AirAsia flies to Melb instead of KL . Navigation error

[garpal gumnut]

https://www.theguardian.com/australia-news/2016/sep/07/airasia-pilot-flies-melbourne-malaysia-navigation-error

[HamishMcBush]

A comedy of errors so hilarious that you couldn't make it up. Thanks for a good laugh to start my day !

[framer]

Hard to believe really.

[Wizofoz]

Cheezus Christos!!!!!

[BuzzBox]

Yep, at one point after getting airborne the FO was going to cycle all three ADIRUs to OFF to 'reset the system'. The captain initially said nothing and the FO got as far as turning ADIRUs 1 & 3 off before the captain told him to stop! :ooh:

[Xeque]

"when autopilot engaged at 410 feet," says it all really. As many have suggested in this forum on previous occasions, what is wrong with hand flying the aircraft to (say) 10,000 feet or even transition altitude? Had one of them done this the other might have had time to reprogram the FMC from scratch providing, I suppose, the inbuilt systems in an A330 would allow the flight crew to do this once they were airborn?

[CurtainTwitcher]

Video from the original thread (http://www.pprune.org/australia-new-zealand-pacific/557869-airasia-issue-syd-today.html)

ATSB Final report (https://web.archive.org/web/20160907084846/http://www.atsb.gov.au/media/5771318/ao-2015-029_final.pdf)

n2lFwyWVinI

[IcePack]

Obviously unable to fly using "old conventional " nav. & totaly "untrained" in the fmgc as their is an position update facility.

[paperHanger]

One small planes, a normal check is to make sure the DI and compass are aligned as part of the start-up checks. When flying with a GPS a similar "does the reported position agree with where I am" sanity check during start-up is useful.

Now, colour me stupid, and I am not familiar on type, but if the inertial nav initialisation position is significantly different from the current GPS coordinates, should that not bring up a very large warning that one (or both) are not to be trusted? Or was that the initial warning that they ignored?

[Algol]

And what were the Aussie ATC'ers doing when the flight departed radically from its filed route?

[BuzzBox]

ATC was the saving grace in this whole sorry episode. They picked up the aircraft's incorrect tracking immediately after take-off and then provided assistance to help the hapless crew get the aircraft back on the ground. The crew weren't able to do a visual approach back in to Sydney due to the weather, so ATC provided vectors all the way to Melbourne.

[EGPFlyer]

Getting everyone else out of the way!

A review of the ATC response to this occurrence identified that the controllers carried out several tasks that reduced the risk to both the occurrence aircraft and other aircraft in the area. They were the first to notice and alert the crew to the tracking problem, and provided assistance to identify that the aircraft’s main heading indicators were erroneous. Additionally, ATC quickly resolved a possible conflict with another aircraft lined-up and ready to depart on the parallel runway.
Subsequently, coordination with several ATC units and the availability of continuous radar coverage provided the crew with a safe diversion alternate and vectoring from Sydney all the way to final approach in Melbourne. The captain reported that ATC had prevented the situation becoming a ‘dire emergency’ and that in many ways they had ‘saved the day’.

[CurtainTwitcher]

And what were the Aussie ATC'ers doing when the flight departed radically from its filed route?

From the report (https://web.archive.org/web/20160907084846/http://www.atsb.gov.au/media/5771318/ao-2015-029_final.pdf) page 3
A few seconds later, ATC observed the aircraft turning left, contrary to the SID, and tracking towards the flight path for the active parallel runway, runway 16L. In response, ATC contacted the crew and requested confirmation that they were tracking via the SID and would be maintaining a heading of 155° before turning right. At the same time ATC held another aircraft in the line-up position for departure from runway 16L
More on the ATC response on page 5.

[oldpax]

Don't these big jets have the equivalent of the RAF P12 compass ?How in laymans terms did a large airplane "get lost" after taking off from a large airport!!

[wiggy]

paperHanger

I am not familiar on type, but if the inertial nav initialisation position is significantly different from the current GPS coordinates, should that not bring up a very large warning that one (or both) are not to be trusted? Or was that the initial warning that they ignored?

I would have thought so. Some types certainly will warn " check present position" or similar if the entered position is wildy out from GPS.

It's interesting to see (from the report, if I'm reading it correctly), that they manually entered the gate position as a lat/long even though it seems they had GPS. Certainly on a similar type I know of if GPS is available then you must line select the GPS position into the IRS/ADIRU present position "boxes" at the start of the align process ( basically copy and paste), without amendment, precisely to avoid the sort of finger trouble described in the report.

TBH regardless of the above I'd have though there were plenty of other opportunites to catch the foul up before start, let alone takeoff (Gross distance checks, map display checks, etc....).

[Sailvi767]

Don't these big jets have the equivalent of the RAF P12 compass ?How in laymans terms did a large airplane "get lost" after taking off from a large airport!!

They had all kinds of options available to them. They had full VOR and ILS capability and could easily have flown the approach back to Sydney. I guess not having a magenta line scared them.

[slowjet]

Children of the majenta line, of course, but will this industry ever wake up to the solution ? It is going to get much worse.

[chimbu warrior]

The report also states that after almost 3 hours on the ground in Melbourne (during which time system checks found no unserviceability with the aircraft), the same crew flew on to Kuala Lumpur, for a total duty of 14 hours 53 minutes and total flight time of 10 hours 27 minutes.

