The Great Ongoing Aviation Thread (general and other)

Miscellaneous and Personal Stuff I Must Share
8723

Tail mounted cameras are non-essential systems. They can fail and there would be no consequences. Such systems still take a couple of years to conceptualize and approve. I know, I’ve taken part.

A system like you propose would be part of the pilot’s toolkit during critical phases of flight. If it fails, and the crew is delayed in taking other actions because they tried to use it, the failure is catastrophic. It’s a major design change in any airworthiness regime I’m aware of, including military.

I’m a certification and airworthiness expert and I know for a fact you’d have to involve the regulatory agencies in each country this configuration will operate. If not the agency directly, you’d have to go through a delegated individual or organisation. It’s baked into the airworthiness laws, at the federal level.

You can “imagine” it being easy, and if you come to my shop we will gladly take your money, but don’t complain when it costs orders of magnitude more than you’re thinking.

Of course it can be done, but it cannot be done cheaply or easily. The physical design is trivial, the compliance is not.

8724

So the pilots are not able to see what the passengers can see? That makes sense somehow?

8725

Often, no. Such systems are, in my experience, not interfaced to the cockpit monitors. They also aren’t watching the in-flight movie on the primary flight displays.

And adding anything to display on these monitors is an incredibly intensive and expensive certification process, it’s not as simple as just downloading a new app.

8726

ISTM airline safety has been many incremental steps over the last 75 years (or so). You could claim any single step wasn’t worth the effort or money. Yet we did it and here we are today with remarkably safe air travel. Why stop now? Why is this step one too far?

8727

Cost vs return.

But go ahead and make the business case that this enhances safety. And go ahead and foot the bill to get it done, for each and every aircraft variant operating (because they all need to be approved individually).

You don’t have to believe me that there isn’t much benefit, I’m open to being convinced.

It it will not be cheap and easy which was my position to begin with.

8728

ETA: @mnemosyne slipped in while I was bloviating. I see we’re of like minds.

The issue is two-fold. First off, as @mnemosyne says, you’re talking a LOT of money. You’re handwaving away 100% of the difficulties = expense. All of which must be gone through by law.

Kind of like drug discovery, it costs billions to invent a pill, then they can be manufactured in quantity for 47 cents each. Aviation is as safe as it is because not because we’ve festooned airplanes with cheaply designed and implemented ad hoc knee-jerks following each accident, but because we haven’t.

Once you’re talking real money per airplane, the second point kicks in. In effect there’s a fixed finite budget every year for buying more safety through engineering changes. The number is ncessarily vague, but the airlines, the public, the regulators, and the manufacturers all know it’s out there and have a decent feel for how big it is.

Unless your idea has the best bang for the buck, we can buy more incremental safety with our budget by spending it elsewhere. If your idea isn’t the best bang for the buck, switching from what it best to your idea instead is reducing, not improving safety compared to what it would have been.

It is always tempting to imagine that one gizmo (whatever it may be) would have saved this airplane and so of course it’s worth it to prevent future similar crashes. But there’re two huge assumptions there:

  • That had this gizmo already existed and been installed, it would have altered the outcome of this accident enough to matter. I personally do not believe that is true for the UPS MD-11 crash.
  • That there are enough general scenarios every year where this gizmo would alter the outcome. I personally also don’t believe that’s true.

The pros at NTSB and FAA and their equivalent agencies all over the world absolutely do perform these analyses before acting. They don’t depend on rash assumptions. Yours or mine.

Might we see action on the topic of exterior cameras with this accident being the straw that broke the logjam and moved this gizmo to the “best bang for buck” slot? Perhaps. But I for one heavily doubt it.

Right now NTSB has a list of most-wanted safety improvements for the airline industry. Has had for decades. Some things have been on that list for decades. The ones that are still languishing are the ones with cost/benefit problems.

8729

I think one plane saved is benefit enough but I can’t think we could ever know that the camera was the deciding factor. But that is probably true for all sorts of safety systems on a plane. Who can say it was that one thing that saved the day? The plane is a system with loads of working parts.

8730

The NTSB said they found fan blade parts along the runway. A WAG at this point would be a catastrophic by-pass fan failure taking out the cowling followed by an unbalanced engine quickly tearing itself off the pylon.

8731

Again…the past is filled with little incremental safety changes. So many could have been bypassed because you’d say they were iffy and not worth the cost/effort.

I, for one, like our well regulated airline industry and have confidence when flying today because of it.

YMMV…maybe your plane ticket would be a bit cheaper without all that regulation but would you be as confident you would arrive safely?

8732

It is impossible to know but, as we talked about earlier, maybe the pilot would have rejected the takeoff if he knew his plane was truly fucked (by seeing the engine is gone with a good part of the wing with it).

Likely, even with perfect knowledge, that plane was done for and all the people on it. The important part is not leaving the airport and crashing into the city beyond it. Maybe…maybe…having a camera to see how bad things really were the pilot could have chosen to keep the crash on the airport.

We can never know of course.

8733

There are committees, public consultations, studies and tests and expert analyses and reams of data to support new regulations and design approvals with the primary goal being to either maintain or enhance aviation safety.

I am part of that system, I work with it every day. An aircraft is more than the physical plane; it includes it’s maintenance, operational instructions, limitations, etc. You can’t have physical solutions to everything, sometimes the better approach is through other means such as inspections and training. Cost isn’t the deciding factor most of the time, but of course it’s a consideration. A quick visual inspection of a gauge maybe easier than a complex sensor and reporting system that doesn’t reduce the frequency of an event either way. Prevention is worth much, much more than immediate detailed (and possibly overwhelming) information in an emergency.

