Care to back that up with some statistics and examples?

 

Aeroflot Flight 593 - Flaw in Airbus FBW logic causes partial autopilot disconnect with zero indication to crew.

 

Air France Flight 447 - Flaw in Airbus FBW logic fails to provide indication of pilot input.

 

Airbus Industrie Flight 129 - Flaw in autopilot causes crash.

 

I could probably find even more... but the point remains... with FBW the pilot says "do this" and the computer says "do that"... whether they are they same "thing" or not is not guaranteed and there is no method for the pilot to override the computer's decision.

 

FBW is fine... but it relies on computers being smarter than humans... which is not always the reality of the moment.

FBW was Airbuses hopefull solution to incompetant pilots. Since that didn't work too well (think AirFrance), their next fly by wire will involve pilots sitting in a cubicle on the ground and sending control commands by Western Union (wire) to the aircraft. :rolleyes:

Aeroflot Flight 593 - Flaw in Airbus FBW logic causes partial autopilot disconnect with zero indication to crew.

 

Air France Flight 447 - Flaw in Airbus FBW logic fails to provide indication of pilot input.

 

Airbus Industrie Flight 129 - Flaw in autopilot causes crash.

 

Flight 593 was an A310, there was no FBW as we know it in the A310 series.

 

Flight 447 was not a FBW issue, the FBW worked exactly as it was supposed to. I have posted a number of responses on the cause of the AF447 crash, FBW was not one of them

 

Flight 129 was a test flight where a flaw with the auto pilot not the FBW that arose. The problem arose as they were pushing the boundaries of what the aircraft was capable of by trying it out with extremely aft Center of Gravity.

 

FBW the pilot says "do this" and the computer says "do that"... whether they are they same "thing" or not is not guaranteed and there is no method for the pilot to override the computer's decision.

 

They are the same thing and the pilot has the ability to override anything he wants to, but the fact of the matter is, that a pilot will almost never feel the need to over ride what the computer thinks is a good idea, as it almost always is. In fact, I can't think of a situation where a pilot would want to over ride the flight envelope protections. The computer in FBW aircraft is not some intelligent being, it takes in the control inputs, processes them and sends them to the control surfaces, it is an inanimate object.

 

If you have specific questions about the workings of FBW or Airbus Flight Envelope Protections I'd be more than happy to answer them for you, but please don't start posting misinformed opinion on here as fact.

 

Regards,

Ró.

I know you're an airbus fan... that's fine.

 

I also know political b.s. when I read it. Air France shouldn't have crashed... it's primary cause which Airbus fans refuse to admit: The concept that two pilots can input pitch commands and that there is zero indication of what the computer actually does is a flaw. Period. Bad design, going to get more killed sooner than later.

 

The Russian Airbus shouldn't have crashed either. Once again, zero indication to the flight crew as to what decision the computer made. Bad design. Period.

 

As for the test flight... the autopilot was clearly willing to take the aircraft into a stall condition... which it shouldn't ever do.

 

I realize you don't like me saying bad things about Airbus... but as far as I'm concerned... it's like driving a car with a blindfold and expecting nothing to go wrong.

Well, you're right that FBW is not some "HAL", but the reality is that the control laws are fairly complex and the gains and feedbacks are are adjusted based on various sensor inputs and flight phase. The pilot needs to be aware of how this happens. I don't agree that a computer is "making a decision". It takes inputs, applies a control law, and generates outputs.

 

scott s.

.

I know you're an airbus fan... that's fine.

 

I also know political b.s. when I read it. Air France shouldn't have crashed... it's primary cause which Airbus fans refuse to admit: The concept that two pilots can input pitch commands and that there is zero indication of what the computer actually does is a flaw. Period. Bad design, going to get more killed sooner than later.

 

The Russian Airbus shouldn't have crashed either. Once again, zero indication to the flight crew as to what decision the computer made. Bad design. Period.

