After due consideration by the Captain Autolands done in VMC without bothering to advise ATC are most certainly not stupid.:D

You do keep your finger close to or indeed on the A/P disconnect button don't you!!

A rejected landing due to bounce or float is a go around maneuver which includes retraction of Flaps to 15 at a speed not less than Vref flaps 30.

A rejected stop....also called rejected landing, when ALL (not only MaIN) gears are on the ground, a touch and go maneuver should Be used unless thrust was deployed,in which case,well hope for the best.

Nope, wrong.

If you bounce and decide to go around its just that, a "go around"

If you have landed already and before reverse decide to reject the landing then it's called a "Rejected Landing"

Flaps/Trim are not touched. They don't need to be for goodness sake mate, the whole idea is to get the Aircraft airborne again ASAP.

Save the Flaps/Trim changes for planned trained briefed Touch and Go manoeuvres.

I suggest you look up your FCTM. It's all in there bud :ok:

In fact here it is for your reading pleasure:--

REJECTED LANDING

A rejected landing is defined as a go-around manoeuvre initiated after touchdown of the main landing gear.

Once the decision is made to reject the landing, commit to the go-around manoeuvre and do not be tempted to retard the thrust levers in a late decision to execute a landing.

Apply TOGA thrust. Ground spoilers will auto-retract and autobrake will disarm as a consequence. A CONFIG warning will be generated when the aircraft is still on the runway, with thrust applied and the flaps at FULL. Disregard this warning. If the AP was engaged, it will disconnect. If on the ground, continue de-rotation. Rotate only when the PM has confirmed the thrust is set and the speed is above VAPP . When clear of the ground, with a positive ROC, select Flaps 3 if approach was made with Flaps FULL. The landing gear should be retracted when a positive ROC has been established with no risk of further touchdown. Thereafter, proceed as for a standard go-around.

If reverse thrust has been applied, a full stop landing is mandatory.

NP330, you can do autoland in any weather (limitations observed of course), but you cannot do a LVP in any weather. LVP autoland is autoland with very low minimas. Practise autoland require high minimas, if you have not all installations in place (no redundancy on airport installations, sensitive zone not protected). You cannot just come and say "let's make an autoland" just because the visibility is 2000m. Autolands are reasonable if LVP in force (<550m visibility) and in good weather (say a few 1000m). The "marginal" weather case is exactly the weather you should NEVER make an autoland!!! SIA's Op philiosophy goes exactly the other way around.

Yes i do keep my fingers where they should be.thanks.:hmm:
Advising ATC for one would allow them to know your intensions so they can advise traffic to hold at the cat2 hold line rather than cat 1.
For two,performing an auto land implies than you will Be fully configured hence slower much earlier therefore letting ATC know is good practice.

Technicallywise,increasing the possibility of my aircraft veering off the side at low level including in the flare in a nose up trim,because some aircraft is at the cat 1 position is a risk that i find highly unnecessary.
In that case a GA should be performed,INCLUDING retraction of FLAPS.

Advising ATC for

It does? News to me. We fly the autoland profile exactly as any normal ILS approach, there is no difference in regards to configuration. Might be different on the big Boeing, but i somehow doubt it.

Well I can tell you that quite a lot of highly respected safe International Carriers in addition to SQ don't "require" their crew to advise ATC when Autolands in NON LVP conditions are done. They only advise that particular care be taken as protected areas are not protected and that be ready to take over manually.

Thousands of safe Autolands are conducted each year all over the Globe in such conditions. :ok:

Ok DAni,
My Sops stated that full config 3nm before glide in auto land.
During normal ops fully config by 1000ft.

To the other,
Flaps on a 737 to be retracted!!! Minimum speed Vref,also for a rejected landing.
Thrust,pitch maintain,vref then flaps 15,pitch increase.
Please post your info here if otherwise.
Thanks

NP330, so your airline does also not state a minimum for autoland practise approach? Then I shall count you also as a member of the warm weather ligue ;)

Just looked at a 737 FCTM and here it is:--

Go-Around after Touchdown

If a go-around is initiated before touchdown and touchdown occurs, continue with normal go-around procedures. The F/D go-around mode will continue to provide go-around guidance commands throughout the maneuver.

