UPS cargo crash near Birmingham AL

[mm43]

@ Ian W

If you were to extend the flight path back where the 'runway in sight' call would have been made from - that's adding under half as much of that dotted flight path, how would the runway have been 'in sight'? Certainly not the touchdown end, PAPIs and approach lights they would be hidden by the crest of the hill.

Would you agree it is looking more like the crew suffered a visual illusion pf some sort?

Not so sure about that! They could have been visual earlier, and as previously posited, got tempted for the Drive and Dive scenario. OTOH we don't know whether they acquired the PAPIs at a refracted angle, which lead them to believe they were high.

My overall impression of this approach is that it can be deceptive, be it daylight or night.

[valvanuz]

vertical speed

The issue is that we do not know the plane vertical speed.

If they had a high descent rate, 13 seconds before tree impact(+/- 850ft), they could be high enough to see the runway assuming normal visibility. If they has a low descent rate, they would not see the rway as the top of the hill (800-820 ft) is only 4 500 ft from threshold (645ft).

[roulishollandais]

@roulette,
... and I don't like to play "roulette russe".
That is the reason I wanted to emphasize the counterparts of "continuous descent". They are more than one. Once again poor automation had a part in these choices of design philosophy, not only the size and inertia of wide-bodies. You need a high rate of stabilization to realize autoland and have benefit of very low minima, generalized Cat II, Cat III.
RWY 18 in Birmingham,AL is another challenge aswell as many approaches in the middle of hills and mountains.
The last limitation against CFIT is the crew, AFTER GPWS... Dive&drive needs to know at any moment where you are. The different mandatory distances and heights are locked in the pilot's mind. Near of mountains wind may change during approach and with continuous descent method way of thinking you have illusion of ground stability, and you just ignore your real minimal marge, you are descending just like in a simulator with low concern. Only dive&drive way of thinking gives you insurance to be in the protected volume. If we use a distance/height table it is to be able to do a pitch/power correction depending if you are fast or late in your descending path.

Since they built continuous descent approaches they suppressed both the notion of the limits and the table, and sometimes worth : the FAF ! (LFST VOR-DME 05, AIR INTER 20. JAN 1990). But sold aircrafts with such commercial lies.

Freightdogs are used to be asked to land at time in every weather, and to take greater risk (it is a fact) as they have no passengers. The Transport letter LTA given to the crew show that everything has its own insurance. When I was freightdog the total to pay by insurances in case of crash was around 10 times what they would pay with aircraft full of passengers under Warsow/Montreal convention. Freight airlines have no concern with crash risk and costs. So their pilots are alone to protect their own safety, LOCKING in their mind exactly where they cannot go lower and in which hotizontal protection segment they are. That is what is safe in the dive&drive logic. But I know the risk exists to be one step forward in descending, (easy to avoid with good formation) but mean continuous descent is not enough to help if you forget the LIMITS.
Mean continuous descent was built to help automation who are still untrustful in TRANSIENT dynamic.

I do not say more than that when I say dive&drive is safer, as that method gives us more conscious pilots of the ground limits and respect of MDA.
Roulishollandais

[Capn Bloggs]

Not so sure about that! They could have been visual earlier, and as previously posited, got tempted for the Drive and Dive scenario.

Perhaps not. The autopilot was flying; they could have been distracted by the poor vis or confused about what they were actually seeing, all the while the autopilot is taking the aircraft lower and lower.

Nice Distance/Altitude table there, Dadanawa. Great minds think alike!

Originally Posted by Bloggs, Post 210

18 LOC approach according to Bloggs:

350ft per nm/3.28°.

21 7580
20 7230
19 6880
18 6530
17 6180
16 5830
15 5480
14 5130
13 4780
12 4430
11 4080
10 3730
9 3380
8 3030
7 2680
6 2330 (limit 2300)
5 1980
4 1630
3.3 1385 (MDA)
3 1280
2 930
1 580

Get on it, stay on it. End of story. Too hard to work out? Get the FAA to put it on your charts.

[deSitter]

I wonder if it's possible that the PAPI was obscured by the hill, but a mix of red and white lights was seen in the rush hour traffic on the highway (Interstate 59) just beyond the opposite end of the runway? A crew that was very tired could possibly have been pulled in like this, particularly in the type of light one has at that time of morning.

Such an illusion could be even more convincing if most of the runway lights were also obscured by the hill.

