Go Back  PPRuNe Forums > Flight Deck Forums > Accidents and Close Calls
Reload this Page >

Another Southwest close call

Accidents and Close Calls Discussion on accidents, close calls, and other unplanned aviation events, so we can learn from them, and be better pilots ourselves.

Another Southwest close call

Old 18th Jun 2024, 11:58
  #61 (permalink)  
 
Join Date: Mar 2002
Location: Seat 1A
Posts: 8,651
Received 123 Likes on 69 Posts
Originally Posted by EXDAC
I did not see any statement that suggests there was 16 seconds of data missing
31sec to 47sec = 16 seconds
The next line down:
Originally Posted by AVHerald
What happened in these 16 seconds is described in an internal memo circulating in Southwest Airlines stating...
Clearly, you don't buy my hypothesis. What do you think happened?
Capn Bloggs is online now  
Old 18th Jun 2024, 12:26
  #62 (permalink)  
 
Join Date: Apr 2009
Location: Phoenix, AZ
Posts: 766
Received 63 Likes on 43 Posts
Originally Posted by Capn Bloggs
31sec to 47sec = 16 seconds
That is not in dispute but it does not say there was no data between those times. It only provides the aircraft state at those two times. I do not know what data set they were looking at or whether there was data missing in that dataset. ADS-B Exchange data is all sourced from amateur enthusiasts and does not get included in other ADS-B data repositories. It is very likely that different ADS-B tracking sites will have different data points.

Originally Posted by Capn Bloggs
Clearly, you don't buy my hypothesis. What do you think happened?
I didn't spend much time working on 737 avionics and it was some time ago. I don't know the potential for mode confusion. The profile suggests to me that the FO was flying an FD that was still in approach mode but I'm not saying that is what happened.
EXDAC is offline  
Old 18th Jun 2024, 12:32
  #63 (permalink)  
 
Join Date: Dec 1998
Location: UK
Posts: 303
Likes: 0
Received 0 Likes on 0 Posts
https://www.gov.uk/aaib-reports/aaib...737-8k5-g-fdzf

not the first time this has happened.
Propellerhead is online now  
Old 18th Jun 2024, 13:00
  #64 (permalink)  
 
Join Date: Mar 2002
Location: Seat 1A
Posts: 8,651
Received 123 Likes on 69 Posts
Originally Posted by Exdac
ADS-B Exchange data is all sourced from amateur enthusiasts and does not get included in other ADS-B data repositories. It is very likely that different ADS-B tracking sites will have different data points.
You're looking at something we can't see. That graph is not useable for analysis of the 16 seconds. How about posting the data? All that is relevant is height and speed for each time point.

ADS-B exchange captures data from the ADS-B output, just like all the other "repositories". I doubt that it is anything special or more accurate. There might be a user, close-by, who can capture the data when no FR24 user exists; that would be the only reason you have the data for that 16 seconds whereas FR24 doesn't.

Originally Posted by Exdac
The profile suggests to me that the FO was flying an FD that was still in approach mode but I'm not saying that is what happened.
That is not possible. If you are implying they just continued on down the approach, the achieved/reported numbers do not make sense. They achieved 4400ft/min descent rate and still didn't go below 400ft. That just could not happen from the MDA. There is no way in the world that, from ~1000ft, you could push the nose down, achieve -4400fpm, then recover by 400ft in this size aeroplane.
Capn Bloggs is online now  
Old 18th Jun 2024, 14:00
  #65 (permalink)  
 
Join Date: Apr 2009
Location: Phoenix, AZ
Posts: 766
Received 63 Likes on 43 Posts
Originally Posted by Capn Bloggs
You're looking at something we can't see. That graph is not useable for analysis of the 16 seconds. How about posting the data? All that is relevant is height and speed for each time point.
My level of membership does not allow a dump of the raw data. What I did was step along the flight path manually recording the data for each available point. I have not yet attempted to plot any of the data and maybe someone will beat me to it. I cannot see how to post a spreadsheet file but pm me if you'd like it.

time mm:ss // baro alt ft // gps alt ft // alt rate ft/min // gps gs kt

12:21 1100 1250 -1024 151

12:27 1000 1150 -896 150

12:35 900 1025 -768 149

12:39 875 1000 -320 151

12:40 875 1000 -128 154

12:41 875 1000 256 158

12:42 900 1025 256 161

12:43 900 1025 256 165

12:44 900 1025 128 169

12:45 900 1025 0 172

12:46 900 1025 -256 176

12:47 875 975 -1216 186

12:49 800 975 -2752 191

12:51 700 825 -3712 200

12:52 600 725 -4160 204

12:58 400 500 -64 231

13:00 475 550 2752 235

13:01 525 575 3904 236

13:02 625 675 4928 235

13:03 700 775 6080 231

13:04 875 925 7424 221

13:06 1100 1100 8192 216

EXDAC is offline  
Old 18th Jun 2024, 14:09
  #66 (permalink)  
 
