Utair ATR 72 Crash in Siberia

[up_down_n_out]

"This take-off allegedly happened on Jan 1 this year, approx at this time:"

That's SU flight SVO to LED that time of the day.

http://cdn-www.airliners.net/aviatio.../4/2089416.jpg

Correction here:-

A/c ID
http://www.avia.ru/photo/aeroflotA32...vich/pano1.jpg

and here is a view the other up the wing:-

http://www.avia.ru/photo/aeroflotA32...h/IMG_2899.jpg

Here is video from Tiumen allegedly:-

Life News

[Sassy91]

Would have been better if we could see the leading edge of that wing. Maybe that was the only thing de-iced?

[dvv]

up_down_n_out, the author's comment on the youtube page identifies the aircraft as Aeroflot's Mstislav Rostropovich (A320 VP-BKY) en route to LED (St.Petersburg Pulkovo).

[liider]

Link to the news programme video (in Russian):

Авиакатастрофа под Тюменью: последние 42 секунды ATR-72

At 190m Capt. turns the autopilot on, 5 sec later orders "flaps up".
Aircraft starts buffeting.

FO: O--ohh..
Capt.: What's that?
FO:Why this buffeting?
Capt.: Autopilot disengage

Aircraft banks 35 degrees right.

FO: Quiet, what is this?
Capt.: Report to him.
ALERT: DON'T SINK!

Aircraft banks 50 degrees left.

FO: What should I report, f*k? What kind of failure?
Capt.: I don't understand.
ALERT: DON'T SINK!

FO: (cursing)
Capt.: UTAir 120, falling down!

Last phrase was not heard by ATC, probably because the pilot pushed the wrong button on the yoke. Aircraft collided with the ground 4 seconds later.

[vovachan]

"White" and "Red" bug speeds are determined from the charts dependent on take off weight.
Typically for the ATR 72 at, say, 20 tonnes the speeds are:
White Bug = 134 Knots
Red Bug = 161 Knots

Also from bits and pieces of CVR in the media:

Sounds like they never got to "red bug" - I figure this means they retracted flaps too soon for the icing conditions?

Decoding starts from the moment when the commander turns on autopilot and removes the flaps. After that the ship starts to shake, and the commander takes control, but the plane lurches - first right, then left, losing altitude and falls. Report to ATC, neither the commander nor the copilot have time.

Voice informer: Autopilot on. White Bug
Pilot 1: What is that shaking?
Voice informer: Autopilot disengage.
Pilot 2: Silence!
Pilot 1: What is it? Report!
Voice informer: Don't sink.
Pilot 2: What's to report? What's kind of failure?
Voice informer: Don't sink.
Pilot 1: UTair 120 we are falling

[FuntSole]

The latest info from Russian media (seems to be confirmed) that the aircraft in question was not de-iced/anti-iced after being exposed to icing conditions overnight. If it is true the outcome is unfortunately not very surprising...

[henra]

Originally Posted by liider

Capt.: Autopilot disengage

Aircraft banks 35 degrees right.
...
Aircraft banks 50 degrees left.

If there is one common denominator in ATR Icing accidents it is this.

In every single one you somehow find exactly this chain of events leading to a very unfortunate end.

[up_down_n_out]

So according to another thread here:-

Tailplane Icing


tail icing/rear stabiliser stall was to blame?

From what little CVR info released here....

PIO?
Buffetting and heavy stick pressure?

All seems to add up.

FO doesn't understand symptoms, does exact opposite of what is required, thinks it's a wing stall, has very little time to do anything about it,

Autopilot already took away all the early clues, and by the time he works out it's going bad...

..guess what the aircraft goes into a steep dive at too low altitude for any form of recovery.

(In these cases it's always the cockpit hits the ground first).

It's very possible the wings looked clear of ice on the ground, leading to this false sense of security.

Maybe:-
Tail icing wasn't even brought into the equation.