-drl

[WillowRun 6-3]

Book & GroupThink (Board style)

Two questions with comments embedded;
1. What interest would there be amongst those PPRuNe-ers who have been active on thread (or following it closely) in a new thread for discussion amongst and between members who are reading the "Rethinking" book (Kindle, Majorly Tall Woman, Mortared Brick or from whatever source derived)? I hesitate beyond pausing to start a thread that would just draw contempt ridicule derision and derisive scorn from the cognoscenti for whom it would be intended to be interesting / useful / informative / maybe entertaining (though not while driving but I digress). If Oprah can have a book club, well, the million-plus PIC hours -(aggregate) frequent drivers club could have very stimulating and learned discussion of what those who are reading the book - promoted by the ghost of 1354 - think of particular arguments developed therein, and so on. Kind of like recurrent training, but not on the dynamic motion base six-degrees-of-freedom Magenta Line queen. Maybe it might even build toward consensus for a return to hand-flying as an emphasis in CAAs. What sayeth the Community herein?

2. Does anyone know whether, and if so to what extent, the attitude and expertise very manifest in-thread -- the drill into it, hit all the details, keep going until the facts are assembled Complete(ly), speak your mind freely on all matters of interpretation or conjecture, and just a general refusal to accept the Big Shrug (in my steel mill youth, I recall, a Safety placard read: Accidents Don't Just Happen. They Have Causes. no shrug allowed) -- this level of expertise exists and is applied at NTSB, right? Right? But do they (the Board staff) have a process like the GroupThink that has produced this thread? Yeah, that too: I hope the Board staff is right over yonder, lurkin' away.

I'm WillowRun 6-3, which BTW does not stand for "verbose lawyer". Good Day and "be careful out there"

[tubby linton]

WillowRun 6-3, perhaps you could send the NTSB an email and ask them to have a look at this thread.

[WillowRun 6-3]

Email?

tubby, do you think really that would be a good idea? I mean what with dispatching Go Teams a lot lately, they're already pretty busy. Besides, I wouldn't want them to call me a non-profit and then audit my taxes (all those gov't agencies work really closely together, you know). But I might use a safer method of communication: I'll go back to where I used to perch my rack in the summertime during my Zen Vagabonding at O'Hare trip, and mark the message up in chalk on the flat vertical concrete sides of the 15-foot (approx) semi-circular vents left and right of the CT (the old one), you know alongside the walks between T-2 and the hotel - just as I weekly chalked the University of MICHIGAN football scores in Fall Term 1983 (until I returned to Ann Arbor, anyway). True story. Do you think NTSB knows where to look? I mean at O'Hare, for chalk signs??

[PEI_3721]

Capn Bloggs
‘Nice Distance/Altitude table there, … Dadanawa’
But please ALTITUDE before range; incorrect altitude kills you, usually before incorrect range.
See previous link:- incidents, particularly #3.

For the Dismukes searchers (and Reason):-

http://human-factors.arc.nasa.gov/fl...ISAP01_417.pdf

Human error: models and management

http://www.signalcharlie.net/file/vi...s%20Summit.ppt

[Capn Bloggs]

Originally Posted by PEI_3721

But please ALTITUDE before range; incorrect altitude kills you, usually before incorrect range.

Fair comment but not accepted (I did read the report you linked to). Before the approach-proper commences (further out than the FAF, we use 10nm if not doing a VNAV approach) the aircraft would be protected by the Altitude Selector set at the MSA.

Approaching the commencement point as shown by the distance, the ALT SEL is wound down to MDA and the approach commenced. Thereafter, it is easier to monitor your profile error by checking the altitude against upcoming DME distances eg "8 DME, 100ft Low" "correcting!". You know exactly what the error is and how much you need to correct to be back on profile by the next mile. If it was "4000ft, .4nm too far out", it's harder to conceptualise and fix the profile error, high or low.

That said, one-time checks such as checks at the FAF should be as you say: "FAF check 2500ft" (obviously called before you get there!) is more logical because it is a catch-all check, not a profile monitor/correction technique.

Done many hundreds of these; Distance/Altitude is easier than Altitude/Distance.

[Coagie]

Coagle,

There is a book you need to read...like yesterday.

The Limits of Expertise: Rethinking Pilot Error and the Causes of Airline Accidents (Ashgate Studies in Human Factors for Flight Operations):Amazon:Books

737er, Thanks. I'll see if I can dig it up and check it out. You seemed to have peaked interest in the book among those in this thread. 737er, you may rival Don Imus or Oprah Winfrey, in getting a book on the New York Times Best Seller list! Good work! I'm sure everyone, who reads the suggested book, will get something useful out of it. The whole Human Factors thing is fascinating, and can be applied in a broad array of endeavors.

