Hong Kong - why 5NM intercepts?

[Flaps none]

If any HK approach controller is out there, would love to know why we keep getting these close-in intercepts.
They cause almost more safety reports in our airline than any other location worldwide.
It can't be for separation reasons as surely the separation/speed control can all be done before Limes.
Occassionally when there is no traffic early in the morning, there is a certain male controller that does indeed give "Direct Limes, cleared ILS 07L" whch makes life sooooo much easier!
"Direct to Limes, cleared for the ILS, maintain 5NM seperation from proceeding/180 kts to 8miles" would be much easier after a 16 hour flight and I am too tired to see my instruments. :)

[On the beach]

Why do you consider 5 mile intercepts to be close in? Provided that you are at 1500 ft it should be no different to intercepting at 10 miles at 3,000 ft.

I would suggest that a look at the airport statistics will show you that capacity at CLK coupled with actual traffic movements have increased by 30% in the last 5 years. In fact capacity is set to increase by another 15% over the next 4 years.

So, 5 mile intercepts are probably going to be common practice as ATC utilise all the available airspace available to achieve and sustain this increase in movement rate.

"Direct to Limes, cleared for the ILS, maintain 5NM seperation (sic) from proceeding/180 kts to 8 miles" would be much easier after a 16 hour flight and I am too tired to see my instruments

Your comments above, perhaps, show the true problem, specifically ...I am too tired to see my instruments.

I would respectfully suggest that you cite this on your "safety reports" and not try to shift the blame onto others who are trying to get the most benefit for the most aircrews in a severely limited amount of airspace.

And by the by, 180 kts to 6 miles would be much easier for Approach controllers to achieve 32 landings an hour. :)

Try and visit the centre for a visit some time. I'm sure you'll be surprised to see life from the other side. I know that most of the visitors on Command Courses have had their eyes opened.

P.S. You may get more replies if you cross-post this on the "Fragrant Harbour" forum.

[HEATHROW DIRECTOR]

You don't see too many 5nm intercepts at Heathrow where they're whacking in very considerably more than 32 an hour! Short final turn-ons come when it's quiet and the pilot accepts it, not when it's busy.

[DFC]

Why do you consider 5 mile intercepts to be close in? Provided that you are at 1500 ft it should be no different to intercepting at 10 miles at 3,000 ft.




Indeed you are 100% correct.

They both fail to meet the applicable requirements.

The flight should establish on the centerline at a distance which enables 2nm of level flight before final descent. i.e. in your examples the flight should be on the LLZ at 7nm (1500ft) or 11nm (3000ft).

The exception to the above would be when procedures for a continuous descent are applicable and that is not the case for the 07L or 07R.

[On the beach]

HD,

Hong Kong has a unique set of limitations. To the north and west there are airspace limitations due to Macau and Shenzhen airports and their respective approach and departure procedures and to the south there is terrain over 3,000 ft high within 4 miles of the runway.

This means that even if you fly the full ILS procedure on 07L you still only establish on final at 8.3 DME at 1,700ft. The 90 degree turn onto final from base leg means that if the approach controller has less than 5nm separation on base leg then radar separation has been lost (3nm in HK) in the turn onto final between 2 successive aircraft. Tromboning, as you carry out at LHR, is simply not an option due to the above airspace limitations. All sequencing has to be carried out well before base leg with minor adjustments being made through speed control and a small 60 degree closing leg turn onto final which is generally only done if the aircraft are close to or at 1,500ft.

Hong Kong does not have the luxury of a left base or straight-in option for 07 and is limited to vectoring all aircraft onto 07L via a right base only. Similarly, only a left base for traffic landing on 25, however the final is a little longer stretching out to 15DME on 25.

