Ethiopian airliner down in Africa

[yanrair]

Originally Posted by deltafox44

I know that, thank you

At low speed and with a limited amount of AND untrim, increasing speed may reduce or cancel aft column required, due to static stability of the aircraft. But in the case of ET302, even at 500 kt IAS and still both pilots pulling as much as they can, the aircraft has a -10° AoA and -2g ! No hope to reach any in-trim airspeed..

What I mean is that manual trimming was impossible due to loads on elevator (which vary as the square of airspeed) ; that the only thing that could have save the day is to slow down at a speed where loads would have permitted manual trimming.

Instead, they let full power up to VMO and above... I could not imagine why, now I know that it was what was recommended by Boeing...

my apologies. Misunderstanding. I was just referring to 737 needing AND trim all the time during acceleration.
all you say is correct. The NNP for runaway stab. Is predicated on staying ahead of trim requirements. To avoid the last ditch heave up , and unload- TRIM TRIM TRIM NU routine. Use trim switches NU until stab back in trim - not blip blip , but major NU input over several seconds. STAB OFF. TRIM Manually thereafter. In runaway stab in previous variants by the time you notice, the stab. will be a couple of divisions AND. Or more and requires a sustained ANU input.

[yanrair]

Originally Posted by fdr

back on planet earth, where Newtonian physics still reigns, you may find that for an aircraft to be statically stable, mass is always ahead of the center of pressure. Now your B2 may be unstable but pretty much no airliner is unstable, including the Max8. It has a reduced stick force gradient in part of the envelope, which MCAS addresses, just as STS deals with another gradient issue in a different part of the envelope.

Nect time you you are in a pub with a dartboard, throw a dart feathers first and observe status stability play out.

The acceleration issue as quoted before this comment, where in the excessive out of trim AND case is so that the aircraft approaches the speed that it has been miss-trimmed to, and near that point, the stab forces between section Cm and the elevator countering load will approach a minimum, allowing for movement of the stab with less force.

Now the the non linear longitudinal stability arises from the component of lift arising from the engine nacelle, which happens to be forward of the center of mass, so as additional lift is generated, that results in a pitching moment being added to the total body pitching moment. The fuselage is a lifting component, as are the wings. The stabiliser normally generates a downforce to counter the pitching moment of the total aircraft.

end

Well spotted. You trim nose down during acceleration because centre of lift moves forward. From memory.

[GordonR_Cape]

A0283

So, Mr Sinnett’s statements add something but are not clear enough and not deep enough. Many would welcome more.

I posted earlier in the thread that my impression was that only a detailed public presentation and publication by the ‘chief engineer (with FAA delegated certification signature authority)’ of the MAX would do. I wonder if Mr Sinnett fits that bill? The text has a ‘nice and sunny taste’, rather more commercial than tech savvy …

Way back on 27 Mar, the Seattle Times mentioned that Sinnett was trying to win back confidence. As VP of development, this is a double edged sword: https://www.seattletimes.com/busines...iginal-design/

Boeing details its fix for the 737 MAX, but defends the original design

Edit: As a computer programmer (not a pilot), those 3 key questions were ones I asked repeatedly after the "fix" was announced. It has taken a long time for details to emerge, so AvWeek needs to be thanked for that.

[Goth]

“In a situation where there is erroneous AOA information, it will not lead to activation of MCAS,” says Sinnett. He underlines that the entire speed-trim system, including the MCAS, will be inhibited for the remainder of the flight if data from the two vanes varies by more than 5.5 deg. If an AOA disagreement of more than 10 deg. occurs between the sensors for more than 10 sec., it will be flagged to the crew on the primary flight display.

Is there a reason for the different thresholds for AoA disagree and STS/MCAS inhibit?

So if an AoA vane sticks so that it's (say) 8 degrees offset from the other for a period, STS and MCAS will be disabled for the remainder of the flight, with a change to handling characteristics but no crew warning - or is there an STS/MCAS warning indication as well as AoA disagree?

[Fortissimo]

You trim nose down during acceleration because centre of lift moves forward

I think the larger effect is not so much any centre of pressure movement (which does contribute) but the fact that your overall lift is a V-squared function. When you accelerate you have to reduce AoA to maintain the same lift (for level flight), so lower nose attitude and more AND trim.

