Atlas Air 3591 NTSB Public Docket Opened

[Ray_Y]

Originally Posted by Uplinker

...Are there force sensors on each yoke to record who was pushing or pulling, and if so how do they work when the shear pin is in?

The FDR file in the public docket only contains Control Column Left (Pitch Control), Elev L and Elev R. All values in Degrees, no forces. So roughly assume Ctrl Left=Ctrl Right before the split, after that you need to guess from the ELEV R what the First Officer did. I was close to provide a higher resolution Graph of the final seconds, but it's difficult.

It also seems to say in the report that the Captain accidentally pressed the TOGA switch. How is it known who pressed the switch?

I remember from the hearing the most probable scenario was the F/O guarding the SPD BRK Lever as per procedure, bringing his hand+wrist close to one of the TOGA switches. Then 5-1 seconds before G/A mode those light turbulences affected the aircraft. Means they don't really "know". It's very likely to be related to the "turbulence", it's not known who touched it.

[Ray_Y]

Originally Posted by OldSowBreath

I have been a trial attorney in Texas for 40 years this year. ...
The case will never go to trial, and it won’t be a decision made by Atlas, Amazon or Boeing, et al. The insurers will settle it out of court and probably during court ordered mediation. ...

Thank you. Your post surfaced late due to your status as new user.
Means it's important to address the wealthy ones who are able to shoulder their share of payment - or is it just the insurances that pay?

[Ray_Y]

Originally Posted by Check Airman

I’ll accept that the CA may have been distracted by something else, but he wouldn’t have been writing for very long. Isn’t this the accident where the pushover was well into the negative g’s?

If I’m not mistaken, and it is, the CA would have noticed immediately. The question is, why did it take him so long to realise the pitch and the control column position were inappropriate?

A combined timeline from all sources could help judging. I can't recall all, is this sequence close to reality?
Assumed: Both miss the FMA announced G/A
The pitch up is quite gentle, short after light turbulences
The engines spooling up are the first real alarming surprise
The CPT was busy constructing an answer for ATC and transmit it?
He wants to get an overview, but now the bloke right of him mentiones speed. He compares what he sees with what the other calls out and can't find a connection. This consumes time.
He wants to summarize and maybe act, but then bloke right of him now mentiones "we're stalling". Another seconds wasted in getting an idea. And what's those forces suddenly?
No wonder if there's a delay in response. And they didn't have that much time until they were doomed.
Final bad is, the CPT not only acted late, but lonely. Last chance gone.

EDIT: Alright, 4 timestamps:
12:38:31 G/A
12:38:37 N1>50%
12:38:47 Pitch going Through -20° ,
12:38:50 Pitch ~-40; 290KN, at this increasing Speed & current Power they were most probably doomed here

Fill the gaps with the other events

[OldSowBreath]

Originally Posted by Ray_Y

Thank you. Your post surfaced late due to your status as new user.
Means it's important to address the wealthy ones who are able to shoulder their share of payment - or is it just the insurances that pay?

The wealthy ones, if not self-insured, will pay out through their primary and umbrella insurance policies. Often, big self-insureds will hire an insurance company and their lawyers simply to handle the negotiations. Today, almost all cases are settled within policy limits. We've gotten attuned to the quick, guaranteed payment of an agreed settlement, as opposed to trying a case that will take years to get through the trial and appellate level, and then trying to seize and sell the assets of the liable parties. I don't see any reason these cases won't settle within policy limits. As you are sure to know, all settlements will be confidential, but between the attorneys, the amounts and percentages of the pay-outs will give them them an idea as to culpability between the parties.

Last night, I got to thinking, and there are probably ten or more defendants who will be sued in addition to the big three. I don't know if Boeing makes its own in-flight cockpit instrumentation, or if that is subbed out to someone like Honeywell, TI, Garmin, etc., etc., but if so, they will be potential defendants.

[alf5071h]

Check A, A320, re #158 '… the CA would have noticed immediately. The question is, why did it take him so long to realise the pitch and the control column position were inappropriate?'

Rephrasing the question #158 with stick-force (trimmed state vs 'pitch') as the key 'awareness' parameter (poor flying technique, training or aptitude) would provide an alternative view of the crews actions.

Opposed to what the pilot 'saw', what was 'felt' and acted on; particularly with respect to configuration change - flap, airbrake, change / reversal of thrust, and speed. In addition, any change from overpowering / CWS function from the autopilot / auto trim ( if it remained engaged ).

