ATR, Nepal, Jan 2023

[Gabriel]

ATR levers.jpg (642.99 KiB) Viewed 1321 times

[3WE]

[Gabieee’s quadrantphoto]

[3BS muscle memory hiccup comment]

Maybe some truth there.

Still, I can’t believe that it’s a simple “move to the next notch.”

And Evanie will cite that CRM generally calls for “30 degrees selected and indicated” callouts.

[3WE]

Avhearaldie gossip of pilots wearing big gloves…

[elaw]

It looks like those condition levers have buttons in the middle that unless pressed, would prevent moving them from certain positions to certain other positions. Does anyone know if that's the case?

If so, and that's not enough to discourage confusing them with the power levers, it seems it it would take some really special technique and concentration to move both of them at the same time.

[3WE]

It looks like those condition levers have buttons in the middle that unless pressed, would prevent moving them from certain positions to certain other positions. Does anyone know if that's the case?

The old Evan would have read the ATR-72-212A QRHFCOMPOH and cited the feather checklist to learn when and how to select feather; however, it must be short on FBW and acronyms and therefore not_interesting.

I went to PPRUNE and did not_note any ATR jocks saying how to feather and how easy or hard it was to do…

Parlour talking has become more difficult in recent years.

[Gabriel]

It looks like those condition levers have buttons in the middle that unless pressed, would prevent moving them from certain positions to certain other positions. Does anyone know if that's the case?

If so, and that's not enough to discourage confusing them with the power levers, it seems it it would take some really special technique and concentration to move both of them at the same time.

The levers are very different: Different shapes, different colors, different heights, different operations (squeeze latch under the handle vs lift the full handle).

However, we need to understand 2 different kinds of brain fart: Bad aim (you aim for the right stuff but you miss) vs aiming for the wrong thing.
The features I mentioned above are very powerful to stop the first kind. If your brain is looking to lift a flap-shaped handle and your hand touches 2 square knobs with latches underneath your brain will jump at once. But if the flying pilot says "flaps 30" and the copilot's brain says "where are my square levers with the squeeze latches underneath", it will never detect the mistake. At least not initially. If the airplane's (or car, or boat) reaction is crisp and clear, at that point the pilot may connect the dots and see if it is because the thing he just did, and then realize.

When a 747 took off, the PM called "positive climb" and the PF called "gear up", all the difference in the world didn't stop the the PM from grabbing a flap-shaped handle and moving it forward, instead of garbing a round handle on the instrument panel, miles away from the correct one, and moving it up. It took the stickshaker to activate for the PM to say WTF and check what he had just did, discover his mistake in horror, and correct it, all the while the PF masterfully treaded the needle by lowering the nose just enough to prevent the stall and prevent contacting the ground at the same time, managing to keep the AOA at the onset of the stickshaker.

Jumbo levers.jpg (138.37 KiB) Viewed 1303 times

[3WE]

Quote= Gabriel [Blah blah blah flap handles look like flaps]

Quote=Averaldie Parlour Talkers [Indian pilots often wear heavy gloves]

3BS continues to be redundant: Hopefully flaps aren’t pull up and back in contrast to feather: pull up and back…hopefully there’s a bit of an interlock and bright flashing red light to feather a prop. (I know there’s a warning- it’s been alluded to). BUT still looking for an ATR jock to say, really easy TO miss or really hard NOT_TO miss.

[Gabriel]

Quote= Gabriel [Blah blah blah flap handles look like flaps]

Quote=Averaldie Parlour Talkers [Indian pilots often wear heavy gloves]

3BS continues to be redundant: Hopefully flaps aren’t pull up and back in contrast to feather: pull up and back…hopefully there’s a bit of an interlock and bright flashing red light to feather a prop. (I know there’s a warning- it’s been alluded to). BUT still looking for an ATR jock to say, really easy TO miss or really hard NOT_TO miss.

It doesn't need to be easy to miss or hard not to miss for one person missing it every God know how many hundreds thousands or perhaps million take-offs. It hard to make something right better than 99.999% of the times.

[elaw]

[It hard to make something right better than 99.999% of the times.

Especially considering that that level of correctness means in a group of 10 pilots with 10,000 flight hours each, you have one full hour of totally effing up.

[Gabriel]

[It hard to make something right better than 99.999% of the times.

Especially considering that that level of correctness means in a group of 10 pilots with 10,000 flight hours each, you have one full hour of totally effing up.

Well but something like setting flaps 30 is something that each pilot only does once every 2 flights.

But still, this is something that is overlooked especially by people that is not closely familiar with human factors.

Imagine an almost ideal scenario where all pilots have identical training, skills, discipline, commitment, honesty, etc. You cannot tell one from the other. And say that some specific critical mistake is something that any of those identical pilots can randomly make in any flight, at an average rate of once every 10 million flights. No pilot comes anywhere close to flying that many flights in their lives, or 1/10 as many, or 1/100 as many. Most don't even fly 1/1000 as many flights in their entire life, so the vast majority of the pilots will not make that mistake ever. And yet, that mistake will happen once or twice every year in the US alone. And it is not because the pilots that make that mistake are any less good, more distracted, less responsible, dumber, or possesses worse skills than the rest. Because the hypothesis was that that was not the case.

