A320 autoland : what’s the significance of ILS course check at 350ft RA?

[C.M]

If you are flying the Airbus then you are more than familiar with the fact that during an autoland approach we are suppose to check for correct ILS selected course at 350ft RA.( just after the LAND mode appears)
What’s the significance of the 350 ft since a check of the ILS course is done during prior checks? Does something happen internally that could alter the ILS course when the LAND mode appears?

[LW20]

below 700 ft RA data coming from the FMS is frozen. So your check could be done anywhere below 700 ft.

[gearlever]

Oooh many moons ago, 19 years or so.
But I think it has to do with AP decrab.

[C.M]

below 700 ft RA data coming from the FMS is frozen. So your check could be done anywhere below 700 ft.

You have a point there ,but still it doesn’t give a justification of the overall concept of checking the selected ILS course especially at such low point . The course is generally checked as part of the briefing for the approach . Would there be any point subsequently where it could change ? If there was such a case , then assuming the last point were this would happen was at 700ft then all common sense would dictate to do the check right at 700ft RA . At least there you may have some time to revert to CAT 2 ( which is what you are suppose to do if the course selected by the FMGS differs by more than 5 degree from the nominated course)

[Goldenrivett]

The course is generally checked as part of the briefing for the approach . Would there be any point subsequently where it could change ?

I agree. I can't see how the course could possibly be "accidentally changed".

However on a previous aircraft type, the Omni Bearing Selector (OBS or QDM selector) was next to the Heading and Speed selectors and it was very easy to move the wrong one. The last chance we had to correct the error was by 350' else the Auto Land system would miscalculate the amount of Align to apply during the flare.

I think the "350 feet" rule is a hang over from the old days and needs to be updated with the modern equipment in use.

[pineteam]

Message deleted due to wrong information.

[Jesse Pinkman]

You won't get indication of localiser deviation if the aircraft is following the actual beam, the selected course is irrelevant in this sense.

During Loc Intercept, the ac uses the selected course to discern direction of turn when it detects a localiser. Once pointing in roughly the right direction, 90Hz/150Hz beam differentiation is all the ac needs to track the localiser but when it gets its first 'sniff' of a beam it has no idea what direction to turn in or what course to turn on to.

Why we check it again before autolanding I do not know, but I suspect it's something to do with the transition from wingborne flight to wheelborne rollout.

[gearlever]

YEP:ok:

Decrabbing?

[Goldenrivett]

Affirm.

DSC-22-30-80-30-10 Precision Approach Modes:
"ALIGN SUB-MODE
ALIGN is a sub-mode of LAND mode that lines up the aircraft's axis with the ILS course.
It is not displayed to the flight crew."

[gearlever]

So waiting for dementia (A320 was 19 years ago for me) :rolleyes:

[Jesse Pinkman]

Great find Goldenrivett! I'd never seen that before.

[C.M]

However on a previous aircraft type, the Omni Bearing Selector (OBS or QDM selector) was next to the Heading and Speed selectors and it was very easy to move the wrong one. The last chance we had to correct the error was by 350' else the Auto Land system would miscalculate the amount of Align to apply during the flare.

I think the "350 feet" rule is a hang over from the old days and needs to be updated with the modern equipment in use.
I don’t think Airbus intended to adopt this procedure based on how other aircraft used to be . Prior to the 320 family was the A310 whose MCP (FCU) was again not the same as a typical Boeing (the CRS selector on the 310 was on the pedestal so you couldn’t accidentally change the OBS if you were changing the heading bug) . One thing you will definitely have seen is that Airbus in all likelihood is obsessed with avoiding doing things and naming things the same way Boeing does .
The 350ft check remains a mystery

[Goldenrivett]

(the CRS selector on the 310 was on the pedestal so you couldn’t accidentally change the OBS if you were changing the heading bug)

But you could accidentally change it when attempting to change a radio frequency.
The point I'm trying to make (rather unsuccessfully) is that you can't accidentally change the course in a modern glass cockpit. The 350' rule could be moved to 700' (data lock).

[applecrumble]

This is for decrab purposes but I also agree 350 feet seems to be stuck in the past. This should be done at localiser capture or before so you can change it before the FMGC freezes below 700'.

[Jesse Pinkman]

Affirm.

DSC-22-30-80-30-10 Precision Approach Modes:
"ALIGN SUB-MODE
ALIGN is a sub-mode of LAND mode that lines up the aircraft's axis with the ILS course.
It is not displayed to the flight crew."

