Jet goes down on its way to Medellin, Colombia

[DaveReidUK]

Originally Posted by lemme

I have stated my prime interest is to use ADS-B data as a tool for air-safety investigation.

Out of interest, can you think of any accident investigation where ADS-B has provided more of a clue to probable cause than the evidence provided by the FDR and CVR?

[lemme]

Originally Posted by safetypee

lemme, et al.
Most of the diagrams presented are too simple for this discussion.

agree. Your description is marvelous, thanks! Designing a four-engine airplane with loss of all engines is a fascinating challenge. The devil lives in the details. I dig deep to understand these issues in my blog.
Satcom Guru: LMI2933 LAMIA AVRO RJ85 Medellín Deadstick

Originally Posted by safetypee

The engine windmill characteristics involve a relatively fast N1 but slow N2, even to the point of stopping (170kts?).
Hydraulic and elect services require high N2. Normal operations use a flight idle stop.

I believe they established about 125 KIAS from 16,000 down. From your statement, that would not generate much windmill hydraulic power (if any).

Originally Posted by safetypee

Some of the calculated speeds appear very slow for what could have been a low or zero flap setting. Thus unless the flaps were selected before all power loss, it would be more consistent speculation to consider stall or near stall conditions.
From deep and failing memory, I recall that the stall warning was available on essential AC and possibly emergency DC.

I think that they configured for landing under engine power ahead of flameout. It does not look like a deep stall scenario, rather a steep, controlled descent. They had 11,000 feet to lose by the FAP in about 16 nm, and they started clean.
from FCOM: There are two electronic stall protection channels: channel 1 and channel 2. Either channel can provide stall warning and stall identification. With one channel failed, the working channel can provide complete stall protection. Each channel requires power from the ESS AC and the ESS DC busbars.

Originally Posted by safetypee

flynomore; fungus, yes it can, first hand experience of gauging problems, world wide; and corrosion (Columbia!). Also there were reports of possible fuel pump blocking and engines stopping - far east.

noted, I have one referenced.

[lemme]

Originally Posted by DaveReidUK

Out of interest, can you think of any accident investigation where ADS-B has provided more of a clue to probable cause than the evidence provided by the FDR and CVR?

No, and I can't imagine a case of where it ever would. But I will say that having ADS-B data from surrounding flights, and from preceding flights, is an added dimension. On EK521, for example, I could watch the tail-wind shear evolve on each successive plane on the approach. I could compare the Pence plane path in space to those ahead of it.

But what if we don't recover the recorders?
MH370. All we have are a handful of indirect position estimates.

Or what if it takes nearly two years to recover the recorders?
AF447 made a lot of hay over a handful of ACARS messages that implicated air data faults.

MS804 ADS-B data was catastrophic, and an ACARS message showed smoke. Six months later forensics are describing evidence of explosives.

We started flight data recording over 50 years ago with altitude, airspeed, heading, and vertical accelerations as a function of time. That is about where we are with ADS-B telemetry, and I foresee a chance to build out an acceptable data set.

[misd-agin]

As engines spool down you can try and move flight controls. HYD pressure on other jets is typically present with approx 5% N2. At approx 8% N2 it appears to have enough pressure to perhaps move the flight controls on the ground with no air loads.

[Chronus]

More I read of lemme`s post the more I begin to understand that his interest in this particular accident is academic. It appears to me that he is trying his level best to engage with pilots, rattle some inadvertently of course, in the process, in order perhaps to expand and advance his knowledge on his chosen field of endeavour. Namely ADS-B and Flight Data Monitoring as a tool for telemetry in aviation. Nothing new really, it has been going on for some time now. However I would suggest that this particular accident involves relatively few and fairly straight forward circumstances. At this juncture, known facts may be few but are overwhelmingly persuasive as to cause. I therefore cannot quite see how an all dancing singing FDM system with ADS-B could possibly have made a difference. Look at the Azores Glider event. A much larger aircraft , A330 and dead stick 65 miles out from 34,500 ft. at dead of night with nothing but ocean all around. What was the difference but the two guys up front, nothing else. In this particular case they were near enough on top of the rwy and what did they do, they stuck the nose right down and headed for the VOR. So what`s all the hydraulics and windmilling fan blades got to do with anything. For me the olde proverb says it all, thrust is a must , but lift is a gift. So no use just sitting there hammering at the top of the instrument panel insisting, demanding I must have thrust, look N1 says I do and all the time looking the gift horse in the mouth.

