Ethiopian airliner down in Africa

[Chesty Morgan]

Originally Posted by OldnGrounded

. . . I wanna fly with reamer, not one of those ordinary pilots who are affected by work overload, startle factor, baffling PFD readings and alarming acft behavior, at high speed and low altitude. I mean, who wouldn't want to?

Simply writing a few words saying do it this way does not make anyone a better or worse pilot than anyone else.

Trying to understand every implication, cause and problem of a particular accident does make you better. As does trying to see this from the point of view of those dead pilots.

[RickNRoll]

Originally Posted by Chesty Morgan

Simply writing a few words saying do it this way does not make anyone a better or worse pilot than anyone else.

Trying to understand every implication, cause and problem of a particular accident does make you better. As does trying to see this from the point of view of those dead pilots.

irony alert! Don't take what he wrote literally.

[OldnGrounded]

Originally Posted by RickNRoll

irony alert! Don't take what he wrote literally.

Uh, right. That said, Chesty's contributions are polite and level-headed; he may expect the same from the rest of us -- which shouldn't be an unreasonable expectation.
I apologize. The "any good pilot [engineer/physician/whatever] would have . . ." attitude just really annoys me. And I'm pretty sure it doesn't contribute to safer operations in any discipline.

[GlobalNav]

Originally Posted by DaveReidUK

You're probably right.

But what's really needed is an in-depth look at the (changed) relationship between the regulators and the industry (by no means confined to the USA).

Maybe. But so much of the certification work is done by designees (Boeing employed) and the truth is that the FAA executives who put that system in place are mostly retired, some employed in the private sector.

[Chesty Morgan]

Sorry OldnGrounded I misunderstood.

Still fuming about the Calcutta Cup, that's my excuse and I'm sticking to it.

[StuntPilot]

My thoughts about MCAS as a hardware designer:

1. Aircraft are nonlinear/unstable systems that can only be stabilized by control laws in a small (linear perturbations) part of the parameter space. A deep stall is an example of non-linearity.
2. Complexity explodes exponentially with state (autopilot mode, AOA vane failure detected etc.), an important design goal is to reduce state. State includes any if... then... in software.
When software initiates a state change on its own (autopilot switches off, systems disabled because of a broken sensor, stall recovery deployed) this should be announced to the pilot by aural warning. A pilot should always know exactly what the control state of the aircraft is, 'what is it doing now?' is not a good thing to wonder about up in the air.
3. There is unavoidable state that relates to the physics of flying: flaps, trim, gear = configuration. If automation is allowed to mess with configuration there must be cutouts and self-checks (cross checks against other sensors, stick position, whether data is consistent) to prevent instability.
4. MCAS has a very specific control function in a specific part of the flight envelope. It is easy to cross check with other sensors whether this part of the flight envelope is entered and how large a control input is required.

It puzzles me that, at complete odds with this, MCAS was given an integrating control function without bounds on a crucial flight control surface. Without data validity check. Without aural deployment warning. Without aural and visual AoA disagree warning. Without control input cross check. Without adiru cross check.

With safety critical subsystems there should always be at least 2 barriers before handing things over to a human as the last line of defense to prevent an incident.

[Dave Therhino]

Originally Posted by dingy737

Thanks. Therefore and for argument sake, if these incidents were in fact “runaway stabilizers” we could expect the same tragic result. MCAS should have been interpreted as a “ RUNAWAY STAB. “ for which pilots are repeatedly drilled in the simulator.
MCAS: parameters for operation.
1. Autopilot OFF
2. Steep Bank/ Turn. ( what angle ?)
3. Flaps retracted.
4. High AOA
If the above 4 are correct, how many of those parameters were NOT met in Ethiopia. Nevertheless, prolonged trimming with the AP off instinctively requires stab switches OFF. I believe with the flaps NOT retracted, the Trim rate is increased as compared to flaps retracted. A runaway trim with flaps extended will very quickly overpower the pilot, the time frame allowed to identify and react appropriately, by selecting both switches OFF is extremely short if this occurs just after lift off. Once the pilot starts fighting the Yoke with both hands I suspect his mind will become focused on the physical deminishing the ability to keep thinking.