If a go-around is initiated after touchdown but before thrust reverser selection, auto speedbrakes retract and autobrakes disarm as thrust levers are advanced. The F/D go-around mode will not be available until go-around is selected after becoming airborne.

Once reverse thrust is initiated following touchdown, a full stop landing must be made. If an engine stays in reverse, safe flight is not possible.

There is a difference, this 737 FCTM doesn't get specific but I know the 777 FCTM my company has does get specific the same as the Airbus FCTM.

I'll keep that in mind next time I'm landing in -20c in ANC or YYZ or ORD :ok:

Perhaps the industry should reflect on why some operators believe that it is necessary to fly ‘practice’ autolands. With modern, high reliability systems it should not be necessary to prove technical integrity.
Normal procedures are best learnt in simulation, normal (without failure) operations are routine which should only require minimal refreshing. The critical procedures involving reconfiguration, go around, or lack of visual cues, all involve assessment and decision making (and the use of normal procedures); these must be taught and practiced in simulation.

Any automatic landing must be aware of the need for protection. In real conditions large safety margins are applied with Cat 2/3/ holding zones (LVP), approach and takeoff spacing, dual transmitters, etc. However, in clear conditions, any effect of potential interference has to be anticipated and mitigated by the crew. Some aircraft / situations are easier to manage than others. Thus the risk in managing these vs the risk of a manual landing after a long flight or similar judgment has to be considered.
When there are incidents / accidents it probably indicates a mismatch in the risks. A key safety issue is to identify where the risks may have changed or which were mismanaged. Whatever is concluded, knowledge of the subject and the risks is always required, and this is a function of education – operator, airport, individual.

True the 737 fctm isnt specific,a bounced landing recovery should be dealt using a standard go around procedure and a rejected landing in the same way.
Boeing is not as precise and informative as airbus in many areas...including perf issues but you may understand that applying ga thrust while maintaining pitch until VREF is to cover the chance of losing control if the engine fails in the process.
Obviously boeing believes that full thrust and vref,retraction of flaps is ok,the only perf issue would be the landing gear up after rate of climb,in case you do a secondary touch(pilot handling more than perf issue i think).
Airbus or your company fctm may think otherwise which is ok.
My european licence issuer advised airlines via an AC to retract flaps once vref is obtained during a rejected landing.No mention when clear of ground.
There is material about low energy go around which covers this case.

Ah, now i see where your comment came from. Very non-boeing procedure, but if it is your SOP you have to follow that of course.

Anyway, we do not know any "practice Autoland" limitations or rules. An autoland can be done anytime conditions are within the autoland capability of the plane, if the minima are above CAT I and better both pilots may do an autoland at their discretion but care must be taken to take over manually very fast where LVPs are not in force, with weather below CAT I only the captain may do a landing. We used to do that in MUC quite often, if advised early enough ATC was usually able to protect the CAT II/II safety zones, if not they told us so. In that case rapid localizer deviations were normal and it was usually better to do a manual landing.

Since simulator training focuses on real CAT IIIb conditions in which only the captain may do an autoland the only chance at practicing an autoland for any FO in my outfit is during CAT I or better conditions during normal line operations. It is however extremely rare that one chooses to do so, but it is perfectly possible and even encouraged by our flight OPs department to do so from time to time.

This is the key statement.

The actual case showed that it might be very difficult to react on such deviations in "poor" weather (slightly above Cat I). They didn't have enough visual cues to determine that they where off the center line until it was too late. Otherwise they surely would have disconnected.

A minimum weather for practise approach is a no brainer in my opinion. If you don't agree, walk along the skidmarks in MUC...

Dani, your concluding assumptions may be hard to substantiate (#294).
A PF in Cat 3A should have sufficient cues to assess aircraft position, flight path, and deviation from the norm – it’s the point have having visual requirements. Thus it is difficult to understand how Cat 1 or better results insufficient cues.
However, it is possible that the cues available were either not used, or not understood. Alternatively if the situation was known then the action was delayed or incorrect; the latter could be affected by alertness etc.
Thus, a decision to use autoland after a long flight, or when landing in lower than normal visibility, relies on good alertness. Perhaps as much as for manual flight?
IMHO, this incident has contributing factors originating much earlier;- in management, SOP advice on vis limits, not requiring declaration of A/L intent, and crew, an awareness of preceding traffic and the hazard/risks involved.

