There are some highly intelligent questions asked by the press during the press conference that are worthwhile listening to (or perhaps they are not) It will be interesting to read the press coverage in the next day or so.

During the press conference, in reply to a question, it was said that fatigue had not been found to be an issue

Hours Limit?
 

Is 98.9 hours for a 30 day period near to the limit that is allowed by Emirates?

From the press conference again

A typical take off weight for the aircraft would be of the order of 362 tonnes

The takeoff weight that was entered was 100 tonnes less.

The takeoff weight on the paperwork given to the cockpit was correct.

The error was made in the cockpit whether human or by the laptop in computing the figures.

Comment as SLF, but with a very much a technical backgound. A takeoff weight that is almost 40% more than that entered into the systems is a hell of a difference. The professionals would be able to say what that would mean in terms of the accident.

I see "The Australian's" leader writer got it utterly wrong with his headline:

Emirates A340 that scraped its tail was 100 tonnes too heavy | The Australian

One could say that - it was 100 tonnes heavier than the figure that finished up in the plane's computer systems :)

Anyway enough said from SLF for now - the professionals should take over, but I will be fascinated to follow the debate on systems design that allowed the figure to get into the system in the first place, and how the systems will be modified to decrease the likelihood that it will happen again.

The ATSB report is interesting in that it provides considerable detail on the accident, however there is no mention of any human factors mentioned or impact of same in causing the accident.... rather odd.

As you say: averages are misleading. Bear in mind that flight duty limits are designed with public safety in mind rather than flight crew lifestyle. What is commonly overlooked by the casual observer comparing flight duty to office duty is the cumulative effect of time zone changes, environment, and the irregular 24/7 nature (ie usually involves flying all night and weekends). Imagine how you felt afterwards last time you flew as a passenger on a 7, 10 or even 17 hour flight. Now do again the day after tomorrow. Repeat twice a week or more for the rest of the month. And year. Continue till made redundant or medical lost. Relocate to other side of world. Repeat. Now you are gaining an insight!

Flight time is calculated blocks off to blocks on, ie actually operating the aircraft.
EK limit is max 100hrs in last 28 days in order to commence a flight. Pretty much industry standard. This crew were at the upper end but theoretically ok to fly.

Report


Looking to get that part of the report

Just an interested observer (please be gentle). If figures mentioned before in this post are correct, the typical take off weight is around 360 tonnes and the weight entered into the aircraft systems was 100 tonnes less - around 30%.

I assume that, as the take-off was not a '100%' power (please insert correct terminology), the computer systems calculate the amount of thrust required for safe take-off based on, amongst other factors, the entered aircraft weight? Would not such a large discrepancy - 30% - be noticeable, in terms of aircraft performance, early into the the take-off? Putting it simply, if my car tried pulling away from traffic lights using 30% less power, it would be noticed immediately.

Actually, I heard listening to the press conference that the investigation was honing in on that as one of the aspects that they were certainly concentrating on.

Somewhere there has been an error made either in input to the laptop, or in the laptop software (very unlikely), or in the transfer from the laptop output to the aircraft's systems, and the design of the system cross checks for such a fundamental input, whether human or computer hasn't picked it up. As I said above, a 40% difference in takeoff weight is fairly fundamental.

ATSB report
 

The full report can be found here - it takes a while to download.

http://www.atsb.gov.au/publications/...012_Prelim.pdf

The report gives an account of what happened - with pictures and graphics. But it gives no consideration to why it happened.

There are no comments on the content of the CVR.

I suspect the press q and a would be a little more revealing.

The Sun Herald asks the "WHY" question in its report:

How was the wrong weight entered in Emirates cockpit? | Herald Sun


RAS

ITman, it's only a preliminary report.

p.s.

http://i87.photobucket.com/albums/k144/h3mpy/Scrape.jpg

:ouch:

Cross checks
 

I would like to see the program used for the T/O calcs. Especially the cross checks built in to make sure data is near correct. For instance, if it asked the flight duration, then some basic assumptions can be made about weight. Or am I assuming there are cross checks??

There has been stuff all media on the press conference so far but I notice the guy who writes Plane Talking points to a 60 second delay from brakes off to any sort of reaction in the cockpit, and that TOGA wasn't slotted into position until the plane was hitting stuff at the end of the runway.

How Emirates flight EK 407 struck it lucky three times leaving Melbourne and no-one died - Plane Talking

Not a pilot in any respect however...

As much as the documentation on this incident points to a manual error and whilst I am glad it was not a tragic outcome - I cant help but feel that more could have been done to avert this situation from a systems point of view.

