Moderators: jsumali2, richierich, ua900, PanAm_DC10, hOMSaR

 
User avatar
par13del
Posts: 10350
Joined: Sun Dec 18, 2005 9:14 pm

Re: Boeing 737MAX Grounding, General Discussion Thread, Q3 2020

Thu Sep 10, 2020 12:25 pm

737max8 wrote:
FWIW, even if these planes are due to be delivered to WN eventually, they are Boeing A/C until the day they are delivered. Found the term "borrowed" interesting about the 737-8. It's Boeing's airplane.

So we are sure this is not an a/c that was already delivered to WN, was in service, and was flown to storage after the grounding?
Is there any merit in taking an a/c that was already in commercial service and using it to proof the modification process?
 
User avatar
Polot
Posts: 10708
Joined: Thu Jul 28, 2011 3:01 pm

Re: Boeing 737MAX Grounding, General Discussion Thread, Q3 2020

Thu Sep 10, 2020 12:27 pm

par13del wrote:
Is there any merit in taking an a/c that was already in commercial service and using it to proof the modification process?

Not really, since aircraft already delivered and previously in service should be identical to the aircraft built but not delivered. From my understanding most of the current aircraft modifications are just software changes with the only hardware changes being the wires that are being fixed.
 
User avatar
SheikhDjibouti
Posts: 2278
Joined: Sat Sep 30, 2017 4:59 pm

Re: Boeing 737MAX Grounding, General Discussion Thread, Q3 2020

Thu Sep 10, 2020 1:02 pm

Polot wrote:
kalvado wrote:
737max8 wrote:
FWIW, even if these planes are due to be delivered to WN eventually, they are Boeing A/C until the day they are delivered. Found the term "borrowed" interesting about the 737-8. It's Boeing's airplane.

Airline expects to get the plane delivered with single digit number of cycles and hours. These test flights increase both. Not sure how much that affects the value or spelled out in a contract, though. Bit I am pretty sure I would be not too happy if my "new" car had couple hundred miles on delivery.

Airlines get compensated when their plane is used for testing. Boeing doesn’t just grab a random plane, they ask the intended customer and negotiate a settlement- lower price, discount on new feature if that is what they are testing, included in any publicity (eg testing new biofuels) , etc. It may technically be Boeing’s plane still but it usually has the customer’s trademarked name and branding painted on it so it’s not like Boeing can just use the planes willy nilly for whatever they want.

Boeing often first asks airlines they have a close relationship with if they have a suitable aircraft in the pipeline-WN, UA, etc.
So... it's a bit like asking if you may "borrow" something. :duck:
Nothing to see here; move along please.
 
Noshow
Posts: 1669
Joined: Wed Jun 15, 2016 3:20 pm

Re: Boeing 737MAX Grounding, General Discussion Thread, Q3 2020

Thu Sep 10, 2020 1:48 pm

It's in the contract. It's like buying a prototype for later airline use. You get guaranteed the number of flight hours it will have. If not you can decide to not take it. Like LH with their final 747-8. It had more hours than expected and would have needed different heavy maintenance cycles compared to the rest of their fleet.
 
User avatar
Revelation
Posts: 24589
Joined: Wed Feb 09, 2005 9:37 pm

Re: Boeing 737MAX Grounding, General Discussion Thread, Q3 2020

Thu Sep 10, 2020 3:58 pm

Virtual737 wrote:
Polot wrote:
Nope. Raw test data will never be released, that is a commercial secret. You can gain a lot of information about the plane beyond just what MCAS is doing from raw data, and of course the government is not going to hand that out so competitors worldwide can see and analyze it. Then no company (in any industry) would submit information to the US.

My personal opinion is that a company should lose those commercial protections the moment it releases a product with serious, life endangering flaws. It's also not likely that another manufacturer would be particularly keen to copy anything from the MAX.

Understood that there are a myriad of other reasons why it wouldn't be in Boeing's interest, but that's not near the top of my worry list.

Then data for pretty much every plane, train and automobile would be released. My own car has had three recalls in the 15 years I've owned it, and it's not a cheap car. People don't seem to understand getting everything right is a big challenge. MCAS is a particularly egregious screw up, but it's far from the only product in use every day that has serious life threatening flaws in it. For instance most airlines were in the process of replacing the pitot tubes that were the root cause of the AF447 crash.

737max8 wrote:
FWIW, even if these planes are due to be delivered to WN eventually, they are Boeing A/C until the day they are delivered. Found the term "borrowed" interesting about the 737-8. It's Boeing's airplane.

That's why I put "borrowed" in quotes.

kalvado wrote:
Airline expects to get the plane delivered with single digit number of cycles and hours. These test flights increase both. Not sure how much that affects the value or spelled out in a contract, though. Bit I am pretty sure I would be not too happy if my "new" car had couple hundred miles on delivery.

These days new MAXes will show up with more than a single digit of hours since most are getting ferried to MWH or various other distant airfields and some are making multiple trips to get maintenance or to get the wiring fix installed.

Polot wrote:
Boeing often first asks airlines they have a close relationship with if they have a suitable aircraft in the pipeline-WN, UA, etc.

Agreed, although there must be something more to it, since at this point Boeing has several NTU MAXes to choose from.
Wake up to find out that you are the eyes of the world
The heart has its beaches, its homeland and thoughts of its own
Wake now, discover that you are the song that the morning brings
The heart has its seasons, its evenings and songs of its own
 
Dogman
Posts: 119
Joined: Fri Mar 16, 2018 4:47 pm

Re: Boeing 737MAX Grounding, General Discussion Thread, Q3 2020

Thu Sep 10, 2020 4:03 pm

I am not a pilot, so I apologize if my question is a bit silly. If you have only two AOA sensors, and you suspect that one of them is malfunctioning, couldn't you just point the nose of the plane up and down and see which sensor changes its output accordingly to the pilot's actions?
 
737max8
Posts: 640
Joined: Thu Dec 10, 2015 4:13 am

Re: Boeing 737MAX Grounding, General Discussion Thread, Q3 2020

Thu Sep 10, 2020 4:32 pm

par13del wrote:
737max8 wrote:
FWIW, even if these planes are due to be delivered to WN eventually, they are Boeing A/C until the day they are delivered. Found the term "borrowed" interesting about the 737-8. It's Boeing's airplane.

So we are sure this is not an a/c that was already delivered to WN, was in service, and was flown to storage after the grounding?
Is there any merit in taking an a/c that was already in commercial service and using it to proof the modification process?


Yes we are sure, all 737-8 MAX A/C that were delivered to and operated by WN are sitting in Victorville, California.

It appears the test flights are with Boeing test A/C that will/may be delivered to WN down the road. It would be interesting to know how Boeing chooses what A/C to use, but maybe it does have something to do with strong relationships and contracts. No clue!
The thoughts and opinions expressed in my comments do not represent that of any airline or affiliate.
Flown on: 717 733 734 735 736 737 738 739 7M8 744 744ER 752 753 762 763 772 773ER 788 789 A220 A319/20/21 A332 A333 A339 A343 A346 A359 A388
 
AndoAv8R
Posts: 55
Joined: Thu Feb 06, 2020 4:29 pm

Re: Boeing 737MAX Grounding, General Discussion Thread, Q3 2020

Thu Sep 10, 2020 5:48 pm

As we get closer to (hopefully) the MAX certified again, does anyone know where most of the upgrades will be taking place? I'd assume the airlines maintenance hubs will be able to or will they all need to go to a Boeing facility?
 
User avatar
Revelation
Posts: 24589
Joined: Wed Feb 09, 2005 9:37 pm

Re: Boeing 737MAX Grounding, General Discussion Thread, Q3 2020

Thu Sep 10, 2020 6:01 pm

Dogman wrote:
I am not a pilot, so I apologize if my question is a bit silly. If you have only two AOA sensors, and you suspect that one of them is malfunctioning, couldn't you just point the nose of the plane up and down and see which sensor changes its output accordingly to the pilot's actions?

It's an interesting question.

http://www.boeing.com/commercial/aeroma ... story.html is a pretty old but interesting article from Boeing on AoA. It seems to date from the late 90s or early 00s. One interesting quote:

Dedicated AOA indicators have been used on military aircraft for many years, but this form of display has not been used often on commercial airplanes. On Boeing models currently in production, AOA is used to drive stall warning (stick shaker), stall margin information on airspeed indicators, and the pitch limit indicator (PLI) on the primary attitude displays. AOA information is combined with other data and displayed as an integral part of flight deck displays.

