PJ2,

You have all my respect and admiration for your wise words, technical knowledge and pointing out the morbid in some of our colleagues....

Marsh Hawk asked: how instrumental was the muddy field in increasing the survivability of this accident? It seems to me that the yielding dirt cushioned the vertical forces just enough to make this accident survivable for most

I remember some tests done by US Authorities years ago using actual airframes which determined that - contrary to what one might think - impact on soft ground results in higher G forces as the aircraft tends to dig in and stop; on hard surfaces the aircraft tends to slide. I don't think this is particularly relevent in this case, however, just giving an answer to a query.

Once again looking at the ADS-B Google Earth track plot,

assuming the 370 ft altitude correction mentioned in my previous post is correct, it appears the plane remained on the glideslope all the way to the second last data point at 9:25:53 which is 1.3 NM from the threshold. The ADS-B plot shows a height of 445 ft (75+370) at that point. This fits well with the approach chart descent profile which says 355 ft at 1 NM from the threshold.

Did the autopilot "pull" the plane to remain on the glideslope, while the power was insufficient, causing it to slow down until it stalled? In that case a very similar-looking scenario to BA038... ???

Assuming the above is correct, the next interesting question is then whether the autothrottle was engaged. FDR will answer that. Please correct if you find this theory impossible....

The roof was hacked by the fire and rescue team.

Early on there was TV coverage showing a rescuer on the roof attacking the fuselage with an axe. The shot was from the starboard side.

It seemed odd at the time that they had not tried entry through a side window; had there been sign of life it is perhaps more likely that a window would have been used for entry or exit.

Engines stalled on finals
 

A report in the Times of London - link below - states that one of the Netherlands aviation authorities suggests that both engines stalled on finals.

Engines stalled before Turkish Boeing plunged into Schiphol field - Times Online

It has happened before, and depending on the actual QNH it can make a significant difference in altitude reading. But in that case the aircraft wouldn't have followed the normal glide path before things went wrong, so probably irrelevant here.

Times item
 

That appears to be the product of a journalist who has failed to understand the distinction between aerodynamic and engine stalls. :suspect:

I'm only a PPL so can someone estimate what the landing speed should have been, roughly? Looking at those OpenATC screenshots I find it strange how they went from a normal -700ft/min descent at 09:24:23 to -1344ft/min and -1216ft/min at 09:24:38 and 09:24:53, respectively. They then recover this descent back to a normal rate after that, though with an increasing loss in speed. Am I right in thinking they were above the glidescope initially, then tried to reintercept it, which seems to be when the problems started. So maybe the question is why were they high initially, and did they just do a bad job of flying the rest of the approach?


Please read the thread. This was asked and answered earlier on.

Duck
Moderator

Basically, Vref should be around 120KIAS. Plus or minus depending on weight and flap setting.

I read the Times article by David Charter.
He doesn't seem to have done any research at all. He picked up the word stall as a possibility and made up the headline " Engines stall " :ugh::ugh:

Just remember when you read an article about a subject your not familiar with it's written by idiots like him, with even less knowledge. :rolleyes:

Why is no one discussing the possible recurrence of one of the malfunctions (such as master caution light) that occurred previously?

Do you just trust that it was repaired properly?

Quote:

---

Originally Posted by Pontious Navigator

The roof was hacked by the fire and rescue team.

---

The rescuer was hacking at the structure protruding through the crown skin probably in an attempt to release the flight deck door not to gain entry directly.
I don't recall if there is an external release for the dv window on the 737. I think you may have to cut in through the skin and release the window at the top.

As previously pointed out, ADS-B doesn't transmit airspeed at all - only groundspeed (and the GS values are near-real-time, not averaged)

It's also highly unlikely that the speed readouts were from the aircraft once it was on the ground. Mode S uses 1090MHz, which makes it even more line-of-sight than VHF - unless the enthusiast with the Mode S receiver was very close, they would only have received transmissions while the aircraft was airborne and above their radio horizon (as indeed the height readouts indicate).

there IS an external release for the R/H DV window. however with that much damage on the airframe it might have been blocked.
===
John R:
there might be not a lot of (actively flying) airliners that do not have any entries in the tech log within a week!

A bit of graph paper and google earth + open ATC =

4.2 D 150' above the glide correcting 159 gs.(wind was reported as 210/10 so lets use 8 head for airspeed that gives 167kt
3.6 D 30' above 156kt

2.9D on glide 153kt
2.4D on glide 137kt
1.8D on glide 126kt
1.4 D on glide 94kt
1.2D wrekage

with a gs of 86kts as the last known speed thats 11 seconds for the last 450'

This really looks like the jet captured the glide and decelerated until sometime after 1.4D where it stopped flying and hit the deck 1/4 of a mile later.

Visual illusion
 

Dear readers / collegues,

First of all I would like to apologize for not reading every posting here. I did read about 50% though.

