Was MCAS needed?
 

Perhaps MCAS was not needed to all on the 737Max. Patrick Ky of EASA has said that "they pushed the aircraft to its limits during stall tests, assessed the behaviour of the aircraft in failure scenarios, and confirmed that the aircraft is stable and has no tendency to pitch up even without MCAS."

https://theaircurrent.com/aircraft-d...37-max-at-all/

I obviously don't know all the ins and outs of the MAX design process, but since the beginning I have wondered why: if the longer engines caused uncommanded additional pitch-up at high AoAs, then why not just modify the artificial feel in pitch and leave the control surfaces alone? Or have the stick shaker start at a slightly lower AoA to warn of the impending pitch-up?

Good point, Bergerie.


I don't know why* Boeing didn't bring in FBW on the B737. Even if they did so only in pitch, it would have brought benefits such as auto pitch trim, no pitch/power couple, turning without needing to pull back; and MCAS would not have been needed - just a change of some variables in the software.

* well, of course, I do.

Making of MAX was a ruthlessly comercial act by Boeing. They weren't going to spend a penny extra if it could have been saved. If it could be certified they would have sold the dummy without MCAS, May be with some luck they would have gotten away. Afterall it was a false activation that exposed them.

I realize that the Boeing Company has a diminished reputation, but there is one thing it is noted for. It is extremely hard to add a new “feature” to a Boeing design because in Boeing’s philosophy, “it has to buy its way on”. A lot of good ideas have been rejected over the years. Not saying MCAS was a good idea, but Boeing had to have thought it essential or they would not have expended the resources to add it. The need would have been identified by Boeing engineers. So I don't doubt there was a technical need, even if the solution leaves something to be desired.

It would just be yet another big patch for a very old airplane, with questionable benefits. They were better off pushing the airplane out as it was (in hindsight, with better MCAS redundancy), and rather focus on developing a 737 successor from scratch.

Didn't they add "FBW" for the spoilers in the MAX? Anything else probably would have required a re-design of the impacted flight control system that simply was too expensive, and would have taken too much time, always worth remembering that Boeing was in quite a desperate game of catching up to the order advantage Airbus had.

That said, what puzzled me at the time was the difference in testing program, although Boeing had to change quite a lot more on its new variant than Airbus had, Airbus did more than twice the number of test flight hours, which in itself does not say all that much of course, but it was a point i noted in passing.

According to the FAA it was needed.Source https://www.faa.gov/news/media/attac...ummary-v-1.pdf (page 10)

Stick force gradient must not become lighter towards a stall. But this is what supposedly happened.

As #9 MCAS was required to meet stick force requirements; it was not directly related to the stalling characteristics. The comments about stalling could relate to acceptability if MCAS had been inhibited; thence a comparison with the NG stall, a certification consideration for the failure case.

MCAS was also driven by the Boeing focus of type commonality, minimum differences training; thus a stick force tweak to achieve a 'NG like' feel (#10).

FBW might not have resolved the issue(#4). Electrical signalling was not required; the associated computation and sensor redundancy could not be justified - why not add third AoA to the existing 'mechanical' system.
Also, the Boeing FBW philosophy of keeping the pilot in the loop - stick force feel - negates the auto trim follow up. The C*U algorithm is based on trim for speed, which would not provide an additional stick force change to rectify the deficiency in manual flight. Hence the need for separate computation to drive the trim directly - MCAS.

Awww man, this statement bugs me:

The MCAS implementation was not some gentle input to the existing column "feel". Good grief. It put in gobs of nose down trim and did it for "x" bumps and then did it again if the trigger parameters still existed. It was not well thought out and was poorly implemented, at that.

If the MCAS could make the Max "feel" like the NG, then I guess the STS in the NG is very harsh, as that system seems to be another kludge to make the pilot do something that should be part of basic trim action by the pilot as the plane increases or decreases speed. GASP!

We need to stop making excuses and simply admit the MCAS was a poor solution to an aero design that changed the stick force gradient so much that the jet would not meet the criteria.

Gums sends...

