Computer-assisted dependency

Looking at the AF447 crash report and the continued adoption of more and more FBW assists in contemporary aircraft leads me to suspect CY+3 militaries are in the process of becoming overly dependent on semiconductors for basic actions.
How bad is it Zig Forums?
Do NATO militaries have any contingency protocols for HEMP-tier SHTF scenarios?

Attached: F-35 cockpit.jpg (850x551 655.33 KB, 249.05K)

Other urls found in this thread:

law.cornell.edu/cfr/text/14/61.57
twitter.com/SFWRedditVideos

AF447 was almost entirely because one of the FOs went full retard and was pulling back on the stick the entire time without anyone realizing until it was too late. Afraid I don't have an answer to your actual question though.

If anything it's not enough.
Had the AF447 been automated it wouldn't have crashed.
That timeline is wrong the so called pilots are 100% responsible for it.

The only technical error that happened is the icing thing (which didn't lasted more than a minute) but the pilots reaction to that was completely disproportionate AND wrong.

If tomorrow you're on a freeway going at 130 km/h and suddenly your speedometer suddenly fall to 0 (which is a similar breakdown), do you assume it's broken or do you untie your belt and get out because it means you've parked?

In that plane crash the morons in command did exactly that, they saw their speed fall from 509 km/h to 111 in an instant and instead of assuming that was a gear malfunction (since you know… they hadn't hit a fucking wall in mid air!) they believed the plane was in free fall so they tried to gain altitude… while they were already near the max for such a plane.

The fact that airbus was blamed for it is 100% internal french aviation politics and Air France cannot burn quick enough.

Attached: The attitude of the General Staff.jpg (500x375, 167.17K)

But that's the issue.
They were so accustomed to automation that they relied on the faulty stall warning instead of simply looking at the flight instrument readouts and going from there.
AF447 was not the only accident like that, there a few others were the AP disconnected and the Pilots pulled straight up completely ignoring their instrument readouts.

Are military fighter pilots these days still trained in stall&spin recovery outside of trainers?
I can't really imagine it with FBW dependent relaxed stability aircraft like the F-16, F-22, Mirage 2000 etc. even though it's an essential piloting skill.

Stall recovery is one of the most basic things pilots are trained in. Really, while is correct in that the pilot acted improperly, I can't fully blame them all. Robert acted properly and tried to push the nose forward, but in Airbus aircraft, there is no visual feedback system with the controls. Bonin was still pulling back, but Roberts couldn't know that. So he (somewhat reasonably) came to the conclusion that something was wrong with the aircraft itself.

Retarded pilots are retarded, tech has nothing to do with it.
In the case of Air France, the one actually flying the plane was a trainee, the one in command was a desk jokey that was making up lost hours (and really had no idea what he was doing since he didn't even qualified on modern pilots posts of that type so he never actually realized the trainee was the one with the commands), the one commanding the plane was sleeping off his drunken party with the chief hostess of the night before…

You cannot take on such a clusterfuck and textbook idiocy and present that as a example that "tech makes bad pilot" or whatever your point is.
Bad pilots are bad pilots, there will always be some but Air France has a company culture that encourage such layabout attitude.
Are you fucking kidding me, yes there are, there are 3 OF THEM, one per seat 2 shared.
Is just that since they had a speed error on ONE of the FOUR speed indicators (which is what disabled auto-pilot) they decided that all the instruments were faulty!

Apologizes, I read "the Airbus side stick controls give little visual feedback and no sensory or tactile feedback to the second pilot." and misread it as no to all.

As is the rest of the western populace.

Holy shit I hadn't heard of af447, some retard basically stalls it for 30000 feet. Too bad about the passengers, but that pilot deserved to die and everyone who trained him or let him fly needs to be shot.

Just to add, it's impossible to shield aircraft against EMP.

EMP shields are essentially two layers of copper tubing over every single wire, with insulation between them. Copper acts weird in presence of a magnetic field, almost repels it, when it's properly arranged. The problem is that this arrangement would add a metric tonne to a fighter.

