View Full Version : Missed Approaches beyond the Missed Approach Point.

24th May 2004, 04:21
This thread is a continuation of a post I made in the thread relating to "What is the real minima?".


This discussion reminds me of an issue that I fear is mostly overlooked regarding MDA's, DA's and missed approaches.

Here we are talking about how much of a big deal it is whether or not the momentum of your aircraft will take you below the MDA if you initiate a go-around at rather than above the published MDA.

That's fine. But who here has thought about this. You arrive at the MDA. You're "Visual", so you continue the approach below MDA toward the landing runway (I'm thinking of a runway aligned approach here). Then, for whatever reason, you decide to discontinue the approach, you press TOGA and head skyward again.

Reasons for the late Go-Around / Aborted Landing?

1) Becoming visual but subsequently losing visual reference again due to cloud drifting across the flight path.

2) Flying into ever increasing precipitation that makes it increasingly difficult to see through the windscreen.

3) Windshear and/or turbulence. (Not necesarily asociated with microburst/thunderstorms).

4) Previous landing aircraft too slow vacating the runway.

5) Previous departing aircraft Rejecting Take-Off and remaining on runway.

6) Animals on runway.

7) Crosswind component above Operations Manual "limits"

So here you are, going around not that far off the deck. Makes the previous concerns about MDA vs RA rather academic! You have now possibly progressed 1 or 2 nm beyond the Missed Approach Point, at an altitude 100 ft, 200 ft, 500 ft below MDA. What the @#$%$ are you going to do now?!?! Have you even thought about what you'd do if or WHEN this happens to you? :uhoh:


Just an observation. Once you have safely landed on the runway, There's a good chance that you (or at least your aircraft with another crew) are going to be taking-off from that same runway. It'll be starting out at ground level, at zero knots, possibly at MTOW! Think how it's curent potential and kinetic enery levels (zero) compares to the aircraft deciding whether or not it's ok to descend below MDA [140 kts at 500 ft AGL (for example)]. :rolleyes:

So in one way it is a big deal that you are initiating a go-around so late, and in another way it isn't because you most likely going to be taking-off from that same runway shortly afterwards.

And there I think might be the answer.

If you have to go around your choises are either:

1. Climb to the Circling Minima and join a visual circuit while remaining inside the Circling Area, or

2. Climb back in to the cloud and not level off until you reach a safe altitude.

Now if you have to take option 2. then you have to think about the path along the ground that will afford you the greatest terrain clearance. For my thinking, the published missed approach tracking is no longer valid because you are well and truely below it's designed climb gradient requirements.

I would be thinking more along the lines of where I 'd be tracking during a take-off. And why not?! You're lined up with the runway. The authorities have surveyed the area along the extended centreline up to 15,000 metres (in Australia in any case for Catagory 4 runways).

So if it is OK to take-off straight ahead, surely straight ahead is your best way out of your predicament.

If your take-off procedures for a particular runway require a turn in the case of a normal all-engine take-off, or if in the case of an engine failure on take-off requires specific alternative tracking, I would suggest you follow that procedure until you reach a safe altitude.

Also keep in mind the minimum accelleration altitude used for take-offs from that runway. Might be more than 1000 ft AGL or 1000 ft above MDA.


24th May 2004, 11:43

At least one operator of large aircraft with whom I am familiar specifically deals with the scenario you are pondering during every proficiency check and IRT.

First, a word of warning about relying too heavily on the take-off path planning or the straight ahead option.

Your normal large aircraft take-off planning in Oz is based on CAO 20.7.1B and commences at the departure end of the clearway with a total width of 500' or 152.4m expanding at 12.5% or 7.125 deg. That is not much more than a couple of wingspans in your 737-800 and given that the centre point of that splay is often 1-1.5 nm from your go-round point and is under the nose completely outside your field of view even in an OEI missed approach (during which I suspect you will predominantly be on the clocks maximising performance and configuration management), then there is a high probability that you will not be within the containment area. The normal survey is wider at 180m, but that is only an extra half a wingspan or so.

A better consideration would be to fly the missed approach flight path for a relevant runway aligned approach if available, simply because you are normally already contained within the lateral confines of the design area. How far you are below the relevant minima is then your major problem.

What are the vertical options?

Generally, but not always, your OEI approach climb gradient capability will exceed the take-off surface survey gradient at normal landing weights. However, it is worth consulting your AFM to get a handle on your aircraft capacity in various environmental conditions.

At Canberra for example, on runway 35 you would need a missed approach gradient capability of 2.5% from the ILS missed approach point and the DA minus sink-through allowance (2340-150=2190 or 320' AGL) or 3.2% from about 180' AGL or, assuming for simplicity a gradient capability increment of 0.1% per 20' above threshold elevation, about 4.1% from the flare.

