Breaking news: Ethiopian Airlines flight has crashed on way to Nairobi

[pegasus]

A straightforward question. Is there a difference between the AoA and the attitude of an aircraft? If not, why can't the artificial horizon be used to verify AoA sensors?

[Black Ram]

So, no third ADIRU? No sdditional, backup AoA vane? Oh, but you get a light!

You thought a third ADIRU was coming?

[Black Ram]

A straightforward question. Is there a difference between the AoA and the attitude of an aircraft? If not, why can't the artificial horizon be used to verify AoA sensors?

There is a difference. AoA measures the airflow relative to the wing. It can be very different from attitude.

[Gabriel]

A straightforward question. Is there a difference between the AoA and the attitude of an aircraft? If not, why can't the artificial horizon be used to verify AoA sensors?

AoA and pitch attitude are not the same.

Pitch attitude is the angle pitch angle of the plane measured from the horizontal.
AoA is the pitch angle of the plane measured from the feestram (i.e. airspeed vector or relative wind).
Add one more angle, air flight path angle (the angle between the airspeed vector and the horizon), and you have that AoA = pitch attitude - flight path angle.

For example, if an airplane after take-off has a pitch of 15 degrees nose up and the trajectory is climbing at an angle of 10 degrees (no wind), then the AoA is 5 degrees.
If an airplane is approaching in the glide slope of an ILS that has a slope of -3deg and the airplane is pitched up 2 deg, the AoA is again 5 degrees.

As you can see, 2 very different pitch can give the same AoA.

[Black Ram]

What do you guys think of this? Talks about how MCAS and trim runaway are different, again, and then adds something new:

https://leehamnews.com/2019/03/22/bjorns-corner-the-ethiopian-airlines-flight-302-crash-part-2/

[pegasus]

AoA and pitch attitude are not the same.

Pitch attitude is the angle pitch angle of the plane measured from the horizontal.
AoA is the pitch angle of the plane measured from the feestram (i.e. airspeed vector or relative wind).
Add one more angle, air flight path angle (the angle between the airspeed vector and the horizon), and you have that AoA = pitch attitude - flight path angle.

For example, if an airplane after take-off has a pitch of 15 degrees nose up and the trajectory is climbing at an angle of 10 degrees (no wind), then the AoA is 5 degrees.
If an airplane is approaching in the glide slope of an ILS that has a slope of -3deg and the airplane is pitched up 2 deg, the AoA is again 5 degrees.

As you can see, 2 very different pitch can give the same AoA.

Thanks to you and others for the explanation.

[Evan]

You thought a third ADIRU was coming?

Did I think it was coming? No. Do I think it's needed on an aircraft with FBW functions that oppose pilot commands. Yes. So did the FAA and EASA when the A320 was being developed.

The best Boeing can do without adding a third data source is fail-safe, meaning that if a sensor fails, the system becomes inop and thus does nothing dangerous. However, that leaves the crew with an aircraft that is unstable near the edge of the envelope under certain conditions for the remainder of the flight, and thus somewhat dangerous. I'm concerned that, somewhere down the lifespan of the Max, another crash will occur due to a stall that resulted from pilot error plus a faulty AoA sensor taking the MCAS offline. Is that likely? No. Is that greater than a one-in-one billion possibility? Certainly. The FBW approval standard for failure-leading-to-loss-of-control rate on the A320 was one-in-one-billion.

Providing a third data source would give the Max fail-passive redundancy, meaning a sensor could fail without removing the augmented stability provided by the MCAS system. If Boeing were truly safety-focused, their solution to the current problem would include this additional hardware.

[3WE]

What do you guys think of this? Talks about how MCAS and trim runaway are different, again, and then adds something new:

https://leehamnews.com/2019/03/22/bj...-crash-part-2/

I stumbled on this in your link:

He flew the aircraft and asked the First Officer (FO) to go through the Quick Reference Handbook (QRH) which contains the Emergency checklists, to find the relevant emergency procedure for the strange behavior of the aircraft.

As the FO couldn’t find anything which fitted to their situation

For some reason, I am taken to Evan's ad-nauseum pontification of "procedure procedure procedure" along with "disdain for solid fundamental thought" that is not a named procedure...

Is this a crew with a brain full of procedures and all common sense removed?

The thing is: If the procedure is consistent with common sense, you don't have to waste much effort learning the procedure...just point out where it diverges, and then have some space left over for thought.

[Evan]

For some reason, I am taken to Evan's ad-nauseum pontification of "procedure procedure procedure" along with "disdain for solid fundamental thought" that is not a named procedure...

And that reason is that you continue to resist resist resist understanding the relationship between CRM procedures, situational awareness and stealth factors.

Is this a crew with a brain full of procedures and all common sense removed?

No, this is a crew that first tried common sense and, when the stealth factors confused the hell out of them, went to the QRH for answers. But the QRH is useless if you don't understand where the problem is. Procedures are useless if you lack proper ground-based training. 56 minutes on an iPad isn't proper ground-based training. Especially if the manufacturer is hiding things from you.

At some point you, Boeing and a lot of pilots are going to have to give up the idea that this was a simple, recognizable problem that should have been easily dealt with through common knowledge. This was a stealthy, confusing trap that introduced a very type-specific scenario with type-specific behaviors.

[Black Ram]

And what about the elevator blowback? Speculation, or is there something to it? The original source for this is not very clear.

[Peter Kesternich]

(...) All this implies something was not right.

