraft. To my knowledge the MAX does not display any of these handling issues as there have been no reports of these happening.

            2) It is impossible for a system like MCAS as it was implemented, or as it is going to be implemented, to stabilize a naturally unstable aircraft. This can only be done with a closed loop full authority control system. MCAS is not even close to this. First, it is an open loop control law. Second the stabilizer cannot be driven nearly fast enough. Third, it does not have full authority as evidenced by the pilot’s ability to fight it, albeit unsuccessfully in two of three cases. If MCAS is not a closed loop full authority flight control system, then MCAS can only be used to intentionally mistrim the aircraft under very limited high AoA conditions to compensate for an uneven column force per g profile. This tracks with how the KNKT report describes it.

            3) During the several flights before LNI043 (LNI043 was the flight before ill fated LNI610) the aircraft suffered the intermittent loss of the left side AoA vane reading. These intermittent signal losses got progressively worse until the flight right before LNI043 where a significant portion of the flight was flown without the left side AoA signal. This means MCAS was essentially switched off for a significant portion of this flight. No control issues were reported. If the MAX truly lacked natural stability and MCAS was being used how Philip claims it is, then there definitely would have been control issues.

            So, for these reasons I cannot read the words of the KNKT and JTAR reports the way Philip reads them. To do so would fly in the face of what I know to be true about aircraft stability and control, what experts like Bjorn say about aircraft stability and control, all the data I know from news reports about the flights involving MCAS failures, and every detailed description of MCAS that I’ve ever been able to find.

          • Thank you for clarifying, Mike. I agree with your analysis and comments about precision.

            I think we all fall into the terminology trap at times. These are complex issues with many ways to describe and think about them. It’s easy to be imprecise because in your mind, you know exactly what you are saying, but others reading your words may not. Precision definitely helps to address that.

      • Philip, I have tried to address the evidence, numerous times. I have never simply told you that you’re wrong, I’ve tried to engage you in discussion by presenting my reasoning based on the evidence I have available.

        If you disagree that’s fine, you’re entitled to form your own opinions and state them here. It’s best not to express opinions as fact, because then others will feel compelled to speak up. I’m not singling you out, I’m sure I’m guilty of that as well.

        Taking the JATR report as an example, it says that the use of stabilizer is novel in the context of stall prevention. I’ve agreed with that in many posts, have explained their reasoning and defended it.

        But I also gave the other side, as repeated by Bjorn and Mike, that Boeing did not conceive MCAS in the context of stall prevention, but rather as handling characteristic augmentation. The evidence for this is in the reports of the early high-speed flight testing, where the test pilot reports that MCAS had adjusted the handling well, it felt much better to him. Note that he was not testing for stall, just behavior at high AoA.

        I personally believe that to be the more correct interpretation of what actually happened. MCAS came about because testing showed it was needed to meet regulations for handling at high AoA. It was not meant to take control of the aircraft from the pilot, as a stick pusher or other stall prevention would. The fact that it behaved that way in the accidents, was due to errors in input, design, and implementation

        Since stall also occurs at high A0A, and MCAS was meant to help the pilot avoid that area of the flight envelope, it’s valid to say that maybe this is getting into a grey area, where MCAS is coming close to being active at stall.

        With this in mind, the recommendation of JATR was that Boeing could have written an issue paper to have the matter clarified, and in so doing might have found the flaws in MCAS. I have agreed with that finding, in several posts. It would have been the abundantly cautious thing to do.

        I realize you probably don’t agree with this view, but it’s the most balanced and fair conclusion that I can come to, given all the evidence I have. It includes findings of the reports as well as statements by Boeing and expert opinion. It doesn’t absolve Boeing, nor does it trash them. It just attempts a clearer and better understanding of what went wrong.

        • Rob, There was MCAS 0.5 and MCAS 1.0. The first was after test pilot Ray Craig was noticing the handling issues at high speeds. That was in the 2012 time frame, I believe in a simulator. In 2016, test pilot Ed Wilson was flying some low speed stall testing, and that’s when the dangerous changes were snuck into MCAS. MCAS 0.5 would probably be OK, as it was lower in authority, and slower, designed as more of a augmentation system. But, MCAS 1.0 was the low speed extension where the increased the trim authority and response rate of the system were added without proper oversight. The new MCAS 2.0 seems to be unchanging back to the original MCAS 0.5 version with more safeguards. This article points out the changes. I really hope a good book with chapter and verse comes out about the MAX/MCAS crisis.


          • Richard, I totally agree with that view. The risk associated with MCAS increased substantially from 0.5 to 1.0, but the safeguards went the other way. The g-force sensor requirement was removed. leaving one AoA vane as the sole activation sensor.

            Boeing informed the FAA of this change as required, but not in a way that drew attention to the increased risk.

            This is why cooperation and participation of the regulator is key, they need to be aware of what’s on the drawing table and monitor stages of development. Not interfering or telling the manufacturer what to do, but making them aware of what is expected and the level of scrutiny that will be applied at the end.

            In most industries, this is done by writing an RFP and then evaluating proposals by those specifications. Aircraft manufactures are different in that they are in a leadership position, they are writing their own specs, based on what they think the market wants.

            That’s why we need regulation, to have an independent view of those specs, and a hand in the design and development process, so we don’t get to the end and find a huge problem.

          • Rob:

            I think that is pretty well put. Philip of course will not get it because he does not want to, it does not fit his agenda.

          • Rob, The safety analysis aspect, as to who or how the decision is made on how dangerous this is to the plane. (minor / major / catastrophic / hazardous etc), seems to be a grey area. On one had it’s a judgment call, but, it has to be based on the known past defect rates, and failure modes. MCAS 0.5 woke the pilots up, but, wasn’t anything that pilots hadn’t been dealing with (bad AOA values), successfully in the past. MCAS 1.0 grabbed the plane away from them, using the same control surfaces, too fast for them to deal with. Also, the on/off cycle of MCAS 1.0, had to lead to confusion in the cockpit. I’m trying to think of any other system that cycles like that on an aircraft. Maybe the ice boots?

        • Rob your opinion, not mine and others. Don’t say people are wrong on your opinion. That’s Mike’s world.

          When you look at my posts I leave it open if I can. The most obvious example is the CoG versus CoP. I know Bjorn is responsible. He won’t like that. I gave him a year to correct himself. He hasn’t. That’s on him.

          I do my best to keep the debate open. After all Boeing still won’t make public when MCAS engages. We are all guessing.

          Don’t tell me or anyone else they are wrong when you don’t know.

          • Philip: No, you are totally closed minded and ignore people who really do understand the material.

            Somehow I think the only thing you do for Bjorn is give him a good laugh. I have a mental picture of an angry ant lecturing an Elephant.

            Good luck with that of course. You don’t even register in the Elephant world.

            Philip: You are plane (pun intended) so far off base you arn’t even playing in the same stadium as the rest of the world is.

            OCD on when the MCAS kicks in, all you need to understand is its had high AOA at stall (slightly before if its going to smooth the pilots controls the way its supposed to.

