PPRuNe Forums

PPRuNe Forums (https://www.pprune.org/)
-   Tech Log (https://www.pprune.org/tech-log-15/)
-   -   UPS cargo crash near Birmingham AL (https://www.pprune.org/tech-log/521370-ups-cargo-crash-near-birmingham-al.html)

tubby linton 18th Aug 2013 17:12

The following is from an FAA chart and lists the obstacles for a departure from RW36, i.e in the RW18 undershoot.

Rwy 36: Fence, tree, and vehicles on road beginning 130’ from DER, left and right of centerline up to 28’ AGL/662’ MSL. Tree and pole beginning 467’ from DER, 470’ right o fcenterline, up to 63’ AGL/726’ MSL. Trees beginning 637’ from DER, 420’ left of centerline, up to 13’ AGL/696’ MSL. Pole and tree beginning 1353’ from DER, 669 left of centerline, up to 37’ AGL/739’ MSL. Trees, poles, and fence beginning 1279 from DER, 492’ right of centerline, up to 19’ AGL/713’ MSL. Terrain beginning 43 feet from DER, left and right of centerline, up to 794’ MSL

DER=Departure end of runway.
On an FAA website I found an old version of the GPS approach and the descent angle was originally 3.04 degrees. Chart is dated July 2008.

A Squared 18th Aug 2013 17:28

Tubby, that distance in the last item appears to be in error. It doesn't follow the trend in the list of getting further from DER, and 150 ft obstacles 43 feet from the runway??? :eek:

Airbubba 18th Aug 2013 17:32


What does your fraternity of PIC and civil aeronautics do to honor their memory? Will a scholarship fund be established for their children or nieces or nephews or the kids at the local church temple mosque or yoga studio if they have none of their own? or at the Boy Scouts or Girl Scouts or 4-H or Explorers or the entity which the US ought to form combining all four of these groups? I began my career as a labor attorney and I don't even know if they were organized into a labor union! What can be done that will honor their memory, and at the same time help the taxpayers of this country realize that aviation safety doesn't just happen. It does not just happen by itself.

I'd gladly volunteer my legal services, saving to suitors the proverbial dollar (Sound as a Dollar), to set up a trust or scholarship. Or to press for legislative reform of how FAA updates approach corridors. I'm WillowRun Six-Three, good day.
Some memorial information on one of the pilots is here:

UPS plane crash: Pilot Shanda Carney Fanning's funeral, life celebration planned | al.com

Both pilots were members of this labor union: Independent Pilots Association

tubby linton 18th Aug 2013 17:35

I have re-read that a few times and that is what is says . The reference is here:
https://charts.aero/airport/KBHM and the specific chart is Birmingham 5 cont.1

PJ2 18th Aug 2013 17:36

BOAC, re "Yes thanks - do we have an initial impact elevation? "

Even given the inaccuracies described regarding Google Earth's topographical data, using the "ruler" tool in G.E. to draw a line from the 1000ft point on 18 along the approach path over the two houses (photographs showing broken trees), and using the Altitude control, "Clamped to ground" feature available when saving the line to "My Places", one can obtain a slightly better appreciation for the character of the terrain.

And using the Altitude feature, "Relative-to-ground" and placing the same line at the runway elevation, one can see that the line "disappears" anytime it is going through terrain higher than the altitude selected by moving the slider on the "ground - space" control.

Viewing from the side, (perpendicular to the approach course, looking north for example), the same line "Clamped to the ground" shows a rough approximation of the undulation of the terrain.

The limitations are obvious (for example, the actual descent path obviously cannot be described at this time) but with the cautions expressed, it does give one something by which one may roughly examine the terrain.

PJ

WillowRun 6-3 18th Aug 2013 17:49

I liked Harrison Ford's ambulance driving but, no
 
Sorry this was left obscured. Volunteer my legal services, no charge, free, except for a nominal One Dollar U.S. (unfolded). But not pro bono, which is a slightly different concept of service. I am interested in what sometimes is called public advocacy, not the same thing as pro bono publico, and quite a lot not the same as politics or running for public office. No relation whatsoever to the fee-chasing lawyers who try to sign up injured parties or their surviving kinfolks after an accident - thank you, not why I went to law school.
Thanks for the info on the memorial and the union.
See ya.

BOAC 18th Aug 2013 18:11

Thanks, PJ - I do not have time (I am familiar with the process from wireless broadband mast siting) but if, as you say, the line 'disappears', it presumably means the airfield would have disappeared largely or completely from sight (allowing of course for cockpit height.) Viewing 4 reds through trees is not a preferred option.