Am I the only one who thinks it very unwise to continue a duty, obviously into discretion, after a significant incident?

[RAT 5]

It will be interesting to learn of Air Asia's actions with the crew. Not only did they show a huge lack of procedural knowledge in the flight preparation, and discipline with disregard of warnings and mis-use of checklists, the F/O then exhibited huge lack of technical knowledge in his suggestion and subsequent attempt to realign the IRS's in the air. Not to mention that the view out of the side window after 400' is not what was expected.
"Oh, a/c is turning left and we expect to turn right. Why?" Heads down FMC-ing? Solution: disconnect, ask for vectors and clear critical airspace.
As someone said, why can they not fly an instrument approach to Sydney. They still had 1 IRS it would seem for attitude information.
It seems a little more than a re-training session is required.

[framer]

Are the Australian public happy with their government departments decision to allow this operator to share airspace with them?
It's beyond a joke.

[compressor stall]

No but if we ban them they ban us. And we need their airspace to get anywhere more than they need ours. Same for Indo.

[gazumped]

Wow! And this happened 15 March last year. This should be being shouted from the tree tops by our press. How do the paying public get on these aircraft?

It seems the over-riding modus operandi of some airlines is: IF THE AUTOMATICS FAIL THE AIRCRAFT CANT OPERATE.

Let alone all the warnings this crew overlooked to get airborne in the first place. the operational decision by this airline to allow the same crew to continue to their original destination, beggars belief!

How can anyone describe Raw Data as a "dire emergency"?

I say again Wow!

[BuzzBox]

Ben Sandilands is on to it, although I doubt that very many of the 'paying public' read his blog:

ATSB reports on cockpit cockup in AirAsiaX flight to KL (https://blogs.crikey.com.au/planetalking/2016/09/07/atsb-reports-cockpit-cockup-airasiax-flight-kl/)

[flash8]

It could have been worse... put them in a clockwork airplane and they would have ended up in London..

attempt to realign the IRS's in the air.

wtf.

[deadheader]

Dodgy earmuffs => u/s GPWS

Could easily have led to more than just tea & biscuits, e.g. mountainous terrain IMC etc

[n5296s]

Sad story. But why were they entering coordinates in the first place, rather than airport designators and waypoints? (Not to mention the bit about them reading them from a piece of card held in front of the aircraft - reminiscent of Three Men in a Boat).

And why did entering the wrong coords cause a terrain alert?

Luckily this just turned into an amusing story. But I wouldn't be in a big hurry to fly with Air Asia.

[wiggy]

why did entering the wrong coords cause a terrain alert?

Unless the EGPWS was solely using GPS data I wonder if it was a predictive warning, based on the corrupt position?

Edit to add, having just dug into the report:

"An EGPWS alert activated at approximately 600 ft during the initial approach at Melbourne,
seconds prior to the commencement of the go-around. An assessment of recorded parameters
indicated that, as with the EGPWS alert departing Sydney, this was also a spurious warning
associated with the aircraft’s incorrect position information."

[Herod]

I think the EGPWS thought they were heading for Table Mountain. Without looking at it, I would suggest KL is a track in the region of 320 degrees. Actually about 298, but my guess would have had the aircraft going in the right direction. Set heading on the a/p, continue climbing, and sort it out when time permits. Or have ALL the old, basic skills gone?

[GordonR_Cape]

Cape Town is 18.5E, 15E is far out in the Atlantic Ocean, and flying SW goes even further offshore (nowhere near Table Mt). The lack of airports matching that location in the database would be more likely to trigger a warning (with gear down).

A rather similar navigation error (insufficient decimal digits) in Brazil in 1989 led to a much less happy outcome: https://en.wikipedia.org/wiki/Varig_Flight_254

[PAXboy]

The ATSB said “even experienced flight crew are not immune from data entry errors” and advised AirAsia to upgrade its flight systems to assist in preventing or detecting such errors in future. Does this indicate that the ATSB think it better to make the machine smarter - rather than attempt to make the crew smarter ...? Had it have been QF/or other, might they have said something different?

[springbok449]

Faulty earmuffs? Seriously is that the only excuse they could come up with? Sounds more to me like the skipper didn't fancy doing the walk around in SYD because of the bad weather...

[peekay4]

Does this indicate that the ATSB think it better to make the machine smarter - rather than attempt to make the crew smarter ...? Had it have been QF/or other, might they have said something different?

ATSB is saying that it's impossible to 100% eliminate human error.

The Captain who made the initial data entry mistake had over 22,500 hrs. No matter how experienced you are, people must be expected to make mistakes at some point. There must be cross-checks, and in this case human & automated cross-checks / procedures all failed.

While both pilots' performances were inadequate, there were also numerous issues with the Airbus system:


When there's GPS and IRS disagreement, the GPS position is supposed to be invalidated and a GPS NAV FAULT warning should have appeared on ECAM. But this warning was never displayed.
Similarly, upon entering the wrong IRS coordinates, a separate FMS/GPS POSITION DISAGREE warning should have activated. No evidence that it did.
When takeoff thrust is applied and GPS PRIMARY is not active, the aircraft coordinate is supposed to auto-update based on the entered runway location. But this didn't happen either; the update just silently failed.
And instead of flagging the bad data, the system displayed the wrong magnetic heading to the crew on the ND and PFD. After takeoff the displayed heading was 193 degrees when the actual heading was 155 degrees. So they turned left to intercept the SID instead of turning right like they were supposed to.
On a couple of occasions there were warning chimes, but without any ECAM messages or indication of what (if anything) was actually in error

I'm sure software updates are being made...

