PPRuNe Forums - View Single Post - Spanair accident at Madrid
View Single Post
Old 29th Aug 2009, 08:44
  #2559 (permalink)  
justme69
 
Join Date: Aug 2008
Location: Canary Islands, Spain
Posts: 240
Likes: 0
Received 0 Likes on 0 Posts
I generally agree with the assesements. Copilot was pbbly the PHYSICAL "prime responsable" of both, the most important actions that made them get into the situation and the most important actions that ended up not allowing them to get out of it.

In that sense, the copilot perhaps has a larger percentage of the "blame". But, as we all know, copilots can be a bit "rookie" and the captain is the final responsible for just about anything going on in the cockpit, and thus, the final culprit, and in this case he helped very (very) little, so he pbbly deserves to get many of those "guilt" points back.

Personally, I give the maintance crew (in Madrid as well as in Barcelona the day before), a slightly (say a couple of percentage points) more "shame" than 5%, but who am I to say? It's just a personal opinion.

The Lanzarote case, further disccussed previously in this thread, like here in page 107 http://www.pprune.org/rumours-news/3...ml#post4440133 , is also due its CIAIAC investigation report soon. The draft is already made and will be finalised anytime between now and the next 5 months or so.

In it, it becomes clear how the most important part in these types of situations is the crew. By extension, also their training, selection and cockpit management and safety culture.

The crew in Austrian Mapjet case of Lanzarote flight to Barcelona, had a 60 years old experienced captain (21000 hours, 7700 in the type) paired with a 31 yo copilot (2180, 2000 in the type) from different nacionalities and training backgrounds (Argentinian and Venezuelan), both rested and holding appropiate training courses and certifications, in the cabin. They had been flying together for about 2 weeks. Mapjet selection and additional training etc wasn't up to par around the time, though, as wasn't their crew subcontractor.

The procedures in the Mapjet case, DID call for TOWS verification check before each engine start, unlike Spanair's obsolete recommended procedure of "once a day" or whenever pilots entered a new cockpit for the first time or had been away from it for "too long". This check was not carried out by Mapjet's crew none of the two times the airplane flew that day (in the flight of the incident and the previous flight of the day from Madrid to Lanzarote), in spite of being required to do so by the SOP both times.

The TOWS in Mapjet case were proven in full working order, except that they had been rendered effectively inoperative by a maintenance technician pulling the circuit breaker K33 in the panel behind the pilot's seat during a routine strobe lights check the night before and forgetting to push it back. This rendered the whole Left Ground Control circuit (part of the air/ground sensor circuit) to effectively remain in the fixed "air mode" logic state, which included the TOWS. In Spanair's case, the TOWS lack of action seems it was most likely caused by an intermittent electrical fault, most likely caused by a malfunctioning single relay (R2-5).

I guess the circunstances leading to the crew forgetting to deploy the slats/flaps are pretty universal. Very lax procedures and checklists operation and radio distractions (in this case, asking ATC for some clarification of nearby traffic priority). They also had your usual conversations about manual/flex/auto/etc takeoffs decissions. They also had the additional workload of trying to deal with the "alarms and warnings" that resulted from the "faulty" ground/air logic.

Some different lessons learned in Mapjet case, prompted changes to the maintance procedures, with Boeing recently issuing new Check Cards for maintenance to clear up the difference between daily/every 3 days checks of external lights being different from checks of strobe lights, which were prompting maintenance facilities all around the world to routinely pull the K33 circuit breaker every 3 days, or daily in many cases, causing a non-intended overuse. Also, the recommendation of placing flags and collars on breakers was added to check cards whenever a breaker is pulled to reduce the chances of them being inadvertently left open.

In this case, i.e, the crew DID carry a check of the breaker panel, but because the K33 white collar was dirty from so much use, the visual inspection wasn't enough to make its (wrong) possition noticed by the crew.

And, of course, there is a reminding for crews to not forget to carefully check for open breakers as part of their checks when they enter a cabin, as required by procedures for the type in all operators.

The pulling of the K33 c/b, which practically left ~a dozen systems in air mode logic while still on the ground, gave the Mapjet crew an additional 5 o 6 alarms and warning lights which, in Mapjet's case, were also mostly ignored (they were each attempted to be "reset" by the crew and that was pretty much it). In Spanair's case, the fault resulted in very minor symptoms for maintenance or pilots to figure it out: the TOWS wouldn't work, which is not obvious unless a TOWS check is carried, and the RAT probe heater is activated in the ground, which was detected in that case, but unfortunately the maintenance actions didn't solve the matter, effectively just bypassing the problem rather than fixing it. Pretty much no other tell-tale signs.

The breaking of the Left Ground logic circuit in the type, on the other hand, can cause many signs such as:

3 AHRS lights on in Basic mode
No Flight Director FD, left or right
STALL IND FAILURE light ON
EPR LIMIT/TARGET flag
Amber NO AUTOLAND light in both Flight Mode Annunciators
RAT probe heater turns on, indicating excessive External Air temperature readings
TOWS would fail an operative test
AUTO SPOILER FAIL light goes Amber in the overhead pane
Cooling fan off (equipment ventilation, like radio, etc)

On the other side, the crew handling of the actual stall situation was better in Mapjet's case, where the engines were quickly firewalled, the attitude maintained low, the wing drops controlled (as per FDR, 63º right, 60º left) and the plane controlled until it reached ~200kt and started flying normally once again. Needless to say, flaps/slats retracted is NOT an approved configuration for take off in the MD-8x series and they were lucky they didn't find tall obstacles in their way.

Meterological circunstances were also better in Lanzarote than Madrid's, with ~15kt of front wind (instead of Spanair's ~7kt of tail wind). Temperature was 23º, sea level, and Mapjet's MD83 was carrying less weight (140.000lbs, 64ton) than Madrid, which was at ~97% of its capacity. The MD83 was also equipped with more powerful engines than Madrid's MD82.

The crew did retract the landing gear before full "normal" flying conditions were reached which, it seems, may not have been the best course of action as it temporarily offers worse aerodynamic performance while the gear's doors are being closed at a critical time, and perhaps the preferred method of action for the type is to leave the gear's configuration "as is" until sufficient lift is achived to get out of the stall warning entirely while being able to climb at the needed rate. I'm not sure about this, so please check with someone who knows how it would be best to proceed in your type with the gears in case you ever get caught in a similar situation (stall with landing gears down and no possibility of landing on that terrain, like while flying over a city, i.e.).

Last edited by justme69; 30th Aug 2009 at 22:23.
justme69 is offline