Join Date: May 2024
Location: Kaupuala
Posts: 131
Received 9 Likes on 9 Posts
Pesky

There is that pesky physics thing. Instrument problems either recorded or not, and (gasp) another uncommanded descent..."inadvertent" and robust push, followed by thrust levers full forward. There could be many billions of dollars riding on "inadvertent"

To the linguist, vocabulary is not matching up to reality. Too many Boeings giving up controlled flight to gravity
BugBear is offline  
Old 18th Jun 2024, 14:17
  #67 (permalink)  
 
Join Date: Apr 2009
Location: Phoenix, AZ
Posts: 766
Received 63 Likes on 43 Posts
Here is a plot of the raw ADS-B data:


EXDAC is offline  
Old 18th Jun 2024, 14:27
  #68 (permalink)  
 
Join Date: May 2024
Location: Kaupuala
Posts: 131
Received 9 Likes on 9 Posts
Yikes. So the a/c climbed 700 feet in nine seconds. From level at 400RA to a Climb rate of 8000/minute. Must have been some GLOC in the cabin
BugBear is offline  
Old 18th Jun 2024, 14:41
  #69 (permalink)  
 
Join Date: Mar 2002
Location: Seat 1A
Posts: 8,651
Received 123 Likes on 69 Posts
OK, that's great, and supports my supposition that they didn't "bust" the MDA while descending. The FO actually levelled off (probably thinking he was climbing because of the thrust trying to pitch up, and the acceleration illusion). Looks like he started the GA at around 12:41, with GA thrust and speed starting to increase. As the speed increased, so would the sensation of climbing (steeply) and the pitch force so he has started pushing the nose down... from level flight. To go from level at 900ft to 600ft and -4000fpm over 12 seconds would be doable. Difficult, but doable.

Southwest has some work to do, based on those numbers.
Capn Bloggs is online now  
Old 18th Jun 2024, 14:57
  #70 (permalink)  
 
Join Date: Apr 2009
Location: Phoenix, AZ
Posts: 766
Received 63 Likes on 43 Posts
Plot of GPS GS and Baro alt:


EXDAC is offline  
Old 18th Jun 2024, 15:08
  #71 (permalink)  
 
Join Date: May 2024
Location: Kaupuala
Posts: 131
Received 9 Likes on 9 Posts
Ca

CapnBloggs

"As the speed increased, so would the sensation of climbing (steeply) and the pitch force so he has started pushing the nose down... from level flight"

That's alarming to say the least...nothing to see outside, IMC... He should not be flying... Sensation.
Something wrong with FD? VS? Alt?

At this point, Climbing is part of the plot, no?
This meets criteria for JetLOC, YES?
Was DFDR read? What was trim set? What were elevators deflection to recover? G loads? Time enough for NTSB initial...anyone?

Eleven days to report the dutch roll accident, now this? Calhoun got some splainin' to do...

Didn't Reuters use to read PPRune??

Bugged Bear


Last edited by BugBear; 18th Jun 2024 at 15:26.
BugBear is offline  
Old 18th Jun 2024, 15:09
  #72 (permalink)  
 
Join Date: Mar 2002
Location: Seat 1A
Posts: 8,651
Received 123 Likes on 69 Posts
I've seen enough. I'll have nightmares tonight!
Capn Bloggs is online now  
The following users liked this post:
Old 18th Jun 2024, 15:17
  #73 (permalink)  
 
Join Date: May 2011
Location: Metropolis
Posts: 1,281
Likes: 0
Received 0 Likes on 0 Posts
That is extremely alarming. Not everyone is cut out to fly.

Also disturbing is why the captain didn’t say, “I have the aircraft”, and take control.
Rozy1 is offline  
Old 18th Jun 2024, 15:28
  #74 (permalink)  
 
Join Date: May 2024
Location: Kaupuala
Posts: 131
Received 9 Likes on 9 Posts
Originally Posted by Rozy1
That is extremely alarming. Not everyone is cut out to fly.

Also disturbing is why the captain didnít say, ďI have the aircraftĒ, and take control.
Maybe he did, and both had to arrest the descent!!

​​​​it is beyond reason to assume Captain "just let this happen'" Beyond counterintuituve...Comes now whistleblower First Officer X?
BugBear is offline  
Old 18th Jun 2024, 17:12
  #75 (permalink)  
 
Join Date: Apr 2009
Location: Phoenix, AZ
Posts: 766
Received 63 Likes on 43 Posts
I found another data point at 12:57 and improved the axis presentations:





EXDAC is offline  
Old 18th Jun 2024, 19:52
  #76 (permalink)  
 
