B737 Go-Around tragedy. Pilot error all over again
 

https://mak-iac.org/upload/iblock/45...vq-bbn_eng.pdf
Suggestion.
Those readers who prefer to quickly get to the nexus of accident reports should skip the peripheral information which forms a great part of this first class investigation. This is where speed reading skills come to the fore.
Otherwise by the time you gets to the guts of what went wrong in what should have been a normal go-around in a Boeing 737, you will suffer from reading fatigue, despite numerous cups of coffee.

One way or another, depending on your bent, eventually you will reach that part of the report where the investigation sets up a series of go-around scenarios in the simulator and had ten 737 pilots fly the same procedure in IMC. Those results are seriously thought provoking and I wouldn't mind betting a high percentage of current 737 pilots today would likely have similar results. Some would even crash.

Designers of simulator training syllabi would do well to learn from this report and ensure ample time is allotted to teach IMC low level go-around procedures as part of type rating training and proficiency checks.

Pages 96-100...
Worthy of notice, the seemingly lack of/ fake qualification of the accident PIC and the mention of "dubious flight schools" for some pilots in the experiment.

How about this non-Max crash. Seems awfully suspicious: Flydubai Flight 981. Is there a problem with A/P in these birds?

No. But there is a problem with pilots suffering from Somatogravic illusion during high energy go arounds - which is impossible to demonstrate in present simulators.

See https://mak-iac.org/upload/iblock/45...vq-bbn_eng.pdf page 219.
& https://mak-iac.org/upload/iblock/3d...a6-fdn_eng.pdf page 166.
"2.4. On the possible impact of somatogravic illusions.
Somatogravic illusions have been noted repeatedly as a contributing factor during investigations of aviation accidents and incidents which involved go-arounds."

Hi Golden
But isn't the basics of instrument flying is to only trust your instruments and not your feelings. Humans have no instincts in the air. Somatogravic factor comes in when the pilot is trying a pitch change without looking at the PFD and as a result it is too shallow. In FlyDubai copilot didn't suffer it because he was monitoring the pitch and guiding the PF perfectly. Unfortunately the PF had gone into the realm of the irrational. AF447 no somatogravic illusion but again commanding a pitch without looking at PFD. This loss of scan causes the problem. These are also procedural errors.

Hi vilas, I agree - but sometimes when trying to explain why a crew did something very unusual and there wasn't any technical problem with the aircraft - then accident reports often mention the possibility of somatogravic illusion. It doesn't matter how often you you may have practiced in the simulator (all at 1g and within the pitch limit geometry of the motion) apparently the illusion is powerful and can take some crews by surprise.

How many times have you seen crews in the simulator performing something mildly stressful and they are both shouting over the "C chord" altitude alerter - neither of them even notice it is on until the work load is reduced? I suspect the somatogravic illusion stresses the crew member to such an extent that they don't read the PFD correctly or may even disbelieve it.

I don't think pilots are trained to tackle any illusions at all because they cannot be created in the simulator. In the simulator they are simply trained to handle the plane with reference to instruments. The somatogravic illusion of which the pilot has no experience hits him whenever the technique that is taught and practiced by him to execute a GA i.e. the pitch, power is not followed and the community comes to know through a fatal accident. I am not sure if anyone has experienced it and overcame to survive. So all that is in pilots hand is to follow correct technique and if it is not happening may be someone out of the two takes control to effect it. In FlyDubai case the copilot knew at all times what is happening and what should be done, unfortunately he didn't physically intervene.

Hi vilas, I agree.
This 2016 powerpoint from Cathay Pacific Group Safety department https://flightsafety.org/wp-content/.../11/Ludlow.pdf sounds like their Military pilots have received SGI training experience in flight. (Slide 38). It also has some very good advice about briefings for SGI avoidance, use of Autopilot and using reduced thrust during Go Arounds when possible.

FlyDubai had lots of holes in the cheese lined up that night: Tired crew after long holding delay, light aircraft, full GA power due WS warning, single AP approach resulting in AP disconnect on initiation of the GA, etc.
If only they had briefed SGI as a threat.....?

Funny that aircraft carrier pilots are not crashing all over the world during thousands of catapulted takeoffs and wave-offs (go arounds) every year. After all, you can not imagine a more severe acceleration in these aircraft.

