It does look like the port side stabilizer and elevator are misshapen but it may just be some kind of an optical illusion.

Even if part of the stabilizer were missing I don't believe there is enough missing to account for what we see happen to the aircraft. To me it just looks like a stall with a wing drop.

Lowering the nose would reduce the AOA on both wings and would prevent a stall. Are you are suggesting that a smaller port side stab would reduce the AOA on only the starboard wing and cause a spin to the left? That just doesn't make any sense to me at all.



http://i2.photobucket.com/albums/y17...e.jpg~original

Nope. Firstly, regarding spinning, a reduction in AoA will reduce the likelihood of a stall/spin, not increase it. (It would appear that had the crew reduced the AoA prior to the roll commencing, the result would have been similar; it would have hit the bridge, but probably at a slightly different - and maybe less survivable - configuration. They simply didn't have the altitude to play with, and probably not the required power to prevent it.

I doubt that anything the crew did made the accident more or less survivable. They were clearly out of options, and just prolonging the inevitable (as you do. Poor buggers.) (This is called ''stretching the glide''. It didn't quite work. It often doesn't. It may have set up - inadvertently - an accident geometry that made the impact survivable for some on board. Having seen the videos, it is not the sort of accident sequence one would expect survivors from. Those few were darned lucky.)

The accident sequence was already established; the roll to the left was merely part of it.
The aircraft was already committed to crash into the river/bridge prior to any power line contact taking place.

Appears a classic stall/incipient spin to me.

Post 116 - One can understand why they would stretch the glide as the terrain was as hostile as it gets.






There is no "stretch the glide" ability by going slower than your best L/D speed. Going slower reduces how far you fly.

Further, to stretch the glide into wind the airspeed should be increased above best LD by around half the wind speed.

Stretching the glide..yes,I was sure I mentioned that..I just can't found out where...Alzheimer's
..or Mr.Moderator!!

This assumes that all you care about is at what point along a flat surface you will hit the ground. There are often other considerations besides best distance. Like perhaps a large chunk of concrete in your way.

From the close in pics of the aircraft clipping the car, can anyone see if the rudder is being use - I would have expected to see a whole lot of right rudder but to me it looks central??

May I throw a spanner in the works, Does this company use Selatar in Singapore for their proficiency checks? If so is this company one of the Asian carriers that were continuing to check their pilots in a simulator with broken rudder peddles? I know for a fact that this simulator was still been used by company's in this broken state in the second half of last year. If so did these two pilots conduct a check and pass in a simulator with broken rudder peddles?

From that image it is appears that the left prop was rotating at much slower rate, hence the different motion blur that both props have got on the image - it is 2-2.5 times longer on the right propeller.

If my assessment is correct, it looks like the right propeller was rotating faster - (roughly) 2 2.5 faster, than the left one.

Here - i made a small test to confirm that - the left object rotates at 2.5 times faster rate than the right one.


http://s21.postimg.org/7enx6n9jb/atr..._blur_test.png

Where do I start?

If there is damage to the port side elevator balance tab that some of you see in the video stills, where did it come from? Ground damage missed on preflight is my only guess. Even if a turbine wheel burst, it would most probably have been contained by the shroud. The engine parts would not go downstream and hit the T tail, no alignment. Prop blade? I see all four blades on both engines. Only "damage" I see prior to impact is caused by camera angle and the inherent distortion of a digital image expanded beyond the limits of its pixels. No, the aircraft was in good shape before impact.

In the best of the two dash cam videos, the aircraft appears from the left and is wings level, apparently slow, in a nose high attitude at a high rate of descent. At one point after takeoff they declared a "Mayday, engine flame out".

But at one point in the flight they had altitude (1250 feet), near sea level (fat air), and adequate airspeed. It was not that long a flight, so probably a light fuel load and well under seat capacity. The were not "heavy".

The aircraft should have been able to fly after V1 , climb and return. Minimum airspeed, no altitude, the aircraft should have been up to the task.

They pissed it all away and when they saw that they were not going clear the roadway, panicked and deep stalled the aircraft; the left wing stalled first. Lift normally generated by thrust over the wing that was absent due to the failed engine determined the direction of roll. Stalled. Not a "spin". A spin would need altitude.

The crew mismanaged a manageable incident. They "screwed the pooch". Whether it was their shortcomings or the lack of good training or training standards, is not mine to speculate. But when I transitioned to several different aircraft I was trained to "approach to stall" recovery. It wasn't till several crews actually stalled the aircraft (thankfully with sufficient altitude) that we received true stall avoidance/recognition/recovery training.