          • Philip,

            Perhaps you should live by your own words. When you say things like:
            “one day you will get there, but not today”
            “keep offering the gibberish”
            “doesn’t understand engineering principles”
            “JATR used the word “novel”, I would say “stupid”. Not surprisingly they are getting it wrong. The same words applies to Bjorn as to you.
            you are saying someone is wrong.

            If you can’t take the heat, get out of the kitchen.

        • “” The fact that it behaved that way in the accidents, was due to errors in input, design, and implementation “”

          Rob, that were not errors, safety wasn’t their intention and not in their calculation.
          Boeing is not a normal company, they only care about profits, they rushed everything to reach EIS as fast as possible. That’s why they pressured engineers not to find problems and forced them to keep their mouths shut, kept them in small groups with little information.
          They didn’t care about safety, it’s cheap to pay victims. They didn’t want to put much time and effort into safety, finding ways to classify hazards “minor”, finding ways to get excemptions from regulations and keeping them secret, even cheating, changed what was certified without new certificaton. This culture was supported by politicians over many years.
          This is not a normal company, their calculation is that profits are more important than some deaths are cheap if they even have to pay something, mostly they have to pay nothing, only blaming pilots.

          Some posters here act as if Boeing is a “normal” company. Nobody would care much if there were mistakes made unintentional, like the company in Florida with the AoA sensor.
          The big difference is Boeing doesn’t care about mistakes, only care about profits which is supported by politics.
          That’s why Boeing is in the headlines and FAA too.

  38. If MCAS 1.0 was designed for slow speeds, at high speeds it must have been four times overdesigned, meaning that to stay in the turn, it requires a huge amount of backforce all of a sudden.

    Kind of like if there was automated turning on a car with the same design for 15mph or 60mph. Physics is saying that is not an optimum solution.

    • Ted, MCAS 0.5 was designed for high speed first. The low speed case extended the MCAS 1.0 authority by a factor of 4 from the high speed case. So the original high-speed case was probably ok.

      You’re right about the impact of speed, at low speed MCAS needs more authority because less aerodynamic force is available. At high speeds, less authority because more force is available..

    • Ted:

      You con’t get it or understand stalls.

      They can occur at any speed and MCAS was active across the range with varying amounts of authority.

    • Ted, From what I understand, there is a high speed and a low speed setting on the speed trim system, of which MCAS borrowed. If the plane was flying at a high Mach speed, then the lower speed setting of the STS was used, and at a low Mach speed, then the higher speed setting of STS was used. The dangerous portion of MCAS was at the low speed stall Mach speed, so it was using the high speed STS setting after the ‘extension’ of the STS system to be a near stall prevention. If they hadn’t put in the last minute change, the two accidents probably wouldn’t have happened. Although, the plane would still experience speed warnings etc. but, the nose trim wouldn’t have gone down so quickly. It really does lead one to ponder how someone cobbled together this last minute change. They really had blinders on, to focus so closely on solving the one issue, without asking how it effected everything else.

      • TW and Richard,

        Are you saying that MCAS had two inputs? AOA and speed?
        And it had two different settings based on those two different inputs?

        I assumed it was a binary on/off based on one input, some AOA number. And, the rate was jacked up for the worst case, the low speed stall test, but that rate was then used by default on the high speed turn.

        • Ted,
          I wondered the same thing a while back ..

          There is a high speed part of MCAS, which seems to be working OK, but, who knows without details. Then there is the last minute change to MCAS that added an ‘extension’ to MCAS in low speed stall situations. That’s the one that folks are focused on in terms of things going wrong. In the Air Data computer, that feeds the Flight Data computer, I think the speed is corrected by AOA. That’s why, with the AOA signal messed up, the speed also was messed up giving all the strange warnings and alarms.

          • So in the two crashes, which one of the two speeds of MCAS was operating in each?

            Since ET was going pretty fast, I’m guessing only the .06 units per second rate?

  39. Ok. Let’s say Muilenberg steps up to the plate and says: ‘This whole fiasco came about when Boeing was controlled by MBAs and not engineers. Boeing’s an engineering company. I’m an engineer and I’m in charge. In order to make the Max the aircraft it needs to be — something we can be proud of — and return Boeing to its blue chip status, we are going to have to do something drastic, expensive and right. Something we should have done in the first place! We need to re-engineer the tail.’

    I’ve been trying to imagine what giving the 737 Max an L1011 Tristar (not a much younger plane than the 737) stabilizer would look like. I’ve been googling around for images. What does the Tristar stabilizer look like on the inside? Is there room inside the Max stabilizer to replace the jackscrew with “high speed/high precision servo hydraulic actuators”? I’m also trying to imagine the servo hydraulic actuators already on the elevators. I haven’t found images of them. Can this be done? Or at least, can it be imagined?

    • Nothing wrong with the tail so why would you suggest change it?

      Unless you don’t understand the aspects of what goes on and is going on.

      You may well need to make changes to manual trim and or add a stab motor back but changing the tail?

      Frankly that is nuts. Gut the whole bird and change the whole control system to FBW?

      The MAX was done so they did not have to do a new SA.

      • Vasco. as I understand it the BBJ is a strengthened frame because it trades-off normal passenger and cargo weight for fuel & range. It has to be able to land with that fuel load on short hops, but also have the range to fly 5,000 miles or more. Wings and landing gear are strengthened as well, so my guess is the tail was strengthened too, and probably enlarged for that purpose, and the different weight distribution.

        I have not seen the tail structure of the MAX raised as an issue in any of the reports. A larger tail would give greater pitch control authority, but would not alter the basic problem MCAS was intended to address, which was the shift forward of the CoL at high A0A. So I think you still would end up with some kind of augmentation needed.

        The regulations require specific handling characteristics at high AoA, so as to help the pilot and correctly represent flying conditions at the approach to the edge of the flight envelope. They are designed to discourage the pilot from pulling back too sharply on the control column and stalling the aircraft, by intentionally increasing the column force at high AoA. That is what MCAS was meant to do.

        If the larger tail provides greater authority, it might actually cause the opposite effect, it might reduce the control column forces at the approach to stall, because the pilot has greater authority margin available. The regulations would not permit this, regardless of the size of the tail. Again the intent is to discourage the pilot from using that authority to enter a stall.

        I know those statements will be unpopular, so I apologize if anyone is offended. I’m offering an alternative viewpoint and am not telling others they are wrong, just asking them to think about it in a different way.

        • Rob, I can’t find the source right now, but, I remember seeing that the 737-MAX had stiffened the elevator, and thickened the stabilizer. I assume because of the larger engines.

      • Is that drawing right? The 800 BBj versus the 800, the horiz stab and vert stab are very different?

        I thought they were same? What’s up? Is this a Boeing drawing?