 

As for the test flight... the autopilot was clearly willing to take the aircraft into a stall condition... which it shouldn't ever do.

 

I realize you don't like me saying bad things about Airbus... but as far as I'm concerned... it's like driving a car with a blindfold and expecting nothing to go wrong.

I am not an Airbus fan, I am an Airbus pilot, I've been flying them for 19 years at this stage. I'm also a Boeing fan if that's your definition having worked with them for the first 4 years of my career.

 

I'm a fan of pretty much any thing that has two wings and can guzzle fuel like the new time.

 

I agree that AF447 should not have crashed, it was a tragedy and was very easily avoidable by having competent, well trained pilots behind the controls. The fact is that in AF447 the computer did exactly what the pilots told them to do, unfortunately they told it to do the wrong things. If I got into a stall in a B737, B767, ATR72 etc and pulled back on the stick the end result would have been the exact same.

 

The Russian Airbus had zilch to do with FBW, it was disregard of SOP and a poor pilot machine interface, nothing to do with FBW as there was none on that particular aircraft.

 

The auto pilot in the test flight case was obviously not meant to do that, that was what they were testing. The purpose of the test was to examine how the autopilot would react in that situation. The pilots should have taken control earlier and rescued the aircraft if possible. Again though you seem to be confusing FBW with the Auto-pilot.

 

You are free to say whatever you want about Airbus aircraft and I take constructive criticism on board. There are many many things about Airbus aircraft that I don't like, or find annoying, or think could be improved, and I'll put my hand up right now and admit that. I'm no fan-boy, but what I do expect are reasonable arguments about the subject at hand, with proper facts and not just opinionated trolling.

 

I fly in Airbus aircraft day in day out and I have no issues with their safety, I've also flown Boeing aircraft day in day out and had no issues with their safety. At the end of the day I will fly in whatever aircraft I'm paid to fly in, because both are extremely safe and both make great airplanes.

 

Regards,

Rónán O Cadhain.

Good debate.... interesting reads.

The first FBW aircraft was actually, suprisingly, the Avro Arrow. Yes, I know it never made it to production for a variety of reasons but... the technology was in place LONG before the F-16 came along...

I know you're an airbus fan... that's fine.

 

I also know political b.s. when I read it. Air France shouldn't have crashed... it's primary cause which Airbus fans refuse to admit: The concept that two pilots can input pitch commands and that there is zero indication of what the computer actually does is a flaw. Period. Bad design, going to get more killed sooner than later.

 

The Russian Airbus shouldn't have crashed either. Once again, zero indication to the flight crew as to what decision the computer made. Bad design. Period.

 

As for the test flight... the autopilot was clearly willing to take the aircraft into a stall condition... which it shouldn't ever do.

 

I realize you don't like me saying bad things about Airbus... but as far as I'm concerned... it's like driving a car with a blindfold and expecting nothing to go wrong.

 

The test flight you are talking about, is that the A330 involving Nick Warner? (Lovely man, former RAF test pilot).

 

There were a lot of mistakes in that flight and Airbus got panned for performing the engine failure on autopilot during an actual take off. Boeing does the test either at altitude or "on paper". The issue was that the aircraft was empty, performed the test brilliantly, landed, and when they tried again, they neglected to select flex, they put toga instead. When they pulled the circuit breaker to simulate engine failure the aircraft, at max power on one engine, rolled violently. Warner decided not to take over as the data was being sent to a telemetry station. By the time he did, it was too late, not enough altitude to recover. I would hardly blame the fbw, this would have resulted in a crash even in a DH Dragon!

 

I realize you don't like me saying bad things about Airbus... but as far as I'm concerned... it's like driving a car with a blindfold and expecting nothing to go wrong.

If that was the case, there would be Airbus crashes on a daily basis, wouldn't it? :Nerd:

If the buses are so "bad", why do you think airlines keep on buying them? & why is Airbus safer then Boeing statistically?