I am not for a second criticising the crew although it would appear from the report that a large element of the blame seems to lay there. My concern is simply mis-typing one digit into a laptop or during transferring the data to the FMC is what is alledged to be what has been the fundamental underlying issue.

As a software developer myself and not a pilot this seems to be a case of the software and procedure biting the crew in the ass. In my own personal opinion the crew is not so much to blame as the procedure. Humans may not intend to make the odd typo here and there but we do, and yes we should be double checking and triple checking things but it is a fact of life when we look at things we tend to see what we expect to see not what is really there (check this if you want to see how good the mind can be at correcting what is actually there with what you expect to see Do You Have An Amazing Mind?) - we can be our own worst enemies without even knowing or realising it quite easily.

I think the aviation industry should not just be concentrating on what the errors were with what was entered in the FMC but more with how the software can deal with such erroneous entries. For a start eliminating the step of transfer from a laptop to the FMC would be good but it would be better if the the FMC could (maybe it can I dont know) check that the parameters entered are acceptable with axel weights - Pressures on the axels would not have to be that accurate and could probably be easily measured on oleo movement all that would be required would be an acceptable margin.

Apologies for my tuppence worth but I think the problem highlighted is alot bigger than pilot error. As the saying goes - to err is human, to really cock things up requires a computer (add in a few stages in between and your asking for trouble).

There's a really east to implement procedure that would be a great aid to avoiding a gross error, and that would be for the optimum takeoff figures for the expected runway (which is already included in the flight plan) and the planned ZFW and fuel load to be included in the printed flight plan immediately above where the actual figures are to be hand written.

These figures would not be used by the operating crew other than to provide a crosscheck with the figures they come up with after recieving the latest information (actual ZFW and runway/weather conditions.)

It wouldn't be fool-proof. (For instance, the runway actually used is frequently different to the one planned, and notams may or may not become an issue.) However, it would provide a gross error check, and I can't help but think this would be more useful than both pilots doing separate calculations immediately before engine start.

What a load of nonsense!
The procedures to ensure these mistakes do not happen have been in place for the last 40 yrs!
Apply standard operating procedures, as per your company's manual, and standard ops will occur.
The crew stuffed up, that's obvious, let's find out why! :ok:

Why did it take right up until crossing the threshold before the PF decided the situation warranted TOGA thrust?

The A345 came into service in 2002 so where did the other 33 years of procedures come from ? I am sure they were not calculating weights etc with laptops and typing them into the FMC 40 years ago especially with the A345 in mind.

These pre-emptive 33 years took into account human error I guess - your assertion is that the crew did not follow SOP's - people do however make mistakes, that you will never eliminate - but you can at least try to spot them.

One question here, how many similar events have occured using paper calculations as opposed to laptops? 2 events using laptops spring straight to mind, this and MK, none from paper.

I should add that I've never used a laptop for TO calcs only ACARs and far too many years of performance manuals.

One doesn't need a laptop plus an Fmc to work out that EW+Fuel+Pax+Freight=TOW.

If you do then I suggest you shouldn't be in the job! :ok:

Simple GIGO. The EK procedure is designed to try and trap the errors via Gross Error Checks and a final cross check of the lap top Green Dot, Minimum Clean speed and the Minimum Clean Speed in the FMS.

At 360 tons GD would have been about 265 ish, at 260 tons 30 kts or so less. If they had followed the mandated procedure they would have caught the error. However even if they hadn't followed the procedure I would hope most pilots would have the awareness to recognise that a 360 ton A345 needs Flap 3 not Flap 1, 170kts ish to rotate and 265 kts for minimum clean with a flex somewhere in the 30s not 74. That a crew did not know that is a concern and the lack of awareness maybe points to a complacent rather than questioning culture. Also reading the report I wonder how many of us would have selected TOGA having reached the 900m to go markers without any obvious signs of getting airborne.

Earlier someone questioned the level of thrust reduction. The max reduction from Flex is around 25% depending on the airport so they would have had probably 20% less thrust than expected during the take off roll.

Some may think I am harsh but I feel sorry for the guys on a personal level but not professionally.

EkwhistleBlower,
There but for the grace of God go you or I.
A professional crew that makes a fundamental error = contributing factors.
None of us are immune fom error.
And by the way, the error in your post is that at 360 tonnes the flex off Rwy 16 in Mel is 45 - not in the 30's.
So much for your awareness - now I'm concerned for you!

ekwhistleblower

Not necessarily so. I'd be interested to see the figures for that.