      • Ted, these are official drawings down to scale, with an accuracy of 6inches pre overlay editions, of the three models, originally in industry standard .dwg format (shared in common vectorial .svg). These are publicly accessible at the Airport Compatibility CAD page. I apologise, I had sourced it before in a previous comment to RealSteve, when the CoG issue was raised (estimated/generic, drawn in black), I had shared them in order to obtain a higher accuracy about the differences between 800 and MAX8.

        Mike, yes, unfortunately, not all models include inner contours to be able to show the stabilizer/elevator individual areas, even in the Airplane Characteristics for Airport Planning model PDFs which have additional schemes (ie Ground Servicing, etc), I was not able to find those contours to transfer on. I am sorry but I have no treat for us.

        Richard, Boeing states these CAD drawings have an accuracy of 6 inches, I did my best not to interfere with that when overlaying (scale and translation), so if the differences you refer are over that accuracy they should show. I don’t see any between 800 and MAX8.


        A larger tail would give greater pitch control authority, but would not alter the basic problem MCAS was intended to address, which was the shift forward of the CoL at high A0A.

        This is undoubtedly where there appears to be different opinions. I am no aerodynamics engineer, so I can’t objectively tell. But as an observer I can understand the logic underlying this point of contention. In addition Boeing is the manufacturer, the decision is absolutely theirs, only constrained by the certification process.

        I may be wrong, but while on the drawing board of MAX developement when the pitch-up tendency was first identified, these options must have surfaced, and a decision was taken in favour of MCAS. This BBJ2 empennage only shows that there must have been a development head start against MCAS, which nevertheless, along with additional criteria must have not passed the cut: an increased drag would reduce the gains of the re-engine effort.

        My conclusion: if an increased stabilizer constitutes a solution to the initial problem, as some expert opinions attest, Boeing had the solution at hand with reduced engineering implementation effort (BBJ2 tail on same airframe), with the added bonus that the constituent parts were already certified 20 years ago. Which reinforces the conclusion MCAS was a high management type decision, not a mere stumbling upon an unfortunate solution type of thing.

        • Vasco, I can’t offer anything new in response to this. It depends on two contentions:

          1. a larger stabilizer/elevator/tail was an alternative solution to MCAS for the MAX handling characteristics problem.

          2. the modification of the MAX tail structure was a realistic and justifiable engineering task. considering all the other consequential design changes that would accompany it.

          I don’t have the knowledge to evaluate these, beyond what I have already posted, so can only really say they are alternative viewpoints. I’d be interested in getting Boeing’s perspective on these, or someone qualified with experience in commercial aircraft design and construction. I obviously am not that person.

        • …and are not required to, Rob, I think we can handle the remaining doubts persisting, I do feel privileged just engaging so much experience and knowledge found here, so much so that I felt the need to offer back my best in return.

          Out of curiosity, let me tell you my first approach to this whole thing originated in my interest on user interface. Why, oh, why, would that thing deny the user multiple input repeatedly to the point of no return? What kind of absurd automation is this? What do specialist press say? Here I am now waist deep in aeronautics (which is both a professional as much as an entertainment interest).

          This thread is humongous, I’ve almost lost track of all the points of interest raised.

          Fehrm, Lemme, Gates, many commenters here will keep us in the loop.

  40. “To give an example, both reports say the MAX has a pitch up tendency. That means it isn’t naturally stable. To simply just tell me I’m wrong needs explanation as opposed to just telling me I’m wrong. If it’s naturally stable provide the evidence.”

    Philip: You really need to get a dictionary and look up the word tendency.

    Much like over-steer or under-steer by a car in a corner

    Just because a car under-steers does not mean you are going over the cliff at Dead Mans Curve (look up Beach Boys and the song Dead Man’s Curve)

    Its a tendency that is easily adjusted for.

    They try to make it as neutral as possible for (dang, smoothing purposes just like flying an airplane)

    Unlike your average driver a pilot is better trained.

    We don’t panic at a bit of handling not perfect at extremes, that is why we are pilots and not baggage handlers.

    When our water heater broke my wife did not run out of the house yelling ans screaming, its the end, its the end.

    She turned off the main water supply just the way I showed her.

    You tell me a pilot can’t fly an airplane on a moonless night in a storm! When in fact blind instruments came out in the 30s to do just that.

    Being uninformed and wanting to learn is one thing, being ignorant and refusing to learn is a whole different aspect.

  41. This may have just hit the all time high posting

    309 or so.

    I want to thank Rob for some very well presented responses.

  42. What is the total duration in seconds of the low speed stall test and the high speed stall test, from in trim no elevator, to final AOA? How does MCAS and its 9 seconds overlay onto those durations?

    • Ted, We don’t know the details of the changes to MCAS 1.0, even now. At some ‘high value of AOA’, MCAS kicks in for approx 10 seconds, cycles off for 5 seconds, rinse repeat, until the AOA lowers to another safer AOA. If it’s at the high or low speed trim setting, I’m not sure. We do know it’s at the high speed trim setting during a low speed stall. If this changes at a higher Airspeed, I’m not sure. If you’re researching this, you can get deep rather quickly. For example

      • I believe it has lower authority at the higher speeds as the stab is more effective and less would be needed.

        I don’t have the link for that.

    • Ted, it’s important to note there were not stall tests associated with MCAS. There were tests of aircraft handling at high AoA, but before stall.

      Also 9 seconds was the maximum cycle time before reset of MCAS. In a true as-intended MCAS activation event (handling not stall), MCAS would be active only until the nose came back down, probably a few seconds at most.

      MCAS only ran in 9 second cycles during the accidents because of the failed AoA sensor, that told it the attitude had not changed, even though that was not true.

      Not defending the MCAS poor implementation or lack of safeguards here, just pointing out how it was supposed to work, and how it would have worked, if it had been implemented well & correctly.

      • I never thought about that, that it would shut off before full duration if the critical AOA switched it off as well as on.

      • What is the official name of the two tests? and
        ‘min speed approach to stall test’?
        (accomplished by level flight light load)
        ‘max speed approach to stall test’?
        (accomplished by high G, heavy load)

        What’s the worst (limiting) case condition for pitch stability for the stab in this test? Max aft load or max fore load? I’m guessing max fore load, since the stab would be more nose up, opposite of the function of MCAS.

        • Ted, I think the tests would be with increasing A0A values above 12 degrees, but below the critical AoA angle of stall, at varying airspeeds, with a clean wing (no flaps or lift devices deployed). Initially the handling problem for those conditions was found at high-speeds, eventually a similar problem was found at low speeds as well.

          Since MCAS is designed to counter the effect of CoL moving forward toward CoG at high A0A, I would think the worst case would be aft-most CoG, as that creates the quickest onset of handling issues.

    • In short: We don’t know.

      In detail:
      While MCAS 1.0 rates are fixed at 0.27deg/s-flaps extended/down and 0.09deg/s-flaps up, it is triggered at unknown AoA (guesses at ~10deg), duration/magnitude varies as a function mainly of Mach and AoA (but also Air/Ground, Flap position, Pitch rate, True Airspeed).

