Wizz Air A321 CG was off the chart

[Check Airman]

Report: Wizz UK A321 at London on Jan 16th 2020, late rotation on takeoff (http://avherald.com/h?article=4dd9ae59&opt=0)

[PilotLZ]

A Vr of 112 knots for a not fully loaded but well and by far not empty A321 should have rung a bell as early as at performance computation stage. The whole set of performance data was invalid.

Also, as people well acquainted with the matter say, Wizz air have a highly customised normal checklist which does not include an after takeoff checklist. Surely reading the standard Airbus checklist to the line after flaps retraction wouldn't have allowed them to climb to 5000 with the gear down?

[giggitygiggity]

Does an after takeoff checklist appear on what Airbus actually produces? Genuine question. At my large A320 operator we don't have one.

[PilotLZ]

After takeoff/climb checklist:
Landing gear - up
Flaps - retracted
Packs - on
--------------------
Baro reference - STD set

That's the Airbus version of it, as far as I am aware. Everything else, including removing this section, is an airline-specific customisation. And that's the trap with deleting it - probably because whoever designed the SOP and checklist thought that raising the gear was too intuitive to forget.

[Roj approved]

We have this situation a couple of times a month, and my first thought is always pax seating and underfloor loading. Followed by fuel required.

I always discuss the pax seating with the Cabin Manager and get them to check if it “looks ok” before we close the door. (Ie: pax distribution across all 3 zones.

My company has their own iPad based loading system, which is far from perfect and doesn’t populate any of the Airbus performance data.

So it is always a challenge, and requires extra vigilance for loading and operations.

That doesn’t stop these similar events from happening, but touch wood, I’ve managed it up to this point.

They were lucky this time, it could have been a lot worse.

[macdo]

Pretty sure there is no published after take off check list in the standard Airbus SOP. We used a customized one about 15 years ago, then ditched it to align more closely with the Airbus SOP. Seem to remember that the aircraft will warn you if you've left the packs off.

[Qwark]

Airbus most definitely has an AFTER TAKE-OFF checklist. See Pilot LZ reply!

Airbus are currently revising many items of the normal SOP and I believe the AFTER TAKE-OFF checklist will be removed. These changes are delayed but due in the second half of next year.

[ManaAdaSystem]

As far as I know, this is the new CL, due to be released late next year. No after takeoff checklist.
https://cimg9.ibsrv.net/gimg/pprune.org-vbulletin/2000x1504/16d54f8e_0041_4e0e_a319_5a788ac2c2b6_251b6714d7a429e1772df5e a0a29fe5c29cb4ac3.jpeg

[Check Airman]

Interesting. No more Autothrust on the landing checklist?

[Flying Clog]

How does that old saying go?

You pay peanuts...

[DaveReidUK]

Apropos the above:

"Following the incident, the operator carried out an internal investigation. It identified safety actions it would take to prevent a reoccurrence, which were to:

... Improve Ground Handling Agents' awareness of the implications of a change in aircraft variant.

... Provide additional training for cabin crew on weight and balance distribution and its affects [sic]."

[CW247]

Macdo, TCX were operating a very slick and cut down Airbus SOP. Every other operator out there has an After TO and Approach checklist that is straight from the official Airbus books.

[mcdhu]

Indeed, the low V1/Vr of 112kts should have rung alarm bells! I seen to remember from my A320 days at LTN that the Vmcg/Vmca was around 115kts, but surely the giveaway should have been the aircraft registration on the load sheet handed to the captain. Yes, a good SCM should have flagged the unusual distribution to the pilots, but no more than that. I'm afraid the fault lies firmly with the dispatcher and the pilots.
a narrow squeak!

[sonicbum]

Airbus has an after takeoff/climb checklist definitely since my initial type rating in 1993 and I honestly do not understand why any operator would want to skip such an important part of the checklist. L/G left down is highly common for many distractions occurring during takeoff, including EFTO slightly mismanaged with some sweating where, once the a/c is under control, the crew focusses on the EFP and forgets the gear.
Rotating an A321 with 157 pax at 112 kt means a very very low understanding of basic aerodynamics...

[Nil further]

10000hrs on type. Never seen an after T/O check list nor is there one anywhere in the manuals that I can see (worlds largest Airbus single aisle operator)

[TheEdge]

?? well that sounds awkward to say the less

[tubby linton]

The manuals for your operator have had a huge amount of Airbus material removed from them, not all of which has made it into the OMB.

