Ethiopian airliner down in Africa

[Cows getting bigger]

One of the problems with a safety system is that when you find an omission/error/oversight/process concern you (should) start to wonder what else has gone wrong or been missed. if I were the FAA, I would be taking a big sample of other sign-offs and having a really good look at them. This MCAS affair has cast doubt on the integrity of a Organisation's Management System; that's going to take a lot to address and far more than a software tweak.

[HarryMann]

Originally Posted by Dark Knight

Why if one has a significant problem such as may have been the case why allow the aircraft to accelerate to 382kts?

Wouldn't an experienced crew at least reduce thrust to maintain around 210-250kts and a safe altitude attempting to get things under control?

DK ... Speed isn't all about thrust alone. It is climbing that controls speed in the early stages of flight once flaps and xart are cleaned up.
if somethings trying to pitch you into the ground it isnt your first course of action.

[FCeng84]

Originally Posted by BRE

The Seattle Times article says FAA has briefed lawmakers that the software fix will:
- use input from both AoA vanes
- limit travel
- limit retriggering

Sorry if this has been covered before: how do the MAX of SWA and AA differ? Do they just have an AoA display or do they actually have an extra vane, and their MCAS is already configured to use that extra input?

Difference is display only. 2 vanes and exactly the same MCAS functionality as all other 737MAXs.

[wiedehopf]

Originally Posted by DaveReidUK

Interesting graphic.

The altitude spike where the aircraft dips below the runway surface(!) accompanied by a single, instantaneous 2000+ fpm ROC value is clearly an artifact (unless Newton got it all wrong), probably coinciding with rotation.

This altitude dip below 0 is connected to erroneous high AoA readings.
It can be found in the altitude graph of the FDR readout of the Lion Air accident flight.
Check page 14 of this document:
https://www.flightradar24.com/blog/w...ary-Report.pdf

[infrequentflyer789]

Originally Posted by FCeng84

Thoughts on Seattle Times article published Sunday morning Seattle time

This article is generally well written and seems quite accurate.

Interesting as I think Boeing described it as "mis-characterisation"... I thought (having less other info to corroborate it, I suspect) it was plausible and made sense, your clarification makes a lot of sense too.

Question: If MCAS has (sort of) a speed-dependent "gain", what value does it use when airspeed/mach is unknown (or known-bad)? Does it, perhaps, use the higher value (more trim) to be "safe"?

The accident flights and the made-it-back Lion previous flight all reported unreliable airspeed as first problem. I don't think airspeed is voted on 737, although there are three pitot/statics one set is used for standby only so at the FCCs airspeed must be either known (two inputs agree) or unknown (disagree). Have I got that right?

[CodyBlade]

Originally Posted by wiedehopf

This altitude dip below 0 is connected to erroneous high AoA readings.
It can be found in the altitude graph of the FDR readout of the Lion Air accident flight.
Check page 14 of this document:
https://www.flightradar24.com/blog/w...ary-Report.pdf

Must interpret the graphs with some common sense eh.

[Aloha_KSA]

Boeing has been here before with LSAS...

Originally Posted by WHBM

Now those with long memories may recall another overstretched aircraft, the MD-11. That, too, got a "clever" bit of software to overcome design aspects. LSAS. Longitudinal Stability Augmentation System. Even the name sounds similar to MCAS. Did one of the onetime McDD engineers who stayed on after the Boeing takeover have anything to do with the more recent concept.

And again, those with long memories will recall that the MD-11 had a hull loss rate substantially out of kilter with norms, and was well known for ending up on its back and burned out alongside the runway on landing, exactly at the point where LSAS had been designed to kick in.

There were even discussions about it on PPRuNe at the time. One who seemed to understand its technicalities wrote "I seriously wonder if the FAA would be as accommodating now". Hmmm ...

Md-11 Lsas

The parallels to the MCAS and the design philosophy of attempting to make an airframe with errant handling characteristics manageable by using software are stunning. What is also stunning is the (1) lack of training on the MD-11 to handle related issues, and (2) the FAA's complicity in the problems.

In a nutshell *Boeing and the FAA have been here before and did not learn from their mistakes.*

As I read some of these MD-11 accident reports, and the author's comments in light of the recent Max incidents and accidents, my jaw literally dropped: The McDonnell Douglas MD-11 Accident History

[averow]

This, exactly this!

Originally Posted by Cropduster

I think Boeing is probably facing a potentially bigger issue than the failure of the MCAS system. The bigger issue is that it is becoming increasingly apparent that the FAA failed to provide adequate oversight of delegated certification functions. What Boeing really needs to fear (apart from the punitive damages from the lawsuits) is that the airplane will have to undergo a complete re-certification. And even if senior FAA officials are satisfied, I am not sure that other world authorities, especially the European JAA are going to take the FAA's word on it this time.

