Cat III autolands on non-Cat III approaches

[AnthonyGA]

Disregarding the regulatory and procedural restrictions, is it technically possible to do an autoland on an ILS approach that is not certified for autoland? Will the on-board automation catch this somehow and refuse to do it, or will it autoland on any ILS to which it is pointed?

If the latter is the case, do pilots ever autoland for currency in clear weather on an approach that isn't certified for it, or do they always wait until they actually have a Cat III approach to use?

Also, if anyone knows what technically distinguishes a Cat III ILS from a lower level ILS, I'd be interested in knowing about it. Are the tolerances tighter for Cat III approaches, or are there other, qualitative differences in the way the electronics and radio signals work? From what I've read, I get the impression that the only thing that makes a Cat III a Cat III is that it has been tested and certified to provide a specific accuracy, whereas other more ordinary ILS approaches have simply not be tested (even though they might still qualify in terms of accuracy).

[terrain safe]

Some of the more technical stuff I'll leave to others more qualified to answer but for a full cat 3 approach the runway has to be safeguarded, i.e. all aircraft at the cat 3 holding points but if they are doing it for training then the aircraft have to told that the runway is not safeguarded.

what distinguishes a cat 3 approach. lots of things including both approach lighting and also lights on the ground, qualification of the pilot and aircraft etc. Also cat 3 ils has much tighter tolerances than cat 1 or 2.

[BOAC]

Technically the 737 will 'autoland' on any ILS - BUT the runway should have the 'L' (or 'E') autoland status to be correct, and if not the results may be erratic. In certain conditions like obscured vision, 'nowhere-else-to-go-and-weather-duff' it is an option, but I'm not sure where you would stand if it went wrong otherwise. There is no special 'signal' involved in a Cat II/III ILS

[BYALPHAINDIA]

Slightly off the thread, Many Years ago there was always a 'rumour' at LBA that R14 was a 3.5 Deg App, But only on Cat 1, It did however state on most charts 3.0 Deg.

I would have thought 3.5 Deg was correct taking in the chevin terrain??:ouch:

Regards.:ok:

[411A]

The glidepath at LBA is just a tad steeper than most, due to the terrain.

As for CAT III autolands during and ILS approach which is normally CAT I, yes it is done from time to time, generally no problems encountered.
However, some CAT I ILS approaches have a notation on the approach chart that says (in effect)...not suitable for autoland.
This is generally the result of the glidepath signal having a bit of roughness at the runway approach end, for a variety of reasons, and in this case, an autoland should not be considered.
In addition, threshold crossing height needs to be considered as well.
For example, the minimum TCH for the L1011 standard body airplane is 42 feet, due to ICAO standard approach light clearance considerations.

[Kliperoo]

I can only speak for Canada, but assume it is the same worldwide.

The basic difference between the CAT's is simply the height to which you can go down to before executing a missed approach (if you dont have the runway visual). In most cases you can take the runway to mean the runway lights, lead-in lights, etc. If you see any of those, you can proceed below the Decision Height (DH)

To be designated as CAT III though, the equipment itself (the localizer and glideslope) must be certified for use beyond the threshold of the runway. This is what seperates a CAT I and II from III. In CAT I and II installations, the LOC and GS are not tested beyond and below the path past the threshold.

Now some figures.

A CAT I approach can be conducted down to 200' AGL (in most cases). Visibility must be RVR2600 (Runway Visual Range of 2600ft) or 1/2 sm.

CAT II approach is good from any height from 200' to not below 100', with a visibility from RVR 2600 down to RVR 1200.

CAT III is dependent upon the installation, but can have DH's below 100'. There are also 3 different types of CAT III installations, A, B and C of which your normal instrument rating only covers A. B has more stringent heights and visibilities and CATIIIC is strictly Autoland only (pretty much 0 Visibility, 0 Cloud height)

I hope this helps!

[AnthonyGA]

All very informative replies, thanks.

So an instrument rating covers only up to Cat IIIa? How much training and what sort of certification is required to legally conduct Cat IIIc autolandings? Are all airline crews trained and certified for Cat IIIc?

[Kliperoo]

Any aircraft that has dual autopilot (such as an A320, B767, etc) has the capability of doing an autoland. In the case of completing the approach, both autopilots must be engaged and functioning to within approved tolerances - which I dont have the figures for what those tolerances are but they will be VERY tight.

As for Flight Crew it depends on the airline itself. Quite a few airlines will certify only the Captain for CAT IIIc Operations, while others will certify all their crew in the interests of saftey.

When it comes to the training the crew conducts the approach over and over again with various emergencies (engine failures, hydraulic failures, depressurization issues, etc) in a simulator, then goes out and does it in real life.