[Ian W]

Originally Posted by bsieker

That is possibly the worst suggestion so far.

Assembly code is almost impossible to analyse for correctness in any meaningful way. It is far better (and provably so) to write in a well-specified (i. e. not C) language, prove the source code correct (for which scalable and practical techniques exist today), or define and prove correct a finite state machine and have code generated from it. That still leaves one with a need to have reasonable confidence in the compiler, but in many cases the service history for the most-used language core, and, in some recent cases, formally verified compilers, take care of that.

Just because you have one hero programmer who claims to have done it "Right" in assembly does not help you in any way because you need to demonstrate that it does what it is supposed to do (reliability), and never does what it is not supposed to do (safety), and ideally also never fails (availabilitiy). And this cannot be demonstrated by testing alone to the extremely high requirements needed in aviation. Assembly and machine code are avoided like the plague in safety-critical programming, and rightly so. Where some parts require it, extreme care must be taken to get it right, and the amount must be kept to a minimum.

Besides, as threemiles has pointed out, the implementation is not the problem (as far as we can tell, it may be flawless), but the specification. "Working as specified" can also mean that it did the wrong thing.

Bernd

Almost all the major 'programming errors' I have seen have been perfectly implemented errors in understanding due to poor systems analysis/design. Coding these days rarely has errors as there are many tools that can be used to verify the code it will also validate correctly as the code does just want the designer erroneously asked it to do. This is the shortcoming in mathematical approaches and formal proofs they do not find these errors in understanding but they do limit the avionics; it is one of the reasons FMCs are beasts of little brain as anything with any power is beyond formal proof.

.

[fdr]

Originally Posted by Fortissimo

I think the larger effect is not so much any centre of pressure movement (which does contribute) but the fact that your overall lift is a V-squared function. When you accelerate you have to reduce AoA to maintain the same lift (for level flight), so lower nose attitude and more AND trim.

section pitching moment, CsubM tends to increase on symmetric and cambered sections with increasing AoA. A section with a reflex camber may maintain low moments, which is why they are used occasionally on rotor sections, where they reduce pitch link loads. As Mach increases, pitching moment tends to increase. At very high Mach, (above MMO, approaching MD,), CsubL for a given AoA reduces, when a normal shock develops on the underside of the section. As washout results in a higher AoA inboard, and section T/C is greater normally inboard, the shock development is most pronounced inboard, and to maintain a constant total lift, the tips become more loaded, resulting in a shift rear wards of the total lift relative to longitudinal station, with increased lift distribution outboard. Separate to all that, the aircraft is required to have a nose down trim needed for increasing speed above trim speed (AoA) and vice versa, the result of inflow to the tail, tail volume and section moment design. A change in speed alters the normal force from the section as a function of the differences of the squares of the speeds, therefore level flight requires a lower AoA which is achieved by a lowering of attitude. To achieve that outcome the slope of CsubL to AoA, the a-slope need to be equal or the tail has to have a lower slope than the wing, when the inflow. Volume and section coefficients are considered. You can fly a plane with neutral static stability, they are fun to fly, they just are higher workload on instrument flying for extended periods, ask any Helo pilot ( helos are mildly statically stable, but are dynamically unstable in pitch in forward flight and hover and also in yaw in hover)

[Ian W]

Originally Posted by diclemeg

And yet...it's the MD95, also known as the Boeing 717, that never has had a fatality..... McDonnellDouglas made a great plane...

Of course to be pedantic it was the Douglas Company 9. They had problems with the screw jack and stab jamming too didn't they in the 717 older brother MD-8# series

[fdr]

Originally Posted by yanrair

Well spotted. You trim nose down during acceleration because centre of lift moves forward. From memory.

nope, Re-read the quote, and get a white board and draw out the forces to see the couples.

read my follow up comment, any further questions feel free to PM.