While MCAS was required to address column force characteristics in a turning flaps up stall maneuver, its logic does not include any consideration of bank angle or any other detection of turning conditions.

[slip and turn]

Originally Posted by Chesty Morgan

Dear Mods, any chance you could give the altimetry geeks their own thread? It's been done to death and we don't need to see reams of calculations cluttering up the technical stuff.

As one such 'geek' (if what happens in modern cockpits doesn't actually need much human mathematical verification?), I don't entirely disagree. I do well appreciate most PP readers here do want to try to understand the foibles of AOA sensors and stab trim runaway with MCAS, and also to be able to scope the data inputs to MCAS so they have some idea of "what's it doing now". PPs will want particularly to be able to answer the related big questions about how MCAS really behaves armed with dodgy AOA and other data, and how to disarm it.

But the reason perhaps why some of us are questioning altimetry may well be because we see very low altitude reports from the outset, and wonder if this accident began with something else entirely e.g. an error during the take-off roll.

Fact is, unless we toss in the benefit of a couple of hundred fictitious feet for doubt, there otherwise seems precious little evidence from the tele-altimetry - which looks very consistent and accurate in so many respects all the way up to rotation - that the aircraft then left the ground at all normally.

If there was a static pressure increase at rotation (if that is a known feature of the aerodynamics at high AOA leading to the "below ground" 05:38:47.714 altitude of 7075') it would affect cockpit IAS, and VS indications too, would it not? And if that is evidence of an aerodynamic anomaly, and it is not a spurious measurement, then might other static pressure anomalies exist at the static port positions at high AOA with this airframe?

That said, the first ADS-B altitude report that has to be a truly airborne report - because it may be the first that can be truly validated as being above aerodrome level - is timed at 05:39:04.028. Yet that one is well over the dirt which has sloped some 150ft down from the runway by that point. Putting full credence in ADS-B 1013 altitude data, and taking Luc Lion's reminder of 420' differential from QNH, a quick look at that particular datapoint using the "Viewshed" tool in Google Earth, raises a question as to whether a spectator standing behind the other end of the runway could even still see the aircraft. Was it indeed so low that it disappeared off the end of the runway to such a spectator at the 07R threshold end? The previous datapoint is 5 secs earlier at 05:38:59.102, also over dirt not tarmac, and located 1542 feet further back. There is nothing in any of the datapoints along the centreline of the runway to suggest that they are anything other than spot-on location-wise. The two data points indicate an average ground speed of 1542 feet (0.2538nm) in the 4.74 secs between them, or 0.2538 * 3600/4.74 kts = 193kts. The actual reported speeds transmitted as those two datapoints are 200 kts and 207 kts (not sure how they are computed onboard the aircraft before transmission - sourced from onboard GPS or from INS?).

And at that 05:38:59.102 datapoint - the first just beyond the runway - ET302 may have been 100ft (+ or - 25ft) above the by then already downward sloping dirt. The VS between the two points assuming static pressure was then lapsing in book expected linear fashion should have indicated (7300'-7225' + or - 25') * 60/4.74 ft/min = (100' or 50') * 60/4.74 ft/min = between 633ft/min and 1266 ft/min. The lower would seem to reflect what was given in ADS-B reports and that in turn may suggest the aircraft was indeed still fighting for altitude down in the weeds.

IF it was, why was it in the weeds in the first place? It was still only 12-14 secs after initial signs of a rotation. MCAS doesn't explain weeds so soon on its own, does it? Did something happen first on the runway, which has not been much discussed (if something did happen), and did it line up a couple of holes in the cheese? And then did MCAS offer the third or fourth bad thing and the eventual gotcha? Or was the take-off normal and only afterward went pear-shaped?

[4runner]

Originally Posted by reamer

It's supposed to when the autopilot is on and after takeoff when the autopilot is off and the flaps are not up.
When you are reading erratic indications on your instruments and it is consistantly trimming down, switch it off, you might save some lives.