P.S. FBW computer solution is not the answer without a sh!!!tload of waivers from the certifying authorities. The 'bus is a very stable plane, as the AF crew demonstrated. They didn't even recognize that they were stalled! The military planes using FBW use it for variour reasons, and not all are to correct for a poor aerodynamic characteristic. As with the 'bus, FBW computer flight control systems can reduce workload and smooth things out (things we old farts did without HAL doing a lot except autopilot funtions when in level flight)

For an aircraft that is only manually flown for the best part of 30 seconds per sector, is there really a justifiable need for FBW?

Would you not say the opposite is true?

I fly a 737 manually for a lot longer than 30 seconds every sector...
And FD and AP are MEL items.

Back to your Cessna 150.

Swatting bugs around a cowpat
 

Yo gums, #13 Let's swat some 'bugs'
The design concept of MCAS was fine. It met the objectives - as misguided as they appear to be with hindsight.
However, it was the engineering implementation that was flawed, together with weak and failed processes in testing, airworthiness, and certification; all of the holes in the Swiss Cheese lined up - they were already there just waiting for MCAS (it could have been something else).

The system operation which you describe was a failure case, not normal operation.

Yes, in hindsight, MCAS was not an ideal solution, but alternatively if the system has been well engineered, tested, certified, etc, then we would not be having this discussion - MCAS could have been hailed as a neat low-cost fix for an old design pushed too far.

Whilst the rule-makers and regulators fix, patch, kluge, the holes in regulation - swatting bugs around their cowpat, the greater concern is that they might not see the 'cowpat' which they are still standing in; that should 'bug' us.

Hindsight; 'sometimes things go wrong, so that we will know when they go right'.
Aviation assumes that the wrong things will be minor, manageable by limited capability humans, but sometimes those assumptions are wrong.

John Leahy ex CCO of Airbus said A380 was undone because when they launched in 2000 the engine manufacturers offered them engines which according to them was the most economical and anything better was atleast ten years away. But in three years the engine manufacturers turned out new engines for B787 which was 12% better in SFC. B737 was also undone by new engine technology in a different way. The 1960s low wing aircraft was unsuitable for high bypass engine era. Had Boeing anticipated this they should not have made 800 series itself but built a new aeroplane which being later generation aircraft may be even outlasted Airbus neo. Unfortunately they had gone for widebody 787 and by the time it turned profitable Boeing was left with no gumption to invest in another new aircraft. They hurriedly did another plastic surgery on the old lady which turned out very nasty and apart from financial damage has even destroyed their credibility and reputation.

Far being for me to debate a certified test pilot with all the degrees and actual diplomas from the "schools".

I cannot find the "#13" ref document.... I gave up looking for technical, non-political, non-management and so forth once pPrune closed discussion here on the Tech Log. Whatever document had "#13" ? It must have been buried in all the "return to service or certification" threads that were not worth the time to look at due to legal and non-pilot posts. I tried to get tech stuff on Tech Log to no avail.

I do not take completely agree with:

I agree that moving the stab would change the control input gradient in order to require more aft stick for more AoA or whatever. It would not necessarily require more pounds per stick movement unless it influenced other systems in the control implementation. I do not agree that it was an acceptable concept to make the plane "feel" like the NG model. So meeting the "objective" of the Part "x" that FAA required seems to me to be the biggest bone of contention I have. How would moving the stab angle of incidence and then remain there at other parts of the envelope? "Oh, Gums, we just return to some other angle when ?" What angle?
I appreciate the implementation of moving the stab "gradually" to help with "elevator" pitch authority/movement. But my experience was with three jets that did not have "elevators" (and one that was a delta with elevators/elevons). We moved the stab for pitch control! In the "neutral" position, we followed basic FAA stability and control requirements. On autopilot, they trimmed the slab very slightly to maintain either mach hold or altitude. When we pulled hard it got harder approaching higher AoA, except for the last few degrees in the VooDoo. In the VooDoo we definitely got a "light stick" in the end game, and I wish you could have flown one to see what I mean.
----
and then I completely agree with:

You nailed it there, PEI. And do not forget some management issues that Big B is now paying for.

