Airbus ILS course

[michelda]

Hi guys,

at 350ft, during cat 2/3 approach, pilot flying has to check ils course.
I didn't find the tolerance. Any idea?

Michelda

[9.G]

if my memory serves me right the figure is 5 degrees. The reason being the alignment during de-crab maneuver. :ok:

[PantLoad]

9.G is correct..... 5 degrees or less. Will have to look for the reference....can't
find it at the moment.


Fly safe,


PantLoad

[nitpicker330]

Huh?
I haven't heard of this requirement before?

[michelda]

thanks guys,

I remember 3 degrees, but I'm not able to find any reference

Michelda

[rudderrudderrat]

Hi michelda,

I think the tolerance check is the last opportunity to confirm the correct ILS QDM is in the box (it's too late to do anything about it by this stage except GA - but I think the "logic" *1 is:

FCOM 1.22.30 P50:
ILS APP Mode Arming Conditions ... "The aircraft is above 400 feet RA," etc.

FCOM 1.22.30 P 53.
"Align sub-mode.
Align is a sub-mode of LAND also known as "decrab" that lines up the aircraft's axis with the ILS course at approximately 30 feet. It is not displayed to the crew."

On TriStars - it was quoted as being acceptable if ILS QDM was within 6 degs of Runway QDM - but I cannot find any Airbus equivalent limit published.

*1 = oxymoron.

[aterpster]

rudderrudderrat:

On TriStars - it was quoted as being acceptable if ILS QDM was within 6 degs of Runway QDM - but I cannot find any Airbus equivalent limit published.

Where was this quoted? It certainly wasn't a procedure at my airline. Rather, an autoland below specified visibility minimums was limited to a crosswind component of 10 knots. That took care of things nicely without having to observe a compass heading that might not be all that accurate.

I did quite a few 1011 autolands in clear weather with a 25 knot crosswind. It did just fine. But, the margins in that strong of a wind were outside the statistical envelope for a low-visibility autoland, one where the runway might not be seen until after the main landing gear had touched down.

[DutchOne]

in LAND mode LOC and GS are locked. Anything more than 1/4 dot (LOC) or 1 dot (GS) will give you trouble when passing 200', so you check that it is locked within these margines.

1 dot LOC = 0.8 degrees
1 dot GS = 0.4 degrees

[Coldbear]

Hi Michelda,

I found this in my Instructor support manual:

"Incorrect ILS CRS (Δ 5° with runway CRS) leads to incorrect autopilot DECRAB.
Check ILS CRS at 350 ft. If incorrect, continue approach down to CAT II minima (AP must be set OFF at 80 ft latest)."

Kind Regards,

Martin Dahlerup

[michelda]

Hi Martin,

thanks for your replay

Michelda

[salamanderpress]

I made some calculations to relate drift angle to groundspeed.

5 deg drift angle corresponds to crosswind well below the 20 k limit for autoland. At a high ground speed of 160 k, 5 deg would correspond to just 13.3 k CW. To reach 20 knots crosswind and 5 deg drift, the GS would have to be 240 knots.! At a "normal" groundspeed of say 145 knots, 5 deg would correspond to 12 knots.

The 5 deg limit seems to be advisory in nature? the extract from Instructor Manual is

"Some requirements are specific to CATIII with NO DH, because an Autoland is compulsory. These are not
necessarily monitored and thus not indicated.
• Incorrect ILS CRS (Δ 5° with runway CRS) leads to incorrect autopilot DECRAB."

The FCOM extract (PRO-NOR-SRP-01-70 P 14/32) merely calls for the PF to "check ILS Course", and does not actually ask you to DO anything about it. Nor does it refer to a limit.

I found a limit in the AFM saying "rudder travel limiter required to be serviceable for autoland with crosswind over 12 knots".

If everyone who flew a Cat III approach and saw a drift angle over 5 deg (how many of us have looked?) went around, there would be a lot fewer landings. I guess that the Instructor Manual is advisory in nature. The decrab not being completed would mean that the aircraft would touchdown with some drift still on. This would mean a landing gear structural limit coming into question. Does anyone have an idea what that is? In terms of drift angle or crosswind limit for landing without decrab, possibly?

