Tuesday, March 04, 2008

Crosswind Landing Decisions

I haven't posted for a few days because I've been enjoying the discussion about this wingtip strike that has cropped up in the comments to my last blog post.

In the clip, you can see that the airplane is approaching the runway in a strong crosswind, strong enough that the pilots have had to turn the longitudinal axis of the airplane to the right, to a significant angle with respect to the runway, in order to keep the path of travel aligned with the runway centreline. The airplane is travelling through the air in the direction that it is pointing, but the air is moving from right to left such that compared to the ground, the airplane is not moving to the right. Depending on factors like the position of the engines and the construction of the landing gear, the pilot usually needs to straighten the airplane out to some extent before touchdown, so that the wheels line up with the direction of travel. This should be done with rudder, because doing it by banking has two major problems: banking while very low to the runway risks striking the wingtip or engine pod, and banking away from the wind gives the wind a chance to pick up the into-wind wing as the wheels touch down.

As you straighten the airplane out, the wind continues to push it sideways, making it want to drift with the wind, in this case to the left. And the act of straightening it out does two things to make the airplane want to roll away from the oncoming wind. First, as you yaw the airplane with the rudder, you're turning it about the vertical axis, making the into-wind wing go faster. And second, as you turn the airplane away from the wind, the fuselage shields the downwind wing from gusts. Lift is generated by air flowing over the wing, so the into-wind wing gets more lift, making the airplane want to roll away from the wind.

When I straighten out an airplane in the flare for a crosswind landing, I put the into-wind wing down, and then use the rudder to keep myself straight. The airplanes that I have been certified on have had either high wings, or significant enough dihedral (wings that slope up from root to tip) that I haven't had to worry about striking a wingtip or pod, but I know this is a concern in some types. I need to know that as I touch down I will have enough rudder authority to keep the airplane straight while maintaining the wing down no more than it needs to be. And I need to follow through after touchdown, turning the ailerons such to kill the lift on the into-wind wing and make sure the airplane stays down.

The pilot needs to decide on final approach whether she will be able to maintain control of the airplane, despite the wind, as it slows in the flare. If the approach is stabilized, such that the rate of descent is constant and the speed is on schedule, the pilot can judge the consistency and strength of the gusts to make the decision. Usually there are a few things to guide the pilot in making the decision: the manufacturer's maximum demonstrated crosswind performance, the company's crosswind restrictions, and the pilots personal limits. While there is no law prohibiting a pilot from landing an airplane in harsher crosswinds than the manufacturer's test pilot demonstrated, usually company limits don't exceed the manufacturer's. A pilot or the company may set an individual limit lower than the company-wide one. This helps to counteract pilot machismo. The pilot can say, "oh I could have landed in this, but it's over the limits, so I won't." Company SOPs will also specify that if an approach is not stabilized, the pilots must go around.

This is all fun and challenging, and I imagine the co-pilot (yes, so soon after my entry on pilots and co-pilots, it was the co-pilot who was flying) had a big grin on his (or, if Syrad's information is correct, her) face right up until the point when it all went quite literally sideways. A lot of the discussion on this has centred around the fact that both the media and the company are praising the pilots for a job well done. I doubt the pilots see it quite that way.

I once watched an extremely similar thing happen at a company where I worked. There was a nasty rainstorm and the winds were so strong they had knocked out power at the terminal. I had loaned my handheld radio to the dispatcher so he could continue his duties, and was waiting for this last flight of the day to come in so I could take it the radio home with me. I remember standing in the dark terminal looking through the glass door. Beside me was another pilot, and the wife of the arriving captain. The airplane came into view on what should have been an into wind approach. But strong winds are rarely steady winds and it is especially common for winds to change abruptly in strength and direction very near the ground. There was a sharp yaw and one wing went down, the tip struck, caught, and for a moment I thought the airplane was going to flip over. So did a jumpseating captain on board, as he told us later.

The airplane didn't go over its centre of gravity, so the crew were able to right it, and taxi in. There were no injuries and only wingtip damage, not even enough to constitute an accident. The passengers, including company, were all somewhere between white and green as they disembarked.

No one was filming it, and it wasn't a jet. so it didn't make international news, but I remember the captain beating himself up about it. You don't blame a gust, except in jest, and you know that when you survive something like this with minor wingtip damage, you were lucky, not good.

16 comments:

syrad said...

I was just watching the youtube video, and I have to agree with aviatrix's analysis. I didn't see any aileron input on this landing attempt. Instead of the left wing coming up as it should have, the right one did (most likely as a result of the increased airflow during the yaw as aviatrix said). I'm sure that a strong gust hit just then, because of the speed at which the aircraft floated sideways. However, with the proper correction the aircraft would most likely have had a safe landing, if not a smooth one.