I absolutely agree. Something is not right on many levels and in many areas. Without having read each and every post here concerning the LionAir and Ethiopian Airlines crashes, I think the following questions need to be answered:

- If the stabilizer problem was more or less easily solvable by the flight crew IF they realized what was going on and how to deal with it, why did the Ethiopian Airlines crew fall in the same trap as the LionAir crew, even after the problems encountered by the LionAir MAX were widely publicized? I would expect any pilot to be up to date on what's happening operations- and safetywise anywhere in the world - not only after I have been briefed and trained by my airline.
- Did Boeing and/or the FAA cut any corners when designing, testing, and certifying the MAX? If that is the case then it is a scandal of enormous political, economical, and technological proportions and consequences.


One last aside:

(...) The UTAir one was a runway overrun in difficult meteo conditions. Runway overruns happen like every day. Yes, UTAir was a very serious overrun resulting in a fire, but it doesn't seem to be on the level of AF358. (...)

Well - without knowing more specifics about UTAir 579 (other than what is written in the Aviation Herald article http://avherald.com/h?article=4bd19050&opt=0 ) I do think it seems to be pretty much on the level of AF358, even if the fire damage is not as extensive.

[Gabriel]

Did I think it was coming? No. Do I think it's needed on an aircraft with FBW functions that oppose pilot commands. Yes. So did the FAA and EASA when the A320 was being developed.

The best Boeing can do without adding a third data source is fail-safe, meaning that if a sensor fails, the system becomes inop and thus does nothing dangerous. However, that leaves the crew with an aircraft that is unstable near the edge of the envelope under certain conditions for the remainder of the flight, and thus somewhat dangerous. I'm concerned that, somewhere down the lifespan of the Max, another crash will occur due to a stall that resulted from pilot error plus a faulty AoA sensor taking the MCAS offline. Is that likely? No. Is that greater than a one-in-one billion possibility? Certainly. The FBW approval standard for failure-leading-to-loss-of-control rate on the A320 was one-in-one-billion.

Providing a third data source would give the Max fail-passive redundancy, meaning a sensor could fail without removing the augmented stability provided by the MCAS system. If Boeing were truly safety-focused, their solution to the current problem would include this additional hardware.

There is another possibility. A sensor-less AoA indicator. It is not really sensor-less, there are sensors, but no AoA sensors. It uses other sensors already installed in the plane to calculate the AOA indirectly form the orientation (pitch, roll and heading) and trajectory (3D inertial speed vector) of the plane. It already exists in General aviation and I understand that the concept was developed by Sperry many decades ago before the technology needed to make it real it existed.

[Evan]

There is another possibility. A sensor-less AoA indicator. It is not really sensor-less, there are sensors, but no AoA sensors. It uses other sensors already installed in the plane to calculate the AOA indirectly form the orientation (pitch, roll and heading) and trajectory (3D inertial speed vector) of the plane. It already exists in General aviation and I understand that the concept was developed by Sperry many decades ago before the technology needed to make it real it existed.

I don't know how reliable that is. I mean, you have an AoA vane disagree. That means you can't be sure of airspeed, but you are need airspeed and IR data to calculate AoA, right? I don't know how it all comes together. I guess you can use that data to determine which of the two is most likely faulty. But what if both of them are? I just think, with anything overriding pilot commands you always need TWO sensors in agreement. Nothing beats a third physical sensor and it seems do-able on a $100M airplane.

BTW, the A320 that crashed due to frozen AoA sensors had three vanes, but, if I remember correctly, only two were affected and, because they were is agreement, the third, functional one was ruled out. But I guess any level of safety can be overcome by total stupidity.

[Gabriel]

I don't know how reliable that is. I mean, you have an AoA vane disagree. That means you can't be sure of airspeed, but you are need airspeed and IR data to calculate AoA, right? I don't know how it all comes together. I guess you can use that data to determine which of the two is most likely faulty. But what if both of them are? I just think, with anything overriding pilot commands you always need TWO sensors in agreement.

Actually, ideal redundancy requires different designs and even different principles of operations and different codes of software so the different sources are not susceptible to the same failure mode at the same time. We discussed that in the context of the AF accident. What's the point of having 3 or 30 Pitot tubes if all of them will freeze at the same time when exposed to the same conditions?

I also don't know how accurate or reliable that sensor-less AoA is. I know it uses information from the AHRS (which is basically a cheap but very good and reliable version of an IRS, which instead of laser gyros uses MEMS accelerators and gyros). I don't know how much it relies in air data or GPS if at all. Perhaps it is excellent, perhaps it isn't. It is approved in GA but it is only an indicator not something that will make control inputs.

[Evan]

Actually, ideal redundancy requires different designs and even different principles of operations and different codes of software so the different sources are not susceptible to the same failure mode at the same time. We discussed that in the context of the AF accident. What's the point of having 3 or 30 Pitot tubes if all of them will freeze at the same time when exposed to the same conditions?

You no powerwash the AoA vanes, they no freeze. It takes an act of spectacular stupidity to freeze the AoA vanes. Otherwise, there is no need for disimilarity.

I also don't know how accurate or reliable that sensor-less AoA is. I know it uses information from the AHRS (which is basically a cheap but very good and reliable version of an IRS, which instead of laser gyros uses MEMS accelerators and gyros). I don't know how much it relies in air data or GPS if at all. Perhaps it is excellent, perhaps it isn't. It is approved in GA but it is only an indicator not something that will make control inputs.

The gyros will tell you your attitude relative the to horizon, but not relative to the relative wind. You need pitot data for that. How are you going to get accurate pitot readings if you don't know your AoA? It just seems like a fool's errand to me. Actually, the entire 737-MAX feels like a fool's errand to me. Why not do things right in the first place?