Spanair accident at Madrid
Join Date: Jun 2009
Location: US
Posts: 497
Likes: 0
Received 0 Likes
on
0 Posts
Yes the pilots should have set the flaps for take off but they didn't because the mechanics disabled the take off warning system when they pulled the circuit breaker for the TAT. They forgot and had no back up. Both screwed up. We have learned from this.
Join Date: Mar 2009
Location: Feriton
Posts: 73
Likes: 0
Received 0 Likes
on
0 Posts
Investigators believe the faulty external temperature gauge may have been caused by a mechanical fault that also affected other parts of the aircraft.
Join Date: May 2004
Location: 3.5 from TD
Age: 47
Posts: 1,042
Likes: 0
Received 0 Likes
on
0 Posts
Yes the pilots should have set the flaps for take off but they didn't because the mechanics disabled the take off warning system when they pulled the circuit breaker for the TAT. They forgot and had no back up. Both screwed up. We have learned from this.
Nothing makes me go slower than someone who is rushing or says "we need to get going".
Planes can't be built to prevent all human mistakes. There will always be a weakness in every system. But we can prevent commercial factors from putting pressure on the humans which we rely on to avoid those system weaknesses.
Join Date: Aug 2002
Location: Europe
Posts: 4
Likes: 0
Received 0 Likes
on
0 Posts
Commercial factors--->pressure on humans !!!
"Planes can't be built to prevent all human mistakes. There will always be a weakness in every system. But we can prevent commercial factors from putting pressure on the humans which we rely on to avoid those system weaknesses."
"we can prevent commercial factors from putting pressure on the humans " This holy sentence should be engraved in each cockpit, in each Airline Boardroom, in each Airline CEO office, including all LowCo.
Excellently worded indeed.
"we can prevent commercial factors from putting pressure on the humans " This holy sentence should be engraved in each cockpit, in each Airline Boardroom, in each Airline CEO office, including all LowCo.
Excellently worded indeed.
Join Date: Aug 2008
Location: Spain
Age: 82
Posts: 490
Likes: 0
Received 0 Likes
on
0 Posts
Las Provincias has reported today tha the judge in charge of the investigation considers that he now has sufficient evidence to produce parameters for a flight simulation of the accident. He is also to examining the MD80 incident at Lanzarote, as metioned earlier in this thread, which took off under almost identical conditions and flew off successfully. His job is to apportion blame and, between the lines, it seems that he wishes to determine whether the crew of the Madrid aicraft could have flown out of the stall. He has said that he will pronounce by November (this year, that is!).
Join Date: Jun 2009
Location: US
Posts: 497
Likes: 0
Received 0 Likes
on
0 Posts
The difference between making it and not making it on an unintentional flaps up take off is the initial rotation rate and initial deck angle. Most any aircraft will take off with flaps up if the runway is long enough and clean stall attitude isn't exceeded providing tires don't blow first. A friend of mine did one in a 737 out of LAX many years ago and made it after initial stick shaker with no problem. If they look at the initial deck angles of both flights the FDR will tell the story very quickly.
Last edited by p51guy; 18th Aug 2010 at 14:23. Reason: error
Pegase Driver
Join Date: May 1997
Location: Europe
Age: 74
Posts: 3,689
Likes: 0
Received 0 Likes
on
0 Posts
Absolutely . By coincidence I am working on a serie of accidents regarding a perticular type of high perforfance ultra-light .
On this type ( or better said this perticular wing profile aimed at attaining high speeds during cruise ) the rotation speed without flaps is much higher than with flaps 1. If you unitentionally attempt to take off without flaps at the "normal" usual speed. you will take off due to ground effect, then if you keep the nose up as you normally do, , the a/c will stall at around 10-15 ft , normally sharply with one wing down, the result being always fatal for the occupants.
On a reported incident ( unitentional falpless take off ) on this perticular type, accident was prevented, when the pilot regognised the pre-stall condition and lowered the nose to level above the runway , until more speed was gained. Obvious but not that simple.
Regognising a pre stall condition will be the focus of training on this perticular type, but you have 2 -3 seconds max to play with. I am not sure how much time the Spanair crew had on the MD80 , and if they received a similar training to detect pre-stall conditions, or if indeed a MD80 do "warns" before it stalls.
On this type ( or better said this perticular wing profile aimed at attaining high speeds during cruise ) the rotation speed without flaps is much higher than with flaps 1. If you unitentionally attempt to take off without flaps at the "normal" usual speed. you will take off due to ground effect, then if you keep the nose up as you normally do, , the a/c will stall at around 10-15 ft , normally sharply with one wing down, the result being always fatal for the occupants.
On a reported incident ( unitentional falpless take off ) on this perticular type, accident was prevented, when the pilot regognised the pre-stall condition and lowered the nose to level above the runway , until more speed was gained. Obvious but not that simple.