One unconfirmed comment I've read says that the pilot flying, the first officer, was female. Naturally, the "women drivers" jokes weren't far behind. It's always irritating that when a female pilot messes up people start deriding all of us as a group. I don't know about other female pilots, but that's always in the back of my mind when I'm being evaluated. I don't think being female has any impact on my performance as a pilot, but I have this nagging worry that if I mess up the examiner/checkairman/sim partner/coworker will see it as a mark against all female pilots and not just against me. I'm definitely being unfair to all my male colleagues when I think this way, since I'm positive most of them have flown with enough women to realize that we're as diverse as men when it comes to flying skills. Besides, the vast majority of the men I've flown with don't appear to think twice about working with a female pilot. Unfortunately, the general public does still take notice.

Christopher said...

Here are a couple of links with photos of the "touchdown" and the wingtip damage:

Airliners.net

Hamburg-Airport-Friends-Forum

Also, this was apparently the reported weather sequence at the time:

EDDH 011220Z 29028G48KT 9000 -SHRA FEW011 BKN014 07/05 Q0984 TEMPO 29035G55KT 4000 SHRA BKN008

CanuckPilot said...

Although it may be legal to continue an approach with crosswind in excess of the demonstrated maximum if your COM doesn't have a limitation. I doubt it is a good idea in a big transport category airplane and is certainly beyond my personal limits with passengers on board.
From my armchair perspective although there was a huge gust as the plane entered the flare (watch the increasing crab angle close to the ground) the crab angle through the entire approach and the wind recorded during the approach should have been aborted.
If there aren't any pax I quite enjoy the challenge presented by landing a high wing small plane in a strong crosswind and nailing the landing. But only if everyone on board is a current pilot and knows what we are doing.

Aluwings said...

Aviatrix wrote: "...When I straighten out an airplane in the flare for a crosswind landing, I put the into-wind wing down, and then use the rudder to keep myself straight. ..."

Most jet transport drivers use a similar technique most of the time too - with some concern for wingtips and engine pods, as you mentioned. fwiw...

However some airlines do teach (and the manufacturer accepts) that the aircraft can touchdown without "kicking off" the crab angle. Check out some of the B777 crosswind landing test videos on the net where the Boeing pilots demonstrate it.

This technique has the undesirable side effect of adding yet another force lifting that upwind wing.

And to make it more complicated, if the runway is wet and so the wheel spin up signal doesn't happen right away, the wing spoiler panels may not deploy - making the airplane more vulnerable to a skip/bounce... giving the crosswind even more opportunity to make things "interesting" ...

Watching that video again, I'd say all these things were going on at once: Landing with crab angle on; no right wing down input (or not much) during the impact; a pilot-induced skip; slow or no reaction to the rapid crosswind-induced roll to the left; ... yikes!!

And if the less-experienced FO was flying, I wonder what sort of "I have control" things were going on at the same time...?

Happily they lived to be (a little?) embarrassed by the whole fiasco. I'm so glad none of my screwups have appeared on Youtube! (yet...)

Aviatrix said...

Some airlines do teach (and the manufacturer accepts) that the aircraft can touchdown without "kicking off" the crab angle. Check out some of the B777 crosswind landing test videos on the net where the Boeing pilots demonstrate it.

Yeah, if I did that in what I fly now I'd be afraid that the gear would collapse or the tire roll off the rim, but I know that B747s do it, because the wing down position causes a podstrike for the outboard engine. That's what I was getting at with "Depending on factors like the position of the engines and the construction of the landing gear."

A Squared said...

"Although it may be legal to continue an approach with crosswind in excess of the demonstrated maximum if your COM doesn't have a limitation. I doubt it is a good idea in a big transport category airplane"

Disagree. I do it frequently. We have two destinations at which it is not uncommon for the winds to be directly across the only runway we are allowed to use, in excess of the maximum demonstrated crosswind componnent of my airpalne. I know from repeated experience that both the airplane and I am quite capable of landing in such crosswinds. It's neither poor judgement nor machismo, it's just the environment in which I operate.

Ed said...

Syrad: http://xkcd.com/385/

"First, as you yaw the airplane with the rudder, you're turning it about the vertical axis, making the into-wind wing go faster. And second, as you turn the airplane away from the wind, the fuselage shields the downwind wing from gusts."

Isn't there a third effect on swept wing aircraft in that, as the fuselage is kicked straight, the into-wind wing is turned more perpendicular to the airflow whereas the downwind wing is turned nearer to parallel to the airflow? AIUI, swept wings increase the roll-yaw coupling because of this effect, which is why yaw dampers are required.

Anonymous said...

What do you do if crosswinds are too strong at the destination and surrounding fields? Or is there always so much variation (and reserve fuel) that you can find some place to land?

Paul said...

Yea, but if you screw-up with winds over the max x-wind component, you're gonna meet some nice folks from the FAA (CAA?).

If you do that in a part 135 or 121 operation, I would think it would be deserved.

I was always taught the POH (for small planes) or your SOPs is not a set of recommendations.

--paul

A Squared said...