8734

Has there ever, ever, been a new regulation that airlines and airplane manufacturers have supported? My sense of it is they bitch and moan about every new “needless” rule.

I might be wrong. Maybe they have embraced a new rule here and there.

8735

Have you spent a lot of time reading and participating in rulemaking?

I’ve submitted personal and consolidated (team, including OEM and third party suppliers) responses to proposed regulations that are supported and welcome, but perhaps need nuance or clarity for a broader range of applications than how the proposed text was written. We’ve also objected to some that were too broadly written as to be impractical, or which fail to consider different types of operations.

It’s a collaborative process.

8736

Nope.

And to be sure, I very much respect your expertise in this area and really like your responses. Please do not think I am just being an ass when responding to you.

Truly…thanks! (same to everyone else…the list is too long to list…read I am too lazy to make the list).

8737

198 sets of comments were submitted to the first proposal of this rule (I don’t know how many rounds it went through, I just opened one proposal and then the final). The final rule lists ways that they listened and adapted the rule in response to feedback. The process was open and transparent and will have significant impact to aviation safety. It was welcome by my peers who specifically work the affected disciplines.

8738

In this case had they stomped on the brakes they’d still have left the airport. Stopping on the runway was already impossible or they’d have done so by current procedures with current gizmos.

Now there is a difference between leaving the runway going e.g. 100 knots, crossing the highway, and plowing into the first of the buildings, versus getting barely airborne, flying 3 blocks, then settling into the buildings three blocks in at 180 knots.

The total collateral damage of the first scenario is clearly less than the second. But not game-changingly less. The accident doesn’t “stay on the airport”. At least not the airport in Louisville.

There are some few airports around where there is enough land within the airfield boundaries to run off the end at e.g. 100 knots, have the airplane disintegrate into a huge fireball, and never endanger the public on the far side of the distant fence. But those airports are comparatively rare.

8739

About the cameras…

My bizjet actually has a manufacturer’s option for a set of external cameras, the feeds from which can be displayed in the passenger area and in the cockpit. I consider them mostly useless in terms of operational benefit.

First of all, this system is considered entertainment and the manual has specific language about, “Not to be used for navigation” and such. It could provide extra (re: “advisory”) information about say, the landing gear or flap position if it were in question. If an engine were on fire up at cruise altitude I might take a glance at the feed once I was done with a dozen other things and had a spare brain cell to recall that we have cameras.

But this system would make absolutely no difference with an engine failure at takeoff, such as the UPS situation. And even in the engine fire at cruise scenario, I’d think twice about telling an investigator I even referred to the camera feed because it wouldn’t be part of the relevant procedures.

8740

I think there was a big concrete wall at the end of the runway at that airport. Not sure what that would have stopped though. That was a whole-lotta plane going fast.

ETA: I think we see the UPS plane barely limp over it as it tried to takeoff.

8741

As a general matter, hard obstacles are not permitted on airports off the end of runways. The airport (or FAA) cannot much influence, much less control, what exists or gets built later once you’re off airport property.

There was a lot of worldwide criticism about the airport facility design at Muan, South Korea that turned this problem Jeju Air Flight 2216 - Wikipedia into a mass casualty event when the same bad landing absent a tall concrete berm would have been far more survivable.

Looking at Google Maps, the end of Louisville runway 17R shows the runway end followed by:

  • About 500 feet of “overrun”, a concrete pavement area painted with a few yellow chevrons. It contains a few approach lights on short poles. Which are designed to shear off easily at ground level when hit, much like modern road signs do. The overrun is intended to be stout enough pavement to withstand an occasional excursion by a not-too-heavy airplane, but is not stout enough to routinely be taxied on. Decent bet a fully loaded MD-11 at speed would have plowed furrows through that with each gear. Most likely ripping the gear off in the process.

  • Then comes about 200 feet of grass

  • Then a localizer antenna, which looks like a red-painted comb of 14 posts, each with multiple side arms, set in a row crosswise to the runway. These posts are casting prominent black shadows in the Google image. Those posts are ~5’ tall and also frangible.

  • Then more grass and ~75 feet later there’s a blast fence. You can see by its shadow that it’s a slatted design. They’re also frangible. It’s role is to deflect jet blast from any airplanes taking off the other way, to protect vehicles on the road just behind the blast fence. Here’s some pix of similar fences: airport blast fence - Google Image Search

  • Just past the blast fence we see the two lane airport perimeter road which is used by airport vehicles.

  • Then comes the airfield boundary cyclone fence, a public 4-lane road, ~40 feet of low trees and bushes, a railroad track, a drainage canal, and the first civilian building out in public. Which was probably the same building that we saw the airplane’s rump had carved a furrow through some of the roof before (briefly) climbing high enough to clear the rest of the building.

8742

At other airports it may have been possible to keep it on the field but the end of the runway at SDF is at the very end of the airport. It would have meant running across a road and into/through a warehouse. Maybe a better tradeoff or worse depending on how many people in the warehouse.

Another 25 feet of altitude and this may have been a recoverable event. I’m wondering if maybe they should design planes with enough power for an engine out AND structural damage that adds drag. Even if it’s an emergency power setting.