And even then, radar and radio would be exposed to the burst.

the future belongs to those with no electrical systems

These guys shouldn't have been allowed in a Cessna, never mind a passenger jet. Entry and recovery of stalls, spins, and spirals is one of the most basic things you learn in pilot training. And they fucked it up in a way even a ground school trainee still waiting for his first time in the cockpit shouldn't fuck it up.
That kind of stupid could never be cured by the presence nor absence of FBW.

I suspect these Pilots had been properly trained at some point in their lives, but after they got their jobs the airlines didn't want to waste any precious shekels on simulator exercises for the next 30 years.
A-any USAF/NATO pilot anons here?? I'm curious as to the frequency of basic stall&spin recovery refreshers, especially for large aircraft that don't maneuver all that much
T-they still do those, right?

use magnetic amplifiers or vacuum tubes instead of transistors so they are emp impervious

Infantry level of air force experience here, why didn't he point the nose downward and increase thrust at the same time as to increase speed in more ways then one? If he wants to go fast why is he decreasing his odds of success by going up? If he wants to gain altitude, why doesn't he feel the g force of him going down to determine altitude based on his last position? If he has not felt his per peer tingle, hes close to where he was, if he did, OH SHIT NIGGER MAKE IT STOP PLEASE I HATE THAT SHIT!

Attached: e1abbc403349cc31bacc904d247d95a83d237166d39980ff2779241717dc6063.jpg (400x300, 11.97K)

Nah just more resistant, a stronger pulse would still take them out. Besides this comes again at a huge weight cost, a vacuum tube radar weighs tens of times more than an equivalent range PESA, not just raw mass but the cooling requirements as well.

By the time you're done rad proofing everything, your "fighter" is a pig that will be shot out of the air with artillery.

Fiber optics.

I don't know about Germany, but US law is strict and specific on what a pilot has to do to stay current.

law.cornell.edu/cfr/text/14/61.57

I'm pretty sure (c)(3)(iv) was because of the Air France incident.

that seems pretty standard to stay current, but I don't know when that was put in place. Regardless, it's basic shit and they were still retarded.

of course everything now is about as much flying as someone fucking with an iPad and talking on a radio, so maybe they're all retarded.

Plane was giving overspeed warning by airspeed indicators and stall speed indications by the control wheel vibrating.

Correct action would have been to set 40% throttle and level it according to the navball and ascent/descent indications and ignore the known conflicting information.

Not for a specific incident, no, but you can certainly use it as an example of the culture of incompetence that's affecting the entire developed world. Without all the fancy bells and whistles in the cockpit to fly the plane for them, those retards would've never even been able to get off the ground. Instead, an entire plane full of people had to die because the sophisticated technology in use was able to smooth over almost all of their mistakes, meaning that they were already in a very dangerous situation when they finally fucked up too much. If you make something idiot-proof, someone else will just make a better idiot.

To be honest, I would feel much safer travelling in a plane where the pilot only has a compass, a paper map, and a callous disregard for his own mortality than I would travelling in a modern airliner.

If I recall correctly, this is a failure in how training manuals are written. As planes have become more complex, many countries have sacrificed actual piloting skills in favor of being able to find where the fuck-up is in the manual and follow it step-by-step. Without the intuition they would gain from piloting, a pilot would look at the manual/recall what the manual said, do exactly that, and likely be in a "you're fucked" situation.

I think the US Navy and Airforce had an issue with their pilots doing this about 5-7 years ago and realized they needed to bring back some actual common sense training to get pilot's shit together. A lot of air vehicles have started including things like "take deep breaths for ten seconds" in their steps so pilots don't jump to the next step too soon and hopefully think with their head for a second before continuing to make a fatal automated mistake.

Yeah, not sure why no one mentioned these. Biggest issue is that fiber acts weird under high G-forces.

Some Airbus BS where the junior co-pilot's station somehow overrides input from the captain's stick. Bonin panicked and forgot to release his stick which kept registering full back through the entire event.

It wasn't automation's fault as much as poor flight crew management on the captain's part.