I agree with everything you say but an ATR-72 has more than four prop blades. I suppose the photos only look like four blades due to motion blur of the dash cam video.

http://i2.photobucket.com/albums/y17...C9IAAEb1g6.jpg

http://i2.photobucket.com/albums/y17...16/Taiwan2.jpg

http://i2.photobucket.com/albums/y17...o16/taiwan.jpg

Bloom you are happy to speculate that "they pissed it all away", "deep stalled" (sic), "mismanaged a manageable incident" and "screwed the pooch" on what little evidence there currently is yet it's not for you to speculate on training standards...wtf?

As for the old stretching the glide scenario, we need to consider that this probably saved those few. Turning that last bit of airspeed into momentum away from buildings and admittedly losing glide distance afterwards may have prevented a 911 type impact.

To deep stall it they would have had to defeat the stick pusher.

There is nothing in the alt profile to suggest that it activated

I think an engine failed at ~1,300 ft
Look at the graph on page 6 of this thread.
The flight lasted 3 min, 45 sec

00:00 Takeoff roll commenced
01:00 Rotate
01:45 Lower the nose, accellerate, retract flaps (maybe), all normal, autopilot may have been on
02:10 Engine fails at 1,300ft, pilots level off, airspeed decays
02:20 Wing stalls, aircraft drops 500 ft in 15 sec
03:45 Impact

I think the wing stalled in level flight after the engine failed. Maybe late going TOGA or MCT on the good engine. I think the aircraft remained stalled, or at least on the far back side of the power curve, until impact.

The crew were likely task saturated trying to secure a failed engine, turn away from terrain, and recover from a stall, almost simultaneously. Throw in GPWS alerts to add to the mix at the end. And we know from the final seconds of the video that this ATR can surely drop a wing in a stall, so the initial stall that lost them 500’ in 15 seconds may have well been similar. Very disorienting.

We all practice V1 cuts. Rarely do we practice engine failure in initial climb or in the cleanup phase, possibly in a terrain avoidance turn. My last airline had a sim scenario with engine failure at 700’ in IMC with the autopilot off. Much harder than a standard V1 cut.

Apparently one of the air stewardesses who survived was also on board the last crash in July 2014. Lucky girl, depending on whether you're a glass half full kind of person! Not sure how anyone could return to work after going through this twice though.

58 souls, 15 survivors, 12 stilll missing.

as they never reached VFTO they didn't complete memo items ( SHUT OFF engine )....if engine 1 was not fully flamed out ...propeller was still spinning.

Interesting comments as Selatar in Singapore is in fact ATR's own training Center and until last month was the only operating 72-600 simulator within the region. Would find it somewhat difficult to accept an aircraft manufacturer would compromise themselves this way!

Another explanation is that when the elevator is in full-up position, the mass balance horns protrude from the shape quite prominently. The starboard side seems to look similar.

In that situation, maximum pull would be a natural, if not optimal, pilot action.

Fishy ...
 

Can't spell pedals and doesn't know it's Seletar not Selatar.

Do apologize for my spelling as I was on a IPhone and in a rush. But if anyone of you do the research you will find that it is fact that the sim was still operating with inoperative rudder pedals for up to 4-6 weeks. ATR were still allowing companies to use it in this condition.

Yes the SIM in Singapore operating for about 2 weeks without rudder pedals on the Captains side not 4-6 like you say. You are correct in saying many companies continued to train and check their crews in it, all apart from the Kiwi and Ozzy companies that refused. It is completely at the airlines and TRE/TRIs descretion if they want to continue training in a faulty product. Read into that what you will!

With all the assumptions and speculation in your post, and the condemnation of the crew, maybe you just lobby to replace all accident investigation operations with your superior ability and random deductive skills :ugh:

Or, better still, make some considered observations without coming across as a thoughtless and arrogant prat by drawing such absolute and as yet unfounded conclusions.

A couple of questions from a dumb F$%, and they are not to imply in anyway this is whats happening, just technicality and enlightenment.

Roll rate, yes we know it can't roll that quickly commanded. But what is the roll rate with one engine out near stall maybe with some rudder kicked in?

Glide distance, yes pilots know the drill, optimum glide distance, too fast or too slow you fall short. But does that not imply a speed that the aircraft is under some sort of reasonable control, even up to landing. Not maybe opps we have some power, now we don't. Or the thing is flying when it really shouldn't be at the moment speed.