Have a look at the charts here and you'll see what I mean:

http://http://www.hkatc.gov.hk/HK_AIP/AIP/AD/HK_AD2-93A.pdf (http://www.hkatc.gov.hk/HK_AIP/AIP/AD/HK_AD2-93A.pdf)

http://http://www.hkatc.gov.hk/HK_AIP/AIP/AD/HK_AD2-101.pdf (http://www.hkatc.gov.hk/HK_AIP/AIP/AD/HK_AD2-101.pdf)

Unfortunately, Hong Kong airport was not built in an ideal location. The alternative to "short-final turn ons" as you call them is less capacity/more holding. I agree, it's not an ideal situation but then the airport is not in an ideal location and you have to adapt to what you've got. Also, Hong Kong only got a dedicated final approach director a couple of years ago. I'm now retired so can only look on from afar and admire the skills of those who remain and continue to increase the capacity of the airport year in, year out despite its limitations.

[throw a dyce]

Onthe beach,
I think the gweilos left are all older than you as well.Most would be travelling around the UK on mega bus for nothing.:)

Bring back Kai Tak.:oh:

[HEATHROW DIRECTOR]

On the beach... OK. Fully understood thanks to your explanation. Sounds like a nightmare!

[LapSap]

Occassionally when there is no traffic early in the morning......."Direct Limes, cleared ILS 07L"

I think you just answered your own question.

DFC
The flight should establish on the centerline at a distance which enables 2nm of level flight before final descent.

You mind quoting the relevant International standard that relates to the figure of 2 miles?

[DFC]

You mind quoting the relevant International standard that relates to the figure of 2 miles?


Sure.

ICAO DOC 8168, Volume 2, Table II-1-1-1 Minimum Distance Between Localizer and Glidepath Interceptions

To quote ICAO 8168:

1.3.3.2 Segment length should be sufficient to permit the aircraft to stabilize and establish on the localizer course prior to intercepting the glide path, taking into consideration the angle of interception with the localizer course.
1.3.3.3 Minimum values for distance between localizer and interception of glide path are specified in Table II-1-1-1; however, these minimum values should only be used if usable airspace is restricted.

(My emphasis)

[LapSap]

Thanks DFC.

I note you are quoting the PANS-OPS procedure design criteria for the intermediate approach segment.

However, from an operational controller's point of view, PANS-ATM DOC 4444 is not so specific:

8.9.3.6 Aircraft vectored for final approach should be given a heading or a series of headings calculated to close with the final approach track. The final vector shall enable the aircraft to be established in level flight on the finalapproach track prior to intercepting the specified or nominal glide path if an MLS, ILS or radar approach is to be made, and should provide an intercept angle with the final approach track of 45 degrees or less.
(my bolding)

While I fully agree that aircraft should be able stabilize before commencing the approach, and in fact I was taught that 2 miles was the requirement, my experience is that the vast majority of modern aircraft operating here tend to fly a continuous optimum descent profile, regardless of the altitudes we give them i.e. very few will fly a level segement even if put on the LOC 2 miles before nominal G/S intercept.

[DFC]

All one is doing when one is vectoring to the localizer is performing the initial approach segment.

i.e. under vectors, initial approach = vectors; intermediate approach = established on localizer; final approach = localizer and glidepath.

Yes, you are correct that most operators will aim to fly something like a 5.2% descent (3 degrees) all the way from cruise to touchdown.

However, the result is no different if they fly a procedural initial and intermediate approach or a radar vectored initial approach to intermediate approach - they should have the opportunity to fly level (or very close to it) for a suitable amount of time in the intermediate approach phase.

If they have no need to do so then that is good for them but just in the same way that I would be unhappy if you told me to turn-in tight on a procedural approach and establish at the FAP, the same applies when radar is replacing the initial approach leg.

Practically speaking, I look at the lowest platform altitude / published glide-slope intercept altitude and expect to join the localizer 2 miles outside that when cleared to that altitude unless on a continuous descent approach. If you tell me otherwise (tight approach) then I may accept it or I may not - it is my choice.