Eg.-
"Investigators have said the Colgan Air aircraft seemed to be flying okay to the moment that the pilots lowered the wing flaps in preparation for final approach as the airplane dropped below 2,000 feet. In short order, the autopilot disengaged, the airplane's nose pitched downward, then upward, before a final pitch downward.
The aircraft rolled left, rolled right, dove sharply and hit the ground, in short order.
The NASA Lewis video cites four warning signs of ice build-up on the horizontal stabilizer: a lightening of the controls, particularly stick lightening in forward direction; a difficulty trimming the airplane; the onset of pilot-induced oscillation; and buffeting felt in the controls, but not the airframe.

"Individual pilots may perceive these warning signs at different times depending upon the pilot's experience, the icing conditions, the workload in the cockpit and the intensity of the situation. Now, it should be noted that if you are flying on autopilot you would almost certainly miss these symptoms because you would not get any tactile feedback from the controls."

The Colgan pilots were operating on autopilot until the autopilot disengaged.
Said NASA:

"Remember, many times, these symptoms are encountered when flaps are at full extension. In extreme cases there may be a sudden pulse forward stick movement, possibly very strong. The nose of the aircraft may sudden pitch down, and it is very possible that this may not be recoverable on final approach because of the low altitude of the aircraft."

Tail-plane stall caused by icing often is first noticed when the pilot increases the flaps, the power or the speed, NASA said. In the Buffalo crash, investigators said the pilots had increased the flaps. When it began going out of control, the Colgan pilots increased the airplane engines on full throttle.

On a normal wing stall, pilots usually should increase power and relax back pressure on the yoke or push the yoke forward. On a tail-plane stall, they should pull back on the yoke, return the flaps to their former position and, depending on the aircraft type, ease off on power.
In other words, the remedy for a wing stall in some cases is the opposite of the remedy for a tail-plane stall, NASA warned."

Also:-

The Real Reason Behind Regional TurboProp Icing Crashes

Who maintain:-
The ATR was designed and manufactured in France by Aerospatiale. Mr. Fredrick's book, Unheeded Warning, reveals how political considerations overruled the safety mandate that is supposed to govern the FAA's design certification decisions on aircraft of foreign manufacture.

Fredrick details strong evidence that some experts in the FAA, knew the plane would be dangerous in icing conditions, but they were overruled by higher officials because the French might have been offended if certification was denied by the FAA.

He also demonstrates that numerous "close-call" incidents and one accident (10-15-87, near Lake Como, Italy with no survivors), which preceded the Roselawn crash, were known to be the result of in-flight icing, but Aerospatiale effectively covered-up and did not circulate that information, lest the reputation of its ATR aircraft be damaged.

[henra]

Originally Posted by up_down_n_out

So according to another thread here:-

Tailplane Icing


tail icing/rear stabiliser stall was to blame?

Honestly, I'm not really convinced it was the tail plane.
The Sequence of disengaging autopilot followed by significant roll excursions would rather lead me to believe it was primarily the main plane. Maybe with some contribution from the back.
But it seems it occured upon flap retraction. Flap retraction reduces the load on the tailplane, thereby it should be reducing the effect of any icing.
but the contrary appeared to be the case:
Apparently the effect increased/set on after retraction, therefore I would rather look to the main wing.

Edit: Tail plane icing would be particularly critical on the underside.
Leaving the plane outside on the ground in snowing/freezing conditions would rather lead to build up on the upper side. This is where the main plane is much more susceptible.

[Clandestino]

If there is one common denominator in ATR Icing accidents it is this.

Or Fokker 100. Or anything with highly loaded wings. Primary design concern of transport aeroplane wings is cruise efficiency. When it comes to stall, they are not supposed to display the benign characteristics of C-150 and ice accretion generally does not tend to improve them. ATRs have both shakers and pushers, clearly showing their natural stall is somewhat vicious, which doesn't imply we should be deeply worried about it and start another "ATRs are dangerous!" frenzy.

tail icing/rear stabiliser stall was to blame?

Absolutely not. Tailplane stall results in very rapid pitchdown, not roll. During NASA tail stall tests with Twotter, intention was never to achieve stall, just to get to brink of it and record the aeroplane behaviour. Plan worked every time but once. Nose went from horizontal to 80° AND in about two seconds, despite the two test pilots initiating recovery action as soon as the control was lost.