So, it hadn't been usual for the AoA to be displayed on commercial aircraft at that point in time, but the article goes on to say they've developed an optional AoA indication for 737 and other models that looks like:

Image

So, one way to address your comment was that while AoA data was input to many other things that were displayed, AoA data itself typically did not have its own indicator, and even in recent times it was an optional feature.

The other thing to note is it is not displaying two AoA indicators, which means you could not compare the two readings.

It is well known that Boeing made a mistake with the AoA disagree indication. It was supposed to be available to all MAXes, but instead it was tied to the selection of the AoA indication option I described above, and only 20% of the operators chose that option, so 80% did not have it. Boeing made the argument that it was not critical to flight so they deferred the fix to the next software update, which did not happen before the two crashes.

Ref: https://www.aviationtoday.com/2019/05/0 ... ree-alert/

The flight configuration FAA certified presumed there would be a working AoA disagree indication but 80% of the aircraft did not have this. This is one of many findings in the FAA's review of the MAX ( ref: https://www.faa.gov/news/media/attachme ... ry-v-1.pdf )

It's not clear if AoA disagree would have triggered early enough to help either crew understand what was going on, because the JT pilots did not know of MCAS and the ET pilots didn't have much info about MCAS. It is clear Boeing botched the implementation and FAA is holding them to task for it.

Bottom line: I don't see why your idea would not work, but the approach Boeing took was to have the computer compare the two AoA indications and flag a disagreement on the pilot displays. Unfortunately they botched the implementation and didn't fix it before the two crashes.
Wake up to find out that you are the eyes of the world
The heart has its beaches, its homeland and thoughts of its own
Wake now, discover that you are the song that the morning brings
The heart has its seasons, its evenings and songs of its own
 
User avatar
Revelation
Posts: 24589
Joined: Wed Feb 09, 2005 9:37 pm

Re: Boeing 737MAX Grounding, General Discussion Thread, Q3 2020

Thu Sep 10, 2020 6:04 pm

AndoAv8R wrote:
As we get closer to (hopefully) the MAX certified again, does anyone know where most of the upgrades will be taking place? I'd assume the airlines maintenance hubs will be able to or will they all need to go to a Boeing facility?

For the aircraft not yet delivered, I think it's all stuff that the airlines can handle in house or contract to MROs rather than going to a Boeing facility.
Wake up to find out that you are the eyes of the world
The heart has its beaches, its homeland and thoughts of its own
Wake now, discover that you are the song that the morning brings
The heart has its seasons, its evenings and songs of its own
 
Vicenza
Posts: 142
Joined: Sun Apr 19, 2020 3:21 pm

Re: Boeing 737MAX Grounding, General Discussion Thread, Q3 2020

Thu Sep 10, 2020 6:12 pm

Revelation wrote:
AndoAv8R wrote:
As we get closer to (hopefully) the MAX certified again, does anyone know where most of the upgrades will be taking place? I'd assume the airlines maintenance hubs will be able to or will they all need to go to a Boeing facility?

For the aircraft not yet delivered, I think it's all stuff that the airlines can handle in house or contract to MROs rather than going to a Boeing facility.


If they haven't yet been delivered, then surely Boeing should be doing the work, no? Why would airlines be doing it?
 
StTim
Posts: 3732
Joined: Thu Aug 08, 2013 7:39 am

Re: Boeing 737MAX Grounding, General Discussion Thread, Q3 2020

Thu Sep 10, 2020 7:16 pm

737max8 wrote:
par13del wrote:
737max8 wrote:
FWIW, even if these planes are due to be delivered to WN eventually, they are Boeing A/C until the day they are delivered. Found the term "borrowed" interesting about the 737-8. It's Boeing's airplane.

So we are sure this is not an a/c that was already delivered to WN, was in service, and was flown to storage after the grounding?
Is there any merit in taking an a/c that was already in commercial service and using it to proof the modification process?


Yes we are sure, all 737-8 MAX A/C that were delivered to and operated by WN are sitting in Victorville, California.

It appears the test flights are with Boeing test A/C that will/may be delivered to WN down the road. It would be interesting to know how Boeing chooses what A/C to use, but maybe it does have something to do with strong relationships and contracts. No clue!


Also, surely, this aircraft will need to have the proposed software changes applied before the test flights. This will not be don on customer owned planes at this time as it is not certified.

I do presume they don't need the wiring harness changes applied for these EASA flights. That would seem superfluous - although they may want to see exactly an as implemented example.
 
User avatar
Revelation
Posts: 24589
Joined: Wed Feb 09, 2005 9:37 pm

Re: Boeing 737MAX Grounding, General Discussion Thread, Q3 2020

Thu Sep 10, 2020 7:44 pm

Vicenza wrote:
Revelation wrote:
AndoAv8R wrote:
As we get closer to (hopefully) the MAX certified again, does anyone know where most of the upgrades will be taking place? I'd assume the airlines maintenance hubs will be able to or will they all need to go to a Boeing facility?

For the aircraft not yet delivered, I think it's all stuff that the airlines can handle in house or contract to MROs rather than going to a Boeing facility.

If they haven't yet been delivered, then surely Boeing should be doing the work, no? Why would airlines be doing it?

Sorry, I got that backwards, my bad.

Boeing can't deliver a plane without the fixes in place, it can't deliver a plane that can't legally fly.
Wake up to find out that you are the eyes of the world
The heart has its beaches, its homeland and thoughts of its own
Wake now, discover that you are the song that the morning brings
The heart has its seasons, its evenings and songs of its own
 
BEG2IAH
Posts: 971
Joined: Thu Apr 22, 2004 3:42 pm

Re: Boeing 737MAX Grounding, General Discussion Thread, Q3 2020

Thu Sep 10, 2020 7:57 pm

They are pulling some tight turns at very high speeds: https://flightaware.com/live/flight/BOE701
Flying at the cruising altitude is (mostly) boring. I wish all flights were nothing but endless take offs and landings every 10 minutes or so.
 
Dogman
Posts: 119
Joined: Fri Mar 16, 2018 4:47 pm

Re: Boeing 737MAX Grounding, General Discussion Thread, Q3 2020

Thu Sep 10, 2020 9:17 pm

Revelation wrote:
Dogman wrote:
I am not a pilot, so I apologize if my question is a bit silly. If you have only two AOA sensors, and you suspect that one of them is malfunctioning, couldn't you just point the nose of the plane up and down and see which sensor changes its output accordingly to the pilot's actions?

It's an interesting question.

http://www.boeing.com/commercial/aeroma ... story.html is a pretty old but interesting article from Boeing on AoA. It seems to date from the late 90s or early 00s. One interesting quote:

Dedicated AOA indicators have been used on military aircraft for many years, but this form of display has not been used often on commercial airplanes. On Boeing models currently in production, AOA is used to drive stall warning (stick shaker), stall margin information on airspeed indicators, and the pitch limit indicator (PLI) on the primary attitude displays. AOA information is combined with other data and displayed as an integral part of flight deck displays.

So, it hadn't been usual for the AoA to be displayed on commercial aircraft at that point in time, but the article goes on to say they've developed an optional AoA indication for 737 and other models that looks like:

Image

Thanks for your answer. I've suspected it won't be that easy.
So, one way to address your comment was that while AoA data was input to many other things that were displayed, AoA data itself typically did not have its own indicator, and even in recent times it was an optional feature.

The other thing to note is it is not displaying two AoA indicators, which means you could not compare the two readings.

It is well known that Boeing made a mistake with the AoA disagree indication. It was supposed to be available to all MAXes, but instead it was tied to the selection of the AoA indication option I described above, and only 20% of the operators chose that option, so 80% did not have it. Boeing made the argument that it was not critical to flight so they deferred the fix to the next software update, which did not happen before the two crashes.

Ref: https://www.aviationtoday.com/2019/05/0 ... ree-alert/

The flight configuration FAA certified presumed there would be a working AoA disagree indication but 80% of the aircraft did not have this. This is one of many findings in the FAA's review of the MAX ( ref: https://www.faa.gov/news/media/attachme ... ry-v-1.pdf )

It's not clear if AoA disagree would have triggered early enough to help either crew understand what was going on, because the JT pilots did not know of MCAS and the ET pilots didn't have much info about MCAS. It is clear Boeing botched the implementation and FAA is holding them to task for it.

Bottom line: I don't see why your idea would not work, but the approach Boeing took was to have the computer compare the two AoA indications and flag a disagreement on the pilot displays. Unfortunately they botched the implementation and didn't fix it before the two crashes.


Thanks for your answer. I've suspected that it won't be that easy.
 
