Well ..., in these THEORETICAL scenarios we are working with, it COULD've been avoided in many different ways.

1) -If the crew would've made very sure the flaps/slats were down.
2) -If the crew would've tested the TOWS system prior to take off.
3) -If the TOWS system wouldn't have failed.
4) -If the crew or the engineers would've figured out that a seemingly unrelated issue with the probe's heater potentially also made TOWS inop.
5) -If the weight/air/wind conditions would've been different.
6) -If a more skillfull handling of the situation on the air would've ocurred.

Etc, etc.

All those, of course, are very big "ifs" and none of them actually "guarantees" that the accident could've been avoided. Except, perhaps, number 1.

After all, the LAPA flight took off clean and the TOWS did sound ...

And a couple of flights took off clean, w/o TOWS, and they still made it alive ...

So there is no single unavoidable cause for the accident. Just a main and very important factor: the flaps/slats weren't out. This big problem means that everything else has to play in your favor if the accident is to be avoided. And that day, the stars weren't aligned, I'm afraid. :rolleyes:

Justme69, i read it (i read all this thread from the very begining, but thanks anyway). And more and more evidencies about human factor (not only pilots, but also technicians and ground workers...) :ugh:

UPS! Same time posting, just couple seconds later! :cool:

Can I dramatise this IF'S a bit?

1) If the crew would read checklist properly?
2) If Spanair would not missed boeing recommendation?
3) -
4) If the engineers would ask WHY before HOW?
4a) If ground worker wouldn't hesistate to ask WHY collision lights are ON?
5) -
6) -
etc etc
:ugh:

Sure, but it's a game that can be played both ways. I.e.

1) If the crew would've carried out procedures as trained.
2) If Boeing would've made sure all operators of its airplanes knew about recommendations after manufacturing.
2a) If air safety regulation bodies would've made mandatory those recommendations.
3) -
4) If engineering manuals gave more clear information to assist diagnosis ...
4a) If Boeing would've developed a knowledge database that could be downloaded on a laptop, type in: "RAT probe heater activates on the ground" and it would respond: check LEFT GRD CTRL RELAY C/B pushed in, check R2-5 relay operative, run additional TOWS check.
5) -
6) -

etc etc

BTW, just to make clear, I think the anticollision lights on the ground is only a sign of a distracted crew not completing parking checklist. It doesn't seem to me related to faults in the circuits, etc.

Also, ultimately, it's all in the hands of the crew. Whether TOWS checks are mandatory or not, if the crew fails to carry out the checklist (i.e. Northwest), then they could as well just skip the TOWS test.

Also, even if not mandatory, and knowing their own lifes are at stake, nothing prevents the crew from making a test more than once a day or once every change of pilots or long stop overs.

Lastly, even if the check is carried out and the TOWS works fine, the crew could still "ignore it" (i.e. LAPA accident). Or the TOWS could fail at any given moment in time, like anything electric. I.e. you test it with the flaps in ... you taxi for 20 minutes to line up ... TOWS fails around that moment ... you forget to deploy flaps ... you take/off.

Of course testing TOWS before each takeoff is the best "solution" as it decreases chances of clean TO a lot ... but still, the only true way to avoid this accident, is making sure the pilots just do not ever attempt to take off without the flaps deployed for sure (as sure as humanly possible).

If they fail to check this, they can just as well fail to test the TOWS, and then all recommendations (or requirements) by Boeing, Spanair or the FAA won't matter at all.

My PERSONAL take on this as an ignorant layman not related to the industry:
-Remind pilots/industry of the "killer items".
-Improve operational procedures so that they are harder to miss.
-Make Boeing build in some more redundancy on this system/more diagnosis (even a little "TOWS inop" light could help).
-Make a requirement that crews perform a TOWS test before each flight.
-Warn/train crews on take offs with wrong wing configurations.

It would also be nice if Boeing improved the engineering support of these older planes (if they are not doing it now) and trained technicians to better recognize TOWS failures.

So we may all be a little safer ... hopefully.