[TheEdge]

PilotLZ

Cannot understand how this did happen...I mean you get a different aircraft, a 321 is different in everything in terms of ZFW, TOW, MLW; You have your FlySmart tool and Inflight performance application and do your loadsheet calculatioin, you insert your cargo and pax weights and distribution check the CG, the THS data, the TOW, the ZFW, double check with your colleague, insert in MCDU INIT page, then you do your TakeOff calculation, PERF page, then check again same distribution numbers with Ramp Agent, SCCM and so on and so forth.
Even if it was a manual calculation you would have immediately spotted the CG out of range.

Am i missing something ?

[booze]

I think two reasons could lead to this (stand to be corrected): misloading of aircraft by loading staff and flight crew complacency entering and crosschecking perf.data.

[tubby linton]

Am i missing something ?

You are missing the fact that the loadsheet was generated external to the aircraft . I would imagine that it was probably just a print out on poor quality paper using a life expired print cartridge with no envelope depiction printed on it.
“When all the passengers have boarded, the Load and Trim Sheet is printed, and a copy is passed to the flight crew for them to complete their performance calculations.

[TheEdge]

So it wasn't the Flight Crew doing it ...and no quality/safety check done afterwards. They were lucky this time.

[Sidestick_n_Rudder]

For those who are quick to jump the gun and blame the crew. If you care to read the actual AAIB report, you will find that:

- the crew was presented with a A321 loadsheet indicating a CG within limits. The loadsheet assumed even distribution of PAX throughout the cabin, whereas in reality everyone was seated in the front.

- there is no mention of incorrect takeoff weight or performance calculations. By all indications the V-speeds were correct

​​​- forgetting to raise the gear after barely leaving the ground (mind you intersection t/o at LTN means around 1800m TORA) is a minor error, completely irrelevant to the incident

[Time Traveller]

Don't most airlines include a final closing up cross check from the cabin crew confirming approximate passenger seating distribution? Mine does.

[seventhreedriver]

Sidestick_n_Rudder

Finally someone understanding English or actually taking time to read the report before calling the crews something connected with peanuts... Thank you Sir!

[andrasz]

All those rows empty in the back should have prompted any properly trained cabin chief (who are there primarily for safety, not to serve passengers, as we are always reminded when their service attitude is questioned...) to pop their head in the cockpit and query whether that fits with the information at the pointy end. Of course, as someone above aptly said, if you pay peanuts...

[tubby linton]

‘By all indications the V-speeds were correct”

Have a look in the fcom and there are some speed tables based on VMU and VMCA and compare them to the figures given in the report.

[sonicbum]

Sidestick_n_Rudder

There is no way a Vr of 112kt on an A321 makes any sense unless You are ferrying the aircraft on a short sector (and also in that case probably with Conf 2 or 3).

Being a Captain means also having enough experience and gut feeling to understand that something’s definitely not right and you must break this Swiss cheese that is forming up just before your eyes. They were also very lucky they didn’t leave half of the tail on the runway.

[bentbanana]

tubby linton

Doesnt the departure system flag a warning if the aircraft is overloaded or out of trim? Could the load controller have changed the seating to free seating in order to produce the loadsheet - and forgotten to tell the cabin staff - or if he did they didnt action it?

[aviationvictim]

Seems to be a very large split between the Vr and V2. A V1 speed of a 112 would not be uncommon in Ltn from the intersection but as you say a Vr speed of a 112 is virtually impossible unless it’s empty. Could there be some confusion in the report regarding V-speeds? Seems very odd to me.

[Flying Clog]

You can make all the excuses that you want, but the fact that this happened to Wizz, as opposed to Ryan or Easy, doesn't surprise anyone.

Sure, at better, safer, more experienced 320 operaters things happen. But when it does, it's surprising, and dealt with.

With Wizz, nope, no way I'd put my family on there.

[A321drvr]

All euro-locos had their fair share of cockups without any serious outcome, fortunately. Just think about EZY's intersection incident in Portugal, or RYR's NDB approach in France. WZZ's recent stint in LTN is no different. However their EIN-SKP departure and continuation of flight, following a birdstrike causing temporary unreliable airspeed indication raises a few questions about the company culture...

[Check Airman]

For what it’s worth, I can see why the crew left the gear down. They were obviously distracted by what had just happened. That’s more or less inconsequential in the grand scheme (in my opinion).

I’m curious about the V speeds though, as they seem way too low. The report doesn’t mention anything though.

[TheEdge]

oooh yes thanks sonicbum

[Sidestick_n_Rudder]

I don't have acess to A321 performance data, as I have only flown the A320, and that was some time ago. However:

- if anything, the Vmc speeds on the A321 should be lower than on the A320 due to longer fuselage. Correct me if I am wrong

- the takeoff was conducted with a light-ish airplane on a short, wet runway. It is very possible in such conditions to have the V-speeds limited by Vmca/Vmcg/Vmu and be lower than what we are normally used to see. On longer runways the V-speeds are usually higher in order to improve climb performance.