WillowRun 6-3 18th Aug 2013 18:20

Terrain
 
PJ, sir, is it reasonable and accurate - based on the data actually in hand and the review, analysis or other cogitation by learned professionals especially and particularly PJ2 - to say that there appears to be a significant mismatch between the terrain measurements or depictions (or some similar element) upon which the approach corridor was designed - and the actual terrain? It appears that a major factor here will be a deficit in reliable and accurate topographic data? (subject to the above-noted qualification of data preliminary in nature and probable cause analysis yet to be stated). :ugh:

Speed of Sound 18th Aug 2013 18:21

Latest topographical map of area.
 
Although unfortunately the contours aren't labelled but you can see that the initial impact point was to the west of the highest point on the ridge although that's where the cockpit section of the aircraft came to rest as the rest of it continued down the hill.

http://i1280.photobucket.com/albums/...ps5af1082f.jpg

A Squared 18th Aug 2013 18:44


PJ, sir, is it reasonable and accurate - based on the data actually in hand and the review, analysis or other cogitation by learned professionals especially and particularly PJ2 - to say that there appears to be a significant mismatch between the terrain measurements or depictions (or some similar element) upon which the approach corridor was designed - and the actual terrain? It appears that a major factor here will be a deficit in reliable and accurate topographic data? (subject to the above-noted qualification of data preliminary in nature and probable cause analysis yet to be stated).
No. There is nothing in this discussion which should be construed to mean that the approach or PAPI was designed using erroneous terrain data, nor is there any suggestion that the LOC 18 approach and PAPI installation do not meet the applicable standards and specifications.

aterpster 18th Aug 2013 18:57

Speed of Sound:

The dark contour line is 700 feet, msl. Trouble is, it is meaningless because so much surface material has been moved around since that chart was issued.

A Squared 18th Aug 2013 19:00


Latest topographical map of area.
Although unfortunately the contours aren't labelled but you can see that the initial impact point was to the west of the highest point on the ridge although that's where the cockpit section of the aircraft came to rest as the rest of it continued down the hill.
No, that isn't the latest topo map of the area, it's the same vintage as the one previously posted by aterpster:

Aterpster's post with topo map.

And your view does not include the initial impact area, that was well to the north of your picture. The houses and streets in the north portion of your view have been removed. aterpster's map has a much close indication of the initial impact point. The aforementioned Ms Benson has stated that the airplane hit trees in her yard, and dropped airplane parts on her property which is located at the intersection of Tarrant-Huffman road and Treadwell road. That would be the intersection of the red and white striped roads, just west of the "impact" symbol on aterpster's map.

Sorry Dog 18th Aug 2013 19:01


Originally Posted by A Squared
Quote: do we have an initial impact elevation?

The terrain in the vicinity of where the airplane started hitting trees is between 750 and 800 ft MSL. Trying to define it any closer than that without actually being there and measuring thing would be futile.

Comparing it to the topography map the first trees hit are on an elevation of 790 to 810 and the first trees are the smaller 40-50 foot tall variety. Google Earth appears to estimate 20 feet higher compared the topo. The second larger tree in front of the house is probably 30-40 feet lower in elevation, but nearly that much taller too. By that time the elevation had deteriorated another 20-30 feet based on it being topped off there. The hill elevation is 790 to 810.

This is based on the topo, and me actually taking a look at the site... so they were not below rwy elevation, but MDA is another matter....

PEI_3721 18th Aug 2013 19:03

Re Terrain profile / EGPWS alerts;
I recall a briefing by Honeywell at an operator’s conference (8 yrs ago ?) which indicated continuing improvements in the terrain clearance warnings to help avoid undershoots.
The terrain clearance floor (TCF) ‘cut off’ distance originally ended at 1nm from the runway threshold, but was to be reduced to 0.5nm assuming accurate runway detection and availability of survey data, which should be the case in the US.
A further reduction to 0.25 nm might be available with GPS based aircraft position. The effect of these changes would provide a warning boundary of 150 ft above threshold datum at 1.25nm to 0 ft at 0.25nm.

Does anyone have further details of this, and if the modifications were adopted which software mod package (SW 218-218)?
w.r.t. this accident what was the EGPWS mod state, and what navigation source was used for EGPWS position?

barit1 18th Aug 2013 19:18

Smilin_Ed:

Does this aircraft not have a radar altimeter?
The undulating terrain at KBMH would make any use of RA problematic. Recall that the Polish presidential Tu-154 that crashed in Russia 10 Apr. 2010 was using RA, which misled the crew as they approached over a ravine.

In that case, the ravine was below the airfield elevation; at KBMH the ravine (or other low terrain) was below the intevening hill.

