Spanair accident at Madrid

[agusaleale]

May be possible that the pilots somehow got confidents about the TOWS when testing it that morning, and later the same system turned against them when not working.
I mean, if the confidence in the system may have produced that prior to the take off, the checklist has been done in a relaxed way.

[GearDown&Locked]

for justme69

Is there any thruth in this information?

YouTube - Conversacion pilotos avion Spanair accidente Barajas

[justme69]

GEarDown&Locked:

No, those are false statements on that youtube video, blantently made up by some news program in Argentina even before the CVR recordings had been extracted by the labs.

Bears no resemblance whatsoever with the actual conversation. It has been discussed way earlier in the thread, some 50 or so pages back. The CVR recording is reportely of very bad sound quality in the moments previous to the impact. It did record all the way to the end, though, and a substantial partial transcript has been made part of the judiciary investigation.

The reporter, Guillermo Lobos, of Argentina TV channel "Todo Noticias" (TN, owned by Canal 13 of the Clarin media group), admited having fabricated the whole thing, which he shamelessly presented as "The last words of the pilots in the cockpit extracted from the black box" just a few days after the accident.

[justme69]

This is likely the last state of the preliminary CIAIAC report to be published any time now:

-The flaps probably weren't out.
-The investigation still hasn't been able to proof the position of the handles to activate flaps and slats.
-The R2-5 relay is still suspected of malfunction and the culprit that the TOWS didn't activate the alarm to signal that the flaps/slats weren't out. But its performance state still hasn't been established.
-From Sep 22nd til 26th the QAR (Quick Access Recorder) disc data was to be extracted at the manufacturer's facility in the UK.
-The manuals and procedures used to "repair" the RAT probe heater are being investigated.

You can see scans of the letter sent to the judge a couple of weeks ago about the advances in the investigation and main changes from the Sept 4th draft to the september 23rd revised draft:

http://img66.imageshack.us/img66/379...ardlastta5.png
http://img66.imageshack.us/img66/432...rdlast2ue3.png

[Finn47]

Just came across this article:

Pilot error not to blame for Madrid air crash, say investigators - Telegraph

[Dutch Bru]

"flaps/slats set"

I'm wondering indeed what exactly the prelim report will say about the earlier reported "flaps/slats ok" or "....set", allegedly retrieved from the CVR (see LUALBA post 1478 and jutme69 posts 1487 and 1714).

If that would be confirmed the key issue is whether the pilot not only "just" called this, but at the same time really/consciencly had verified that what he called was in fact in accordance with the settings in the cockpit. In other words was it really a "check"-call or a call without the actual "check".

[justme69]

A large segment of the press refuses to speak of pilot error, but they have no problem whatsoever suggesting system malfunctions.

The preliminary report doesn't point fingers and doesn't mention either pilot error or system malfunction, other than the TOWS.

The "facts" to the best of their knowledge are:
-The pilots followed (reasonably) the checklist. They seem to have called flaps and slats as "ok". It is not known if they were sure, as it wouldn't be the first case found of a "rotten response".
-But the flaps were NOT out at that time, as per the DFR. It is not known why yet, but the procedures to determine the possitions of the handle are being studied right now.
-The TOWS alarm didn't sound, and it should've if the flaps weren't out. It is not known why it failed, but the workability state of the R2-5 relay is being suspected and investigated right now.

Now, think for a second. If the pilots called the flaps and slats as "ok" and, LET'S ASSUME FOR A SECOND that they were not "ok" (either system malfunction or forgetting to deploy them) ... how do you think it could've happened that, when the pilots "lowered the handle" to the first detent:

-Both, the left and right slat didn't go out to mid possition in the front of the wings.

-That the lighted indicator that signals that the slats are in takeoff possition lighted up, while the rest of the slats indicators (autoslat fail, etc) remained off.

... and when they set the handle to the second detent:

-Both, the left and right flaps didn't deploy the 11º commanded on the back of the wings.

-That the ANALOG dial that shows the position the flaps are "positively" in (and independently of the possition of the handle), showed 11º for each wing and yet the flaps remained at 0º (as shown in the DFR).

That would imply that at least 8 independent mechanisms (4 for each wing), all very simple and extremely reliable, decided to fail at the same time (slats actuators, slats sensors, flaps actuators, flaps sensor) in such an incredible way that they signaled "slats ok for takeoff" and "flaps at 11º" for both wings while they were actually not.