According to the report mentions V1 of 112kt and Vr of 123kt @68.6T. Does it really sound so low?

Can anyone post published Vmcg and Vmca for the A321 and stall speed in Conf3 for 68T? I remember there used to be a stall speed graph in the FCOM. Min V2 can be as low as 1.13*Vs1g

Or even better, do a calculation in FlySmart? The wx/rwy conditions are described in the report.

[Check Airman]

You know, you’re right. Avherald says V1/Vr were 112 and v2 was 123. The actual report says v1 was 112 and vr was 123. That’s more reasonable.

[DaveReidUK]

Several of the reader comments in the Avherald report also query the quoted Vr. Hopefully the article will be corrected in due course to use the figures from the AAIB report.

[sonicbum]

I believe they did correct it now.
Definitely a wrong Vr of 123 becomes tricky.

[pineteam]

Interesting. No more Autothrust on the landing checklist?

Never understood why it was there at the first place. Good thing they finally removed it. This aircraft can be flown entirely without checklist in my opinion especially the newer models where you have an Ecam warning for everything. All the important items are displayed on the Ecam memo. Gotta love Airbus for that.:)

[CW247]

So accounting for all the errors in the reporting, understanding this was a lightish A321 off a wet runway (therefore not totally unexpected speeds), we now have a situation where the only way this could've been trapped was by having knowledgeable and safety aware cabin crew. At Wizzair and most LCCs, this means you are expecting three 18 year olds and one 25 year old with barely any life experience, let alone aviation experience to speak up. At my previous legacy airline, at least 2 of the crew would've picked this up in a flash but they're in their 30s and 40s and cost far too much.

[vilas]

The first comment on the report in Avherald makes it very scary. It could have ended in disaster. The RW is short with no overrun the crew wouldn't consider abort takeoff. The aircraft which is in direct law on ground got airborne only because of the thrust/weight couple created by TOGA. Otherwise they would've overrun with takeoff thrust. Providential escape.

[Rodney Rotorslap]

bentbanana

Do I understand correctly that the A320 has 3 cabin zones and the A321 has 4 cabin zones?

This being the case, the late reallocation of an A320 load to an A321 would have resulted in Zone D being empty. The 42 pax shown in Zone D on the loadsheet would have previously been seated across Zones A, B & C on the A320. During the change of gauge to A321 there must have been 44 pax moved into Zone D to achieve trim. Evidently these pax were not issued with new boarding passes and a communications breakdown has resulted in their remaining in Zones A, B & C. With additional seats in the first three zones of the A321 there were probably no seat dupes to raise alarm and if there were they were probably just told to sit in the vacant seats in Zones A, B & C. What should have raised alarm was the vacant Zone D but the cabin crew probably assumed that it was deliberate for trim.

[DaveReidUK]

I believe they did correct it now.

Yes, it's been corrected to 123 KIAS Vr, although the byline still shows it as "last updated" Thursday afternoon, in the hope that nobody will notice. :O

[DaveReidUK]

Do I understand correctly that the A320 has 3 cabin zones and the A321 has 4 cabin zones?

This being the case, the late reallocation of an A320 load to an A321 would have resulted in Zone D being empty.

You understand correctly - see post #25.

From the report:

"Passengers were boarded with their seat allocation for the A320 and therefore were seated within cabin Zones A, B and C. This left the seats at the rear of the A321 aircraft, which has a fourth zone, Zone D, unoccupied. The unusual passenger distribution was not noticed by the cabin crew or dispatcher. The aircraft commander was unaware of the passenger distribution in the cabin but was passed a Load and Trim Sheet for his A321 aircraft, G-WUKG"

[john_tullamarine]

A couple of thoughts ..

(a) This being the case, the late reallocation of an A320 load to an A321 would have resulted in Zone D being empty.

Unless a trimsheet be designed to cover multiple models (and that may be both reasonable and feasible for some aircraft) a change in model surely requires a reworked loadsheet ? One presumes an adequate level of competence in those charged with the execution of load control. We have checks and balances and multicrew operations to provide for a high probability of success, providing everyone does a reasonable job. In the simplistic view, the crewmember signing off the completed loadsheet ought to be able to detect such gross errors.

(b) I see some comments regarding the non-intuitive nature of loadsheets. This is especially so with the typical electronic summaries we have seen for many years, now. However, there are various simple ways to get around Murphy's antics ...

I was most impressed with a very simple system we came up with to defeat a lot of these sorts of problems which was implemented for IPEC freight operations in the 80s.

The freight shed did their thing using a normal paper trimsheet - but my comment would be equally pertinent regardless of whatever system might have been used - even an electronic ACARS style sheet such as I was routinely familiar with flying for another operator.

