Go Back  PPRuNe Forums > Flight Deck Forums > Rumours & News
Reload this Page >

Atlas Air 767 down/Texas

Rumours & News Reporting Points that may affect our jobs or lives as professional pilots. Also, items that may be of interest to professional pilots.

Atlas Air 767 down/Texas

Old 25th Feb 2019, 16:08
  #141 (permalink)  
 
Join Date: Oct 2017
Location: Vienna
Posts: 143
Likes: 0
Received 0 Likes on 0 Posts
Originally Posted by SeenItAll View Post
Not at all. Statistics is about accepting or rejecting hypotheses that are made in advance of knowing the data. Claimed confirmation of a hypothesis that was generated after seeing the data is just cherry-picking. Given there are so many parameters to cherry-pick from, it is sure that you can find some that will all align between the incidents -- but this has no probative value as statistical inference. Indeed, each of these crashes is so different from one another in terms of the stage of flight, that I doubt any similarity will be found. And note that trim problems generally don't cause an immediate nosedive into the ground. You generally see the flight crew struggling with the imbalances for some minutes (or at least multiple seconds) before control is finally lost. That doesn't seem to be the case with the altitude profile shown here.
Not at all. Indeed the p value (if one is to use the dreaded p value in order to accept hypothesis) should be adjusted and fairly small. In essence, we could go in blind over all data available, but adjust the p value according to degrees of freedom available to avoid false positives.

Now given that the flight data sample is huge (say, all flights in the past 3 years), the probability of false positive is actually small anyway. So is there a statistically significant difference between # of flights that did/didn't crash in the group of flights which are Boeing in low visibility on final/approach?

And btw, picking a hypothesis up front makes no difference. Imagine 20 different statisticians would decide to test a hypothesis over flight data, each unbeknown and independent of each other. Each would choose a different parameter, but p of only 0.05. Those results would not be different, but there would indeed likely be at least 1 false positive due to stupidly large p value (0.05 is really astronomical when you think about it).
derjodel is offline  
Old 25th Feb 2019, 16:31
  #142 (permalink)  
 
Join Date: May 2010
Location: Usually on top
Posts: 135
Likes: 0
Received 0 Likes on 0 Posts
ADS-B data from FR24 gives GPS derived speed, which is the ground speed, i.e. the speed of the aircraft in the horizontal plane only as seen from straight above. If you're tossing a stone off a tower by throwing it away from the tower, it will retain most of the forward velocity but vertical velocity will continue to increase until it hits the ground, thus creating parabolic flight path. An aircraft without (sufficient) elevator authority will do the same. The impact velocity is far greater than the horizontal velocity would indicate, it's the vector product of both the horizontal and vertical speed components.

Lion Air was a very different case. Fly Dubai and this Atlas crash look a lot more similar in the end phase - which bears absolutely no meaning on the root cause. All it says is: elevator authority was impaired. Period.
physicus is offline  
Old 25th Feb 2019, 16:48
  #143 (permalink)  
 
Join Date: Feb 2019
Location: Alabama
Posts: 3
Likes: 0
Received 0 Likes on 0 Posts
Going to be lots of human factors focus here. Interactions between pilots, between controllers, between pilots and controllers.

Why the CA/PF was on the radio descending through 17,800' when Approach cleared them for the 26L transition (they were actually told that they "can" fly the 26L, pilot appeared to say "91" but did not read back the transition change).
What was the FO/PM working on.
Why they were approaching RDFSH at 6000' and not GARRR at 7000' (sure, Approach thought 3591 was deviating for weather, although they hadn't cleared them for it, and Approach didn't ding them for being at 6000' instead of 7000' for GARRR).
Why the constant 240KTS in the descent with the parabolic drop-off at the end, were they in speed mode, what altitude was set.
Did they have a high rate of descent result from pulling power while trying to decelerate, did the 30,000 FPM final rate of descent result from a departure from controlled flight-- maybe with parts coming off-- in the effort required ("PULL!") to bring the nose up from 18,000 FPM through 11,000 FPM, which is where the bottom fell out.

Not throwing stones here. I live in a glass house of my own making. Just looking for the most obvious/likely causal factors. Occam's Razor. Whatever you do, don't ask how I can envision this scenario happening.

I've worked through it a number of times and have a more lengthy write-up, that is, if you want some more conjecture as to what happened. Guys just out trying to make an honest living, and the whole thing went to hell on them. Whole thing makes me want to puke.
AnyOldPilot is offline  
Old 25th Feb 2019, 17:24
  #144 (permalink)  
 
Join Date: Aug 2002
Location: Zone of Alienation
Age: 78
Posts: 9
Likes: 0
Received 0 Likes on 0 Posts
The 240 ground speed would roughly equal 210 KCAS at 6000’, which is the published speed and altitude at RDFSH point (if in fact that was being navigated to and autothrottle managing speed).