Most people are reasonable flexible and ATC get it just as wrong as pilots from time to time and we all have experienced the "request lower to establish on localizer before glideslope" situation either due to pilots or ATC cutting it too tight but in the end whoever cut it too tight has to realise that while they got away with it, it is not the standard way of doing it.

[throw a dyce]

Flaps none,
How are you supposed to ''Maintain 5nm separation from preceeding/180kts to 8nm''..Use TCAS,or do you have a secret radar consol on board.Then you can't read your instruments,so that will be an airprox then.You certainly won't hear that phrase from me on radar anywhere.
Are you hand flying these 5 mile intercepts,or is the plane doing it? :hmm:

[LapSap]

No argument with any of you last post DFC.
We have been reminded not to cut aircraft in so tight by safety managers, particularly in light of it being one of the contributing factors in the Turkish Airlines accident at AMS.
Some people are obviously still not taking heed.

[Flaps none]

thanks for the great info OnTheBeach. If we had those types of charts onboard then you wouldn't get any more queries asking "what speed for the arrival"!, as all we have is the company "translation" of the charts which are not exactly that clear compared to the originals.

Although I probably didn't phrase the original question particularly clearly, I was really wondering why we often get vectored tight around Lantau. Is it because of Macau traffic or is it to reduce track miles? On a busy day it seems to really overload the controllers and increase everyones radio workload.

With regard to 1500ft intercepts, well sure we could intercept at 100ft. The problem with a 2NM level segment is that with full gear and flap, we really don't want to have to go from a descent to a level segment as then we have to spool the engines right up because of all the drag of the gear/flaps, which also requires a large aircraft pitch change at level off and then again as we re-intercept slope to start going down again - this is "untidy' flying, and leaves less time to produce a "stable" approach. We try for continuous descents which means a near simultaneous intercept of LOC and GS whilst descending at around 1000fpm, which can be a bit sporty to get right at 1500ft.
If we can intercept at say 3000ft, at this point we will have just half the flaps out and no gear, so it is less of a big deal to fly level if we have to, or even better we have more time/distance to finesse the intercept by adjusting the vertical speed without having to level-off. Additionally the auto-pilot can fly the full approach and profile beautifully if we are not vectored, which is a real workload saver, especially in a typhoon.

Ultimately, with a 1500ft intercept, there is no time to recover from an incorrect intercept, or inadvertant "capture from above", which either results in the requirement to go-around, or the requirement to visit the chief pilots office to tell him why you didnt go-around but should have. (company requirements are very specific about speed, position and configurartion on approach, and if not met require a mandatory go-around)Yes I know it's 2 minutes before touch down but you chew it up very quickly when things don't work out as they should.

So although 2NM may be the minimum, respectfully, it shouldn't be used as a target.
And besides, it would be great if we could avoid it, if for no other reason than to stop every single person I fly with to stop whinging about it :)

[HEATHROW DIRECTOR]

Tightest one I saw was a visual approach by a 747, turned off the downwind leg abeam the field and straightened up on final at about 2 miles. I said: "Wow, some approach". Reply: "Coupla 'Nam vets up here boy!"

[BOAC]

How on earth did we finish up talking about vertical profiles when the question was about lateral positioning?:ugh:

[DFC]

The problem with a 2NM level segment is that with full gear and flap, we really don't want to have to go from a descent to a level segment as then we have to spool the engines right up because of all the drag of the gear/flaps, which also requires a large aircraft pitch change at level off and then again as we re-intercept slope to start going down again


Remember that the whole reason why the intermediate phase of an instrument approach is level or has very little descent is so that you can slow up and configure the aircraft. If you are already configured then this is an indication that the 2nm+level segment is in the wrong place.

As I said earlier, while the design is level or clsoe to there is no mandatory requirement to fly level - it simply gives you the opportunity to do so if you want (need) to.

We have all operated with the dive and drive jockey who descends rapidly to the cleared level with no awareness of the ideal profile to go and failing to recognise that at 7000ft cleared to 3000 with 23 miles to run descending at 2000ft per minute is going to cause a level-off at some stage.


