AirAsia Airbus 320 flight missing

[muppet]

Reading more about it there are some keys issues.

There was a genuine fault, which was detected by the relevant system and a warning was sent. Of course the warning message simply notifies the crew of a general fault within a system. They would have known nothing about cracks in soldering.

On an Airbus all of the systems are computerised and like any computers, when there is a problem, turning it off and on again usually fixes it. This is not always a good idea while airborne though and needs some careful thought before doing it. For example, simple as it sounds, you need to consider all of the functions being controlled by that (or those) computers before satining it (them) off. Further to that, it would be very unusual to consider turing off two computers, especially when one is a back-up of the other.

In this case the captain got out of his seat and turned (by pulling circuit breakers) off and then on, two key computers. What he didn't appear to know is that, even though he reset the c/b of the first computer before turning off the second one, he needed to also reset the master pushbutton of that computer to re-activate it. So when he turned off the 2nd one, neither were working and the status of the aircraft was significantly downgraded, meaning a number of controls and protections (usually provided by either of those computers) were lost.

This still left the aircraft perfectly flyable. However, when the two computers were lost, the downgrade caused the auto-pilot to disconnect suddenly. The captain was out of his seat (to reach the circuit breakers) and the co-pilot was presented with a sudden need to fly the aircraft. Again this shouldn't be a problem, but amazingly he did exactly what the co-pilot did on the AF447. He pulled the stick back, held it there and stalled the aircraft. The captain didn't help by telling him to 'pull down'. You either 'pull up' or 'push down'. Language barriers were obviously an issue here along with the 'startle factor'.

To keep this post short, if they had done nothing, they would probably have been fine. The would have had to tolerate an annoying warning coming on every few minutes, but obviously that was better than what happened. If in any doubt, don't trouble shoot in the air.

For anyone interested, here is the final report (I haven't read it properly yet so feel free to correct any of the above): http://kemhubri.dephub.go.id/knkt/ntsc_home/ntsc.htm

[DennistheMenace]

It's usually a combination of factors that lead to the a crash like this.

You hear people say why didn't the pilot communicate any issues to the tower but this is his least priority in a dangerous system with his main objective being, aviating, then navigating and finally communicating. If you are in your car and in danger of crashing, your least concern is to call an ambulance. Simplistic anaolgy but gets the point across when explaining to some.

That is true normally Dennis.

But there are two pilots. When you suddenly can't maintain your altitude, for example, one pilot must send a distress call so ATC can route other traffic away from your position, above and below. Otherwise you may recover control of the aircraft, but risk a fatal mid-air collision. Also, when you are about to crash your car the impact is just a second or two away. When you are in an aircraft at over 30,000' the impact is minutes away.

The lack of a distress call could hint at major confusion, just like AF447, whereby the crew were unable to comprehend what was happening, but at the moment that is just speculating.

Muppet I think it's something very similar to AF447, they had minutes to send a mayday to ATC when the plane had stalled and had 3 pilots in the cockpit for the latter part of the crash. in the end though it's the least priority to them. Anyway the black boxes should be found relatively easily in this instance in comparison to AF447 which took 2 years IIRC..

Sadly it does indeed look similar to AF447 and whilst Learmont was premature he was correct in his assumption of pilot error playing a part.

[johnneycool]

Reading more about it there are some keys issues.

There was a genuine fault, which was detected by the relevant system and a warning was sent. Of course the warning message simply notifies the crew of a general fault within a system. They would have known nothing about cracks in soldering.

On an Airbus all of the systems are computerised and like any computers, when there is a problem, turning it off and on again usually fixes it. This is not always a good idea while airborne though and needs some careful thought before doing it. For example, simple as it sounds, you need to consider all of the functions being controlled by that (or those) computers before satining it (them) off. Further to that, it would be very unusual to consider turing off two computers, especially when one is a back-up of the other.

In this case the captain got out of his seat and turned (by pulling circuit breakers) off and then on, two key computers. What he didn't appear to know is that, even though he reset the c/b of the first computer before turning off the second one, he needed to also reset the master pushbutton of that computer to re-activate it. So when he turned off the 2nd one, neither were working and the status of the aircraft was significantly downgraded, meaning a number of controls and protections (usually provided by either of those computers) were lost.

This still left the aircraft perfectly flyable. However, when the two computers were lost, the downgrade caused the auto-pilot to disconnect suddenly. The captain was out of his seat (to reach the circuit breakers) and the co-pilot was presented with a sudden need to fly the aircraft. Again this shouldn't be a problem, but amazingly he did exactly what the co-pilot did on the AF447. He pulled the stick back, held it there and stalled the aircraft. The captain didn't help by telling him to 'pull down'. You either 'pull up' or 'push down'. Language barriers were obviously an issue here along with the 'startle factor'.