Actually, Boeing was planning a completely new aircraft to replace the 737 (part of, IIRC, was called 'project 2016'). Problem was when Airbus announced the A320 NEO, it caught Boeing completely off guard. Not only where they resource limited (both 787 and 747-8 still a year away from type cert/EIS), if Boeing had launched the new aircraft immediately, it would have been five to six years before initial cert. Worse, both Boeing and Airbus were pushing 737s and A320s out the door at ~50/month - it would take another four to five years to get a brand new aircraft up to that sort of production rate. It would have meant giving the A320 NEO a near monopoly and the market for about five years of production - roughly 3,000 aircraft. That was too bitter of a pill to swallow, so Boeing cobbled together the MAX - which kept the A320 NEO advantage down to around 12 months.
With the benefit of 20-20 hindsight, doing a completely new aircraft looks much better than what happened on the MAX, but no one could foresee that 10 years ago...

BTW, the A380 had much bigger problems than it's engines (although they didn't help) - even now Leahy is unwilling to admit he screwed the pooch when they launched the A380 just as the market was moving to big, long range twins.

I've posted this before, but the problems with MCAS all go back to two related assumptions. First, that an MCAS malfunction was no worse than a stab trim runaway, and two that the crew could recognize and accommodate such a failure in four seconds (this second assumption pre-dating the MAX, possibly going back to the original 737-100/200). Based on these assumptions, an MCAS malfunction was judged as no worse than "Major" - same as a routine engine shutdown - and single failures are acceptable.
Turns out that both assumptions were wrong - four seconds to recognized and correct for a stab trim runaway was too optimistic (at least for some pilots), and it was not recognized as a Stab Trim malfunction. Had MCAS been originally identified as a Critical system (as it is now), it never would have been designed the way it was.

OT: With some new from scratch design Boeing would have been able to match or top the A321neo.

Wonder how much longer "blancoliori" will continue to defend Boeing's negligence and tortious actions? Hiding the existence of MCAS until after the first of two crashes? Sounds legit.

Thanks, TD

The assumption that MCAS loss of sanity would be seen as runaway trim is one thing that bugs me. Most runaway trim would not be corrected by beeping the trim switches on the yoke, unless there is alot of folks out there that never had to trim using switches on the stick or yoke.

I do not put a lot of credence to the four second assumption of a big problem because the MCAS activation stopped briefly if the pilot just beeped the trim switch. Added to the implementation was undocumented changes to the trim cutout switches. Oh oh oh, how about telling the pilots that they had this thing called MCAS? How about telling them what it was supposed to do? As a pilot, I am pissed about the hold thing, and feel betrayed.

I had my share of "firsts" in new planes, and some hairy moments in others. But this MCAS implementation in both aero design and procedural implementation simply scares me. Imagine if you were on nuclear alert back in the sixties and they changed the pre arming and actual arming of our weapons without telling us and having us practicing a few sim missions?

I relinquish the floor.

Gums
Shortly after the second 737 MAX crash, I was at an event at the Seattle Museum of Flight where I ran into some old friends from Boeing (including a test pilot) who'd been involved in the MAX. One thing they all agreed on was that they'd all assumed that MCAS would be covered in the differences training from the NG.
That it wasn't was admittedly bad, but may not have made a whole lot of difference in the long run - the Ethiopian crew certainly should have known about MCAS yet still couldn't handle it.

gums, re your post #19,
Apologies for my my cryptic #13 which was a reference to your previous post '13' in this thread, minimising thread clutter with endless quotes.

We are dancing around the same pole; but without the details of the various factors which could have influenced the goals of the people involved we cannot know their reasoning.

A minor deficiency in certification stick-force requirement could be argued with the regulator, normally the first choice - low cost. We don't know the extent of the deficiency, likely the wrong side of the line with Mach effects.

However, the second goal - NG compatibility, minimum training, then at $1M per aircraft there would be a strong incentive to improve the MAX. Particularly where the cost of MCAS as an extension of the existing STS would be small beer in the overall certification costs.

Amongst these goals it is feasible that MCAS might not be required for certification, but for commonality - training, MCAS was essential. A typical situation of conflicting goals, commercial pressures, and time constraint, cobbled together and thrown at the regulator.
Maybe the 'MAX did not required MCAS, but it had it.'