      If elevated AoA condition persists (angle unk.) it resets bellow a threshold(angle unk.) or upon use of electric stabilizer trim switches, after a 5 second delay MCAS commands another stabilizer nose-down.

      Maximum authority of 2.5 at low Mach (=~9secs@0.27deg/s) and 0.65 at high Mach (~7.2secs@0.09deg/s).

      To the best of my knowledge, having looked upon official reports and other official data, nothing describes how this function integrates its various inputs into a final duration as a part to the STS.

      We can confirm both high rates (when MCAS) and low rates (when STS) and do know that it varies duration, as attested by both accident flights DFDR reports. Most common (flaps-extended@high-rate) MCAS signals had durations of 3/4secs with ~7secs the lengthiest observed, speeds between ~250kts-375kts (0.38-0.57 Mach). All MCAS resets observed are preceded by electric stabilizer trim switch activation. Normal (flaps-down@low-rate) STS was observed with much shorter bursts of activations of ~1/2secs. Ethiopian flight shows one STS activation nose-up overlapped with one MCAS nose-down as soon as flaps end motion from extended to up position.

      • Vasco, you are right that we don’t know, because we don’t have data. My comments were based on the assumption that MCAS would be responsive to decreases in AoA as well as increases. That would be required to limit MCAS to its role as augmentation.

        If instead MCAS was based on cycle time, that would not create smooth control forces on the column as intended, and the aircraft might even porpoise up & down.

        Since there has never been a case of legitimate MCAS activation, we don’t know, we can only go by what the test pilots reported when they tested high A0A. They did not report porpoising or varying forces, they said it felt right to them.

        Admittedly that is a qualitative result and we don’t have any actual data, so we don’t know for sure. Those issues have not been reported in the testing of MCAS 2.0 thus far.

        • Rob, I may be splitting hairs here but, it’s one of the great unknowns about MCAS.
          “Since there has never been a case of legitimate MCAS activation, we don’t know, we can only go by what the test pilots reported when they tested high A0A. They did not report porpoising or varying forces, they said it felt right to them.”
          Do we actually know that there has never been a legitimate MCAS activation? What if MCAS goes on at 10 degrees and off at 9.5? Would it not be hidden in speed trim changes? There’s no light or buzzer that activates when MCAS is active. Just the trim wheel spinning. So, how would a pilot know if the trim wheel is spinning because of MCAS or because of speed trim? Could MCAS be active on every flight? (I admit that’s a long shot, but, other than ‘high value of AOA’ to activate and some other value of AOA to deactivate, we really don’t know how many times MCAS has been active in real life. Do we? Not every Boeing test pilot knew about MCAS. I’m sure some know the general values of when MCAS activates and deactivates, so you’re probably right, but, do we really know? It’s so hidden in it’s activation, that it could have activated and the pilots didn’t notice? When MCAS 2.0 comes out, will it have an indication of activation on the panel? If both AOA’s get out of calibration by the same mechanic checking both of them, will the pilots notice that MCAS is firing at 8 degrees instead of 10? This is what bothers me about MCAS. No feedback to the pilot, and no control by the pilot.

          • Richard, I think if you are correct and MCAS was silently active in the background and completely unnoticed by pilots, that would justify Boeing’s whole rationalization for designing MCAS as they did. Your description is how they claimed and expected it to work.

            I don’t know if that was happening, my speculation would be, based on Boeing’s assertions, that it would take an AoA extreme enough to compromise handling. Boeing said they did not expect it to be needed in the life of the airframe.

            From Bjorn’s diagram for general pitch stability, it looks like the divergence begins around 12 degrees. with a clean wing. (obviously this point is extended to higher angles for takeoff and climb with flaps extended).

            You raise a good point though, if you have 3 or more automatic systems moving the stabilizer, should there be some indication of which one is active?

        • That was my error:
          Most common (flaps-extended@high-rate) MCAS signals
          Should read:
          “Most common (flaps-up@high-rate) MCAS signals”

      • MCAS 1.0 had two specific speed settings or variable at all speeds in between?
        MCAS worked nose up? Really? Like to hear more about that.
        If this is all from the NTSB report, and coming out eight months after the grounding, who’s on first?

        • Ted, Boeing has not released the specs to MCAS. Peter Lemme, a very experienced former Boeing flight controls engineer has tried to surmise as much as he can about MCAS. But, we’re all surmising from what little Boeing has said and what can be fathomed from the accident data published. Simply put, we have a lot more questions than answers. Here’s a link to a Nov 2018 article that gave some good info on MCAS activation rules, I believe from Boeing statements but, not very much else has come out since then. That’s been a major frustration for those of us who would like to know the details of what happened, and to try and avoid another failure.


          • Just responding to the comment above which my comment is indented to, ” While MCAS 1.0 rates are fixed at 0.27deg/s-flaps extended/down and 0.09deg/s-flaps up”

            Is this correct, or misinformation?
            Flaps down? .09deg/s? This is the first I have seen either of these.

          • Ted, All, sorry again. I reverted the rates in that statement too.

            0.27deg/s-flaps extended/down and 0.09deg/s-flaps up
            Should read:
            “0.27deg/s-flaps up and 0.09deg/s-flaps extended/down”

          • Ted, I think VdG is quoting the rates for the Speed Trim System. From what Boeing has been said about MCAS, it should only work when flaps are not deployed. But, we don’t know for sure. When they did the ferry flights from Singapore to Australia, they ordered the planes to fly with flaps deployed all the way, to avoid having MCAS activate.

          • VdG,
            The .09 rate makes more sense to me, since .27 is a multiple of .09, although I have seen .06 previously referenced.

            Interesting that the NTSB says, in the original MCAS function, it was programmed to retrim nose up once it hit the trigger value as AOA decreased. I wonder if that function is part of the new version of MCAS?

          • Ted, yes, that MCAS ANU behaviour also surprised me. It is mentioned in context of hysterisis, which I had to look up to understand the concept. It did make sense then, Wikipedia:

            In aerodynamics, hysteresis can be observed when decreasing the angle of attack of a wing after stall, regarding the lift and drag coefficients. The angle of attack at which the flow on top of the wing reattaches is generally lower than the angle of attack at which the flow separates during the increase of the angle of attack.[7]

            (Linked paper as some good illustrative graphs)

            Regarding the 0.06deg/s, I have never seen it, but NTSB confirms in a footnote that the high-rate is three times the low rate. Speculation follows: if Lemme’s figure of 0.2deg/s manual trim, flaps up (hence high-rate), which Richard points out bellow, is anything to go by, applying the same factor of three for the corresponding low-rate (flaps-down) would give us 0.067deg/s.

      • Ouch… I had one job, and I botched it. Sorry for the confusion raised here. This had been my attempt at compiling very synthetically the pieces of official information I found. I stand corrected, on the flaps-up vs flaps-down confusion.

        Ted, 0.09deg/s is solid data (p33).

        I have other comments somewhere in this thread where I have been sourcing this data as I was collecting it. I just botched it here.