Regognising a pre stall condition will be the focus of training on this perticular type, but you have 2 -3 seconds max to play with. I am not sure how much time the Spanair crew had on the MD80 , and if they received a similar training to detect pre-stall conditions, or if indeed a MD80 do "warns" before it stalls.
For Smilin_Ed: from about thirty pages ago, the Saberliner is T-39, not T-34. One of each has crashed in the past year (CNATRA training flights) with fatalities. Crew (IP) errors certain in the case of the T-34, near New Orleans. Have not heard from my old friends on the T-39 loss.
For ATC watcher: Your point on stall risks at takeoff and training seems a valid area for improvement.
From the transcript of cockpit conversations, the CRM / Mission focus in that two man crew looks to have eroded before takeoff clearance, with the problem of maintenance events weighing heavily enough on the captain's mind (in re answering to the company) that there looks like a non-trivial "company cultural" (we used to call it Command Climate in the Navy) influence to this mishap.
Why?
If he was preoccupied with his record, or recent string of maintenance delays, I think a human factors analysis would point to contributing factors that led to attention not on checklists, and seven's "killer" lists as one takes position on the runway.
Make sense?
Based on the transcript, it appears the captain made a misdiagnosis of why he had an alert on or shortly after rotation, or lift off. (Engine failure versus stall?). I am not familiar with the cockpit, so I may have guessed that incorrectly. Can any MD-82 drivers explain the differing alerts/sounds so that I can understand better?
Is the common theme of the three points above all related to training, or am I blurring important distinctions?
It appears that the punitive/criminal and safety investigations were conjoined, which strikes me as a process foul. Maybe Spanish/EU law is different from the rules I am familiar with.
Comments on that? Enlightenment?
Read through the whole thread today ... and I am deeply saddened by the loss.
For ATC watcher: Your point on stall risks at takeoff and training seems a valid area for improvement.
From the transcript of cockpit conversations, the CRM / Mission focus in that two man crew looks to have eroded before takeoff clearance, with the problem of maintenance events weighing heavily enough on the captain's mind (in re answering to the company) that there looks like a non-trivial "company cultural" (we used to call it Command Climate in the Navy) influence to this mishap.
Why?
If he was preoccupied with his record, or recent string of maintenance delays, I think a human factors analysis would point to contributing factors that led to attention not on checklists, and seven's "killer" lists as one takes position on the runway.
Make sense?
Based on the transcript, it appears the captain made a misdiagnosis of why he had an alert on or shortly after rotation, or lift off. (Engine failure versus stall?). I am not familiar with the cockpit, so I may have guessed that incorrectly. Can any MD-82 drivers explain the differing alerts/sounds so that I can understand better?
Is the common theme of the three points above all related to training, or am I blurring important distinctions?
It appears that the punitive/criminal and safety investigations were conjoined, which strikes me as a process foul. Maybe Spanish/EU law is different from the rules I am familiar with.
Comments on that? Enlightenment?
Read through the whole thread today ... and I am deeply saddened by the loss.
Join Date: Aug 2008
Location: Spain
Age: 82
Posts: 490
Likes: 0
Received 0 Likes
on
0 Posts
You are correct in surmising that the accident investigation procedure in Spain is quite different from that of the UK of USA. Accident investigations are undertaken by a judge who calls in expert witnesses to assist him or her in the case. The sole purpose of this judicial investigation is to allocate blame.
It is unusual in such inestigations in Spain to take human factors into account. During the long investigation of the Valencia Metro crash in which 43 people died in 2006, the judge determined that the crash had been caused by the excess speed of the train. The investigation was re-opened twice by the relatives of the deceased. At no time was the reason why the driver might have been driving at excess speed ever investigated.
It is unusual in such inestigations in Spain to take human factors into account. During the long investigation of the Valencia Metro crash in which 43 people died in 2006, the judge determined that the crash had been caused by the excess speed of the train. The investigation was re-opened twice by the relatives of the deceased. At no time was the reason why the driver might have been driving at excess speed ever investigated.
Join Date: Sep 2008
Location: Ottawa, Canada
Posts: 27
Likes: 0
Received 0 Likes
on
0 Posts
Virus in maintenance computer
Reports that failure correlation computer was infected with trojans and also Spanair took 24hrs to enter failures in to the system.
Spanish original:
El ordenador de Spanair que anotaba los fallos en los aviones tenía virus · ELPAÍS.com
Google translation:
Google Translate
The third failure of pitot heat on whilst on the ground should have triggered action.
Spanish original:
El ordenador de Spanair que anotaba los fallos en los aviones tenía virus · ELPAÍS.com
Google translation:
Google Translate
The third failure of pitot heat on whilst on the ground should have triggered action.