I have 16 years of 737 experience, of which most as test-pilot and instructor.

I 've seen no reactions so far regarding possible visual illusion during cloud break. Low cloudbase at 700' and visibility of 4000m tempo 2500m can give a momentary impression of extreme low pitch. This is a scary sensation when you are at low altitude. An agressive reaction to this sensation could have been an agressive pitch-up, resulting in an accellerated stall. A pitch down movement would be normal after such a stall, this is per aeodynamic 'stable' design. As an instructor, you would have no options left to counteract such an act of your student....

@AstonMartin

Very sensible suggestion but unlikely. The last thing any pilot would on gaining visual contact from a stabilised approach is to punch the AP out and make a sudden pitch input. I'm not convinced.

I am now getting rather fed up with the number of 'experts' posting here. There are a great number of contributors who pose as ATPL holders whose theories and suggestions give away the fact they have no experience whatsoever in aviation.

Seriously considering not contributing here anymore.

Now the mods have a better grip on this thread, could I ask that posts which imply negligence by dead pilots be reviewed also? Rainboe's 'suggestion' that wing fuel 'mismanagement' may have caused the crash are, I think, offensive.

Some points on that process:-

a) It has been an 'approved' procedure at some airlines for a while on the 737NG
b) I was given to understand it had a 'no objection' from Boeing which is the right side of 'don't do it'.
c) We do not know if THY approved the procedure or not
d) There are several considerations in use:-
1) ZFM limits (for wing bending relief)
2) Competent crews ensure a sufficient div fuel in the wings on approach in case of centre tank feed issues.
3) Trim issues are insignificant
e) IF the flight return tanked from IST there was no option for 'ground transfer'
f) Several warnings would have to have been 'ignored' by the crew. and this is where I baulk at this suggestion without evidence
1) Wing tanks reducing below expected
2) Low fuel warnings (906 or 403kg per side depending on setting ) - still enough on approach to land.
g) I have done it many times and it is a 'no problem' techiqueAll the recorders will tell us whether there was 'negligence' or not, and we should not rush to speculate.

I am lost in the crud here on this thread, but has one possible cause been mentioned ie wind shear/downdraft? It looks extremely unlikely from the TAF but................

Aston martin,

Good and well put suggestion.

I would imagin though looking at the data on top of this page that a control input like that would have thrown them well above the glide at such a low altitude (very sensitive needles that close in) and it doesn't seem to show such above?

Any thoughts?

N

It seems that the report ACTUALLY suggests the engines were not delivering power- as would happen if the engines were at idle.

And of course, as soon as the word "Stall" appears in the lexicon, journalists relate it to there mini-metros and assume engine failure.

One of the most challenging things to do in a Boeing (and I would assume other makes) is do an Auto-pilot on GS capture from above.

SOPs almost always mean the Platform Altitude is set in the MCP. Descend in VNAV SPD or FLCH with the GS armed and reach the MCP altitude fiirst, the aircraft captures the alt and levels off, putting it FURTHER above the glide.

I've seen this many times, both Aircraft and Sim. Already hot, high and loaded, the pilot is suddenly faced with an aircraft not heading down, with TLs increasing thrust.

One (Pefectly valid) reaction is to disconnect AP and increase descent rate to capture the GS. In 737s, many operators SOPs require (and, indeed, Boeing reccomends) that AP off means AT off as well.

I'm a couple of years off the NG, but I seem to think that in the above senario, there would be no Autothrottle Wake Up if speed decayed. Is that correct?

If the engines are at flight idle-they are still delivering thrust.

WizofOz, I totally agree!
There would be no warning from the A/T (as it would be off).

Autopilot
 

Everybody seems convinced that the autopilot was engaged. Was it?
Personally, when I intend to do a manual landing, I fly a manual approach.

In those conditions, I think its fairly safe to say the autopilot was on-although of course one must wait for the report to confirm it.

No one here knows, yet.
I'd speculate, however, that in a manually flown approach the pilot would have noticed the speed decay earlier. Unless he was distracted of course.

Astommartin,
Where do you fly? Doing a manual approach into a busy airport (eg AMS) can really load the other guy up with MCP inputs to keep up with your flying. Virtually all my landings are manual. Most are done done following an automatic approach. It maintains SA in very busy airspace IMO.

What of a prolonged glide in icing conditions....with no anti-ice selected.
Conditions are visible moisture.. Thrust response would definitely be a problem if ice is present..

I dont buy windshear and I think Rainboe's fuel theory is crazy.Although Dani is technically correct about passengers not surviving true stalls,he is splitting hairs somewhat.This aircraft crashed as it approached the stall.Lemurian's post about dissipation of the energy forces in the crash sequence was very well thought out and perhaps explains why we have so many survivors.
Survivor testimony of engine acceleration in the final moments is very telling in what it implies and what it rules out.
The most plausible thing we have on the table right now is automation complacency/confusion(ie.AT status) brought about by a deadly mixture of distraction,haste and inexperience(training flight).I know that Schiphol controllers are excellent but they can be highly demanding as well.Sometimes their demands can lead to hurried approaches where callouts and careful monitoring are absent and where checklists are completed in the most critical phase(below 1000').May not be the case here but at this stage it should be considered.