Yes Boeing are paying for this misjudgement - hard cash; the FAA can only 'pay' with loss of reputation, credibility, and worldwide trust. Which of these two, Boeing / FAA, will be back in 'profit' first ?

Flown the F101 Voodoo?! But for the toss of a coin I would have. However, the long-stab characteristics were closely studied, being typical of that generation of transonic aircraft. Extensive experience in 'the' Lightning which had similar, mild characteristics, showed how tolerant or unaware pilots can be of adverse characteristics.
These aircraft required 'MCAS', but didn't have it.

I fly the 737 too actually, the only hand flying we ever do is to 1,000' AGL and it comes off at the 500 foot stabilisation gate... what's the point in having FBW for stages of flight where everything is already trimmed out and all done in straight lines?

For all to see, PEI and Gums can discuss some of the fine points about design and aero without calling names and raising cultural aspects of aviation.

We both share some "exciting:" times in various planes, and are still here to talk about it.

I am not sure PEI has as much experience training and learning from foreighn countries as I have, but I can unnerstan those here who have flown with many of the foreign carriers and all that that involves.

I sure hope we can contribute to a great resolve not to repeat the MCAS debacle.

I think this has all been done to death on other threads but it is worth reiterating the simple facts.

- MCAS was created to meet certification requirements of control column pitch force at the extreme edge of the performance envelope.
99.999 % of Pilots in normal operations would never expect to get anywhere near those conditions in 99.999 % of their total time on the aircraft.

- The inadequacy of the design was exposed ONLY after two subsequent events occurred;

- an angle of attack sensor failure
PLUS
- the incorrect application of the Non- Normal checklists expected by Boeing to resolve the issue.

Boeings assumption was that the procedures it had in place were adequate to mitigate the risk.
Subsequent events proved them wrong.
But not unreasonably so.
Took lots of cheese slices to line up to reach the outcome.

Boeing had a NNC to address MCAS failure ?

MCAS was necessary though....
 

From my reading of the Boeing email in the AirCurrent article, flight test data indicated that there was not sufficient warning to the pilot that one would be entering a stall condition. The issue thereby appears to not be one of control column feel but instead that the aircraft must provide tactile feedback that a stall is imminent. If this is the actual physical behavior of the aircraft than by all means MCAS would be required.

I always wondered why it took until flight test to identify the need to use MCAS in low speed flight. This observed behavior would explain why wind tunnel data and simulation would not have predicted this beforehand.

With regards to the criminal citation, my understanding of the law is that even if failure to comply is an honest mistake, e.g. Forkner did not know of the decision to change functionality, that it would still be criminally prosecutable. This would explain why the Company is criminally liable but not the individual. It would also explain why the criminal penalty was not as severe in only ~$200 million.

Does this seem like a reasonable interpretation?

Yup.

Initial failure of Angle of Attack sensor leads to NNC “Airspeed Unreliable.”

Continued motion of Stab Trim leads to NNC “ Runaway Stabilizer.”

Challenging, but doable. Fly thrust and attitude.
But if you leave the autothrottle engaged at T/O thrust , try to re-engage the autopilot AND try to re-engage stab. trim you are screwed.
As a matter of interest , the flight data recordings from the initial Ethiopian report seem to have disappeared from the internet.
Why ?

However it used to be some intermittent motion of stab trim. Starting over and over again. MCAS had been kept unknown to the pilots. And those AoA indicators were not displayed as advertised.
Not only the training got changed but the hardware and software.

The question was did Boeing had NNC for MCAS? When MCAS itself was not disclosed how will they have a procedure for its abnormality?

I'm not sure exactly what you mean - the Preliminary Report, complete with the FDR traces that were widely discussed here on PPRuNe, is still downloadable from the Ethiopian CAA website.

For the record, as well I find G.G.'s summary markedly off the centre-line. That's from a person who agreed on all observations made by 737driver and fdr. For the latter it is certainly not a surprise, however, the former had received a lot of friendly and undeserved fire IMHO. With so many lives lost unnecessarily, the aim for everyone is taking as many lessons as possible for their own job.

