UPS 1354 NTSB Investigation - CVR
Join Date: Jun 2001
Location: Rockytop, Tennessee, USA
Posts: 5,898
Likes: 0
Received 1 Like
on
1 Post
Well, there are former Blue Angels flying cargo, and you can talk to quite a few fine aviators who did not get a thumbs up when they interviewed, so I do not agree with your United law suit version of this view of training failures. It might very well be that, although, unusual, he saw a spanking new 757 training program and thought it not up to standards for his ability to jump from 727 right seat to 757 left seat.
And yes, like United a few years before, Coke was under an EEOC monitored lawsuit settlement to 'promote diversity in hiring':
COCA-COLA SETTLES RACIAL BIAS CASE - NYTimes.com
I do feel that EEOC lawsuit settlements like those at UAL and Coke have dramatically influenced hiring standards, for some folks at least, in years past. Maybe I'm wrong or being judgmental but that's my view. Obviously, many other factors affect hiring standards and the forever 'coming pilot shortage' may further drop requirements to absolute mins.
The poor training history, both before and after being hired, of RS in one of the FedEx MEM Mad Dog crashes was met with calls from the NTSB for better remedial training. However, after the BUF Colgan crash, hiring standards seem to be under NTSB scrutiny as well.
In fairness to the BHM 1354 crew, due to the strong corporate commitment to recruit underrepresented demographics, they were quite possibly hired with significantly less experience than many of their classmates. This might account for some of the captain's difficulties in training years ago. He reportedly had little multiengine fixed wing time when hired at UPS, and what multi time he had was in a Shorts SD-30.
Also, the recent assessments of both crewmembers presented in the docket interviews were glowing. They were both warm, professional and wonderful people by these accounts.
Join Date: Jun 2001
Location: Surrounded by aluminum, and the great outdoors
Posts: 3,780
Likes: 0
Received 0 Likes
on
0 Posts
Yes the second profile appears to be more accurate...I was having problems with the first one, because it shows them above profile above IMTOY, which would have not been possible given their rate of descent from 2500'....
@jstflyin
When the Captain said "and we're like way high" (4;46:53.7) the aircraft was actually above (~135ft) the nominal flight path, though characterized by the Captain as "about ... a couple hundred feet" (4:46:57.1).
When the FO called "thousand feet instruments cross checked no flags" (4:47:02.9) the aircraft was about 60 ft above the nominal flight path.
However, as we all know, at 1,000 ft above touchdown the VS was about 1,500 FPM, when the UPS maximum VS below 1,000 ft for a Stabilized Approach is 1,000 FPM.
The high VS persisted long through MDA 1,200 ft MSL (or DA if you will) (with no Minimums call), and through the required, but not called 500 ft above touchdown, until after the Sink Rate alert at around 200+ ft RA and 1000 ft MSL.
Too bad there were no automatic callouts, though I am not absolutely convinced the crew would have recognized the situation, even then.
Also too bad, in my opinion, that VS mode past FAF, though not preferred by UPS, is even permitted. The AP remained engaged in this basic mode until practically the last few seconds of flight, much below MDA.
When the Captain said "and we're like way high" (4;46:53.7) the aircraft was actually above (~135ft) the nominal flight path, though characterized by the Captain as "about ... a couple hundred feet" (4:46:57.1).
When the FO called "thousand feet instruments cross checked no flags" (4:47:02.9) the aircraft was about 60 ft above the nominal flight path.
However, as we all know, at 1,000 ft above touchdown the VS was about 1,500 FPM, when the UPS maximum VS below 1,000 ft for a Stabilized Approach is 1,000 FPM.
The high VS persisted long through MDA 1,200 ft MSL (or DA if you will) (with no Minimums call), and through the required, but not called 500 ft above touchdown, until after the Sink Rate alert at around 200+ ft RA and 1000 ft MSL.
Too bad there were no automatic callouts, though I am not absolutely convinced the crew would have recognized the situation, even then.
Also too bad, in my opinion, that VS mode past FAF, though not preferred by UPS, is even permitted. The AP remained engaged in this basic mode until practically the last few seconds of flight, much below MDA.
Last edited by GlobalNav; 28th Feb 2014 at 19:51.
Guest
Join Date: Apr 2009
Location: On the Beach
Posts: 3,336
Likes: 0
Received 0 Likes
on
0 Posts
GlobalNav:
As well as a couple hundred feet below the PAPI slope at point of impact. Red lights only.