[clonecity1]

On a CAT 2 approach, incorrect inbound course at Land Green.
FCOM note stipulates that AP should be disconnected by 50ft(ie before Flare where the aircraft would be aligned to the incorrect runway QDM).
Question I have is that if it's going to be a premeditated manual landing from 350ft why isn't the normal 80ft Auto Flight limit applied?

[PENKO]

Salamander, I am under the impression that you think we have to check the drift at 350 feet. Am I correct?

Anyway, I always wondered why we have to check the course at 350 feet (busy time!). Why not before the approach? It is not going to change unless you change runways. Or am I missing something?

[salamanderpress]

About AP disconnection, ie aborting the autolanding. This is required only if the red AUTOLAND light comes on. This does not happen regardless of the strength of the crosswind. Drift angle is not ever monitored, only beam drift or loss of signal. And below 100 ft, even a single autopilot failure will not change the landing status and an autolanding would be executed without any capability downgrade being indicated.

At 350 ft we ARE required to check the drift angle (refer Task Sharing, "check ILS Course"). This is not the ILS Final Approach Track which is fixed, but the actual aircraft course to keep you tracking the ILS beam. This may not correspond EXACTLY to drift angle, because the aircraft might possibly be correcting for a (small) beam deviation caused by changing winds. Also, this drift is only indicative of the crosswind at landing, because the winds would change as we go down.

Thanks a lot, Clonecity and PENKO.

[PENKO]

Sorry salamander, but check ILS-course means what it says: check the course of the ILS in the FMGC. How do you interpret that as check drift angle? The only action the FCOM describes at 350 feet is: 'incorrect selected course at 350ft revert to CATII minima-disconnect ap at 50FT at the latest' or 'check ILS course on PFD'.

Course, not drift. This has, as far as my knowledge goes, nothing to do with the 20kt crosswind limit!

[9.G]

one needs to check the drift angle between ILS FC and aircraft's kinetic vector e.g track as it's simple logic. If you have the correct DA then de-crab isn't a issue in case the aircraft believes to have it's vector aligned with CL but's not you end up like SQ in MUC. Aircraft will try to to align it's ground track with the CL and guess where you'll end up with the ground track diverging more than 5 degrees.:ok:

[Madp1lot]

have you ever thought that it simply means:

- ILS course? A/C course? are they the same? If so... then the A/C decrab algorithm will work... If not? well... it will not decrab normally, then maybe you should go with Max CAT II manual landing... or GO Around...

The FMGS is supposed to update your BIAS as soon as a LOC signal is caught, but even so, if you're on the LOC beam, with the correct ILS course set, but your real course is +7º, then I think your IRS just got busted in att mode.

Will you really continue APP for landing, flare and roll-out when something's not smelling right (lets say... with 150m visibility?)

[PENKO]

Ok, do you have a reference for this? What does your company SOP say?

[Weapons_Hot]

The correct Q code for runway heading is QFU. Period! (ICAO Doc. 8400 or, ACP 131E)

ILS QDM? How about LOC's QDM, to be correct.

So, it should read: a check of the LOC's QDM against the (runway) QFU.:hmm:

[9.G]

PENKO, does you company SOP forbids you to auto land on RWY with LOC offset? I'd be extremely surprised if it did. Neither there's such limitation in the Airbus AFM. No one in the right state of mind would ever do it. It's self explanatory mate. Aircraft's ground track and LOC CL must be coinciding otherwise you go mow the lawn. Whether it's aircraft's internally produced failure or LOC offset on the ground makes no difference the result is the same you become a gardener. :ok:

[737Jock]

He is asking for a reference in the books.

So are you saying that the aircraft doesn't line-up with the localizer signal during decrab, but with the IRS track?

And that the aircraft track should therefore be within 5 degrees of the localizer course?

This sounds all highly unlikely, the IRS track is not the same as the aircrafts ground track it's only the calculated ground track. When locked on to the localizer you know you are following it's ground course, no matter what the IRS thinks.

So I guess it must be related to the IRS heading, not track. (the logic might be that if the track has drifted the heading might have drifted as well, and track difference is easier to spot As a symptom while on the LOC) Which could cause the aircraft not knowing where to put the nose during the line-up with the localizer.