The critical thing was that the IPEC crew then ran an independent check using the load summary to check the trim with a whizz wheel trimsheet. One pilot called the loads in sequence and summed the totals as an independent check. The other pilot ran the whizz wheel to check the trim and the whole exercise took no longer than the time it took the first pilot to call the numbers at a measured pace. In addition, as was typical for freight operations then, one pilot physically ran a can check on weights to detect any gross misloading sequence errors. This is entirely analogous to the cabin senior's running an eye over the cabin and (ideally) then running a quick brief with the cockpit to make sure that the reality matched the storybook to a reasonable degree.

Really, folks, there is no reason, or excuse for, why the load control answers can't be reliably accurate and consistently so.

In the same vein, there is no valid reason why operator training should not ensure that the flightcrew is alert to gross speed schedule it-doesn't-look-quite-right calls. I have no background playing with Airbus but, in another age, we would have picked up alarm bells (on Mr Boeing's products) such as speed schedules encroaching on Vmca/Vmcg regions of the envelope without any undue problem or delay.

[FlexibleResponse]

The first comment on the report in Avherald makes it very scary. It could have ended in disaster. The RW is short with no overrun the crew wouldn't consider abort takeoff. The aircraft which is in direct law on ground got airborne only because of the thrust/weight couple created by TOGA. Otherwise they would've overrun with takeoff thrust. Providential escape.

The Captain saved the day by selecting TOGA which provided an UP pitching moment as pointed out by vilas .

The other thing they could have done was to move the trim wheel to increase UP stabilator (the stabilator is frozen at the TO Trim setting in direct law until airborne).

When you run out of elevator authoriy (or up/down pitch stick control authority) in an all-moving-tail (all-flying-tail) aircraft (most airliners), in any phase of flight, you must manually move the stabilator to regain control.

The same applies to datum trim aircraft (which then need to be re-trimmed).

[Roj approved]

You understand correctly - see post #25.

From the report:

"Passengers were boarded with their seat allocation for the A320 and therefore were seated within cabin Zones A, B and C. This left the seats at the rear of the A321 aircraft, which has a fourth zone, Zone D, unoccupied. The unusual passenger distribution was not noticed by the cabin crew or dispatcher. The aircraft commander was unaware of the passenger distribution in the cabin but was passed a Load and Trim Sheet for his A321 aircraft, G-WUKG"

Interesting, we only have 3 zones for both 320 (1-10, 11-20, 21-31] and 321 (1-10, 11-25, 26-39).

So we could use all the 320 data (Zone, Underfloor, Fuel) and enter it into our 320 e Loadsheet by mistake, and if pax sit in their assigned seat, this forward CofG is still a problem as the zones might appear correct, but 5 rows of zones 3 are actually sitting in zone 2 and the last 8 rows are still empty.

Once again it comes down to the Cabin Crew to eyeball the cabin and recognise a problem and the Pilots to check the paperwork, carry out the cross checks and be vigilant to the 320/321 differences.

I haven’t flown for 6 months, or studied, and am racking my foggy brain to try to remember any other mitigators for this problem

[krismiler]

Strikingly similar to this: Incident: Jetstar A321 at Melbourne on Oct 29th 2015, difficulty to rotate aircraft for takeoff (http://avherald.com/h?article=4904383d&opt=0)

Incident: Jetstar A321 at Melbourne on Oct 29th 2015, difficulty to rotate aircraft for takeoff
By Simon Hradecky, created Thursday, Dec 3rd 2015 16:12Z, last updated Friday, Jan 18th 2019 17:46ZA Jetstar Airbus A321-200, registration VH-VWT performing flight JQ-976 from Melbourne,VI to Perth,WA (Australia), was accelerating for takeoff from Melbourne's runway 16 when the crew needed to apply nearly full back pressure on the side stick to rotate the aircraft. The aircraft climbed out to safety, the flight crew requested cabin crew to count the passengers on board and note down their seat distribution. The results were recalculated and the flight crew found, that the aircraft had been outside the aircraft's loading limits. The passengers were redistributed within the cabin to bring the center of gravity back into the aircraft's envelope. The flight continued to Perth for a safe landing.

On Dec 3rd 2015 Australia's TSB rated the occurrence a serious incident and opened an investigation, the final report is being estimated for October 2016.

On Jan 18th 2019 the ATSB reported that the investigation was discontinued. The ATSB argued that the organisational context within the operator had changed significantly in the last 3 years and any finding would thus no longer match the current situation.

In the event summary the ATSB stated: "It was later established that passenger seating allocations had been determined using the seat map for an A320 instead of an A321." After reseating 6 passengers to the aft cabin the crew recomputed weight and balance and proceeded for an uneventful remainder of the flight.