BEG2IAH
Posts: 971
Joined: Thu Apr 22, 2004 3:42 pm

Re: Boeing 737MAX Grounding, General Discussion Thread, Q3 2020

Fri Sep 11, 2020 4:03 pm

Looks like the tests are completed.
https://www.easa.europa.eu/newsroom-and ... st-flights

Can't wait for next week.
Flying at the cruising altitude is (mostly) boring. I wish all flights were nothing but endless take offs and landings every 10 minutes or so.
 
Spetsnaz55
Posts: 224
Joined: Wed Feb 20, 2019 2:38 am

Re: Boeing 737MAX Grounding, General Discussion Thread, Q3 2020

Sat Sep 12, 2020 7:03 pm

https://www.seattletimes.com/business/b ... -now-safe/


Good read. The plane is stable with Mcas off
 
User avatar
Revelation
Posts: 24589
Joined: Wed Feb 09, 2005 9:37 pm

Re: Boeing 737MAX Grounding, General Discussion Thread, Q3 2020

Sat Sep 12, 2020 8:05 pm

Spetsnaz55 wrote:
https://www.seattletimes.com/business/boeing-aerospace/as-the-faa-finalizes-the-737-maxs-return-is-boeings-jet-now-safe/

Good read. The plane is stable with Mcas off

One part I found interesting:

The Boeing senior engineer said the company hopes that with MCAS fixed, the stick shaker silenced and pilots trained with new alert checklists, that will prove enough to satisfy EASA, too.

The way I read it, the implication is that Boeing hopes EASA will let the 3rd AoA sensor issue drop, yet we've read here that EASA wants to have a plan in place for the 3rd AoA sensor before they will let MAX RTS and will not certify MAX-10 till the 3rd AoA is implemented.

In a statement, the FAA said that in collaboration with three major foreign aviation safety regulators it has extensively evaluated the MAX redesign.

The modified aircraft will be fully compliant with the applicable rules, using the most conservative means of compliance,” the FAA said.

That's what they said about the MAX with v1 software too, isn't it?

Fully compliant means no 3rd AoA sensor is needed, hmm...

Dropping a new software system such as Ewbank’s synthetic airspeed into the middle of an old system architecture is a complex task that would require significant design and test resources. The former senior Boeing engineer who worked on the MAX said it would risk unintended consequences in some unforeseen flight scenario.

“You’ve got to be very careful anytime you want to add some new feature to a very proven reliable airplane,” he said.

That's not the way to look at it, though. Either it is needed to met regs, or it is not. Either Boeing knows what its systems will do at the edges of the flight envelope, or it doesn't. If the architecture is too old to support the requirements, it has to go, as does any other plane that can't meet the regs.

For current airplanes, including the MAX, the manufacturers and the airlines will have to gear up training and skills checks and sort out pilots who are demonstrably not apt at handling an aircraft that requires more from the pilot.

Good luck with that. Pilot training is now a profit center for some airlines, and it's bad business to weed out paying customers.
Wake up to find out that you are the eyes of the world
The heart has its beaches, its homeland and thoughts of its own
Wake now, discover that you are the song that the morning brings
The heart has its seasons, its evenings and songs of its own
 
User avatar
Stitch
Posts: 27226
Joined: Wed Jul 06, 2005 4:26 am

Re: Boeing 737MAX Grounding, General Discussion Thread, Q3 2020

Sat Sep 12, 2020 8:23 pm

I would guess be it a synthetic or physical AoA sensor, the fact that the existing architecture is designed for one or two inputs only means adding said third adds possible complexity and integration issues. But as others have noted, while two is better than one, you can still have a situation where one would be offering erroneous data and how does the system determine which sensor data to trust.

I wonder if it would have been possible to have three physical sensors and isolate that data outside the system. The three then round-robin to ensure at least two are in agreement, which is then sent into the system, which would operate in a single-AoA mode as designed to support, but would always have trusted data to work from.
 
User avatar
Revelation
Posts: 24589
Joined: Wed Feb 09, 2005 9:37 pm

Re: Boeing 737MAX Grounding, General Discussion Thread, Q3 2020

Sat Sep 12, 2020 9:02 pm

Stitch wrote:
I would guess be it a synthetic or physical AoA sensor, the fact that the existing architecture is designed for one or two inputs only means adding said third adds possible complexity and integration issues. But as others have noted, while two is better than one, you can still have a situation where one would be offering erroneous data and how does the system determine which sensor data to trust.

There's an old joke from computer science that has an element of truth: All complexity can be solved by adding another layer.

In this case, it suggests have a "virtual AoA" that reflects the consensus output of the N different AoA sensors, or if a consensus can't be reached, a failure is signaled. Then you feed this virtual AoA result to all the places that need the real AoA result, deal with the inevitable mismatches and side issues, and all should be OK. Then the issue becomes do you have the CPU resources to implement the virtual AoA. If so, you're golden. If not, you need to add CPU resources and that opens a can of worms since it's a flight critical instrument, but they are all solvable problems if you're willing to spend the money to solve them.

The issue here seems to be a preference to avoid spending the money. All the responses I've seen all lead back to that basic issue.

It should be black and white: either FAA/EASA require the third reference for MAX and all other aircraft in its category and Boeing must find a solution, or not. I'm not understanding the greyness, other than it's a result of a desire to save money.
Wake up to find out that you are the eyes of the world
The heart has its beaches, its homeland and thoughts of its own
Wake now, discover that you are the song that the morning brings
The heart has its seasons, its evenings and songs of its own
 
2175301
Posts: 1893
Joined: Wed May 16, 2007 11:19 am

Re: Boeing 737MAX Grounding, General Discussion Thread, Q3 2020

Sat Sep 12, 2020 10:58 pm

Revelation wrote:
Stitch wrote:
I would guess be it a synthetic or physical AoA sensor, the fact that the existing architecture is designed for one or two inputs only means adding said third adds possible complexity and integration issues. But as others have noted, while two is better than one, you can still have a situation where one would be offering erroneous data and how does the system determine which sensor data to trust.

There's an old joke from computer science that has an element of truth: All complexity can be solved by adding another layer.

In this case, it suggests have a "virtual AoA" that reflects the consensus output of the N different AoA sensors, or if a consensus can't be reached, a failure is signaled. Then you feed this virtual AoA result to all the places that need the real AoA result, deal with the inevitable mismatches and side issues, and all should be OK. Then the issue becomes do you have the CPU resources to implement the virtual AoA. If so, you're golden. If not, you need to add CPU resources and that opens a can of worms since it's a flight critical instrument, but they are all solvable problems if you're willing to spend the money to solve them.

The issue here seems to be a preference to avoid spending the money. All the responses I've seen all lead back to that basic issue.

It should be black and white: either FAA/EASA require the third reference for MAX and all other aircraft in its category and Boeing must find a solution, or not. I'm not understanding the greyness, other than it's a result of a desire to save money.


I actually see several issues. It may not be just about spending a bit of money. The existing flight computers may not have the calculation capacity (space) to handle an additional input. It may be that all that could be done with a 3rd sensor is to just disable the MCAS system and let the pilots know that there has been a failure and the aircraft will not respond exactly the same. They are trained to do that for other failures.

I'm also curious how EASA will be able to demand a change for something that is acceptable to US certification standards and is not on the current exception list where Boeing (and any other non-EASA manufactures from Brazil, Canada, and the USA have to provide additional data for certification) under the current treaty. Not honoring the treaty would certainly have grave consequence for the EU for all international trade treaties. The EASA may want a change... But, on this issue I'm not sure if they have a legal leg to stand on if the FAA accepts that the current version meets US standards.

It will be interesting to watch this all play out...

However, for now the Max appears to be able to return to commercial service in the next few months.

Have a great day,
 
User avatar
Stitch
Posts: 27226
Joined: Wed Jul 06, 2005 4:26 am

Re: Boeing 737MAX Grounding, General Discussion Thread, Q3 2020

Sun Sep 13, 2020 12:00 am

2175301 wrote:
I'm also curious how EASA will be able to demand a change for something that is acceptable to US certification standards and is not on the current exception list where Boeing (and any other non-EASA manufactures from Brazil, Canada, and the USA have to provide additional data for certification) under the current treaty. Not honoring the treaty would certainly have grave consequence for the EU for all international trade treaties. The EASA may want a change... But, on this issue I'm not sure if they have a legal leg to stand on if the FAA accepts that the current version meets US standards.


That is an interesting point. EASA did certify the MAX without three AoA sensors under the Airworthiness Approval section of the current Technical Implementation Procedures agreement between them and the FAA. I've not read the TIP, so I cannot say it would prevent EASA from forcing Boeing to add a third AoA sensor for models operated by European operators or allow them to not certify the MAX10 unless it adds them as I believe they have stated is their intent.
 