The AP remained engaged in this basic mode until practically the last few seconds of flight, much below MDA.
Join Date: Feb 2014
Location: USA
Posts: 11
Likes: 0
Received 0 Likes
on
0 Posts
"As well as a couple hundred feet below the PAPI slope at point of impact. Red lights only."
Take a look at Document 42 ATC 3 - Attachment 1 - BAA - ATC Airport Emergency Operating Procedures Letter of Agreement Filing Date December 16, 2013 4 page(s) of Image (PDF or TIFF) 0 Photos
Open it up and look at Figure 9b on page 47.
They popped out of the cloud base about the same time they were hearing "Sink Rate!" Less than 4 secs later the captain said he had the runway in sight but from about then onwards, the 4 reds of the PAPI were obscured by terrain. Precious little time to make sense of it.
Take a look at Document 42 ATC 3 - Attachment 1 - BAA - ATC Airport Emergency Operating Procedures Letter of Agreement Filing Date December 16, 2013 4 page(s) of Image (PDF or TIFF) 0 Photos
Open it up and look at Figure 9b on page 47.
They popped out of the cloud base about the same time they were hearing "Sink Rate!" Less than 4 secs later the captain said he had the runway in sight but from about then onwards, the 4 reds of the PAPI were obscured by terrain. Precious little time to make sense of it.
Guest
Join Date: Apr 2009
Location: On the Beach
Posts: 3,336
Likes: 0
Received 0 Likes
on
0 Posts
They may have never seen the PAPIs because of weather obscuring their ability to see them. But, they had a sight line to the PAPIs until very near the end when the small hill and perhaps trees would have blocked the sight line.
Nonetheless sighting of the PAPIs was required before they departed MDA, which this crew apparently didn't understand.
Nonetheless sighting of the PAPIs was required before they departed MDA, which this crew apparently didn't understand.
Join Date: Feb 2014
Location: USA
Posts: 11
Likes: 0
Received 0 Likes
on
0 Posts
They may have never seen the PAPIs because of weather obscuring their ability to see them. But, they had a sight line to the PAPIs until very near the end when the small hill and perhaps trees would have blocked the sight line.
Reference....
http://dms.ntsb.gov/public%2F55000%2...2F550789%2Epdf
....."Aircraft Performance Study" pages 24 and 25.....
It is of interest to know whether the PAPI lights would have been visible to the crew of flight 1354 as it approached runway 18, and if so, what light pattern would have been perceived. This problem amounts to knowing the position of the airplane relative to each of the four PAPI light beams, and identifying whether any obstructions (such as terrain) penetrated the line of sight between the PAPI and the airplane. The angle of these beams relative to the runway are defined by the (actual) PAPI glide path angle, and the (actual) aiming angles of each of the beams relative to the glide path angle.
PAPI beam solutions for both the design value of the PAPI glide path angle (3.20°), and the nominal aiming angles for a height group 3 PAPI are depicted in Figure 9 (per Table 7, the actual aiming angles are all within 0.5 minutes of arc of the nominal values, so the beam
paths depicted in Figure 9 are representative). The altitudes of the beams depicted in Figure 9 take into account the curvature of the Earth, which will increase the height of the beams as the distance from the PAPI increases, compared to the height obtained assuming a flat Earth.
Figure 9 indicates that prior to IMTOY, flight 1354 was above the PAPI glide path, and crossed below it about 0.1 nmi north of IMTOY, though at IMTOY, the PAPI display to the crew would still have been 2 white and 2 red lights (indicating on the glide path). The display would have changed to 3 red lights and 1 white light about 0.05 nmi south of IMTOY, indicating that the airplane was deviating below the glide path. The PAPI display would have changed to 4 red lights (indicating a position well below the flight path) about 0.3 nmi south of IMTOY, or about 1.7 nmi north of the runway threshold. Of course, the video analysis described in Section D-VI indicates that at this point the runway (and PAPI) would still have been obscured by clouds, and would remain so at least until the airplane descended below 1000 ft. MSL, about 1.4 nmi north of the threshold.
When the crew reported the runway in sight at about 900 ft. MSL and about 1.24 nmi north of the threshold, Figures 9 and 24 suggest that the PAPI would have been visible for less than a second, becoming obscured by terrain almost immediately.