I know that the real world is more complicated because we do have normal variation (as well as outliers) from pilot to pilot and from day to day in each pilot. Not all pilots are the same every day, and there is no pilot that is the same mental and physical condition every day. But we need to be careful when we judge people's performance, that we are judging their way of doing things and not one particular outcome, good or bad, that can be just the result of random chance and that could have been done by any other person.

So, in this accident, more than looking at the mistake itself of grabbing and moving the wrong level, I would compare what they were required to do, what they were really expected and motivated to do by their company (this is about safety culture), and what they did. In terms of monitoring flight parameters, stabilized approach criteria, landing checklist, crosscheck of each other's actions, etc.

Having a brain fart and grabbing the wrong lever? Can happen to any of us. Moving "the lever" but not checking that the flaps moved accordingly? Flying for 1 minute without realizing that the engines were producing no power? Or that the engines were indicating zero torque? And letting the airspeed go way down without realizing? And maybe not running the landing checklist? Or doing a formal stabilized approach assessment at a defined threshold? I am more curious about these things, and I think that there is more to be learned and corrected in that field than in blaming the stupid pilot for pulling the very obviously wrong lever.

[3WE]

[It hard to make something right better than 99.999% of the times.

Especially considering that that level of correctness means in a group of 10 pilots with 10,000 flight hours each, you have one full hour of totally effing up.

You guys are right…but as we know, this crash was the result of the crew’s cowboy monkey insistence of making a steep turn just to avoid a little tailwind.

[elaw]

...And letting the airspeed go way down without realizing?...

To me, this right here is the crux of the matter. Yeah I know AoA yada yada, but how many air accidents could have been avoided by the following:
1. Look at airspeed indicator. Note that it's reading a bit low.
2. Wait a moment (like, look at some other things).
3. Look again at the airspeed indicator. Not that it's not only reading low, it's lower than it was before.
4. Think to yourself "That's not right" and work the problem.

???

[3WE]

...And letting the airspeed go way down without realizing?...

To me, this right here is the crux of the matter. Yeah I know AoA yada yada, but how many air accidents could have been avoided by the following:
1. Look at airspeed indicator. Note that it's reading a bit low.
2. Wait a moment (like, look at some other things).
3. Look again at the airspeed indicator. Not that it's not only reading low, it's lower than it was before.
4. Think to yourself "That's not right" and work the problem.

???

Are you 3BS?

Zero disagreement here, but I'm TRYING to understand that maybe feathering the props generates very little seat-of-the-pants sensation, thus they "forgot" to follow airspeed...

[Gabriel]

...And letting the airspeed go way down without realizing?...

To me, this right here is the crux of the matter. Yeah I know AoA yada yada, but how many air accidents could have been avoided by the following:
1. Look at airspeed indicator. Note that it's reading a bit low.
2. Wait a moment (like, look at some other things).
3. Look again at the airspeed indicator. Not that it's not only reading low, it's lower than it was before.
4. Think to yourself "That's not right" and work the problem.

???

Forget AoA. You are totally right. The problem is that it is not easy for us humans to do it, and when we do it sometimes we don't do it right. We tend to focus our attention in what seems to us to be the highest priority at he time. And thus we miss that some other thing has become highest priority. Because you need to do the same thing that you mentioned for the airspeed indicator, with the artificial horizon, the ILS needles, the altimeter and vertical speed indicator, the engine instruments, the flight mode annunciator, the ball to keep the plane coordinated, all while you push buttons and move levers to configure the plane, and check their indicators to confirm that the action performed had its intended result (i.e. that the flaps did extend, that the gear did go down), and talk on the radio, and run checklists, and assess if stabilized approach criteria is met. It is not that easy. And while that is exactly what pilots do, they also make mistakes.

[Gabriel]

Zero disagreement here, but I'm TRYING to understand that maybe feathering the props generates very little seat-of-the-pants sensation, thus they "forgot" to follow airspeed...

If they were flying already with about zero thrust, then they would not feel the loss of thrust. However... the propellers much have reduced the RPM a lot, and I expect that the sound would thus change a lot, I suppose that that is absolutely not normal, RPMs don't go down in constant-speed props.

[Gabriel]

Blancolirio brings an interesting point of a captain flying in the right sear and the muscle-not-memory of operating the flaps lever with the wroing hand and at the wrong distance from the body. In the captain seat, he would use the right hand to extend the flaps and he would need to reach all across the center console to reach the flaps lever. On the left seat, he used his left hand and the flaps lever was the closest one, actually very very close to his body.

[3WE]

…we’re going to talk this in circles, forever, and where is the CVR?

And its layerS of Swiss Cheese.