But I can't find any reference for this.

Or is it simply that the LOC course has to match the rwy QFU? Which would then mean that the aircraft can perform an Autoland on a 5degree offset LOC (max Δ 5° with rwy QFU)? Which is unlikely and not something I would (like to) perform.

[PENKO]

This must be one of the weirdest threads I have ever encountered on PPRuNe!
First of all, my company only allows us to auto land on pre-approved runways. Offset courses are not mentioned as far as I know, but it is irrelevant as it would not be pre-approved!

Now back to the original confusion. In my company, one of the largest Airbus operators in the world, SOP is to check the ILS course on the PFD. This is in full accordance with the Airbus FCOM. You check this data to ensure that the aircraft has the correct reference when it comes to aligning the aircraft with the runway during the flare.

Now please tell me if you do things differently in your company for I have never heared of any drift limitations. Why are we even talking about drift? The only crosswind limitation in the FCOM is 20 knots. Again, I might be missing something here so I am more than happy to be corrected, but please give me a reference to this drift limitation as I have never heared of it and I doubt any Airbus 320 rated pilot has. (mind you, This is a simple pilot's SOP question, no software designer's theories please :ok:)


In answer to 737jock, the aircraft indeed aligns the IRS/FMGS computed track with the course you check at 350. Which other track can it use?Align sub-mode

Align is a sub-mode of LAND that lines up the aircraft's axis with the ILS course. It is not displayed to the crew.

[737Jock]

I don't get it, Penko. From what I see people seem to think that the aircraft can fly a ground track that is different from the localizer course while the aircraft is locked on to the Localizer and that this is evidenced by the track not matching the Localizer course. Which would then lead to the aircraft mowing the grass.

While in fact it is just evidence that the FMGS is not calculating the ground track accurately, thus that the GPIRS has drifted and is not giving accurate information.
In answer to 737jock, the aircraft indeed aligns the IRS/FMGS computed track with the course you check at 350. Which other track can it use?

Penko: The drift is the difference between the ground track and the heading/axis (nose), drift is what is removed during decrab maneuvers through the ALIGN SUB-MODE (as you said DSC 22-30-80-30 Precision Approach Modes)
Thus the aircraft aligns the Nose of the aircraft/axis (heading) with the course of the localizer during decrab!

OR does it align the nose with the FMGS computed track which is assumed to be equal to the LOC course? In thiscase it would obviously be very important that the FMGS track matches the LOC course, as the aircraft would not use the LOC signal during decrab.

However if the aircraft would align the FMGS track with the course of the localizer it would not be doing anything (certainly not decrabbing), as on serviceable aircraft the FMGS computed track is equal to the LOC course. .

Once the aircraft has captured the Localizer it adjusts the heading to maintain the localizer course/localizer ground track.

The FMGS updates it's lateral position when using the LOC beam during an ILS approach and for quick update in GPS/IRS mode(DSC 22-20-20-10 Radio Position and DSC 22-20-20-10 Navigation Modes)

So my concern if any would be that if the FMGS track does not match the LOC course at 350ft is my displayed heading actually correct? Does the aircraft know the direction the axis is facing? Or is it off by X degrees as well?

People are saying that if at 350RA the FMGS track deviates more than 5 degrees from the RWY QFU/LOC course (we are not discussing offset approaches here) than we should go-around or fly to CAT2 minima for a manual landing. I don't get it:

PRO-NOR-SRP-01-70:

AUTOLAND
CHECK that the FMA displays the aircraft capability (CAT 2 or CAT 3) for the intended ILS approach. MONITOR the radio automatic callout.

At 350 ft RA: CHECK that “LAND” is displayed on the FMA. CHECK ILS course. If LAND is not displayed or if the ILS course is not correct, do not perform an autoland. The flight crew should perform a go-around, if visual references are not sufficient.
Between 50 and 40 ft RA: CHECK that “FLARE” is displayed on the FMA.
At approximately 30 ft RA: CHECK that “IDLE” is displayed on the FMA, and that autothrust starts to reduce thrust toward IDLE.
At 10 ft, “RETARD” callout comes up: MOVE the thrust levers to IDLE. Autothrust disconnects.