Like someone posted above too there are also then recency requirements, which leads us back to your original posting (see Aviation is a circular business :ugh:) that there are more good days then bad, so they have to use CAT I and II installations to complete their autoland recency requirements, even if it is not "suggested" by the ruling authority.

[t-bag]

Kliperoo, do you fly with an airline?
Your post above is a load of cobblers.

[Kliperoo]

Cobblers? Im assuming (not being rude) that you disagree. Rather than getting into a slugging match PM me with what you mean.

[BOAC]

..........erm not quite sure why a 'load' of cobblers?

To answer Ant's last questions:-

1) specific training is always required to operate BELOW CATI, ie IIIa, b, c or whatever.

2) Ground studies/exam and simulator for above

3) Only if a/c is CATIIIc capable and there is time in the programme to do 2) above

[5150]

We've autolanded an aircraft on a Cat I ILS, but you must notify ATC as you will not get the protection on the ground. We did it as the aircraft needed to be checked for it - the crew get checked every 6 months in the sim.

Regarding Kilperoo's comments,

All our crew our CAT3 trained - I don't understand why only certifying the Captain is enough in a multi crew operation. You have to have both crew qualified or you can't make the approach. In our operation, the Captain is always the handling pilot for Low Vis approaches - perhaps Kilp's is referrring to that. . . ?

[BOAC]

This is a world-wide forum - always prepare to be surprised!

[Bus14]

An unusually reasonable and well presented question, but the usual mixture of good and bogus answers,

BOAC is mainly right. However, to the best of my knowledge, there are no Cat3c (no DH and no RVR limit) appoaches authorised. Even Cat3B with no decision height has a visibility (RVR) limit to enable safe clearance of the runway under manual control. Also, some aircraft are authorised for Cat2 approaches, but to a manual landing. So not all Cat2/3 approaches end in an autoland.

In the UK, subject to the airlines rules, any 'in line' ILS can be used in VMC for practise CAT2/3 appoaches to an autoland. You do not need to tell ATC (again, subject to your own airlines rules) since you will be able to monitor the quality of the guidance visually and can disconnect the automatics for a manual landing if the flight path errs. On a busy VMC day most ATC units will decline a request for them to provide Cat2/3 safe guarding due to time, manpower, and landing rate constraints. Some military ILSs are not 'in line' with the runway so that there is room for a jet barrier at the end of the runway where the ILS localiser aerials usually live. An autoland to such an 'offset' ILS would be technically possible, however, shortly after touch down the aircraft would gracefully steer itself off the side of the runway as it continued heading towards the localiser array. Indeed it may also land short of the other side of the runway. On any runway the autoland touchdown zone is a box rather than a point due to the inevitable tolerance in lateral and vertical guidance.

There are some Cat1 ILS installations that are technically CAT3 capable but for which the airport operator chooses not to go down the very expensive route of qualifying them for Cat2/3. There are also some Cat 1 installations that cannot generate a tight enough beam to deliver the Cat2/3 flight path tolerences. If you look at an ILS array, the fewer aeriels, the wider the beam. Don't quote me, but Cat 1 has something like 7 or 9, and Cat2/3 has many more. There are also ground lighting, beam protection, and physical security issues that conspire to put the bean counters off the idea of providing Cat2/3. The airport, the aircraft, the airmen, and the airline all require qualification and certification in order for Cat2/3 ops to be approved.

Under JAR (euro) regs the pilots can achieve and maintain currency in the sim. Nevertheless, as the same rules require the aircraft to be regulalrly checked, it results in practise cat2/3 approaches being conducted in VMC, although a real approach in fog would, of course, also count towards the aircraft monitoring programme.

Apart from the requirement for an 'in-line' ILS, there are, as already noted, a few reasons why some Cat1 runways are not suitable for practise autolond. At BRS (Bristol, EGGD), the upslope before the threshold and the 'hump' on the runway on 09 denies the aircraft system the chance to correctly calculate the flare. However, the other end (27) is Cat3 capable. At MAN (Manchester, EGCC), a steep valley on finals for 05L precludes Cat2 ops for large aircraft, but does allow Cat3, ironically, as the system (particularly the rad alt) can get it's act together for an autoland once the valley has been overflown.

Hope this helps:)

[t-bag]

777/747/757/767 have 3 a/ps. There is a requirement to "fail operational" to operate down to DH below 50` and RVR below 125m. 2 a/ps can only "fail passive" therefore cannot.
Under Jars cat 3 C i.e. 0/0 does not exist therefore no JAR operators are so certified.I am not aware of any company in the world that is so certified (stand to be corrected). The minimum is cat 3 B no DH and 75m required RVR.
Both crew members ,the company and the aircraft have to be qualified to operate below cat 1.
A Jar instrument rating only qualifies you to Cat 1.
Depressurisation "issues" are not part of LVO training.