The junior pilot panicked. Your comparison to getting out of the car going 130km/h is faulty because sitting in a car you can still see the rest of the world zooming by. These guys were going by IFR because they were flying;

1) At night
2) In cloudy conditions

As says, it was more an issue of improper crew management.

Sage for double post.

Ok since you faggots doesn't seem to understand how deeply retarded the whole thing was I will do a fucking play by play.

The A330 is one of the most safe airplane ever built, there are 1398 of them and at that time besides a crash takeoff in trials (during a one engine take off trial, to be exact) it had NEVER known a crash and had like 2 major incidents (which were explained by maintenance crew literally installing the wrong parts, in full knowledge they weren't the right one and being ordered to jury-rigging them…).
And since then there was only another one in Libya (typical short landing piloting error, plague of 3rd world airlines). The A340 variant is still the safest airplane ever built (with no one ever dying in them, there are a "long haul" airliner despite being only around 400 in service, they have a stupid amount of flight hours, comparable to the total of the far more numerous 777).
AF 447 is also the worst crash in the history of Air France.

The crew arrived in Rio three days before the accident and stayed at the Sofitel of Copacabana Beach (5 star hotel), this was (and still is) the most desirable layover of the company since everything is on company dime.
There were three pilots in the plane.
The captain, Marc Dubois, 58, he was a veteran pilot, with nearly 11,000 flight hours on many different planes, more than half of them as captain. He had brought an off duty flight attendant with him for the (leisure) trip.
The senior co-pilot, David Robert 37, has around 5,000 flight hours which could seem decent but being a "company baby", it means most of those are sitting on it's butt in an airbus in auto-pilot, worse he's not a pilot of the company but a desk jockey and is doing that long haul specifically to get enough hours to keep his license. Before the trip it had been 3 months since he touched a plane.
The junior co-pilot, Pierre-Cédric Bonin, 32, is also a "company baby", with 2 936 flight hours again, those are not quality hours, most of them is just sitting on it's butt in a airbus on auto pilot. In fact it's highly unlikely he has ever actually "flown" an airliner since flight school. He also had brought along his wife for the (leisure) trip.

That's the background that should give you an insight to what the company culture at Air France is…

The events are:
The junior pilot is piloting, the Captain is co-piloting.
The captain is CLEARLY out of it (answer the wrong radio calls to the point ATC give them another frequency, doesn't read the weather report properly)
There is a storm coming, the junior copilot is worried, anxious, makes bad jokes and then talk about the "dead zone" and want to try getting above the clouds (despite the fact it would get the plane beyond the safeties margin), the Captain doesn't clearly shoot down the idea and just say a polite "we'll see" (while clearly not planning to do it, being a very experienced pilot he probably knows a bit storm turbulence is nothing).
Then get up to go to sleep openly joking on the fact: HE ONLY SLEPT ONE HOUR THE NIGHT BEFORE.
He then says he leave the junior-copilot in charge, the senior-copilot comes in and ask who will land the plane, the captain then ask if the junior-copilot has his license (NOT AS A JOKE§§§) because he's realized he doesn't know…
The junior-copilot just says "yeah". The senior co-pilot seats in the co-pilot seat. The way the conversation goes… all the three men can perfectly think they're the one in charge.
They go into the storm, the junior copilot, that was uneasy before, is already in full panic sounding off with exclamations over nothing and bitches again about not being able to climb (despite the fact it would change nothing).
The junior copilot suggest they reduce speed, the senior copilot approves.
The ice freeze the pitot probes. The speed indicator marks a silly slow speed.
However, both co-pilots DON'T NOTICE IT.
Why because the alarms are ringing off and the plane is switching to manual.
The throttles automatically switch to "real" (engine trust) speed mode and also display the GPS speed, the plane stops compensating automatically, however the plane is steady and fine. I no command are inputted the only way it crashes is around Paris, when out of fuel!