So where does the FCOM refer to checking the Track vs the ILS course?

[PENKO]

Apologies jock, I should have said that the aircraft heading (axis!) is aligned with the ILS course, which results in a nicely decrabbed landing. This heading info must come from the IRS. I think we are getting hopelessly confused over something that is very simple.

[737Jock]

9.G definitely needs to come with some references, cause his gardening references simply don't make any sense.

My guess is that the aircraft simply aligns the IRS heading with the localizer course no matter what the calculated FMGS track is. And that we only need to check if the ILS course is displayed correctly.

The FCOM does not provide any more information than this.

The notion that the aircraft is not following the correct localizer (ground) course because the calculated track is off is simply ridiculous. The aircraft adjusts the heading to maintain the localizer course and is thus on the correct ground track.

[9.G]

I wish it was as simple as you try to picture that Penko, well it's not. To start with IRS deliver a true heading using gravitation equation. FMGC applies then magnetic variation dependable on current a/c position. It's not unusual for those magnetic variations to be outdated thus resulting in incorrect magnetic heading without the system detecting it. Airbus is supposed to publish the list of airdromes with outdated variations. On top of that WGS 84 non compliant countries like China only add fuel to fire as your accuracy gets screwed. It's not unusual to see being on LOC and the diamond triangle track being coupla degrees off. Let's say ILS 07 with QFU 070 with left cross heading will be in the range of 060-065 ILS 070 and track 078. Obviously a/c will align it's nose along the LOC using stored magnetic course and actual LOC signal. Ideally all 3 indications must be perfectly aligned after de crab a/c MH,MC and ILS. It's not gonna happen if the a/c's magnetic variation is wrong. One of indication for that is when the track is offset with the ILS LOC on final. Just my 2 cents. Can't find the reference though. :ok:

P.S. And that we only need to check if the ILS course is displayed correctly. That check is done before TOD during briefing, at least in my airline.

[flash2002]

9.G what does the track diamond have to do with flying a localizer? The aircraft will adjust it's heading to maintain the localizer.

It doesn't really matter if that is a true heading or magnetic heading. In fact you could fly limited panel ILS's in a simple SEP without heading information at all. LOC deviation to the left, fly left, to the right, fly right. In the middle keep the wings level etc.

During decrab the aircraft only needs to lineup its axis with the localizer, and the drift angle is solely dependent on wind.

Now if there is 0 wind and the heading is 5 degrees off the localizer course, than that would lead to a problem as it indicates a non-existing drift. But that is heading, not track diamond!

There could be various other problems with the track diamond, as it is a calculation based on more variables.

I guess that MH problems could happen with an out-of-date magnetic variation database, but as you said yourself regulated by airbus and also by the airline through pre-approval of airports at which an autoland can be made. It does not require a check at 350ft in my view. And again the limits you mention are nowhere to be found.

That check is done before TOD during briefing, at least in my airline.

Here too, but that is what the FCOM says. The other limitations you mention are nowhere to be found.

[9.G]

I'll throw this in. I just can't imagine an ILS contained in an NDB being accidentally coded with a front course more than 5 degrees in error, so It's not the ILS which is wrongfully coded but the magnetic variation as it's is part of the data base. Logically if the variation is wrong everything else is wrong. on the 330 you can simply check it by switching to TRUE HEADING not on 320 as far as I remember. I clearly remember to have read the explanation in the old FCOMs but it was 20 years ago. However there's a FCB 15 on 320 saying AUTOMATIC LANDING WITH OUT-OF-DATE ADIRS MAG VAR TABLE Some Airbus SA and LR aircraft continue to fly a number of years with ADIRS part numbers fitted with out-of-date magnetic variation tables. If the ADIRS magnetic variation differs by more than 2 ° or 3 ° (depending of aircraft type) compared to the airport current magnetic variation, the lateral performance of the automatic landing and rollout is significantly affected, which prevents from using the automatic landing system on these airports. Airbus publishes in the FCOM for each year, the list of airports where automatic landing is no more authorized with these ADIRS part numbers. Last but not least what can you check at 350 ft you haven't done or seen before? ILS course check is part of approach briefing and done before TOD. :ok:

[flash2002]

Last but not least what can you check at 350 ft you haven't done or seen before? ILS course check is part of approach briefing and done before TOD.