If the pilot erroneously used stick force as a key 'awareness' datum, then the rapid and unexpected change in operating configuration (control force is out-of-phase with that required) could result in inappropriate control input opposed to what we, (with hindsight and normal attitude reference), would expect. Thus the CA had to re-establish the situation from an unexpected attitude, control, feel, condition. P2 presents a severely out of trim aircraft to CA, and continues to make inputs - crew use different datums.

View the simulated video of events, without pitch attitude, replace with with stick force - true value has to be calculated from FDR parameters and aircraft type characteristics. How would we fly with our eyes closed.

[misd-agin]

The pitch up after clicking G/A is noticeable. Seat of the pants sensations. Acceleration, positive G's, throttles advancing, engines spooling up (less obvious).

Vertigo issues come into play with advancing thrust and pitch up.

[Ray_Y]

Then have a look at this, regarding positive g's related to the turbulence before. This is from FDR data, 20 seconds around the GA mode activation
The A/P G/A mode recordings have an 1 sample per second interval, so I marked the 1 second time range where the mode must have activated
The vertical accel had 8 samples per second.
The pitch was constant at -1.4 for the last 4 seconds up to 12:38:31.6



(Analysis and Graph Intellectual Property with me)

I'm not saying the CPT behaved perfectly. But the Vertical Acceleration wasn't that clear.

[parkfell]

Originally Posted by OldSowBreath

.......Last night, I got to thinking, and there are probably ten or more defendants who will be sued in addition to the big three....

Might the family lawyers attempt to sue the FAA for issuing him with a licence and/or Boeing type rating?

Now that would set an interesting precedent.....?

[Uplinker]

Originally Posted by ehwatezedoing

Uplinker

Assuming the Captain would have recovered relatively quickly the upset then what?
What about the F/O, given another life extension under the cover of "More training!?"
Or finally fired?.......

Yes, exactly. It would appear that within a normal routine flight and the framework of the SOPs, this F/O could operate a B767. However, when things became non-normal, or unexpected, he couldn't, and his instinctive inputs to the flight controls were massively inappropriate. He appeared to have almost no SA and no instrument scan, or appreciation of what his primary instruments actually told him.

So yes; had the flight not crashed, the captain should have sent a formal report or a private word to the chief pilot: "F/O xxx reacted and behaved in a very strange way to an unexpected event. I think you need to look at F/O xxx very carefully, and you should probably check his reactions to such events and other upsets - unbriefed - in the SIM. Over to you".

[parkfell]

Originally Posted by Uplinker

.......... However, when things became non-normal, or unexpected, he couldn't, and his instinctive inputs to the flight controls were massively inappropriate. He appeared to have almost no SA and no instrument scan, or appreciation of what his primary instruments actually told him........

A question for those familiar with FAA / NTSB protocols:

Will the FO’s FAA medical records etc be subject to scrutiny with disclosure of any relevant aspects?

[Ray_Y]

Digging in the NTSB docket ...File #36! I quote segments from accident flight CVR:

Between approximately 10:54:48 and approximately 11:04:38, the captain gave the first officer advice on material and study strategies for future captain upgrade training.

Around 11:07:28, the first officer was briefly audible reading items consistent with training material.

Around 11:09:33, the first officer and the jumpseater began a conversation regarding aviation career related topics. The conversation topic changed to discuss how the first officer transitioned to the 767 airframe around 11:13:48. Comparisons were made to previous airplanes, the first officer discussed the transition being “not hard” and that “the box [Flight Management System (FMS)] is easy.”

Around 11:21:48, the first officer discussed the jumpseater’s new job at a different airline. The first officer and the jumpseater discussed upgrade opportunities, basing, scheduling and salaries at different carriers.

??:??:?? The three personnel continued to discuss the aviation industry and upgrade opportunities. The first officer made favorable comments about upgrade opportunities and the lifestyle of his current employment when speaking to the jumpseater.

[WillowRun 6-3]

Lawsuits: 1 over, 1 pending (apparently), 1 unfiled (uncertain)

Case No. 3:19-cv-00170, U.S. District Court for the Southern District of Texas, filed mid-May 2019, dismissed upon a confidential settlement (after referral to a Magistrate Judge) in mid-May 2020. Plaintiff was surviving family member of the jumpseater, Sean Archuleta, as administratrix of his estate and as next friend of two minors. Two heavy-hitter aviation litigators were involved in the case (Mary Schiavo (Pl), Andrew Harakas (D)). The court documents (available on the court's website) contain an interesting item: the release of claims includes a list of related parties, which in turn includes "George Bush Intercontinental Airport, the Federal Aviation Administration, or any other airport or aviation authorities, . . ." (emphasis added).

