Spanair accident at Madrid

[philipat]

Wow/tocw

NOD:

Sorry - cannot recall it, so either not read it, or dismissed in my mind as not really relevant, or too MD-80/DC-9 specific for me...

As Follows:

Just thinking out loud and to ask for clarification from the experts. If the A/C thought it was flying because of a faulty WOW switch, would the TOCW still operate? If not, then the C/B issue is possibly a red herring?

It seems hard to believe however that such a simple problem could lead to such a disaster and there there is no redundancy inbuilt?

[Finrider]

Flapless T/O

I tried timing and measuring the ground roll to 150 kts both flapless and with T/O flap/slat (we take-off in mid flap/full slat, which is also one of our landing configurations and is actually quite high drag - theory is to minimise ground roll and rely on reasonable excess power, and option to jettison in the event of engine failure, for initial climb) in the Tornado sim last week. The result support NoD's statement - there was negligible difference. Admittedly not an airliner, but operating using the same laws of physics!

[NigelOnDraft]

Re

Just thinking out loud and to ask for clarification from the experts. If the A/C thought it was flying because of a faulty WOW switch, would the TOCW still operate? If not, then the C/B issue is possibly a red herring?

It seems hard to believe however that such a simple problem could lead to such a disaster and there there is no redundancy inbuilt?

I did reply in fact

Errrr.... If the Flaps were not set for takeoff, I am afraid there it is probably SOPs / Human Factors that led to the loss of the aircraft.

Things like TOWS are backstops that ideally should work, but it is difficult to attribute the (primary) cause of the accident to the TOWS not working. IMHO. A bit like blaming a faulty airbag for a road accident fatility... it would be better to avoid the accicent in the first place.

In short, if the takeoff was Flap/Slatless and the TOCW was not working, the accident was not "caused by" the TOCW, IMHO. It might be a "contributory cause" but if you have to reject a takeoff because the TOCW stops you, there should be a form of inquriy into just this... because the TOCW likely saved a lot of lives, and the rest of the "system" had failed.

Admittedly not an airliner, but operating using the same laws of physics!

Not too sure about that - an airliner can get above 20,000'

NoD

[Finrider]

Very true...but our wings would probably stay attached for longer at 600kts, 100ft!

[matkat]

Sorry if this has already been asked but let us assume that the TOW was inop would it not be in the T/O checklist to check the flap position indicator? the reason I ask is because some years ago I was the flight mech aboard an American trans air L1011 out of Philadelphia the check were down to the line and the check of the posn indicator was called and the flaps were set at 0 deg. we discovered that the hyd lock-out had activated and the flaps did not deploy even though the handle had been correctly positioned.

[ppppilot]

@ justme69

Hmmm ... now Spanair declares that, because both pilots left the cockpit for several hours, they were suppossed to check the TOWS alarms again for this flight.

And they said they did, according to Spanair. But they did it before the first return to gate.

I am sorry but I have never heard of that. As far as I know, the only TOWS check performed was at the first flight BCN-MAD.

@sevenstrokeroll

the RAT probe being heated on the ground meant that the plane ''thought'' it was in the air. Pulling the RAT heater CB didn't do anything to the ground control relay etc. THE PLANE ALREADY THOUGHT IT WAS IN THE AIR and another CB didn't need to be pulled...

That is what I believe the more relevant to the accident.
The famous P40 breaker at Detroit was pulled out due to "unknown reasons"
The same cause wings1011 is talking about at the Lanzarote's "almost accident" and I believe something has to see at Madrid. Seems to me the P40 "the hand that rocks the cradle" is still ignored by the authorities.

[HarryMann]

Please also NB that a stick shaker on departure is an "interesting event". History might teach one to ignore the stick shaker and just fly the aircraft - see TWA Tristar @ JFK and even a recent VS747 incident. At a critical phase of flight, a stick shaker could be very distracting - and a good chance spurious. As above, you can only assume you are taking off with the correct performance and configuration.

History might also teach one to trust it, it's a system designed for a purpose... and at least one 'All Souls Lost' accident resulted from just that, ignoring it and even fighting the resulting 50lb stick push... Trident/LHR/Staines reservoir

Unless it is common knowledge specific to an aircraft type and situation that spurious s/s action can occur, aircraft mfrs and certificating authorities would take a very dim view of ignoring it. More so on 'T' tailed aircraft.