Justme,

exactly, and ground worker (they might decide to start over without refueling) is just another small flagpost God or Fate placed on the (grim) way... :sad:

PS and I belive that this

must be telegrafed to each DC-80 owner RIGHT NOW

PPS me too, but more than 20 yrs ago I used to work on Sukhoi ;)

strobe lights
 

Thanks, Justme69 and others for the enlightning info on the RAT-heater and TOWS relay issues.

Concerning the strobe lights, I thought the MD-80's strobes would only come on on change to "air mode" of the signal in the N/G strut, as discussed in earlier posts. I have been able to verify that this is indeed the case as a passenger during 4 SAS MD80/90 operated flights earlier this week: N/W lift-off: strobes on; weight on N/W on landing: strobes off.

So, the earlier question on which relay the strobes are is indeed an interesting one, if (as a mechanic on the ground seems to have declared after the accident) he saw the strobes of HFP working when it taxied away from the ramp for the fatal take-off.

Dutch Bru,
no, look post #1280.

Also there is no circuits which marked a/c lights on posted diagram.

This is UNOFFICIAL information, otherwise known as rumor: Sources leaking investigation details confirm that investigation is centered on failure of R2-5 relay and pretty much specifically discard (don't know the reasons) c/b GND CONT being pulled.

Here is a photo of part of Spanair's official declaration to the judge on the airplane's content:

http://img502.imageshack.us/img502/2...rcargo1ew0.gif

It basically just gives the known number of passengers, crew, PAX+cargo+luggage weight, empty plane's weight, weight w/o fuel, fuel's weight, take off weight, fuel expected to have been used in the flight (17.000 lb), expected weight upon landing (124.863 lb) (1 lb=453gr). Besides the 400kg of fresh fish cargo, there was also 1.095kg of textile products (mostly Timberland).

Landing gears on the plane were frequently serviced on the craft. On Aug 9th parts of the lighting system were also serviced, including "lights at the sides of the landing gear". "Inspection lights of both, left and right side, of the main gears found light bulbs out". "Portable flashlight in cabin, also with a blown bulb". "Light at the end of left wing: also blown bulb.".

Other maintenance actions during August: cockpit security door not able to open with key. First aid kid change requested. The list goes on and on ... Nothing seems relevant, though.

Good analysis justme69. I agree with you that if the FDR signal was received at rotation the L gnd control relay CB was in. R2-5 is in air mode when not energized so a failure of the relay would put it in air mode. One of the contacts goes to the position lights, obviously strobes in both position. If the strobes were on on the ground, not anticollision light, R2-5 wasn't energized. Also I had a nose oleo strut overinflated one night and along with strobe lights on the ground I could not reduce power on taxi below flight idle. I am sure they would have noticed that on taxi out. Another relay controls flight idle off the same circuit breaker.

DFDR.......
 

justme69,

Many thanks for your many enlightening posts!:D

I agree with you 100% in your reasoning regarding the (possible) failure of relay R2-5 being a factor in this regrettable accident.
I also is convinced that relay R2-212 was working.

I flew the DC-9/MD80 series aircraft for many years - although it is 16 years since I flew the types, my memory is still more or less intact - I think!

So, therefore this comment/clarification:

You wrote in post #1937:

"It's the one ultimately feeding data flight recorder 31-31, which I'm guessing is the one recording the change of air/ground logic state (therefore, R2-212 was getting the 115VAC, sensed the oleo switch and changed the state of 31-31, so it was also working."

It appears to me that you think(?) that the DFDR starts operating when an airborne condition is sensed by the L oleo switch (and relay R2-212).
This is not so.
The only thing that happens to the DFDR is the reception of a signal indicating an airborne condition.
The DFDR starts operating when the parking brake is released and continues to operate until parking brake is set again - in order to have data whenever the aircraft is moving on the ground and in the air.
At least it was so on the DC-9.
On the MD80 I am a bit uncertain - I think that it started to operate when fuel was selected to on in the start sequence of the first engine - somebody correct me!