B777
Posts: 384
Joined: Fri Sep 17, 1999 9:52 am

Re: Boeing 737MAX Grounding, General Discussion Thread, Q3 2020

Sun Sep 13, 2020 3:33 am

Test aircraft for the EASA flights was a Max 7 in Boeing house colors, registration N7201S (destined for Southwest Airlines). Tests were conducted on September 8,9 and 10. Every test day, the aircraft departed BFI for YVR to pick up the EASA personnel. The aircraft then flew to Moses Lake to conduct testing. After the test, the aircraft flew back to YVR to drop off the EASA personnel. Then it returned to BFI.
 
djpearman
Posts: 19
Joined: Thu Feb 06, 2020 9:02 pm

Re: Boeing 737MAX Grounding, General Discussion Thread, Q3 2020

Sun Sep 13, 2020 9:31 am

2175301 wrote:
I actually see several issues. It may not be just about spending a bit of money. The existing flight computers may not have the calculation capacity (space) to handle an additional input. It may be that all that could be done with a 3rd sensor is to just disable the MCAS system and let the pilots know that there has been a failure and the aircraft will not respond exactly the same. They are trained to do that for other failures.


Firstly, if a third AOA input is required and the existing flight computers can't accommodate this, then the conclusion has to be a replacement of the flight computers with more capable ones (which in turn does make it about spending a bit of money), not the omission of the third AOA input. It would set a dangerous precedent for the future to let an implementation dictate the regulation.

Secondly, Boeing designed the 737 MAX with MCAS, which implies that it is a necessary system for the aircraft to operate. Otherwise, why bother designing it in? So, in the case of two AOA sensors and a disagreement between the two, since there is no means of distinguishing the correct value, the only course of action is to disable MCAS. Is the aircraft now still airworthy? If a disagree occurs during a flight, continue the current flight or divert to the nearest airport? In my opinion, if the answer is that the aircraft is still airworthy and is able to complete its flight, then that begs the question as to why not just remove MCAS altogether? As far as I can tell, the only reason MCAS was installed was to make the MAX behave like the NG so as to not require pilot sim training. Since that is no longer the case, I don't see why MCAS is still needed. Maybe someone with more knowledge on this can enlighten me :)
 
Opus99
Posts: 1080
Joined: Thu May 30, 2019 10:51 pm

Re: Boeing 737MAX Grounding, General Discussion Thread, Q3 2020

Sun Sep 13, 2020 9:35 am

djpearman wrote:
2175301 wrote:
I actually see several issues. It may not be just about spending a bit of money. The existing flight computers may not have the calculation capacity (space) to handle an additional input. It may be that all that could be done with a 3rd sensor is to just disable the MCAS system and let the pilots know that there has been a failure and the aircraft will not respond exactly the same. They are trained to do that for other failures.


Firstly, if a third AOA input is required and the existing flight computers can't accommodate this, then the conclusion has to be a replacement of the flight computers with more capable ones (which in turn does make it about spending a bit of money), not the omission of the third AOA input. It would set a dangerous precedent for the future to let an implementation dictate the regulation.

Secondly, Boeing designed the 737 MAX with MCAS, which implies that it is a necessary system for the aircraft to operate. Otherwise, why bother designing it in? So, in the case of two AOA sensors and a disagreement between the two, since there is no means of distinguishing the correct value, the only course of action is to disable MCAS. Is the aircraft now still airworthy? If a disagree occurs during a flight, continue the current flight or divert to the nearest airport? In my opinion, if the answer is that the aircraft is still airworthy and is able to complete its flight, then that begs the question as to why not just remove MCAS altogether? As far as I can tell, the only reason MCAS was installed was to make the MAX behave like the NG so as to not require pilot sim training. Since that is no longer the case, I don't see why MCAS is still needed. Maybe someone with more knowledge on this can enlighten me :)

I’m assuming without MCAS there would be even MORE pilot training needed than with MCAS. The reality if MCAS is needed for the max to fly you shouldn’t be able to turn it off.
 
djpearman
Posts: 19
Joined: Thu Feb 06, 2020 9:02 pm

Re: Boeing 737MAX Grounding, General Discussion Thread, Q3 2020

Sun Sep 13, 2020 9:39 am

Opus99 wrote:
djpearman wrote:
2175301 wrote:
I actually see several issues. It may not be just about spending a bit of money. The existing flight computers may not have the calculation capacity (space) to handle an additional input. It may be that all that could be done with a 3rd sensor is to just disable the MCAS system and let the pilots know that there has been a failure and the aircraft will not respond exactly the same. They are trained to do that for other failures.


Firstly, if a third AOA input is required and the existing flight computers can't accommodate this, then the conclusion has to be a replacement of the flight computers with more capable ones (which in turn does make it about spending a bit of money), not the omission of the third AOA input. It would set a dangerous precedent for the future to let an implementation dictate the regulation.

Secondly, Boeing designed the 737 MAX with MCAS, which implies that it is a necessary system for the aircraft to operate. Otherwise, why bother designing it in? So, in the case of two AOA sensors and a disagreement between the two, since there is no means of distinguishing the correct value, the only course of action is to disable MCAS. Is the aircraft now still airworthy? If a disagree occurs during a flight, continue the current flight or divert to the nearest airport? In my opinion, if the answer is that the aircraft is still airworthy and is able to complete its flight, then that begs the question as to why not just remove MCAS altogether? As far as I can tell, the only reason MCAS was installed was to make the MAX behave like the NG so as to not require pilot sim training. Since that is no longer the case, I don't see why MCAS is still needed. Maybe someone with more knowledge on this can enlighten me :)

I’m assuming without MCAS there would be even MORE pilot training needed than with MCAS. The reality if MCAS is needed for the max to fly you shouldn’t be able to turn it off.


Which in turn necessitates a third AOA input.
 
morrisond
Posts: 2773
Joined: Thu Jan 07, 2010 12:22 am

Re: Boeing 737MAX Grounding, General Discussion Thread, Q3 2020

Sun Sep 13, 2020 10:55 am

That's great that the MAX is Stable even with MCAS turned off. I think we wasted pages on debating what stable is. Without MCAS the MAX is not an unsafe aircraft.

As I wrote which now seems like years ago - while MCAS was needed to meet a reg - sometimes regs need to be looked at and exceptions made in certain cases as they can add too much complexity for very little to no safety benefit(the stick force requirement decreeing that the force must always be increasing and can't plateau).

In terms of AOA - The NG has been flying very safely with 2 AOA sensors for decades now - why add unneeded complexity? Let it be. Even with three sensors there are examples of those planes almost going into the ground.

One of the downsides of making a plane too perfect is Pilot's start assuming that it is perfect and that it can do nothing wrong. Instead of trusting their own training (assuming they had half decent training) when something goes wonky with the flightpath they start analyzing the systems vs just flying the aircraft like they are supposed too.

Until such time that AI can be perfect and take over the whole flight - I suggest we reexamine how we train commercial pilots and no more 100 hour wonders which end up in an 737 without having any real experience actually flying and dealing with adverse situations.
 
Noshow
Posts: 1669
Joined: Wed Jun 15, 2016 3:20 pm

Re: Boeing 737MAX Grounding, General Discussion Thread, Q3 2020

Sun Sep 13, 2020 12:24 pm

What do we know about flight behavior with MCAS off? Is there any new source please?
 
Spetsnaz55
Posts: 224
Joined: Wed Feb 20, 2019 2:38 am

Re: Boeing 737MAX Grounding, General Discussion Thread, Q3 2020

Sun Sep 13, 2020 2:21 pm

Noshow wrote:
What do we know about flight behavior with MCAS off? Is there any new source please?


https://www.seattletimes.com/business/b ... -now-safe/
 
User avatar
Revelation
Posts: 24589
Joined: Wed Feb 09, 2005 9:37 pm

Re: Boeing 737MAX Grounding, General Discussion Thread, Q3 2020

Sun Sep 13, 2020 3:11 pm

djpearman wrote:
Firstly, if a third AOA input is required and the existing flight computers can't accommodate this, then the conclusion has to be a replacement of the flight computers with more capable ones (which in turn does make it about spending a bit of money), not the omission of the third AOA input. It would set a dangerous precedent for the future to let an implementation dictate the regulation.