One thing noticeable in many accident reports involving a go-around accident at night or in IMC, is that authors of these reports are quick to use somatogravic illusions as contributing factors when they know there is no way of proving these were positively identified as present. A get-out-of-jail free, card

What many investigators fail to realise, if the truth be known, is that the majority of these accidents are simply caused by poor instrument flying skills on the part of the pilot - not necessarily by the little hairs in your ear canals.

Investigators at loss to explain the cause of an accident, can simply blame the spectre of somatogravic illusions, as if that explains everything. Their report can now be signed off because no one can prove otherwise..

After having flown and taught hundreds of aerobatics flight hours, I'd like to point out that somatogavic and spatial disorientation do occur, especially when one is not accustomed to rotations and accelerations, or when something unusual occurs (aircraft upset, etc.) or when visual cues are lacking.
It even happens to well trained fighter pilots.
There is no means of simulating rotations and accelerations in the sim, and I doubt many airlines do practice aircraft upsets in a real plane, so I wouldn't be so affirmative re this "poor instrument flying skills".

Somatogravic illusion could be avoided or at least minimized by performing a soft go around. Many skippers believe they need to keep MAX power until thrust reduction which leads to having way too much energy with all engines running. On Airbus a small tap to TOGA then back to climb detent when appropriate (or Flex/MCT detent if equipped with the Soft Go Around) will make the go around much safer. Less chance of illusion, overspeed and altitude burst.
The recommendation to perform soft go around can be found in the Airbus Safety First magazine #23.

Fly Arpt
. During Instrument flying traning pilots are trained to manoeuvre the aircraft accurately purely according to the instrument indications. When a GA requires 15° pitch up but is executed at 4 or degrees isn't it poor flying? I will again say that it all starts with giving pitch inputs without looking at the attitude indicator. In FlyDubai the FO who was monitoring the attitude did not suffer any illusion and perfectly advised the captain. In Atlas air case (though to early) the FO was opposite. He didn't check the attitude but voiced his illusion loudly misleading the pilot who wasn't monitoring either. The aim here is not blame but find a way out of this. Which is to imbibe that pitch inputs require attitude reference.
Pineteam I would say feel what you will but do only what you are supposed to.
​

​

pineteam, if you are talking about Airbus then this is normal. Stick free Airbus FBW tries to maintain 1g path. So it only permits pitch change as a g demand through the stick. The thrust weight couple pitch up is resisted. In Airbus pilot has to raise the nose himself otherwise it will not go beyond 5°/7°. In a 737 GA thrust will cause a strong pitch up which needs to be held by forward pressure and later trimmed off. May be that becomes a habit with some pilots and they push forward without reference to the attitude.
​​​

I'd go to manual thrust, or v/s, and reduce the rate of climb to 1,000 FPM, on Boeings. It reduces all the drama.

Goldenrivett, # 4
However, from page 169 “the complex analysis of all the information available, stated above, is the evidence that, most probably, both the intense forward repositions of the control column up to the initiation of the stabilizer trim to nose-down and the trim itself, had not been attributable to the PF (the PIC) having been affected by the “pitch-up” illusion.”

“… the investigation team is of the opinion that the potential occurrence of the somatogravic “pitch-up illusion” did not have crucial importance as far as the onset of the accident situation is concerned. In the progress of the accident situation the pitch-up illusion might have had some impact on the long-time keeping the trim switches pressed to nose-down.”

The issue of trim and continued forward pitch is discussed in the primary thread on the accident - https://www.pprune.org/rumours-news/...eport-out.html
and specifically at
https://www.pprune.org/rumours-news/...l#post10642086

Centaurus, is the issue of trim / pitch feel something which can be identified in simulation in the 737.

well aerobatics was part of my initial training(long time ago and never again)and I have done go arounds in piston, turboprop and three comercial jets including a FBW. Did I experience SI? You tell me. All I did was pulled aircraft to required pitch and set the thrust. Am I immune to SI? Definitely not. Had I done anything differently and not monitored the attitude before and after then I would have reacted the same way. Solid training? You cannot be trained against SI in a simulator. You only get trained as I said to manoeuvre to required pitch and bank. Awareness? Everyone has been made aware of SI. We are made aware so that we don't react by feelings but to reality on the instruments which according to you is almost impossible. So back to square one.