I don't know what it's like at your airline, but at ours we check items multiple times during a single flight. So it doesn't have to be new, it can be a last minute safety catch in case it was forgotten during the approach briefing.

[PENKO]

9G, I see what you are getting at, and I know the reference you provide. I would have never connected magnetic variation errors with the SOP to check the ILS course on the PFD. Are you certain that this is what Airbus means with check ILS course? Airbus certainly does not mention it in their SOP and as you rightly quote Airbus, autoland would be prohibited anyway on the affected fields!

Are you sure that this is not a case of flying a 'bus like a Boeing? :ok: (which happens a lot!)

[9.G]

flash 2002, ILS course is derived outta DB it's the same from the PB till TD and it won't change a wee bit but of course you can check and double check it endless times if you wish. However it's stored value in the computer and doesn't change. I however admire your diligence.:ok:
Penko, isn't ILS course equal with MH in wind calm? There's your answer. The ILS course check must have some technical relevance that's what we all agree upon. ILS is stored in the DB and doesn't change in any phases of the flight thus the check must have some other background. Perhaps AI has changed all that over time with modifications however 5 degrees value is still there. Whether it's between MH ILS or GT or all of them I stand to be corrected. :ok:

[flash2002]

I'm not sure if it has technical relevance, I still am at a loss why we check the position of the ignition during every approach checklist iaw with airbus sop's.

I thank you for admiring my diligence but I'm just referring to airbus SOP's from the FCOM. I haven't seen you quote from any reference material yet?

Where do those 5 degrees come from?

I guess airbus could have designed something that eliminates this problem. After all in no-wind conditions the fmgs track and MH should be the same. The difference between MH and Loc can then be deducted from the correction that is applied during decrab. But again this would assume that the track calculation is correct and does not drift.

I think this FCTM quote reveals that the track line is subject to extra unreliablility's beyond magnetic variation. But maybe I'm reading into it.
SI-020 FPV
The FPV appears on the PFD as a symbol, known as "the bird". The bird indicates the track and flight path angle in relation to the ground. The track is indicated on the PFD by a green diamond on the compass, in addition to the lateral movement of the bird in relation to the fixed aircraft symbol. On the ND, the track is indicated by a green diamond on the compass scale. The difference in angle between track and heading indicates the drift. The flight path angle is indicated on the PFD by the vertical movement of the bird in relation to the pitch scale.

RELIABILITY The FPV is computed from IRS data, therefore, it is affected by ADIRS errors. An error may be indicated by a small track error, usually of up to ± 2 °. This can be easily determined during the approach.

[9.G]

I still am at a loss why we check the position of the ignition during every approach checklist iaw with airbus sop's. This is off the topic but if it helps, airbus aircrafts are fitted with different types of engines as a matter of fact. Some of'em like V2500 automatically select auto ignition during approach while the others don't. The check list is a generic one and dependable on engines mounted one needs to select ignition manually if required or desired. Hope that helps. :ok:

[salamanderpress]

Many thanks for all the inputs. I now realise that "check ILS Course" means exactly that and NOT drift angle between heading and track.

The 5 deg course difference limit is probably required to ensure a successful autoland in case of a loss of LOC signal after flare (the AUTOLAND light will not come on for this eventuality below 15 ft). In this case the aircraft is going to align ground track with the IRS (runway) heading. So if this is off by any number of degrees, you would touch down with that amount of drift?

Your thoughts?

[rudderrudderrat]

Hi salamanderpress,
In this case the aircraft is going to align ground track with the IRS (runway) heading.
It will do that even when the LOC signal is present. The automatics simply align the heading with the ILS QDM ready for touch down. If there is an error, you'll feel the initial sideways acceleration on touch down and it should then track the LOC during roll out - else you'll take over manually.

[salamanderpress]

Possibly, Airbus wants the check at 350 ft, because the IRS position would slowly get updated by the localiser from the moment of localiser intercept?