[ATC Watcher]

Lemme :
I think I too start to understand your motivation in doing this ( although you choose not to reply directly to my questions on my first post on this subject).

Yes ADS-B down-link parameters can be useful in the early stage of an investigation if you do not have other surveillance means which are independent like radar, and if CVR/FDRs are not recovered/usable or simply absent ( small a/c types)
However in most cases , and in this one in particular, there is multiple radar recording available, and FDR and CVR were recovered in good shapes.
FR24 provides some form of data, and one can use this to draw some conclusions before the Investigation team does its work. Nothing wrong with that.That is why we come to this forum .
What I object in your case is that you present this FR24 data, correlate it with info collected on Internet to present your version of the " facts " ahead of the preliminary report of the official Investigation team and fool people with such detailed and specialized info that most will believe that what you present are " facts".
Lot of persons died in this accident and for me at least, you should have respect and refrain from assuming or worse stating what the crew did or did not do because there you have no idea.
Finally a word of caution , never forget the " D" in ADS . It is only dependent of what it reads and what is manually inputted . BS in , BS out.

I know I am probably not going to convince you to call it a day and wait a year for the preliminary report, but you are not going to convince me that what you do is " scientific" either.

[MartinM]

Originally Posted by lemme

No, and I can't imagine a case of where it ever would. But I will say that having ADS-B data from surrounding flights, and from preceding flights, is an added dimension. On EK521, for example, I could watch the tail-wind shear evolve on each successive plane on the approach. I could compare the Pence plane path in space to those ahead of it.

But what if we don't recover the recorders?
MH370. All we have are a handful of indirect position estimates.

Or what if it takes nearly two years to recover the recorders?
AF447 made a lot of hay over a handful of ACARS messages that implicated air data faults.

MS804 ADS-B data was catastrophic, and an ACARS message showed smoke. Six months later forensics are describing evidence of explosives.

We started flight data recording over 50 years ago with altitude, airspeed, heading, and vertical accelerations as a function of time. That is about where we are with ADS-B telemetry, and I foresee a chance to build out an acceptable data set.

AF447 had very usefull ACARS messages from the start. No one believed them until the day of the report came out.

I don't remember that short haul aircrafts, like the RJ100 are zrasnmitting ACARS. I never heard of it. As such I. an be very wrong, i admit.

[DaveReidUK]

Some RJs do, for example the Brussels and Swiss RJ100s, but CityJet's don't, so it's very unlikely that the LAMIA did.

[Airbubba]

Originally Posted by lemme

I had responsibility for both flight data recording and communications while at Boeing, continuing most of my career on communications. One area we, as an industry, have been seeking is telemetry for use in both operational quality assurance and for air-safety investigation. Like it or not, the public has little tolerance for the glacial pace of aviation evolution, especially in the area of accident investigation.

Mode S transponders and ADS-B open up an new generation of analysis unrelated to air traffic control or collision avoidance.

Public companies like flightradar24.com have crowd-sourced a very comprehensive set of data and make this available in real-time. Not only can we examine the flight of interest, we can look at all the other flights too. This adds a whole new dimension of insight.

Our planes, even older ones like the RJ85, are spewing streams of data that can be harvested for forensic and a lot of other purposes. And much of it, like ACARS and ADS-B is sent totally in the clear with little authentication to prevent spoofing, interception or hacking. I can see where security is being added as an afterthought in some of these protocols. In the meantime we amateurs can track Air Force One and Tyson One going south for the holidays.

Originally Posted by lemme

I was lead engineer (control laws) for Boeing automatic flight controls, Thrust Management, for 757, 767, 747-400 and am very familiar with airplane systems. If you fly those planes, or 777 or 747-8, you are using thrust management, data link, flight data recording, and satcom systems I once led, or are directly based on developments I played a principal role in.