[peekay4]

How can anyone describe Raw Data as a "dire emergency"?
Nothing to do with raw data. They had the wrong heading displayed on both ND & PFD due to incorrect magnetic variation being applied, so they were unknowingly flying the wrong way even when they tried to follow ATC instructions. That could have been disastrous.

But why were they entering coordinates in the first place, rather than airport designators and waypoints?
We're talking about two different things here. Airport designators & waypoints are entered into the flight plan in the FMGS. The crew did this part correctly. The coordinates are needed to update the IRS and require greater precision. The crew did this part incorrectly.

[Herod]

The lack of airports matching that location in the database would be more likely to trigger a warning (with gear down).

GordonR. Yes you're right, I hadn't thought it through properly. I had just that when flying into a new airfield in Africa. Night-time, perfect vis, runway lights in sight. EGPWS started blaring, because it thought we were trying to land in the desert. SOP at night? Go-around. That would have gone down well with management.

[Enos]

Didn't Polish LOT do something very similar in an old generation 737 possibly a 400 out of LHR.

Entered East into the FMC because that's where they mostly do all their flying on the 737, instead of West and with the IRSs not working correctly your down to STBY instrumentation.

ATC talked them down with turn right 5' left 10' etc.

We are all sky gods at home on the couch but are all just one flight away from a Disaster at work!

Correct me if I'm wrong gents.

Cheers Enos

[peekay4]

Yup: http://www.pprune.org/rumours-news/330847-lot-737-incident-june-2007-crew-s-poor-english-blamed.html

[atlas12]

Did the FO even check the FMC entries before departure I wonder? And why did the skipper manually enter the lat/long into the FMC from the sign at the gate? At my company we just copy the GPS coordinates and verify them against the airport/gate reference from the database or JEPS if required. I have stuffed up lat/long waypoint entry before on a long sector with about 10 manual waypoints to enter because ACARS didn't download them, but this is why the second crew member checks the FMC. The error was detected during the LEGS check and it turns out I had some finger trouble during entry and the waypoint was about 50 miles off track :O very stupid on my behalf and I learned a valuable lesson that day, now I go over my own work with a fine tooth comb before passing it over to the other guy. Having said that, departing with an incorrect alignment in this day and age should be impossible and ignoring all of the warnings is just insanity :confused:

[ImbracableCrunk]

Not to mention that the view out of the side window after 400' is not what was expected.

In my experience in Awesome Asian Aviation Excursion, the ground doesn't exist above 400' AGL.

[NSEU]

They had full VOR and ILS capability and could easily have flown the approach back to Sydney. I guess not having a magenta line scared them.

Sydney now has no VOR. Also, the magnetic heading data fed into the VOR systems is based on IRU present position (which appears to have been wrong). I'm surprised the aircraft reached Melbourne with vectors (unless they were using their whiskey compass).

Turning off intertial reference systems at the wrong time hasn't only happened to budget airlines.

[CurtainTwitcher]

One very subtle point that I've found is that alignment of the IRS on my aircraft calculates its own latitude as a reasonableness check, but not it's own longitude. The manufacturer FCOM is very hazy on alignment and error codes (none are listed any more).

However, a well known third party publication states:
Reasonableness tests
- The IRU compares entered long with the last LAST POS stored in non-volatile memory
-The IRU does not calculate its own longitude
-If the the difference you enter is greater than 1°, the ALIGN annunciation flashes, status code 4 shows and ENTER IRS POSITION displays on the CDU
-This could legitimately occur (i.e the entry is correct if the IRU was newly installed
...

The IRU also compares entered latitude with the LAST POS latitude
-The IRU calculates its own latitude
-If the difference is greater than 1°, the ALIGN annunciation flashes

This failure mode (for us humans) is very subtle.
Incorrect latitude correct longitude will be picked up through the IRS's own internal error checking (excluding LAST POS check)

however
Correct latitude, incorrect longitude will not (excluding LAST POS check).

I was unaware of this subtleties until this thread forced me to have a greater look at my own system knowledge and defences. The FCOM would not lead you to this understanding, and that is quite a large hole in the Swiss cheese.

[BuzzBox]

Originally posted by NSEU
Also, the magnetic heading data fed into the VOR systems is based on IRU present position (which appears to have been wrong). I'm surprised the aircraft reached Melbourne with vectors (unless they were using their whiskey compass).

The report says the FO subsequently carried out the IRS ALIGNMENT IN ATT MODE checklist from the QRH. That would have restored attitude and heading information once the aircraft's magnetic heading (from the standby compass) was entered into the MCDU. The heading would then need to be updated periodically, just like an old-fashioned gyro compass.

[NSEU]

Thanks, BuzzBox... I should probably read the report (*^ ^*). From the initial posts, I didn't get the impression that the F/O knew what he was doing. I recall that on the 744, if you have one IRU in NAV and the others in ATT, the IRU in NAV will feed magnetic heading data to the IRUs in ATTitude (I can't recall if the HDG entry boxes still appear with one IRU still in NAV).

Of course, we are talking about an Airbus....