Yes, that's my point. The regs should drive the implementation, not the other way around. If there aren't enough computes left to do the small calculations needed to support a 'virtual AoA' then we could also create an auxillary processor "complex" to do this task. I say complex beause it too would need redudancy, and would in turn create small issues with space and power and heat as does every processor, but these can all be solved with money. The amount of processing we're talking about here can be done with an Arduino after all, since we won't be constrained to using 80s vintage CPUs.

djpearman wrote:
Secondly, Boeing designed the 737 MAX with MCAS, which implies that it is a necessary system for the aircraft to operate. Otherwise, why bother designing it in? So, in the case of two AOA sensors and a disagreement between the two, since there is no means of distinguishing the correct value, the only course of action is to disable MCAS. Is the aircraft now still airworthy? If a disagree occurs during a flight, continue the current flight or divert to the nearest airport? In my opinion, if the answer is that the aircraft is still airworthy and is able to complete its flight, then that begs the question as to why not just remove MCAS altogether? As far as I can tell, the only reason MCAS was installed was to make the MAX behave like the NG so as to not require pilot sim training. Since that is no longer the case, I don't see why MCAS is still needed. Maybe someone with more knowledge on this can enlighten me :)

It's a bit more than making MAX feel like NG, it's the 'force gradient' requirement in the FARs. The change in stick force does decrease at high angles of AoA because the repositioned nacelles generate lift as the AoA increases. The reason the reg exists is because this may induce the pilot to compensate by pulling back the stick even more and causing a stall just because the stick force decreased too rapidly. It's not a Boeing preference, it's a Federal Aviation Rule.

morrisond wrote:
As I wrote which now seems like years ago - while MCAS was needed to meet a reg - sometimes regs need to be looked at and exceptions made in certain cases as they can add too much complexity for very little to no safety benefit(the stick force requirement decreeing that the force must always be increasing and can't plateau).

That would have painted Boeing into a corner, because asking for an exemption would be an admission that MAX did not fly like NG so sim training would have been needed. Boeing wasn't going to go there, thus the invention and certification of MCAS v1. Once Boeing released a very broken MCAS implementation and after the two crashes, I think it they weren't going to then ask FAA to fix it all via pencil whipping nor would FAA be in the mood to do such.

morrisond wrote:
Until such time that AI can be perfect and take over the whole flight - I suggest we reexamine how we train commercial pilots and no more 100 hour wonders which end up in an 737 without having any real experience actually flying and dealing with adverse situations.

Yes, the ST article did loop back to the pilot training topic later in the piece. Surely we haven't heard the last on it. You will note that the new training updates need to be approved before RTS. Then we had Boeing's CEO suggesting that FAA was pushing back on 787 level automation for the NMA cockpit and improvements would be needed.

Personally I think the training issue may be sold short. We have evidence that the combination of pilots and checklists weren't working well in the two crashes and in some of the post-crash sim checks. Personally I think every 737 pilot should be given a sim check focused on the challenges revealed via MCAS such as stab trim runaway and AoA disagree, and if they can't sort them out, they should be retrained till they can or till the company decides they need a better pilot. There's no excuse not too, other than putting money over safety. All the MAXes are grounded and most of the NGs too. Sim training will be mandatory for RTS. It seems like a great time to throw a bunch of testing in at the same time and use this opportunity to improve compliance on familiarity with and use of checklists.
Wake up to find out that you are the eyes of the world
The heart has its beaches, its homeland and thoughts of its own
Wake now, discover that you are the song that the morning brings
The heart has its seasons, its evenings and songs of its own
 
TropicalSky
Posts: 512
Joined: Fri May 05, 2017 1:37 pm

Re: Boeing 737MAX Grounding, General Discussion Thread, Q3 2020

Sun Sep 13, 2020 3:20 pm

i agree with the pilot training....during my years flying the SF340 and doing recurrent,I've had some instructors that would run you thru the ringer with faults that i never saw in 8yrs of flying the actual plane. TO me it made you better knowing that you had to be prepared for anything and not expect a scripted sequence of events (training) and it made you better while actually line flying knowing more about the systems.

Revelation wrote:
djpearman wrote:
Firstly, if a third AOA input is required and the existing flight computers can't accommodate this, then the conclusion has to be a replacement of the flight computers with more capable ones (which in turn does make it about spending a bit of money), not the omission of the third AOA input. It would set a dangerous precedent for the future to let an implementation dictate the regulation.

Yes, that's my point. The regs should drive the implementation, not the other way around. If there aren't enough computes left to do the small calculations needed to support a 'virtual AoA' then we could also create an auxillary processor "complex" to do this task. I say complex beause it too would need redudancy, and would in turn create small issues with space and power and heat as does every processor, but these can all be solved with money. The amount of processing we're talking about here can be done with an Arduino after all, since we won't be constrained to using 80s vintage CPUs.

djpearman wrote:
Secondly, Boeing designed the 737 MAX with MCAS, which implies that it is a necessary system for the aircraft to operate. Otherwise, why bother designing it in? So, in the case of two AOA sensors and a disagreement between the two, since there is no means of distinguishing the correct value, the only course of action is to disable MCAS. Is the aircraft now still airworthy? If a disagree occurs during a flight, continue the current flight or divert to the nearest airport? In my opinion, if the answer is that the aircraft is still airworthy and is able to complete its flight, then that begs the question as to why not just remove MCAS altogether? As far as I can tell, the only reason MCAS was installed was to make the MAX behave like the NG so as to not require pilot sim training. Since that is no longer the case, I don't see why MCAS is still needed. Maybe someone with more knowledge on this can enlighten me :)

It's a bit more than making MAX feel like NG, it's the 'force gradient' requirement in the FARs. The change in stick force does decrease at high angles of AoA because the repositioned nacelles generate lift as the AoA increases. The reason the reg exists is because this may induce the pilot to compensate by pulling back the stick even more and causing a stall just because the stick force decreased too rapidly. It's not a Boeing preference, it's a Federal Aviation Rule.

morrisond wrote:
As I wrote which now seems like years ago - while MCAS was needed to meet a reg - sometimes regs need to be looked at and exceptions made in certain cases as they can add too much complexity for very little to no safety benefit(the stick force requirement decreeing that the force must always be increasing and can't plateau).

That would have painted Boeing into a corner, because asking for an exemption would be an admission that MAX did not fly like NG so sim training would have been needed. Boeing wasn't going to go there, thus the invention and certification of MCAS v1. Once Boeing released a very broken MCAS implementation and after the two crashes, I think it they weren't going to then ask FAA to fix it all via pencil whipping nor would FAA be in the mood to do such.

morrisond wrote:
Until such time that AI can be perfect and take over the whole flight - I suggest we reexamine how we train commercial pilots and no more 100 hour wonders which end up in an 737 without having any real experience actually flying and dealing with adverse situations.

Yes, the ST article did loop back to the pilot training topic later in the piece. Surely we haven't heard the last on it. You will note that the new training updates need to be approved before RTS. Then we had Boeing's CEO suggesting that FAA was pushing back on 787 level automation for the NMA cockpit and improvements would be needed.

Personally I think the training issue may be sold short. We have evidence that the combination of pilots and checklists weren't working well in the two crashes and in some of the post-crash sim checks. Personally I think every 737 pilot should be given a sim check focused on the challenges revealed via MCAS such as stab trim runaway and AoA disagree, and if they can't sort them out, they should be retrained till they can or till the company decides they need a better pilot. There's no excuse not too, other than putting money over safety. All the MAXes are grounded and most of the NGs too. Sim training will be mandatory for RTS. It seems like a great time to throw a bunch of testing in at the same time and use this opportunity to improve compliance on familiarity with and use of checklists.
 
User avatar
Revelation
Posts: 24589
Joined: Wed Feb 09, 2005 9:37 pm

Re: Boeing 737MAX Grounding, General Discussion Thread, Q3 2020

Sun Sep 13, 2020 3:20 pm

Dominic Gates quotes a new WSJ article:

"Mr. Teal, who as the MAX’s chief engineer personally signed off on MCAS ... told investigators he approved MCAS without knowing that a malfunction of a single sensor could trigger the system to repeatedly push down a plane’s nose and possibly lead to a crash."

Wow!

Ref: https://twitter.com/dominicgates/status ... 8538324997

No surprise this got a 'Wow!' out of Gates. Teal is the same guy who decided synthetic air speed was too costly and complex to make it onto the MAX. Now he's saying he didn't even know that MCAS only used one sensor?

Anyone got a WSJ subscription and can tell us more about the article at https://www.wsj.com/articles/former-top ... 9?mod=e2tw ???