Sorry Dog 18th Aug 2013 19:20


Originally Posted by Aterpster
The dark contour line is 700 feet, msl. Trouble is, it is meaningless because so much surface material has been moved around since that chart was issued.

Most of the grading was done south of the final impact. Also, the house existed before then as well as the hill across the street where the trees were initially hit. Assuming the old topo is reasonably accurate, then that gives us a an approximate benchmark. Even if you say a +- of 40 feet then you can still start to draw some conclusions...

PJ2 18th Aug 2013 19:31

WillowRun 6-3, I concur with A Squared's response. AFAIK, one would never use Google Earth's terrain data for such work or verification of such work.

I've plotted the approximate approach path described above but won't post it as it cannot be verified without proper topographic mapping data. The only intention was to view an approximation of the approach terrain, (hilly, vice flat, etc), possible visibility of the runway when lower than the IMTOY minimum altitude, etc. We may expect that the NTSB will do the precise work and the actual approach profile will be known then.

PJ2

WillowRun 6-3 18th Aug 2013 19:45

Terrain, contd
 
PJ, I regret having interrupted the thread with an incorrect thought process. I shall continue with an effort (very inexperienced though it may be) to acquire a better understanding of the subject matters discussed herein, I mean, on this thread.

aterpster 18th Aug 2013 20:01

A Squared:

That would be the USGS Birmingham North and Irondale 1:24,000 topos.

I have no idea of the date of mine because they are WSG84 seamless bitmaps from Delorme. For my purposes they are usually fine because I am interested primarily in terrain.

Speed of Sound 18th Aug 2013 20:26


No, that isn't the latest topo map of the area, it's the same vintage as the one previously posted by aterpster:
My mistake.

I downloaded it from the US Geological Survey website and assumed it was a contemporary map. It is actually from 1978.

But as Sorry Dog (who is local to the area) says, the grading was done between the foot of the hill and the end of RWY18. In that case we can assume that the ridge shown is 794’ MSL as the FAA notice will have used information from an actual survey of the terrain and the hill top remains unaltered from 1978.

When I said initial impact I meant with the ground rather than the trees as I was referring to the area contained within the portion of the map I posted.

tubby linton 18th Aug 2013 20:31

How often do the trees on the approach path get measured and have their tops cut?
There is a point just to the southeast of Imtoy that shows on the original GPS chart of 915ft with a + and - next to it. On the later chart this has been changed to 921 ft again with the + and -. Interstingly the same point is depicted on the current GPS 36 chart as 915 ft +-. This to me shows some sloppiness in the editing of the charts.

Speed of Sound 18th Aug 2013 20:40

Tubby Linton
 
I would say not at all.

The trees in the area look fairly mature which means they will remain at that height until they fall or are felled.

Coagie 18th Aug 2013 20:58

Thanks for posting that, Airbubba. From reading this funeral notice for Shanda, it seems like she had the passion for flying, that folks might want in a pilot. I get the feeling that she was a competent pilot. I haven't seen much about the other pilot, other than he was rotor-wing in the Marine Corps, as stated in the article. It seems like there must have been some bad data, an oversight, or a random error, maybe coupled with some fog (that time of morning, with all the rain this summer, you see it clinging to the hills), leading to this crash, as this thread seems to be narrowing down, rather than incompetence, like in the Asiana crash. It's just the feeling I get. Of course, having a passion for and being good at something, don't necessarily follow one another. For instance, I have a passion for Golf ...

tubby linton 18th Aug 2013 21:02

The current LOC18 chart also shows this point as 915+-.. It is only six feet but it is only that last six feet which will kill you.Again to me it shows poor editing.

A Squared 18th Aug 2013 21:06

At this point it seems we can be fairly certain that the airplane started striking trees about a mile from the threshold, at an altitude somewhere around 300 feet or more below MDA, 150 feet or more below the PAPI, and at least 100 feet above airport TDZE.

So the question becomes "how did they arrive at that point, that far below the MDA and that far below the commissioned slope of the PAPI?"

Not to disparage anyone's efforts, but trying to refine those numbers any more precisely doesn't seem to change the overall scenario.

Coagie 18th Aug 2013 21:16


The trees in the area look fairly mature which means they will remain at that height until they fall or are felled.
I have a place, maybe 40 miles southeast of Birmingham, and the mature pine trees are over 100 ft (30 meters or so). I wonder if the trees involved in the crash are only 40 or 50 feet? I guess, if the soil isn't that deep, as on some hills, or if they aren't pine trees, the trees may stay as low as 40 or 50 feet at maturity, but it'd be worth double checking. I mention pine, as they are common around the south and grow very quickly, but there are others just as tall.

