UPS 1354 NTSB Investigation - CVR
Guest
Join Date: Apr 2009
Location: On the Beach
Posts: 3,336
Likes: 0
Received 0 Likes
on
0 Posts
DMJ618:
Why did they bust minimums and descend at 1,500 FPM?
Seems to me they had no concept of the hazards unique to a NPA and they just wanted to get on the ground.
I have to admit, if they were at all nervous about this approach, why didn't they go around when they heard the tower and ground workers saying that the work was done and 6/24 will open soon?
Seems to me they had no concept of the hazards unique to a NPA and they just wanted to get on the ground.
Join Date: Mar 2014
Location: New Hampshire
Posts: 10
Likes: 0
Received 0 Likes
on
0 Posts
DMJ618:
Sorry, DMJ, you're not getting many takers!
Your comments are all valid, and yes, I wouldn't have been offended by a "heads up" on the LOC18.
One of the risks you list....."- low level patchy clouds (they may not even see the runway)".... appeared to be an item of particular interest in the NTSB's hearing.
The ATIS they got told them, "Visibility one zero. Sky condition, ceiling one thousand broken. Seven thousand five hundred overcast."
Sorry, DMJ, you're not getting many takers!
Your comments are all valid, and yes, I wouldn't have been offended by a "heads up" on the LOC18.
One of the risks you list....."- low level patchy clouds (they may not even see the runway)".... appeared to be an item of particular interest in the NTSB's hearing.
The ATIS they got told them, "Visibility one zero. Sky condition, ceiling one thousand broken. Seven thousand five hundred overcast."
One question, if you only had the ATIS that they did, wouldn't 1000 ft clouds be a possible concern when your decision height is 1250? Even if they had the glide path, they need to be visual at 1250, (or was it 1380?) correct?
Join Date: Mar 2014
Location: New Hampshire
Posts: 10
Likes: 0
Received 0 Likes
on
0 Posts
Why did they bust minimums and descend at 1,500 FPM?
Seems to me they had no concept of the hazards unique to a NPA and they just wanted to get on the ground.
Seems to me they had no concept of the hazards unique to a NPA and they just wanted to get on the ground.
Join Date: Feb 2000
Location: Alaska, PNG, etc.
Age: 60
Posts: 1,550
Likes: 0
Received 0 Likes
on
0 Posts
the airport is at 650 MSL so the 1000 ft cloud layer would be at 1650 MSL, above the MDA
@ A Squared. I agree.
However, you might find the Aircraft Performance Group Report's treatment of this to be interesting. The analyst viewed numerous videos from the airport, noting both position, intensity of the aircraft's landing lights, time and so forth. The apparent conclusion being that there was cloud obscuration of the landing lights anyway as the airplane was passing approximately 1,000 ft MSL. Can't really conclude what the crew would have seen and when, if they were looking. No indication in the CVR transcript that they were looking until after the Sink Rate alert. All of this, of course, way below the 1,200 ft MDA.
However, you might find the Aircraft Performance Group Report's treatment of this to be interesting. The analyst viewed numerous videos from the airport, noting both position, intensity of the aircraft's landing lights, time and so forth. The apparent conclusion being that there was cloud obscuration of the landing lights anyway as the airplane was passing approximately 1,000 ft MSL. Can't really conclude what the crew would have seen and when, if they were looking. No indication in the CVR transcript that they were looking until after the Sink Rate alert. All of this, of course, way below the 1,200 ft MDA.
Join Date: Mar 2014
Location: New Hampshire
Posts: 10
Likes: 0
Received 0 Likes
on
0 Posts
GlobalNav, you have cleared up some confusion for me. Thanks! But... now I have more questions.
They may have been aware of the terrain and had no concern since they expected to descend through the clouds at around 1000 AGL/1650 MSL, as I understand now? The fact that they didn't may have contributed to their lack of vertical awareness?
They may have been aware of the terrain and had no concern since they expected to descend through the clouds at around 1000 AGL/1650 MSL, as I understand now? The fact that they didn't may have contributed to their lack of vertical awareness?