The CVR (23-71) is operative as long as there is electrical power on the aircraft.
The CVR gets a signal when airborne via R oleo switch and relay R2-08.

BRGDS
grebllaw123d

We forgot to set T/O flap once,a long time ago full pax, and from an intersection.Rotated,obviously weren't going anywhere.If it hadn't been for the FO's clear and anxious communication I often wonder how it would have turned out.Not much time to play with!

BRGDS,

I was just going to clarify that. I believe the FDR starts functioning on first engine start and the FDR contact on R2-5 is a signal saying the oleo strut is extended. Otherwise aborted take offs wouldn't be recorded.

Strobe lights....
 

bubbers44,

If you look at the diagram in post #1936 you will see that the position lights/strobe lights are controlled via relay R2-308 and NOT via relay R2-5, which is suspected to be a factor in this accident.

I also think that it was clarified(?) many posts ago, that it was the anti-collision lights (and not the strobe lights) that was observed to be in ON position, while the RAT fault was dealt with.

Brgds
grebllaw123d

bubbers, there is no R2-5 contacts as far as i can see, which goes to FDR. Did you mean R2-212?

BTW it's interesting which circuit is feeded from R2-5 24-22 contacts.

Raredata
 

So how did you save the day ?

Pilots were never going to work this all out..it needed engineering input.A GO team comprises specialists in many differing fields.I think Pichu17 was obviously an engineer judging by his posts early on but the help from Wodrick and the schematic from SPA83 have been invaluable.

You are correct. I was thinking of what the CB controlled and typing without having the electrical schematic to reference. A lot of people were calling the strobe lights anticollision lights in this thread. Obviously only the strobes are controlled by the oleo switch so would be the only ones relevant on the ground as to gnd/air sensing. If anticollision lights were on during the return it was because they forgot the switch on.

grebllaw123d, thanks, i see...

Strobe lights - AC cross-tie
 

I went back into the 1200's series of posts and found that there is indeed no agreement about which exterior lights were on as observed by the ground-technican when the A/C was still on the ground prior to t/o: either the anti-collision lights or the strobes on the wingtips. The first are just put on or off from the cockpit, but the strobes will only start working on N/G lift-off (provided that the cockpit switch is on BOTH (pos+strobes).

If the R 2-5 relay installed were to be of a certain type, it has already been subject to Boeing Alert SBs and FAA AD's in the early 2000's because of unreliability issues with potentially serious consequences (E/B fire or in case of engine failire not connecting certain circuits linked to the generator of the failed enginewith the generator of the still working engine, i.e. the essential function of the AC cross-tie). The latter makes me still wonder about the earlier reported witness statements indicating engine failure and what is stated in the draft report, namely:
1.the apparent excessive length of the runway covered by HFP before rotation,
2.the apparent fact that both engines were producing thrust during take-off run and for the remainder of the flight and
3.the statement that the first visual inspection of the engines didn’t show pre-crash anomalies.

One of the issues of a non-functioning AC cross-tie relay in case of engine failure seems to be that the electrical circuits fed from the generator of that engine may cease to work, freezing e.g. the EPR reading of that engine in the cockpit and not giving any master caution warnings in the cockpit either, making it very difficult for the pilots to notice any malfunction with the affected engine (as it apparently did in a DC-9-32 accident in 1994 in Canada, as reported on in TSB A94C0034).

AC cross-tie on the ground
 

grebllaw123d

The report on the DC-9-32 1994 accident Canada (TSB A94C0034) does not seem to preclude AC cross-tie activation on the ground. On the contrary, AC cross-tie activation would have been essential for correct EPR readings in the cockpit during the take-off roll, giving earlier clues to the f/c that in fact they had an engine failure, the continuation of the functioning of FDR and CVR, as well as the functioning off anti-skid during the subsequent rto

Looking back a few pages.........
The fault which caused the return to stand seems to be air/ground sensing failure and not probe heat failure because the probe heat was working. Shouldn't have been working on the ground though.
In that case you could say that the MEL should not have been applied for probe heat failure and that the defect should have been investigated further.
I agree with an earlier post that the FIM (fault isolation manual) should caution about possible other effects associated with the failure observed but having seen how often the FIM isn't used I fear it wouldn't have helped.

ground/flight shift
 

Many ppruners still seem to go out from a ground/flight shift fault. However, in the leaked draft CAIAC report it said that the FDR recorded the shift from ground to flight when the NLG lifted off on rotation...