In this case, we could propose the following, admittedly hypothetical (as we don't know crew control inputs yet for sure) and assuming the most +ve outcome imaginable...

Stick-shaker noticed > Immediate reduction of stick back-pressure > Reduces AoA > Returns better roll-control > Prevents (2nd) bad roll to starboard > Aircraft stays more nearly on runway centreline > Crew have more options due less panic (compared to a/c diverging off-heading into unknown territory)

>> Possibly, whatever the outcome >> more people survive!

Because if the aircraft does prove to be correctly configured, and airspeed is correct, an immediate 'trial' reduction in AOA would not necessarily produce ground contact - and would increase margin speed over stick-shaker sensor speed, whether erroneous or not.

The first left-wing drop here was of course, a very strong indication that aircraft had been over-rotated (for speed/configuration) and needed flying 'gingerly' to retain control...

[NigelOnDraft]

HarryMann

History might also teach one to trust it, it's a system designed for a purpose...

It's a balance. Read the NTSB L1011 report, it was fairly harsh on the crew for believing it and clearly stated (as above) they should have realised it was false.

Stick-shaker noticed > Immediate reduction of stick back-pressure > Reduces AoA >

First Officer was pretty well slated for doing exactly that. First Officer was not permitted to initiate an abort, but by relaxing on the back pressure he was deemed to have effectively aborted the takeoff after liftoff

Because if the aircraft does prove to be correctly configured, and airspeed is correct, an immediate 'trial' reduction in AOA would not necessarily produce ground contact

It just about did in their case - certainly led to the illusion the aircraft would not fly... and the resulting overrun and hull loss.

It's a difficult call, as the report states... but as quoted above, there is no training in stall recoveries on liftoff, and very difficult to devise a drill that covers all situations...

NoD

[sevenstrokeroll]

DEAR NIGEL ON DRAFT:

My airline trained for takeoff departure stalls to recognize the stall and apply firewall power / flaps 15

my airline is a very large airline and you may e mail me for more details.

I still recall performing this as part of my checkride in the sim.

[bert0503]

Facts: zero flaps/slats take off attempt. Cnfig Warning did not work. Question: can FDR determine if the flap/slat lever was moved out of the "up" position? In other words has the possibility been ruled out already that the crew in fact could have set the flaps/slats correctly but that the flaps failed to operate followed by forgetting to check the FPI. In our company (I fly Avro) it is customary to check the correct setting by calling: FLAPS (NUMBER) (points to T/O datacard), SET (points to flap lever position) AND CHECKED (points to FPI).

[tyro]

Maybe slat deployment should be made fully automatic below some defined speed and/or above a certain AoA. The Tiger Moths I've flown have such slats over which the pilot has no control. I believe this design feature was included to protect student pilots from themselves.

[wings1011]

selected or not

If the slats were in the up position you can with the most certainly say the lever were in up position.This aircraft slat is controlled by a single drum dual actuator mechanism mounted in the aft center wing box section.This unit the is linked to each slat section with VERY thick control cables that via a simple solution with wires and pulleys from both sides actually pulling the the slat either in or out and they are all linked together unlike say a 737. Thats why no separate slat section indicator is needed like the 737.If one is going out ALL is going out if not a catastriophic failure accured to the cotrolcable system. That is highly highly unlikely though, basicly non existant if looking on how its constructed. The good thing of an old wires and pulley aircraft is that you can be very sure of the inputs done from cockpit actually is gonna happen in the back no matter what.It weighs but sure as hell.
And that the wires from the leaver to the mid section of the aircraft should be broken off without any notice in feeling of flap handle is even that highly unlikly-Just basicly never happens

Regards

Wings 1011

[ppppilot]

@wings1011
Someone told me that the MD in air mode doesn't have steering, neumatic from the APU neither TO mode. Is that correct?

[justme69]

The judge in charge of the judiciary investigation will hear personally the CVR recordings tomorrow at CIAIAC's facilities.

He is not going to call any more witnesses or experts until the commissions preliminary report is finalized and becomes public, probably monday.

[RatherBeFlying]

It is natural for pilots to try to think up a way to salvage a bad situation.