  43. TW, I’m going to give you the same friendly advice that Vasco gave me. When you address these issues, it doesn’t have to be personal, even if you are upset at what has been said. All viewpoints should be heard, people can make up their own minds about what is true and false.

    You’ve made some great contributions here and I hope that continues. You can keep doing that without implying anything about other commenters. The true value in what you say is to bring your knowledge and experience to light. So please take a breath, think about what is important in the issue being presented, and address those aspects. That is what’s most valuable to the community.

    Sorry to bust you, but like Vasco said, I just don’t want to lose your input due to a confrontation style. That would be a real loss. Hope you understand and forgive.

    • Well put, will take the response seriously, thank you.

      Aviation growing up was our whole life, my dad’s job was directly in support (FAA).

      I get too carried away at times.

      I am not sure I would call it a passion as in the DNA. Last 25 years were in support of air freight facility.

      At the end they had shifted from fact based ops and decisions to voodo mechanics.

      I refer to my current state as PTWS (Post Traumatic Work Syndrome)

      • Thank you TW. PTWS, that’s a good one!! It describes many of my experiences too. Will remember it for the future.

    • Rob:

      I need to remind myself that there are also a posters on board now who are technically informed and in the case of Mike – can discuss the aerodynamics end of the theory beyond my ken.

      Richard while I don’t agree with all is not trying to force a closure.

      Forced Closure: That is a term from my surveying days (one of many careers!) Mathematically you run a survey back to its staring point , close up at the end and there is always some error.

      If is withing tolerance you can adjust the miss out throughout the various distance and angles and its non existent as far as being able to see accuracy wise. Like machinery, there is some tolerance in it as a stake in the ground and a tack in the top are not .001 precision.

      If its out of tolerance, professionally and ethically you have to go back and find your mistake. A surveyor is technically a witness to the courts as these are legal document and to not have it correct is the same as a stenographer putting down false information (or labs lying on DNA etc) .

      A bad unethical surveyor will fudge numbers and make it look right, ie Fore the Closure.

      When another surveyor picks up the plat and try to survey off it you run into a bust and anything set is now legal you can’t just undo.

      Personally I ran into one that was 13 feet off. That meant major boundary shift not just in that subdivision, but the next one over was 13 feet short. All legal and recorded. The only good part was we were on the side and not the end, we simply surveyed in our monument right and ours was correct.

      I know of one that was hundreds of feet off and huge legal issues.

      That is what I see some posters doing, Forced Closures and then people that participate accept it and it runs off the rails on a very good forum.

      I see Phillip as trying to force a closure on his agenda. Good part is no one gets hurt but there is the misinformation.

  44. TW, I get what you’re saying, you just have to trust in the diversity of the forum, that enough opinions will be tossed around by enough people and eventually vetted, so that validity will surface and emerge.

    Think of it this way, if everyone here posted the same things as you, you’d have a mutual admiration society, but you wouldn’t learn anything. Diversity is required for learning, and thus has a value unto itself, even if some of what’s said turns out to be incorrect. Finding inconsistencies can be a learning exercise as well.

    The nature of that process is that we will all be wrong at some point, so helps to keep that in mind when mistakes of others are discovered. We are all just one mistake away from being in the same position.

    • Pondering:

      I tend to think that technical people all have views, maybe less opinions but not fully in tech depth as we are not in the Boeing centers (all the details )

      No, I don’t want any accolades but I appreciate it when a good point is made or brought up by anyone including myself.

      Add into it my bewilderment when someone that is Bjorns level is cavalierly dismissed.

      It would be like me arguing physics with Einstein.

      Sure I have some lay understanding of the field, but tell him he is wrong about something?

      I was good at my work and at times I was accused of arrogance (mostly when it got into the political area) but I had a proven track record.

      I had a run in one time with a Branch Manger (early days of my work) when I had to finish a job up at her facility so I could do another one at another facility that was an all day job (if not done a plane reservations was lost and a lot of money for another nights stay with room and board just to do a few hours work)

      My boss finally told the complainer that you know he is a technician and not a politician? We used that a lot afterwords.

      False information is the bane of any techs world, you can’t function on that. I never successfully BS ed a piece of equipment into working.

      So yes a difference of view is fine, but when proven wrong ?? Phew

      • TW, in this forum we are all peers (except for Bjorn) so we cannot force each other to accept viewpoints, or change methodology/approach to problems. We can gather facts and present them along with our reasoning and hope for the best. Some things that are said will be wrong, and we can point them out, but that’s the limit of our authority here.

        Also we can look for motivations. I sense that some of the anti-Boeing comments stem from outrage at the loss of life that occurred, and I get that. I wrote a response to it but decided not to post, as it strays philosophically far from the purpose of Bjorn’s column.

        That outrage is not misplaced or necessarily a bad thing, even if it fuels some extreme views. Better to have an outlet for it where others listen, than for it to be denied or invalidated. Denying or shutting down a person’s true feelings closes the door to ever cooperating with them, which is always a loss.

        People tend to listen when they are listened to, and to not listen when they are similarly ignored. That’s just human nature. So part of our role here is to listen, as well as post our views.

  45. Puzzled … I was looking at the ET302 data trace. At 05:40:45, the 2nd to last, Automatic Trim
    command to pitch down occurs and is labeled as “automatic trim command with no change in pitch trim”.
    That would seem to indicate that the pilots had switched the stab cutout switches OFF. But, there are
    no more Automatic Trim commands at all until 05:43:15-05:43:30, when the last Automatic Trim command
    to pitch down occured (when the pilots had switched the stab cutout switches back ON to issue manual yoke trim commands again)? So, what swtiched OFF the Automatic Trim commands between those two events? The stab cutout shouldn’t have, as it just shuts off power to the trim motor? Is there a feedback loop from the stabilizer position to the FCC going to MCAS? I don’t think the pilots had flaps in. I”m missing something obvious, what is it?
    see paged 26 here

    • I’ve taken a reread of Peter Lemme’s analysis of the ET302 and he doesn’t mention this. So, I’m missing something fundamental in the reading of the data trace. Or overlooking something in the text that explains the MCAS commands not activating the automatic trim command for this period. What am I overlooking? Reading Peter Lemme’s analysis again, it is amazing that MCAS activates for the final time with the plane going at a speed that couldn’t be a stall situation as it’s off the stall speed chart. And am I reading it correctly that electric trim from the yoke has less of a range than the autopilot? On his diagram showing the autopilot limits (the red range) and electric trim limits (the green range), it looks like the AP has more of a range down than the electric trim?

      • Yes, I have been looking for that 0.2deg/s manual electric figure, mentioned by Lemme down in that post’s comments, in the various official documents, yet I could not confirm it. The MCAS known 0.27deg/s slope is indeed more pronounced than manual electric trim in the graphs.