Join Date: Feb 2014
Location: USA
Posts: 11
Likes: 0
Received 0 Likes
on
0 Posts
DMJ:
Actually, they may not have been particularly aware of the terrain. The thing about BHM is that it's hilly, not mountainous. I can't remember if the NTSB has established whether either of them had landed on 18 before, but I'm pretty sure that at least the captain had landed on the longer runway a fair few times. The hills for 06/24 are further away from the respective runway thresholds. The thing about 18 is that those small hills come all the way up to the threshold making the tolerances so much tighter in the last couple of miles. That's one of the issues aterpster is so hot about. There is little margin for error. If you look at the charts (pre-crash) the crew had....http://dms.ntsb.gov/public%2F55000%2...2F549718%2Epdf
...... there's little there that might point this out. You could look at similar charts for an airport near mountainous terrain and there would be all sorts of colored contours and warnings plastered all over them.
Probably. You can Google "expectation bias" to do with aviation.
They may have been aware of the terrain
...... there's little there that might point this out. You could look at similar charts for an airport near mountainous terrain and there would be all sorts of colored contours and warnings plastered all over them.
had no concern since they expected to descend through the clouds at around 1000 AGL/1650 MSL
DMJ. I am probably too tempted to speculate so I won't. Remember it was night time, IMC, though visibility reported to be ten miles. I wouldn't expect to see much terrain even though it's there, because of the darkness.
Wouldn't the altimeter be the primary cue for altitude, reaching minimums, and deciding to continue or go-around based on the required visual references? Can we suspect the crew was relying totally on the reported cloud height rather than their own instruments and real time view out the window?
As I have said before, perhaps the pace of the approach affected their anticipation of the event milestones. Twice the normal vertical rate, MDA twice as high above touchdown than they usually experience, means they reach MDA before their minds tell them to expect it. I may be wrong. But something must account for the mental picture they held.
I am not meaning to disparage the crew. What can be done to reduce the vulnerability of other crews to the same factors?
Wouldn't the altimeter be the primary cue for altitude, reaching minimums, and deciding to continue or go-around based on the required visual references? Can we suspect the crew was relying totally on the reported cloud height rather than their own instruments and real time view out the window?
As I have said before, perhaps the pace of the approach affected their anticipation of the event milestones. Twice the normal vertical rate, MDA twice as high above touchdown than they usually experience, means they reach MDA before their minds tell them to expect it. I may be wrong. But something must account for the mental picture they held.
I am not meaning to disparage the crew. What can be done to reduce the vulnerability of other crews to the same factors?
Join Date: Mar 2014
Location: New Hampshire
Posts: 10
Likes: 0
Received 0 Likes
on
0 Posts
I didn't think they would rely entirely on the cloud ceiling, but it may have been one more thing that led them to believe they were higher than they were.
From the beginning, I simply couldn't understand how x could happen, or why they chose to do y instead of z. My "x, y, and z" are things that the pilots here already know. All I had was speculation, and very uninformed and ignorant speculation at that. And so, I really appreciate the depth of knowledge and expertise here. I'll sit back and wait for the NTSB report or any further news and look forward to reading the opinions here.
From the beginning, I simply couldn't understand how x could happen, or why they chose to do y instead of z. My "x, y, and z" are things that the pilots here already know. All I had was speculation, and very uninformed and ignorant speculation at that. And so, I really appreciate the depth of knowledge and expertise here. I'll sit back and wait for the NTSB report or any further news and look forward to reading the opinions here.
Join Date: Mar 2008
Location: Western USA
Posts: 555
Likes: 0
Received 0 Likes
on
0 Posts
Global
Any discussion of the facts in this accident leads to disparagement of the crew, at least in my opinion. Plain and simple.
This was my airline for just short of 20 years, and I was a check airman for the majority of that time (DC-8 & B-747), and as a result I somewhat take it personally when my airline has an accident like this. The training on the two fleets I helped crew was conducted by very experienced, prior airline people with experience on type. As the airline grew, that changed. People were hired into fleet management off the street, in many cases, without previous airline experience or time in type. Crewmembers, filtered by HR hiring goals, were hired who were not necessarily the best technical candidates for the basic requirement of being an instrument pilot. Some were not even considered for interview, even though they had substantial time in type (but minimal or no college) along with a sterling work record, and others were interviewed because they met certain HR profiles and had four years of college but with minimal flight experience.
So you tell me, what can be done to reduce the vulnerability of other crews to the same factors? I think it is fairly obvious when success generally begins with a proper grasp of the basics based on the knowledge and experience of those who have gone before you...and applied to the best candidates available. Simple, but not so simple when exposed to the pressures of political, PC influences.
At least that's how I see it, but then I'm getting old and probably don't know what I am talking about.
I am not meaning to disparage the crew. What can be done to reduce the vulnerability of other crews to the same factors?