It would be a nightmare for Boeing if this were mandated. And yet without MCAS it seems that the 737 Max would not pass muster, in terms of stick and handling behavior. The kludge of using a software loop to make up for inadequate aerodynamic behavior has bitten Boeing very hard and may have lead to these two tragedies. There may need to be a complete rethink (and rebuilding) of the 737 Max.

[wiedehopf]

Originally Posted by CodyBlade

Must interpret the graphs with some common sense eh.

Is this ridicule?
I can't quite tell, please elaborate!

Anyway i've enlarged the Lion Air 610 FDR graph a bit so maybe you can see it as well.
The dip in this image is about 100 to 150 ft, comparable to the dip on rotation we see in the ADS-B data from the Ethiopian flight.

It is only present for the left altitude readout, which was corrected by the faulty left AoA.
FDR graph magnificiation take from preliminary Lion Air 610 report:

[sky9]

A/P autoland needs three a/p channels, why isn't the same criteria used for the AOA indicators? What they really need is 4 AOA vanes so they can dispatch with one unserviceable.

But then using software to correct an inherently unstable design at the extreme of its envelope shows how far the industry has gone away from safety first.

there are only 2 channels on a 73

The 757 and 767 both have 3 channels.

If there are only 2 channels on a 737 A/P how does it do a CAT3 A and B autoland or is this another delegation of certification by the FAA?

[Sailvi767]

Originally Posted by HarryMann

DK ... Speed isn't all about thrust alone. It is climbing that controls speed in the early stages of flight once flaps and xart are cleaned up.
if somethings trying to pitch you into the ground it isnt your first course of action.

Climbing controls speed? Only if you leave the thrust at a high power setting! Yes they should have adjusted power to maintain a reasonable speed. It’s called piloting!

[.Scott]

Originally Posted by El Bunto

With the 737 Max orderbook clocking-in at well over $600 billion I don't think a few million spent for a patch-up fix will matter much.

We don't know how long the MAXs will be grounded. Perhaps 3 months. Aside from lots of missed new orders, what will that cost Boeing?
But I agree, the reengineering and deployment of the fix will not be financial killers. The most financial damage will likely come from the time it takes to work through the process.

The problem with the fix for this is not in figuring out what to do. From the time the flight data becomes available to the time they have a specific design and plan shouldn't be more than a week.
And the implementation should only take a week as well. But then they need to demonstrate that the what they have done works, addresses the problem, and doesn't introduce other problems.
Those involve generating more test procedures, test plans, test execution and documentation, and reviews at many levels.

It appears that there is a close connection between what happened at Lion, so they have some head start. But that also implies a flaw in their reasoning: if they fully understood what happened with Lion, why didn't they ground the MAXs until the fix was deployed? Bad luck? There is only one palatable answer, they did not understand the gravity of the problem - which puts that original fix into serious doubt and scrutiny. If the fix prepared for the Lion was not based on a full understanding of the problem, then it is not going to pass the scrutiny that it must go through before it can be deployed - and especially deployed and deemed sufficient to to restore the MAXs airworthiness.

If the reviewers do their job, this is what will realistically happen: Those at Boeing, FAA and others will be reviewing this change. But before they even look at the change, they will want to look at the design, development, and testing histories for the systems that are involved. They will need to do that simply to make themselves expert enough to understand how this process broke down. Then they will look at the process that was use to design, develop and test this change - and if they do not find reason for increased confidence - the will send the entire process back to Boeing.

[Sailvi767]

Originally Posted by Aloha_KSA

The parallels to the MCAS and the design philosophy of attempting to make an airframe with errant handling characteristics manageable by using software are stunning. What is also stunning is the (1) lack of training on the MD-11 to handle related issues, and (2) the FAA's complicity in the problems.

In a nutshell *Boeing and the FAA have been here before and did not learn from their mistakes.*

As I read some of these MD-11 accident reports, and the author's comments in light of the recent Max incidents and accidents, my jaw literally dropped: The McDonnell Douglas MD-11 Accident History

You could add training pilots to fly the A330 in the various laws with the CG near the aft limit at high altitude.

[Aloha_KSA]

Originally Posted by .Scott

We don't know how long the MAXs will be grounded. Perhaps 3 months. Aside from lots of missed new orders, what will that cost Boeing?
But I agree, the reengineering and deployment of the fix will not be financial killers. The most financial damage will likely come from the time it takes to work through the process.