In total panic, the junior copilot makes very brutal left right maneuvers and voluntarily and with insistence pull the stick towards him (not in a constant manner, but really pumping on it).
And will do so until they hit the water.
In the 6 seconds he has taken real command, the plane is already at an 11° climb which all the instruments are showing properly, and which is enough to trigger the stall alarm.
They now realize that the relative (not the real, nor the GPS) speed indicators are screwy.
The senior co-pilot is doing his co-pilot job, reading out out loud the various indicators, in doing so HE REALIZE THEY'RE CLIMBING TOO FAST.
He warns the junior co-pilot several time, which acknowledges him… but doesn't actually stops climbing and settles for 6° climb, but they're now at 36000ft dangerously close of the 37000 ceiling.
The senior copilot is busy calling back the captain, the junior co-pilot once again, put the plane this time at a 13° degree climb.

Once again: Stall alarm, this times stays for good, the plane is starting to shake displaying clear physical indication of a stall but keeps climbing by inertia.
A WHOLE FUCKING MINUTE.
At that point the shaking has freed the probes of ice and there is nothing wrong with plane… except it's at an angle beyond what it can do and is still climbing despite being well over his max authorized altitude.
WHICH THEY DO NOTHING TO ADDRESS.
They hit 38000ft and what could have simply been addressed by pushing the nose down to a 0° angle becomes a true free fall, which they properly stabilize using their perfectly working instruments, a very clean very flat free fall at a 42° (!!!) angle.

So yeah you're right, I'm wrong when I'm saying: getting out of the car going 130km/h when your speedometer broke down.

It's more of driving on the other line into incoming traffic, then colliding with a truck, then getting out of the car, while it's being projected in mid-air.

The only way you're getting such a catastrophe to such a benign problem is by having retards in the seats.
Also you do realize that that type of probes was on the hundreds of A330 that flown for 15 years…
Meaning them freezing over probably happened every-time one flew into a storm at 35000ft.
Meaning maybe tens, maybe hundreds of thousands of flights likely had the issue and nothing happened to them!
How did those other pilots managed to survive such a big technical breakdown???

All reports blamed the pilots, even in the Air France own ordered expertise they blamed the pilots. Air France lawyers so far have managed to get them removed from the evidence on technicalities, but the new one (n°5) the court ordered is even more damning as the ones before.

The plane worked perfectly as intended, the auto-pilot worked as intended, all the alarms warned properly of the various symptoms for which listening and reacting in a textbook like manner to ANY of them would have prevented the crash, Airbus had amply warned that it was a known issue, easily identifiable and very easy to compensate and had given Air France the guidelines for it (which they had turned into policy… but forgot to actually spread it to the crews), etc…
The trial is gonna be this year and I fucking hope Air France gets what has been a long time comin

Holy shit. I thought the US was lax.
How well is the pay for Air france?
In China it's a 300k USD salary if a white man wants to join any airline usually as long as you dont sperg autism.

For a basic (co)pilot around $200k.
Average is around $250k.
And that's salaries, not the bonuses and the considerable (as already clearly seen) perks.

Wow. In China I've seen 500K with bonus for new pilots recently but they work you hard as fuck.

That reminds me. Every once in a while on 2ch we used to have the crash threads. There we would post chilling tales of immense retardation of the pilots that disregard instruments and flight control and carefully guide their planes into the ground. There was this iconic phrase heard on one of the black box recording, "please don't kill us!" and "Andrey don't panic!". Also the iconic "kurwaaa!" mid-crash. Those were the times, now we don't do it anymore.

Attached: kurwa.webm (640x480, 1.93M)

You can make microscopic vacuum tubes, the same way you make microscopic transistors. And they actually switch faster than semiconductor transistors, too. The problem with semiconductors is that electric breakdown generates broken paths with very low resistance, effectively shorting the unit. Then it overheats and explodes. For obvious reasons that doesn't happen with vacuum tubes and while you can overheat the traces to the point of silicon casing exploding, that would be nigh impossible, definitely not with a single burst.

Have we started the fire?

Attached: B8393A68-383A-4804-B7D9-C90E94457403.jpeg (960x720 1.45 MB, 90.15K)

What?