To keep this post short, if they had done nothing, they would probably have been fine. The would have had to tolerate an annoying warning coming on every few minutes, but obviously that was better than what happened. If in any doubt, don't trouble shoot in the air.

For anyone interested, here is the final report (I haven't read it properly yet so feel free to correct any of the above): http://kemhubri.dephub.go.id/knkt/ntsc_home/ntsc.htm

Turning something off and on again never fixes anything, it masks an issue, intermittent or otherwise, one that happened 23 times in 2014 alone.

That's shocking to say the least irrespective of the intricacies of resetting a computer system mid flight!

[muppet]

Reading more about it there are some keys issues.

There was a genuine fault, which was detected by the relevant system and a warning was sent. Of course the warning message simply notifies the crew of a general fault within a system. They would have known nothing about cracks in soldering.

On an Airbus all of the systems are computerised and like any computers, when there is a problem, turning it off and on again usually fixes it. This is not always a good idea while airborne though and needs some careful thought before doing it. For example, simple as it sounds, you need to consider all of the functions being controlled by that (or those) computers before satining it (them) off. Further to that, it would be very unusual to consider turing off two computers, especially when one is a back-up of the other.

In this case the captain got out of his seat and turned (by pulling circuit breakers) off and then on, two key computers. What he didn't appear to know is that, even though he reset the c/b of the first computer before turning off the second one, he needed to also reset the master pushbutton of that computer to re-activate it. So when he turned off the 2nd one, neither were working and the status of the aircraft was significantly downgraded, meaning a number of controls and protections (usually provided by either of those computers) were lost.

This still left the aircraft perfectly flyable. However, when the two computers were lost, the downgrade caused the auto-pilot to disconnect suddenly. The captain was out of his seat (to reach the circuit breakers) and the co-pilot was presented with a sudden need to fly the aircraft. Again this shouldn't be a problem, but amazingly he did exactly what the co-pilot did on the AF447. He pulled the stick back, held it there and stalled the aircraft. The captain didn't help by telling him to 'pull down'. You either 'pull up' or 'push down'. Language barriers were obviously an issue here along with the 'startle factor'.

To keep this post short, if they had done nothing, they would probably have been fine. The would have had to tolerate an annoying warning coming on every few minutes, but obviously that was better than what happened. If in any doubt, don't trouble shoot in the air.

For anyone interested, here is the final report (I haven't read it properly yet so feel free to correct any of the above): http://kemhubri.dephub.go.id/knkt/ntsc_home/ntsc.htm

Turning something off and on again never fixes anything, it masks an issue, intermittent or otherwise, one that happened 23 times in 2014 alone.

That's shocking to say the least irrespective of the intricacies of resetting a computer system mid flight!

Turning it off and on was the correct procedure.

This appears on the 'ECAM' which is a screen which presents the pilots with the relevant checklist when a 'known' problem is detected.

The procedure for the failure, as it would have appeared on the ECAM for all 23 occurrences, was to turn off FAC 1 and then turn it on. If that didn't work then turn off FAC 2 and turn it on.

Regarding fixing the problem. That is a completely different thing entirely. That would be a maintenance function. Pilots don't fix things. They follow the checklists, to try to restore a system if possible, and make decisions based on what systems are available to them if they can't restore them.

Now having said that turning the FAC 1 & 2 off and on was the correct procedure, cycling the circuit breakers off and on wasn't. Mechanics on the ground will try this and often it works. As you say, it may mask the real problem, as it certainly did in this case. Fair play to the investigators for tracing it to a soldering problem. But it is a function of modern technology that the goto fix is to turn it off and on again. Think of smartphones and tablets for example. Sometimes people get fixated on this and stop thinking beyond it.

[armaghniac]

It seems this is human error. If they were turning off the computer the co-pilot should have been ready to fly manually.
There is a real issue with pilots (maybe younger co-pilots) ability to fly manually on an Airbus when the computer disconnects and there might be a data problem.

[muppet]

It seems this is human error. If they were turning off the computer the co-pilot should have been ready to fly manually.
There is a real issue with pilots (maybe younger co-pilots) ability to fly manually on an Airbus when the computer disconnects and there might be a data problem.

I think this may be a real problem.

But there is a list of other stuff in this one.

* At what point does maintenance stop changing/resetting the computer and actually try to find the real problem?
* When does Airbus instruct them to stop the resets and go after the problem?
* Why does Airbus still suggest there is no need for any upset training, even after AF447?
* Why did the Indonesian Authorities not insist on such training regardless of Airbus?
* Why would an airline risk no upset training, even if Airbus and the authorities aren't bothered? (Cost is probably the answer here)
* Why did the Captain go down the road he did, meaning why get out of your seat and try out procedures heard from maintenance (who only ever deal with an aircraft while it is on the ground) while in the air and apparently without any discussion with the other pilot?
* And, as Armaghniac said, why yet again do we have a relatively inexperienced pilot who does exactly the wrong thing when faced with a sudden requirement to manually fly at high altitude?