[Algol]

Buzzbox and Curtain twitcher, thanks for the link to the report, I hadn't seen it before questioning ATC's actions. The OP only put up the Guardian article which seemed to say the crew were left blindly flying around and found their own way to Melbourne. Happy to withdraw any criticism of ATC, they helped enormously.

Now regarding the criticisms of this crew. Some on this thread have lambasted them for lack of systems knowledge. I think its also rather unfair. My own decades long experience and knowledge of INS systems wouldn't have anticipated the behaviour seen in this event. Its perplexing.

Several yonks ago when I did my ATPL I remember the instructor telling us that an INS will NOT align if you input a grossly inaccurate position. The reason being that the INS platform compares the input position to the local 'vertical' and this would not resolve if the 'expected' vertical didn't match the 'sensed' vertical within narrow limits.

A decade later I flew INS equipped B747's (before the days of GPS) and those INS's behaved exactly as was taught. They were twitchy buggers and would fail to align even if a wind was making the aircraft move around, or the loaders were being a bit boisterous. A few degrees disagreement to the sensed vertical was enough to fail the process.
In this AK case the difference between 'input position vertical' and the real 'sensed vertical' was huge - 136deg - and yet it seems the INS accepted the error and aligned to the wrong location. A very 'robust' INS, no doubt. Is that such a good thing?

When I went on to my first Airbus there was no GPS. The first commercial GPS system was only set up around 1995. Before that it all relied on INS (IRS).
In those days we had to confirm a valid 'Position Update' of the aircraft symbol on the ND to the runway on start of the take off roll. It would happen automatically when thrust levers were advanced. It was reliable. I never saw it 'not work'.

In this AK event the report says the Position Update did not happen at start of take-off, because "the error was too great". Seriously!? You mean Airbus designed a navigation system that simply goes 'idle' when you've created a massive error - exactly when you most need it to work? Or at least WARN you that it isn't working?

When GPS was introduced we were told this would (when available) prevent any more gross errors in Nav, as the GPS would automatically update the 'mix POSN' and eliminate gross errors. Surely a pair og GPS's in agreement are more reliable than even 3 INS's?
But no. Now we discover that if the GPS doesn't agree with the INS's it will effectively vote itself out of the system, and the only warning you'll get is one single chime after it has thought about it for ten minutes (or more)! Incredible.
By the way, its not at all unusual to have 'mystery chimes' go off which never reveal themselves. Perhaps a flicker on the ECAM which you can't read, and which can't be recalled. So what do you do? Stop everything? Go back for engineering inspection? Not likely in the real world.
The lack of a proper clear warning of a primary navigation failure is again unbelievable.

Airbus aircraft seem to regularly do this to their crews - they fail, and dump everything in the crews lap with little hint as to what's really going on. This incident reminds me so much of the Air France AF447 accident - another crew let down by hard to diagnose Airbus systems failures, with obscure (out of normal envelope) systems behaviour contributing to their confusion. They were extremely lucky this didn't end up the same way. Perhaps if it had been night time?
Those saying there was an obvious action - what is it? When the EGPWS starts shouting at you at 400' after liftoff you know you've got problems - but when you realise now your ND compass is giving you erronoeous information (because ATC tell you) what would be expected except confusion and shock?
Reverting to raw data sounds an obvious first step - but the ND is still wrong, and you can't steer unless you realise it and focus on that micky mouse compass. All very easy to describe from the comfort of an armchair. Ever steered by a magnetic compass? If so you'll know its not that simple.

Did the AK crew screw up? The Captain input the position manually. Is that Air Asia procedure? If so - why? If that's what they train them to do, who's at fault really?
In my INS days the position entry had to be done together with a second crewmember, and double checked. This didn't happen here - is that how they're trained?
Why weren't they trained to use the slew buttons? We were trained to use them and we all much preferred it - its so much easier than struggling with conversion of Lat/Long to Airbus formats.
Why were they even required to insert the Gate rather than just accept the ARP (which will be updated anyhow at take-off)?
Why wasn't the maintenance mod done by AK? The Align would have been automatic - as it is in the Airbus I fly now. Human error is largely removed entirely.

What I'm saying is that the armchair heroes are good at boasting about their own knowledge of the systems - after the fact.
The keyboard warriors get some self satisfaction from slating their fellow pilots - inferring to the rest of us how 'skilled' they themselves are? They let the manufacturers and regulators and the airlines entirely off the hook. Haven't you learned by now that the next guy to swing is you!?

[TurningFinalRWY36]

Algol i don't know all the details of the report but it seems that only the Longitude was incorrectly entered. If I remember correctly from ATPL studies an INS platform will happily accept any Longitude entered as long as the Latitude is correct. Latitude is used to level the platform. So in this situation the INS will happily align itself and believe its position to be correct

[Rule3]

To OLDPAX et al

As an ATC working in the Middle East for many years, you would be staggered at the number of times we had to ask pilots "Where are you tracking to"

It was mandatory for ATC to issue a route clearance, and obtain a correct readback, for EVERY aircraft that entered the FIR. On many occasions the readback was correct but they then flew a different route. Go figure.

[wiggy]

If I remember correctly from ATPL studies an INS platform will happily accept any Longitude entered as long as the Latitude is correct. Latitude is used to level the platform.

That's pretty much how I recall it as an end user of early platforms. An incorrect latitude entry will, I think, manifest itself at the levelling and/or gyro compassing stage. An incorrect longitude causes no such problem.