This was my airline for just short of 20 years, and I was a check airman for the majority of that time (DC-8 & B-747), and as a result I somewhat take it personally when my airline has an accident like this. The training on the two fleets I helped crew was conducted by very experienced, prior airline people with experience on type. As the airline grew, that changed. People were hired into fleet management off the street, in many cases, without previous airline experience or time in type. Crewmembers, filtered by HR hiring goals, were hired who were not necessarily the best technical candidates for the basic requirement of being an instrument pilot. Some were not even considered for interview, even though they had substantial time in type (but minimal or no college) along with a sterling work record, and others were interviewed because they met certain HR profiles and had four years of college but with minimal flight experience.
So you tell me, what can be done to reduce the vulnerability of other crews to the same factors? I think it is fairly obvious when success generally begins with a proper grasp of the basics based on the knowledge and experience of those who have gone before you...and applied to the best candidates available. Simple, but not so simple when exposed to the pressures of political, PC influences.
At least that's how I see it, but then I'm getting old and probably don't know what I am talking about.
Join Date: Apr 2008
Location: East Anglia
Age: 83
Posts: 437
Likes: 0
Received 0 Likes
on
0 Posts
Sir, You know exactly what you are talking about....the pity is nobody in the HR department has the background cockpit experience to understand this or if they do they are incapable of action.
If I was the boss of the airline I would be looking at my route structure and looking for the airfields that did not provide an ILS onto a 8000ft long runway that was free of terrain and obstacles and grading the airfields that did not fit this criteria by difficulty.
I would have a department of experienced pilots looking at the risks my line pilots were exposed to and developing training programmes to make sure my line pilots could deal with the risks that had been discovered. I would also be equipping my fleet with TAWS including terrain depiction on the efis map display and a 21st century FMS that was easy to use and capable of flying all approaches easily.
I would also be asking the chart manufacturers to provide distance/altitude tables on non-precision approaches and also depicting minimum safe altitudes for each segment of the approach for every airfield used.
I would have a department of experienced pilots looking at the risks my line pilots were exposed to and developing training programmes to make sure my line pilots could deal with the risks that had been discovered. I would also be equipping my fleet with TAWS including terrain depiction on the efis map display and a 21st century FMS that was easy to use and capable of flying all approaches easily.
I would also be asking the chart manufacturers to provide distance/altitude tables on non-precision approaches and also depicting minimum safe altitudes for each segment of the approach for every airfield used.
@Tubby. I share your sentiments. As a previous poster, Tom Imrich in Post #92 noted, it's time to make use of available technology and put precision approaches at these runways. GLS is here, it needs to be exploited. A single GBAS at an airport can enable precision approaches for practically every runway at the airport, perhaps even nearby airports for the price of a new approach procedure. Then every approach will work and look like practically every other one and the training problems and error prone procedures can be replaced. This is what NextGen should be about.
Join Date: Dec 2009
Location: U.S.
Posts: 8
Likes: 0
Received 0 Likes
on
0 Posts
"The thing about 18 is that those small hills come all the way up to the threshold making the tolerances so much tighter in the last couple of miles."
I'm sure there are lights on those hills, just like there are lights on towers and cranes. I wonder if any lights were NOTAM'd inop. A bright red beacon placed atop of an obstacle located in the approach path to a runway seems like a good idea, especially for pilots flying in there at night.
I'm sure there are lights on those hills, just like there are lights on towers and cranes. I wonder if any lights were NOTAM'd inop. A bright red beacon placed atop of an obstacle located in the approach path to a runway seems like a good idea, especially for pilots flying in there at night.
The Terrain
Granted the terrain profile is interesting. But the terrain didn't grab this airplane, it just interrupted an extremely and inappropriately steep flight path. The procedure, flown properly, met all the obstacle clearance requirements, albeit perhaps a bit less forgiving.
Had the terrain been flat, the sink rate alert would have been delayed, the 1,500 FPM VS would have continued, unabated, whilst the AP remained coupled. The paltry response to the sink rate alert may have been just as inadequate over flat terrain as it was at BHM. Perhaps a careful technical analysis would disprove that assertion, but the cause of this accident was, nevertheless, a badly flown approach.
Will additional training correct the situation? It might help, but, the operator already thought it was doing enough and given the high costs of training, how much more would they be prepared to do? It will not change the relative infrequency of nonprecision approaches, and crews' relative inexperience with them.
Time is marching on. It's time to improve the safety of operations. We may have fewer nonprecision accidents than we used to, but I suspect it's because fewer are flown. Not a bad thing, except crews are even less familiar with all the intricate FMS steps to fly them. We should get nonprecision approaches out of the menu as soon as possible and replace them with GLS precision approaches.