At that speed, clean wing in light to moderate chop (as reported by another arrival), that could be approaching a critical AOA. Of course nothing is known here- conditions, config, etc.; nothing.
FIRESYSOK is offline  
Old 25th Feb 2019, 17:31
  #145 (permalink)  
 
Join Date: Jan 2008
Location: Reading, UK
Posts: 14,633
Likes: 0
Received 29 Likes on 13 Posts
Originally Posted by physicus View Post
ADS-B data from FR24 gives GPS derived speed, which is the ground speed, i.e. the speed of the aircraft in the horizontal plane only as seen from straight above. If you're tossing a stone off a tower by throwing it away from the tower, it will retain most of the forward velocity but vertical velocity will continue to increase until it hits the ground, thus creating parabolic flight path.
Yes, although the aircraft's trajectory, with at least some thrust being generated, would differ slightly from the purely ballistic trajectory of a thrown stone.

The last few points of the ADS-B data would suggest a flightpath angle approaching -50° and a resolved velocity approaching 400 kts at the last recorded point. That's consistent with the relatively small debris field.
DaveReidUK is offline  
Old 25th Feb 2019, 18:10
  #146 (permalink)  
 
Join Date: Oct 2017
Location: Vienna
Posts: 143
Likes: 0
Received 0 Likes on 0 Posts
Originally Posted by physicus View Post
ADS-B data from FR24 gives GPS derived speed, which is the ground speed
Of course! How could I have missed that. Sorry!
derjodel is offline  
Old 25th Feb 2019, 18:50
  #147 (permalink)  
 
Join Date: Nov 2005
Location: Potomac Heights
Posts: 469
Likes: 0
Received 0 Likes on 0 Posts
Originally Posted by derjodel View Post
Not at all. Indeed the p value (if one is to use the dreaded p value in order to accept hypothesis) should be adjusted and fairly small. In essence, we could go in blind over all data available, but adjust the p value according to degrees of freedom available to avoid false positives.

Now given that the flight data sample is huge (say, all flights in the past 3 years), the probability of false positive is actually small anyway. So is there a statistically significant difference between # of flights that did/didn't crash in the group of flights which are Boeing in low visibility on final/approach?

And btw, picking a hypothesis up front makes no difference. Imagine 20 different statisticians would decide to test a hypothesis over flight data, each unbeknown and independent of each other. Each would choose a different parameter, but p of only 0.05. Those results would not be different, but there would indeed likely be at least 1 false positive due to stupidly large p value (0.05 is really astronomical when you think about it).
I really don't want to argue with you, but using a tighter p-value is no cure for improper statistical design. You have no model for the distribution of the errors because you chose a hypothesis to fit your limited choice of data. Indeed, the hypothesis you chose fit the data exactly -- so you have a degenerate error distribution. Further, you seem to have a very limited view as to what the relevant data are. For example, the XL Air crash at Perpignan was also a nose-over -- but that was a A320. Your "model" cannot explain that at all. https://en.wikipedia.org/wiki/XL_Air...ny_Flight_888T
SeenItAll is offline  
Old 25th Feb 2019, 18:53
  #148 (permalink)  
 
Join Date: Feb 2011
Location: Leeds
Posts: 1
Likes: 0
Received 0 Likes on 0 Posts
Why the CA/PF was on the radio descending through 17,800'
There a loads of reasons for this, PM could have been on the other radio talking to Ground services/company or updating the new weather, or anything, how many times in a flight do you end up saying "can you just take "one" a second while i do XXXX"
Livesinafield is offline  
Old 25th Feb 2019, 19:53
  #149 (permalink)  
 
Join Date: Feb 2019
Location: Alabama
Posts: 3
Likes: 0
Received 0 Likes on 0 Posts
Originally Posted by Livesinafield View Post
There a loads of reasons for this, PM could have been on the other radio talking to Ground services/company or updating the new weather, or anything, how many times in a flight do you end up saying "can you just take "one" a second while i do XXXX"
Not disagreeing with you on this. Of course we all do it. Just saying it’s going to be part of the board’s figuring out what happened and why the jet was where it was.
AnyOldPilot is offline  
Old 25th Feb 2019, 20:09
  #150 (permalink)  
 
Join Date: Jun 2001
Location: Rockytop, Tennessee, USA
Posts: 5,899
Likes: 0
Received 0 Likes on 0 Posts
Originally Posted by MarkerInbound View Post
And from the NTSB the debris field is about 200x100 yards.
Actually, what Chairman Sumwalt said is that the main wreckage is approximately 200 yards long by about 100 yards wide. Perhaps he implies that the packages and aircraft parts are scattered over a wider area.