Regardless of the theory, controlled flight into high rises or houses may make you try for the illogical.

A lot of speculation here just like the Air Asia thread. We all would like answers but it may be optimal to hear the outcome of the investigation(s)....before coming up with definitive conclusions based on assumptions.

The 3 cockpit crew
 

Was this likly a line training flight.One of the crew members had iover 16k hours and the other 2 had 6.9k and 4.5k respectively. This was no lack of experience scenario unless all 3 were low hours on type.Giving them all the benfiit of the doubt they probably did all thta was instinctly and procedurally possible to minimise damage. That last minute high bank might have been to avoid clipping the structures on the road across theior flight path with the intent of ditching into the river?Had they not clipped that car the plane might not have hit the road bank?I wish the crew are alive to tell the story in hindsight...but are they?Cockpit was submerged for a long time and they must have been strapped and probably might have been electrocuted with all the electrical and electronic bays behind their backs.

Any news of the recovery of the 3 cockpit crew?

If it was a "glide" as a lot are saying then it was done incorrectly. (As per QRH/FCOM)

I don't pretend to know the weight of the aircraft involved but the target speed for a glide in an ATR is VmHB. If the info out there is correct and its maximum speed was only 113Kts then it was a long way from the optimum glide speed especially at a flap zero configuration. At 80 odd knots with flap zero that airframe was stalled as it went in.

What is extremely obviously to those with experience on type is that the aircraft had an issue with an engine that was either managed incorrectly by the crew (I really hope not) or was completely unmanageable.

Local-ish newspaper reporting that one family switched seats prior to take-off as they were unhappy about engine noise from port engine:

Together with the child's mother, the family had switched seats on the plane "out of a hunch" that saved their lives, the United Daily News said.
"The family originally sat in the heavily damaged left side but Lin Ming-wei felt uneasy after he heard noises before taking-off and requested to switch seats," the report quoted Dai Bi-chin, a friend of the family, as saying after visiting them in hospital.

From
TransAsia pilot hailed as a hero for avoiding populated areas - Regional | The Star Online

Oh BO0M......lets not all get too smart here...
As has been mentioned,we do not have all the facts yet,to confirm double engine failure.
However, we are not talking about perfect conditions with an optimum approach angle and speed to the water surface..Is it not apparent to you that his choice of touchdown/crash point may well have been compromised!..just a little!

.."Done incorrectly"......please!

BTW..have my "Silver C"... And the Lasham Plate!

I flew on B-22816 last year - odd vibrations
 

I'm a longtime reader and finally have something to contribute. I live in Taiwan and fly back and forth between Hualien Taiwan and Taipei almost weekly. Transasia is the only carrier that services our town and really the only domestic carrier aside from a few Mandarin airlines connection flights.

I just checked my photos and I flew on this EXACT plane last year on July 24th from Songshan airport. The flight really stood out to me because the plane was very new with different lighting and interior finishes than the rest of their ATR-72 fleet.

This plane had the oddest vibration in the engine. I normally sit up in seat 2D, ahead of the props since noise is lower. That wasn't the case on this plane. When they started the engine the vibration was intense up front and continually rattled the emergency exit door. The flight itself was uneventful. I've flown on this type over 50 times and this one was unusual. I usually enjoy flying on a new aircraft, but not in this case. Here is a photo from boarding that night on the tarmac. I'm now seriously considering taking the train or driving until TransAsia gets their issues sorted out. Sadly tour-bus and truck filled mountain roads here are pretty dangerous.

http://i.imgur.com/qREfHdIl.jpg

Stretching the glide....undershooting. Very little time to do the right thing, but the BA777 captain on final approach to LHR a few years back, the copilot being PF, and the engines declining to provide power when asked....the captain retracted a stage of flap, they scraped over the fence and all survived.

That was a power failure on approach. Much more difficult if you are faced with EFAT. Too many problems - how to improve the performance of your aircraft which has suddenly turned into a very unhappy glider. No flaps to put away. A useless windmill on one wing at least, acting as asymetrical airbrake. Poor choice of landing areas immediately ahead. No time to restart an engine. They did very well to make it to the river.

When undershooting in a simple glider, with no other problems, you have to lower the nose to INCREASE your speed over the ground, and if done properly this may get you over the hedge. The worst thing to do would be to raise the nose to clear obstacles.