    • That is a good catch, I figure the Automatic Trim commands were not exactly “switched OFF” but were simply not reset according to either reset conditions, during this time the left AoA vane reading persists elevated (~75deg) but may have not pass bellow the threshold required, while the pilot also did not activate the trim switch (flaps remain up too). As soon as the pilot switched back the CUTOFF switches and activated the electric trim, the latter satisfies the reset condition and MCAS awakes pending the 5 sec delay. This would be consistent with known MCAS logic.

      • Bingo! This would’ve been my explanation as well.

        MCAS had reset logic that was based on a safety analysis ridden with faulty/outdated assumptions.

        • Vasco, Mike, are you two suggesting something like the Speed Trim Resets to a new speed setting after the pilot trims up or down via the yoke switch, that the AOA setting for MCAS may also reset to a new AOA value? That it may be based on a relative value rather than an absolute value of AOA? I hear that the activation of MCAS sleep timer is reset every time the yoke trim switch is used. Could that logic path be faulty?

      • Richard, it is harder to go more in-depth into MCAS logic with the data available. Beyond the numbers that have been published, we have to turn over to statements of function that may not reflect fully its programmed logic. For example, there appear to exist two distinct AoA thresholds (activation and reset threshold) for each MCAS regime (high Mach and low Mach). Boeing mentions literally an “MCAS activation region” but maybe they just mean the region between these the high Mach and low Mach AoA activation thresholds, in which case it would be irrelevant in terms of MCAS functioning logic. I rather think the former is the case, but we really don’t know. It would indeed explain away MCAS behaviour for the section of the flight you pointed to.

        Regarding this reset characteristic: I am assuming MCAS is monitoring its input values constantly while it is “dormant”. The system activates (AoA threshold) and begins its internal programmed routine, while inside this routine it either does it job, checks the threshold and persists or “exits” or it is interrupted while doing its thing (2nd? AoA threshold or manual electric trim) and resets. Notice now that normal exit and its alternative, a reset, would leave MCAS simply monitoring for the AoA threshold again, the only difference being that with a reset, an eventual reactivation is put on hold for 5 seconds.

        Regarding the relative versus absolute question, makes no difference in context of a 2nd “deactivating” threshold. If the activating threshold is absolute we can refer to the deactivating threshold in both terms meaning the same value. It would make sense to refer to a relative threshold for both high Mach and low Mach which would likely employ different AoA values.

        Do notice how the bulk of this comment is speculation.

    • Richard, we don’t know, but since the low-speed problem was found during flight testing, it follows that the solution also would have been subject to flight testing, and the authority selection also may have been influenced by that testing.

      Bjorn has said that one justification of that authority was the need for MCAS to respond quickly as the nose is rising towards the edge of the envelope. Reaction time is a factor for MCAS, just as for the pilots.

      The thing we know for sure, is that MCAS was not tested in the malfunction scenario that occurred, or likely in many other possible scenarios. That was a major failing that you have alluded to in past posts.

      Even with the higher authority, MCAS was not necessarily a major problem if it had been designed and implemented correctly. Exceeding 12 degrees of AoA, outside of takeoff/landing, would be a rare event, which was valid, but that was used as a rationale to not consider what might happen in other scenarios, which was not valid.

      • Rob,
        Isn’t the first thing a 737 pilot does when the stall shaker goes off is to add thrust?
        The following article seems to think that M=td .. or that the sudden added thrust will
        pitch the plane up. M=Td
        Is MCAS in there to avoid the sudden pitch up from the sudden thrust of the engines?
        Should the recovery procedure for the 737 be changed to 1) pitch down, then 2) increase thrust?
        I’m not sure how fast jet engines ‘kick into gear’ these days. Years ago, there was quite a
        time lag between throttle up and additional thrust.

        • Richard, I think the MCAS function was meant to increase the column force at high AoA, becoming effective before the stall alarms or stick shaker. It’s not a response to engine thrust, as far as I know. It’s meant to satisfy regulations that discourage a pilot from using authority to stall the aircraft, or at least inadvertently doing so.

          Stick-shaker is part of stall prevention, sensing an impending stall based on several factors. MCAS was not intended for that purpose.

          In the accidents, stick shaker activated for the same reason as MCAS, faulty AoA reading. It actually was active before MCAS pushed the nose over, but that would not normally be the sequence.

          It’s true that adding engine thrust generates a pitch-up moment. This is partly due to engine location and partly due to increasing lift with airspeed. Even tail-mounted engines will cause the nose to come up as power is increased, and to drop if power is decreased (all other things being equal). But the effect is enhanced by torque if the engines are below the CoG (M=Td).

          • Rob, Picture yourself as test pilot Ed Wilson, testing the low speed stall characteristics of the 737-MAX with MCAS 0.5. (before the big changes). You’re slowing gradually to approach stall, and
            then you reach the point that you’re supposed to count to 3 and recover. You add thrust at a high AOA,
            low airspeed, and I’m assuming the plane would pitch up because of the added thrust, and the added airflow of the nacelles, and whatever else. You’d probably try and hold the column down, towards
            level flight. I’m assuming something in this time frame was not happening as it did on the 737-NG, and/or not within the test parameters for certification. The test pilots reported back to the designers to do something, to fix ‘it’. The ‘it’ hasn’t been described that well by Boeing. But, we know the fix was a change to increase the speed and authority of MCAS. Now, after the accidents, Boeing seems to be leaning towards a less powerful MCAS authority. Without a good description of what ‘it’ is, we’re in the dark as how the new MCAS fix will work when it is rolled out. We need a better definition of ‘it’. I want to know what ‘it’ is or what ‘it’ was, from the test pilots description and data.

          • I am tring to follow this but getitng pretty deep into the weeds.

            What I can tell you from a pilots perspeicv that was recenifn confimre for a 737 type stall.

            Unload, unload unload is the mantra being taught.

            That is dump the nose.

            Yes, you also add thrust but the first move is to dump the nose.

            Its much quicker and more effective in recovery.

            In the case of a power on stall (accelerated) I believe we took power off to avoid an over-speed but would have to look it up.

            But we dumped the nose first as it would take the engine a short time to spiral up through rpms and easily caught.

          • Richard,

            As far as I know, Boeing has only claimed that the new MCAS:

            … can never command more stabilizer input than can be counteracted by the flight crew pulling back on the column.

            and that the new MCAS will:

            … only provide one input for each elevated AOA event. There are no known or envisioned failure conditions where MCAS will provide multiple inputs.

            This does not mean that the new MCAS will only utilize the lower stabilizer rate.

            I think the primary reason MCAS (as implemented on the crash flights) had too much authority was because Boeing’s faulty assumptions that drove the MCAS reset logic. If this reset logic would’ve been based on correct assumptions, only one MCAS activation would’ve occurred during the crash flights. I think this would’ve allowed the pilots time to recover.

          • Richard, you may be thinking in terms of intentionally stalling the aircraft, as a test of MCAS.

            I think the test regime would be more like a powered climb that becomes too steep. You might be turning as well, as that would be more consistent with an emergency maneuver or an aborted landing.

            The condition where MCAS was originally expected to be used was a steep climbing turn under power. In those circumstances airspeed and thrust might not be a problem, but you are approaching the critical AoA.