CsubP moves forward on a section as AoA increases, however it is dependent on section camber, and also to an extent the section LE radius. A section that has a laminar separation bubble will have a slight wobble in moments for lift and pitching but not a big deal. Mach tuck occurs due to reduction in lift inboard, on a swept section this loads the tips giving a pitching moment, as well as a movement rearwards of Cp and the resultant pitching moment from that change, Cp follows normal shock movement on the section. As the wing section and T/C result in shock formation being more pronounced inboard than outboard, the total result is an increase in pitching at higher Mach. At speeds below Mcrit, lift and section a-slope, and inflow to the tail result in pitch moments leading to nose down trim being needed as speed rises above trim speed. Shock formation is not relevant to JT or ET’s events, they had normal pitching moments going on other than the MCAS establishing a major trim error for desired speed.

The annoyance of MCAS is that it has authority to schedule trim for a period of more than 3 times the miss-trim case considered for certification, per 25.255, which seems like a lousy concept.

[LEOCh]

Originally Posted by fdr

back on planet earth, where Newtonian physics still reigns, you may find that for an aircraft to be statically stable, mass is always ahead of the center of pressure.

Agreed with the rest of your post, but as a counter-example to the statement above, consider a 737 NG loaded to it's absolute rear limit (36%MAC), and trimmed to fly at high but nonstalling AoA. It's CP will be relatively forward and close to but not reaching the wing aerodynamic centre of 25%MAC (will be approximately 32%MAC). CP is hence ahead of the CG and the tail is lifting slightly to maintain trim. However, the aircraft is still longitudinally stable, as although the tail is lifting it's local AoA is less than the wing. Of course this is an edge case for stability, more often CP is behind the CG in all flight regimes and the tail is always in downforce.

[fdr]

Originally Posted by LEOCh

Agreed with the rest of your post, but as a counter-example to the statement above, consider a 737 NG loaded to it's absolute rear limit (36%MAC), and trimmed to fly at high but nonstalling AoA. It's CP will be relatively forward and close to but not reaching the wing aerodynamic centre of 25%MAC (will be approximately 32%MAC). CP is hence ahead of the CG and the tail is lifting slightly to maintain trim. However, the aircraft is still longitudinally stable, as although the tail is lifting it's local AoA is less than the wing. Of course this is an edge case for stability, more often CP is behind the CG in all flight regimes and the tail is always in downforce.

nope.

25% is a convention for measurement of moments, so a wing with zero pitching moment will have a Cp at 25% chord. Refer to Abbot and VonDoenhoff to look at the moments that occur on a section. Next time walking around your brand A or B plane have a look at the section of the stab, it is an inverted cambered section, which means it has a zero lift line, ZLL, that is considerably beyond a zero stab LE up limit, which is usually around 2 to 3 degrees up dependent on flavour. The stab resides in an area of down wash on standard tails, less so for T tails or cruciform tails. The stab on an A or B brand does not get to a point in normal use of trimming to an up force. The elevators of course may result in a change above the stab limit, but would be untrimmed. For your vanilla flavoured brand, the neutral point is way, way further back, around a center of mass aft of 40% for a plane with an aft envelope limit of 32%. Long time back we looked at a B744 that achieved 43.5% in flight... vs a 32% envelope. The plane was marginally statically stable (being generous), and appeared to be slightly dynamically unstable, the autopilot coped with the mis load, but dang if the elevators weren’t working their passage, they oscillated for the whole flight. To achieve that level of load error, the nose wheel was not on the ground at 80kts, and on landing, the plane sat with the nose wheels off the ground, unable to steer for taxi. Even at that case, the tail was producing a slight down force.

[3rd_ear]

MCAS... “not a stall-protection function and not a stall-prevention function,” says Mike Sinnett, Boeing Commercial Airplanes vice president of product development and future airplane development. “It is a handling-qualities function. There’s a misconception it is something other than that.“

Eeeeew! That gives me the heebies, it reeks of disingenuity. Is he justified in what he says?

[GordonR_Cape]

Originally Posted by 3rd_ear

Eeeeew! That gives me the heebies, it reeks of disingenuity. Is he justified in what he says?

As a computer programmer and former engineer, the time-delay between the AOA input, and the 10-15 second activation of MCAS, could lead to oscillatory behaviour. Pilot input may still be needed, along with some training.

Edit: I meant it is not an 'elegant' solution, more a band-aid. And the nose-down trim still needs to be explained to pilots.

Edit #2. Forgot to add that when MCAS 'unwinds' the trim that it previously applied, this is another time delayed effect.

