Circling approach procedure

[mitsupini]

I am just an aviation enthusiast and simulator user, probably not very popular kind in pprune but nevertheless I felt that this is the best place to ask.

The issue is real-life circling procedures, compared to some gross math I ran regarding a typical scenario (ATR72 circling to land following an instrument approach; circling minimums is 1500ft AGL, vis 2.4km).

Basic assumptions:
1. Turn 45deg off RCL for 30s, gear down flaps approach (as in ATR FCOM; Airbus also I think). Assumed KIAS for this leg 140knots.
2. Downwind at 130KIAS, start timing when abeam landing threshold. Timing assumed 45sec (no wind) (as in FCOM). Keep circling minimum of 1500AGL until next phase.
3. Turn base and start descending at the same time at 120KIAS; continuous turn until finals (according to calculations there is no room for straight base leg or delayed descend).
4. Cat B aircraft figures used.

The math show that:
1. From turn-base to finals a constant 27deg bank angle is needed.
2. The descend rate needed is 1027fpm.
3. If the timed downwind section is flown for 60sec (40sec/1000ft rule), then bank angle needed is again 27deg and descend rate 866fpm.

So, questions are:
1. Since circling starts with 30sec/45deg turn, the distance from rwy while downwind will be 0.82nm, which dictates a rather steep bank angle at turning base and final. Moreover, PANS-OPS are supposed to be based on 18-20deg bank angles which does not seem to be applicable in this case (which I suppose would be much better for the pilot and the passengers).
2. The descend rate is also quite steep, more than expected standards I suppose (less than 1000fpm). At this descend rate, PAPI seem unusable for the whole finals, so it will be a pure visual steep approach. This seems to be a considerable additional burden to the pilot in an already difficult procedure. And as far as I understand, descending before turning base is not recommended with MOC at just 295ft.
3. Even if the 40sec/1000ft rule is used for the timed section, descend gets somewhat better (866fpm) but not much and still PAPI seem useless. So, what happens in reality? An even steeper initial descend, so PAPI's will be captured at final?
4. In many relevant approach plates, vis minimum is something like 2.4km while the max distance from the threshold will be 2.0-2.6nm (apogee of the base turn), ie. 3.8-4.8km. So, even if visibility is above minimums the runway will be out of sight and , I suppose that pilot has to abort landing. In such a case what is the meaning of such a minimum? Shouldn't be someting like 4-5km in order to be more realistic?

Summarising, is circling in real life so awkward as it seems in a spreadsheet? High bank angles and step descend rates needed? No PAPI's used?
Or I miss something altogether?

Rather long post, but the issue seems very interesting, so feedback would be much appreciated.

[radu2000]

The circling minima you present seems rather high (1500' AGL). Usually on the bottom right hand side of an approach chart you will see that circling minimas or MDA are a function of the approach speed for you type of aircraft, eg. LGTS for an approach speed of 135KTS the MDA is 1028' AGL. A circling height of 1500' usually applies to a visual landing maneuver.
I am not sure about the Airbus or ATR FCOMs but I can tell you that the B737 FCOM or FCTM does not specify timing for the break off or abeam threshold position. It does however specify the protected area in which the circling maneuver should take place (ICAO 4.2NM radius measured from each threshold and joined by a tangent). The most important thing is to stay in the protected area at all times where obstacle clearance is assured at the MDA.
You make a very interesting point about visibility. Again visibility on such a maneuver is a function of approach speed. The slower the speed the lower the visibility required. I guess that the main thing to keep in mind is that this is a visual maneuver and the runway environment and threshold should always be kept in sight and the maneuver should not be attempted at visibilities lower than prescribed. You are right though, the value seems a little low, I will ask my TRE next week as I am due for my LPC.
With respect to the visibility, you must also take into account the approach lighting system which can extend several hundred meters from the threshold and if visible does allow you to continue the approach to land.
I hope this helps.

[Dan Winterland]

This is a topic where there is no straight answer. You have to remeber that from the start of the circling approach, it is now a visual manouvre and the PANSOPS guidelines no longer apply. A's:

1. I go straight to 25 degrees at the start of the final turn and see how it goes. Wind will have an influence on the turn.

2. You shouldn't have to exceed 1000fpm. You will find most pilots try to position so that the last part of the approach, say from 500'will be on the 3 degress from the PAPIs. Not many circling approaches will be fown from 1500'. The minima will invariably be lower than this. (My company has a lowest circling minima of 1000') so most pilots will fly the first part of the procedure level, until the 3 degree path is intercepted.

3. The speeds flown depend on the type and weight of the aircraft. In the Airbus, we fly the F speed until in landing config, then it's VApp.

4. Your figures seem to be a bit high. I think you are assuming you fly the procedure at 1500' whereas in reality, it will be flown lower. If the minima was 1500', the the vis requirement will be in the order of 5kms.



But as the first respondant hinted, it's not designed to be flown like a precision approach. The most common circling approach I fly for real is at an airifeld in Korea where TERPS apply and we have to stay within 2.4NM of the runway. The downwing section is flown at 1100' and there's alway a crosswind blowing you into the runway. If you don't go straight to 30 degress and 1000fpm, you're not going to get in! We have to have our wings level by 300' on that approach and that's hard to acheive.