Hi vilas, Next time you have the chance, try flying the sim with the motion off. If you can fly it initially as smoothly and accurately as you can with the normal motion on - then I'll be impressed.

We get used to the sensation of pitch from pressure in our back and butt which agree with the ADI. In the real world, with an unexpected acceleration from full TOGA power and a light aircraft - then there can be a sensation / visual mismatch which can cause confusion.

Perhaps SI affects some pilots and not others? I recall my first night takeoff in the P51 Mustang. At the time I only had 240 hours total time. Now that aircraft really accelerated on take off. I glued my eyes on the artificial horizon and directional gyro once airborne. I certainly don't recall any problems during the initial and subsequent climb which according to my log book was to 15,000 ft.
I have experienced "the leans" in many aircraft but never SI and that includes flying jet transports for decades. I suppose the question needed to be answered if some pilots are more prone to SI than others, could that be proven medically? if so, should they be denied a pilot licence or perhaps restricted to day VMC flying only?

To change to a climbing flight path you have to generate upward acceleration (expressed in g's).

dh, agree.
SI involves acceleration; generally longitudinal, and lesser in non-turning flight while pitching up. A GA contributes little speed change nor great rotation acceleration, usually very short duration and thus insufficient for disorientation.
This appears to to be the reasoning behind the conclusion in the accident investigation.

The pitch - body feel - stick / trim feel aspects could create a different illusion.
I recall that there was a very interesting 757 event during GA where this could have contributed.

Golden R, All of that is known and is supposed to happen to all pilots and yet literally hundreds off GAs are performed all over the world without a problem. How is that? In fact GA is considered so mundane that even a report is not required. And then we have a fatal GA and SI appears almost as an obituary. And every time the nose is pushed down to a bizarre -40/50° pitch. Nobody ever stalls by over pitching. Centaurus has raised a good point. Are some humans more human than others? Is it not always a wrongly executed manoeuvre that opened up a Pandora's box of human factors? Knowledge of SI is not going to help unless it is imbibed that a pitch input must be made by looking on PFD and not by feeling.

dh, ‘There are plenty of cocked up GAs in the 757 lifetime!’
and 737 ?
GA Trim ?
https://www.skybrary.aero/bookshelf/books/2627.pdf

Is this the incident you had in mind?
REPORT 7/2003 - Date: 22 January 2003
SERIOUS INCIDENT TO ICELANDAIR BOEING 757-200 AT OSLO AIRPORT GARDERMOEN NORWAY 22 JANUARY 2002
Aircraft type: Boeing 757-208
Registration: TF-FIOOwner:Flugleidir h/f, 101 Reykjavik airport, Reykjavik
Operator: Flugleidir h/f, 101 Reykjavik airport, Reykjavik
Crew: 2/5
Passengers: 75
Incident site: Over RWY 01L at Oslo airport Gardermoen
Date and time of incident: 22 January 2002 at time 1049 hrs.

I'm not allowed to post URLs yet, but I'll gladly PM you the link and maybe you can share it for everyone to read.

Get a hold of this: "the lowest altitude in the recovery was 321 ft radio altitude with a peaked load factor of +3.59 g’s."

Seconds from being another tragedy, just like all these nosing-over in IMC incidents we're reading about. Confused Captain is PF, gets behind the aircraft (due to illusions or whatever other reason), puts the plane in a steep dive; only this time the PM manages to snap him out of it and they both pull for their lives and barely escape before hitting the ground. Some passengers get a glimpse of the earth at the lowest point of the recovery, the cabin is a mess, everybody (except the sheepish flight deck crew) is traumatized for life, but thankfully everyone is alive.

Pretty sure there are some obvious problems that all these incidents and accidents have in common, and that they need to be addressed in training. Urgently.

Depends on the operator. In at least one major Chinese airline the company fined the captain for doing a go-around. That is because in some cultures real men don't go around.

https://www.skybrary.aero/index.php/..._2002_(HF_LOC)
Click on the red link.