Join Date: Apr 2009
Location: bristol
Age: 49
Posts: 8
Likes: 0
Received 0 Likes
on
0 Posts
malware
Trojan-ridden warning system implicated in Spanair crash ? The Register
Its being reported spyware / malware might have been a possible cause towards the crash.
Its being reported spyware / malware might have been a possible cause towards the crash.
By John Leyden
Speculation. Based on what evidence, one wonders?
(I don't read Spanish well enough to parse the linked article, sorry to say. ) Allow me to be skeptical, given what the professionals at PPRUNE have discussed on this matter in the past year or two. Simpler malfunctions can lead to just as tragic a crash ...
Who are these independent crash investigators? I thought a judge in Spain was in charge. 
Do any of our MD pilots have a comment on this? Do the computers abort a takeoff on a three strikes rule? (I have a hard time believing this, but as I don't know the aircraft type in question ... )
From the Spanish article, did some IT techs fire up the mission computer on the airliner and discover it to have bad code in it during the forensic part of the investigation? Again, I don't read Spanish well enough to understand what that article was pointing to. Something has the judge's attention, at any rate.
This is an interesting claim, which would be best supported by some technical evidence. What sources is this journalist, Mr Leyden, relying on for his information, one wonders?
Malware may have been a contributory cause of a fatal Spanair crash that killed 154 people two years ago.
Spanair flight number JK 5022 crashed ... The airline's central computer which registered technical problems on planes was infected by Trojans at the time of the fatal crash and this resulted in a failure to raise an alarm over multiple problems with the plane, according to Spanish daily El Pais.
) Allow me to be skeptical, given what the professionals at PPRUNE have discussed on this matter in the past year or two. Simpler malfunctions can lead to just as tragic a crash ...
The plane took off with flaps and slats retracted, something that should in any case have been picked up by the pilots during pre-flight checks or triggered an internal warning on the plane. Neither happened, with tragic consequences, according to a report by independent crash investigators.
If the airlines' central computer was working properly a take-off after three warnings would not have been allowed, thereby averting the tragedy.
Do any of our MD pilots have a comment on this? Do the computers abort a takeoff on a three strikes rule? (I have a hard time believing this, but as I don't know the aircraft type in question ... )
Investigating judge Juan David Perez has ordered Spanair to supply data on the state of its systems at the time of the crash. An investigation commission is due to report on the case by December. ®
This is an interesting claim, which would be best supported by some technical evidence. What sources is this journalist, Mr Leyden, relying on for his information, one wonders?
Per Ardua ad Astraeus
Join Date: Mar 2000
Location: UK
Posts: 18,579
Likes: 0
Received 0 Likes
on
0 Posts
Lonewolf - I think you have slipped 'off target' here. I do not see reference to the a/c 'computer' in the article. The reference as I see it is to the maintenance computer at HQ which should have triggered action, possibly a grounding pending investigation, after 3 similar failures on a particular airframe. It is suggested the computer did not issue the action so the a/c continued in service. This does NOT mitigate the errors by the crew.
I am sure the initial investigation of the crash was undertaken by 'investigators' and I believe the judge is there to oversee the whole enquiry.
I am sure the initial investigation of the crash was undertaken by 'investigators' and I believe the judge is there to oversee the whole enquiry.
Join Date: May 2005
Location: uk
Posts: 573
Likes: 0
Received 0 Likes
on
0 Posts
sunnyjhon
This is not correct. The investigation of air accidents in Spain is always carried out in by the CIAIAC, equivalent to the US NTSB or the UK AAIB, with the same objectives, and in compliance with guidance from ICAO.
Relación de accidentes e incidentes - Investigación - CIAIAC - Órganos Colegiados - Ministerio de Fomento
Parallel to this sometimes you can find that on request of affected parties a court investigation/case in brought forward and this is led by a jugde. This investigation is not always as detailed as the one from the CIAIAC, but most times a jugde will wait until the official investigation by the CIAIAC is released to the public before giving their veredict.
You are correct in surmising that the accident investigation procedure in Spain is quite different from that of the UK of USA. Accident investigations are undertaken by a judge who calls in expert witnesses to assist him or her in the case. The sole purpose of this judicial investigation is to allocate blame.
Relación de accidentes e incidentes - Investigación - CIAIAC - Órganos Colegiados - Ministerio de Fomento
Parallel to this sometimes you can find that on request of affected parties a court investigation/case in brought forward and this is led by a jugde. This investigation is not always as detailed as the one from the CIAIAC, but most times a jugde will wait until the official investigation by the CIAIAC is released to the public before giving their veredict.