But this. ALIGN is part of LAND mode. LAND mode doesn't annunciate till about 400'. I guess we can't really check ALIGN's reference until ALIGN is active. Who knows what it might be doing to that course in the background.

This is really a question for an Airbus FMGC IT programmer, isn't it?

[Goldenrivett]

I guess we can't really check ALIGN's reference until ALIGN is active
But that's why we check the ILS QDM by 350 feet - because the aircraft will ALIGN to that reference on touchdown then track the centreline using the LOC. If the ILS QDM is in error, then you'll land with a crab angle.

[Jesse Pinkman]

Goldenrivett, I think you and I are saying the same thing. I was answering Applecrumble.

[C.M]

Goldenrivett

With this we are in total agreement .

[ANCIENT]

Can someone give me a reference for the requirement to check LOC QDM at 350ft.
My manual is current(5/12/17) and I can not find anything like what is discussed here.
Quote: AT 350 ft RA LAND mode CHECK ENGAGED/ANNOUNCE.
That is all we are required to do.

[C.M]

There are a few places in the FCOM such as the sources I am pasting below :

PRO-NOR-SOP-19 ( Under AUTOLAND )
“ AT 350 FT RA ILS /GLS /MLS COURSE on PFD........................................CHECK
If the ILS /GLS /MLS course pointer and the runway track differ by more than 5 °, autoland is not authorized.”

PRO-NOR-SRP-01-70 (under APPROACH MODE ACTIVATION (LOC -G/S) )
“ 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.”

[ANCIENT]

Thanks C.M
We don't do autoland so had not dug that deep.

[seven3seven]

In my previous 320 job, I have never seen the track/course differ by more than 1degree.

[applecrumble]

Just to confirm this as I have checked on a simulator. The aircraft will change the localizer dagger on the PFD to the front course loaded into the MCDU (so long as this is changed before the MCDU data lock at 700AGL, anything below 700AGL has no change).
The aircraft will continue to track the localizer like normal showing the lozenge central on the PFD (still tracking the correct localizer beam) but the dagger will be whatever is loaded in the MCDU. Then in the flare the aircraft will align with the course entered into the MCDU. The 5 degree limit (before go around is required) between ILS course and course on the PFD is mentioned elsewhere in the FCTM where in a crosswind you can land with up to 5 degrees of drift. Anything greater than this then you may risk damage to the gear due to torsional forces.

[FlightDetent]

Thanks for that, mighty good work!

[Bendig]

Hi Guys. If I’m not mistaken, the FMS data is frozen at 700ft but the FCU is only frozen at 400ft hence the check at 350ft and not earlier.

[Escape Path]

That's quite a nice find! Ok, now, we check it for a reason. If the course it's wrong, then what? You can't change it because the mcdu is locked. So, manual landing? Revert to fail passive if in CAT III ops?

[Denti]

If it differs by more than 5 degrees its a mandatory go around.

[Escape Path]

Thanks Denti. Do you have a reference?

[Denti]

Well, the basic reference has been given in post #20, however the specification to 5 degrees i found only in the OM B of my current airline.

[Escape Path]

Ah thanks for that one. I sort of missed it

[C.M]

applecrumble

That deepens the mystery .... Its commonly said and understood that the roll-out mode simply tracks the LOC beam ...... in the same way it does during the approach... now it would appear that a difference course in the FMGS acts as a track command on ground .

[EGPFlyer]

The track is used to align the flare down the correct course (remember you can lose the localiser below 15’ and the aircraft will still autoland). You won’t get the yaw bar if you don’t have a LOC signal at this stage.

[Goldenrivett]

Hi C.M
now it would appear that a difference course in the FMGS acts as a track command on ground
No. The course in the FMGS will be the aircraft's touch down heading.
The LOC signal will guide the aircraft along the centre line. That's why there is a 5 degree error limit.

[C.M]

Hi goldenrivett
You are right , in the flare mode it should be heading so a de-crab and align the nose with the runway.

[C.M]

Hi goldenrivett
You are right , in the flare mode it should be heading so to decrab and align the nose with the runway. And once ROLL OUT mode engages , whatever is in the FMGS should be irrelevant since after touch-down it is the LOC signal that guides the aircaft . To sum -up the whole issue should be for correct decrabbing - runway alignment to avoid main wheel torsional stresses.