[bsieker]

Originally Posted by 3rd_ear

Eeeeew! That gives me the heebies, it reeks of disingenuity. Is he justified in what he says?

Yes, and this has been discussed a dozen times here. Of course one reason for the certification requirement for certain control forces at increasing angles of attack is there to make it harder to stall the aircraft inadvertently, but the primary reason for MCAS is to fulfill very specific criteria for control forces (14 CFR, §25.173), and even more specific instructions on how to demonstrate their fulfillment (14 CFR, §25.175).

Bernd

[catch21]

Sadly I haven't been able to keep up with all the posts in this thread so apologies if this is a duplicate.

How did the MEL change with respect to AoA sensors with the introduction of MCAS?

My understanding from this:

http://fsims.faa.gov/wdocs/mmel/b-737_rev%2057.pdf

is that only one AoA sensor is required before despatch?

[bsieker]

Originally Posted by GordonR_Cape

As a computer programmer and former engineer, the time-delay between the AOA input, and the 10-15 second activation of MCAS, could lead to oscillatory behaviour. Pilot input may still be needed, along with some training.

Edit: I meant it is not an 'elegant' solution, more a band-aid. And the nose-down trim still needs to be explained to pilots.

I don't recall anyone having said that there is a delay between fulfillment of MCAS activation criteria and the (first) trim input. There is only a 5 second inhibition of further trim inputs after manual electric trim input.

Bernd

[Lost in Saigon]

Originally Posted by Chris2303

SLF here

Could the pitch up problem caused by the new engine position be ameliorated with the use of strakes such as those on the CF6 attached to the DC10?

The problem with the MAX is that the larger, more forward, engine nacelles create too much lift at high angles of attack. Strakes would create even more lift. Why would that help?

[bsieker]

Originally Posted by catch21

Sadly I haven't been able to keep up with all the posts in this thread so apologies if this is a duplicate.

How did the MEL change with respect to AoA sensors with the introduction of MCAS?

My understanding from this:

http://fsims.faa.gov/wdocs/mmel/b-737_rev%2057.pdf

is that only one AoA sensor is required before despatch?

The line referring to the alpha vanes explicitly only lists Original and Classic models (-100/-200/-300/-400/-500). Not NG (or MAX).

Bernd

[3rd_ear]

Originally Posted by bsieker

Yes, and this has been discussed a dozen times here. Of course one reason for the certification requirement for certain control forces at increasing angles of attack is there to make it harder to stall the aircraft inadvertently, but the primary reason for MCAS is to fulfill very specific criteria for control forces (14 CFR, §25.173), and even more specific instructions on how to demonstrate their fulfillment (14 CFR, §25.175).

Ok, I'm with you - I've read most of this thread and yes, the intent of MCAS seems clear. Perhaps I wouldn't have reacted if he'd said "intended to be a handling qualities function" instead of "is...". Maybe wrongly, I mentally separate emulation from underlying functionality (computer background) and that doesn't apply here? As if, in this case, "sure, it's just like the NG but boy, we had to screw with the kernel to get that working".

[deltafox44]

Originally Posted by yanrair

my apologies. Misunderstanding. I was just referring to 737 needing AND trim all the time during acceleration.
all you say is correct. The NNP for runaway stab. Is predicated on staying ahead of trim requirements. To avoid the last ditch heave up , and unload- TRIM TRIM TRIM NU routine. Use trim switches NU until stab back in trim - not blip blip , but major NU input over several seconds. STAB OFF. TRIM Manually thereafter. In runaway stab in previous variants by the time you notice, the stab. will be a couple of divisions AND. Or more and requires a sustained ANU input.

Here is what I was referring to (published by satguru) :

Recent training handbook also recommend using the speed for which the aircraft is trimmed, i.e. AND => increasing speed

In the case of ET 302, it is clear that INCREASING speed had no chance of success and lead to loosing control.

REDUCING speed would have caused a pitch down moment induced by static stability and by reduced thrust, which AMHO could have been counteracted by elevator, until the moment when the aerodynamic loads would have been alleviated enough to permit manual trimming. If manuel trim were still jammed, aft forces required on the column to maintain level flight would have been lower (due to logic of feel and centering system) with no risk of elevator blowback anymore.