It might be counterintuitive to you that practise autoland above Cat I WX requires quicker reactions than Cat III autoland, but this is exactly the case, and the actual case shows us the reason why: In practise landings you have no protected zones, thus the deviation is much faster and heavier.

Of course you also need cues in a Cat III approach and must be able to disconnect or abandon the approach, but there are many more redundand systems in place to avoid such deviations. I agre that a deviation like in MUC might be possible (if for example someone violates the sensitive zone), but in real Cat III WX you can't do anything about it, you are more or less doomed or can hope you get away with it. Chances are relatively low though that something like this happens. TWR should get a warning if someone violates the sensitive zone.

So, to reiterate, yes, I strongly suggest that everyone should have relatively good WX - several 1000m visibility for a practise autoland. Autoland to relief a tired pilot is surely not intended by the inventor.

The german report specifically refers to an avroliner that took off just before the 777 and was a few hundred feet over the localiser antenna around the time all this happened. Obviously nobody yet knows for sure what happened. But this incident has raised some interesting points.

Safetypee,
I mostly agree, technically, but out in the real world I like to know exactly how the aircraft actually performs and it's not quite like the digital simulation you get indoors.

It's good to get a 'feel' for how the aeroplane flies the last 100' so you can get a heads-up when it's having a bad day and not be totally surprised when you have to take over.

I've done 'practice' i.e. no LVPs autolands when approaching directly into-sun late in the day on wet runways - I think the A/P has a better chance of a getting it right than I do after a 12hr flight.

Also, CATII, III training is but a small part of the syllabus and there is only a limited amount of time to practice normal operation. We get one landing a year in the sim. Not really enough to be familiar...

Absolutely right Dani!!!!!!! I think some of these muppets posting here have no concept of how and why we pros use AUTOLAND:=:=:=

But it is! For a proper LVP Cat II/III autoland to be carried out there are three requirements:

1. The runway in use must be equipped Cat II/III and LVP in force.

2. The crew must both be current Cat II/III

3. The aircraft must be cleared and current Cat II/III, a satisfactory autoland having been carried out within a previous subscribed period, possibly 28 days, operators will vary, (also, if an aircraft has been 'snagged' and lost it's Cat II/III capability, when rectified, a practice autoland must be executed before Cat II/III capability is restored).

Bit strong White Knight! Having a bad day/beer?:)

Morons.

I've been a "pro" for nearly 30 years flying in the real world.

I've never done an Autoland for "practice" that's what the Sim is for. Any I've done are for a good reason, weather related or not for weather.

Boeing and Airbus spent zillions perfecting their fantastic reliable Autoland systems and I for one intend to make good use of it whenever I feel safety is improved for whatever reason.

My company agrees and has done it this way since Autoland was first brought in over 30 years ago AND HAS NEVER HAD AN INCIDENT DOING ONE. All runways we Autoland on must be approved for it, some like WMKK are not.

Like the poster above said, the Captain must always be in a position to assume control if the autopilot runs away!!

- a bit of a disjoint in your thinking there? Since all the 'excitement' of which you speak will be in the last 100 feet or so, why the need for '1000s' of mtrs vis? Surely we are all 'prepared' for that unexpected A/P quirk during a real Cat III, so what is the difference?

I am also surprised at the lack of autolands carried out by some operators here in Cat I or better. I must have done 10 or 15 in my career to clear Cat III tech log restrictions. How do the airlines involved do this?

The Muppets were the SQ ones who weren't on the ball!

I Was There.........
 

I have come to this topic late, and I haven't waded through all the posts above, but I thought I'd put some of the latter comments into perspective.

I actually landed on the parallel runway minutes after the incident. Trust me, the weather was nowhere near as good as the weather mentioned above. We were flying an Airbus A320 and briefed to carry out an autoland due to the weather. On the radio, we heard the tower hassling aircraft to move up to the CAT I holding point. On very short finals the aircraft began to veer off the ILS centreline - ILS scales flashing, etc; we were - just - visual and the autopilot was disengaged and the aircraft landed manually. Sure enough, an aircraft was seen to be taxying on to the runway as we passed the threshold.