            The loss of control at critical AoA can be sudden, since it’s based on separation of the airflow from the control surfaces. If the aircraft also begins an acceleration in pitch rate, due to the shift forward of CoL from the large engine nacelles, that could be unexpected to the pilot, as the additional lift & moment are not coming from the control surfaces he is operating, so he doesn’t necessarily sense them.

            So, the purpose of MCAS is to represent those forces in the control surfaces, so the pilot is getting the correct feedback that the aircraft pitch is accelerating and becoming harder to control. At the same time, it is countering the non-controlled accelerated pitch moment using a control surface (the stabilizer).

            This counteraction is not meant to be complete, MCAS is not closed-loop in that sense, it doesn’t know what the pilot’s objective is, and is not trying to force or guide him. Just trying to represent the forces on the aircraft appropriately so the pilot is sufficiently aware of them.

            This the difference between handling augmentation and stall prevention. I realize it’s a very subtle point. I didn’t understand this for a long time, reading repetitively through Bjorn’s pitch stability series has helped a lot.

            As Bjorn said, finding yourself in an accelerated pitch state, without augmentation or compensation, is very uncomfortable. You can lose control authority quickly, and just as you begin to sense it. This is why the regulations require certain handling characteristics in this regime.

            I hope that makes sense, I think understanding the difference between this situation and a stall due to lack of airspeed or insufficient thrust, is critical to understanding MCAS.

        • Richard,

          Although there is a sudden pitch up tendency due to a sudden thrust increase, I doubt that was the major reason for needing MCAS. All aircraft, where the thrust line acts below the CoG, experience a sudden pitch up tendency when the thrust suddenly increases. Very few of those aircraft need MCAS, including the NG.

          • Mike, The thrust difference looks like about 10% between the NG and MAX.
            737-NG .. range from 22,800 lbs to 27,300 lbs
            737-MAX.. range from 25,958 lbs to 29,317 lbs
            With that and the nacelle placement, maybe it’s enough to need MCAS. I don’t know if the high
            bypass engine might have faster response time or

  46. I wonder if any Airlines with 737-MAX’s grounded, or the FAA and/or EASA have thought to backup a copy of the MCAS 1.0 load program directly from a FCC? I assume the code is in an EPROM (electrically programmable read only memory) chip? And any other software that is currently in the MAX?

  47. Forced Closure. That’s a good term. But I think it applies to much more than one person’s musings on this site. To me, it’s a narrative term. It’s a struggle to reconcile ideological positions to actuality. And the two tragic crashes of the Max evoke a lot of ideological impositions: the inferiority of non-Europeans before operating advanced technology, the end of American exceptionalism, the follies of neoliberal hegemony, corporate greed, income inequality, etc.

    And while engineers and workers (and politicians) build the world, the rest of us live in it — never questioning what went into the sausage — and always assuming that a sleek airliner (though it may have some of the shape) is nothing like a sausage. Butchers throw meat no one would imagine eating into a sausage — it’s basically spiced up garbage. For the regular traveler, however, the airliner, must be like top sirloin or rib eye. We risk our lives in them.

    So, while Boeing refuses to tell the aircraft engineers and academics of the world what went into its sausage — forcing them to guess — the rest of us, with little technical expertise, are no longer fooled. We may not know or care what they are, but ugly, nasty things went into the making of the 737 Max. It can be anti-stall or flight smoothing, but it’s not what we had been led to expect of advanced, blue chip American technology that holds our lives in its hands. So, we can impose all of the forced closure on this debacle that we desire, but it won’t change what looks like a slow moving and absolutely devastating train wreck of epic proportions that Boeing has forestalled by spending lavish amounts of money on PR — money it could spend on engineering. It’s forced closure to say that the US and Boeing can’t afford to ground the Max forever.

    Meanwhile, the rest of the world looks on in stupefied awe and tries to figure out how this company and the country got itself into such a mess. Two good — and sad — recent articles:

    (Please note, I’m still wondering about the vilified airmanship of the ET302 captain and his never reducing the speed of the aircraft. I think about Patrick Smith discussing the terrible takeoff performance of the 737, and imagining what it must have been like to pilot that aircraft — accumulating 8,000 hours — out of a high heat and high altitude airport that sucked performance. And then I think about Richard Davenport’s question about that flight and his statement that: “MCAS activates for the final time with the plane going at a speed that couldn’t be a stall situation as it’s off the stall speed chart.”)

    • Real Steve: From a pilots perspective, its a struggle in this.

      I agree fully that the aircraft should not do bizarre things which MCAS 1.0 did due to more than one failure type on the AOA. I think what they did was criminal even if they don’t get charged for manslaughter.

      On the other hand, from a pilots perspective, the mishanlind g of the crissi is an anathema.

      Clearly in both cases they failed to manage the situation. Loss of speed control is one, that is so basic as to just be imponderable.

      Boeing should have owed up to their end and left the comments about pilots out of it.

    • ..(Please note, I’m still wondering about the vilified airmanship of the ET302 captain and his never reducing the speed of the aircraft…”

      Seems to this SLF that when on a hot daym a few thousand feet AGL and the nose all lof a sudden pointing down, and knowing that thrust reduction usually ( always? ) results in nose down- with all the lights hjorns going off , phony airspeed ,- that thrust reduction would be a bit lower on the list of WTF happened and what do I do first, second, third, etc

      • This 3 minute video gives an idea of the roller coaster ride, with alarms going off, etc
        pulling back the throttles does seem low on the list
        (yes, I know it’s not the most realistic video, but, it gives you an idea)

  48. If the new MCAS has only one activation at a time, then what was the thinking the first time with the MCAS program, unlimited activation with the 5 sec delay in between? Where multiple activation needed to pass the either of the approach to stall tests?

    What will the new MCAS look like? What angular trim speed(s) and for what duration?

  49. Ted, I think MCAS 1.0 was considered mostly in the context of a legitimate activation. If the pilot did not respond to alleviate the high A0A condition, it would reset and continue modifying the behavior of the aircraft, to encourage recovery.

    Obviously that programming could be disastrous in the context of non-legitimate activation. MCAS did not receive the testing that was needed for failure scenarios.

    We know only general behavior of MCAS 2.0, not specifics. It can engage only once in a high A0A event, and it’s authority has been limited so the pilot can counter it with the control column. Also it will use multiple sensors so can’t be triggered be the failure of a single sensor.

  50. I”m still trying to figure out the missing MCAS automatic trim
    commands for 2 minutes on the ET302 data trace. Specifically
    from 05:41:00 to 05:43:20. (Page 26 of the preliminary report)

    (the very last sentence of this 737-MAX training manual page)
    under the section “Maneuvering Characteristics Augmentation System”
    “Only the F/O’s column cutout switch module is affected because
    it is the only module that interfaces with the FCC’s”

    I’ve been trying to decipher that sentence and the only thing I
    can come up with is that there is something different between
    the Captains (Left side) column cutout switch and the First Officers
    (Right side) column cutout switch. These are the limit switches
    that cutout the Stabilizer Trim System simply by pulling back on the
    column. MCAS supposedly runs through the override switch column switch,
    bypassing these column limit switches.

