Go Back   English Forum Switzerland > Off-Topic > Off-Topic > International affairs/politics  
Reply
 
Thread Tools Display Modes
  #141  
Old 12.04.2019, 03:52
Newbie 1st class
 
Join Date: Apr 2016
Location: Anchorage and Seattle
Posts: 22
Groaned at 2 Times in 2 Posts
Thanked 43 Times in 22 Posts
AlaskaGuy has earned some respectAlaskaGuy has earned some respect
Re: Ethiopian Airlines ET 302 crashed near Addis Ababa

From today/Thursday's New York Times:

https://www.nytimes.com/2019/04/11/b...-faa-mcas.html

"After the agency’s initial safety review, Boeing decided to quadruple the power of the anti-stall system, which could push down the plane’s nose. The company also expanded the use of the software, known as MCAS, to activate in more situations.

"Although officials were aware of the changes, none were fully examined by the F.A.A., according to three people with knowledge of the process.

"A new review would have required F.A.A. officials to take a closer look at the system’s effect on the overall safety of the plane, as well as to consider the potential consequences of a malfunction. Instead, the agency relied on an earlier assessment of the system, which was less powerful and activated in more limited circumstances."
Reply With Quote
This user would like to thank AlaskaGuy for this useful post:
  #142  
Old 12.04.2019, 18:53
me.anon's Avatar
Forum Veteran
 
Join Date: Jan 2012
Location: thun
Posts: 1,808
Groaned at 36 Times in 25 Posts
Thanked 2,291 Times in 1,108 Posts
me.anon has a reputation beyond reputeme.anon has a reputation beyond reputeme.anon has a reputation beyond reputeme.anon has a reputation beyond reputeme.anon has a reputation beyond repute
Re: Ethiopian Airlines ET 302 crashed near Addis Ababa

Quote:
View Post
From today/Thursday's New York Times:

https://www.nytimes.com/2019/04/11/b...-faa-mcas.html

"After the agency’s initial safety review, Boeing decided to quadruple the power of the anti-stall system, which could push down the plane’s nose. The company also expanded the use of the software, known as MCAS, to activate in more situations.

"Although officials were aware of the changes, none were fully examined by the F.A.A., according to three people with knowledge of the process.

"A new review would have required F.A.A. officials to take a closer look at the system’s effect on the overall safety of the plane, as well as to consider the potential consequences of a malfunction. Instead, the agency relied on an earlier assessment of the system, which was less powerful and activated in more limited circumstances."
The article itself doesn't seem to bring much that is new, but the reader comments, especially the later ones, are extremely unsympathetic towards Boeing with many readers vowing never to fly in a Max again (not entirely surprising). And those comments which combined negative statements against Boeing with the Republican "less government" (therefore less FAA) mantra seemed to get the highest recommendation.

I'm guessing after reading that lot that the Boeing CEO is soon going to be announcing another cut in production, naturally and of course to be presented as allowing even more resources to be freed up for solving the current problems.

On a technical level, there appears to be some lack of consensus between the various articles about MCAS and whether it reads only one sensor (the left hand one I've seen somewhere) or whether it reads both and acts on either one if it shows a reading above a threshold. Whatever, that system (MCAS) itself and the design methodologies (if any) which were used in its development are going to be an important case study for engineering students for years to come.

It is also dawning on me how bad that whole MAX design is. The oversized engines making the plane inherently unstable and the badly executed MCAS design to attempt to mitigate this are one thing. But there are 2 other things that have become clear to me. One is the lack of integration between the various systems and the other is the unclean layering of the systems in varying degrees of pilot autonomy. Example1: one system can scream "Terrain, Terrain" and another can be simultaneously forcing the nose further towards terrain. Example 2: The pilots can take manual control of the aircraft by switching off the autopilot but then another layer of automation kicks in (MCAS again) subverting the manual control, and even without any indication that it is active.
Neither, it appears, is any other input considered by MCAS. Surely if it read an altimeter or accessed an attitude sensor or other similar device, it could have rejected implausible data from a faulty angle of attack sensor.