Join Date: Dec 2006
Location: uk
Posts: 783
Received 49 Likes on 16 Posts
However you look at this it is not a good day at the office. I train on the 737, the latest variants. I would never claim to be the fount of all wisdom but the data from this incident is alarming. A 4k/ min descent followed by an 8k/min climb is not the hallmark of a well briefed, trained or structured go around. Regardless of somatographic issues, press TOGA, smoothly pitch to 15 degrees nose up, call go around flaps 15, positive climb, gear up, select a roll mode, follow the flight director accurately. If you have to engage the autopilot make sure that you are in trim. Remember with alt acquire and / or autopilot engagement the speed window opens. PM should monitor thrust application which is also dependent on auto throttle engagement or not. First push should be enough for a reduced and sensible climb rate or approx 80 / 85 pc N1 dependent on variant. I don’t know why or how that was not trained @ Southwest, an airline I have long admired. I hope very much that these issues are not becoming a trend. The 737 go around has its peculiarities not least with the pitch power relationship but it is not difficult if trained properly, briefed and rehearsed if a go around is anticipated.

Last edited by olster; 18th Jun 2024 at 20:11.
olster is offline  
Old 18th Jun 2024, 20:00
  #77 (permalink)  
 
Join Date: Sep 2022
Location: Perpetually circling LAM for some reason
Posts: 152
Received 9 Likes on 5 Posts
Originally Posted by olster
However you look at this it is not a good day at the office. I train on the 737, the latest variants. I would never claim to be the fount of all wisdom but the data from this incident is alarming. A 4k/ min descent followed by an 8k/min climb is not the hallmark of a well briefed, trained or structured go around. Regardless of somatographic issues, press TOGA, smoothly pitch to 15 degrees nose up, call go around flaps 15, positive climb, gear up, select a roll mode, follow the flight director accurately. If you have to engage the autopilot make sure that you are in trim. Remember with alt acquire and / or autopilot engagement the speed window opens. PM should monitor thrust application which is also dependent on auto throttle engagement or not. First push should be enough for a reduced and sensible climb rate or approx 80 / 85 pc N1 dependent on variant. I donít know why or how that was not trained @ Southwest, an airline I have long admired. I hope very much that these issues are not becoming a trend. The 737 go around has its peculiarities not least with the pitch power relationship but it is not difficult if trained properly briefed and rehearsed if a go around is anticipated.
All sounds very familiar even though itís been a very long time!
Speed_Trim_Fail is online now  
Old 18th Jun 2024, 20:22
  #78 (permalink)  
 
Join Date: Dec 2003
Location: Tring, UK
Posts: 1,891
Received 34 Likes on 19 Posts
Originally Posted by olster
However you look at this it is not a good day at the office.
You can say that again! -4,000fpm below 1,000RA is deep into GPWS hard warning territory - at least that appears to have been heeded. What the PM was thinking during the whole sequence IDK but from what has been unearthed so far this was LOC, as they were so far off a nominal profile in terms of height and speed. No wonder itís under investigation and Iíd have thought it would have been caught by FOQA anyway.
FullWings is offline  
The following 2 users liked this post by FullWings:
Old 18th Jun 2024, 20:48
  #79 (permalink)  

Only half a speed-brake
 
Join Date: Apr 2003
Location: Commuting not home
Age: 46
Posts: 4,324
Received 7 Likes on 6 Posts
Originally Posted by Propellerhead
https://www.gov.uk/aaib-reports/aaib...737-8k5-g-fdzf not the first time this has happened.
Thank you. In a similar fashion, I find this thread lacking any notions of the Rostov tragedy: Accident Boeing 737-8KN (WL) A6-FDN, (flightsafety.org).

Not wishing to stir the pot too much:
+ speedtape confusion (pitching down to avoid the overspeed barberpole descending from atop the PFD onto the IAS index)
+ negative g dis-orientation (untrained against, the sensation of being lifted into the harness might cause a subconscious push to "get down" or "stop the levitation".

Both work in nasty harmony if you just busted your G/A alt, compounded by the FDs pointing to pitch-down. In case you already had started to trim (high speed) nose-down to overcome the NU force on the yoke and the finger becomes a little suck on the PT rocker ... you are riding the tiger on the tail in no time at all.



FlightDetent is offline  
The following 2 users liked this post by FlightDetent:
Old 18th Jun 2024, 20:54
  #80 (permalink)  
 
Join Date: Jan 2004
Location: Here and there
Posts: 3,120
Received 19 Likes on 15 Posts
Originally Posted by olster
I donít know why or how that was not trained @ Southwest, an airline I have long admired. I hope very much that these issues are not becoming a trend. The 737 go around has its peculiarities not least with the pitch power relationship but it is not difficult if trained properly, briefed and rehearsed if a go around is anticipated.
I'm sure something similar would've been trained at Southwest. An isolated event doesn't necessarily indicate any flaws in a training system.
AerocatS2A is offline  

Thread Tools
Search this Thread

Contact Us - Archive - Advertising - Cookie Policy - Privacy Statement - Terms of Service

Copyright © 2024 MH Sub I, LLC dba Internet Brands. All rights reserved. Use of this site indicates your consent to the Terms of Use.