[lederhosen]

Very similar incident happened to LOT with a 737 at Heathrow some years ago. They entered east instead of west and flew round not very convincingly on the standby instruments before landing back. It is fairly basic system knowledge that you can force the IRS to accept a wrong position (by repeating the false data entry) for reasons explained above. What is not so simple is to fix the problem once in the air. The captain in the Sydney incident had 22000 hours so rather than filing this under children of the magenta line I think the CRM and problem solving aspects are the interesting bit. The idea that you are going to fly a long haul leg without the primary instrumentation and autopilot is a bit out of touch with reality and indeed regulation. Problem solving and recovering the systems (as far as was possible) is the bit that went seriously wrong.

[DaveReidUK]

If it's not a silly question, why can't FMS data providers include airport gate coordinates in their data packages?

Even if crews were still required to input the lat/lon manually from the coordinates displayed at the gate, at least the system could verify that they corresponded to a known location rather than being in the sea 300km from the nearest airport.

In fact even a simple automated check that the input coordinates are within a couple of miles of the flight plan origin would do, at a pinch.

In an era where the aircraft is increasingly expected to protect itself from the actions of the crew, it's not rocket science.

[macdo]

Sounds like a simple failure to follow company SOP's, which are almost certainly the same as or very close to Airbus SOP's. This error should have been easily trapped before departure, during either the briefing or running the checks.
Poor tech knowledge by both pilots explains the rest.
Had they managed to turn off all three ADRIRU, gulp!!

[wiggy]

If it's not a silly question, why can't FMS data providers include airport gate coordinates in their data packages? .

Some (many?) do, or failing that at least hold the lat/long of the airport reference point. FWIW our preferred option is to align using the current GPS position but as there's still a verification process.

On some types/SOPs the manual longhand entry of Lat/long for alignment is seen as something to be avoided if at all possible( for obvious reasons). If that it has to be done then there's usually the need for extra checks to trap errors.

[framer]

If it's not a silly question, why can't FMS data providers include airport gate coordinates in their data packages?

At my outfit you check the actual position coordinates against the airport coordinates as part of the preflight checks. The airport coordinates are part of the data package obviously.
I think this is a case of there being multiple chances to trap the error but none of the opportunities were taken to do so.

[Meikleour]

Surely the map could have been corrected by overflying a known VOR and entering the co-ordinates into the Fix Info page. This was taught by Airbus back in the early '90s.

[De_flieger]

Meikleour - Airservices Australia has just spent a fair bit of time and money removing lots of navaids to leave only the Backup Navigation Network, so this might have been practical in early 2015 but it is less practical now. The nearest VOR to Sydney looks to be Canberra, roughly 130NM away, and almost exactly halfway to Melbourne where they ended up. The idea behind the Backup Navigation Network is that all IFR aircraft will primarily rely on GPS for navigation, with only a skeleton network of ground based aids left. Luckily all the GPS-based navigation systems as seen here are so robust!! :eek:

[wiggy]

Luckily all the GPS-based navigation systems as seen here are so robust!!!


Maybe, but perhaps a better description of modern airliner nav systems would be "computer based with inertial/GPS/Air data/radio input, blah blah..etc etc".

In this incident's case it looks to me as if the GPS side of the nav system was robust enough, but it's data was being disgarded because of system logic.

If the crew had been using a pure GPS nav system I somewhat doubt the incident would have happened.

[RAT 5]

From early comments on here it would seem they still had No.2 ADIRU. Will someone say if this is enough to give correct info to either PFD, or whatever ABus call it.

[WingNut60]

What has my attention right now is, what EXACTLY did they tell the passengers when they landed in Melbourne.

I can imagine that the first advice would have been wrapped in "technical problems". But at some point the truth must have become obvious to the flight crew. Did they at that time fess up and give those who might have wanted to get off the damned aircraft the opportunity to do so, or did they then continue to conceal the real nature of the diversion and in so doing stamp their behaviour as dishonest and misleading?

Keeping the customers calm is not an excuse for feeding them bullsh...t. Not when you're already on the ground.

[Hogger60]

How many cockups is this for AAX in Aussie airspace in the past few years??:ugh::ugh::ugh::ugh:

I would hope CASA would be taking a very close look at them and soon.

[RVF750]

Of course, if their ND HDG was so far out, didn't they query it on line up? I certainly would with a 30 degree error on the runway....

[Heathrow Harry]

It's easier than you'd think to enter/copy the wrong co-ordinates into any spreasdsheet/computer/ whatever - has happened since the year dot..... (I always worried about Christopher Columbus's booze cruise to Morocco....)

You'd be better off if you just typed in the FULL name (not a set of letters like MEL) of your destination

[Meikleour]

Der fleiger: thanks for your update. The procedure I mentioned also works if you simply type in the GPS co-ordinates!! Perhaps the crew here disbelieved their GPS as well as their FM data?

[PersonFromPorlock]

Whatever happened to "It's all an aid to DR"?

[LeeJoyce]

How many cockups is this for AAX in Aussie airspace in the past few years??:ugh::ugh::ugh::ugh:

I would hope CASA would be taking a very close look at them and soon.

Bloody hope so

Just 25min ago local XAX213 almost taxied over the pushout tug and engineer who was still standing by to show the bypass pin, they got taxi clearance and were on their way, engineer had to do the bolt, it was very close

[Nick 1]

What about FMGC Fuel prediction , or expected landing fuel at destination ?