Also I'd be interested in seeing statistics on the quality of the angle of attack sensors used and the number of replacements that have been fitted. Those aircraft which had been fitted with the option of an indicator to warn of a discrepancy in the angle of attack sensor readings will have shown something concrete to report, others probably not. Many modern devices, for example PC hard disks, are capable of monitoring their own quality and can give an indication of an impending failure. I guess the Boeing Max with is early generation electronics is incapable of collecting quality statistics from attached devices and producing a maintenance "todo" list.
__________________
If you have difficulties with a post which contains a link to a site in one of the Swiss languages, use Google Translate or your own favourite translating browser.
Reply With Quote
This user would like to thank me.anon for this useful post:
  #143  
Old 12.04.2019, 21:41
Forum Veteran
 
Join Date: Oct 2014
Location: Nyon
Posts: 1,407
Groaned at 53 Times in 35 Posts
Thanked 1,296 Times in 661 Posts
bowlie has a reputation beyond reputebowlie has a reputation beyond reputebowlie has a reputation beyond reputebowlie has a reputation beyond repute
Re: Ethiopian Airlines ET 302 crashed near Addis Ababa

But, let's face it. The powers that be have allowed there to be only two aircraft manufacturers. Airbus and Boeing are both the results of mergers and more mergers with smaller, competing, producers to give us the behemoths that we now have.

Sure the Canadians, Russians, Brazilians and Chinese have some production, but in niche markets.

In the US, Boeing is critical to the military and the space program amongst other things. The US is not going to let Boeing fail, no way, no how.

Something similar in Europe, although Airbus is less tied to the military but they have long arms into numerous industries across the continent.

Lots of jobs, and votes, at stake. We are stuck with this Duopoly

And we likely will be flying on B737 variants for decades to come, but likely under a different 'new' name.
Reply With Quote
  #144  
Old 13.04.2019, 08:26
Newbie 1st class
 
Join Date: Apr 2016
Location: Anchorage and Seattle
Posts: 22
Groaned at 2 Times in 2 Posts
Thanked 43 Times in 22 Posts
AlaskaGuy has earned some respectAlaskaGuy has earned some respect
Re: Ethiopian Airlines ET 302 crashed near Addis Ababa

These voice warnings:

Excessive descent rate ("SINK RATE" "PULL UP")
Excessive terrain closure rate ("TERRAIN" "PULL UP")
Altitude loss after take off or with a high power setting ("DON'T SINK")
Unsafe terrain clearance ("TOO LOW – TERRAIN" "TOO LOW – GEAR" "TOO LOW – FLAPS")
Excessive deviation below glideslope ("GLIDESLOPE")
Excessively steep bank angle ("BANK ANGLE")
Windshear protection ("WINDSHEAR")

are from a system called EGPWS (Enhanced Ground Proximity Warning System) as described here:

https://en.wikipedia.org/wiki/Ground...warning_system

They're just warnings -- EGPWS can't fly the airplane.

As far as which AoA sensor MCAS monitors, everything I've read is one, BUT I've read several places that it alternates from side to side on successive flights. That may seem odd, but I vaguely recall something similar on some system on some Boeing airplane in the past. I never heard a reason (nor did I ever ask) why anything would alternate side-to-side.

I've read that AoA sensors on the MAX are made by Rosemount. When I was at Boeing, I regularly heard about ice-over problems with Rosemount TAT (Total Air Temperature) probes. TAT has something to do with calculating TAS (true airspeed).