[john_tullamarine]

Interesting, we only have 3 zones for both 320 (1-10, 11-20, 21-31] and 321 (1-10, 11-25, 26-39).

Entirely a matter for the operator and the weights engineering folks to determine how their loading systems are designed. The problem here appears to be that the relevant folks didn't recognise the mistakes made which ought to have been recognised at the time. Mistakes are part of the experience and we have to accept that they will occur - trapping mistakes before they become somewhere between embarrassment and catastrophe is the essence of what we do.

[Check Airman]

Never understood why it was there at the first place. Good thing they finally removed it. This aircraft can be flown entirely without checklist in my opinion especially the newer models where you have an Ecam warning for everything. All the important items are displayed on the Ecam memo. Gotta love Airbus for that.:)

I understand it’s there because a few people have tried to land in OP DES.

[Longtimer]

AAIB investigation to Airbus A321-231, G-WUKG Slow response to aft side stick control input at rotation, London Luton Airport, 16 January 2020. Evidently the rotation problem was due to not correcting the passenger seating when the 321 was subbed for the 320.]
https://assets.publishing.service.gov.uk/media/5f5a0251d3bf7f7238f22fd2/Airbus_A321-231_G-WUKG_10-20.pdf

[PilotLZ]

Check Airman

The original design concept of the A320 was that it would never have any QRH or paper checklist and that it would only fly in managed modes. For self-explanatory reasons, none of that materialised, especially given that all the 1980s technology available back in the day was not even close to what we have now.

[Denti]

Have flown the A320 for a while with an operator SOP that had no check lists between after start and parking checklist, and generally a silent flightdeck philosophy. Works very well indeed. But there is many ways to fly the same plane.

[DaveReidUK]

Interesting, we only have 3 zones for both 320 (1-10, 11-20, 21-31] and 321 (1-10, 11-25, 26-39).

Entirely a matter for the operator and the weights engineering folks to determine how their loading systems are designed.

Zone D on the Wizz A321ceo is, in effect, a virtual zone that simply designates those row numbers (31-39) that are present on the A321 but not on the 30-row A320.

It's hard to see how there can be any physical demarcation between the 28" pitch seats in Zones C and D, respectively, other than at most a curtain - which would be pretty pointless in an all-Y layout.

In which case, the criticisms being levelled at the cabin crew amount to them failing to twig the significance of the fact that only the first 6 rows of the 14-row cabin aft of the E/E were occupied, which seems a tad unfair.


https://cimg5.ibsrv.net/gimg/pprune.org-vbulletin/347x1102/wizz_a321ceo_7b30a08aee2942723c784ccba0c623692955382a.jpg

[john_tullamarine]

The critical consideration, though, is that the OM loading system protocols must define, adequately, differences between, and procedures for, different models to avoid screw ups. How this might best be done in a matter for the ops management, tech services, and weights engineering folks.

Point is that, if there be no formal protocol to avoid problems, then problems will arise as surely as day follows night.

I can't comment on the Airbus as I have no background, either flying or engineering, with the aircraft. However, when it comes to weights engineering work, an aircraft is an aircraft is an aircraft and differences are unremarkable.

[Flocks]

In one of my previous operator (UK), for every flight the cabin crew were doing a counting of number of passenger per zone and the number one was giving me a paper with the 4 zone and each number of pax per zone.
it was really easy to check now with the load sheet the number of total passenger and each zone, easy to spot if passenger moved or was a mistake.

Most of the time some passenger had moved and sometimes when it was clearly impacting the CG to my experience we would ask the CC to move some passenger, for example, 3 pax from zone A to D ...

My last operator, we don't do that and it is even not mandatory to tell me the full number of pax on board (it depends of the destination, what I don't really understand)

If it would be mandatory for all EU (world) operator to do what I just said above with mandatory CC counting, this kind of incidents with CG would be unlikely to happen. But as always in the world, until we don't have a big crash at the end of the runway with media coverage and the world been shocked, nothing will change ...

[BDAttitude]

So, who else is doing load sheets with Excel tables?
(Very intrested for my personal no fly list)

[scr1]

Fly Be used to a few years ago

[FlyingStone]

Would you prefer manual loadsheets? Those are definitely error-proof.

[BDAttitude]

In a way yes, because I would expect a second pair of eyes looking over those.
Over the years I have seen many Excel "applications". I have even seen a closed loop simulation in Excel - albeit only with fixed step solver.
They all had one thing in common: The were perfecly good when they were working. However if somebody inserted a row, broke a link or Excel would not update a cell for some unknown reason (which I have experienced) they would fail - and the failure could only be discovered by critical awareness of the individual before the screen.
Add the possibility of these copy and paste errors as we've seen here.
So yes, perfecly inadequate for loadsheets.
QED

[Flying Clog]

The mark one eyeball, and experience, comes into play here.