As supervisor for data link and satcom, I led the ability to for satellite voice and for ACARS/FANS to connect pilot and controller, to which I spend considerable time dealing with human factors and having to invent "comm messages" as a whole new category.

Many of us here have used those systems over the years, thanks for your efforts. Trying to do a user interface for messaging and ATC with the installed base of text screen user terminals and FMS boxes must have been a challenge.

Originally Posted by lemme

Having examined a lot of data throughout my career, I have found that it is far easier connect the dots when you know what happened.

And, here on PPRuNe, we have sometimes found the smoking gun in an accident investigation long before the report comes out a year or two later. Or, at least, some of us think we have.

In the 1997 SilkAir 185 crash a poster here correctly gave the probable cause within a few days. Was PPRuNe still an e-mail list back then? Or had Danny started the web page by then? I can't remember. For some of us, server crashes and lost passwords in the early days have given us an 'adjusted date of hire'.

Similarly, the likely cause of the 1999 Egyptair 990 crash was revealed here long before the official reports were published. I was assured by others on PPRuNe in those pre-911 (and pre-Germanwings) days that it couldn't be suicide because the captain was a Muslim.

A poster named Uncle Jay posted his early speculative analysis of the 2009 Colgan Air 3407 mishap as a stall-spin accident and nearly nailed it in my opinion. For some reason his post was summarily removed by the mods.

So, why not just wait for the accident report to come out? Why look at the publically available information and data trails and form opinions and have these sometimes heated discussions?

For some of us in the flying business the lessons from a mishap may have practical application long before the final report is published. A suspected battery fire, fuel exhaustion mishap, or long landing runway excursion causes us to think twice about how we would avoid or handle those undesirable situations as we operate aircraft in just another day at the office.

And sometimes the final report just doesn't past the smell test based on what we have learned and discussed here and elsewhere. The Egyptair 990 report issued by Egypt's ECAA concluded that Gamal El-Batouti did not crash the plane despite fairly persuasive published evidence from the wreckage and investigation. Similarly, the Indonesian NTSC was unable to find the cause of the Silk Air 185 crash. In both of these cases the NTSB concluded that the probable cause was intentional control inputs by one of the pilots.

As Professor Feynman famously observed, often we know much more than we can prove. And as President Reagan's favorite Russian proverb dictates, trust but verify.

[TowerDog]

Well spoken Bubba, agree with everything.
Egypt Air was pretty obvious, the F/O piece of was flying home to be fired, too many complaints about him, and he knew it.
Many of these accidents being discussed on PPRuNe seems to be spot on long before the report comes out.
Gulf Air Bus: Those guys did not know how to do instrument flying, they lost control of a perfectly good plane.
Same with the bus in Libya, they cocked it up and crashed a brand new airplane.
I remember those discussions and threads very well.

[lemme]

Originally Posted by ATC Watcher

Lemme :
The first question that come to my mind is why are you doing this? Is it just a hobby or is it professionally motivated, like selling a technology or a project/ idea ? or just wanting to show what people can do with FR24/ ADS ?

(apologies, off thread) This is an unpaid personal quest looking at options for telemetry (not system health monitoring). I am on pprune.org seeking help to answer questions (crowd sourcing). I have responded to questions I have studied thinking to be helpful, and that has brought some interesting new aspects to light. Challenging assumptions, looking at the data again, running down a dead-end...maybe we don't find something new or wrong, but sometimes we pick up a gem that becomes instrumental in unexpected ways.

I keep assembling more and more information in my analysis. I am sharing the blog because I think others might have similar questions and may find the information helpful. The fine line is to not represent it as a factual report, which of course, it is not. I cannot attest to the flightradar24.com data as factual, just for starters. But I am mindful of correlation in data sets too, so agree have to tread carefully. My stock is in the flightradar24.com ADS-B data, the BAE FCOM Vol 1, photos of the crash site, the approach transcript, Google Earth and Excel. I have great trust in the BAE marketing brochure and information relating to engine off glide matching from two different sources.