"When initiating the auto “Go-around”, the auto throttle became engaged, and increased automatically the engine trust to the EPR limit. The application of the under wing engine power also gave a pitch up movement. The flight director pitch bar commanded a pitch attitude of 15o. (The AFDS commanded a level off at 2 500 ft (the last assigned altitude by ATC). The AFDS calculates the high closure rate to 2 500 ft and captures that altitude almost immediately; causing the A/T to change from Go-around mode to retard power to MCP selected speed (150 kt)). The aircraft therefore climbed very rapidly through the MCP selected altitude of 2 500 ft and with the aircraft pitch increasing to 21o. AAIB/N considers that the Commander at this time had lost situational awareness (being “behind the aircraft”).

2.2.3 As the Commander noticed the speed to be rapidly decreasing, he pushed the control column forward. This was in order to follow the command of level off at 2 500 ft. Pushing the control column forward is also an elementary flying procedure to increase the speed and to prevent the aircraft from entering a stall.

2.2.4 The First Officer at this time called for “Bug up”. The Commander reached for and reset the MCP speed. This was contrary to company Standard Operation Procedure. Simultaneously the Commander continued to push the control column even more forward. The aircraft reached a maximum altitude of 2 895 ft and the load factor reached a negative g-value of –0.6.

2.2.5 The aircraft then entered a rapid dive, and the speed increased. Different warnings were given of ground proximity and the command of pull up by the aircraft systems, but not registered by the crew. The A/T reduced the trust from 98% N1 (full power) to 45% N1 (idle power). The negative pitch reached a maximum value nose down of 49o. Up to this time the First Officer had been somewhat passive and confused. Now he acted as an active and co-operative crew member and asked: “What are you doing” and next, he called out: “Pull up!” - “Pull up!”. Both pilots pulled back on their control columns, and the aircraft, after reaching a maximum speed of 251 kt, recovered from the dive with a clearance of 321 ft (radio height) over the north end of the runway 01L. During the pull-up the load factor increased to positive G-value of 3.59."

What common effect causes some disoriented crews to push to 49o nose dow?

Complete panic with both pilots frantically pulling and pushing at the same time and neither "seeing" what is happening on the ADI? Blindly following inappropriate FD commands is also a common factor in incidents of this nature. Have seen this in the simulator where there is no SI.

Just wondering why "speed tape confusion" does not get any mention in these, unfortunately recent, discussions.
Together with a reduced/negative G confusion and the improper trimming technique (explained in the Rostov report).

In other words, combined with a high-rate trim with flaps out and possible TE auto-retraction, the pushovers once the brain trails behind the A/C are almost inevitable - in the absence of visible horizon if you are a human.

Stuka Child, thanks.
That was the specific incident, however we should not jump to any conclusion just because there appears to be many events. Aircraft type specific, rarity of event, different training, … but no smoke without fire.
- - -
Accidents usually involve a combination of issues, different aspects of disorientation - conflicting spatial senses. The stick force issue might contribute via the points in OP #1, what pilots feel in relation to what is seen - IFR manual instrument flight. This aspect of force can be demonstrated / experienced in simulators; - to what extent do pilots use stick force in forming the ‘big picture’ when manoeuvring aircraft.

From an old 737 (classic?) incident: fuel cross-feed balancing selected, overlooked by distraction, route change, the AP had difficulty in turning. The crew disengaged the AP, falsely judged roll control with respect the mis-trimmed, unbalance lateral stick feel, the aircraft rolled thought 90 deg with crew holding zero force - stick offset creating roll. N.B. lateral trim may not exhibit the same characteristics as pitch trim.

Pilots may be unfamiliar with situations which require flight with offset stick force, having to ignore the haptic sense and concentrate on the instruments - vision is the primary sense.

Golden, a potential ‘common’ contribution is misuse of trim in those aircraft with susceptible systems; see link in # 27. The amount of nose down is not important; the issue is that the nose is pointing down, and why.

e.g. a GA involving a high nose up pitching moment - low engines, max thrust. If the compensating stick forward control continues to be held forward, and still trimming with the expectation that trim will result in zero stick force, then it is conceivable that over correction will continue to lower the nose and the trim positioned nose down, effectively controlling the aircraft. These aircraft require that the stick is moved towards centre to identify in-trim conditions, whereas other trim systems will reduce stick force at the existing control position.

A very strong possibility that this happened in some of them. Coupled with habit of pushing on the stick and trimming only by the tactile feel without reference to ADI it can be a dangerous combination.