[Bus Driver Man]

Wizz doesn’t use Excel loadsheets for operational use. Only a computer generated loadsheet from ground handling or a manual loadsheet.

[PilotLZ]

There's no going back to manual loadsheets. In the race for efficiency that the LCCs lead, every little bit of the process which can be optimised will be optimised - and that involves time spent doing calculations with a pen and paper. Not to mention that use of performance software allows for greater integration of the process into the overall operational system if done correctly. And, frankly, manual loadsheets are not immune to errors either.

So, I don't think that it's a matter of whether the software should be there. It should and it will. But the lessons learned from this case should result in more and better safeguards within it. The computer programme used for loading is not an Excel sheet cooked up by someone off the street. It's a very mature piece of software which has passed all the necessary approvals and has an excellent track record globally. However, that shouldn't mean that the support team shouldn't have some thoughts about how to prevent this from happening again.

This is not entirely a matter of mass and balance. It wouldn't have happened if the cabin crew had noticed the unusual load distribution. Having 160 people in the first 30 rows and nobody in the last 9 rows should ring a bell - and be immediately reported to the Captain as unusual. So, improved cabin crew training should also be on the cards.

[A320LGW]

You can make all the excuses that you want, but the fact that this happened to Wizz, as opposed to Ryan or Easy, doesn't surprise anyone.

Sure, at better, safer, more experienced 320 operaters things happen. But when it does, it's surprising, and dealt with.

With Wizz, nope, no way I'd put my family on there.

Absolute horse manure.

Report: Easyjet A320 at Lisbon on Sep 16th 2019, wrong performance calculation results in late takeoff (http://avherald.com/h?article=4daf8fe0&opt=0)

Report: Easyjet A319 at Nice on Aug 29th 2019, takeoff with insufficient thrust, both pilots made same error (http://avherald.com/h?article=4d5af6cf&opt=0)

Report: Easyjet Europe A320 at Lisbon on May 7th 2019, takeoff with insufficient thrust (http://avherald.com/h?article=4d2081e6&opt=0)

Report: Easyjet A320 at Lisbon on Apr 24th 2019, takeoff with insufficient thrust (http://avherald.com/h?article=4d208110&opt=0)

Incident: Easyjet Europe A319 at London on Sep 30th 2018, error in takeoff calculation (http://avherald.com/h?article=4bed5415&opt=0)

[Nick 1]

To those that state that ‘d be cabin crew task to check the distribution of the pax , according to my flights with wizz ,they are very prone and aggressive to check and remove pax that did not pay for the proper seat instead . And probably most of them dont’ even know the danger related to improper pax distribution CG wise in a plane. But is not their fault , is company culture.

[vmandr]

"unusual passenger distribution"
nothing unusual per se here. the lc-system recalculated per new registration and redistributed pax to all 4 zones I read 42 in 0D.
having said that, definitely someone failed to notify the gate supervisor to change (paper ?) boarding passes for 42 known pax, (time consuming, creating frustrations and delays),
or allocate 42 seats at random, again creating frustrations and delays,
or if free-seating was decided, (justified by the excess capacity) someone failed to alert the cabin crew supervisor (to block seat-rows) and advise all pax.
the report is unclear in this.

"not noticed"
'by the dispatcher'
meaning the person preparing the loadsheet, load-controller/planner (usually far from the gate/aircraft) ? Well they never come on board to verify their work nor they are required to.
supervision of loading and pax boarding is supervised by the ramp agent / loading supervisor, usually handling 3-4 flights simultaneously !

'by the cabin crew'
no EASA or FAA legislation - to my knowledge - exist, mandating this short of checks by cabin crew.
it is a matter to be dealt with by the CC Training Dept and that points to company culture.
in the old days cabin crew were better trained and also received a copy of the loadsheet, even LCC, so there was better awareness.

[dc9-32]

When I was a despatcher during the 1980's, and as well as checking the holds after loading, I made it my job to check the cabin for pax distribution and cross-check, then head into the cockpit for the load sheet sign off. I wasn't required to check the cabin, I just did it because I was taught by someone who took me right back to basic math and the effects of weight versus balance. I carried over that mentality to everyone I trained.

Empty aircraft prevent challenges sometimes simply by design. If they had a T tail, they were worse than the engine/wing configuration we see these days. BAC1-11's were notorious for needing at least 400kg ballast in the forward hold in order to get the ZFW just within limits. Once fuel is added to the centre tank, the C of G comes forward, but not by much. The centre fuel was often "trapped" and not used in trip fuel calculations.