Had the terrain been flat, the sink rate alert would have been delayed, the 1,500 FPM VS would have continued, unabated, whilst the AP remained coupled. The paltry response to the sink rate alert may have been just as inadequate over flat terrain as it was at BHM. Perhaps a careful technical analysis would disprove that assertion, but the cause of this accident was, nevertheless, a badly flown approach.
Will additional training correct the situation? It might help, but, the operator already thought it was doing enough and given the high costs of training, how much more would they be prepared to do? It will not change the relative infrequency of nonprecision approaches, and crews' relative inexperience with them.
Time is marching on. It's time to improve the safety of operations. We may have fewer nonprecision accidents than we used to, but I suspect it's because fewer are flown. Not a bad thing, except crews are even less familiar with all the intricate FMS steps to fly them. We should get nonprecision approaches out of the menu as soon as possible and replace them with GLS precision approaches.
Guest
Join Date: Apr 2009
Location: On the Beach
Posts: 3,336
Likes: 0
Received 0 Likes
on
0 Posts
Wood:
I doubt there were any obstruction lights. It is a grassy field that was subjected to earth moving after the houses were removed, probably a long time ago. Nothing about that terrain meets obstruction lighting standards.
The PAPI was mandatory at night, at least for the two IAPs, which is all the obstruction lighting that was required for the Runway 18 close-in approach path.
I'm sure there are lights on those hills, just like there are lights on towers and cranes.
The PAPI was mandatory at night, at least for the two IAPs, which is all the obstruction lighting that was required for the Runway 18 close-in approach path.
Guest
Join Date: Apr 2009
Location: On the Beach
Posts: 3,336
Likes: 0
Received 0 Likes
on
0 Posts
GlobalNav:
FAA close-in obstacle clearance standards simply will not permit an approach with vertical guidance (FAA-speak these days for ILS, LPV, GLS, and LNAV/VNAV).
First, there is a TERPs criterion called the GQS (glide-slope qualification surface) which must be met from the DA point to the runway. Second, there is the wider visual segment, which ideally should be 34:1 on the sides of the GQS. The visual segment can have 34:1 penetrations, which can affect minimums. But, if a 20:1 is violated the FAA is "on a roll" to deny night minimums altogether unless there is a flight inspected VGSI (such as at BHM 18) that serves to mitigate the hazards at night in the visual segment. The Runway 18 PAPI (one type of VGSI) was not approved for some period of time. Then, it was flight inspected at the NOTAM at issue was issued to permit night minimums provided the VGSI was operating.
There are a lot of runway ends in the U.S. that have been disapproved for night IFR landing. This all came about when a Lear Jet hit some trees on short final that penetrated the 20:1 in the visual segment of an IAP.
There are an increasing number of GA runways that cannot meet the GQS requirements, so instead of having LPV, they have LP.
Time is marching on. It's time to improve the safety of operations. We may have fewer nonprecision accidents than we used to, but I suspect it's because fewer are flown. Not a bad thing, except crews are even less familiar with all the intricate FMS steps to fly them. We should get nonprecision approaches out of the menu as soon as possible and replace them with GLS precision approaches.
First, there is a TERPs criterion called the GQS (glide-slope qualification surface) which must be met from the DA point to the runway. Second, there is the wider visual segment, which ideally should be 34:1 on the sides of the GQS. The visual segment can have 34:1 penetrations, which can affect minimums. But, if a 20:1 is violated the FAA is "on a roll" to deny night minimums altogether unless there is a flight inspected VGSI (such as at BHM 18) that serves to mitigate the hazards at night in the visual segment. The Runway 18 PAPI (one type of VGSI) was not approved for some period of time. Then, it was flight inspected at the NOTAM at issue was issued to permit night minimums provided the VGSI was operating.
There are a lot of runway ends in the U.S. that have been disapproved for night IFR landing. This all came about when a Lear Jet hit some trees on short final that penetrated the 20:1 in the visual segment of an IAP.
There are an increasing number of GA runways that cannot meet the GQS requirements, so instead of having LPV, they have LP.
@aterpster. Thank you, I grant that there can be a host of details, maybe even impossibilities that prevent establishing precision approaches at every runway. In addition to the obstacle criteria you mention there are also infrastructure requirements for runway and approach lighting to account for.
In my judgement it was not the night nor the terrain/obstacles that created this accident, but your points are well taken.