PAPI beam solutions for both the design value of the PAPI glide path angle (3.20°), and the nominal aiming angles for a height group 3 PAPI are depicted in Figure 9 (per Table 7, the actual aiming angles are all within 0.5 minutes of arc of the nominal values, so the beam
paths depicted in Figure 9 are representative). The altitudes of the beams depicted in Figure 9 take into account the curvature of the Earth, which will increase the height of the beams as the distance from the PAPI increases, compared to the height obtained assuming a flat Earth.
Figure 9 indicates that prior to IMTOY, flight 1354 was above the PAPI glide path, and crossed below it about 0.1 nmi north of IMTOY, though at IMTOY, the PAPI display to the crew would still have been 2 white and 2 red lights (indicating on the glide path). The display would have changed to 3 red lights and 1 white light about 0.05 nmi south of IMTOY, indicating that the airplane was deviating below the glide path. The PAPI display would have changed to 4 red lights (indicating a position well below the flight path) about 0.3 nmi south of IMTOY, or about 1.7 nmi north of the runway threshold. Of course, the video analysis described in Section D-VI indicates that at this point the runway (and PAPI) would still have been obscured by clouds, and would remain so at least until the airplane descended below 1000 ft. MSL, about 1.4 nmi north of the threshold.
When the crew reported the runway in sight at about 900 ft. MSL and about 1.24 nmi north of the threshold, Figures 9 and 24 suggest that the PAPI would have been visible for less than a second, becoming obscured by terrain almost immediately.
As noted in Section D-VII, non-precision approach procedures specify how the airplane is to descend on the final approach course to the Minimum Descent Altitude (MDA), which in this case is 1200 ft. MSL. To descend from the MDA, the crew must have the airport / runway environment in sight, and complete the descent and landing visually. However, the security video analysis described in Section D-VI indicates that the airplane was not below the clouds (and the runway would not have been visible from the cockpit) until about 1000 ft. MSL, 200 ft. below the MDA. This finding is consistent with the crew’s reporting the airport in sight at about 900 ft. MSL.
Furthermore, the PAPI visibility analysis described in Section D-VII indicates that when the crew reported the runway in sight at about 900 ft. MSL and about 1.24 nmi north of the threshold, the PAPI would have been visible for less than a second, becoming obscured by terrain almost immediately. Consequently, it is likely that the crew never saw or recognized the PAPI lights, which could have alerted them to the airplane’s dangerously low altitude.
Furthermore, the PAPI visibility analysis described in Section D-VII indicates that when the crew reported the runway in sight at about 900 ft. MSL and about 1.24 nmi north of the threshold, the PAPI would have been visible for less than a second, becoming obscured by terrain almost immediately. Consequently, it is likely that the crew never saw or recognized the PAPI lights, which could have alerted them to the airplane’s dangerously low altitude.
Nonetheless sighting of the PAPIs was required before they departed MDA, which this crew apparently didn't understand.
Guest
Join Date: Apr 2009
Location: On the Beach
Posts: 3,336
Likes: 0
Received 0 Likes
on
0 Posts
showbo:
I don't believe there is sufficient information to conclude that the crew didn't understand the PAPI requirement. But, their actions make it reasonable to conclude that they disregarded this and the other visual cues requirements of FAR 91.175. Similar reasoning applies to the other omissions you cite.
Correct...... but I seriously doubt the NTSB will ever conclude that the crew didn't understand this. Nor that they didn't know they were supposed to sequence the FMC and "extend" the approach; nor that they didn't know that the autopilot was supposed to be disconnected by 50 ft below the MDA; nor that they didn't understand that 1500 fpm at 1000 ft afe is an unstable approach requiring a go around; nor that they thought the "minimums" and 500 ft callouts were optional. The question will be, what human factors caused them to make all these omissions?
No particular reason to think the crew didn't know the visual requirements, per se, but pretty clear they were unaware of their vertical situation. I say this based on no calls for 500 ft, approaching minimums and minimums.
Most of the approaches they flew were to 200 ft DA at 700 FPM, this one to 560 HAA at 1500 FPM, so things were happening faster than they are used to. The VS mode is no help with awareness. No automatic call outs, yet AP coupled and the crew missed their required calls. They apparently didn't even look for visual cues until the sink rate alert. Behind the airplane.