I'm somewhat tempted to go out and find a copy just to read this one article. Maybe my library has it.
Last edited by Revelation on Sun Sep 13, 2020 3:24 pm, edited 1 time in total.
Wake up to find out that you are the eyes of the world
The heart has its beaches, its homeland and thoughts of its own
Wake now, discover that you are the song that the morning brings
The heart has its seasons, its evenings and songs of its own
 
Sooner787
Posts: 2719
Joined: Thu Jul 18, 2013 1:44 am

Re: Boeing 737MAX Grounding, General Discussion Thread, Q3 2020

Sun Sep 13, 2020 3:24 pm

AndoAv8R wrote:
As we get closer to (hopefully) the MAX certified again, does anyone know where most of the upgrades will be taking place? I'd assume the airlines maintenance hubs will be able to or will they all need to go to a Boeing facility?


AA has already flown their Max 8's from storage in ROW to their huge MX base in TUL.

I've read UA will perform their Max 9 updates at their MCO MX base.

AS for WN, I seem to recall they stored all their Max 8's together in VCV so the mods
could be performed there.
 
kalvado
Posts: 2896
Joined: Wed Mar 01, 2006 4:29 am

Re: Boeing 737MAX Grounding, General Discussion Thread, Q3 2020

Sun Sep 13, 2020 3:24 pm

Spetsnaz55 wrote:
Noshow wrote:
What do we know about flight behavior with MCAS off? Is there any new source please?


https://www.seattletimes.com/business/b ... -now-safe/

What I see is a perfectly rounded PR statement - truth, only truth, nothing but truth - but god forbids telling entire truth!
Stability statement comes from Bjorn - not from certification officials or pilots - and in a different context:
“If MCAS is deactivated, you can still fly the aircraft and it is not unstable,” said Fehrm. “The MAX without MCAS is a perfectly flyable aircraft.”

Bjorn statement could come in a different context that we read it, I don't see a source expanding the statement - or if Bjorn is privileged to testing data at all. The fine print here is the scope - there is no doubt that regular flight envelope doesn't need MCAS, MAX is perfectly flyable etc etc. It is the edge of the envelope where the issue is. So entire truth is still pending.
 
djpearman
Posts: 19
Joined: Thu Feb 06, 2020 9:02 pm

Re: Boeing 737MAX Grounding, General Discussion Thread, Q3 2020

Sun Sep 13, 2020 3:43 pm

morrisond wrote:
That's great that the MAX is Stable even with MCAS turned off. I think we wasted pages on debating what stable is. Without MCAS the MAX is not an unsafe aircraft.


If that is the case, why aren't Boeing simply removing MCAS?

morrisond wrote:
As I wrote which now seems like years ago - while MCAS was needed to meet a reg - sometimes regs need to be looked at and exceptions made in certain cases as they can add too much complexity for very little to no safety benefit(the stick force requirement decreeing that the force must always be increasing and can't plateau).


Two points to this: Firstly, if MCAS was needed to meet a reg, which one do you mean? Secondly, if it was needed to meet a reg, why was it not known before the Lion Air crash? Anything installed in an aircraft in order to fulfil regulatory requirements must be traceable and thus disclosed as part of certification. To my knowledge, Boeing's argument at the time of certification was that MCAS impacts the aircraft's characteristics to such a small extent that noone needs to know about it, even the pilots.

morrisond wrote:
In terms of AOA - The NG has been flying very safely with 2 AOA sensors for decades now - why add unneeded complexity? Let it be. Even with three sensors there are examples of those planes almost going into the ground.


Again, two points: Firstly, as far as I'm aware, the NG only ever flies using one of the two flight computers, alternating between the two available units for each flight. As such, the NG would only ever be using one AOA sensor. Also, the NG never had any flight control system feeding off it - it was purely informational - and thus requires no further complexity. On the MAX, however, a system with the ability to significantly influence the aircraft's flight control was installed that used the AOA as its (only?) input. This requires additional complexity to avoid it acting on a wrong input. On most aircraft with similar systems, the common practice is to use three AOA sensors to be fully immune to the random failure of a single sensor.

Secondly, )'m very interested in what examples you have of aircraft with three sensors almost going to the ground. Can you name them please? The only incident I know of that is directly related to AOA sensors is LH1829. I'd like to know of more these types of incident.

morrisond wrote:
One of the downsides of making a plane too perfect is Pilot's start assuming that it is perfect and that it can do nothing wrong. Instead of trusting their own training (assuming they had half decent training) when something goes wonky with the flightpath they start analyzing the systems vs just flying the aircraft like they are supposed too.


In my opinion, if we start assuming that aircraft are perfect, then we have already lost before we start. Any system not only can but will fail. If we do not start from that assumption, then we will not make good decisions, both in terms of aircraft design as well as pilot training. The problem with MCAS on the MAX is that it was not included anywhere in pilot training so that when it activated, the pilots could no longer just fly the aircraft like they are supposed to.

morrisond wrote:
Until such time that AI can be perfect and take over the whole flight - I suggest we reexamine how we train commercial pilots and no more 100 hour wonders which end up in an 737 without having any real experience actually flying and dealing with adverse situations.


As mentioned earlier, any system will fail - the same goes for AI. So, I wouldn't put too much hope in that eliminating the need for pilots. We humans have a much greater array of inputs (eyes, ears, nose, sense of balance, experience, intuition) and can apply these to unforseen situations, something any programmed system is incapable of.
 
sphealey
Posts: 321
Joined: Tue May 31, 2005 12:39 am

Re: Boeing 737MAX Grounding, General Discussion Thread, Q3 2020

Sun Sep 13, 2020 3:44 pm

BEG2IAH wrote:
They are pulling some tight turns at very high speeds: https://flightaware.com/live/flight/BOE701

Wow - watch the 10-Sep flight starting at 11:18 PDT from Vancouver, Canada on 10x replay. I assume this is the Canada regulatory agency since the flight begins and ends in Canada. Those are some tight circles, mostly at altitude but the last set before landing lower. They didn't fly the course changes filed in the final quarter of the flight so either the regulator was satisfied at that point or someone was too sick to continue!
 
djpearman
Posts: 19
Joined: Thu Feb 06, 2020 9:02 pm

Re: Boeing 737MAX Grounding, General Discussion Thread, Q3 2020

Sun Sep 13, 2020 3:50 pm

Revelation wrote:
It's a bit more than making MAX feel like NG, it's the 'force gradient' requirement in the FARs. The change in stick force does decrease at high angles of AoA because the repositioned nacelles generate lift as the AoA increases. The reason the reg exists is because this may induce the pilot to compensate by pulling back the stick even more and causing a stall just because the stick force decreased too rapidly. It's not a Boeing preference, it's a Federal Aviation Rule.


Ah, didn't know that, thanks! That explains it.

Edit: small side note - angles of AOA? :D
Last edited by djpearman on Sun Sep 13, 2020 4:04 pm, edited 1 time in total.
 
sphealey
Posts: 321
Joined: Tue May 31, 2005 12:39 am

Re: Boeing 737MAX Grounding, General Discussion Thread, Q3 2020

Sun Sep 13, 2020 3:55 pm

"@Revelation: Yes, that's my point. The regs should drive the implementation, not the other way around. If there aren't enough computes left to do the small calculations needed to support a 'virtual AoA' then we could also create an auxillary processor "complex" to do this task. I say complex because it too would need redundancy, and would in turn create small issues with space and power and heat as does every processor, but these can all be solved with money. The amount of processing we're talking about here can be done with an Arduino after all, since we won't be constrained to using 80s vintage CPUs."

To my mind that is a problem that has been revealed that affects the entire industry: they are using 80286 processors not just because they are certified and/or out of inertia, but because those are the last mass market CPUs available that do not use out-of-order and speculative execution. Much of the enormous process power gain of the last few generations of CPUs have derived from design techniques that make them very difficult to certify for realtime systems (we got a hint of this with the MELTDOWN and SPECTRE attacks on CPU shared memory ~2 years ago, and the firmware patches for those vulnerabilities significantly affected performance).

The automobile industry used to design (and for a while manufacture) their own auto-specific CPUs and associated logic chips - perhaps it is time for the aviation industry to commission a custom CPU with better performance that is specifically designed to be validated for realtime use and less vulnerable to radiation and cosmic rays.
 
Noshow
Posts: 1669
Joined: Wed Jun 15, 2016 3:20 pm

Re: Boeing 737MAX Grounding, General Discussion Thread, Q3 2020

Sun Sep 13, 2020 4:06 pm

Thanks for the ST link. But there is no actual observation about flying a modified MAX with MCAS set to off in MCAS conditions.
Waiting for the FAA and EASA flight test protocols.
 