How on earth did we finish up talking about vertical profiles when the question was about lateral positioning


Becasue both are linked when flying an approach?

[BOAC]

Becasue both are linked when flying an approach? - not in my book when I'm looking ONLY at localiser intercept range as was the OP. Now, IF the question had been about glidepath intercept you talk sense..........................................

[DFC]

- not in my book when I'm looking ONLY at localiser intercept range as was the OP


So ignoring the glide-slope of the ILS for a minute.

When ATC vector you onto the localizer only approach they should have you established 2nm or more before the point at which you will descend fro landing

When ATC vector you onto the final approach track of a VOR approach they should have you established 2nm or more before the point at which you will descend for landing

When ATC vector you onto the final approach track of a NDB approach they should have you established 2nm or more before the point at which you will descend for landing

When ATC vector you onto the final approach track of an SRA approach they should have you established 2nm or more before the point at which you will descend for landing

When ATC vector you onto the final approach track of a Visual approach they should have you established 2nm or more before the point at which you will descend for landing

The idea is to give you the opportunity to fly level to configure if you want

Hows that? :) :) :)

[BOAC]

I'll take your word for all those but the localiser which I do know. The idea of that has nothing to do with 'flying level' or 'configuring' but all to do with having a chance to establish fully on the LOC BEFORE engaging G/S or 'APP'. Even if you were 'flying level' or 'configured', intercepting at 5D at 1500' (ie exactly at G/P intercept point) is too tight for airlines that require LOC capture before Approach mode. In the UK especially you would have no chance to 'call established' and get clearance to descend on the GP before you were above it!

This is a known 'event' at AMS as previously mentioned.

[Flaps none]

so can we conclude this thread with HK ATC promising to clear all aircraft for the full arrival via Limes, thereby providing separation after Limes with just speed control instad of vectors? :D

[throw a dyce]

Flaps None,
You will never get that from any Approach Radar controller,because he/she has basically lost control in an area where there is almost no room to play with.Approach radar use headings,vertical separation,and finally speed control to achieve whatever final spacing is required,according to differing criteria.I think that you need to visit an Approach radar unit rather quickly,because you don't seem to understand what is required from controllers.Worrying if you are long haul.:eek:
I would vector you on a 7-6 mile final,descending to 1500ft,and good-day.If you complained about that then I give up.
Ex HK ATC.:)

[Flaps none]

ThrowADyce,

occassionally we get exactly that instruction from some of the better controllers. Think you should visit a flight deck one day.
Seems to me that if you have to rely on the last 10-15 miles to get the required spacing then the controlling upstream is not very effective.

What is the point of designing the arrival if you never use it :ugh:

[throw a dyce]

Flaps none,
The instrument approach chart has speed control built in ,so you can't separate using speed control.Therefore if you request flying the arrival as published,the radar controller has no means of separating you from proceeding or following aircraft.The only way is build in a lot of extra spacing and hope.Then if someone gets it wrong or slows up to early,you can kiss your licence goodbye.
I use the term radar vectors ILS/DME approach.That is exactly what you get and the final director has to be very accurate to get it right..
Also these approach charts have an element of procedural in them.That approach can be flown from Limes without controller vectors.However with everything procedural it slows things down.At my old unit in the UK there is a published procedural ILS approach,used without radar at night.If we did what you are requesting here,and the subsequent aircraft requested the same,the landing rate goes down to 8 an hour.No vectors,full approach flown,speed control as per approach chart.OK an extreme example but you can look up the plate if you want.So translated to HK when you get to CLK after 16 hours airborne,you would be faced with extensive holding,instead of direct Limes,which in itself is not flying the approach procedure.:uhoh:
I have been on flight decks,and it was quite interesting,pre 9-11 when most airlines stopped fam flights.
So that would be 10 aircraft types and 5 heils that I've been on or flown in the sim,excluding the ones flown with a ppl,IMC,night rating.:ok:Have you been to 15 approach radar units?