And remember, some 45 minutes earlier, the flaps had come down 11º upon the first attempt to take off.

Besides being obvious that it was probably the pilots forgetting to deploy them (as in Northwest, LAPA, MAP and other "proven" cases), even if the flaps and slats failed by some "impossible" reason, the crew would still be at fault for not checking the indicators to make sure the devices were in the correct position before accelerating for takeoff.

The press continues, even as today, to twist and turn the facts to fit their "show".

Today I saw headlines in the like of "The preliminary report shows that electrical malfunction of a relay needed to deploy takeoff systems is the cause of the accident"

[justme69]

The preliminary report has been made available:

http://www.fomento.es/NR/rdonlyres/9...preliminar.pdf

[justme69]

The main "news" are:

-Last Minute Changes were included in the cargo sheet, adding 555lb to the weight (5190lb PAX, 27655 cargo). ATOW reflected 141863lb.

-DFDR recorded RAT probe to 104ºC prior to first line up. After that, RAT didn't show over 30ºC.

-Some of the checklist procedures can be figured from the CVR (prestart and before start) and the taxi checklist. During taxi, they contacted ground control and received takeoff permission. Before take off, they run the take off inminent checklist.

-210º 5knots wind, >28ºC.

-DFDR has reliability problems on the recording of one of the two flight guidance computers (DFGC nº2), the origin of which is unknown. The DFRD doesn't use the DFGC to read the RAT temperature or the flaps angle setting.

-The DFGC selected by the crew was nº2 (there goes the investigator's luck).

-Part of the data on the QAR has been extracted and is being analyzed.

-3 right wing and 2 left wing flaps' actuators have been identified. 4 had lost hydraulic pressure and could be freely actuated. The fifth was heavily damaged from the fire and is stuck.

-2 slats control cilinders were found, as well as the mechanical elements for 3 of the sections. All fire damaged and consistent with RETRACTED positions.


The report concludes that ongoing investigations are being centered in finding out the state of wing configuration and cockpit controls as well as further analysis of the black boxes, etc.

And that's it. If it weren't for the leaked draft, we would know even less than we do know. So much for the wait-and-do-nothing until the official information is released.

You can (perhaps) get a google translated version of the report here:

Translated version of http://www.fomento.es/NR/rdonlyres/97C78A32-1428-4BC4-B70C-18DA3474E92C/38073/2008_032_A_preliminar.pdf

[grebllaw123d]

Digital Flight Guidance Computers 1/2

I have read the "translated" preliminary report - mostly "old news"!

The report mentions Digital Flight Guidance Computers (DFGC) 1/2.

A short clarification for the benefit of non MD pilots:

The MD series aircraft have 2 DFGCs - one is in operation, while the other one is in a standby state.
The chanceover is done via a selector on the glareshield panel (below the autopilot ON switch).

In my airline the standard procedure for operation of the 2 computers is:

Left pilot as handling pilot: DFGC 1 (selector left)
Right pilot as handling pilot: DFGC 2 (selector right)

Spanair procedure? (Acc. the report DFGC 2 was operative)

According the report the selected DFGC sends info about slats position to the flight recorder - I wonder why the system works in this way.

grebllaw123d

[justme69]

It looks like Spanair did basically the same.

Since the copilot was the one in control, they selected flight computer 2.

Unfortunately, the Data Flight Recorder had problems reading the data everytime that computer was selected.

As a result, most of the flight recorder data of interest for this accident is not readily available.

No idea why the slats' position goes through the flight computer and then to the data flight recorder.

But as a result, only things like the flaps position and the RAT temperature can be obtained from the FDR "as is", since they went directly to the recorder without going through the computer first.

But the slats position can not be conclusively obtained from the DFR in this situation, since it didn't properly record anything coming from computer nº2. It worked fine with computer nº1, but that was not the one in use during this accident.