Last edited by AlaskaGuy; 13.04.2019 at 09:09.
Reply With Quote
This user would like to thank AlaskaGuy for this useful post:
  #145  
Old 13.04.2019, 10:25
marton's Avatar
Forum Legend
 
Join Date: May 2008
Location: Kt. Zürich
Posts: 10,597
Groaned at 342 Times in 295 Posts
Thanked 13,604 Times in 7,476 Posts
marton has a reputation beyond reputemarton has a reputation beyond reputemarton has a reputation beyond reputemarton has a reputation beyond reputemarton has a reputation beyond reputemarton has a reputation beyond repute
Re: Ethiopian Airlines ET 302 crashed near Addis Ababa

Apart from other issues, it is always a nightmare trying add new functionality to IT legacy systems.
Especially if the legacy systems were not well documented or if the documentation has gone astray!

Looks like a long haul, even if they manage to come up with a solution soon then the testing and certification is likely to run for a long time!
Reply With Quote
This user would like to thank marton for this useful post:
  #146  
Old 13.04.2019, 11:02
me.anon's Avatar
Forum Veteran
 
Join Date: Jan 2012
Location: thun
Posts: 1,808
Groaned at 36 Times in 25 Posts
Thanked 2,291 Times in 1,108 Posts
me.anon has a reputation beyond reputeme.anon has a reputation beyond reputeme.anon has a reputation beyond reputeme.anon has a reputation beyond reputeme.anon has a reputation beyond repute
Re: Ethiopian Airlines ET 302 crashed near Addis Ababa

Quote:
View Post
These voice warnings:

Excessive descent rate ("SINK RATE" "PULL UP")
Excessive terrain closure rate ("TERRAIN" "PULL UP")
Altitude loss after take off or with a high power setting ("DON'T SINK")
Unsafe terrain clearance ("TOO LOW – TERRAIN" "TOO LOW – GEAR" "TOO LOW – FLAPS")
Excessive deviation below glideslope ("GLIDESLOPE")
Excessively steep bank angle ("BANK ANGLE")
Windshear protection ("WINDSHEAR")

are from a system called EGPWS (Enhanced Ground Proximity Warning System) as described here:

https://en.wikipedia.org/wiki/Ground...warning_system

They're just warnings -- EGPWS can't fly the airplane.

As far as which AoA sensor MCAS monitors, everything I've read is one, BUT I've read several places that it alternates from side to side on successive flights. That may seem odd, but I vaguely recall something similar on some system on some Boeing airplane in the past. I never heard a reason (nor did I ever ask) why anything would alternate side-to-side.

I've read that AoA sensors on the MAX are made by Rosemount. When I was at Boeing, I regularly heard about ice-over problems with Rosemount TAT (Total Air Temperature) probes. TAT has something to do with calculating TAS (true airspeed).


That I didn't understand fully before, that the way that MCAS took control of the plane was not consistent with normal aviation practice. Normal would have been a simple "augmention" of the feel of the pilot's control stick [if I understand correctly from the bit you appear to have edited out since I first looked]. However, the chosen method was an unnatural direct control of the flight surfaces. I guess that taking the pilot out of the loop goes back once again to the commercial decision to minimise (a) pilot re-training requirements and (b) certification activities. This had the direct effect that normal design and verification practices and processes were circumvented, and that in the case of the sensor malfunction the pilots were totally unprepared for the behaviour of the aircraft.


Quote:
View Post
Apart from other issues, it is always a nightmare trying add new functionality to IT legacy systems.
Especially if the legacy systems were not well documented or if the documentation has gone astray!

Looks like a long haul, even if they manage to come up with a solution soon then the testing and certification is likely to run for a long time!
I have some experience of this in the telecoms world. If new functionality is to be added, it is inevitable the old software blocks have to be updated. This is a time consuming and costly procedure because huge processes kick in. The tendency is to open the absolute minimum number of old blocks and pack in as much functionality as possible into these open blocks no matter how inappropriate it is to the natural function of the open block and irrespective of the fact that that functionality better belongs to another block. You can only do that a few times and you have a huge mess.
__________________
If you have difficulties with a post which contains a link to a site in one of the Swiss languages, use Google Translate or your own favourite translating browser.
Reply With Quote
This user would like to thank me.anon for this useful post:
  #147  
Old 15.04.2019, 01:09
Newbie 1st class
 