Pax day trip flights on 1-11's were also a challenge simply because the pax had no luggage.

Everything in those days was done with a pen, ruler, calculator and a piece of paper they call a load & trim sheet. The only computer aid we had was the calculator (and our brain).

[andrasz]

When I was a despatcher during the 1980's...

In those good old days the dispatcher was fully responsible for ramp and load control. Now in the name of efficiency load control is sitting somewhere in a windowless office, possibly off-airport, and the ramp agent gets a loading instruction, completely disconnected from W&B. There are less and less dinosaurs around who actually understand the process from start to end...

[pineteam]

I understand it’s there because a few people have tried to land in OP DES.

That’s a serious violation of Airbus auto flight limitation and lack or situation awareness to find yourself in that situation. Never heard that one before.:}

[Check Airman]

That’s nothing. A few months ago, a crew landed with the thing in OP CLB!

[Dannyboy39]

I'd be curious to find out where the exact rotation point was from RWY 26 - there is not a lot of room for error at LTN, with terrain at the end of the runway also. Considering in the last few years where there has been a significant increase in air traffic from the airport, crew / ATC are not requesting full length departures in most cases reducing the available run to around 1900m if I recall correctly.

[DaveReidUK]

In the absence of an FDR/QAR trace, we're unlikely to know the answer to that.

If it's any help, FR24 shows the aircraft passing over the 08 (as was) piano keys at approximately 125 ft AGL.

[vilas]

pineteam

Nor did the pilots who were trying to land knew they were in OP DES.

[vilas]

Check Airman

What is this story now? Triggered OP CLB while setting MA altitude, disconnected everything and tried to land?

[Bus Driver Man]

Dannyboy39

1771m to be precise. Full length is 2162m.
However, full length doesn’t necessarily mean that there would be more stopping margin or runway available after VR. The flex temperature will probably be higher, resulting in an increased TOR and ASD. Selecting TOGA will give you more margin on full length compared to an intersection of course, but in this case, the crew selected TOGA after VR. The remaining runway length could have been the same with a full length TO.

[Icarus]

I’m surprised there hasn’t been more focus on the (to me) lack of procedural integrity within the ground handling processes taking place here, this ‘oversight’ has been around for decades when disparate systems are used at check in and at load control.

A gap very much appears to exist between the check-in side and the ground operations (load control) wing in that there seems to be two ‘systems’ in play and that either one computer system is managing only one portion of this integrated/overarching process or, two separate computers are managing each side independently and without sufficient integrity of ‘hand off’. It does seem to be the latter as it is accounted for in the first two of the seven ‘Safety Actions’ noted in the report coupled with the airline offering online check-in and ‘software’ was used to produce the loadsheet.

It is reasonable to deduce from the report is that what was missing here was a ‘passenger distribution’ communicated by passenger services to load control when they handed over the details of the passenger breakdown and baggage numbers nor was it solicited from passenger services by load control. How that handover was executed is also not apparent in the report; was it a phone call, radio call, was it the physical passing of a print out from the check-in system with passenger data on, was it a view of the passenger data held in the check in system taken by load control? I note there are no safety actions directed at any computer system improvement which may indicate that the data was available but somehow failed to be communicated across the parties.

It would be interesting to see the relevant sections of the Ground Operations Manual and/or Passenger Handling Manual to see if this process is outlined adequately.

This simple information exchange, if conducted with integrity, should -would- have discovered the seating problem causing the undesirable balance condition and providing some opportunity to get ahead of the curve of this problem and being able to take appropriate action in advance of departure.

It also means that load control quite probably produced, not necessarily intentionally, but certainly by omission a flight safety document based upon assumption and not fact (I’m trying to be careful with language here) and as such could be subject to UK ANO Section 256(6) and its affixed scheme of penalty (L5). The balance of the evidence suggests load control assumed a passenger distribution. It is of course possible that the Excel solution automatically ‘assumes’ a certain passenger distribution (personally I fail to see any good reason as to why it should do) yet on the basis that no software demands have been made in the safety actions this does not seem to be likely; a look at the Excel ‘screen shot’ in the report suggest those cells are vacant for input and therefore not automatically calculated within Excel to determine an assumed seating distribution from the total passenger number - that data was most likely entered via a keyboard and based upon assumption and not fact.

Let’s also not overlook that fact that this same scenario and therefore exact same risks can extend into all versions of a ‘system’ other than a fully integrated one; those being (a) manual check in, manual load control, (b) manual check in, automated load control and (c) automated check in, manual load control not just this scenario of automated check in, automated load control performed on two different (and not integrated) computer systems. Yes I will refer to the use of an Excel spreadsheet as an automated system despite my surprise at the fact of its use for these purposes, which therein lies another problem in most part created by the LCC solution providers - they generally ignore this significant component of airport operations yet still still call their check-in and boarding solutions a ‘Departure Control System’, when they really are not.