[jack11111]

Follow your passion

Lemme,

As you describe your background in aviation, it sounds well rounded and solid.
I don't believe you are trying to fool any of us here on the forum.
Follow your passion and worry not about the static in the air.

[ATC Watcher]

Ok last try :
Jack11111:

worry not about the static in the air.

I've been called a lot of things in my days but this one is new

Airbubba :

In the 1997 SilkAir 185 crash a poster here correctly gave the probable cause within a few days

I was there also and remember well. You correctly say "probable cause " because it still is the the theory that best fit the event but no hard evidence of it was ever produced as far as I know,.
If you look at the CRJ/Sweden report being discussed right now, the final report gives a totally different answer as what was discussed here , and also perhaps add another possibility at those accidents of 20 years ago.
Similarly the discussion on Asiana 777/SFO . Those "idiots" that could not fly a visual, remember ? but the final NTSB report gives a very different picture that rather aim at the manufacturer and the airline than to the " idiots "pilots up front. Sometimes we are our worst enemy here.

Towerdog :

Egypt Air was pretty obvious, the F/O piece of [...]
Many of these accidents being discussed on PPRuNe seems to be spot on long before the report comes out.
Gulf Air Bus: Those guys did not know how to do instrument flying, they lost control of a perfectly good plane.
Same with the bus in Libya, they cocked it up and crashed a brand new airplane.

Thanks for proving my point . No further comment .

Finally Lemme :
Thanks for replying to my questions . I get it now . It is a fee space here, and you have followers, Jut be careful not to hurt anybody in the process because you are swimming in murky waters and I may add , at night . .

I cannot attest to the flightradar24.com data as factual

and :

I have great trust in the BAE marketing brochure

Good luck. see you back when the report is out.

[portmanteau]

Preliminary report will be published on December 22 says Director of Colombia Civav.

[aterpster]

I'm curious whether they had the required nav charts on board.

[n5296s]

Those "idiots" that could not fly a visual, remember ? but the final NTSB report gives a very different picture that rather aim at the manufacturer and the airline than to the " idiots "pilots up front.

I hadn't seen the final report for this, thanks for pointing to it. But I don't think it supports what you say here. Both the probable cause and 3 out of 5 contributing factors relate directly to the specific performance of the crew, not to the manufacturer or the airline. Yes, the 777 autothrottle design seems highly perverse, and the airline clearly didn't encourage the kind of in-flight experience that would have equipped the crew to treat this approach as routine. But still, most of the reports fingers point at the crew.

Report conclusions below for those who don't want to wade through 129 pages. The Recommendations section is about training to prevent a recurrence, and therefore by nature says nothing about the specific crew - I'd imagine they won't be flying airliners again anyway so it would be irrelevant. Or if they are, they will have received a LOT of training.

The National Transportation Safety Board determines that the probable cause of this accident was the flight crew’s mismanagement of the airplane’s descent during the visual approach, the pilot flying’s unintended deactivation of automatic airspeed control, the flight crew’s inadequate monitoring of airspeed, and the flight crew’s delayed execution of a go-around after they became aware that the airplane was below acceptable glidepath and airspeed tolerances. Contributing to the accident were (1) the complexities of the autothrottle and autopilot flight director systems that were inadequately described in Boeing’s documentation and Asiana’s pilot training, which increased the likelihood of mode error; (2) the flight crew’s nonstandard communication and coordination regarding the use of the autothrottle and autopilot flight director systems; (3) the pilot flying’s inadequate training on the planning and executing of visual approaches; (4) the pilot monitoring/instructor pilot’s inadequate supervision of the pilot flying; and (5) flight crew fatigue, which likely degraded their performance.