See his comments at 4:46 in the NTSB media conference video:


Here's cellphone video of the debris field on the Chambers County Sheriff's Office Facebook page:

https://www.facebook.com/watch/?v=2305590612836823

Last edited by Airbubba; 25th Feb 2019 at 20:25.
Airbubba is offline  
Old 25th Feb 2019, 20:17
  #151 (permalink)  
 
Join Date: Feb 2019
Location: New jersey
Posts: 130
Likes: 0
Received 0 Likes on 0 Posts
I’ve flown into IAH many times, but as a Night Cargo Pilot, it’s much less busy and relatively easy. My Son, however, absolutely hates IAH. He flies for a passenger airline and he says the controllers constantly change the arrivals and runways on them during the arrival phase of flight. He’s come close to being violated there a few times and tries to avoid IAH if possible. Perhaps this is why some posters mentioned the crew at an incorrect altitude. I know my airline has just issued a bulletin about changing STARS and Runways while on an arrival.
Chiefttp is offline  
Old 25th Feb 2019, 20:20
  #152 (permalink)  
 
Join Date: Aug 2013
Location: South Alabama
Age: 72
Posts: 337
Likes: 0
Received 0 Likes on 0 Posts
I know nothing about the 767 stabilizer and elevator system, and I realize this is a completely different type of airplane, but does anyone remember what happened with Alaska Airlines Flight 261 and the jack screw assembly?

https://en.wikipedia.org/wiki/Alaska...nes_Flight_261
Old Boeing Driver is offline  
Old 25th Feb 2019, 20:29
  #153 (permalink)  
 
Join Date: Mar 2007
Location: under the sea
Posts: 2,742
Likes: 0
Received 1 Like on 1 Post
Airbubba the link to the facebook page led to a rather meaningless video.
tubby linton is offline  
Old 25th Feb 2019, 20:51
  #154 (permalink)  
fdr
 
Join Date: Jun 2001
Location: 3rd Rock, #29B
Posts: 1,960
Received 20 Likes on 12 Posts
Originally Posted by derjodel View Post
So Fly Dubai, Lion Air and now Atlas all share:
- Boeing airplane
- Nosedive

Lion Air is probably an outlier, but then again, what if all 3 have trim management in common? Because at least Lion and FlyDubai do.

In Addition Fly Dubai and Atlas both share low visibility and approach/final, right? And somebody mentioned constant speed. It seems Atlas air was actually slowing down and FlyDubai was just slightly accelerating, both despite nose down, if FR data is to be trusted (can't post links, sorry).

Seems like some kind of a pattern, doesn't it?

your examples also share other characteristics:

1. built after Dec 17, 1903
2. had paint on the outside
3. had people and stuff on the inside
4. took off
5. had wings
6. had engines

yep, theres a definite pattern there.

FZ981 was a disorientation event following a GA in wind shear.
Lion Air 610 has indications of sensor failure leading to loss of control for unknown reasons at this time.

FZ approach was not low viz, with an GA due wind shear according to the interim report. Fedex was at 6000', starting the approach, and the weather at the field itself was clear, the line of weather on the approach that the aircraft entered was well away from the airport.

fdr is online now  
Old 25th Feb 2019, 20:54
  #155 (permalink)  
 
Join Date: Jul 2007
Location: Virginia
Age: 56
Posts: 3
Likes: 0
Received 0 Likes on 0 Posts
I wasn't "trying" to hear it....

Originally Posted by B2N2 View Post


1. I don’t hear anything on that tape
2. Any perceived noises or sounds out of context don’t mean anything.
Even if “pull” was said and picked up by any of the microphones it could be dozens of things.
A comment about seat adjustment, seat belts, folding table tops, crew bags, door handles anything.

”Hey could you grab me the sunshade?”
”Go ahead just pull”

Not trying to be facetious but if you’re trying to hear “Pull” then consider the Youtube videos of cats sounding like they say hello.