A Squared 18th Aug 2013 21:21


Originally Posted by Tubby Linton
The current LOC18 chart also shows this point as 915+-.. It is only six feet but it is only that last six feet which will kill you.Again to me it shows poor editing.

Keep it in perspective. It's 279 feet (or 285 feet) below the MDA. If you find yourself in a position where that 6 feet makes a real difference to how your day turns out, you had much larger problems which resulted in you being there at that altitude.

Speed of Sound 18th Aug 2013 21:35


I wonder if the trees involved in the crash are only 40 or 50 feet?
Sorry Dog was actually at the site of the tree strike at Mrs Benson's house.

I think he said the trees were 60-70 feet tall. There is a photo somewhere.

tubby linton 18th Aug 2013 21:43

My point was about the general editing standards of the FAA charts. Comparing them to the Jepp charts on the LOC 18 chart a point is shown at 910 ft SouthWest of the approach course ,but on the GPS chart a point is shown at 921 ft but further to the Southeast than the point shown on the FAA chart.
I have always been told that Jepp basically replicates the chart published by the state aviation authority but with their own formatting so why is my 915/921 feet point not shown correctly.
The Jepp chart for 36 does not show the point at all.

aterpster 18th Aug 2013 22:04

Here are the official FAA runway and PAPI data for Runway 18-36. Have at it math gurus:

KBHM

Coagie 18th Aug 2013 22:10


Sorry Dog was actually at the site of the tree strike at Mrs Benson's house.

I think he said the trees were 60-70 feet tall. There is a photo somewhere.
Sorry Dog is, I think, who I got the 40 or 50 foot height from. I figured, since he's near the site, he could take out an old camera range finder or something and double check the height, since trees in central Alabama are often over 100 feet tall.

A Squared 18th Aug 2013 22:19


but I think he has a valid point directed at the quality control of reasonably important documents.
Alright, I'll bite:

How would you describe "acceptable quality control" for sets of measured data?

I would suggest that it would be something like "correct to within the accuracy tolerances demanded by the purpose"

Seem reasonable?

OK, now, if you were preparing to fly a localizer approach to 18 and you noted the highest obstacle between the FAF and the runway showed a charted elevation of 921' How would you fly that approach differently than if that same obstacle had shown an elevation of 915' ?

If your answer is "exactly the same" then perhaps the quality control is within the standard I suggested?

tubby linton 18th Aug 2013 22:53

Surely a pilot should have information as to all of the charted obstacles relating to their approach? At this airfield we have two chart sources that cannot agree as to which obstacles they are going to depict or the accuracy of the data.
Having cleared the undershoot area why has the airport not provided approach lighting? Perhaps they never intended it to be used regularly at night or in poor weather.
Standard minima as used in Europe does not allow the LOC approach to be used at night.
Another airfield I can think of that has steep terrain in the undershoot is Lanzarote on RW21. Even this airfield has HIALS and a papi set at 3.7 degrees for that runway.

A Squared 18th Aug 2013 23:06

The runway has a PAPI. Assuming it was functioning as commissioned, (Unknown at this point) the accident would not have happened if they were on the PAPI.

I'll ask you: how would you fly the approach differently if you knew the obstacle was 921 ft vice 915?

Again, I would suggest if you wouldn't fly ten any differently, they you're getting wrapped around the axle over something of no consequence.

Did you happen to notice that the elevations in question are annotated with a plus-minus symbol. What is the significance of that?* and how does that affect how you would fly this procedure.

Jet Jockey A4 18th Aug 2013 23:07

Latest info from the NTSB says ...
 
The autopilot was on until seconds before the impact and so were the auto throttle. The wording in the two different sentences is a bit confusing but the quote from the NTSB says it was on at impact.

For those type rated on the A300, is there an AFM restriction on how low you can fly with the A/P engaged on a non precision approach?

Copy and paste from the article...

BIRMINGHAM, Alabama (Reuters) - The UPS cargo jet that crashed in Alabama this week, killing its two crew members, was flying on autopilot until seconds before impact, even after an alert that it was descending too quickly, authorities said on Saturday

"The autopilot was engaged until the last second of recorded data," said Robert Sumwalt, a senior official with the National Transportation Safety Board.

He said information retrieved by investigators from the flight data recorder aboard the United Parcel Service jet showed that its auto throttle also was engaged until moments before the fiery crash.

The Airbus A300 jet was approaching the runway at Birmingham-Shuttlesworth airport before dawn on Wednesday when it clipped the trees in an adjacent residential area and crashed into a steep embankment well short of the runway.