These are approximations, since to do this analysis properly, you actually need to analyse the obstacles within the missed approach splay (a copy of which, by the way, the money-grubbing ar$$holes in AirNoServices will sell to you for $1500 per approach design!!). The above figures compare with about 3.9% if you could stay within the take-off splay, which excludes some significant terrain that is included in the missed approach splay. All this is approximated without provision for an acceleration altitude or consideration of go-round thrust time limits to reach a suitable MSA! However, it is a much better deal than the more generic cases described below.

What do we need to achieve if we want to abandon the missed approach at the circling minima and stay within the circling area?

Well, given that this is often the most commonly briefed plan (and performance delusion!), your climb performance needs to be good enough for you to achieve circling minima before the relevant circling distance minus your turn radius for the relevant speed so that you can remain within the containment area.

Continuing our Canberra 35 example, for Cat C you need to reach about 1600' AGL (for an OCH of 400') by about 3.5 nm from the runway departure end, a gradient of about 7.5% for the take-off case or about 5.3% from the flare (4.7% @ 180' AGL and 4.2% @ 320' AGL) at the landing threshold, both of which are fairly severe requirements, even if it were possible to satisy the circling manoeuvre requirements (which I totally doubt!!).

If you can't make those gradients, then you are going to end up having to consider the next safe height which is 10 mile MSA of about 3200' AGL in the general case but only 2600' AGL in the NE sector. The OCH is 1000' in both cases. In any event, if we arrive at, say for Cat C, 4.2nm from the departure end at 1600' AGL (the circling minima) we have no terrain clearance for the NE sector and minus 1600' for the general MSA. The gradient to get to that highly undesirable situation is about 6.3% for the take-off case or about 4.6% from the flare at the landing threshold. Even our previous gradients of 7.5%-5.3% for the circling provide us only with 240' of terrain clearance for the sector and 1360' underground for the general MSA.

If we want to get at least 400' between us and the ground, we need 7.8%-5.8% for the NE sector or 10.2%-7.5% for the general MSA. If we want the full Monty, then the required gradients are 10.2%-7.5% for the NE sector or 12.5%-9.2% for the general MSA.

Fortunately, most of our destinations are much less severe so the requirements are less limiting. My main aim was merely to highlight the need for good understanding of what you face, so that the resulting plan offers the greatest risk reduction.

Oh, and for those of you that like to forget that all missed approach procedures are three-dimensional, there is a sound reason for commencing missed approaches at or before the missed approach point if you wish to avoid some very complex mathematics and shetloads of uncertainty.

Stay Alive,

27th May 2004, 07:58
Our SOPīs state: When performing a Go-Around below minimums(DA, MDA) follow a relevant SID or engine-out SID. Your performance along an SID during a go-around is far better than during a T/O so no problems with performance. But continuing along the missed-approach procedure will NOT give you clearance to terrain or obstacles.

27th May 2004, 14:54
On any go-around or missed approach, whether it's initiated at or after MDA, D/H, MAP, or in the case of a rejected landing, even after the main wheels have touched the pavement,...we fly the full missed approach profile as published,...unless radar vectored otherwise. We don't: "1. Climb to the Circling Minima and join a visual circuit while remaining inside the Circling Area; or 2. climb back into the clouds and don't level off until you reach a safe altitude..."

30th May 2004, 06:01
I did some sums based on an aerodrome with the following specifications :-

MSA : 3000 Ft
Landing THR Elev : 98 Ft
DER THR Elev : 101 Ft
RWY Length : 2 NM, including stopway/clearway.
There is a SID for departure from this RWY.
There is also a non-precision approach with MDA 700 feet, aligned with the runway and based on a VOR situated 1 NM prior to the landing THR. Missed approach point at the VOR.

I then assumed that a missed approach would be initiated 2 NM beyond the missed approach point at 500 Ft AMSL.

To climb away safely in this situation, a minimum nett gradient of 2.85% will keep you above the climb plane for the SID. Note that this gradient is based on no delay between decision to go, application of thrust and aircraft climb. Taking the delays associated with those factors into account, the actual climb gradient required will be considerably greater than that.

But, in any event, even at 2.85%, you're aircraft must be able to climb more steeply than the missed approach (2.5%) gradient. So, if you have a performance limitation imposed by, say, an engine inoperative, you must be sure that your residual climb performance is much better than 2.5%.

This will probably be the case with FAR 25 aircraft, especially if the landing weight has been significantly reduced by fuel burnoff enroute. But, as with all such things, it's a really good idea to consult the AFM to determine the climb performance that you can expect with OEI at the appropriate weight, with the existing environmental conditions.

Thus, this comes under the heading of "Contingency Procedures" or "Emergency Procedures", depending on which term your company uses. That means that they need to work out such a procedure for themselves, or by hiring a performance engineer to work it out for them. This is not the sort of thing that you can, or should, try to work out on the fly, at the time when you are in such a serious situation.