Yea, but if you screw-up with winds over the max x-wind component, you're gonna meet some nice folks from the FAA (CAA?).

If you screw up, you're going to meet some folks from the FAA. Regardless of crosswind componnent.

This isn't a case of forging into uncharted territory, wondering all the while if the end of the earth will appear and you'll drop off into bottomless space, metaphorically speaking. Nor is it exceptional heroics. It's simply that a proficient pilot with decent stickhandling skills and an understanding of how to correctly land in a crosswind can easily land this particular airplane under control in a crosswind well above the maximum crosswind which was demonstrated during the certification of the aircrft.


I was always taught the POH (for small planes) or your SOPs is not a set of recommendations.

It is neither agasinst the company operations manual nor contrary to the aircraft flight manual. There is a reason that it is worded the way it is; "maximum demonstrated crosswind componnent" That is to make it clear that it is *not* intended to be a limitation. If it was a limitation, it would be worded as such.

As a point of fact, the only portion of a POH which *is* limitations is the "limitations" section. If your POH instructs you to use one stroke of the primer before starting, and through experience in cold conditions, you know that it will start much more readily with 2 strokes on a cold morning, are you guilty of a violation of the regulations if you use 2 strokes of prime to start your airplane?

Lord Hutton said...

I can just see the co pilot saying "Fuck that! Let's go somewhere else!"

Paul said...

Thanks A**2. I'm going back over my POH again...

--paul

Jim Howard said...

Here is a portion of a post on flightinfo.com (paid subscription). The author is an airline pilot qualified in both B737 and A320 aircraft:


Instead of the 1.3 Vso = Ref that everybody else uses the bus uses 1.23 Vso+5kts or a wind correction component, whichever is higher.

On top of that a system called ground speed mini maintains a constant groundspeed to keep the approach speed/aircraft inertia constant regardless of windspeed and gusts. (GS MINI is installed in different configurations on 319/320/321 aircraft and can differ between like types as well - I have seen two types on 319s)

The GS MINI provides no protection on direct cross winds and quartering tailwinds. In direct crosswind and quartering tailwind situations the aircraft will be at 1.23 Vso+5 for target approach speed.

The 319/320 does not have adequate roll control in many cross wind and gust conditions. Even if a pilot utilizes the recommended flap setting, (CONF3 slat 22/flap 20 - it looks like they were using CONF Full, slat 27/ flap 40) and manually adjusts the target speed up 15kts it is possible to encounter rolling moments that continue to trend opposite of full sidestick deflection.

Though it is in no manual, most pilots, who have been on the plane long enough to encounter the situation, know that to arrest the roll it is necessary to hold full sidestick and to apply upwing rudder to regain roll control.


Perhaps memories of the American Airlines Airbus that lost its rudder kept the Lufthansa pilot from aggressively using the rudder.

50N30W said...

Jim,

I think the Bus control peculiarites below 100 AGL have much more to do with this than anything else. Good post.

Not rated on any Airbii, but anecdotal evidence and crew lounge griping indicate the A319/20 can be a real handful in agusting wind. FWIW.

Kevin said...

I have a question. I transitioned to multi-engine last year, and I find that, in crosswinds (and yes, I have landed in crosswinds greater than the "demonstrated maximum") I can get a lot of help from using asymmetric thrust.

In this technique, you establish the crab on final. You then increase power on the upwind engine so as to straighten the fuselage with respect to the extended centerline. Then, you use rudder, wing down on the upwind side and the usual techniques to arrive safely at the runway.

Of course, when you retard the throttles in the flare and to landing, you do it symmetrically, compensating for the upwind engine being at a higher power setting. On my most recent foray into this kind of crosswind, I had about 17" on the left engine and 25" on the right.

I find the benefits of this technique include the following:

- Passengers are much less likely to freak out, or to realize that the landing conditions are more challenging, because the runway is visible out the front, not out the side

- Using the engines to balance the fuselage gives you more articulation on the rudder, giving you an additional margin of safety should you need more rudder deflection

My question is - is there some reason why you can't use this technique on Airbii and other transport category aircraft with the engines mounted on the wings? I understand that fuselage-mounted engines wouldn't give you the same moment of yaw - but surely increasing the thrust on the right engine of that Airbus in the video would have resulted in a more stabilized approach and less of a goat rodeo?

majroj said...

Speaking as one of the "beasts in the back", I too prefer not to see the approaching runway out my window, but in front of and soon underneath us.

PS: living ON the flightline in Nebraska, on a quiet hot spring Sunday afternoon, we would watch desk officers keeping their "heavy" skills (and flight pay) by touch-and-goes with the local versions of the C-135 (aka Boeing 707), the uneven thermals and breeze gusts causing them to touch down one wheel at a time and occasionally not ever planting all three tires down solid simultaneously before ignominously pouring on the JP-4 and circliong around. I kept waiting for one to try too long.
Of course, a Naval or MArine aviator would never have such a problem...BAM every time.