Tail plane icing would be particularly critical on the underside.

Errr... yes, but that's not the way it works in real life. To have tailplane stall, first you have to have severe ice accretion on leading tailplane edge (failed boot in heavy icing), then flaps have to be very powerful and tail has to be in their wake. Failing on any of that, you'll pick up ice, lose some performance, increase stalling speed somewhat but you won't make vertical dive when the flaps are lowered on final.

FO doesn't understand symptoms, does exact opposite of what is required, thinks it's a wing stall

Provided official reports confirm the leaked info, it was the correct thing to assume.

Investigators have said the Colgan Air aircraft seemed to be flying okay to the moment that the pilots lowered the wing flaps in preparation for final approach as the airplane dropped below 2,000 feet.

They have got it correct. Surprise, surprise

In short order, the autopilot disengaged

Because stall warning fired.

the airplane's nose pitched downward

Losing speed in approach config with engines on idle, it tried to regain the trimmed one. Perfectly natural.

then upward

Because the panicked pilot pulled so hard that he overrode the stick-pusher!

before a final pitch downward.

As the aeroplane stalled. Q400 is very simple: you stall it, you die.

Regarding The real reason behind regional turboprop icing crashes, it is very refreshing to read misinformed drivel about the aeroplane which is not Euro-made FBW airliner. Sad, but refreshing nevertheless.

[henra]

Originally Posted by Clandestino

Or Fokker 100. Or anything with highly loaded wings. Primary design concern of transport aeroplane wings is cruise efficiency.

What I wanted to point out was the involvement of the disconnection of the AP immediatly followed by violent roll.
That is something which happened in all ATR icing accidents. The AP covered the unfolding drama relatively long up to a point where the aileron forces could not be handled by the AP and then the Pilot himself anymore.This could be to some extent due to the unpowered ailerons in connection with icing on the horns.
Therefore manual flying when encountering icing conditions might be one of the better life assurances in an ATR (and the other TP's for that matter).

doesn't imply we should be deeply worried about it and start another "ATRs are dangerous!" frenzy.

Was not my intention, please see above

Errr... yes, but that's not the way it works in real life. To have tailplane stall, first you have to have severe ice accretion on leading tailplane edge (failed boot in heavy icing), then flaps have to be very powerful and tail has to be in their wake.

Exactly my point. I wanted to figuratively show how unlikely tail plane icing would have been in the given case.
Anything short of massive ice accretin on the tail with flaps fully extended should have no appreciable effect on the flyability of the aircraft.

[Clandestino]

Relax, Henra, it's nothing personal.

What I wanted to point out was the involvement of the disconnection of the AP immediatly followed by violent roll.

Correct. Violent roll comes from stalling (or aileron hinge moment reversal at Roselawn). ATR picking up a ****load of ice feels very mushy - recommendation to fly manually in heavy icing is quite sensible.

This could be to some extent due to the unpowered ailerons in connection with icing on the horns.

Heaters do their work just fine. I don't recall any incident or accident because of failed flight controls horns heaters. In my 6 years on ATR, I lost them only once, which prompted mightily rapid descent to MSA, below freezing level.

[liider]

According to preliminary results of the investigation, the accident with the UTair aircraft ATR-72 was the result of the deterioration of aerodynamic characteristics of the wing and stabilizer as a result of the ice not being removed while preparing the aircraft for the flight after a long (8h) overnight stay in icing conditions (freezing rain, and the transition of temperature below 0°C, the wind 7, gusts of 10 m/s).

The investigation of the crash of the ATR-72 VP-BYZ identified deficiencies that lead to the conclusion that the preparation of the aircraft for flight was made in violation of the rules by ground staff of "UTair-Technik", which was to conduct de-icing of the aircraft, and by the crew, which did not assess the state of the aircraft surfaces and meteorological conditions during the overnight stay of the plane at the airport Tyumen.