Book & GroupThink (Board style)
Two questions with comments embedded;
1. What interest would there be amongst those PPRuNe-ers who have been active on thread (or following it closely) in a new thread for discussion amongst and between members who are reading the "Rethinking" book (Kindle, Majorly Tall Woman, Mortared Brick or from whatever source derived)? I hesitate beyond pausing to start a thread that would just draw contempt ridicule derision and derisive scorn from the cognoscenti for whom it would be intended to be interesting / useful / informative / maybe entertaining (though not while driving but I digress). If Oprah can have a book club, well, the million-plus PIC hours -(aggregate) frequent drivers club could have very stimulating and learned discussion of what those who are reading the book - promoted by the ghost of 1354 - think of particular arguments developed therein, and so on. Kind of like recurrent training, but not on the dynamic motion base six-degrees-of-freedom Magenta Line queen. Maybe it might even build toward consensus for a return to hand-flying as an emphasis in CAAs. What sayeth the Community herein?

2. Does anyone know whether, and if so to what extent, the attitude and expertise very manifest in-thread -- the drill into it, hit all the details, keep going until the facts are assembled Complete(ly), speak your mind freely on all matters of interpretation or conjecture, and just a general refusal to accept the Big Shrug (in my steel mill youth, I recall, a Safety placard read: Accidents Don't Just Happen. They Have Causes. no shrug allowed) -- this level of expertise exists and is applied at NTSB, right? Right? But do they (the Board staff) have a process like the GroupThink that has produced this thread? Yeah, that too: I hope the Board staff is right over yonder, lurkin' away.

I'm WillowRun 6-3, which BTW does not stand for "verbose lawyer". Good Day and "be careful out there"

WilloRun 6-3, if you're paid by the word, rather than the hour, you must be a very wealthy lawyer! Kidding aside, although the thin skinned should proceed with caution when participating, it's good to see you appreciate the value of Internet brainstorming.

[olasek]

.To a large extent, IMO AF447 and Asiana were "set up" by their employers.

Total rubbish, in other words thousands pilot are being 'set up' by their employers every day.

Such an illusion could be even more convincing if most of the runway lights were also obscured by the hill.

There was an accident years ago on a night approach to one of Hawaiian airports, crew of a business jet lost sight of PAPI but interpreted this that a small cloud got between them and the PAPI, in fact they were wrong, there was no cloud, they were too low. Everyone perished.

[Coagie]

There was an accident years ago on a night approach to one of Hawaiian airports, crew of a business jet lost sight of PAPI but interpreted this that a small cloud got between them and the PAPI, in fact they were wrong, there was no cloud, they were too low. Everyone perished.

olasek, it's been unusually rainy around Birmingham this summer, and 5 in the morning is prime time for fog. Your example may very well have happened in this crash. The pilots thinking the lights were obscured by a low lying cloud or bank of fog, not realizing it was a hill. There's also a chance of light refracting over the crest of the hill, as sometimes happens in moist conditions, making it seem like they had straight line of sight to the runway, when actually, the light was "bent" over the crest of the hill by refraction.

[Coagie]

No, you got it backwards, blaming institutions is more satisfying than blaming individuals. Yes, I am back to react to nonsense.

Not only more satisfying, but, in most cases more comfortable for people, because they feel less guilty blaming a large, soulless entity, than an individual, who might get their feelings hurt. Sometimes the problem is with the system, but the problem with blaming the system, when it's an individual problem, is, often the solution on a systemic problem involves changing things that are working, thus risking throwing the baby out with the bathwater, otherwise know as the law of unintended consequences, where solving one problem, causes other, perhaps more significant problems. Giving merits and demerits to an individual isn't as risky to what works in the system. That's why it's important to see the big picture, and identify an error as systemic, or an individual anomaly, as the solutions are often different.

[PEI_3721]

Altitude vs Range

Capn’, horses for courses
The objective is to encourage people to think and to aid cross checks during the procedure descent, thus promoting the highest level of safety.
I prefer not to overuse ALT SEL; it has great safety value in setting up an approach procedure and after a missed approach. However, when proposing to use it elsewhere the safety value has to be balanced with increased workload / distraction – another thing to be forgotten, which if depended on (automation dependency) might have serious consequences.
I am further biased by some older systems which resulted in hazardous situations of inappropriate ALT capture during go around (slight dip below ALT SEL).

Perhaps your technique is more focused on conducting the (NP) approach – ‘how go’s it’, correct for accuracy; whereas my view is more of a gross safety check, which in some circumstances (see previous link) requires an immediate climb to a safe alt / profile.

Safety isn’t necessarily ‘easier’ . The debate is like questioning ‘how can we do this’ vice ‘should we be doing this’.