Sumwalt, who spoke at a media briefing near the crash site, had said on Friday that the pilots received a low altitude warning barely seven seconds before the sound of impact. He repeated that in his remarks on Saturday but did not say whether the alert had triggered any attempt by the crew members to disengage the autopilot as part of a last-ditch attempt to abort landing and re-gain altitude.

The pilots did not issue a distress call.

Sumwalt stopped short of saying there was anything unusual about a so-called "instrument approach" to the airport using autopilot.

But he said the NTSB would be looking closely into "UPS's instrument approach procedures" and how it typically went about guiding a large cargo hauler to touchdown on Birmingham-Shuttleworth's Runway 18.

That's the runway the UPS jet was approaching when it crashed and Sumwalt said the investigation would include a flight test at the airport in a UPS A300.

Kevin Hiatt, president and chief executive officer of the Flight Safety Foundation, an Alexandria, Virginia-based international watchdog group, told Reuters in an interview on Thursday that a "full instrument" landing was not highly advisable at Birmingham-Shuttlesworth.

The airport can be tricky to land at because it is nestled among hills and that is especially true of Runway 18, said Hiatt.

Hiatt, a former Delta Airlines pilot, said he had touched down on the runway many times himself.

"It is not a full instrument landing. You have to visually fly into that runway," he said. "Sometimes it takes nuance to land there. You have to realize that hill is there or you could come in too low."

The crash occurred shortly before dawn in rainy conditions as low-lying clouds hung over Birmingham.

So far, Sumwalt said there was nothing to indicate the crash was caused by engine failure or any mechanical issues.

He also said the runway lights were examined and found to have been "within one one-100th of a degree of being properly aligned" at the time of the crash.

UPS has identified the dead crew members Cerea Beal Jr., 58, of Matthews, North Carolina, and Shanda Fanning, 37, of Lynchburg, Tennessee.

As a standard part of any accident investigation, Sumwalt said the NTSB was looking into the physical and mental well being of Beal and Fanning in the 72 hours before the accident.

Beal, the captain of the downed aircraft, had about 8,600 hours total flying experience, including more than 3,200 hours in the Airbus A300, according to the NTSB.

AKAAB 18th Aug 2013 23:10

WAG - High on profile, Open Descent or selected V/S, APPR inadvertently armed instead of LOC (setting mental expectation of G/S capture), no glidepath to capture so aircraft continued in descent until 1sec(?) before impact when A/P is disengaged.

I've seen it happen several times in the past 13 years on the A320. Capture from above with no floor on the descent. Easy to fix if you catch it in time.

Speed of Sound 18th Aug 2013 23:18


Surely a pilot should have information as to all of the charted obstacles relating to their approach?
Yes, but we are also talking about living creatures here.

They grow, they stop growing, bits fall off them they are bigger when they have leaves and smaller when they lose them.

They grow no more than a few feet a year so there is no point in chasing exact heights when you are not supposed to come within 100 feet of them in the first place?

tubby linton 18th Aug 2013 23:25

A Squared I would fly the approach with my head looking out. Having seen a previous link to a video the undershoot are is dark and featurelss and I would be concentrating on trying to build a visual picture of where I was.
This crew really had nothing to help them. The terrain doesn't appear to have had any warning lights lights on it, the DME was showing a meaningless distance to the threshold, there were no approach lights only some very basic runway lights and the papis. The approach may have not been in the kit so it would be LOC and V/S without any electronic vertical guidance.
The area around and to the south of the runway appears to be well lit but they had the localizer for lateral guidance. Were they hanging on to the autopilot whilst they were trying to find the threshold in amongst all of the lights? The call on the CVR "Runway in Sight" may possibly confirm this.
The minimum altitude to use V/S is 160ft agl on approach.

A Squared 18th Aug 2013 23:25


Originally Posted by Tubby Linton
Having cleared the undershoot area why has the airport not provided approach lighting?

Non-precision approach lighting extends up to 1500 ft form the threshold. That seems like it would be of limited utility to someone hitting trees 4 times that distance from the runway.



Standard minima as used in Europe does not allow the LOC approach to be used at night.
Really? A LOC approach with a PAPI or other VSGI would be N/A at night?

A Squared 18th Aug 2013 23:30


I would fly the approach with my head looking out.
That wasn't the question. The question is: how would you fly the procedure *differently* if you knew the obstacle was 921 ft rather than 915 ft.

I guess that I've flown too many non-precision approaches over completely unlighted terrain to runways with nothing more than edge lights and a VASI to view this as recklessly dangerous.


All times are GMT. The time now is 18:27.


Copyright © 2018 MH Sub I, LLC dba Internet Brands. All rights reserved. Use of this site indicates your consent to the Terms of Use.