    But, this sentence, under the MCAS paragraph heading has me wondering.
    What if MCAS only runs through the Captains (Left side) override switch?
    What if the First Officers (Right side) column cutout switch is still intact?
    MCAS supposedly only uses the Left side AOA.
    Then, why not just the Left side override switch also?
    If that is the case, then that may explain why on the ET302 data trace, there were no automatic trim commands between
    05:41:00 and 05:43:15, a good 2 minutes of no MCAS commands, if the F/O’s column was being held aft by the First Officer. According to the Cockpit voice recorder. The F/O starts helping the Captain on the Column, at just about the time the automatic trim commands stop.
    At 05:40:35, the First-Officer called out “stab trim cut-out” two times. Captain agreed and First-
    Officer confirmed stab trim cut-out.

    At 05:40:41, approximately five seconds after the end of the ANU stabilizer motion, a third instance
    of AND automatic trim command occurred without any corresponding motion of the stabilizer,
    which is consistent with the stabilizer trim cutout switches were in the ‘’cutout’’ position

    At 05:40:44, the Captain called out three times “Pull-up” and the First-Officer acknowledged.
    05:40:50 was approx. the last part Automatic Trim Command, that didn’t have a pitch change,
    and is assumed to be when the stab trim cutout switches were stopping the trim motor.

    There are no more MCAS automatic trim commands issued for 2 minutes? The AOA was still very high, flaps up, autopilot off, so what stopped the trim commands from being issued? What about
    the F/O’s column cutout switches being activated by his helping the Captain now, pulling back with force on his Right Side column after the Capt. asks him to “Pull-up”? Previous to this
    time, I’m assuming only the Captain was actively pulling on the column.

    From the preliminary accident report, more references to the F/O “applied to both columns
    simultaneously” / “pitch up together”.
    From 05:40:42 to 05:43:11 (about two and a half minutes), the stabilizer position gradually moved
    in the AND direction from 2.3 units to 2.1 units. During this time, aft force was applied to the
    control columns which remained aft of neutral position.
    The data indicates that aft force was applied to both columns simultaneously several times
    throughout the remainder of the recording.
    At 05:43:04, the Captain asked the First Officer to pitch up together and said that pitch is not

    The only time (the final time) that the Automatic Trim Commands return is after the Stab Trim
    switches are turned back on. And this one doesn’t match with my theory because it mentions
    “simultaneous aft column force”. I assume at this point, both are pulling with all of their
    might on the control columns.
    At 05:43:20, approximately five seconds after the last manual electric trim input, an AND automatic
    trim command occurred and the stabilizer moved in the AND direction from 2.3 to 1.0 unit in
    approximately 5 seconds. The aircraft began pitching nose down. Additional simultaneous aft
    column force was applied, but the nose down pitch continues, eventually reaching 40° nose down.
    So, unless the Stab Trim Cutout switches somehow give feedback to the Automatic Trim Commands,
    I’m still puzzled as to why there were no MCAS commands on the traces for two minutes?

    I didn’t follow through with this theory on for the Lion Air accident, because we may be dealing with different FCC software versions between the accidents, if this P11.1 software
    was updated on the Etheopean Aircraft, after January 2019.
    This sentence that started this thread came from a training manual dated Jan 2019.
    I’m ASSUMING that this training manual was changed for the new Collins FCC software ver P11.1 being released in Jan 2019.
    (The following from Chris Brady’s 737 Technical website)
    P10.0” software – Jan 2017 – This was the basic software for the MAX-8.

    P11.0” software – This was updated software to include the new MAX-9.

    Collins FCC software P/N 2270-COL-AC2-22 (known as “P11.1” software) MAX and NG – Jan 2019
    This has improvements and bugfixes to a total of 17 functions, many of which will give new maintenance messages. I have listed the more interesting ones here:

    Improved IAN backcourse function.
    Improved LNAV to LOC mode transition.
    LNAV to remain armed on the ground while switching from ground power to airplane power.
    Improved A/P pitch response while performing an emergency descent.

  51. Speed Trim System verses MCAS. The Pitot Tube verses the Angle of Attack (AOA) sensor. What happens in the Air Data Computer (ADC) with the data? We know there’s a high Mach speed
    portion of MCAS that was the original MCAS. Later on, during flight testing, a change to MCAS was made to extend it’s functions to a low Mach speed portion of flight, but, with added control movement. In the ADC, the Pitot Tube values are corrected using AOA values. That’s why there were differing speeds on the Left and Right side cockpit instruments.
    In the final Lion Air accident report
    As originally delivered, the MCAS became active during manual, flaps-up flight
    (autopilot not engaged) when the AOA value received by the master FCC exceeded
    a threshold based on Mach number. When activated, the MCAS provided a high
    rate automatic trim command to move the stabilizer AND. The magnitude of the
    AND command was based on the AOA and the Mach.
    MCAS was based on the AOA and Mach. Mach was corrected via the AOA sensor. Is there a feedback loop being established within the MCAS software using the incorrect AOA values and incorrect Mach values? At high Mach settings, I would assume MCAS would only want to use low value AND commands.
    This wasn’t the case in the ET302 accident. They were at high Mach, and MCAS was issuing high
    value AND commands. It would be nice to be able to study the actual MCAS code logic to see if
    there is some fall through logic somewhere. Or even a Truth Table from the design spec sheet of
    MCAS, to see what should have been output to the stabilizer, high or low speed AND commands with
    the inputs (flap position, A/P on or off, AOA value triggers, reset points, Mach trigger values),
    to see if the MCAS software was working as advertised.

  52. Richard, the source of the airspeed (Mach) correction is the effect of AoA on the pitot probes used to determine airspeed.

    The angle of incidence of the airflow impinging on the probe causes a small but measurable error in the dynamic pressure registered by the probe, which is then compared to the static pressure to determine airspeed. For the small angles of attack of a commercial airliner (10 to 15 degrees), the correction is not large, on the order of a few to several percent.

    If an A0A disagree occurs, such that the difference between the two AoA sensors is large, this triggers an airspeed alarm because the correction is no longer being applied, the pilot is seeing the uncorrected airspeed. But I don’t believe there is a cascading feedback-type failure, the Mach values provided to MCAS, even without AoA correction, would have been within several percent of the true values.

    The high-rate MCAS activity was programmed for low-speed (low Mach), whereas the low-rate MCAS activity was programmed for high speed (high Mach). The relatively low Mach of Lion Air could have triggered the high-rate MCAS action.

    The Ethiopian flight was accelerating toward, and eventually exceeded, Vmax for the airframe, but that is not necessarily high Mach. Vmax of roughly 340 knots is about .55 Mach at 10,000 feet. Also that speed is well above the regulation airspeed of 250 knots for that altitude.

Leave a Reply

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