FMC holding patterns

[Kilda Ste Hilda]

Why doesn't the FMC outbound leg correct for crosswind drift of 3 x drift? It seems to correct enough for drift to maintain a perfect race track pattern. Just curious:O

[rudderrudderrat]

Hi Kilda,

Good question.

I think it's because the FMC doesn't "remember" what the drift inbound was. It simply measures the ground speed on the outbound leg, calculates a radius of turn based on that speed, and then aims for a point which permits it to complete a turn to establish inbound.

PS What's the volcano doing now?

[eckhard]

I agree; good question.

The FMC doesn't have to allow 3x drift because it flies constant radius turns at each end of the pattern by varying the bank angle to allow for different groundspeeds due to wind.

When you fly a raw data hold, you have no idea what the groundspeed is doing as you go around the turn, so you fly a constant bank angle. This results in turns of different radius at each end of the pattern (assuming a crosswind component). The only way to compensate for this is to exaggerate the drift effect on the outbound leg.

The FMC flies perfect constant radius turns, so the outbound leg can be flown parallel to the inbound leg.

Hope this helps

Eck

[BOAC]

As eckhard above - the FMC simply follows a ground track - the track of a still air hold. It makes NO allowance for wind effect in the turns other than bank angle variations. IF you ever fly an LNAV hold with an (Icelandic:)) severe wind across you will see 'fly-through' at one end because the bank angle limitation will not hack the turn.

[aterpster]

BOAC:

As eckhard above - the FMC simply follows a ground track - the track of a still air hold. It makes NO allowance for wind effect in the turns other than bank angle variations. IF you ever fly an LNAV hold with an (Icelandichttp://images.ibsrv.net/ibsrv/res/src:www.pprune.org/get/images/smilies/smile.gif) severe wind across you will see 'fly-through' at one end because the bank angle limitation will not hack the turn.

Depends upon the sophistication of the FMS. The current specifications call for calculating the pattern to be flown predicated on the known wind direction and velocity at pattern entry. In strong wind conditions the pattern projected on the moving map should look like an elongated, misshaped egg.

[BOAC]

Good stuff - and sensible. I've yet to see that!.Is it incorporated anywhere?

[aterpster]

BOAC:

Good stuff - and sensible. I've yet to see that!.Is it incorporated anywhere?

It has been implemented in high-end business aircraft with Honeywell avionics for several years now. Two of which I am aware are the Falcon 900EX and Gulfstream G550. I presume it is also implemented in late generation 737s and I would believe the 777 as well.

I know that Garmin has it in their TSO-C145/146 panel mount navigators as well as the G-1000. Below are two screen captures from the Cessna Mustang G-1000 trainer.

Parameters: easterly wind aloft at 85 knots. IAS = 230 knots at 10,300 feet, msl.

The aircraft is setting up for an RNAV IAP to Butte, Montana. In the upper screen shot the course-reversal holding pattern is at the top of the screen and shows as it does on the approach chart. As the aircraft crosses the CPN feeder fix, the holding fix becomes the active waypoint, at which point wind and ground speed are integrated into the holding pattern, as shown in the lower screen shot:

http://tinyurl.com/26g2n7v

[JammedStab]

I'm not 100% sure but one time when doing a hold with a Universal UNS-1F with a very strong crosswind, turned to an outbound heading that was almost directly into the wind and I wondered if it was screwing up the hold. It then gradually came around toward a more realistic outbound heading and then had it aced when turing inbound. I can't remember what the heading was(if it was parallel) as it was a long time ago but it was impressive.

[aterpster]

BOAC:

Good stuff - and sensible. I've yet to see that!.Is it incorporated anywhere?

This got some old memories to surface. The first-generation 767 I flew (1984-86 pre-Pegasus FMS) did positive course guidance holding patterns. I can only recall one hold, because we were there for quite some time. It must have corrected for wind, but I don't recall much about it.

[Kilda Ste Hilda]

Thanks guys, aware of all that but like BOAC said I have noticed that the FMC couldn't hack it with stiff crosswinds. Also with the varying bank angles to compensate for drift, what are the chances of it busting the safe holding area limits. I have noticed that the FMC sometimes do make some very untidy entries into the hold and turns onto the inbound legs.

The volcano is outwardly calm but vulcanologists think that magma below has still a good chance to spew upwards when tectonics favour an upsurge through the crust.

[Rotorgoat8]

Just a note of caution here--I've seen the '57 roll into some aggressive (up to 45*) banks to stay on the ground track in a hold. WATCH your airspeed as the AT's struggle to keep up in this situation.

[tmpffisch]

Depending where you live the rules on holding patterns are different.

Here in Australia the AIP says in flying outbound, you need to parallel your inbound track, so 3 times drift isn't acceptable.

[aterpster]

rotorgoat8:

Just a note of caution here--I've seen the '57 roll into some aggressive (up to 45*) banks to stay on the ground track in a hold. WATCH your airspeed as the AT's struggle to keep up in this situation.

That sounds like a malfunction. The design specs call for a 30-degree bank limit (or 25 degrees with some airframes) for autoflight. The nav logic is supposed to be to accept the deviation from track at 30 degrees of bank and keep changing the course to get back on track.