And a question to the guys in the know: how is a RAT probe heat fault to be 'isolated'? is there a switch on the IP? or can it only be done by pulling the relevant CB, with the known consequences...

borghha, post #1944

Nah, I always had it clear the DFR was working way before the airplane was in "air mode". I had guessed upon release of parking brakes. Incidentally, the DFR seems to have data from many, many flights before this one, so maybe other diagnosis are possible (i.e. did the switch to "fly mode" recorded by it ever failed, etc).

It's just that we don't know for sure what the CIAIAC means by "a sensor in the front wheels changed state to air mode". We can only guess they mean what we think they mean, as discussed before (i.e. R2-212-D).

But good reasoning.

I also agree that a "possitive, reverse logic" TOWS is in call in new airplane designs.

You push the handles and a voice comes up: "TAKE OFF CONFIG OK".

If you don't hear anything: TOWS are not operative, so you are NOGO.

If you hear: "BEEP - SLATS - BEEP - FLAPS" ... you know what you should do.

I also agree that, technically, MEL was probably not well applied in this case.

Although most engineers would've done the same thing, MEL probably only stated that the airplane can fly with inop RAT probe heater in good weather. It didn't say that a WORKING HEATER could just be disconnected. Also, at somepoint, somewhere, I read that the technician (from memory, but I noticed the caveat) "isolated the problem to the RAT probe heater .... more specifically to the circuit that controlled when it was on ..." So, perhaps, the technician DID know that the R2-5 or related circuitry was the one producing the problem ... and I don't know why he could've fail to see that it would also affect other non-MEL items.

Of course, if the problem is isolated to ONLY the heater being permanently on for "no reason", then it can be disconected and be done with it.

But that requires better diagnosis.

It's not a RAT probe's heater failure. The RAT probe heater was working just fine (pbbly). It's something else's failure (i.e. R2-5 relay), that is making the heater do something "ahead of time" (i.e. on the ground instead of in the air-only).

AC X-tie on ground
 

Dutch Bru, kulverstukas,

I just cancelled my post concerning X-tie.

Reason: I am not sure if the information given is correct.

grebllaw123d

"And a question to the guys in the know: how is a RAT probe heat fault to be 'isolated'? is there a switch on the IP? or can it only be done by pulling the relevant CB, with the known consequences..."

Need to see more circuit diagram..........Item like that would normally have it's own protection. If it went short circuit you wouldn't want it taking out all the probe heating when the fuse/cb popped.

But as far as this is concerned the probe heat was working it seems.

RAT probe has it's own cb see my post at #1944 it is overall controlled by the Right stall vane and static port heater switching then via air/ground switching. The only prfactical way to isolate the RAT probe is to pull and cap the cb. Physical disconnection is possible but would take a lot more time.

Guys and girls,
Thank you all for the inputs and thoughts.
But,

The RAT heating working on the ground should NOT have anything (alone) to do with the cause of the accident.
OK, they returned because it was heating on the ground.
They MEL-ed it, maybe the wrong way, because the heater was not inoperative! It was working just fine!
Question should be; did maintenance do the correct procedure?

All of this has nothing to do with a flapless (slats?) take-of! (besides the TOWS) but i don´t think you can blame MD/BOEING/SPANAIR-procedures/MAINTENANCE for pilots trying to get airborne not configured as briefed.

I know, a lot of things will come together and will have some kind of a influence of the whole accident.

For me the question remains: Why did they try to get airborne without flaps without having the proper speeds for that? Rush? Overconfident?