Excel is not an acceptable solution when a fully functional load control solution with an open API (to any/all check in systems) could be developed and on the market for comparatively very little money and competing with the likes of the major ‘DCS” providers who LCC’s usually steer clear of due to cost. Airlines who are customers of most (if not all) LCC solution providers in today’s market should not just note this they should demand it from their “DCS” provider. Far cheaper than a chunk of metal at the end of a runway.

[Check Airman]

vilas


Yes. Something along those lines. Somebody set the FCU too high, and then it all went downhill from there. Only instead of turning EVERYTHING off, they continued to mess around with the FCU and wound up landing very, very long.

Had they disconnected everything, it would have been nothing more than a minor disturbance.

[andrasz]

It is reasonable to deduce from the report is that what was missing here was a ‘passenger distribution’ communicated by passenger services to load control...

Icarus I think you have it the wrong way around. The aircraft change happened after all the boarding cards were issued, and the pax system correctly seated all the pax in the forward cabin, corresponding to the 320 assigned seat numbers. Load Control then had to manually re-distribute pax to get W/B within limits, just forgot to tell gate & crew about it. The problem seems to be that it is possible to manually re-seat pax without re-printing boarding cards (probably by adjusting zone totals as opposed to actual re-seating). Likely this feature was intended to accommodate LMC while pax already on board, and not for any pre-boarding changes, but like any loophole, it is open to abuse.

[Icarus]

However, that still does not negate the passing or asking of the seating distribution when passenger breakdown figures are passed from passenger services to load control when you have a non-integrated EDP system in play - that remains (in my opinion) the fundamental process breakdown and the primary cause of this error and eventual problem.

As for,

"Load Control then had to manually re-distribute pax to get W/B within limits, just forgot to tell gate & crew about it."

If ​​t​​​​his (as you posit) is correct it suggests that a seating distribution was passed and entered in the load control software solution which then presented the out of trim circumstances to load control, if true, then as such the first thing one would do as a load controller is tell passenger services of that issue immediately and the need to reseat or whatever other corrective action was required. Failure to do so would be criminal - for want of a better word. To actually go on to prepare and present a load sheet under those circumstances knowing full well the action you took to produce it and to not tell anyone else ... well.

[zondaracer]

10000hrs on type. Never seen an after T/O check list nor is there one anywhere in the manuals that I can see (worlds largest Airbus single aisle operator)

You work for American Airlines? Because they do have an after takeoff checklist on the Airbus.

[andrasz]

Icarus the handling system IS integrated, any check-in entries will correctly be passed on to the L/C module (as it happened in this case). As for your assessment of the subsequent events, I agree entirely, there is a good reason pinkair is firmly on my no-fly list (actually several good reasons...).

[Icarus]

andrasz If that is the case then I cannot see how this can happen - integration suggests the sharing of the same seat plan within or across systems; even in disparate systems integration is continuous or regular communication; not electronic transfer of data one-time at flight close. This is also the essential dispute in the report. Further illustrated in the report by the outline of W&B being completed on an Excel spreadsheet - unless this is for illustrative purposes only. However that would then raise the questions to why the report does not show the actual documents or simulated screen shots of the condition.

If as you suggest it is fully integrated, then this suggests both parties (passenger services & load control) had the same seat map (320) in the solution up to and including the time load control made the change to 321 from 320. In an integrated solution that process would fail if occupied seats on the original (320) seat map did not exist on the new seat map - to simply enforce a positive reseating of mismatched passengers [seats] and new boarding passes issued. If all seats did map across 320/321 the process would execute without error and load control would know exactly the seating condition on the new aircraft (321) and the resulting out of trim situation.

This is now the most interesting and critical part - if this were the case, which would be the case in an integrated solution; with little 'belly load' to play with, the load controller is left with only being able to re-arrange seating to correct balance now that they are aware of the out of trim condition. That would mean (with high probability) two things then happened.

(1) positive action to change from seat row trim to that of cabin area,
(2) positive action to overwrite the actual cabin area distribution with a new (now required) seating distribution to result in a safe balance condition.

Now - if that is really what happened and those direct and positive actions, taken in and with full knowledge, to correct an unsafe condition were then failed to be immediately communicated to those who would be responsible to execute the physical change in seating to insure the change was effected on the aircraft and not just simply realised within the computer system and on the final load sheet document, well .... honestly, it's not simple, its incredibly negligent.

[DaveReidUK]

As it happens, 4 seats on the Wizz A320 (11A/F, 15A/F) don't exist on their A321ceo (see previous LOPA post). They may or may not have been allocated.