Join Date: Apr 2016
Location: Anchorage and Seattle
Posts: 22
Groaned at 2 Times in 2 Posts
Thanked 43 Times in 22 Posts
AlaskaGuy has earned some respectAlaskaGuy has earned some respect
Re: Ethiopian Airlines ET 302 crashed near Addis Ababa

From today/Sunday's Reuters(UK):

https://uk.reuters.com/article/us-et...-idUKKCN1RQ0G0

Headline: ""Too Many Travelers, Too Few Planes, is US Airlines' 737 MAX Summer Dilemma"

Specifically:

"To compensate [for the MAX groundings], global MAX operators have added a flight or two to other aircrafts' daily schedules and deferred some non-essential maintenance work."

I'd like to know what kind of "non-essential maintenance" will now be delayed, presumably for the whole summer season. And who gave airlines permission to further stretch out their finely-tuned maintenance schedules, which are already under "progressive maintenance," which is "a continuous maintenance program whereby the required FAA and manufacturer inspections are accomplished during the most convenient time, while keeping the aircraft in a state of continuous airworthiness."

FAA? Not likely. Given that FAA is letting airplane manufacturers self-regulate, it's likely letting airplane operators do the same.

Skimping on or delaying maintenance kills people:

https://en.wikipedia.org/wiki/Alaska...ance_intervals

It's well known that you just can't keep short-haul airplanes like the 737 in the air more than about 11 hours/day. Southwest might stretch that a bit with its "cattle class" turnarounds, but adding even one more daily flight is insane.

Last edited by AlaskaGuy; 15.04.2019 at 01:42.
Reply With Quote
  #148  
Old 15.04.2019, 06:10
slammer's Avatar
Forum Legend
 
Join Date: Jul 2009
Location: Lummerland
Posts: 4,656
Groaned at 116 Times in 81 Posts
Thanked 7,842 Times in 2,926 Posts
slammer has a reputation beyond reputeslammer has a reputation beyond reputeslammer has a reputation beyond reputeslammer has a reputation beyond reputeslammer has a reputation beyond reputeslammer has a reputation beyond repute
Re: Ethiopian Airlines ET 302 crashed near Addis Ababa

Quote:
View Post
From today/Sunday's Reuters(UK):

https://uk.reuters.com/article/us-et...-idUKKCN1RQ0G0

Headline: ""Too Many Travelers, Too Few Planes, is US Airlines' 737 MAX Summer Dilemma"

Specifically:

"To compensate [for the MAX groundings], global MAX operators have added a flight or two to other aircrafts' daily schedules and deferred some non-essential maintenance work."

I'd like to know what kind of "non-essential maintenance" will now be delayed, presumably for the whole summer season. And who gave airlines permission to further stretch out their finely-tuned maintenance schedules, which are already under "progressive maintenance," which is "a continuous maintenance program whereby the required FAA and manufacturer inspections are accomplished during the most convenient time, while keeping the aircraft in a state of continuous airworthiness."

FAA? Not likely. Given that FAA is letting airplane manufacturers self-regulate, it's likely letting airplane operators do the same.

Skimping on or delaying maintenance kills people:

https://en.wikipedia.org/wiki/Alaska...ance_intervals

It's well known that you just can't keep short-haul airplanes like the 737 in the air more than about 11 hours/day. Southwest might stretch that a bit with its "cattle class" turnarounds, but adding even one more daily flight is insane.
Sounds like we have reached "peak air traffic"
Reply With Quote
This user would like to thank slammer for this useful post:
  #149  
Old 15.04.2019, 11:09
Forum Veteran
 
Join Date: Oct 2014
Location: Nyon
Posts: 1,407
Groaned at 53 Times in 35 Posts
Thanked 1,296 Times in 661 Posts
bowlie has a reputation beyond reputebowlie has a reputation beyond reputebowlie has a reputation beyond reputebowlie has a reputation beyond repute
Re: Ethiopian Airlines ET 302 crashed near Addis Ababa

Quote:
View Post
Sounds like we have reached "peak air traffic"
And prices will increase. Law of supply and demand.