[Jonty]

Sorry if a repost:
BBC NEWS | World | Europe | Wing flaps caused Madrid crash

[wileydog3]

southernboy

One things I would query is this line, "If they have a license to fly a MD-82 I'm sure they have been trained on how to do it right." We all know it isn't as simple as that. If the atmosphere - safety culture - in which you work is not robust & non standard methods are accepted then it builds a basis for cutting corners. I have worked in airlines where the paperwork was legal but it was a box ticking culture and good solid training was minimal in order to keep the show on the road & costs under control. We all had licences to fly our types but always felt under trained. You are clear that was not the case here & I must accept that as I know nada de nada about the company. However there's world of difference between good training, adequate training & just legal requirements, so I remain curious why such well qualified experienced pilots did - or did not - what they did, there are always reasons in the background & they are ones we really need to understand.

most of us are pretty good at responding to the honking horn and blinking red light and most of us aviators have done the tasks thousands of times without consequence even if we botched the task a bit.

and you can 'swiss cheese' your barriers, train over and over again, create policies and procedures, revise checklists, encourage everyone from the PIC to the plane spotter at the fence to point out errors and... still accidents are going to happen.

sorry to not be typing with capitals as needed by as of right now, I am working with a rebuilt shoulder from a recent bicycle crash which destroyed the shoulder. Now riding a bicycle is a fairly easy task and for someone who has been riding bikes for decades, one could say I was an experienced bicyclist. I have done many 50-75 mile bicycle rides without incident. But then recently I recently turned my head slightly to watch 3 riders go by as I approached the 50 mile point on a 66 mile ride and the next thing I knew I was going over the handlebars after hitting a pot hole large enough to swallow a tank.

One could say I was distracted and the obvious answer is to not be distracted by telling someone to not be distracted is 1) like telling someone they should advise you when they are asleep and 2) it is not realistic.

One could say ride more carefully and have better 'situational awareness'. I didn't see the pothole until I was almost in it. All this to say we make mistakes which are the exception to the rule and sometimes we incur significant consequence up to and including loss of life. or in my case, a new steel plate with eight to 10 screws holding together my reconstructed left shoulder.

[borghha]

CVR

We haven't learned much from the preliminary CIAIAC report that we didn't know already. I was particularly struck by the passage that relates to the CVR which says, and I translate quite literally now from the Spanish report:

"the CVR recordings allow to identify some expressions relating to the pre start, taxi and imminent take off checklists"

'some' expressions? does this mean to say that the recordings are so bad that they become unusable, or do they mean that the checklists were only partly complied with?? That sentence is extremely vague and yet the CVR recordings could shed some light on the sequence of events...

Very disappointing.

harry (PPL, frequent SLF and professionnal conference interpreter)

[TripleBravo]

xolodenko,

I think I can imagine what you feel. I have been to a crash site where a friend of mine fell out of the sky including another person. When I arrived, there were just a crater and some really tiny pieces left, nothing more, nothing spectacular. That event will never reveal so much details like "Madrid", as a FDR was not on board - but exactly that creates an even more restless brain since years. I kept debating with people over and over who knew both and who knew the airplane that brought them to death, as my deepest wish was to just understand what could have been the cause(s).

Since then I am very much interested in valid thoughts about tragic accidents like the one discussed here, as I am eager to learn more about human factors, safety critical engineering design and the like, just in case it can tell me someting about my own flying. E. g. missed flaps you could have on almost every aircraft. I just try to get my bag of knowledge filled more than my bag of luck.

There are different ways to cope with deep impressions, some people keep talking, some shut up entirely. But I think it's ok if everyone tries his own way, so do I when I post here and speculate. This is entirely NO disrespect towards the dear souls missing, indeed perhaps just the opposite. We owe them to try our best to make sure it never happens again. First step would be to understand - and not to rely on just some people whose report is just as thin as in the Lanzarote incident. I for sure learned more about the latter on PPRuNe than from the official report.

BBB

[justme69]

'some' expressions? does this mean to say that the recordings are so bad that they become unusable, or do they mean that the checklists were only partly complied with?? That sentence is extremely vague and yet the CVR recordings could shed some light on the sequence of events...

It seems to be a bit of both.

The CVR is both, of extremely bad quality in the inmediate moments previous to the accident to the extend that most of it can only be made sense of using foreinsic methods (carried out by the police sound lab, and therefore not 100% conclusive) and the checklist (unofficially) seems not to have been done 100% according to regulation, but pretty close. (i.e. flaps and slats may have been reported as "ok" rather than as "11º, takeoff").

It seems, that because the audio can not be fully made out, it can also not be conclusive whether the checklists were done 100% right, 99% right, or what, but they (unofficially) seem to be "possibly not perfect, but reasonably good". In another words, the pilots probably THOUGHT they had gone through it and they probably THOUGHT they had the flaps and slats set.