If there was a considerable IRS drift at the top of descent, say 10 deg for instance, this would gradually get corrected as the aircraft comes down the beam. A check at 350 ft would then allow the IRS more time to update and if the difference had not reduced to less than 5 deg by 350 ft, an autoland is not to be done.

Would welcome any thoughts on this. I could be way off the mark?

It always amazes me that we go about discussion as if we were evaluating a captured enemy aircraft. Surely there is a way to ask Airbus what a specific limitation means or any other technical query for that matter. Every design feature is the result of an Operational Requirement and if we don't know what that is, we can only speculate.

Could anyone shed some light on this? Is there a way to access Airbus with questions like this? I'm sure they would like their "products" better understood? Or is that Not the Airbus Way?

[woodja51]

Heres a question... What if you manually dialled up an ils and put in a course that was incorrect... Would that screw up the system... I will give it atry in the KC 30 sim next week and see what happens...woodja maybe it will kick out some phantom drift?

[salamanderpress]

some suggestions.

1. Manual ILS Course. In case this is inconclusive, maybe you can try it with Localiser signal failing below 15 ft.

2. IRS Drift. Maybe you could try it with IRS drift of some huge magnitude to see if this reduces after localiser intercept?

3. Possibly, a localiser signal that remained available below 15 ft would cause the aircraft to align heading with localiser signal (most accurate input) and not with the IRS heading value for the runway.

Looking forward to your experiment result. (and maybe put in some respectable crosswind since the whole argument is about decrab...?)

[rudderrudderrat]

Hi salamanderpres
If there was a considerable IRS drift at the top of descent, say 10 deg for instance, this would gradually get corrected as the aircraft comes down the beam.
Afraid not. The FMS position will get updated and calculate a position bias to be added to the IRS position. It will interpret the difference between heading and track as a wind vector.
Airbus wants the check at 350 ft, because the latest you can Arm Approach for A/L is 400ft, therefore 350ft is your last chance to check it. FCOM 1.22.30 P50:

Where did 5 degs come from? It's not published in any of my manuals.

[9.G]

Some requirements are specific to CATIII with NO DH, because an Autoland is compulsory. These are not necessarily monitored and thus not indicated.
•
Incorrect ILS CRS ( ∆5°with runway CRS) leads to incorrect autopilot DECRAB.Check ILS CRS at 350 ft. If incorrect, continue approach down to CAT II minima (AP must be set OFF at80 ft latest).

12 - PRECISION APPROACHES - CAT II - CAT III (http://www.scribd.com/doc/51867646/16/PRECISION-APPROACHES-CAT-II-CAT-III) It's in the instructor manual. It's beyond me why anyone would start checking the course at 350 ft on a CATIII approach. It's all done way before. Anyways there's 5 degrees value but not between GT or MH or anything else in this particular case. Sorry boys I was wrong. :ok:

[salamanderpress]

Thanks a lot, Sabenaboy for a very lucid explanation. You're lucky to have such a good source for information.

And so are we, through you.

[rudderrudderrat]

Hi sabenaboy,

Thanks for the explanation.
The reason why we only check this at 350' can be found in the FCOM PRO-NOR-SRP-01:
Sorry for being a bit dense, but we check the ILS course during the briefing before the approach.
I still don't understand why Airbus requests us to do the check at such a late stage during the approach. The appropriate time to correct any error was before we commenced the approach.

[sabenaboy]

Sorry for being a bit dense, but we check the ILS course during the briefing before the approach.

Of course, we also check that the ils course is correctly inserted during the briefing. We don't recheck it at 350' during a normal ILS app, but only during cat II/III ops.

It's highly unlikely that the inserted ILS course would change after you initially made sure it was correctly inserted. However, it's only after reaching 700' that the ils course + freq are frozen, (+ <400' land mode can only be disengaged by "engaging the go-around mode). Perhaps Mr Airbus wants us to check it at that height, because only then is the ils course + freq definitively locked in the RADNAV page and can't be changed any more. (manually, inadvertently or due to some malfunction) Just guessing here, but that appears to be a plausible explanation.