Not that it would keep a lawsuit by the survivors of the captain from trying what would be the novel theory of negligence or other legal responsibility for the FO's - bluntly stated - gross incompetence, asserted against other than his employer (and the chain of related entities -- and in the Texas federal court case, the lists of related entities are pretty substantial). But so far no trace has been found of any lawsuit by the captain's survivors. Maybe I missed it, somewhere.

The lawsuit by the FO's survivors was filed in Sept. 2019 in the 11th Circuit Court for the State of Florida. Not sure whether any electronic access is available to court files (or any other status information about the case).

[OldSowBreath]

Boeing will already be sued, but yes, that is an allegation the estate could make. Whether it goes anywhere would remain to be seen.

FAA will either have sovereign immunity or at least partial sovereign immunity, which would require the government's permission to sue.

All attorneys' mantra - "Plead like Hell, prove what you can."

[megan]

AVweb's Paul Bertorelli raises a a couple of interesting points, one being Atlas had extended the FO's line flying introduction due lack of progress.

You Tube

[parkfell]

This addition line training comes as no surprise. Does the source specify how many additional sectors were required?
The line training file will probably be no more than confirmation that the FO was ‘slow to learn’, together with inappropriate responses to situations.

Just why he ‘pushed’ the control column on that fateful day....?
I just wonder if there any clues during the initial stalling exercises in basic training to explain it?

The NTSB really need to explore his training from day 1 PPL / basic training if important lessons are to be learnt.
And his occupation prior to pilot training?
The clues are out there. They need to be found.

[Check Airman]

Originally Posted by parkfell

Just why he ‘pushed’ the control column on that fateful day....?
I just wonder if there any clues during the initial stalling exercises in basic training to explain it?

Hard to say for sure, but the interviews of the instructors he worked with at previous airlines (linked in the original thread, I think) don't paint a rosy picture.

[Ray_Y]

Ok, I found some time to play with Excel, learned how to master the 54,5 average samples per second timeline and combine parameters. Would someone please crosscheck, cause I had worked a lot with formatting the data and correlate events. Source is the public docket at NTSB, file number 66 containing reformatted FDR data as csv file.

Some had interest in the elevator split.I tried to determine the start and end of it. Here the two Elevator parameters in degree, starting 10 seconds before G/A and ending with the last recorded data.
Note that left and right Elevator were recorded alternating with half a second offset, so you can't precisely compare both at the same time.
12:38:44 there might have been a first and short elevator split.
3 seconds later there's a definite split, confirmed in the report.



Then I added 2 parameters, Control Column Left in Degrees (the only one recorded)
And Horizontal Stab Trim Units (Scale right of diagram).

Controls:
Seems as if the CPT held the Column at -4°, while the F/O broke off by pushing.
After 3 seconds CPT reduced to like neutral. He didn't pull, but was he feeling force from the spring connection mechanism between the 2 systems?
After another 7 seconds both pilots pulled almost simultaniously.

Trim:
Interesting. Dunno 767. Is that A/P's revenge (hasn't been disengaged) or is that pilot trim?



(Graphs and Analysis my property, all rights with me)

[Ray_Y]

Those are the G/A switches in the 767.
(Source: Video of the NTSB Board Hearing)

[megan]

Just why he ‘pushed’ the control column on that fateful day....

If he thought they were stalling you could explain it, even though it may have been an incorrect response, but why did he think they were stalling? That's the one needing an answer to me.

[testpanel]

A simple 757/767 driver here....

Not suggesting/concluding anything, but WHY was the pilot guarding the speed brake?
Wasn't it because the speed brake does not auto-stow when advancing the thrust levers (e.g. on a G/A)?

Have not searched enough but i think after the AA-965 crash in Cali, most/all 757/767 operators world-wide (were told by Boeing?? to) guard the speed brake when extended (since it does not auto stow, compared to, for example an "old" Fokker 70/100 that does auto stow)

Not defending nor attacking anybody, just think out loud...(and very fortunate this pilot was not working in my company......)

RIP