Well the other TP drivers will have seen the coming in of heavy jet guys to regional TP due to lifestyle reasons.

Sometimes its not pretty.

Quite a few don't make it past type rating.

So the linetraining theory may have some weight.

Also nobody has said where that data has come from. It maybe that its ground speed not airspeed. The wind on the day was about 10 knots.

Also is there a noise abatement procedure with a V2+10 off that runway?

I must admit my speed control is very poor when we are meant to do them. My normal 0-5knts tolerance can get as bad as +40. And any excuse such as a bird being seen 24 hours before departure means I can't do it due flight safety reasons.

Never flown a jet, and never had to do one for real, but in the sim it gets very acrobatic in a tp when you get an engine fail at that speed and gear up. You have to instantly drop the nose from 20 degrees nose up down to 9 up Any delay and your speed is below v2. You have to set the attitude before the asi starts moving. You deep into the dirty side of the drag curve.

If they have been doing a v2 + 10 the departed i salute you I more than likely couldn't have done any better if I hadn't cheated with poor speed control.

Cmon Yaw I think you have read to much into what I'm trying to say. I've sqid over and over we don't know the facts!

The point I was trying to get across is that a double engine flame out and glide is highly unlikely! There isn't an ATR pilot out there worth his/her salt that doesn't know VmHB and feathered props are your life in such an event. Stretching the glide can only be achieved with Flap 15 and allowing the VmHB to reduce (and let's face it it's not really stretching it).

Like I said stating some facts to hopefully make people aware a double flame out/glide this was not.

previous engine trouble
 

Is there any documented evidence of previous left engine trouble and what was the nature of that trouble that needed fixing?

Does anyone know what the nature of the 'technical issues' were that resulted in both engines being replaced in Macau on the delivery flight last April,as alleged in the following article ?

Taiwan pilot hailed a hero for pulling plane clear of buildings - World | The Star Online

nnnnn
nnnnnn
nnnnn
nnnnn

Although not reported in the media, while I was watching the recovery of the front cockpit section there was a pause when two stretchers were taken to the area and taken away carrying two bundles wrapped in white cloth.

I don't know the contents of the bundles.

Watching the events of getting the aircraft out of the water, it has to be observed that the local authorities really did a well organised job and achieved the result pretty quickly.

Yes the -600 does have auto rudder trim, which is active as long as the Yaw Damper is engaged, it will hold in the event of an engine failure, even though I have only tried it in the simulator.

Without the yaw damper engaged it just a matter of correcting the yaw with the rudder pedals, and then engaging the yaw damper, which will then trim out any rudder force required.

The -600 is very easy to handle single engine, much easier that for example the -500.

However there is an risk of "lesser" pilots becoming reliant on this system for handling SE flight, and forgetting the basics.

Mayday call issued before deadly crash into river in Taipei | Daily Mail Online

Does anyone else see a feathered prop hub?

The Daily Telegraph says the captain had complained about the left engine...

"The latest so far is the claims that the captain, Liao Jiangzhong, complained of “engine abnormalities” and requested an urgent inspection of the plane shortly before its final take-off but was rebuffed.

Liao Jiangzhong, the plane’s former air force pilot, is among 32 people so far confirmed to have died when the aircraft crashed into a river shortly after taking off from Taipei’s Songshan airport on Wednesday morning.

An unnamed “whistleblower” told Taiwan’s Liberty Times newspaper that Mr Liao requested a thorough inspection of the plane after noticing “engine abnormalities” during its previous flight.

The pilot registered the problem on a flight log, the newspaper added."

TransAsia plane crash: Pilot complained of ?engine abnormality? before take-off - Telegraph

It's not a developed spin (there wasn't the altitude for that) but is sure looks like an incipient spin to me (been there hundreds of times, but at a safe altitude). The aircraft is obviously semi-stalled at least just before the roll-off to the left, nose high with a high rate of descent, very high alpha. Then the left wing dropped as the bottom fell out of it.

What we don't know is why it got into that situation.

If it's true the airline happily used a sim with inop rudder pedals then that says a lot about standards.

Agree that the aeroplane at high alpha as this one was would have come down faster than it would at best glide. But as someone said, the attitude may have been forced on the pilots by having to avoid hitting a building, by sacrificing speed for height. Though from what we can see in videos there are no buildings protruding into the aircraft's flight path that would cause that - maybe just involuntary desire to 'pull back to keep it in the air 'till the river?'