And in cruise as you burn off fuel and your cg changed , and when the Mach transducer says so. It’s constantly moving.

[Sl4yer]

Originally Posted by reamer

Every pilot of the 737 has known since the Lionair crash that the first thing to do in this situation is simply switch off the mcas by cutting off the stab trim switches.

Wouldn't you get the aircraft back in trim with the yoke switches first?

Easy to say what you should do sitting at a keyboard.

[edmundronald]

Non pilot question: Would a despair measure of selecting flaps have disabled MCAS?

Edmund

[ferry pilot]

Old Electra and three holer driver here. In my time, America’s largest airplane manufacturers and the FAA were almost above reproach, making the current public flogging, deserved or not, hard to watch though not unique. The Electra was shedding wings but was not grounded, even after the second one, and the entire industry pitched in to find the cause, fix the problem and get it out of the headlines. There are many reasons why that will not happen now, but why the pendulum has swung so far the other way is a bit of a mystery. To the benefit of no one in aviation, feeding the media and keeping the story alive appears to be more important than encouraging those with a more positive agenda to converse and contribute. Airplane accidents are bad stories but great news, a fact that appears to have sucked us into laying blame before finding solutions. Just an observation, with no intention to overstate the obvious or offend anyone. I come from the past. We do things different there.

[wiedehopf]

Originally Posted by slip and turn

If there was a static pressure increase at rotation (if that is a known feature of the aerodynamics at high AOA leading to the "below ground" 05:38:47.714 altitude of 7075') it would affect cockpit IAS, and VS indications too, would it not? And if that is evidence of an aerodynamic anomaly, and it is not a spurious measurement, then might other static pressure anomalies exist at the static port positions at high AOA with this airframe?

https://www.flightradar24.com/blog/w...ary-Report.pdf

Diagram on page 14:

Large left side AoA overread leads to left altitude underread by an estimated 100-200 ft (no proper scale).
(AoA is a corrective term in the altitude and IAS calculation)

So that FDR readout combined with the FR24 altitude data from ET302 could point to the left AoA sensor reading high just like with LNI610.
This is speculation of course but the whole pattern of the flight just feels much too familiar.

A somewhat shallow climbout is also to be expected when the stick shaker is going off, instinctual reaction is to reduce climb rate.

[deltafox44]

Originally Posted by edmundronald

Non pilot question: Would a despair measure of selecting flaps have disabled MCAS?

Edmund

Normally, yes. But if done at 380 kt flaps would probably also fall into pieces :

[Loose rivets]

As far as we can see, in the first crash, they were deployed and then raised again. It may have been the 5 second time delay that masked the benifit from the PF.

By this I mean, they were lowered again, and for a few moments things looked hopeful. With limited knowledge of the system interconnection, it seems reasonable to raise the flaps again. When they were raised, MCAS had reset but the subsequent delay allowed the PF's mind to go on to other things.

[FullWings]

Originally Posted by edmundronald

Non pilot question: Would a despair measure of selecting flaps have disabled MCAS?

Edmund

Probably. But if you’ve reached the point when you’re pretty sure it’s MCAS gone AWOL, you’d be disconnecting the stab, not deploying flaps at 330kts+

As I’ve pointed out before, how do you know when you’ve got the trim running and it wasn’t you on the switches, that it’s a runaway or normal operation, especially during the cleanup. Is there a simple yes/no path you can follow? Seems not...

[Sl4yer]

Originally Posted by FullWings

Probably. But if you’ve reached the point when you’re pretty sure it’s MCAS gone AWOL, you’d be disconnecting the stab, not deploying flaps at 330kts+

As I’ve pointed out before, how do you know when you’ve got the trim running and it wasn’t you on the switches, that it’s a runaway or normal operation, especially during the cleanup. Is there a simple yes/no path you can follow? Seems not...

Assuming that you knew about the existence of MCAS in the first place.