Yes, but not much. CERTAINLY not enough to stop a landing-configured aircraft on a 3deg glide-slope from stalling.

If he was capturing the GS from above, below platform altitude, I think it more likely that the A/P was off.

the only way to do this with A/P on is-

Select a lower alt on MCP
Select a Vertical mode( Probably V/S)
Set a value for V/s
Wait while the aircraft gently responds
Wait till G/S captures
Reset MCP to Go-around altitude.

OR, you disconnect and shove the nose down.

Most pilots (me included) would opt for the latter.

Just a little thread drift here but I have seen several posters use the term "platform altitude". Is that a JAA term as to my knowledge it is not used in any Boeing or FAA manuals. Now back to our regular programing.

Thanks:}

Am I going a bit daft? Can anyone else see any evidence of the landing gear in any of the crash photos?

dc1968
If you look at fire service photos (link in a post earlier in the thread) you can landing gear

Spook,

I don't think it is even defined in JAR, but is a commonly used term refering to the initial altitude prior to G/S intercept on an Instrument approach.

I'm not so sure he is, its just his words and this black/white stall misconcept I think...

It certainly didn't have a very high vertical speed... my words yesterday were something like '... mushing along with insufficient thrust' at the moment of impact, what preceded that may have, probably was, more extreme pitch and speed deviations, but last minute recovery and control inputs may well have saved over one hundred lives..

The deaths occurred in the forward section experiencing extreme longitudinal decelerations, the centre section behind having many times its mass and hence energy.

I would not be surprised to find that control changed hands within the last minute before impact... nor that there was a degree of inattention shortly before and/or a prior poor decision.

For a low drag approach (flap 30), i would expect a Vref of around 142 or so.....

GS capture from above - likely?
 

Wisofoz,

I just compared the info from Google Earth, info in post#579
and info from the approach chart for ILS 18R, http://www.ais-netherlands.nl/aim/09...-IAC-18R-2.pdf

According the approach chart the crossing altitude when on ILS GS when passing 6.2 NM is 2000 ft. THY was not established until approx. 4.8 NM (acc. Google) and consequently did not pass 6.2 NM on localizer, as still on an intercept heading.
The aircraft passed 4.8 NM at 1900 ft. i.e. high on GS. (approx. 300 ft.)
He realized this and reacted by increasing descent, and he passed approx. 1000 ft. on the GS at 2.9 NM - acc. post#579. Fine so far!

Agree?

What happened thereafter will be guesswork.

brgds

Our SOPs require that the A/T are never fully disconnected. Yes, we do take out the autopilot, and announce it to the other chap, and also ask the other seat for "Speed Off" which requires speed to be deselected on the MCP. Following these actions, an alarm sounds for the AP disconnect, and the red-flashing warning light for A/T flashes. The Speed Mode is then shown as ARM. In these circumstances, thrust will return automatically when the actual speed falls below the selected MCP speed.

However, in the scenario where Vertical Speed is armed, the speed window opens at the selected speed, and the aircraft will attempt to match the specified vertical profile. One can place a lower altitude in the window to arm V/S, or as would make sense in this scenario, the missed approach altitude can be selected, and a vertical speed initiated - now there is no floor to prevent the aircraft acquiring an altitude which can mess everything up in a ILS! We've all done it, when everything is looking dandy, suddenly the ALT ACQUIRE pops up and the descent/re-capture close to being ruined. Fancy handy work, normally by quick hands on the MCP, or manual flight is really the only option then.

Vertical Speed offers absolutely no speed control. Although the autopilot is flying, it concentrates only on obtaining the nominated V/S, ignoring the speed entirely.

@nick 14: such a manoevre would almost immediately result in high sinkrate, regardless of high pitch/AoA.

@28L I fly out of AMS. I don't like late change-overs from automatic to manual. Pilot-not-flying workload during manual flight is not that high....

never discount the subtle changes to the cockpit environment by having someone in the jumpseat.

on the other hand, the jumpseater may have been the first to call a warning.

I'm not sure what the guy was saying about DC8 throttles not moving...unless he meant that they had to be moved by hand. the dc9 had the same sort of speed control/command authothrottles for approach only...they ended up being disconnected to avoid mx issues.

Hand flying making the NFP work harder...too bad. You need to keep your hand current.

subtle incapacitation hasn't been mentioned has it?

Have the Dutch indicated when we might learn something from the recorders? I know the French have them which I fear will no doubt delay any useful information coming forth. Being a Boeing, I doubt the info will be tainted.