Most of the approaches they flew were to 200 ft DA at 700 FPM, this one to 560 HAA at 1500 FPM, so things were happening faster than they are used to. The VS mode is no help with awareness. No automatic call outs, yet AP coupled and the crew missed their required calls. They apparently didn't even look for visual cues until the sink rate alert. Behind the airplane.
Last edited by GlobalNav; 2nd Mar 2014 at 14:20.
Join Date: Feb 2014
Location: USA
Posts: 11
Likes: 0
Received 0 Likes
on
0 Posts
No particular reason to think the crew didn't know the visual requirements, per SE, but pretty clear they were unaware of their vertical situation. I say this based on no calls for 500 ft, approaching minimums and minimums.
Most of the approaches they flew were to 200 ft DA at 700 FPM, this one to 560 HAA at 1500 FPM, so things were happening faster than they are used to. The VS mode is no help with awareness. No automatic call outs, yet AP coupled and the crew missed their required calls. They apparently didn't even look for visual cues until the sink rate alert. Behind the airplane.
Guest
Join Date: Apr 2009
Location: On the Beach
Posts: 3,336
Likes: 0
Received 0 Likes
on
0 Posts
GlobalNav:
Plus, they descended below MDA in clear violation of regs.
All true. Makes them (late) children of the magenta line.
No particular reason to think the crew didn't know the visual requirements, per SE, but pretty clear they were unaware of their vertical situation. I say this based on no calls for 500 ft, approaching minimums and minimums.
Most of the approaches they flew were to 200 ft DA at 700 FPM, this one to 560 HAA at 1500 FPM, so things were happening faster than they are used to. The VS mode is no help with awareness. No automatic call outs, yet AP coupled and the crew missed their required calls. They apparently didn't even look for visual cues until the sink rate alert. Behind the airplane.
@aterpster. Agreed. They descended through minimums unknowingly, I surmise, and hence did not comply with the requirements. I would think they were knowledgable of the requirements though.
I find it hard not to believe that this crew is not very unique from their fleet mates and perhaps many experienced ATP pilots of the day. Lots of flight time, same training program etc. So even if this event is considered an outlier, the vulnerability to such events is not unique to a few pilots.
What allows this chain of events to happen and how can we reduce this vulnerability? I will offer up some thoughts, not because I think they are waterproof (might have said fool proof).
1. Non precision approaches, in spite of CDFA, are flown so infrequently that the level of proficiency and safety just is not equivalent to precision approaches. Therefore, we should equip aircraft with avionics that allow approaches and pilot procedures to instrument runways be as similar to ILS approaches as possible.
2. Provide Automatic call outs for key vertical milestones as a safety net to momentary lack of crew awareness. At least a call out for approaching minimums and minimums, I suggest.
3. Operators should reinforce and mandate the use of Stabilized. Approach criteria, and take away any ambiguity about pilot action in such cases. An unstable approach is a failed approach and should be aborted. It would be preferable that there be an automatic call out for 1,000 ft above touchdown which is point at which a stabilized approach must be verified.
4. It seems to me that the crew response to the Sink Rate alert was less than aggressive. Below 1,000 ft above touchdown every EGPWS/TAWS alert should be considered evidence of an unstable approach and the approach aborted. For Too Low Terrain alert, if it occurs, an aggressive CFIT avoidance maneuver.
Perhaps my suggestions are considered extreme. I think when unanticipated alerts like these occur there is a moment of disbelief and a too lengthy period of assessment and verification of the condition which can delay and diminish the commitment to a positive intervention. So take the wishy washy ambiguity out of the required response.
I find it hard not to believe that this crew is not very unique from their fleet mates and perhaps many experienced ATP pilots of the day. Lots of flight time, same training program etc. So even if this event is considered an outlier, the vulnerability to such events is not unique to a few pilots.
What allows this chain of events to happen and how can we reduce this vulnerability? I will offer up some thoughts, not because I think they are waterproof (might have said fool proof).
1. Non precision approaches, in spite of CDFA, are flown so infrequently that the level of proficiency and safety just is not equivalent to precision approaches. Therefore, we should equip aircraft with avionics that allow approaches and pilot procedures to instrument runways be as similar to ILS approaches as possible.
2. Provide Automatic call outs for key vertical milestones as a safety net to momentary lack of crew awareness. At least a call out for approaching minimums and minimums, I suggest.