Depends what you've been flying recently and on what sectors you've been flying in the a/c and how tired you are. With the mix of types and the ability to be legal on the 340 but not fly the specific "variant" for months that doesn't hold true.

With the introduction of RAAS, I'll be using that as a trigger if the speed trend isn't where I think it should be.

Is there no system on the A330/40 to make a reasonable calculation of aircraft weight/balance based on oleo compression? I'm not familiar with airbus aircraft but they seem to do just about everything else! :rolleyes:

@ No Land 3 - thanks for the answer to my post - much appreciated.

900m to go markers?
 

EKwhistleblower,
There are no such markers on the runway at YMML. Where are they?

Well, if you customarily slammed your foot to the floor on the green light, you probably would. But if you just cruised off from the lights like most, I'm not so sure... I would just put my foot down a bit more, if I did notice.

That's my point though - 'User Input' - would the pilot not increase power accordingly during the take-off when he noticed there wasn't the required power? Would an experienced pilot not have a 'feel' for a take-off (ie. sluggish performance, etc) regardless of what the instruments were saying?

fred_the_red,

Taking off in a large jet, using balanced field length figures for every departure, is not the same as taking off in a light aircraft. At high gross weights all take-offs seem sluggish to start with. Then the acceleration picks up & you rotate with the end of the runway rushing towards you. Sometimes I wonder how we would ever stop from V1 in the runway remaining at that point, but you have to trust the figures.

If you look at the initial tailstrike location in the diagram in the accident preliminary report & allow for the 9 second time interval between initiating the rotate & first tail contact that is also in the report, the rotate position on the runway was about right for a normal take-off.

The extra 100 tonnes over the weight used for the calculations would have some effect on the time to reach rotate speed, but I don't think it would have been that significant. The big problem was that the speed attained was some 20kts short of what was required for the aircraft to fly at the weight it actually was.

The crew would have had very little indication that things were not right until they rotated & the aircraft did not fly. The aircraft 'always' flys when you rotate at Vr & the crew would have been wondering wtf was wrong. With the end of the runway rapidly approaching & not knowing what the problem was, the captain's first reaction was to command more pitch in an attempt to get the thing in the air. When that didn't work the only other option was to push the thrust right up & hope for the best.

When they finally staggered into the air I would be willing to bet that they still didn't know why the aircraft didn't want to fly at Vr. They may have thought of an incorrect flap setting, a screw up by the loaders & extra weight not accounted for in the figures or a number of other possibilities. It would have only been at some stage during this trouble shooting discussion that they would have realised what the problem was.

It is just extremely lucky that it was on 16 at Melbourne & they had the overrun & relatively obstacle free departure path that they had, or the result could have been vastly different.

It appears the thread could now be renamed "EK407 Accident at MEL".

I am surprised the Preliminary Report did not make reference to the crews
A330-200 experience or highlight the 37 tonne overweight landing.

In my view there was little in the report to reflect the time taken to table it.

Hi Oakape - thanks for the reply :)

Another, maybe overly simple idea (for a non-pilot), why not engage full throttle for all take-offs and adjust throttle back when safely airborne, either manually or via automated flight systems? I'll duck for cover now :ouch:.

FULL thrust
 

fred_the_red. When Max thrust is used for take-off that thrust is only used until shortly after take-off when it is reduced to Climb Thrust. Reduced thrust is used when possible to enhance engine life. The turbine section is like an elephant, it has an enduring memory and every time a temperature higher than a previous one is reached it is "remembered". The cumulative effect on the turbine determines it's operational life. Reduced thrust take-off procedures are in themselves not dangerous. What can lead to an incident/accident like Emirates is a reduced thrust setting being used when not appropriate.

Are you sure about that ?

Surely the PNF calling out rotate at a speed that was clearly insufficient would have rung alarm bells.

I can't believe they operate any ULH revenue routes where Vr could be such a low figure.

There are definitely some serious human factors involved in this, I cannot believe fatigue does not play a part.

Wouldn't increasing the pitch aggravate the situation even more, ie, cause airspeed to decay? Would have thought the first reaction would be to apply full throttle if the aircraft didn't lift off at Vr.

Anyway, it's easy for us to say all this in hindsight, but the guys upfront only had seconds to decide what to do. I'm very much appreciative of their handling of the abnormal take-off and so too are my grandparents who live in Keilor not too far off from the runway 16 centreline.