I know you're an airbus fan... that's fine.

 

I also know political b.s. when I read it. Air France shouldn't have crashed... it's primary cause which Airbus fans refuse to admit: The concept that two pilots can input pitch commands and that there is zero indication of what the computer actually does is a flaw. Period. Bad design, going to get more killed sooner than later.

 

The Russian Airbus shouldn't have crashed either. Once again, zero indication to the flight crew as to what decision the computer made. Bad design. Period.

 

As for the test flight... the autopilot was clearly willing to take the aircraft into a stall condition... which it shouldn't ever do.

 

I realize you don't like me saying bad things about Airbus... but as far as I'm concerned... it's like driving a car with a blindfold and expecting nothing to go wrong.

 

lol maybe you should have read Ronan signature think most of us know he is a rw Pilot of the airbus series, that why we take note of what he imputs in here

I often take the example from the automotive sector. Just because it's know to a broader audience. Writing in sim forums.

 

Most of us may already be driving cars which augment (I guess that's the word) what you put in. Those cars are able to take over a lot of control. Full authority. But even old ones bind you to the logics of the systems since you e.g. don't have four separate brake pedals, but one. So you can't alter much on which wheel receives how much force, you are stuck with either the old school cross-feed setup or a modern system, taking care of stability. Just one example.

 

This doesn't actually happen as transparent as it does on a modern plane, giving you altered cues on the flight instruments or even warnings that a mode has changed/degraded. Or even giving you a heads up for approaching some degraded mode. On a modern car, all you have may be a 'sport' switch and a small sign summarizing that either the anti-lock brake system took over, or the stability system or both. Not to mention the stuff on traction control and, on the later things, torque-vectoring, for instance altering it's presets when you pull your trailer. Automated that is. Sensor stuff.

 

So, while we are driving around with the safe guards and while some of them, even on the sports cars, are always on, we raise concerns about a protection looking for the flight envelope on airplanes. Of course, that's the same envelope a good pilot and crew also is concerned about, right? Turns out, after reading reports on accidents and crashes with high losses of lives, that the human being has some limits. Limits it sometimes seems to forget about.

 

The industry, for some reason, has left the path of improving the pilots and went straight to improving the technology. While I don't doubt that this is a hot topic, the fact is that the WWII pilots for example may not have been better or worse than modern trained folks, but their machines are worlds apart.

 

It's about picking the path with the greatest potential and, as some of you may have experienced, one does not simply tune a human being to be twice as intelligent or twice as aware or twice as skilled. This is especially true when it comes to a short amount of time. But, you can develop tech going way above twice the performance and twice the safety of even its direct predecessor.

 

 

There's a another factor. Stress. Add that to any pilot of your choice and leave out the hero stories the media sets up. Stress is able to convert any 'predictable' human behaviour into something being more closer to irrational. Think drill helps? Well, drill repeats patterns, as tech does it. While tech can adjust to certain circumstances, the drill may fail when the circumstances don't match the 'drilled' situation.

 

To give an example. I think they had to alter the reaction to an engine fire to some extent as some crews did tend to shut the hot pod down too soon. Leaving them with a dirty setup (gear, flaps), at low altitude and. now, a seriously degraded performance. Not to mention the unlikely but actually happened event of shutting down the good engine. :unsure:

 

So be careful with the drills. Those are good for clear 1:1 situations but fail or sometimes cause more trouble in the ones needing a flexible response. Psych talks. ^_^

 

So, yes, there are pilots taking off on the wrong runway, landing in the same way or wanting to clear obstacles with the spoilers deployed. There also is the reasonable but sometimes 'fighting physics' skill to get out of a situation by all means, pulling on a stick, sometimes even against a stick shaker and, depending on the plane, stick pusher.