El Pais article 21/9
 

Excellent article, it has it all: Civil Aviation approved Spanair SOP without the Boeing recomendation, Mapjet (incident Lanzarote) informed CAIAC of their conclusions and recomendations in January this year, still no CAIAC report on this incident etc.

Interesting detail: (ARE) because the cb's are being pulled very often by maintenance personnel to put the a/c in air logic on the ground, the white ring got dirty and it being in a remote area of the panel, it was easy for pilots to overlook that it was not put back...

The article cites Mapjet who claim that they got away with their flapless takeoff because of the higher power of the MD83, lower TOW, head wind and crew action (deploy flap)

I m sure a lively public debate will start now in Spain...

R2-5 probably failed in air mode. That would cause RAT heater and take off warning to both fail in air mode.












r

A final horror scenario..
 

I really hope I'm wrong about thisone...

Looking at the video of the final stages of the accident, it has been
puzzling me how far and fast the acf plunged through the terrain.

We know that;
- The pilots engaged the reversers shortly after impact
- The #1 reverser got deployed
- The #2 reverser was stowed/inop
- The reverser buckets/mechanisms detached from the engines shortly
after impact.

Lets say the pilots engaged the reverser(s) shortly after impact in an
attempt to slow the acf down. Then the #1 reverser bucket gets
ripped off shortly after.
- Wouldn't this engine now produce FORWARD thrust?

With regard to the #2 reverser, it was stowed as we know. However,
what would happen if the pilots (in panic) tried to reverse thisone?
Would this #2 engine also produce FORWARD thrust?

The pilots would offcourse not know the reverser buckets were missing
and have no idea they instead of slowing the acf, added thrust!

The obvious reason for speculating is offcourse that this
might have had a significant impact on the tragic outcome of this accident.
If this was the case, they were doomed to disaster.. .

XPM

What about the flap alarm 2 days before the crash ?
 

I did not find discussed in the thread the statement (allegedly based on first-hand evidence) that the maintenance log has an entry, 2 days before the crash, reading "Sistema auto slat falla cuando se extienden los slats" (auto slat failure upon slat extension); a failure that according to the article recurred on subsequent flights.
El sistema de despegue falló dos días antes de la tragedia / EL MUNDO

What exactly is this "auto slat failure" warning?

TIA - a puzzled engineer

That's terrible, XPMorten. I also hope you're wrong, and that the turbine with the blocked reverse had some system that somehow impeded that perverse behaviour...

But you are probably, terribly right, I fear...

XXXavier

No. When you pull up the thrust reverse lever, the throttle is blocked by an interlock mechanism until the reverser is deployed. If the reverser is locked out nothing will happen.

TWT

Grabbed a notch of flap after some serious ar#e twitching.It's not something I'm very proud of and still gives me shivers when I think about it.You asked so there you have an honest answer.

These accidents could be prevented by haveing a third crew memeber on the flightdeck,he is called a flight engineer.

TU-154
 

In the mid-90's I flew Balkan Airlines (Tu-154) from Harare (Zimbabwe) to the UK stopping 4 times. Plane completely full of pax, high altitude runway, though fortunately very long. Plane rotated but did not take off. Nose was lowered, engine noise increased noticeably and we continued for several seconds more before second rotation and liftoff, right at the end of the runway. Most anxious moment I have ever had as a passenger. I wonder now if flaps/slats were incorrectly set, but I am not sure of this type. Anyone know more about this Tupolev type and configuration?

raredata

Kudos to you for bringing that to pprune. That could easily have been the end of you and your pax.

The flaps on my a/c are so slow that your 'recovery' would not have time to work.

I've learnt from this thread.

I always triple check we've got flap out now before pushing the levers up.:ok:

Lots of unwanted failures occur in a crash most of which lead to very high decelleration forces, rather than the other way arround. Once you hit the ground hard enough to break the airplane you no longer consider the consequences as in the causal chain.

Raredata
 

Appreciate your candour.Can happen to anyone.