ILS obstacle clearance area

[greek-freak]

Don't know if this is the right forum to post this.

Does anyone know where I might find the specification for the
required 3D obstacle clearance area that must be around the ILS
centerline such that an ILS can be installed at an airport.

I am just SLF but due to the recent closures of LGIR because of low visbility
caused by Sahara sand I wondered if an ILS could be installed for RWY 27 but
a hill seems to be in the way.

Thanks a lot

[BOAC]

I suspect the problem would be the mast 587'AMSL at around 2 miles on the centre-line which would, I'm sure, infringe the protected area. (NB I am at home and looking at a chart which is not current, so the mast may not in fact be there now).The hills would not pose an obstacle problem for an ILS approach. It is probably down to money! I would suggest that the mast would need to be changed (lowered/moved) to allow an ILS but I'm sure an ILS expert will answer in detail. Perhaps someone can confirm the mast is still there?

[greek-freak]

Thanks BOAC :-) it is still there, the problem is not the mast of 587', but the rather solid hill beneath it. Somehow harder to remove ;-)
Badly enough there are some military facilities on that hill which makes removing those and blowing the hill up harder.

[BOAC]

Not necessarily a problem - the ILS slope CAN go up to about 4.5 degrees for most a/craft. I have no info on the height of the terra firma, just the top of the mast. If it is a small mast on a big hill, the g/s will need to go up.

[greek-freak]

It is indeed a small mast (at most an estimated 60' high).
Forgive me a naive question: Is a 4.5 degree GS really that common?

I am just wondering because, at the new athens airport they blew up large parts of a hill to install a 3 degree GS for 21R (I don't know though if 4.5 would have made much difference in that case).

[BOAC]

No 4.5 is not common. I'll have to 'duck out' now on this discuusion as it is down to the ATC experts to advise on feasibilty. Good luck!

[chevvron]

ILS can be almost any GP angle within reason; London City has GP's of 5.5 deg. ICAO won't allow an ILS to be Cat 1 if it's more than 3.5 deg, and Cat 11/111 MUST be 3 deg. I don't have access to charts for LGIR. Is there a non-precision approach to the runway in question? Does it have a 'nominal' gradient (something like 5% = 3 deg) or RoD for this approach? If so that would give you a rough clue as to the likely GP or angle

[greek-freak]

Thanks chevvron :ok: so that explains why a GS with more than 3 degrees is not very common.
There is a VORDME approach for rwy 27 in Heraklion, it starts at 7 DME and 2,500' (the VOR is located slightly to the north and beyond the threshold of rwy 27) the gradient is 5.8% down to 1,100' (threshold elevation is 82') which is the MDA for that approach (RVR for a straight-in approach is 2000m).
The reason for that MDA is the high terrain to the south of the approach path which runs along the northern coast of Crete.
I just wanted to find out if this terrain is hindering the installation of an ILS.
Thanks guys :)

[BOAC]

A small correction to chevvron's post (and to mine! as I'm not sure about the 4.5 degrees on reflection) - LFML Marseille has a 4 degree Cat I ILS on 32R.

[chevvron]

Don't know how to work it out exactly, but 5.8% would be roughly equivalent to 3.3 deg GP assuming continuous RoD rather than descent at FAP. This is the thing, the minima for non-precision is calculated so that you can if you wish descend straight to MDA (1100ft) at 7DME and still maintain obstacle clearance, which in this case is 246 ft(75m). If there are no obstacles closer in, and there is sufficient room on the airfield to install ILS LLZ and GP, then you might be able to get down to Cat 1 minima. This would also require terrain clearance for a missed approach climb gradient of 2.5%
BOAC: just quoting ICAO re the GP angle, the French may have a 'difference' filed with ICAO.

[OzExpat]

It's a bit difficult to provide a categorical answer but I'll base my remarks on the information already provided. Firstly, the non-precision approach (VOR/DME) to RWY 27 should be designed so that the aircraft will reach MDA at a distance that will allow continued stablised VISUAL descent to touchdown. This is the fundamental basis of Pans Ops design criteria for runway approaches.

