To sumnarize: if it reduces the liklihood of one in a billion mishaps and fails dangerously on one in a million flights, it reduces safety by 1000x.
Because everything is already so safe, many intended improvements are like false positives on tests for rare diseases: far more positive results are false than true.
But why not have the button to save everyone if all else fails?
Sure it is a super unlikely event. But the solution is some code in a computer that runs the autopilot that is already there. Nothing special needed. The code is already written and tested to some degree and uses the stuff already built into the airplane.
If your choice is 100% death or press button and 90% chance of death why not push the button?
The code need not integrate or change any existing code. It only runs when the pilots figure fuck-it…we’re dead otherwise. Let’s roll the dice.
Again, it is like the FDA letting people take an experimental drug when they have no other hope. Will the new drug work? Will it have terrible side effects? No one knows but if you are days away from dying then why not take that chance? It need not be tested and proven safe. There is nothing left to lose.
I don’t even know where to begin.
First off, the rules don’t prohibit such a system. If an applicant wanted to design and certify one, they could, and they would submit a certification plan for it that would identify it as a novel technology which the existing regulations do not consider, and you’d begin the process of getting a Special Condition authored that would define the regulation and means and methods of compliance to get the thing certified.
I guess no one has felt it was necessary to make such a request. But in no way shape or form is is blocked or prohibited simply because it doesn’t fit into the existing rules. You know what also isn’t “permitted” per the rules? Joysticks instead of yokes. Yet they exist; via Special Conditions.
As for “the code already exists”, that’s a laughable claim. It may exist, in theory, for one type of normal category (14 CFR Part 23) aircraft flying private operations (14 CFR Part 91) but that doesn’t mean the “little code” exists for a Transport Category (part 25) aircraft flying commercial ops (121) with entirely different levels of reliability and safety assessments required. The computers are completely different. The flight control laws are completely different. The operational envelopes are completely different. The “autopilot system” is complete different.
You actually have to customize this code and fully demonstrate compliance for every individual aircraft, variant, and software configuration individually. Certifying something on a 737 means absolutely fuck all for being able to install it on an A380. Heck, certifying something on a Challenger 3500 does not mean it is valid on a Challenger 350, and they are the same variant and type certificate. Why? Because there are system and software differences and you have to prove them both.
Then you get into costs, benefits, proportional risk reduction, etc. This magic button won’t prevent crashes of commercial aircraft to any extent, because the types of crash they are meant to prevent simply don’t happen and there are other more effective ways to improve safety without this magic bit of code.
Air India crash Glad to see a bit of backtracking from this guy.
Good summary and just about the situation I called from dot one.
If there are mechanics who would like to chime in, are jet fuel pumps positive displacement or a propeller style?
There are electric pumps in the fuel tanks which move the fuel to the engines. There are 2-stage mechanical fuel pumps in/on the engines that raise the pressure to injector level. My 787 pilot manual is silent on the pump mechanisms, but the tank pumps are usually centrifugal vane type while the engine pumps are PD.
In normal ops the fuel systems and electrical systems are isolated such that any failures on one side could have no effect on the other side.
The engines will run fine at low altitude with all the electrical tank pumps switched off. Assuming a typical takeoff fuel load and takeoff thrust there won’t be (shouldn’t be?) a problem if somehow all the pumps were off / failed.
If I read that correctly the engine PD pumps should overcome any vapor lock situation.
If the problem is downstream of the pump. If somehow there was air in the lines upstream, there’s a problem.
I admit to not following any of the speculation on the Air India crash since a couple hours after the news broke. It was all too painfully stupid and everyone repeating something they think they head somebody else say with nobody, not a single talking head having the slightest idea what they were saying.
Post 7966 has a video of a pilot speculating it might be vapor lock due to 110 temperature on top of even hotter taxiway/runway temperatures and engines heating up the fuel tanks. Apparently there is a temperature where vapor lock becomes a problem. Thus the question about types of fuel pumps used.
So I guess the question is, can the PD pumps pull fuel through the axial flow pumps if they go into vapor lock. Or is the engine PD pump a multiplier that needs X PSI to start with to bring it up to injector pressure.
I can’t help but think of planes in Saudi Arabia. It must be pretty hot there in the Summer.
Without actual research on the issue I place zero credence in the idea of vapor lock.
Pumps don’t pull. They push.
Something I haven’t run across before:
The pilot of a small plane that crashed near a North Carolina airport this month had raised a wheel after landing to avoid hitting a turtle on the runway, according to a National Transportation Safety Board preliminary report.
The pilot of the Universal Stinson 108 and a passenger were killed in the June 3 crash near Sugar Valley Airport in Mocksville, officials said. A second passenger was seriously injured in the crash.
A communications operator looking out the airport office window advised the pilot that there was a turtle on the runway, according to the report released this week.
Priorities, people! You killed a person to save a turtle? Fucking idiot.
I suppose ot depends on the size of the turtle. I’ve seen one near my old office that was about the size of a trash can lid. Hitting that at speed could damage the wheel or the gear strut. I think the Stinson is a tail-dragger, so hitting something big with one of the main wheels could cause an accident.
I’m not sure the pilot was just trying to save the turtle. He may have figured it was the best option for the plane and passengers, too.
The article is short on details, but it sounds to me like a botched go-around that happened to be caused by living FOD on the runway. That it was a turtle made it a good headline.
Could have been funny if nobody got hurt. Or if it was a Galapagos tortoise.
Trying to lift off completely, or steer around, sure. Trying to pick up one main wheel at near-flying speed? A nutty plan in a taildragger.
As always, we’re all victims of the fact nobody but the pilot understands what he was thinking and everything we’re reading is filtered through multiple layers of abject cluelessness.
Stinsons are tail- daggers. My family used to own one, and it’s the plane I learned to fly in.
The Museum of Flight is displaying a MiG-21 that South African artist Ralph Zinman covered in about 70 million (?) tiny glass beads. According to the artist, who was interviewed on the morning news today, there are about 2,000 pounds of beads on the aircraft. The news report said that the aircraft had been flown by a Cuban pilot during the Angolan civil war.
For those who don’t immediately recognize that venue, it’s in Seattle.
I hear some pistons & look up…it’s not every day one sees a C-47 over the house!