[ATC Watcher]

n526s :

This accident report is extremely interesting , especially beyond the 130 first pages. If you have a bit a of time go through the Board members comments . Very interesting personal comments in there .
Especially that one from Bob Sumwalt ( page 136) : a small extract :

Contrary to what some may believe, this accident is not just another “pilot error” accident. Like most accidents, the causation of this accident is complex and involves the interaction of several elements of the system. It involves a set of circumstances that came together on this day to produce a tragic outcome.
I believe setting the stage for the crash was expectancy; the pilot flying expected the airplane to do something that it wasn’t designed to do. Specifically, he expected the autothrottle system to provide speed control for him, but unbeknownst to him, the system would not do so while in a HOLD mode.
This pilot was not the only one who didn’t understand this nuance of the automation,

The Comments from Korea are also worth looking at , but of course will be called biased , however this fact below is interesting , as they say it is in contradiction with FAA certifications requirements :

the B777 low-speed alert was a caution that did not require immediate pilot actions, and generated a master beeper that sounds identical to more than 60 other problems on the aircraft.

The devil is nearly always in the details ..and I would not be surprised a little bit if this is the same in this RJ case. Too much focus on "bad pilots" and fuel imho.

[H Peacock]

I don't dispute that is indeed what was stated, but the guys in the front had a duty to monitor their 777's performance. They got way too slow and didn't notice it = an error by the pilots! Great if we have a system to help us out and control the airspeed, but your job as an airframe-operator is to: Understand its systems fully; Monitor it!

Way too many accidents when, for whatever reason, the system doesn't do what is expected but the aircrew either don't notice or over-react! Inevitably lessons to be learned about the possible lack of understanding of the automation/monitoring, but you can't absolve the aircrew from not doing what is surely their primary role - flying the aeroplane.

Back on thread, unless the Lamia RJ suffered a totally unexpected and simultaneous loss of thrust on all 4 engines, I can't see how their subsequent reaction was anything other than woefully inadequate. And that's before there is any mention regarding their planning and fuel uplift.

[A0283]

Just wondering ... maps ...

Descending (fast) from 21,000ft through 19,000ft under highly stressful conditions ... Maps give Trans Level FL 190 and Trans Alt 18,000ft ... what altitude setting were they on ... (if not reported, then you would of course need CVR and FDR to answer that one)...

Comparing IAC maps with Jeppesen maps ref indication of ground contours/altitudes. IIRC in for example the Guam case there were comments on the contours/alts, the impact area not being properly indicated. In this case the Jeppesen appears to be superior to the IAC map. The Jeppesen contours say 8,000ft by the way.

There is a nasty isolated hump along the approach path around D5 IMDE. On the IAC map it says 8040ft (but you might overlook that). On the Jeppesen it says 8157ft (indicated more clearly, but i had to check if it said 6157 or 8157 because of the contour line).

So pilots who had been there before and pilots who looked at their maps, forgetting the 12,600/13,100 ft MSA, in an emergency would probably make sure they would stay well above 8,200ft.
Which triggers the question who was PF. The FO being reported talking to ATC, would suggest the Capt to be PF. How familiar was the PF with this specific airport. What kind of terrain map would he have had in mind.

IIRC the Lamia flight at some stage reported being 'visual with ground'. They were flying in the dark. The most likely visual might have been La Ceja. If you look at videos from the approach and aerial pictures, then La Ceja has quite typical contours. Wonder if having grating type contours of towns on the maps would be beneficial in this specific area.

Just wondering ...

[Lonewolf_50]

Originally Posted by ATC Watcher

This accident report is extremely interesting , especially beyond the 130 first pages.

If you are 35 knots slow on final, you fail your check ride.
If you are 30 knots slow on final, you fail your check ride
If you are 25 knots slow on final, you fail your check ride
If you are 20 knots slow on final, you fail your check ride
If you are 15 knots slow on final, you fail your check ride.

And you really fail your check ride if you get into those conditions on final and 1) you don't act to correct it and 2) you don't initiate a go around before you get to 35+ knots slow on final. That a training captain was in one of the seats is deeply disturbing from the PoV of the system and culture in that airline. I appreciate Mr Sumwalt's points on how the pilots ended up in a situation where they failed their check ride with a plane full of passengers, in terms of training and culture. There is certainly an element of habits that will hopefully be addressed in that company.


This accident, however, is from a significantly different organizational set up and culture. Asiana is a major airline with some depth. This company, LaMia, was small and has/had no depth. From an organizational aspect, they were more vulnerable to any problems in culture, training, and habits. Fewer opportunities for the 'system' to catch aberration.