At the edge of the envelope, an aeroplane must behave in a predictable manner and provide tactile resistance of sufficient and increasing magnitude.
Encountering unreliable airspeed event at take-off, a pilot shall assure for the necessary power + pitch to continue climbing safely.

Each of them being correct does not make the other one less true. Two gas pedals, two feet.

Is it ?
Last time looked the link wasn’t there.

For those that are wondering I have personally experienced an “ Airspeed Unreliable” event in flight.
After the initial “what the f#ck “ moment the NNC worked just fine.
It will NOT work if you leave the autothrottle engaged.

My understanding is that MCAS was certified on the basis that the existing checklists and procedures would be carried out by a properly trained crew and would mitigate any risk.

Clearly they were wrong.

Aircraft Accident Investigation Preliminary Report Ethiopian Airlines Group B737-8 (MAX) Registered ET-AVJ 28 NM South East of Addis Ababa, Bole International Airport March 10, 2019

I thought that most of us had agreed over this one but the trigger for the runaway trim NNC was “continuous uncommanded movement”. For those who haven’t flown the 737, the trim moves quite a lot in bursts in normal operation, without a command from the pilots, as it is driven by STS, MCAS on the MAX and the autopilot. How could you tell that the trim moving then stopping was a) normal operation or b) a dangerous system failure? Answer: you couldn’t, without (secret) prior knowledge and continued observation. If you disconnect the trim every time it moved under its own accord you might as well pull the circuit breaker on the ground. It’s many years since I flew one but I still remember the “clunk clunk clunk” which was the most common sound on the flight deck.

It was asking too much of pilots dealing with a complex failure plus tactile and aural alerts, during a critical phase of flight where the trim was actually responding to their inputs, to immediately diagnose a failure of a system that wasn’t even documented and work out what to do. We now know that the sim trials were unrepresentative due to prior coaching.

Just a small point on the engine issue. The Airbus A320 and 321 NEO’s suffer from a similar if not the same issue and it’s actually worse on the Airbus because the aircraft could in fact become unstable. This was not caught until the aircraft was in service and it required the CG be restricted by blocking the last seat rows on many configurations. They were attempting to address this and remove the CG restriction via software. I am not sure if this has been completed. MCAS was primarily designed to insure the aircraft handled like the 800 and would not require any pilot simulator training for the transition. In fact Boeing promised SWA that no simulator differences training would be required. SWA wrote a 1 million dollar per aircraft penalty into their purchase contract if even 1 sim session would be mandated by the FAA. The rest is history.

A B737 simulator instructor at a carrier I used to work for used to do an exercise when we had some time left over.
He’d position us at about 7nm on final to a lit runway , at night , in flight freeze , and then turn all the lights off on the forward panel.
When we were happy he’d let us go.
After a while he would stop the sim. and ask us how we thought it was going.
”Bit fast and a fraction low” I said.
”Correct. Continue”
After a normal landing we discussed what had been an interesting exercise.
It demonstrated several things;
1) How well thought out the Boeing cockpit is and how visual cues and muscle memory can get you a long way.
2) That old f@rts like myself know more than we think we do , just by virtue of years of experience.
3) What a confidence building exercise it was for new F/Os , particularly those coming from Airbus , as to what the aircraft can actually do.

Another exercise was to take the aircraft to 20,000 ft and turn off every unguarded switch on the overhead panel.
The aircraft still flies.

The point is that the B737 is an analogue aircraft with lots of nice to have , but not necessary , digital add ons.

Reading all that has been written about the MCAS debacle I now realise that the B737s time has passed , not because it is fundamentally unsafe , but because the world has moved on. I don’t think its going to far to say that the time of Piloting as a highly regarded , valued profession is over.
The accountants have won.
Airbus philosophy has won.
Time to retire the B737 ASAP.
Very sad.

I not sure MCAS was hidden. The Brazilian Authority had a differences document for the MAX that listed MCAS as a cat B training item. Forkner was listed as a contributor to the document. Why other states authorities did not have MCAS listed I don’t know. But it would be interesting to see what the Brazilian training involved.