 

They (should) know that they are doing it wrong by pulling back at low speeds. It won't save them, it may even aggravate the situation a lot. But the are pulling back because.. they fear the impact. Are those bad pilots? No, but they are human beings in the first place and while I don't doubt that one can produce stress levels in a simulator, the ones being reached on the real thing, approaching a stall or 'just' losing all engines is.. different.

 

By this, the same well trained and highly skilled pilot, flying safely for years, is able to react totally wrong. He also is able to see that the fuel gauge is faulty and that the ice warning is wrong. So, we still design him to be responsible and the ultimate decision maker. If he gets out of the flight envelope for whatever reason, we may not be able to trust him any longer though. Hard limits.

 

Remember the automotive paragraph? Does are running hard limits too. Even with the ESP off, most of them come online when you hit the brake. It senses that you have reached your limits. So to speak. While the target audience is different and can not always be called 'rated for that vehicle', the same human limits apply. And the same stress when the air gets thinner than you like it to be.

 

 

Back to the FBW tech. As folks have pointed out. In theory, it means that cables or hydraulic lines got replaced by signals. You can run FBW analogue (think Concorde) and you can pick a digital setup. The mix may also be present.

 

So, apart from a different way of transporting steering commands, there is an option to intercept and augment them. This can, as pointed out , for example help if you are flying an 'unstable' airplane, perhaps just because it reaches its CG limits on a regular basis or because you've designed the control surfaces to cause rather aggressive and therefore quick changes. At low speeds, this may be ok, at high ones, the same stick deflection may cause too rapid changes. A panic stick movement may therefore cause structural damage or total loss of control.

 

Once you are just performing what the pilot commands, you can differentiate between e.g. the actual stick deflection and what it causes on the control surfaces. The pilot will always receive the same response from his plane, while the control surface movement is altered by the computers.

 

And on protections? Well, they are not exclusively bound to FBW tech. You can and will see low speed protections working on more conventional planes too and you can also see that they e.g. take care of the flap speeds and settings to some extent.

 

Same for the augmentation. A MD-11 LSAS performs such tasks with a more conventional control setup. The interesting thing is why this is needed. Well, too much for this post.

 

So what could be a general benefit of looking at a FBW design? Is it weight, is it cost, is it the ability to introduce more radical designs or even those protections, just in case? Well, all of the above, and none. The new implementation of FBW may not be cheap. An aircraft redesign (think 737) just because you now want it to fly without those cables may not be justifiable.

 

GA planes may see FBW very late. I saw that they are testing DA42 setups but the average plane may not show a benefit in efficiency for example. And since the cost factor is the main one, not many manufacturers may aim for small FBW planes instead of the well developed and cheap to produce classic layout.

 

Fun fact, Garmin offers some 'augmented' conventional control setup for the King Air. Garmin ESP™ Electronic Stability and Protection system. Check it out! http://www8.garmin.c...SPSpecSheet.pdf

 

 

So what is FBW? Another way of steering your plane. As pointed out by others, you may actually not feel a difference. But you can alter a lot just by altering the software. From the Boeing techs, they've e.g. aimed for giving the 787 the same feel as the 777. This is not only achieved due to mechanical setup, but also due to how the software in between reacts to your inputs.

 

I guess the code kids love FBW in general. ^_^ At the same time, it seems that not all folks are convinced or even follow some fears which, in my eyes, aren't always that reasonable. :mellow:

When I was really young it was a paper airplane hung from a taut string that went from the top of the room to the bottom of the other side. When I was a bit older it was a flying fox. When I was older still it was a control line plane. Then I got a bit older and found myself under a parachute, that was a bit like fly by wire too.

 

The only thing I haven't done fly by wire is the real thing... Falcon BMS fills the void there.

The FBW of the 777 is very well explained here:

 

http://www.davi.ws/a...book_Cap_11.pdf

 

Airbus FBW Flight Control Laws are very well summarized here :

 

http://www.airbusdri...bus_fltlaws.htm

 

 

Guy

Lets not forget, our sim is pure fly by wire. ^_^ USB FTW!