Flaps…selected…and indicated.

Four eyes and buttocks should have an eye on airspeed. Although I can almost imagine it taking 30 seconds to REALIZE SOMETHING is wrong, followed by 30 seconds to FIGURE OUT WHAT is wrong…and I also think they were about out of altitude.

Do you feather props with the same motions as the NEARBY flap handle (you French aeroengineers did WHAT?)

Evan even has a point- were they doing a HURRIED visual approach/tight pattern?

[elaw]

As if this accident wasn't sad enough already: https://youtu.be/iBpHwhpZMA0?t=284

[Gabriel]

As if this accident wasn't sad enough already: https://youtu.be/iBpHwhpZMA0?t=284

It is clear that that guy watched Blancolirio's video and didn't bother to read the 3-pages interim report. If he did he would totally understand why the investigators issue the recommendation they issued.

[3WE]

I found this:

375DE63B-990C-4501-BBD3-1880288233F4.jpeg (217.31 KiB) Viewed 1252 times

Do I have to read why a flight path analysis is needed? Can you tell us in plain English because that exact word choice reads wrong.

[3WE]

Quote = Gabieee
It is clear that that guy watched Blancolirio's video and didn't bother to read the 3-pages interim report. If he did he would totally understand why the investigators issue the recommendation they issued.

Quote = The Report
Due to the shortened final approach leg for runway 12, in both the flight the stabilization criteria for a visual approach could not be stabilized at the height of 500ft AGL.

Yeah, sure...BUT

Where is the data that these folks are 'regularly' flying patterns that are just a little bit too tight for Evan? Do we really need a broad study that getting tight and time-limited leads to phugoids?

No argument that it's a factor here, BUT there's a few other issues along with it. Not_unlike Hui Theiu Lo and FLCH...Evan wants autopilot training, I want watch-the-airspeed training...I want an ATR where you can't casually reach down and feather both props while turning final. We need some extra guarded switches on "feather"...and more regulation.

Damn, Evan's really going to fixate when he sees that they declared the approach too tight/not_stabilized.

[Gabriel]

Quote = Gabieee
It is clear that that guy watched Blancolirio's video and didn't bother to read the 3-pages interim report. If he did he would totally understand why the investigators issue the recommendation they issued.

Quote = The Report
Due to the shortened final approach leg for runway 12, in both the flight the stabilization criteria for a visual approach could not be stabilized at the height of 500ft AGL.

Yeah, sure...BUT

Where is the data that these folks are 'regularly' flying patterns that are just a little bit too tight for Evan? Do we really need a broad study that getting tight and time-limited leads to phugoids?

I agree. Let's wait for the final report, of course, but I don't think that there were "official" external factors that forced them make a tight approach. I suspect that they just elected to keep it tight.

Now, "not official" external factors may include airline culture. I am curious to see what was the stabilized approach criteria in their procedures and how serious was the company regarding that the stabilized approach criteria MUST be respected (rather than being something that we need to put in the procedures so the local FAA doesn't bother us). In this plane alone, and in a very short of time, there were 2 landings to this new runway that apparently didn't respect that criteria. And I say apparently because it is not clear from the report if the "wings level by 500ft" criteria that they mention was actually in the procedures of this airline and was actually applicable to this type of circling approach.

Notwithstanding the above...

BUT there's a few other issues along with it. Not_unlike Hui Theiu Lo and FLCH...Evan wants autopilot training, I want watch-the-airspeed training...

100% agreed.

I want an ATR where you can't casually reach down and feather both props while turning final. We need some extra guarded switches on "feather"...and more regulation.

Once again, making the difference between these levers more obvious can help when the pilot aims for the right lever and misses, not when he aims for the wrong lever. Gear, flaps and speedbrake handles had been confused several times and not because they were not different enough or separated enough, and that have killed people once or twice.

I like your idea that when you select flaps 30 a voice says "flaps moving to 30" and when they reach there "flaps 30". So if you intend to extend flaps 30 and for some reason your brain decides that to accomplish that you need to pull the condition levers, hopefully the airplane calling "left propeller feathered, right propeller feathered" will be a wake-up call. That is, the machine doing the job of the pilot verifying that their actions have the intended effect and cross-checking each other. It is very easy for us human to make mistakes in this task. The flip side of the coin is that this will cause primmary-backup inversion issues, when the pilots will rely on the plane telling them what they did wrong, instead of that being a backup. This effect, instead of adding a slice of Swiss cheese, it replace a slice with another. So the one time that they make something wrong and some sensor fails and they don't get the message from the plane, they will have about zero chance of finding the mistake by themselves.

[Not_Karl]

Let's wait for the final report

Concur.

I want an ATR where you can't casually reach down and feather both props while turning final.

Ban ALL turns to final, problem solved.

[Gabriel]

Ban ALL turns to final.

Someone over there approves.

[3WE]

Let's wait for the final report

Concur.

I am scared that we are getting old. Please do not say such blasphemy!