Join Date: Jun 2009
Location: US
Posts: 497
Likes: 0
Received 0 Likes
on
0 Posts
Trojans in the computer were not a factor. The R5 relay which senses nose gear strut extention failed causing the logic to say the aircraft was airborn. It inhibited the takeoff warning and also caused the TAT probe to be heated on the ground. This caused the return for maintenance and disabling the TAT also disabling the takeoff warning system. In their attempt to expedite their departure they failed to set the flaps. That is what happened. I had that happen to me one night. The strobe lights were flashing on taxi and the power was at flight idle. I knew why the strobes were flashing so hit the brakes to compress the nose strut and everything returned to normal. I did not realise that the takeoff warning was also inop. The MD80 had a lot of faulty switches when it first came out. We got the first batch and had a lot of false warnings.
Per Ardua ad Astraeus
Join Date: Mar 2000
Location: UK
Posts: 18,579
Likes: 0
Received 0 Likes
on
0 Posts
Trojans in the computer were not a factor.
Join Date: Jun 2009
Location: US
Posts: 497
Likes: 0
Received 0 Likes
on
0 Posts
I can't find the electrical schematic of how the TAT and TWS works but it was posted here hundreds of posts ago. Maintenance disabled the TAT but didn't realize the R5 relay also affected the take off warning system. More knowledge of the systems might have prevented this crash. The R5 relay plug in receptacle might have been corroded and removing it and testing it might have not proven anything. The corrosion in the receptical was probably the reason for the failure. Spending many years working on electronic devices I have found once you move a device everything changes. All the evidence goes away. Sometimes temperature, vibration and moisture can change how electronic things work. They could test it still in it's socket but after the jolt it got crashing it might have jiggled back to working condition. R5 is the most likely cause of the failure of the take off warning system. If someone can find the post with the schematic please post it.
Join Date: Jan 2009
Location: alameda
Posts: 1,053
Likes: 0
Received 0 Likes
on
0 Posts
Let us all just learn some things, OK?
1. Always check the killer items prior to taking the runway for takeoff...flaps/slats, trim, flight controls free and correct and all the stuff we wrote about months ago.
2. On any plane, you need to know certain symptoms. If something isn't right, maybe the plane thinks its in the air. I wonder where the landing gear over ride button was during taxi out.
3. if you aren't flying well on takeoff and all engines are working, and you might be stalling...put down some flaps/slats.
P51 guy is on the right track, whether it is a relay or not...but always check things, just like a doctor with a patient.
I love the DC9 series. I've seen the RAT/EPR gauge show the wrong readings at the gate and lo and behold, the plane thought it was in the sky for whatever reason.
Be careful...it is what you are paid to do.
1. Always check the killer items prior to taking the runway for takeoff...flaps/slats, trim, flight controls free and correct and all the stuff we wrote about months ago.
2. On any plane, you need to know certain symptoms. If something isn't right, maybe the plane thinks its in the air. I wonder where the landing gear over ride button was during taxi out.
3. if you aren't flying well on takeoff and all engines are working, and you might be stalling...put down some flaps/slats.
P51 guy is on the right track, whether it is a relay or not...but always check things, just like a doctor with a patient.
I love the DC9 series. I've seen the RAT/EPR gauge show the wrong readings at the gate and lo and behold, the plane thought it was in the sky for whatever reason.
Be careful...it is what you are paid to do.
p51guy;
You can use the search feature to perhaps find the graphic, using "schematic" as a key search word. Or you can read the first interim report here, and a short second interim report at the same site, both available in English.
I think it is short-sighted to state that the R2-5 relay was "at fault" even though we may know that the TOWS did not perform as designed. A similar claim regarding the Radio Altimeter is made for the Turkish B737 accident at AMS - it may have been a factor, but other, more serious issues were also involved. The logical conclusion to such a proposition is to "focus on MD80 R2-5 relays to ensure their contacts are clean and that they're fastened securely", etc etc, when often these things are far more complicated and causal factors more deeply hidden than first appears.
Checklist interruption, already discussed above, is one such factor. A standard SOP for any interruption during the use of memorized checklists for example, is to start them over again from the beginning. The only exception to this is when using a mechanical checklist where open items may be clearly seen and not merely remembered as "not cleared".
PJ2
You can use the search feature to perhaps find the graphic, using "schematic" as a key search word. Or you can read the first interim report here, and a short second interim report at the same site, both available in English.
I think it is short-sighted to state that the R2-5 relay was "at fault" even though we may know that the TOWS did not perform as designed. A similar claim regarding the Radio Altimeter is made for the Turkish B737 accident at AMS - it may have been a factor, but other, more serious issues were also involved. The logical conclusion to such a proposition is to "focus on MD80 R2-5 relays to ensure their contacts are clean and that they're fastened securely", etc etc, when often these things are far more complicated and causal factors more deeply hidden than first appears.
Checklist interruption, already discussed above, is one such factor. A standard SOP for any interruption during the use of memorized checklists for example, is to start them over again from the beginning. The only exception to this is when using a mechanical checklist where open items may be clearly seen and not merely remembered as "not cleared".
PJ2