At least two other major contributing factors must be addressed as well. And they require that both the FAA and Boeing admit fault.
(1) Why was MCAS not described in sufficient detail in the FCOM/ AFM?
(2) Why was there no training mandated for malfunctions of this System? And not just the failures that potentially caused these crashes, but what about failure of the MCAS system entirely? In that case the aircraft would handle differently, would it not? Given the lack of redundancy in the AOA sensors and possibly elsewhere in the MCAS system, MCAS failure/ inhibition should be a training event, or at least included in the ground school syllabus.
So far Max pilots have been given about 4 hours of CBT and a firm hand shake as their differences training. That will obviously have to change.
Even when the aircraft is deemed “fixed”, (1) the manuals will have to be updated and approved, and (2) the crews and engineers will still have to all be properly trained. That will take some time, as that training cannot commence until the problem is officially identified.

[Discorde]

Speculative QRH excerpt v 2.0:

[GarageYears]

Activation Limits?

The MCAS function becomes active when the airplane Angle of Attack exceeds a threshold based on airspeed and altitude. Stabilizer incremental commands are limited to 2.5 degrees and are provided at a rate of 0.27 degrees per second. The magnitude of the stabilizer input is lower at high Mach number and greater at low Mach numbers. The function is reset once angle of attack falls below the Angle of Attack threshold or if manual stabilizer commands are provided by the flight crew. If the original elevated AOA condition persists, the MCAS function commands another incremental stabilizer nose down command according to current aircraft Mach number at actuation.

I have seen the above description of MCAS function in multiple places, but cannot find the original source, so apologies for the lack of attribution, though I think it came from Boeing.

Can ANYONE confirm there is an ALTITUDE 'limit' on MCAS operation? It has been stated that MCAS is only operational when the flaps are retracted (confirmed by Lion Air data) and AP is NOT engaged (i.e. manual flight).

The statement above includes a 'threshold' that must be exceeded in terms of both altitude and speed. Anyone got anything to confirm this?

- GY

[FGD135]

Speculative QRH excerpt v 2.0:

Ahem.

Discorde, nice try, but it looks fishy. It looks "Flight Simulator-ish". Please stop posting that.

Aloha KSA, how about you read some earlier pages of this thread? All of your questions are fully explored there.

GarageYears, have a read back a few pages. You will find that the word "altitude" is probably a typo. "Attitude" was the intended word.

[Busbuoy]

I think that that statement means that the threshold value of AOA above which MCAS becomes active varies and that that AOA value is dependent upon altitude and airspeed.

[Judd]

It seems both aircraft were porpoising before going in. Is it possible the crews were trying to overcome excessive aerodynamic loads on the stabiliser which had gone towards the forward limit, after cutting electrical power to the stab trim motor and were now trying to wind manual trim during the unloading maneuver called roller coasting? See extract from a Boeing 737-200 PTM date 1982 and migrated from Tech Log
Extract from the Boeing 737-200 Pilot Training Manual February 1982 page 04.80.31. Edited for brevity
Runaway and Manual Stabiliser - Recovery from Severe Out-of-Trim
"In an extreme nose-up out-of-trim condition, requiring almost full forward control column, decelerate, extend the flaps and/or reduce thrust to a minimum practical setting consistent with flight conditions until elevator control is established. Do not decrease airspeed below the minimum maneuvering speed for the flap configuration. A bank of 30 degrees or more will relieve some force on the control column. This, combined with flap extension and reduced speed should permit easier manual trimming.

If other methods fail to relieve the elevator load and control column force, use the "roller coaster" technique. If nose-up trim is required, raise the nose well above the horizon with elevator control. Then slowly relax the control column pressure and manually trim nose-up. Allow the nose to drop below the horizon while trimming. Repeat this sequence until the airplane is trim.

[Luc Lion]

I haven't seen it posted, apologies if it is a duplicate:

From BEA | Bureau d'Enquêtes & d'Analyses, via Twitter

Accident survenu le 10/03 du #Boeing737Max @BoeingAirplanes @flyethiopian / Les données contenues dans le FDR ont été téléchargées avec succès par @BEA_Aero et remises à l’équipe d’enquête éthiopienne / communication en leur nom / 7:43 AM - 17 Mar 2019

It states that the FDR data have been downloaded and transferred to the Ethopian investigation team.
The CVR data have been downloaded and transferred the day before, on March 16th.

And the Ethiopian Transport Minister, Dagmawit Moges, held a press conference on Sunday, stating
"Clear similarities were noted between Ethiopian Airlines Flight 302 and Indonesian Lion Air Flight 610, which would be the subject of further study during the investigation,"
Details here:
https://www.bbc.com/news/world-africa-47605265

I understand that some data revealed by the CVR (or by both the CVR and FDR) confirmed the similarities between the 2 flights.

A video of the press conference can be found here

Facebook Post

The language appears to be Amharic.