[AnthonyGA]

With reference to the touchdown being in a box rather than on a single point, I recall reading that the first ILS systems for autoland were so accurate that they caused runway damage, because heavy jets would always touch down at exactly the same spot on the runway, until finally the runway started to crack at that point. The solution was supposedly to include a random variation in the exact touchdown point that varied within a precisely defined zone, thus spreading the wear and tear around considerably.

Does anyone know if this is true? If I remember correctly, Seattle was mentioned as an airport that had a problem with runway damage from too many touchdowns on exactly the same spot.

[Bus14]

T-bag, as BOAC has already councelled, be prepared to be surprised. A 2 A/P system can be fail operational - if one fails, you still have the other. It is only after the first one fails that the A/C becomes fail passive. if you are used to 3 A/P systems you may have been told differently, but that doesn't make it universally correct.

As to deliberately spreading the touch down box, sorry AnthonyGA, that sounds like moose breath to me.

[Capt Pit Bull]

As to deliberately spreading the touch down box, sorry AnthonyGA, that sounds like moose breath to me.

Well, I was told that by the representative of a certain well known avionic company in the context of DGPS based autolands. Doesn't mean its true of course.

pb

[t-bag]

Bus, prehaps a different context , to my knowledge you cannot autoland on one auto pilot therefore if you start with two and one fails you are fail passive not operational for an autoland .

[Bus14]

T-bag, as you suggest, we may be thinking in different contexts. To clarify my point though, the Airbus A319/320/321 series has 2 autopilots. With both A/Ps working (amongst many other things, of course), the bus is Cat3B, fail operational, and can operate to No DH. With 1 A/P working it is Cat3A, fail passive, 50' DH, but can still autoland. Indeed, my understanding of JARs is that the only mode authorised for a planned auto approach/manual landing is Cat2 100' DH (depending on the reason for the downgrade the A/C may be fail passive or fail operational, but limited to Cat2 for some other reason). The aim of all low vis ops rules is to demonstrate the same safety level as Cat1 ops. The JAR rules interpret 80' ie 80% of the 100' DH as the lowest height to transition to manual flight in poor vis and still achieve a VMC equivalent safety level.
This is true for an Airbus in JARland, but other aircraft and authorities may differ.

[ComJam]

The only way to vary the touchdown point with any success would be to move the glideslope antenna (not possible) or vary the GS angle which has very tight tolerances making it not possible.

[BOAC]

As has been pointed out before, changing the GS angle does not affect the design touchdown point.

[411A]

If I recall correctly, the first jet transport to be capable of automatic approach/land (autoland) ops (the HS.121 Trident) it was found during initial trials that the glidepath had to be 'altered' in some way to prevent pavement damage, from touchdowns in exactly the same place, with each flight.
Just how this was done perhaps others can answer.

And, with regard to two vs three autopilots for autoland CATIII ops, it should be remembered that the best of the best (TriStar) has only two autopilots, but each is dual channel...in effect a quadraplex system, not triplex, as on other aircraft.

Ah....Lockheed.:E

[ComJam]

BOAC, true. Anthony's last post suggested a "random variation" might be applied, i was pointing out that neither option on the glideslope was possible. I should maybe have re-read my post before i posted it! :}

[AnthonyGA]

I only recall reading something about the touchdown point being varied so that the wheels would not always hit the same spot.

I'm tempted to think that I read this in a Time-Life book (The Jet Age?), but it was a long time ago, in any case (obviously still after ILS came into use, of course).

[BOAC]

I only recall reading something about the touchdown point being varied so that the wheels would not always hit the same spot - you may have seen that in my book, "How to fly erratic ILS glidepaths and landings".:)

[Bus14]

There were many factors in the Tristar incident, I don't recall the ILS being in any way culpable.

[BOAC]

AGA has been asking lots of questions around the place and hopefully has learned a lot from the answers. I think we need to clarify the question, as I think the term used, a 'CATIII' autoland, is misleading, and no doubt due to lack of knowledge of the operation. Autoland does not relate to ILS Category. The 'success' of an autoland depends on the quality of the ILS, protection provided and approach terrain.

It is by no means unusual for a CATI ILS to be annotated as 'suitable for autoland' but this has to take place in CATI or better. The ONLY time I would autoland BELOW CATI limits on a CATI ILS would be if there was absolutely nowhere else to land.