[BuzzBox]

Originally posted by RAT 5
From early comments on here it would seem they still had No.2 ADIRU. Will someone say if this is enough to give correct info to either PFD, or whatever ABus call it.

The Capt would have lost all attitude and air data when ADIRU 1 was cycled OFF. The FO's PFD would have continued displaying data from ADIRU 2, but the heading info would have been wrong. The Capt was unable to select data from ADIRU 3, because it had also been cycled OFF. However, he could have selected a copy of the data displayed on the FO's PFD to be displayed on his side by using the EFIS DMC selector. Whether he did that or not is unclear. Attitude, heading and air data from ADIRU 1 could also have been restored to the Capt's PFD when the FO subsequently completed the IR ALIGNMENT IN ATT MODE checklist.

[RAT 5]

Buzzbox: thank you. So their technical knowledge seems to have been lacking. Firstly the attempt to realign in the air, and then how to recover sufficient display data to allow a safe return to Sydney.

[NSEU]

But no. Now we discover that if the GPS doesn't agree with the INS's it will effectively vote itself out of the system, and the only warning you'll get is one single chime after it has thought about it for ten minutes (or more)! Incredible.

I read somewhere that this is the same for 747-400s (at least the part about the GPS being voted out).

According to the report, the aircraft in the incident did throw up a warning about position mismatch during the preflight, but the captain forced through it. i.e. the basic position comparison (which looks at both latitude and longitude).

So their technical knowledge seems to have been lacking. Firstly the attempt to realign in the air, and then how to recover sufficient display data to allow a safe return to Sydney.

That was my initial impression, too, but if you read the report, they did eventually follow the QRH to get the aircraft into basic attitude mode. With limited visibility for a return to Sydney, basic attitude mode may have been insufficient to meet regulatory requirements/airline SOPs (?)

[Sunamer]

And why did entering the wrong coords cause a terrain alert?
Because one of the modes of EGPWS Warnings works of the database with terrain data. That DB is used in connection with the aircraft POSITION that needs to be ENTERED. So, if you enter a wrong position happened to be close enough to some mountains, and then try to fly the AC, EGPWS SHOULD trigger "terrain terrain" warning.

[BuzzBox]

Originally posted by NSEU
With limited visibility for a return to Sydney, basic attitude mode may have been insufficient to meet regulatory requirements/airline SOPs (?)

They certainly couldn't have flown an RNP or any other type of 'new age' approach, but radar vectors to an old-fashioned ILS should have been easily doable once the attitude and heading information was restored. That said, I suspect the crew was a bit 'gun shy', given they didn't really understand what had occurred.

[peekay4]

Surely a pair og GPS's in agreement are more reliable than even 3 INS's?
But no. Now we discover that if the GPS doesn't agree with the INS's it will effectively vote itself out of the system
Some problems relying solely on GPS:

Multiple GPS units share common failure points (e.g., the satellite signals) so they are not independent -- two units may simultaneously fail in the exact same way
Common integrity monitoring algorithms can't detect all failure scenarios -- they can report acceptable integrity when the data is actually garbage
GPS signals are subject to interference, jamming and spoofing

Also when the first aviation GPS systems were being designed back in the mid-1980s and early 1990s, we did not yet have the full GPS constellation launched in space, and the US DOD was actively degrading GPS signals via Selective Availability (SA).

So I think the engineers & regulators decided to take a very conservative approach in preferring the "proven" IRS (incidents like KAL007 notwithstanding). SA was not turned off until May, 2000.

What about FMGC Fuel prediction , or expected landing fuel at destination ?
Prior to takeoff, fuel prediction is only based on the entered flight plan & departure runway -- not the aircraft position -- so it would appear correct to the crew.

radar vectors to an old-fashioned ILS should have been easily doable once the attitude and heading information was restored
I believe they restored (partial) attitude & heading information only after the decision to divert to Melbourne was already made.

[BuzzBox]

Originally posted by peekay4
I believe they restored (partial) attitude & heading information only after the decision to divert to Melbourne was already made.

Possibly so, but the report says "The flight crew continued their attempts to restore the flight guidance and navigation systems while in the Sydney area. The FO then carried out the IRS ALIGNMENT IN ATT MODE checklist from the QRH, restoring some ND and PFD functions..." So it seems they were still in, or very close to the Sydney area when those functions were restored. They were certainly much closer to Sydney than Melbourne! The weather in Sydney at the time was 10k vis, light rain showers and a cloud base of 1,700ft, so they would have been visual well above the minima.

[Ollie Onion]

It is a worry, why can't you just hard tune the ILS and fly a good old fashioned Instrument Approach, all that you need is the compass, altimeter, vor/ils and you are good to go.

[BuzzBox]

It is a worry, why can't you just hard tune the ILS and fly a good old fashioned Instrument Approach, all that you need is the compass, altimeter, vor/ils and you are good to go.

Ollie, that's exactly what they could have done, once they'd realigned the IRSs in ATT mode and restored their attitude and heading info. It'd be a tad difficult without those!

[ACMS]

Well having never tried to depart with the IR's aligned 6,000 nm away!! I for one am glad it was them and not me. We can all learn from their initial stuff up and now have a good refresh on exactly what works in that situation and how to get good data back.