[johnneycool]

Reading more about it there are some keys issues.

There was a genuine fault, which was detected by the relevant system and a warning was sent. Of course the warning message simply notifies the crew of a general fault within a system. They would have known nothing about cracks in soldering.

On an Airbus all of the systems are computerised and like any computers, when there is a problem, turning it off and on again usually fixes it. This is not always a good idea while airborne though and needs some careful thought before doing it. For example, simple as it sounds, you need to consider all of the functions being controlled by that (or those) computers before satining it (them) off. Further to that, it would be very unusual to consider turing off two computers, especially when one is a back-up of the other.

In this case the captain got out of his seat and turned (by pulling circuit breakers) off and then on, two key computers. What he didn't appear to know is that, even though he reset the c/b of the first computer before turning off the second one, he needed to also reset the master pushbutton of that computer to re-activate it. So when he turned off the 2nd one, neither were working and the status of the aircraft was significantly downgraded, meaning a number of controls and protections (usually provided by either of those computers) were lost.

This still left the aircraft perfectly flyable. However, when the two computers were lost, the downgrade caused the auto-pilot to disconnect suddenly. The captain was out of his seat (to reach the circuit breakers) and the co-pilot was presented with a sudden need to fly the aircraft. Again this shouldn't be a problem, but amazingly he did exactly what the co-pilot did on the AF447. He pulled the stick back, held it there and stalled the aircraft. The captain didn't help by telling him to 'pull down'. You either 'pull up' or 'push down'. Language barriers were obviously an issue here along with the 'startle factor'.

To keep this post short, if they had done nothing, they would probably have been fine. The would have had to tolerate an annoying warning coming on every few minutes, but obviously that was better than what happened. If in any doubt, don't trouble shoot in the air.

For anyone interested, here is the final report (I haven't read it properly yet so feel free to correct any of the above): http://kemhubri.dephub.go.id/knkt/ntsc_home/ntsc.htm

Turning something off and on again never fixes anything, it masks an issue, intermittent or otherwise, one that happened 23 times in 2014 alone.

That's shocking to say the least irrespective of the intricacies of resetting a computer system mid flight!

Turning it off and on was the correct procedure.

This appears on the 'ECAM' which is a screen which presents the pilots with the relevant checklist when a 'known' problem is detected.

The procedure for the failure, as it would have appeared on the ECAM for all 23 occurrences, was to turn off FAC 1 and then turn it on. If that didn't work then turn off FAC 2 and turn it on.

Regarding fixing the problem. That is a completely different thing entirely. That would be a maintenance function. Pilots don't fix things. They follow the checklists, to try to restore a system if possible, and make decisions based on what systems are available to them if they can't restore them.

Now having said that turning the FAC 1 & 2 off and on was the correct procedure, cycling the circuit breakers off and on wasn't. Mechanics on the ground will try this and often it works. As you say, it may mask the real problem, as it certainly did in this case. Fair play to the investigators for tracing it to a soldering problem. But it is a function of modern technology that the goto fix is to turn it off and on again. Think of smartphones and tablets for example. Sometimes people get fixated on this and stop thinking beyond it.

The thing is this wasn't the first time the plane manifested this problem and the first time it did, once on the ground the thing should have been fixed, not set up some sort of procedure to mask it.

I'd have thought that the aviation industry would have much higher quality expectations than to accept this bodge on an ongoing basis!

I've worked in a few different electronic industries and whilst you may get away with telling a customer to reboot the thing once or twice, but 20 odd times and you'd be told to pull the thing out and replace it, and there's no lives at risk!!
This I still can't comprehend!

[Tony Baloney]

I'm not going to quote the previous posts but look at the Creasey Fraud Triangle and you'll see why these things happen. Pressure, rationalisation and opportunity.

[muppet]

Reading more about it there are some keys issues.

There was a genuine fault, which was detected by the relevant system and a warning was sent. Of course the warning message simply notifies the crew of a general fault within a system. They would have known nothing about cracks in soldering.

On an Airbus all of the systems are computerised and like any computers, when there is a problem, turning it off and on again usually fixes it. This is not always a good idea while airborne though and needs some careful thought before doing it. For example, simple as it sounds, you need to consider all of the functions being controlled by that (or those) computers before satining it (them) off. Further to that, it would be very unusual to consider turing off two computers, especially when one is a back-up of the other.

In this case the captain got out of his seat and turned (by pulling circuit breakers) off and then on, two key computers. What he didn't appear to know is that, even though he reset the c/b of the first computer before turning off the second one, he needed to also reset the master pushbutton of that computer to re-activate it. So when he turned off the 2nd one, neither were working and the status of the aircraft was significantly downgraded, meaning a number of controls and protections (usually provided by either of those computers) were lost.