In the process of making the decision, that de-icing is not necessary, also participated an aircraft mechanic, who had no aviation education and had been working in "UTair-Technik" for little more than one and a half year. Aircraft mechanic was admitted to ground handling services after 16 hour training courses, of which to the theory of ice control and the de-icing of the aircraft were dedicated only 30 minutes. The supervisor of the shift, who should have controlled the aircraft mechanic's work, was on a nearby parking lot, located next to the plane. Having received the information, that the de-icing of the aircraft was not carried out, the supervisor didn't intervene.

The violations in the ground staff's work also point out that, despite the introduction of airline standards similar to accepted overseas, the necessary quality control system is practically inactive.

[tu144]

You have source where you got this info from

[JammedStab]

http://www.bea.aero/docspa/2012/vp-z...vp-z120402.pdf

[cappt]

Very detailed report.
Conclusion starts on pg192.

[Typhoon650]

I've always wondered, if turbine aircraft typically run some sort of fuel heating to warm up fuel for the engines when cruising at altitude, why not design a system that allows recirculation and heating of fuel in tanks when on the ground?
The fuel wouldn't need to be heated too much, 10c should do it and the excellent heat transfer abilities of aluminium should help to keep ice from forming from at least the majority of the wing?
It seems the systems are already in place to heat fuel and a gas turbine has plenty of excess heat available when running to turn it to this job.
I'm also sure smaller operators would pay for an aircraft that could possibly deice (or even prevent ice buildup) itself on turnaround, or within 30min/1 hour on a cold start with just the APU or a ground cart. Both APU or cart would probably be running anyway, producing large quantities of waste heat.

[Clandestino]

Engines do not produce enough waste heat to warm up all of the fuel. Fuel tanks are almost never full so upper wing skin would not be in contact with heated fuel even if there was a way to keep it warm. Leading edges, control surfaces, high lift devices and empennage have no fuel tanks inside them so can't be internally heated by the hypothetical warm fuel.

Glycol based ground de-icing/anti-icing works when applied properly.

[JammedStab]

Originally Posted by cappt

Very detailed report.
Conclusion starts on pg192.

And a very difficult report to read as are many of the Russian reports. All kinds of unnecessary detail about unimportant subjects.

And.....for a report with very poor English, a large amount of it is spent criticizing the pilots poor English comprehension(and therefore their ability to read about the effects of contaminated wings).

I was very happy to finally finish reading it.

It appears that these pilots were the only ones who did not de-ice that morning. They were told by the ground engineer that it was not needed. Based on the weather conditions that had occurred while they were at this location, a reasonably sharp and safety minded pilot may very well have disagreed with this report from the engineer.

Apparently, there is a culture of bending or breaking rules in Russia. It cannot be ascertained from the report or CVR whether or not these pilots fell into this category. However, for some reason, the de-ice boots were used while on the ground during taxi out not just a short time in case it was desired to check for proper operation but for quite a long period. However, there was no recording of the reasoning for this action.

It appears from the weather reports that wet snow had fallen and temperatures had decreased likely leading to rough contamination on the wing(not just some frost). However, the aircraft did get airborne and reach several hundred feet of altitude.

If one were to decide to knowingly try this, it would seem to be a prudent idea to wait until a reasonably high speed had been reached prior to retracting the flaps. However, the flaps were retracted at white bug which is the minimum allowable speed to do so and the aircraft lost control.

So in the end, perhaps they were just truly unaware of the contamination and had little situational awareness from a weather point of view and were just testing the de-ice boots on the taxi out, or they intentionally took off with contamination(perhaps assuming it was less significant than what it really was) and lost control.

A couple of things to think about. If the maintenance guy says the wings don't require de-icing, it may be prudent to confirm for yourself. If you somehow do get airborne with contaminated wings, waiting for a higher speed to retract flaps could very well be beneficial.

[Clandestino]

Nice advice on takeoff with contaminated wings you have there. In order to clear-up the applicability of it, pray tell us how does one:

1. fail to notice wings need de-icing during walkaround?
2. fail to notice icing conditions during ground ops that necessitate use of anti-icing?
3. suddenly realize that wings are contaminated between Vr and acceleration altitude so postpones flap retraction, or in more extreme case; between V1 and Vlof so gathers speed on runway or in ground effect?

Apparently, there is a culture of bending or breaking rules in Russia.

You don't say!