Thanks Air Asia.....:ok:

[413X3]

"Children of the magenta line" jokes even though the old captain has over 20k hours

Once some people find a scapegoat they always fall back on it.

[Metro man]

PROG page and entering the active runway in the field will show bearing and distance from the threshold. Useful as a gross error check.

[Capn Bloggs]

What would have happened had the FO turned off the No 2 IRS thingee?

[BuzzBox]

What would have happened had the FO turned off the No 2 IRS thingee?

Capt & FO would lose all attitude, heading and air data on their PFDs & NDs. Data could be recovered by realigning the IRs in ATT mode, with the heading updated periodically from the standby compass.

[wheels_down]

I couldn't see anything but strange the Australian regulator has not imposed any conditions on the AOC? They were pretty quick to ground the Tiger Airways operation with numerous restrictions and then the implications were shown.

We're gambling with lives now...

[RAT 5]

What SBY attitude display is there on A330? Is there a good old basic AH lurking somewhere in the corner, or is it all TV screen stuff?

[BuzzBox]

Old A330s have a basic AH, but newer aircraft have an Integrated Standby Instrument System (ISIS). It's basically a souped-up AH plus ASI/altimeter and ILS scales combined into one instrument and displayed on a small screen. The ISIS has its own gyros and air data module, so it's independent of the other aircraft systems.

[sycamore]

Surely the IR systems knew where the aircraft was when it arrived at the gate,and then ground power would be connected so the systems would have been kept `alive` all the time on the ground,and only required a check for a possible bit of `drift`,rather than a reset??
It used to be fairly straightforward with a navigator..we would say ,we are `here`,going `there`,draw a line or two and we`ll be off....!

[Mags on]

I cant help but wonder why there are so many complex instruments on board to do what has become a simple task. Why not put 2,4 or even 6 Garmin1000's on board, there is space for sure, and fly on.

[BuzzBox]

The standard procedure on the A330 is to re-align the ADIRUs before every sector, so they need to be re-initialised with the current position.

[wiggy]

Sycamore

1. Can't speak for the bus but there can be problems in some types if the aircraft is unattended, the ground power drops out and the IRS's end up on battery power.

2. No idea why but where we work on the 777 we moved away from an SOP of leaving the IRSs alone between relatively short long haul sectors and we now always do a full re-align. That might be down to tighter nav accuracy requirements these days but that is only a wild guess.

[peekay4]

When you enter the company route (or manually put in from/to), the IRS initial coordinates are automatically reset to the Airport Reference Point, which is a small distance away from the actual gate location.

In this case the recommended practice is to change the coordinates to the gate location using the slew keys (up/down) -- since the amount of change will be small -- instead of manually entering the full coordinates like the Captain did in this instance.

[FlightDetent]

The standard procedure on the Airbus is also to use the slew (plus / minus) buttons to set a position for IRS INIT, as opposed to punching in the numbers.

This in order to prevent gross entry errors from crew side. Which, especially if only longitude is inserted wrong, could go undetected by the IRS verification algorithms.

Further reading in the book reveals: IRS are automatically initialized with the GPS position. However, the flight crew can override the automatic position initialization.
And there is no recommendation (*) to enter gate coordinates at all.

Without diving in too deep, I would like to underline any and all vilas' repeated post about how non-adherence to manufacturer's S(uggested)OPs brings dire consequences, unforeseen by whoever believes they know better. Unfortunately.


(*) that's actually not 100% true, but I'd take the challenge to see the quote from OEM manuals, full, with the specific conditions when this would be the best practice.

[peekay4]

IRS are automatically initialized with the GPS position. However, the flight crew can override the automatic position initialization.
Not for all MSNs, depending on the software version, and unfortunately not for the incident aircraft.

[cwatters]

Bit surprised they haven't automated the process somehow. Perhaps a QR code on the gate read by a camera on the aircraft or a short range Bluetooth beacon?

[Chronus]

mystery of MH370 resolved!

[sycamore]

Thanks WIGGY,and Buzz...doesn`t help if you do RI-RO,and no gross/cross check of basic distance /time fuel....

[wiggy]

Cwatters

TBH there's no real advantage to inventing Bluetooth or optical processes when you've already got GPS available on many/most airframes and often that can be selected by a couple of key presses as the datum position for inertial align without needing to go down the road of entering figures long hand.

It just seems from what others have said that due to it's modification state this particular airframe lacked even the ability to align off a GPS Lat/long.

sycamore

You are welcome. Pardon the thread drift but I recall the days of our navs babysitting/ nursing an early generation inertial and being quite chuffed if the terminal error was less than 10 miles after an 1 or 2 hour " trip"........how times have changed....

[RAT 5]

IRS are automatically initialized with the GPS position. However, the flight crew can override the automatic position initialization.
Not for all MSNs, depending on the software version, and unfortunately not for the incident aircraft.

Speaking as Boeing NG pilot; why would customer opt for (if that's the case) where the GPS position can not be copy/paste into the FMC for initialisation? Or why would a systems designer create that Gotcha?Punching in Lat/Long manual is an NNC an went out with Omega.

[FlightDetent]

this particular airframe lacked even the ability to align off a GPS Lat/long.Still, even in a non GPS aircraft, the procedure is to enter the aerodrome ID, and align to the database ARP. It makes no difference to the navigation precision or performance of the IRS. And if you need to (ehm, ehm) modify, the technique is to slew the loaded values.