Book early ...
Reply With Quote
This user would like to thank bowlie for this useful post:
  #150  
Old Yesterday, 12:50
me.anon's Avatar
Forum Veteran
 
Join Date: Jan 2012
Location: thun
Posts: 1,808
Groaned at 36 Times in 25 Posts
Thanked 2,291 Times in 1,108 Posts
me.anon has a reputation beyond reputeme.anon has a reputation beyond reputeme.anon has a reputation beyond reputeme.anon has a reputation beyond reputeme.anon has a reputation beyond repute
Re: Ethiopian Airlines ET 302 crashed near Addis Ababa

Quote:
View Post
. . .
As far as which AoA sensor MCAS monitors, everything I've read is one, BUT I've read several places that it alternates from side to side on successive flights. That may seem odd, but I vaguely recall something similar on some system on some Boeing airplane in the past. I never heard a reason (nor did I ever ask) why anything would alternate side-to-side.
. . .
Ok. Thanks. In the mean time I have found this. This guy https://www.youtube.com/watch?v=zGM0V7zEKEQ stated that a single angle of attack sensor is used on a specific flight (around 6:40 in the video). On the next flight, the other one is used and so on. It does appear though, he is quite "Boeing-near" and supports the presentation of MCAS, not as a stall prevention system to address stability problems inherent in the design of the aircraft, but simply as a "handling qualities" function to better emulate the behaviour of older models.

Anyway, what I am curious about is the testing procedure prior to certification process and I'm assuming it would be good (or even a requirement) to test the effects of a sensor malfunction such as the angle of attack sensor, but I'd like to know how this testing is done.

For a real flight test, I could imagine intercepting the data from the sensors and feeding failure test cases into the flight control computer (sensor stuck in various positions, sensor dead, sensor oscillating wildly etc.) is about the only way. It could be risky, though.

In a simulator, the sort that pilots would train on, if the simulator accurately enough models the real environment under which the angle of attack sensor operates (same software level as in the actual flight control computer etc.), I guess some mechanical device could operate the angle of attack sensor forcing it through certain test cases. Assuming, that is, a pair of physical angle of attack sensors exists in a simulator. If not, some computer model of the sensor could be used to create the failure case inputs.
__________________
If you have difficulties with a post which contains a link to a site in one of the Swiss languages, use Google Translate or your own favourite translating browser.
Reply With Quote
Reply




Currently Active Users Viewing This Thread: 1 (0 members and 1 guests)
 
Thread Tools
Display Modes

Posting Rules
You may not post new threads
You may not post replies
You may not post attachments
You may not edit your posts

BB code is On
Smilies are On
[IMG] code is On
HTML code is Off
Trackbacks are Off
Pingbacks are Off
Refbacks are Off


Similar Threads
Thread Thread Starter Forum Replies Last Post
Ethiopian Food! JLF Social events 57 10.03.2019 09:42
Malaysian Airlines MH 17 has crashed over Ukraine telandy International affairs/politics 545 24.05.2018 18:09
Job for (LPIC-302) Cetified System Specialist (Linux) TuttiX2 Employment 3 04.10.2012 16:57
Ethiopian Dinner HashBrown Social events 26 26.08.2010 21:38
Ethiopian Food, Anyone? JLF Social events 9 29.04.2010 13:36


All times are GMT +2. The time now is 20:16.


Powered by vBulletin® Version 3.8.4
Copyright ©2000 - 2019, Jelsoft Enterprises Ltd.
LinkBacks Enabled by vBSEO 3.1.0