With the incorrect TOW entered into the laptop, the computed VR is also incorrect, for starters, so the aircraft will not become airborne properly when that incorrect VR is reached. (ie: a higher than normal body-angle is required or a tail-strike occurs during)

Since the thrust is appropriate for the wrong weight, the aircraft will not even accelerate correctly to that incorrect VR, let alone the correct VR.

Incorrect VR is reached further down the runway than computed. And in this case, even the correct VR was not reached on the runway.

The pilots DID increase power. The question is (or was) when.

In your car, you can determine normal acceleration quite easily. Engine noise and visual cues from being a foot above the ground are easy to judge. Even throttle position with your foot can help. The Airbus throttles go to the same position EVERY time (except TOGA takeoffs).

In an airliner, with eye-height that far above the ground, acceleration oftens looks insignificant. Then there's the added factor of night-time conditions which can reduce the visual cues to a series of points of light.

As for noticing the "feel", pilots are not jedi knights and The Force does not help them. Flying the same type on the same route, yes, you could reasonably expect a pilot to notice this sort of event early on.

But with the variations in conditions experienced on THREE different types of Airbus with many different weight ranges, runway lengths, airport elevations and temperature extremes, you can reasonably question whether or not the "feel" can be relied upon.

If you're ducking for cover, you must have some idea what sort of suggestion you're making but you make it anyway. :confused: I think it's been covered by now that the flex procedure is quite normal and not, in itself, a threat to safety. In the right conditions, this accident could've also happpened using rated thrust if a critical error had been made during performance calculations.

With this preliminary information now available, it seems to me that the airline's claims in the media of having safety as their highest priority are decidedly hollow. Given that the crew involved have been gassed by their employer, how much of a safety culture do they actually expect going forward? I can see many more ASRs going in the bin with a culture like that in play.

JMHO

I have heard some interesting stuff about this, but could someone more knowledgeable answer:

What aircraft were they rated on and when (330/340)?
What sort of ratio of time spent on each type?

Obviously what I'm getting at is, could it be that they were experienced on the 330 and entered a 330 kind of weight when in a 340, and could training/experience of the 340 have had anything to do with this error? Are we back to Kegworth (differences course)?

Apols if off target, but this thinking came from an EK driver.

Ex Cargo Clown,
The PNF called rotate at the bugged Vr, where he was supposed to. This call is a backup call & the PF is required to commence rotation at the bugged Vr, whether the call is made or not. Therefore, he/she is required to monitor the speed during the take-off as well as the PNF. I put it to you that the speed wasn't clearly insufficient, as there are a number of occasions where the speeds are discussed prior to the aircraft commencing the take-off roll, & the crew obviously thought that they were sufficient at these points in time! They are only clearly insufficient with 20/20 hindsight. However, the fact that 4 pilots did not feel that the v-speeds that were generated were somewhat low is something that should be looked at. Many things could lead to this, including cross-type qualification, distraction, inadequate cross check & fatique, to name just a few. To try & pin down a reason at this stage would be pure speculation.

Of course there are human factors involved in this error. I would put it to you that this part of the accident chain is purely human factors related, as opposed to a technical issue.

JulieFlyGal,

This is a very good question. From the preliminary report the aircraft started to rotate at 152 kts. 8 seconds later the aircraft was still accelerating as the speed was 156 kts when the tail contacted the runway. A slow acceleration for sure, but still an acceleration. However, from this point, the aircraft did not get above 157 kts until after it was airborne. So even though maximum thrust was applied, the aircraft did not continue to accelerate - possibly due to excessive pitch, or due to the extra drag from the tail scrapes & the main gear rolling over grass rather than paved runway, or a combination of both. Perhaps someone with a good knowledge of the theory of flight could shed some more light on this.

YoDawg,

I agree, but still feel that the difference to the physical position on the runway that Vr was reached wouldn't have been large enough to have a visual impact on the pilots during the take-off roll. Particularly at night & if they hadn't flown this sector off rwy 16 at Melbourne for a while. My reasoning for this is based on the position on the diagram in the preliminary report that shows the location of the start of the first tail strike & guessing how far before this point the crew commenced the rotation, given the fact that the rotation was commenced 9 seconds before this point.

Perhaps some of the geniuses on this forum could compute the distance from the commencement of the take-off roll to the rotate point, using the figures that the crew used & also using the correct figures. I would be interested to see what the difference in distance would be.

Virtually similar cockpit but so much different aircraft regarding the weights. One guy can enter 262 tonnes and feel very confident as it is pretty close to MTOW for the 340-300 he may have flown the week before ...

Flying so different aircrafts but from so similar cockpits must be part of the problem.
Add up fatigue and that's when you turn complacent.