If in this accident the procedure was commenced at an incorrect range, which resulted in being consistently low (assuming a constant approach), the error might have been detected by an intermediate check of altitude and range. This method has some consistency with crew activity at MDA where altitude must dominate.
When and where the error could be detected depends on the choice and number of alt/range entries, which in this case was 2, BASKN or IMTOY, where the latter might have been too late.

With rwy 18, the FAA could have considered the safety aspects differently. If a procedure is judged not-safe at night what makes it safe by day (cf BOAC comments)?
Whereas day operations might be judged sufficiently safe (acceptable risk with mitigation), is a VGSI mitigation sufficiently acceptable at night – lack of ground plane and textured surface for peripheral altitude checks; not ‘how’ but ‘should’ a night approach be authorised.

Of course it’s easy to question history with hindsight, but the annoying aspect is that in this area of aviation, history keeps on repeating itself; thus what do we require in order to learn from history?
.

[tubby linton]

The Papi and Loc/Dme for 18 were notamed u/s today, the notams show them unservicable for 48 hours.

[Old Boeing Driver]

Just the facts Ma'am

Dear Aviator Friends:

I've been a lurker on this forum for years and have enjoyed reading the information posted. I've come to know that there are a lot of people here that are a lot smarter than me.

However, after 40 years of flying 2, 3,and 4 engine jets all over the world, I have come to a point where I have to make a few comments.

I may get flamed for my opinion, but here goes.

1. The fact that the procedure was NA on some charts does not matter
2. The fact that Mrs. Benson's trees were 921' instead of 915' does not matter
3. The fact that the controller saw their lights does not matter.

What matters is that they were over 200' lower than the MDA for the approach.

Had they leveled off at an MDA of 1,200 and flown until they saw 2 white/2red PAPI's and the REIL's, they would be home with their families, and we would not be having these discussions.

We can discuss mirages, refractions, tree heights, runway in sight call outs, etc., etc., but the basic fact is they were lower than published minimums.

Now, why were they low? The NTSB should have plenty of data to tell us soon.

In the meantime, let me throw out a few more points for discussion.

The NTSB only noted EGPWS warnings and sounds consistent with hitting the trees, however the CVR went on for quite a while. Why weren't there other “expletives deleted” or call outs?

Did the A/P miss a level off? (maybe due to setting the MA altitude before it captured?)

Was there a pilot incapacitation at a critical moment?

Were both heads in the cockpit for some reason? (Remember EAL in the Everglades?)

Thanks to my fellow aviators for your comments here.

Regards

[Passenger 389]

Recording the View from the Cockpit ??

In investigating incidents, it seemingly could be helpful to have a record of what the flight crew might have seen while looking outward from the cockpit at critical times, to help understand why the crew took (or didn't take) certain actions, and to improve future design, training, and procedures.

For instance, on this thread some are busily speculating whether the view of the runway might briefly (or partly) have been obscured by patchy clouds or fog, or perhaps by trees mistaken for patchy clouds or fog. Did they see the entire runway, or only the far end? Could they see the PAPI and the strobe near the runway threshold? (I won't even touch the "mirage" theories. And yes, perhaps what the crew saw is pointless, as they never should have been so low at that location and immediately should have initiated a go-around, but for some reason they didn't and we need to understand why).

There have been other instances in which questions arose about whether a crew might have mistaken a nearby road for the runway, or if a crew's view was seriously impaired by the rising sun or blowing sand or heavy rain. Then there are (alleged or actual) laser beams impairing view or sunlight reflecting, taxiway collisions, runway excursions, various illusions....

I could list dozens of other examples, but the point seems self-evident.
Recording inside the cockpit (showing the flight crew) understandably is very controversial. But would it make sense (and is it technically feasible) to record outward, and capture that data on future data recorders?

A camera cannot exactly replicate what a crew could see -- camera height, angle, positioning, focal length, color, and light sensitivity are just some variables (though software can try to compensate). Unlike a camera the crew also must do more than constantly scan outward, with nary a blink.

There also is the "ghoul" factor. In the YouTube era, the NTSB or other applicable entity would have to safeguard the video as it does the CVR tape.

Still, is this an idea worth considering, or is it unnecessary (or too controversial as potentially the nose of the camel entering the tent)?

[Lonewolf_50]

A camera cannot exactly replicate what a crew could see
-------------
what the flight crew might have seen while looking outward from the cockpit
at critical times

You've answered your own question.

Tubby: being NOTAM'd today means what, exactly, in relation to the mishap in the past?

[A Squared]

Just a guess, they're doing flight checking or other testing on those systems today