This still left the aircraft perfectly flyable. However, when the two computers were lost, the downgrade caused the auto-pilot to disconnect suddenly. The captain was out of his seat (to reach the circuit breakers) and the co-pilot was presented with a sudden need to fly the aircraft. Again this shouldn't be a problem, but amazingly he did exactly what the co-pilot did on the AF447. He pulled the stick back, held it there and stalled the aircraft. The captain didn't help by telling him to 'pull down'. You either 'pull up' or 'push down'. Language barriers were obviously an issue here along with the 'startle factor'.

To keep this post short, if they had done nothing, they would probably have been fine. The would have had to tolerate an annoying warning coming on every few minutes, but obviously that was better than what happened. If in any doubt, don't trouble shoot in the air.

For anyone interested, here is the final report (I haven't read it properly yet so feel free to correct any of the above): http://kemhubri.dephub.go.id/knkt/ntsc_home/ntsc.htm

Turning something off and on again never fixes anything, it masks an issue, intermittent or otherwise, one that happened 23 times in 2014 alone.

That's shocking to say the least irrespective of the intricacies of resetting a computer system mid flight!

Turning it off and on was the correct procedure.

This appears on the 'ECAM' which is a screen which presents the pilots with the relevant checklist when a 'known' problem is detected.

The procedure for the failure, as it would have appeared on the ECAM for all 23 occurrences, was to turn off FAC 1 and then turn it on. If that didn't work then turn off FAC 2 and turn it on.

Regarding fixing the problem. That is a completely different thing entirely. That would be a maintenance function. Pilots don't fix things. They follow the checklists, to try to restore a system if possible, and make decisions based on what systems are available to them if they can't restore them.

Now having said that turning the FAC 1 & 2 off and on was the correct procedure, cycling the circuit breakers off and on wasn't. Mechanics on the ground will try this and often it works. As you say, it may mask the real problem, as it certainly did in this case. Fair play to the investigators for tracing it to a soldering problem. But it is a function of modern technology that the goto fix is to turn it off and on again. Think of smartphones and tablets for example. Sometimes people get fixated on this and stop thinking beyond it.

The thing is this wasn't the first time the plane manifested this problem and the first time it did, once on the ground the thing should have been fixed, not set up some sort of procedure to mask it.

I'd have thought that the aviation industry would have much higher quality expectations than to accept this bodge on an ongoing basis!

I've worked in a few different electronic industries and whilst you may get away with telling a customer to reboot the thing once or twice, but 20 odd times and you'd be told to pull the thing out and replace it, and there's no lives at risk!!
This I still can't comprehend!

It always comes down to an individual's call somewhere. The call to ground that aircraft and find the source of the warning should have been made and wasn't. But the industry is largely self-regulating. Regulators don't have the resources to check up on stuff like this and if they ever did, the airline would threaten to move to a less intrusive jurisdiction. The public generally measures an airline by one single criteria, ticket price.

But despite that defect, the aircraft was still perfectly flyable. There were a lots of errors made in this accident.

[macdanger2]

Surprised there's no mention of this already, looks like 66 dead in this AirEgypt crash. Apparently the water there is deep enough (~4k feet) despite being only a few hundred km off the Egyptian coast so it might take a number of days or weeks to find the black box

[muppet]

Surprised there's no mention of this already, looks like 66 dead in this AirEgypt crash. Apparently the water there is deep enough (~4k feet) despite being only a few hundred km off the Egyptian coast so it might take a number of days or weeks to find the black box

http://www.bbc.com/news/world-middle-east-36349882

BBC reckon there was a fire on board.

Could have been an electrical problem which led to a fire, or it could have been a deliberate/carelessness fire.

Egyptair had a previous cockpit fire in 2011. This one appears to have happened (looking at the alerts) and the one 5 years ago both occurred on the right (co-pilot) side of the cockpit. On both sides of the cockpit there are oxygen masks with high pressure & on demand oxygen available for the pilots. In the fire in 2011, this oxygen fed the fire although they couldn't determine whether the rupture of the oxygens hoses was a cause or a consequence of the fire.

A fault with the oxygen system, along with some form of ignition (a cigarette, lithium battery event etc, etc), could conceivably have caused this. Terrorism seems to be less likely now, but probably still can't be ruled out completely.

[Tony Baloney]

I would still put terrorism down as the likely root cause if the first alarm to go off was in a toilet and it spread so rapidly.

[armaghniac]

I would still put terrorism down as the likely root cause if the first alarm to go off was in a toilet and it spread so rapidly.

Probably a c*** smoker, one of Finian McGrath's followers.