User avatar
Revelation
Posts: 24589
Joined: Wed Feb 09, 2005 9:37 pm

Re: Boeing 737MAX Grounding, General Discussion Thread, Q3 2020

Sun Sep 13, 2020 4:16 pm

sphealey wrote:
"@Revelation: Yes, that's my point. The regs should drive the implementation, not the other way around. If there aren't enough computes left to do the small calculations needed to support a 'virtual AoA' then we could also create an auxillary processor "complex" to do this task. I say complex because it too would need redundancy, and would in turn create small issues with space and power and heat as does every processor, but these can all be solved with money. The amount of processing we're talking about here can be done with an Arduino after all, since we won't be constrained to using 80s vintage CPUs."

To my mind that is a problem that has been revealed that affects the entire industry: they are using 80286 processors not just because they are certified and/or out of inertia, but because those are the last mass market CPUs available that do not use out-of-order and speculative execution. Much of the enormous process power gain of the last few generations of CPUs have derived from design techniques that make them very difficult to certify for realtime systems (we got a hint of this with the MELTDOWN and SPECTRE attacks on CPU shared memory ~2 years ago, and the firmware patches for those vulnerabilities significantly affected performance).

The automobile industry used to design (and for a while manufacture) their own auto-specific CPUs and associated logic chips - perhaps it is time for the aviation industry to commission a custom CPU with better performance that is specifically designed to be validated for realtime use and less vulnerable to radiation and cosmic rays.

Very interesting points. Last I heard, the space industry was using PowerPC 750 chips from the 90s for these same reasons. The 286 are 80s vintage designs, and speculation was coming in during the 90s. From my own knowledge, DEC's first two Alpha chips of the early and mid 90s (EV4, EV5) had multiple issue but did not have speculation or even out-of-order processing, that only came late in the 90s with EV6.
Last edited by Revelation on Sun Sep 13, 2020 4:25 pm, edited 1 time in total.
Wake up to find out that you are the eyes of the world
The heart has its beaches, its homeland and thoughts of its own
Wake now, discover that you are the song that the morning brings
The heart has its seasons, its evenings and songs of its own
 
User avatar
Revelation
Posts: 24589
Joined: Wed Feb 09, 2005 9:37 pm

Re: Boeing 737MAX Grounding, General Discussion Thread, Q3 2020

Sun Sep 13, 2020 4:24 pm

Spetsnaz55 wrote:
Noshow wrote:
What do we know about flight behavior with MCAS off? Is there any new source please?

https://www.seattletimes.com/business/b ... -now-safe/

It says:

To prove the plane’s stability, both Boeing and the FAA test pilots have now conducted extreme flight test maneuvers close to a stall, both with MCAS on and with the system turned off.

So I suppose we can wait to hear more, but we at least know "extreme" flight test maneuvers with MCAS on and off happened, and the tone of the article suggests no surprises were found. Personally I'm not expecting FAA or EASA to say much more than this kind of statement.

kalvado wrote:
Spetsnaz55 wrote:
Noshow wrote:
What do we know about flight behavior with MCAS off? Is there any new source please?

https://www.seattletimes.com/business/b ... -now-safe/

What I see is a perfectly rounded PR statement - truth, only truth, nothing but truth - but god forbids telling entire truth!
Stability statement comes from Bjorn - not from certification officials or pilots - and in a different context:
“If MCAS is deactivated, you can still fly the aircraft and it is not unstable,” said Fehrm. “The MAX without MCAS is a perfectly flyable aircraft.”

Bjorn statement could come in a different context that we read it, I don't see a source expanding the statement - or if Bjorn is privileged to testing data at all. The fine print here is the scope - there is no doubt that regular flight envelope doesn't need MCAS, MAX is perfectly flyable etc etc. It is the edge of the envelope where the issue is. So entire truth is still pending.

Right, but there's also no actual evidence of instability, despite Ralph Nader's dubious claim.

PR-ish statement or not, absence of evidence isn't evidence of absence.

Interesting discussion at https://aviation.stackexchange.com/ques ... ect=1&lq=1 which kind of deliniates what we are talking about and not talking about.
Last edited by Revelation on Sun Sep 13, 2020 4:39 pm, edited 1 time in total.
Wake up to find out that you are the eyes of the world
The heart has its beaches, its homeland and thoughts of its own
Wake now, discover that you are the song that the morning brings
The heart has its seasons, its evenings and songs of its own
 
djpearman
Posts: 19
Joined: Thu Feb 06, 2020 9:02 pm

Re: Boeing 737MAX Grounding, General Discussion Thread, Q3 2020

Sun Sep 13, 2020 4:35 pm

Revelation wrote:
sphealey wrote:
"@Revelation: Yes, that's my point. The regs should drive the implementation, not the other way around. If there aren't enough computes left to do the small calculations needed to support a 'virtual AoA' then we could also create an auxillary processor "complex" to do this task. I say complex because it too would need redundancy, and would in turn create small issues with space and power and heat as does every processor, but these can all be solved with money. The amount of processing we're talking about here can be done with an Arduino after all, since we won't be constrained to using 80s vintage CPUs."

To my mind that is a problem that has been revealed that affects the entire industry: they are using 80286 processors not just because they are certified and/or out of inertia, but because those are the last mass market CPUs available that do not use out-of-order and speculative execution. Much of the enormous process power gain of the last few generations of CPUs have derived from design techniques that make them very difficult to certify for realtime systems (we got a hint of this with the MELTDOWN and SPECTRE attacks on CPU shared memory ~2 years ago, and the firmware patches for those vulnerabilities significantly affected performance).

The automobile industry used to design (and for a while manufacture) their own auto-specific CPUs and associated logic chips - perhaps it is time for the aviation industry to commission a custom CPU with better performance that is specifically designed to be validated for realtime use and less vulnerable to radiation and cosmic rays.

Very interesting points. Last I heard, the space industry was using PowerPC 750 chips from the 90s for these same reasons. The 286 are 80s vintage designs, and speculation was coming in during the 90s. From my own knowledge, DEC's first two Alpha chips of the early and mid 90s (EV4, EV5) had multiple issue but did not have speculation or even out-of-order processing, that only came late in the 90s with EV6.


The work I did on aircraft electronics used microcontrollers with either PowerPC cores (I don't remember exactly which ones, though 750 rings a bell), TI's C2000 architecture or more recently ARM cores, usually fabricated using older lithography processes resulting in better radiation "resistance". In addition, flash-based FPGAs were also being introduced. The only x86 processors involved were in the machines used for development :).
 
kalvado
Posts: 2896
Joined: Wed Mar 01, 2006 4:29 am

Re: Boeing 737MAX Grounding, General Discussion Thread, Q3 2020

Sun Sep 13, 2020 5:22 pm

Revelation wrote:
Spetsnaz55 wrote:
Noshow wrote:
What do we know about flight behavior with MCAS off? Is there any new source please?

https://www.seattletimes.com/business/b ... -now-safe/

It says:

To prove the plane’s stability, both Boeing and the FAA test pilots have now conducted extreme flight test maneuvers close to a stall, both with MCAS on and with the system turned off.

So I suppose we can wait to hear more, but we at least know "extreme" flight test maneuvers with MCAS on and off happened, and the tone of the article suggests no surprises were found. Personally I'm not expecting FAA or EASA to say much more than this kind of statement.

kalvado wrote:
Spetsnaz55 wrote:

What I see is a perfectly rounded PR statement - truth, only truth, nothing but truth - but god forbids telling entire truth!
Stability statement comes from Bjorn - not from certification officials or pilots - and in a different context:
“If MCAS is deactivated, you can still fly the aircraft and it is not unstable,” said Fehrm. “The MAX without MCAS is a perfectly flyable aircraft.”

Bjorn statement could come in a different context that we read it, I don't see a source expanding the statement - or if Bjorn is privileged to testing data at all. The fine print here is the scope - there is no doubt that regular flight envelope doesn't need MCAS, MAX is perfectly flyable etc etc. It is the edge of the envelope where the issue is. So entire truth is still pending.

Right, but there's also no actual evidence of instability, despite Ralph Nader's dubious claim.

PR-ish statement or not, absence of evidence isn't evidence of absence.

Interesting discussion at https://aviation.stackexchange.com/ques ... ect=1&lq=1 which kind of deliniates what we are talking about and not talking about.

The only hard evidence we have is existence of mcas. My gut feeling is that Boeing and authorities are playing with the fire here in terms of public trust.
 
StTim
Posts: 3732
Joined: Thu Aug 08, 2013 7:39 am

Re: Boeing 737MAX Grounding, General Discussion Thread, Q3 2020

Sun Sep 13, 2020 5:44 pm

morrisond wrote:
Until such time that AI can be perfect and take over the whole flight - I suggest we reexamine how we train commercial pilots and no more 100 hour wonders which end up in an 737 without having any real experience actually flying and dealing with adverse situations.