Using this as a basis for hypothesis, the aircraft should reach MDA at a distance of approximately 3.04 NM from the THR. This is the point where the MDA will intersect the 3-degree approach path to the THR. Now then, if the big hill (plus antenna or whatever) is 2 NM from the THR, an aircraft descending on a stable 3-degree approach will be at about 720 feet at that point, which suggests that there might be enough clearance but, without looking at the Obstacle Limitation Surfaces for the aerodrome, I don't want to commit myself to that.

If we consider the lowest protection slope angle to be a gradient of 2.5%, this equates to an angle of 1.43 degrees. The slope from the THR to the 587 feet obstacle at 2 NM would be about 4.2%, or close enough to 2.4 degrees. On this basis, I'd suggest that it would be a serious infringement of the approach surface and, therefore, the only solution would be to increase the GP angle (and therefore, commensurately increase the lowest protected slope).

All in all, I'd say it'd probably be a bit difficult to get an ILS approach to that RWY. I could be wrong, on many counts here - not least of which being the late hour at which I'm making this post while feeling somewhat weary.

[AlR]

3 degree is fairly average and provides for an arrestable sink rate. Transport category Aircraft will consume considerably more Tarmac when reaching min's and trying to arrest the sink rate from a 3.5 degree G/S and greater.

Some Aircraft with lower Vls/Vref speeds are able to compensate for 3.5 degree and greater by the slower Grd Spd. However at the 140 kt Vls / Vref it would take a very harsh round out to match the G/S touch down area plus 100 m.

[greek-freak]

Thanks guys, I will try to find exact topological information about LGIR.

But still, is nobody here who knows the definition of the required obstacle clearance area around the ILS centerline? It should logically be either a circle (ellipse) or a rectangle centered at the ILS axis with decreasing surface towards the threshold, i.e. depending on the distance from the rwy end.

Thanks again :ok:

[OzExpat]

It's not quite as simple as you seem to think it is greek-freak. ILS procedure design involves a complex 3-D analysis of obstacles through Collision Risk Modelling. This is used even when there are no obvious infringements of the approach/take-off surface in the Type A Chart.

[greek-freak]

OzExpat: thanks I was only making assumptions, do you have any link or documentation about how this is modelled?

In real life I am a researcher in bioinformatics, so I should hopefully be able to understand it, even if it's not as simple as I seem to think :O

[Red Four]

If you need to get to sleep, try ICAO Doc 8168
http://dcaa.slv.dk:8000/icaodocs/Doc%208168/

[greek-freak]

Thanks a lot RedFour, I had found that chapter in the meantime :zzz:

[Spitoon]

Annex 14 may also be of interest - particularly chapter 4. Courtesy of our Danish colleagues. (http://dcaa.slv.dk:8000/icaodocs/Annex%2014%20-%20Aerodromes/Vol.%201-%20Aerodrome%20Design%20and%20Operations,%20AMDT%207.pdf)

[discountinvestigator]

The running of the CRM only takes place if the two other obstacle overlay methods have failed. It may take place in some states as a matter of routine. However, for those of us who have spent some time looking at the mathematical basis of it, it does not stand up!

Unfortunately, the obstacle evaluation surfaces are counter risk intuitive, so you get worse obstacle environments when you do not have guidance over the obstacles. Just look at the surface for a slope to a Code 1 runway on a visual approach.

You may well need a step down fix limitation in relation to the obstacle that you discuss.

The Athens rock blowing was for departure slope conditions, from memory.

[OzExpat]

However, for those of us who have spent some time looking at the mathematical basis of it, it does not stand up!
Has anyone informed ICAO of this?

[reynoldsno1]

I know it (the CRM) was based on a relatively small statistical sample, and that the it was restricted by the limtations of Fortran at the time - but I didn't think that much mathematics was involved.....