This is perhaps "out of the box", but I would be happy if the minimums remained the same, for the sake of the terps, but rather than this pseudo vertical flight path nonsense there be an established vertical path a la ILS or GLS that from the flight deck perspective is as straight-forward to set up and fly as today's ILS approaches.
It is a multifaceted problem requiring a multifaceted solution, that certainly includes training and so forth. I do believe its time to address Rube Goldberg nonprecision approaches, no matter how well meaning. I am not satisfied to stop at the notion of "pilot error" without finding truly effective means to prevent it. This crew had a bad night, but I doubt they are all that unique amongst their peers in the business. Things are not entirely like they were 30 years ago and we need to stop trying to shoehorn today's reality into our assumptions from the good old days.
In my judgement it was not the night nor the terrain/obstacles that created this accident, but your points are well taken.
This is perhaps "out of the box", but I would be happy if the minimums remained the same, for the sake of the terps, but rather than this pseudo vertical flight path nonsense there be an established vertical path a la ILS or GLS that from the flight deck perspective is as straight-forward to set up and fly as today's ILS approaches.
It is a multifaceted problem requiring a multifaceted solution, that certainly includes training and so forth. I do believe its time to address Rube Goldberg nonprecision approaches, no matter how well meaning. I am not satisfied to stop at the notion of "pilot error" without finding truly effective means to prevent it. This crew had a bad night, but I doubt they are all that unique amongst their peers in the business. Things are not entirely like they were 30 years ago and we need to stop trying to shoehorn today's reality into our assumptions from the good old days.
Join Date: Dec 2006
Location: Florida and wherever my laptop is
Posts: 1,350
Likes: 0
Received 0 Likes
on
0 Posts
GlobalNav:
FAA close-in obstacle clearance standards simply will not permit an approach with vertical guidance (FAA-speak these days for ILS, LPV, GLS, and LNAV/VNAV).
First, there is a TERPs criterion called the GQS (glide-slope qualification surface) which must be met from the DA point to the runway. Second, there is the wider visual segment, which ideally should be 34:1 on the sides of the GQS. The visual segment can have 34:1 penetrations, which can affect minimums. But, if a 20:1 is violated the FAA is "on a roll" to deny night minimums altogether unless there is a flight inspected VGSI (such as at BHM 18) that serves to mitigate the hazards at night in the visual segment. The Runway 18 PAPI (one type of VGSI) was not approved for some period of time. Then, it was flight inspected at the NOTAM at issue was issued to permit night minimums provided the VGSI was operating.
There are a lot of runway ends in the U.S. that have been disapproved for night IFR landing. This all came about when a Lear Jet hit some trees on short final that penetrated the 20:1 in the visual segment of an IAP.
There are an increasing number of GA runways that cannot meet the GQS requirements, so instead of having LPV, they have LP.
FAA close-in obstacle clearance standards simply will not permit an approach with vertical guidance (FAA-speak these days for ILS, LPV, GLS, and LNAV/VNAV).
First, there is a TERPs criterion called the GQS (glide-slope qualification surface) which must be met from the DA point to the runway. Second, there is the wider visual segment, which ideally should be 34:1 on the sides of the GQS. The visual segment can have 34:1 penetrations, which can affect minimums. But, if a 20:1 is violated the FAA is "on a roll" to deny night minimums altogether unless there is a flight inspected VGSI (such as at BHM 18) that serves to mitigate the hazards at night in the visual segment. The Runway 18 PAPI (one type of VGSI) was not approved for some period of time. Then, it was flight inspected at the NOTAM at issue was issued to permit night minimums provided the VGSI was operating.
There are a lot of runway ends in the U.S. that have been disapproved for night IFR landing. This all came about when a Lear Jet hit some trees on short final that penetrated the 20:1 in the visual segment of an IAP.
There are an increasing number of GA runways that cannot meet the GQS requirements, so instead of having LPV, they have LP.
so no RNP approaches. More needs to be done to take full advantage of the capabilities of GLS the ground installation is cheap and simple and GBAS does not require repeated calibration. The pilot training is not exceptional either and if it was the standard instrument approach then all the flying schools would teach it.
But the major unexploited advantage is that you can be fully 'established' 'stabilized' on a GLS procedure with turns and changes of vertical profile as part of the automated final approach. Yes it will make it different for the procedure designers used to using straight lines at fixed descent rates for final approach and the definition of obstacle clearance surfaces may have to change, but the prize is standard instrument approaches to all runway ends anywhere.