3. Operators should reinforce and mandate the use of Stabilized. Approach criteria, and take away any ambiguity about pilot action in such cases. An unstable approach is a failed approach and should be aborted. It would be preferable that there be an automatic call out for 1,000 ft above touchdown which is point at which a stabilized approach must be verified.
4. It seems to me that the crew response to the Sink Rate alert was less than aggressive. Below 1,000 ft above touchdown every EGPWS/TAWS alert should be considered evidence of an unstable approach and the approach aborted. For Too Low Terrain alert, if it occurs, an aggressive CFIT avoidance maneuver.
Perhaps my suggestions are considered extreme. I think when unanticipated alerts like these occur there is a moment of disbelief and a too lengthy period of assessment and verification of the condition which can delay and diminish the commitment to a positive intervention. So take the wishy washy ambiguity out of the required response.
Join Date: Jun 2009
Location: NNW of Antipodes
Age: 81
Posts: 1,330
Received 0 Likes
on
0 Posts
@GlobalNav,
I have the feeling after looking carefully at the PF's control column graphic, that for about 10 secs prior to the "Sink Rate" warning the trace is looking "relaxed". The reaction time was close to 3 secs and initially in the wrong direction. Could be a "micro-nap" got in the way??
I have the feeling after looking carefully at the PF's control column graphic, that for about 10 secs prior to the "Sink Rate" warning the trace is looking "relaxed". The reaction time was close to 3 secs and initially in the wrong direction. Could be a "micro-nap" got in the way??
Join Date: Feb 2008
Location: Europe
Posts: 24
Likes: 0
Received 0 Likes
on
0 Posts
Quote from GlobalNav:
Perhaps my suggestions are considered extreme. I think when unanticipated alerts like these occur there is a moment of disbelief and a too lengthy period of assessment and verification of the condition which can delay and diminish the commitment to a positive intervention. So take the wishy washy ambiguity out of the required response.
Pretty much SOP in our company, so no, not so extreme as you might think.
Perhaps my suggestions are considered extreme. I think when unanticipated alerts like these occur there is a moment of disbelief and a too lengthy period of assessment and verification of the condition which can delay and diminish the commitment to a positive intervention. So take the wishy washy ambiguity out of the required response.
Pretty much SOP in our company, so no, not so extreme as you might think.
@mm43. Not sure what you are reading into the trace, but they were still AP coupled so I'd think the yoke forces would be pretty relaxed.
I wonder at the pilot's response to the sink rate alert. Twist MCP knob to reduce VS. Way low, yet still on AP and not a very positive/aggressive reaction. Can't say for sure, but I suppose not aware how low they were, and thought a mere VS correction was sufficient.
I wonder at the pilot's response to the sink rate alert. Twist MCP knob to reduce VS. Way low, yet still on AP and not a very positive/aggressive reaction. Can't say for sure, but I suppose not aware how low they were, and thought a mere VS correction was sufficient.
Join Date: Feb 2014
Location: USA
Posts: 11
Likes: 0
Received 0 Likes
on
0 Posts
mm43
But the autopilot was still on until 7 seconds AFTER the Sink Rate warning. That's not to say that micro sleep wasn't a distinct possibility because there was a lot of stuff not happening that should have been.
that for about 10 secs prior to the "Sink Rate" warning the trace is looking "relaxed".
Guest
Join Date: Apr 2009
Location: On the Beach
Posts: 3,336
Likes: 0
Received 0 Likes
on
0 Posts
Global Nav:
They probably were already so equipped had the elected the RNAV IAP to Runway 18 with a Baro VNAV advisory path.
I know the small airplane SBAS navigators provide an advisory SBAS generated advisory vertical path for the RNAV Runway 18. That, and PAPI, makes it a safe approach.
In any case runways like this one should have a tailored air carrier briefing page for Part 121 operators.
The dispatcher didn't do a sterling job either.
1. Non precision approaches, in spite of CDFA, are flown so infrequently that the level of proficiency and safety just is not equivalent to precision approaches. Therefore, we should equip aircraft with avionics that allow approaches and pilot procedures to instrument runways be as similar to ILS approaches as possible.
I know the small airplane SBAS navigators provide an advisory SBAS generated advisory vertical path for the RNAV Runway 18. That, and PAPI, makes it a safe approach.
In any case runways like this one should have a tailored air carrier briefing page for Part 121 operators.
The dispatcher didn't do a sterling job either.