RAT: no gotchas. Just a sloppy SOP set, not aligned with error/failure modes asessment of the manufacturer and designers' intention.

[peekay4]

Still, even in a non GPS aircraft, the procedure is to enter the aerodrome ID, and align to the database ARP. It makes no difference to the navigation precision or performance of the IRS. And if you need to (ehm, ehm) modify, the technique is to slew the loaded values.
The ARP could be many miles from the gate position.

So modification of the init location from ARP to gate may be required to meet RNP, and is the SOP in that case (using slew keys) as recommended by Airbus.
why would customer opt for (if that's the case) where the GPS position can not be copy/paste into the FMC for initialisation?
I don't know for this case specifically. However, as alluded in a post above there has been a shift how we perceive GPS today vs. how GPS was perceived even 10 years ago.

Separating IRS entry from the GPS was thought of a "good thing" as theoretically they cross-check each other. E.g., to detect errors in case the GPS signal was degraded or spoofed. Automatically copying the GPS position would negate this "feature".

Today of course most of us trust a seemingly functioning dual GPS setup 100% of the time, at least until there's a major GPS-related incident in the future. Then people will complain about over-reliance on GPS and why there weren't more checks-and-balances in place.

Having said that, entering the Lat/Long manually from scratchpad was never the recommended procedure.

[woodpecker]

Easy mistake to make. If I go back to the B757 it didn't have GPS so the position was updated via DME/DME or DME/VOR radial. Prior to departure the airfield code was input into the FMC position update which then the appropriate (displayed) stand number. This then updated the IRS.

I enjoyed flying the aircraft using raw data so, on the day in question, NAV was not displayed on the HSI on departure, just the basic VOR radial information. Alas the copilot had earlier entered EGPH (Edinburgh) instead of EGPF (Glasgow) when updating the IRS position. The required stand was not on the list so he accepted the airfield (EGPH) position as "good enough".

All was fine during the departure until well into the climb (using raw VOR data) I engaged the autopilot and reverted to the NAV display. No sign of the magenta line!! Changing the range revealed it well removed from our position. No problem, back to the VOR display!! Continued to Heathrow with the position slowly updating (DME/DME) and when we entered the Bovingdon hold all was back to normal.

Note... The 757 didn't have a position update on entering the active runway contained in the Lnav route which the 777 does (if I remember correctly) so there was no "position error flag" when comparing the actual position to that contained in the IRS brain.

What did it say on the wall of the briefing room wall at Hamble?

"Learn from the mistakes of others, you wont live long enough to make them all yourself"

[Yaw String]

Oh well,
Just one more airline on my personal no-fly list...
Charges...1/lack of adherence to the necessary SOPs..
2/lack of attention to,and cancelling of,crew messages/cautions/advisories....
3/ lack of system knowledge...etc etc..
Crikey,sounds like me as a 17 year old PPL holder!
With retirement just around the corner,I think I will avoid reading the cruise ship forums too!!!!

[edmundronald]

Having at least one navigation system that is autonomous and does not rely on a vulnerable reference seems a good idea. GPS spoofing has been tuned into an art form by the electronic warfare specialists.

Edmund

[RoyHudd]

Local CM1? Malay "boss" in the left seat calling the shots? Inadequate or P2F FO monitoring?

Certainly this event has displayed either a company or an individual arrogance beyond belief that would allow such a deviation from crucial Manufacturer's SOP's. Both MH and Air Asia have demonstrated a shocking list of screw-ups along 3 full-on fatal crashes in the last 3 years.

[wiggy]

Having at least one navigation system that is autonomous and does not rely on a vulnerable reference seems a good idea.


Apologies before hand if I'm teaching you to suck eggs but this might be of interst to others.

Possible problems with GPS is of course why modern long range nav systems en-route don't just use GPS as a sole reference, they'll look at GPS/Inertial/conventional radio aids and in the background whilst en-route are running fancy algorithms/confidence checks, and in extremis can veto out anything it considers to be an erronous source. Of course if at the start of play the PIC is determined to "tell" the system where it is, regardless of where it really is, HAL has a problem................

I think most of us here mentioning the use of GPS are talking about it's use before flight on the ramp, as providing the reference position for aligning the inertial system. If you do that (as we do in our SOPs) there should still be a credibility check to ensure the GPS position isn't wildly in error, for whatever reason ( spoofing, lack of coverage, etc). That usually means in part selecting the electronic MAP display and comparing the real world seen out of the window with the virtual one. For example checking a 16R departure via Dunes from SYD if all was OK I'd at least expect to see the departure runway and the Dunes waypoint displayed on the map display a short range scale.

Apologies again if etc....

[Capn Bloggs]

If you do that (as we do in our SOPs) there should still be a credibility check to ensure the GPS position isn't wildly in error, for whatever reason ( spoofing, lack of coverage, etc).
Get with the program, Wiggy, it's only because of GPS we now have jets ducking down deep valleys to a 350ft minimum, GLS approaches to Cat 1 and soon Cat 2 and, of course, quick, simple, always-accurate IRS initialisations. :ok:

[wiggy]

Get with the program, Wiggy,

It's Sunday, I'll get there eventually ...:ok: (only seen "Unable RNP" on an approach the once :ok::ok: .....)

[costalpilot]

thank God:

robust management systems.....were put in place by the company following this jabberwocky.