As mentioned earlier, any system will fail - the same goes for AI. So, I wouldn't put too much hope in that eliminating the need for pilots. We humans have a much greater array of inputs (eyes, ears, nose, sense of balance, experience, intuition) and can apply these to unforseen situations, something any programmed system is incapable of.[/quote]

My question is would Boeing have told AI developers that MCAS existed and what actions should be taken?

Or are you still following the Boeing line that if the pilots had responded to a novel incident (correctly) in 3 seconds all would have been fine?
 
Noshow
Posts: 1669
Joined: Wed Jun 15, 2016 3:20 pm

Re: Boeing 737MAX Grounding, General Discussion Thread, Q3 2020

Sun Sep 13, 2020 5:58 pm

The plane WILL be flown by average pilots. It must be safe enough for them.
 
Chemist
Posts: 760
Joined: Tue Oct 20, 2015 4:46 am

Re: Boeing 737MAX Grounding, General Discussion Thread, Q3 2020

Sun Sep 13, 2020 6:18 pm

morrisond wrote:
That's great that the MAX is Stable even with MCAS turned off. I think we wasted pages on debating what stable is. Without MCAS the MAX is not an unsafe aircraft.

As I wrote which now seems like years ago - while MCAS was needed to meet a reg - sometimes regs need to be looked at and exceptions made in certain cases as they can add too much complexity for very little to no safety benefit(the stick force requirement decreeing that the force must always be increasing and can't plateau).

In terms of AOA - The NG has been flying very safely with 2 AOA sensors for decades now - why add unneeded complexity? Let it be. Even with three sensors there are examples of those planes almost going into the ground.

One of the downsides of making a plane too perfect is Pilot's start assuming that it is perfect and that it can do nothing wrong. Instead of trusting their own training (assuming they had half decent training) when something goes wonky with the flightpath they start analyzing the systems vs just flying the aircraft like they are supposed too.

Until such time that AI can be perfect and take over the whole flight - I suggest we reexamine how we train commercial pilots and no more 100 hour wonders which end up in an 737 without having any real experience actually flying and dealing with adverse situations.


Re: stick force. Do the FBW Airbus aircraft have stick force gradients, when, say, approaching a stall? (Yes I know there are envelope protections). I assume if so that would be just a spring on the joystick, or is some sort of active synthetic feedback used?
 
User avatar
Revelation
Posts: 24589
Joined: Wed Feb 09, 2005 9:37 pm

Re: Boeing 737MAX Grounding, General Discussion Thread, Q3 2020

Sun Sep 13, 2020 9:05 pm

kalvado wrote:
The only hard evidence we have is existence of mcas. My gut feeling is that Boeing and authorities are playing with the fire here in terms of public trust.

The evidence we have is FAA has done its tests without asking for retests or changing the time line, unlike earlier milestones such as the software review where problems were found and fixes were needed. Then we have both TC and EASA deciding to go forward with testing presumably with some access to earlier test data from Boeing and FAA, and definitely after doing a week of sim tests each with the new software load. The odds of there being a glaring issue with stability (either dynamic or static) at this point in the game seem very remote.

Noshow wrote:
The plane WILL be flown by average pilots. It must be safe enough for them.

True. The narrative in the ST piece still seems to be looking backwards at the NG's record and projecting it forward, rather than taking a baseline of all the current pilots flying the MAX then projecting that forward.

Hopefully we hear more as a side effect of the JOEB training review starting Monday in London.

The next steps are:

After the nine-day review, the results will be incorporated into the draft FAA Flight Standardization Board report, which will be then be open for public comment.

Then, FAA Administrator Steve Dickson will undergo recommended training and conduct an evaluation flight at the controls of a Boeing 737 MAX. He will share observations with FAA technical staff.

The FAA will then review Boeing’s final design documentation to evaluate compliance with FAA regulations. The multi-agency technical advisory board will review the Boeing submission and issue a report prior to a final FAA determination of compliance.

The FAA will then issue a notice of pending significant safety actions and publish a final directive addressing known issues for grounding and advises operators of required corrective actions before aircraft may re-enter commercial service.

Then the FAA plans to rescind the grounding order.

Ref: https://in.reuters.com/article/uk-boein ... KKBN2622VL

So we'll get to review the draft of the FSB report, which should make for interesting reading two weeks or so from now.

That article quotes a Boeing spokesperson as saying they expect to resume MAX deliveries in Q4.
Wake up to find out that you are the eyes of the world
The heart has its beaches, its homeland and thoughts of its own
Wake now, discover that you are the song that the morning brings
The heart has its seasons, its evenings and songs of its own
 
kalvado
Posts: 2896
Joined: Wed Mar 01, 2006 4:29 am

Re: Boeing 737MAX Grounding, General Discussion Thread, Q3 2020

Sun Sep 13, 2020 10:04 pm

Revelation wrote:
kalvado wrote:
The only hard evidence we have is existence of mcas. My gut feeling is that Boeing and authorities are playing with the fire here in terms of public trust.

The evidence we have is FAA has done its tests without asking for retests or changing the time line, unlike earlier milestones such as the software review where problems were found and fixes were needed. Then we have both TC and EASA deciding to go forward with testing presumably with some access to earlier test data from Boeing and FAA, and definitely after doing a week of sim tests each with the new software load. The odds of there being a glaring issue with stability (either dynamic or static) at this point in the game seem very remote.

It is about the scale of the issue and risk analysis more than anything else. Not much about it. As far as I can tell from the tracks, stall recovery was not on testing agenda.
 
Opus99
Posts: 1080
Joined: Thu May 30, 2019 10:51 pm

Re: Boeing 737MAX Grounding, General Discussion Thread, Q3 2020

Sun Sep 13, 2020 10:14 pm

kalvado wrote:
Revelation wrote:
kalvado wrote:
The only hard evidence we have is existence of mcas. My gut feeling is that Boeing and authorities are playing with the fire here in terms of public trust.

The evidence we have is FAA has done its tests without asking for retests or changing the time line, unlike earlier milestones such as the software review where problems were found and fixes were needed. Then we have both TC and EASA deciding to go forward with testing presumably with some access to earlier test data from Boeing and FAA, and definitely after doing a week of sim tests each with the new software load. The odds of there being a glaring issue with stability (either dynamic or static) at this point in the game seem very remote.

It is about the scale of the issue and risk analysis more than anything else. Not much about it. As far as I can tell from the tracks, stall recovery was not on testing agenda.

The FAA carried out no stall tests? Is that correct?
 
kalvado
Posts: 2896
Joined: Wed Mar 01, 2006 4:29 am

Re: Boeing 737MAX Grounding, General Discussion Thread, Q3 2020

Sun Sep 13, 2020 10:23 pm

Opus99 wrote:
kalvado wrote:
Revelation wrote:
The evidence we have is FAA has done its tests without asking for retests or changing the time line, unlike earlier milestones such as the software review where problems were found and fixes were needed. Then we have both TC and EASA deciding to go forward with testing presumably with some access to earlier test data from Boeing and FAA, and definitely after doing a week of sim tests each with the new software load. The odds of there being a glaring issue with stability (either dynamic or static) at this point in the game seem very remote.

It is about the scale of the issue and risk analysis more than anything else. Not much about it. As far as I can tell from the tracks, stall recovery was not on testing agenda.

The FAA carried out no stall tests? Is that correct?

They should as part of initial certification. I don't see that in recent MCAS focused batch. Maybe I just missed those

Popular Searches On Airliners.net

Top Photos of Last:   24 Hours  •  48 Hours  •  7 Days  •  30 Days  •  180 Days  •  365 Days  •  All Time

Military Aircraft Every type from fighters to helicopters from air forces around the globe

Classic Airliners Props and jets from the good old days

Flight Decks Views from inside the cockpit

Aircraft Cabins Passenger cabin shots showing seat arrangements as well as cargo aircraft interior

Cargo Aircraft Pictures of great freighter aircraft

Government Aircraft Aircraft flying government officials

Helicopters Our large helicopter section. Both military and civil versions

Blimps / Airships Everything from the Goodyear blimp to the Zeppelin

Night Photos Beautiful shots taken while the sun is below the horizon

Accidents Accident, incident and crash related photos

Air to Air Photos taken by airborne photographers of airborne aircraft

Special Paint Schemes Aircraft painted in beautiful and original liveries

Airport Overviews Airport overviews from the air or ground

Tails and Winglets Tail and Winglet closeups with beautiful airline logos