[deltafox44]

Originally Posted by deltafox44

Normally, yes. But if done at 380 kt flaps would probably also fall into pieces :

If not, flaps would induce a sudden nose down moment which may be too strong to be countered by the controls

[CONSO]

Of some interest is the Document by Boeing circa year 2000 that suggested an AOA indicator in the pfd would be of assistance when airspeed went bonkers, etc see pages 19-20

of course such a display was/is optional at a no doubt ridiculous cost - and nothing was said that in case of ANY disagreement with pilot- HAL rules unless you quickly pull his plug

With the nonnormalized design, the
position of the needle is a function
only of sensed AOA. The red tick mark
for stall warning may behave erratically
in a pitot or static failure state, as may
stick shaker, PLI, and speed tape amber
and red bands. However, the AOA
needle and digits will remain stable,
and the indicator itself still will be
Improved situational awareness and
flight crew training.There is a desire
to use AOA information to increase
the flight crew’s understanding of the
physics of flight and their general
awareness of the state of the wing dur-
ing normal and nonnormal conditions.
Within certain limitations, the display
provides this indication in a clear,
unambiguous format. The degree to
which AOA can be used to increase
knowledge and airmanship depends, of
course, on the approach taken by the
airline in training its flight crews and
the use of the indicator in training sce-
narios for nonnormal procedures. Some
of the limitations are discussed below.
Pitot or static system failure
requires the flight crew to take several
fundamental steps to resolve the
problem (see “Erroneous Flight
Instrument Information,” Aerono. 8,
Oct. 1999):
Recognize an unusual or suspect
indication.
Keep control of the airplane
with basic pitch and power skills.
Take inventory of reliable
information.
Find or maintain favorable
flying conditions.
Get assistance from others.
useful as a backup for unreliable
airspeed,provided the AOA
vanes are undamaged.
A variety of potential uses for
AOA were examined during the
design of the new AOA indicator:
Improved situational
awareness and flight crew
training.
AOA backup indication
following pitot or static
system failures.
Reference during upset
recovery, windshear
escape, and terrain
avoidance maneuvers.

pages 19-20

And earlier page 12 explains proposed use on boeing models

A dedicated AOA indicator shown
on the primary flight display (PFD)
recently has been developed in
cooperation with airline customers.
The new indicator is offered as an
option on the 737-600/-700/-800/
-900, 767-400, and 777 at this time.
During the development of the
new indicator, discussions with
airlines, the NTSB, and U.S. Federal
Aviation Administration (FAA) pilots
and engineers provided a unique
opportunity to examine potential uses
of AOA and the many existing
uses that have evolved in recent
decades along with advances in
display and indication technology.
This article discusses the following:
1.Basic principles of AOA.Angle of attack
2.Airplane performance and AOA.
3.AOA measurement.
4.AOA indications and flight crewFlight path angle
procedures in current Boeing
production models.
5.Design and uses of a separate
AOA indicator.

http://www.boeing.com/commercial/aer...ero_12/aoa.pdf

[M2dude]

So much has been written here, a lot from some VERY learned folk of the aviation world, some of it justifiably emotive. I suppose at the end of the day we can only hope that some lessons will be learned by the industry, particularly here in North America. Although the DFDR readout is as yet not analysed it HAS been reported that the horizontal stabiliser jack-screw was found in the fully up/nose down position, just POSSIBLY suggesting that MCAS was again the culprit. So one vital lesson be learned would be that never again should a single source failure of a sensor (assuming we're talking about the Alpha vane again) be allowed to cause an aircraft to deviate from it's flightpath. Every aircraft that, as a licensed avionics engineer I've been seriously involved with, Concorde, B747-400 & 777-200/300, had very well designed systems indeed without this absurd fragility of a defective sensor, interface or wiring resulting in such a radical deviation from flight path. (MCAS not having an Alpha disagree inhibit really at best completely ABSURD). Another salient lesson HAS to be in terms of regulatory authority oversight, the FAA delegating such oversight to employees of the manufacturing company, Boeing, is really getting away from the whole idea of such oversight and is, in my opinion, inherently dangerous and certainly irresponsible. And let's all hope that never again shall such a radical design change as MCAS be implemented on any model upgrade without adequate training, including inclusion in the flying manual. Let's all hope that such lessons and more ARE learned.