When on the ground I heard about what had happened to the SIA 777, I had my suspicions immediately that it was an autoland gone wrong due to lack of ILS protection.

FWIW!

Surely a SIA manoeuvre wouldn't be 'expected' during a real CAT 3? I hope you are very 'prepared'...

I think Dani wanted good vis so he could see if any aeroplane was in the protected area...like 146s.

Amazing how ppruners can work themselves up into a frenzy about very little (muppets/morons etc.). Nobody is arguing that one size fits all. For instance if you only do a couple of landings a month and in places like Australia then mandating everyone who is auto land qualified must do practice auto lands in the aircraft is hard to justify.

However if you operate short haul in northern europe it is no big deal. If your OM requires it then you do it. That the sim is the only place to practice might also be used as an argument against raw data flying. Most of us would disagree. I think it is self evident that if you only do auto lands in the sim (in the worst case three approaches twice a year) to be legal, then you are going to be much less familiar and when things go wrong as in this case arguably less able to cope.

Lets share information that makes us all better pilots and try and cut down on the judgmental point scoring stuff (says he being judgmental but hopefully polite).

Yes, I would be, but never experienced it apart from in the sim. There is no absolute guarantee (unless YOU have one) that the system will perform flawlessly.

a) I would 'assume' potential problems with no LVP protection
b) I would probably have picked up the take-off clearance of the preceding and quite probably have seen it during its departure
c) I would have 'noted' the presence of a/c at the Cat I hold.

Is this not just standard airmanship?

Practice autolands
 

In my current company, very few autolands per pilot are performed as the company policy is to perform autoland only if the airport has applied LVPs. Some guys even don't want to perform autoland if the vis is still aceptable for CAT I but LVPs are on ??? Example in CDG you are very likely to see LVPs on with an RVR of 2000m.
This has a consequence: very few pilots go over 10 autolands per year, which to my taste is not enough. I feel it in the sim and in the real world too, when I have to perform CATIII.
In my previous companies, autoland was perfomed for practice about 1/2 times a week, at pilot discretion, unless NOTAM prohibited those. I felt then a lot sharper in detecting anomalies and deviations of the system. In other words, ready to take over.

Question: I could never find a JAR document legislating on the matter. Why are some companies encouraging practice CATIII and others telling you it is prohibited (which I doubt very much) ? Is it a company responsability or a CAA decision country by country ?

Too big a question, Waspy! I could never find any regs that required an Autoland in Cat II either but several companies have that restriction, even when it is cloudbase related with good vis, which often prevented an approach in a strong crosswind and required a diversion. I can fully understand the logic when there is fog around, of course, but then unless you fly at '40kts of fog' RAF Brawdy.......................

SQ have a requirement of 8 auto-lands within 12 months and the aircraft requires one every 28 days. Each Sim (every six months) gives out 4 auto-lands and any real Cat111 is a bonus. I averaged one real one a year, although when the fleet had ZRH it wasn't hard to pick up a couple extra there in winter.
The real lesson from this incident, is to tell or ask ATC if a coupled approach is required when LVP is not in force and get protected.
I find it distasteful to call the crew names. They got caught out very late in the flare. Should have disconnected etc but we were not there. There is a culture to lean on the automatics in some outfits, maybe this will change things.

I can tell you now, in Cat I or better conditions, we are not going to do anything special to protect you. You know the limitations and the mitigating SOPs, so you need to deal with it.

Providing 6 miles spacing between arrivals and holding departures back at the Cat 2/3 holding point is just not feasible unless you want to explain to the company and airport operator why the runway movement rate dropped by 30%.

I doubt it. t doesn't matter what incidents or accidents occur, automation dependency is here to stay.

Most companies boast in their OMA the aspiration of best use of equipment.

Some go as far is prohibit NPA and are extremely repressive concerning visuals, as long as an ILS is available.

Funnily enough it's often the same companies who do and actively demand these practice autolands.

An autoland on a non-LVO enabled runway/ILS is not best use of equipment.