One suggestion in this accident thread is that perhaps lowering the nose to build airspeed in ground effect and/or select flaps might just have saved the day. Simulator trials by the investigation may well address that tactic, but achievable climb gradient is a major factor. Hitting an obstacle at high speed is a bad thing whether in the ground or in the air.

The bigger question is how a presumably well trained and conscientious crew missed this one.

First world air carriers do extremely well. That record perhaps breeds an expectation that you and your crew mates don't make mistakes while in fact you and they are people like anybody else who can screw up at unpredictable moments.


Errare humanum est never goes away.

[bsieker]

Originally Posted by ppppilot

Someone told me that the MD in air mode doesn't have steering, neumatic from the APU neither TO mode. Is that correct?

In this post, and others. pichu17 talked about some aspects of air/ground sensing on the MD-80.

It appears that different functions that change behaviour between air and ground are controlled by individual relays triggered by nosegear (un)compression. So if one or more of them fail, there is no global "in the air"-condition, but some functions behave in their air mode, and others in their ground mode.

So, given that, as per above post,

- ground sensing for both RAT probe heating and Take-off configuration warning is controlled by the same relay,
- Takeoff-configuration warning is inhibited in the air,
- and IF failure-in-air-mode of this relay was the reason for RAT probe heating on the ground
- and the relay as such was not fixed, but only the circuit breaker for RAT probe heating was pulled,

it follows that Takeoff configuration warning was inhibited.

Now if, for whatever reason, the flight crew did not select slats for take-off (others have calculated that flaps were probably not vital, a typical V2 being above slats-only stall speed), they would have been unaware of it during take-off.


Bernd

[sevenstrokeroll]

tiger Moth slats

other planes including some earlier jets had automatic slats. north american rockwell sabreliner 40 and 60...F86, A4

many planes had this nice feature...airspeed increased, slats retracted...based on springs!

ps

nigel, boeing flaps are really slow...douglas slats come out really fast. perhaps one reason boeing bought douglas. 2 seconds seems long for slats in this type.

[NigelOnDraft]

My airline trained for takeoff departure stalls to recognize the stall and apply firewall power / flaps 15

my airline is a very large airline and you may e mail me for more details.

I still recall performing this as part of my checkride in the sim.

Good - if it forms part of the airline's / State's training it can only help the crews.

Problem as I said above is where Airline / Manufacturer / State positively discourage some actions.

Out of interest, which type do you refer to? After all, if you planned a F20 takeoff (?) presumably you would not want to select F15?

The problem comes if, say, SOP is a F15 takeoff, then any "drill" that calls for F15 makes a presumption the crew may have omitted to select Flaps. In this case, if lack of Flaps/Slats was a factor, what if a F15 takeoff was planned? Would they even think to "select F15" since in their minds F15 was already set... etc.

I am unaware, in commercial aviation, of too many drills etc. designed around "coping" with serious crew / SOP errors. The "system" seems designed to prevent these in the first place - which has patently failed in the past, and may have again?

NoD

[sevenstrokeroll]

Nigel

I take it you don't fly douglas. to the best of my knowledge , flaps 20 takeoff can't be done. I don't think there is even a flaps 20 position.

I even checked and stalls in the clean configuration below FL250 may use slats for recovery.

[wings1011]

To PPPilot

Sorry for not really understand you questions, but im trying to answer them as I understood them. Yes the APU bleed is not for flight use only for ground use-APU electrical power can be utilized in flight and also as an extra source of power (electrical) if needed. And yes the steering does not work in flight (if you were refering to nose wheel steering) but that is pure mechanical protected nothing electrical. That is for preventing the nose wheels turning inside the bay when up selected and then get stuck by structure inside.When the strut oleo springs out the planet gear is removed from gearsprockets making the steering cylinders move without bringing the gear strut with it (if I.e rudderinputs are made in flight, or steering tiller turned) -It all cames back to normal when gear is compressed again.Simple solution.
The T.O mode what you are asking about im not sure what you mean though ?? only thing I could think of were the T.O mode of the TRP (thrust rating pnl) and that is just working fine in Airmode.
Maybe You were suggesting how they could start the engines if the aircraft were in partly FLT mode ?? well simple as the APU bleed valve does not go under the same ground/fight relay as the other stuff mentioned earlier in this thread.

Regards

Wings 1011