I did hear it. You are right about the possible context......but given the timing of the crash relative to that word heard, I'd be comfortable making a $20 bet, it wasn't anything to do with a seat belt.
Agent1966 is offline  
Old 25th Feb 2019, 20:55
  #156 (permalink)  
fdr
 
Join Date: Jun 2001
Location: 3rd Rock, #29B
Posts: 1,960
Received 20 Likes on 12 Posts
Originally Posted by Old Boeing Driver View Post
I know nothing about the 767 stabilizer and elevator system, and I realize this is a completely different type of airplane, but does anyone remember what happened with Alaska Airlines Flight 261 and the jack screw assembly?

https://en.wikipedia.org/wiki/Alaska...nes_Flight_261
AZ261 MD80 aircraft had a single screw jack, the B767 has dual jacks which are independent, except that they share structure at the fuselage attachment, which was the subject of AD's in the past related to cracking.
fdr is online now  
Old 25th Feb 2019, 20:57
  #157 (permalink)  
 
Join Date: Aug 2013
Location: South Alabama
Age: 72
Posts: 337
Likes: 0
Received 0 Likes on 0 Posts
@ FDR Thanks for your response.

Here is the "proposed AD

" ....in 2000"On January 31, 2000, there was an accident involving a McDonnell Douglas Model DC-9-83 (MD-83) airplane. The National Transportation Safety Board (NTSB) determined that the probable cause of this accident was a loss of airplane pitch control resulting from the in-flight failure of the acme nut threads of the jackscrew assembly of the horizontal stabilizer trim system.
The NTSB concluded that the thread failure was caused by excessive wear, resulting from insufficient lubrication of the jackscrew assembly.Start Printed Page 58621The drive mechanism of the horizontal stabilizer on McDonnell Douglas Model DC-9-83 (MD-83) airplanes has a jackscrew assembly with an acme screw.
The drive mechanism of the horizontal stabilizer on Boeing Model 767 airplanes uses a ballscrew. Acme screws and ballscrews have some differences in design, but perform similar functions and have the same airplane level effect following failure. The manufacturer's safety analysis of the 767 drive mechanism found no safety problems with the configuration of the drive mechanism, but showed that changes to the maintenance procedures and maintenance intervals are required to keep the drive mechanism properly maintained and operating as designed.
We have received a report indicating that the ballscrew in the drive mechanism of the horizontal stabilizer on a Boeing Model 757 series airplane showed extensive corrosion, which could lead to excessive wear. The ballscrew on Boeing Model 757 airplanes is similar to that on Boeing Model 767 airplanes that are the subject of this proposed AD. Therefore, both of these airplane models could have the same unsafe condition.
We are considering separate action for the Boeing Model 757 series airplanes and other similar Boeing airplanes.Extensive corrosion of the ballscrew in the drive mechanism of the horizontal stabilizer, if not corrected, could cause an undetected failure of the primary load path for the ballscrew and subsequent wear and failure of the secondary load path, which could lead to loss of control of the horizontal stabilizer and consequent loss of control of the airplane."

Last edited by Old Boeing Driver; 25th Feb 2019 at 21:17.
Old Boeing Driver is offline  
Old 25th Feb 2019, 22:34
  #158 (permalink)  
 
Join Date: Mar 2018
Location: Florida
Posts: 12
Likes: 0
Received 0 Likes on 0 Posts
Originally Posted by fdr View Post
AZ261 MD80 aircraft had a single screw jack, the B767 has dual jacks which are independent, except that they share structure at the fuselage attachment, which was the subject of AD's in the past related to cracking.
Are you positive about the 767 having dual jacks? I'm pretty sure there's only one screwjack ( although there might be concentric load path redundancy in the event of a failure) with dual hydraulic motors/brakes. It's been a few years since I've flown it, though...
Zlinguy is offline  
Old 26th Feb 2019, 00:19
  #159 (permalink)  
 
Join Date: Feb 2004
Location: Australia
Posts: 1,302
Likes: 0
Received 0 Likes on 0 Posts
Originally Posted by Zlinguy View Post
Are you positive about the 767 having dual jacks? I'm pretty sure there's only one screwjack ( although there might be concentric load path redundancy in the event of a failure) with dual hydraulic motors/brakes. It's been a few years since I've flown it, though...


Correct. The 767 only has one ballscrew. However, it does use two hydraulic motors which drive the single ballscrew through a differential gearbox.
NSEU is offline  
Old 26th Feb 2019, 02:26
  #160 (permalink)  
 
Join Date: Dec 2001
Location: FL, USA
Posts: 3,045
Likes: 0
Received 1 Like on 1 Post
Even if you personally don’t feel the need to contribute please forward the link.
It has been verified as legitimate.

https://www.gofundme.com/f/atlas-air...ms-family-fund

Thank you
B2N2 is offline  

Thread Tools
Search this Thread

Contact Us - Archive - Advertising - Cookie Policy - Privacy Statement - Terms of Service - Do Not Sell or Share My Personal Information

Copyright © 2023 MH Sub I, LLC dba Internet Brands. All rights reserved. Use of this site indicates your consent to the Terms of Use.