To practice autoland, or testcycle on board equipment in conjunction with ground equipment that is not certified or suited to do so, is oblivious.

Thus practice auto land for CatIII on CatI ILS is futile because it does not prove zilch.

It simply should not be done because it's useless and not very intelligent, quod erat demonstrandum.

Had some extra time on a T-7 sim after a fairly straight forward LOFT session with 2 guys who did really well ( eventhough it's still January they might already had tips about the LOFT scenarios! ). Decided to use the extra time for " SPOT " exercise. Gave them a simple Cat 3 B approach ( with Cat 2 visibility minimums programmed ) with a loss of ND on PM ( first officer's )side; the crew was sharp, continued approach with ND selected on lower MFD.

As our sim had no facility to simulate a scalloping localizer, I programmed in a low level gust of crosswind at 50' AGL. After flare engaged, the aircraft ( sim ) drifted off the centerline, the two very sharp guys were suddenly caught out. I guessed there must have been a lag in the crosswind gust kicking in as the touchdown occurred outside the runway! The PM attempted a balked landing but like the SQ crew, he commanded the normal go around procedures, i.e TOGA, Flaps 20, etc. He did push the thrust levers up, but a quick disconnect of the autopilot and a hasty pitch up caused a tail strike as he flew away.

They came back for an uneventful Cat 1 landing after all the checklist for the tailstrike was accomplished ( they did this admirably ). At the debrief I showed them that during the Cat 3 balked landing, they had taken out some HIRLs and the transmissometers.

They looked utterly chastitized; I then explained to them that I was simulating the SQ B777-300ER incident at EDDM. They then understood how that in a short few seconds things could really go pear shape despite how good one thought one could be. The captain confessed to being quite judgemental about the SQ crew's performance until this exercise!

In my years of conducting sim training and checks, I have seen many many amazing incidences where crews got caught out and I have also had an amazing number of times where crews managed to pull off admirable performances. Many a times, poor management of the automatics led to things going awry, especially when one was too fixated on a task at hand.

Dead right. The other day I lined the 737 crew up (simulator) and cleared them for a visual take off, circuit and landing without the FD or autothrottle. To my astonishment the captain said he had never done circuits and landings in the simulator.. It had always been radar vectors for an eventual ILS. He had some real problems with manual flying without the goodies to help him. Same with the F/O....

I suppose you could call a circuit and landing "unrealistic" in today's world of automation? :sad:

studi - what procedure does your airline use to clear a/c back to CatIII status after downgrading? It would appear you have another way than flying an autoland in CatI or above - this would, I'm sure, be of interest to many operators.

I'm surprised you don't know. If the a/c has a tech log entry 'Cat I only' does your airline just carry on with that or do they carry out a 'practice autoland' to clear the entry? As I said, in my time I must have done 14-15 of these to clear the entries but you have never heard of it? Can you ask someone who does know?

I'm not convinced you understand the procedure in question? This has nothing to do with crew 'practice' or 'number of a/lands in 6 months' etc.

Thanks, Marktabs for your insights, which are completly congruent with my expectations and experiences. MUC is very famous for its "jumping" localisers as well as its inhomogenious fog conditions. Locals or people flying regularly to MUC can completly understand that your localizer will do strange things because seperation provided by local ATC is not such that you are protected - unless in real LVO.

So as a far flown captain from the "warm weather league" you might want to build in some extra margins if you are not completly familiar with the local conditions.

Marktabs, just also for info for you: You are not legally protected if you brief a LVP autoland on a non LVO-in-force airport. Remember that you have to stay on Cat I minima, if you cannot make it, you have to go around. Soon after that, ATC will send some fliers in the holding and issue LVO in force. After that, everyone can do Cat III autoland to their likings. But NOT BEFORE! You might also try your luck by negotiating with ATC before your go around but be reminded that German ATC does have its own way (like they do it in UK, France, Switzerland, Australia...).

My previous company asked crews coming early september to practice cat3 approaches.
We had a form to fill in including weight,speed,wind condition,airport,runway,and touch down location and to write whether the autoland was done successfully or not.
When the mid september fog started to roll in,our aircraft did the required amount of autolands and could be used when LVP were in force.