The "almost exact" transcript of the CVR will probably be known (outside the authorities) in a couple of weeks. It will then shed some light on exactly how distracted the pilots could've been during the maneouver and how closely they followed the checklist.

The judge will spend the next days calling a couple dozen witnesses, 6 on the 17th, 5 on the 27th, 5 on the 31st, some of them for the second time, to ask specific questions and check their declarations against their first ones. He will concentrate on the technicians that did the "repairs", the previous failures of the aircraft and how it could've affected the TOWS.

He'll also call the flight attendant that partially witnesed the accident from the air, the airport ground coordinators, the supervisor of airplane cargo loading, the supervisors of the crew in service on that flight, etc.

The CIAIAC has informed that, as it was expected, they won't speak a word again officially in public about the accident for between 1.5 and 2 years.

Thankfully, there is a public judiciary investigation going on and I'm sure the judge will request updates and evaluations from the QAR data, the flaps handles, and other important data within the next months.

[sevenstrokeroll]

OK:

The posts have trickled down to a few.

Let us all resolve that this crash will at least TEACH us something:

Let every pilot go forth and tell other pilots:

1: Know your plane, inside and out...know the symptoms of problems, and each symptom is not OCAM'S Razor.

2. Do the checklists LIKE YOUR LIFE DEPENDS ON IT...Because it does.

3. ALWAYS CHECK THE KILLER ITEMS : flaps, trim, spoilers, and compass agrees with runway bearing/heading.

IF SOMETHING IS WRONG...STOP, taxi clear of the runway and get it right. Don't try to correct it on the fly as it were.

[larssnowpharter]

and you can 'swiss cheese' your barriers, train over and over again, create policies and procedures, revise checklists, encourage everyone from the PIC to the plane spotter at the fence to point out errors and... still accidents are going to happen.

I have posted here on this before. Allow me to repeat:

All accidents are avoidable (barring 'acts of God' or the enemy); if you think they are not you are on the way to becoming a statstic'.

Forget the 'swiss cheese' analogy. It is not acceptable. Get yourself some Cheddar!

[ChristiaanJ]

Originally Posted by larssnowpharter

Forget the 'swiss cheese' analogy. It is not acceptable.

Aren't you missing the point of the "swiss cheese" analogy? It merely describes that many accidents do not have a single cause, but are the result of a chain of events/causes. It doesn't say they're unavoidable, or indeed acceptable.

CJ

[larssnowpharter]

Aren't you missing the point of the "swiss cheese" analogy? It merely describes that many accidents do not have a single cause, but are the result of a chain of events/causes. It doesn't say they're unavoidable, or indeed acceptable.

Those who use the Swiss cheese anology are saying, in effect, that if you slice the cheese thinly and keep twisting each slice - sooner or later a series of holes will line up which will allow human or mechanical error to cause an accident.

It merely describes that many accidents do not have a single cause, but are the result of a chain of events/causes

Here you are absolutely correct. Well, nearly! Most accidents have more than one causal effect.

Most of my job involves advising people in a - non aviation word now, alas - on how to investigate accidents/incidents/events.

As the threads in Rumours and News demonstate, there are 2 phases to the on line investigation here:

1. The 'What'. By this I mean the answer to the question, 'What went wrong?' The engineers love this one as do the armchair theorists. Most civil accidents involving carriers we have good accident investigators who will give us this answer with a darn sight more evidence than we will get here.

2. Then we get to the 'Who'. Here we arrive at that - oh so fun - finger pointing stage trying to pin the blame on someone. 'They didn't do their checklist'. 'Why didn't the pax alert the flight deck that the slats were not deployed'. 'Did they rush the repair to get the aircraft away?' You will see it all on Rumours and News.

But they tend to miss on the third:

3. We very rarely get to the 'Why'. By this I mean the root cause or causes of a systemic nature. Was it a training issue? Was it too much pressure being put on the crew to 'press on' by the management? Was it a 'culture' (I could do a lot more on 'safety culture' but this isn't the place) of the business? Were there flaws in the recruiting process (believe me, there almost inevitably are)?

I repeat - and will keep on repeating - my original statement:

All accidents are avoidable (barring 'acts of God' or the enemy); if you think they are not you are on the way to becoming a statstic'