[Capt Chambo]

The most common circling approach I fly for real is at an airifeld in Korea where TERPS apply and we have to stay within 2.4NM of the runway. The downwing section is flown at 1100' and there's alway a crosswind blowing you into the runway. If you don't go straight to 30 degress and 1000fpm, you're not going to get in! We have to have our wings level by 300' on that approach and that's hard to acheive.

RKPK, Busan you just gotta love the place:}

[mitsupini]

Thnx to all for answers and comments.
What I had in mind, is something like circling for rwy14 in LGIO (min 1443/2.4 for catB, 1942/4.8 for catC/D).
My basic assumptions assumed catB aircraft like ATR72.

I understand that timing from brake to downwind (assuming an intial straight-in approach to the opposite rwy), timing from abeam threshold to baseturn and speed management all along the circling manoeuvre are crucial. So, it seems to me like careful preplanning for circling is a must when this is a possibility at the arriving airport.

Dan,
I see that TERPS is a much tighter exercise. As far as I can tell, you fly a classD aircraft so it is even tighter.
What puzzles me, is that TERPS design is supposed to be based at 20 deg bank angle (for catD), when this seems to be far from true, as your real life experience also verifies.
BTW, I noticed that while circling min is 1040(1027) for RKPK (circl. classD), the "general" page of the map has a traffic note "maintain 1500ft on downwind". What figure is to be used when cinflicting info appears?

Could you please describe a real-life circling manoeuvre, with actual speeds and timings (eg, brake 45 deg for xx seconds at xxx knots, downwind at xxx knots etc) ?

[Capt Chambo]

BTW, I noticed that while circling min is 1040(1027) for RKPK (circl. classD), the "general" page of the map has a traffic note "maintain 1500ft on downwind". What figure is to be used when cinflicting info appears?

Not wishing to stand on Dan's toes, but...

RKPK R/W18R, is now a "Chartered Circle-to-land" in my company. That is we fly the approach on a prescribed track from the FMS.

Approach will clear you to descend via the ILS to 1500' followed by a circle to land. If the weather is particularly bad then they will clear you to 1100'. There is a hill on the downwind leg that is 1113', so accurate tracking is required.

Th approach is based on TERPS as you say, and we try to fly at a speed below 165Kts. At a ground speed of 160Kts an average bank angle of 21 Degrees is required to fly the approach. However as DW said there is often a very strong tailwind on base leg so it is usually best to use 30 degrees (or more) and 1000fpm RoD to get in. Wings will usually be level and you will be aligned with the runway at about 450' AGL.

The "maintain 1500ft on downwind" note seems to have disappeared, from my Jepps. although I remember it was there! In any case we descend to the level cleared by ATC. The 1500' on the downwind is because there is a lot of military training going on there. If the weather is poor then the military suspend training and ATC may clear you to 1100'

[mitsupini]

On this approach, what would be the lateral distance from runway when downwind?

[Capt Chambo]

On this approach, what would be the lateral distance from runway when downwind?

Transposing diagrams I have as part of our brief onto Google Earth, it works out at about 2nm.

[mitsupini]

ok, this brings my head back on my shoulders :ok:
Using a lateral distance of 2nm and 160knots speed, the math confirms a required bank angle of 20deg (theoretical - no wind correction), as you suggested in previous post.
If circling height is 1100ft, the required descend would be 740fpm, if at 1500ft then 1010fpm (assumed continuous descend from baseturn to touchdown).

Taking into account Dan's earlier note, I presume that when at 1100ft, you begin descend and then level off at some point to follow PAPI down to td.
While on 1500ft, you start descend at a higher than 1010fpm rate until intercepting the PAPI's.
Did I get this right this time:8?

[Dan Winterland]

Wot Capt Chambo said.

We too have FMS waypoints loaded to help us fly the approach, but as far as we're concerned, they are the limits of out manoevering area and we don't go outside them - with the caveat we also don't overfly the highway.

It's hard to explain what ROD's we use. I just fly the thing and if you asked me exactly what I did immediately afterwards, I probably couldn't tell you as I was too busy flying the approach to notice such details! But in answer to your last Q - yes. It's a bit of a dirty dive at first with the aim of getting ''in the groove'' at about 400', 300' at the very latest.

[mitsupini]

ok gents, thanks both very much; I think I now have a much clearer view of the real thing.

[Capt Chambo]

While on 1500ft, you start descend at a higher than 1010fpm rate until intercepting the PAPI's.

We have a vertical profile available to us (Vertical deviation on the ND), basically we start descent from 1500' just after passing abeam the R/W threshold, and 850-900fpm all the way down seems to work a treat.

If we circle at 1100' we start the descent a little later, starting the base turn IIRC, and again about 850-900fpm works fine.

[lederhosen]

Is this the place the 767 crashed going too far downwind some time ago?

[Capt Chambo]

Is this the place the 767 crashed going too far downwind some time ago?

Yes. Do a search for Air China flight CA129 of April 15th 2002.

[mitsupini]

According to the regulations in the general case, at which point an aircraft is permitted to descend below MDA during a circling appoach?

[Dan Winterland]

When visual.


The CA 767 crew failed to appreciate the difference between TERPS and PANSOPS circling minima and ran into the hill.