Wind Check on final
 

Hello all,

Let's say your MAX TAIL WIND LIMIT IS 15Kts. (B777), at 500ft, you request wind check and tower reports 14Kts, you can officially land, as the only official wind is the tower reported wind.
But, your FMC PROGRESS 2/4, reads a tail wind above 15Kts, you continue, and land. This time, you have been unlucky and skidded off the RWY.
How would you be judged by your company?

The bellow was extracted from an overrun incident investigation:

FDR data was found to indicate that "the average tail wind component during the 15 seconds prior to initial touchdown was 12.95 knots" - in excess of the AFM maximum 10 knot tailwind component for landing. It was noted the same information was available on the MFD, but that Operator SOP for landing and takeoff limitation purposes, was to rely on wind velocity communicated on ATIS or direct from ATC.

The crew followed the SOP but was unfortunate.

Question:1. Does your company recommend or suggests any other wind information for landing that is not the REPORTED TOWER WIND?

2. Do you base your land/go-around decision on the FMC wind information?

Happy flights!

1. no
2. Whenever I'm sure that landing distance will not be a problem, I base my decision on the tower reported wind only.
If there's any doubt, I'll use the highest reading.

Surface winds are picked up about 10 meters above runway...this wind is essential,not winds at 500 ft.
You enter the flare around 30 or 500ft?
Do you look at the FMC at 30'feet?
The tower reported winds would be used in a court of law id think.
If you land with only a few knots to spare,id suggest to reconsider.

Since when winds at heights above your flare height have anything to do with your landing distance?
If atc is untrustworthy,PM at flare initition (50'feet for general landing calculations)may call out TW component.

I would suggest that a 12.95kts tailwind, instead of 10kts, was not the cause of an overrun. I suspect there were other parameters at work.
Most AFM's have -10kts as standard, but airlines can research individual runways and apply -15kts. So 3 kts does not cause an overrun.
Every airline I've flown for uses ATC wind. However, it was not uncommon, years passed, for certain Spanish airfields to apply 'calm' when the convenient duty runway was decided. It was calm on landing but -7kts on takeoff.

Continuing, a little in the same vane, I ask for opinions. There are calculations for the adjustment of Vfly for headwind and gusts. There are calculations for the crosswinds including gusts to be compared to the legal limits. There are legal tailwind limits, but I've never read a calculation to include gusts for tailwinds. This aspect has safety implications for over runs. I diverted once for what I considered tailwind gusts to be outside limits. Afterwards I asked for guidance on the tailwind/gust limits. No reply, ever. The performance dept for RTOW's use 150% for takeoff. I asked what about landing? No reply, ever.
I find it strange that at a flight phase where legal limits are in force, and a phase which cause the most accidents, there is no clear answer.
Opinions?

Unfortunately, despite what many companies write in their manuals and what you all expect they will hang you on your EFIS wind

Warsaw - Luftansa A320 Accident

[/URL] [URL="http://www.rvs.uni-bielefeld.de/publications/Incidents/DOCS/ComAndRep/Warsaw/warsaw-report.html"]

There is a fantastic book which would shock many called "Air Law for Commanders and their Crews" - it's based on case law i.e. what will apply in court if you have an accident - unfortunately, long out of print.

There are plenty of places in this world where they prolong the obviously required runway change until it's too late and then they force the aircraft already on final approach to continue with landing by advising exactly 10 kts tailwind, despite it's more like 15 or more when you cross the threshold.

From what I know, only the ATC reported wind is valid. Of course, it's good airmanship to briefly look at the wind indication on ND close to ground (but not at 500ft of course) - and to act respectively if major discrepancy is noticed.

Unless you floated during the last 15 seconds before touchdown, there is no way a normal FDM program would even mention this unless tailwind component below 50ft would be significantly higher than AFM limit. The AFM limitations are for landing, which in this case would count from screen height to touchdown - and that doesn't take 15 seconds.

As some of you correctly mentioned, in the overrun incident I used as example, wind was just a factor, but not the cause of the overrun. There was a chain of events. But it was the initial trigger and the crew could have decided not to land.
I mentioned 500 ft for wind check, but could be any altitude. About looking down at progress 2/4 near flare, I agree it's not where our eyes should be, but once you figured the component/ direction from progress, you can refer to the wind displayed at your ND, that is in your normal scan area.
The fact is, we are not told exactly what to do, or what information to use, and when having to decide for a landing or go-around you will end up deciding based on many factors, not only tower reported wind, but most importantly, how comfortable you feel. If you are comfortable, use lower number and land, if not, use the higher and fly away. In case nothing goes wrong, as in most times, you keep your job in either case, but if you get unlucky, then every small detail counts, and it may be against you.

Is the FMC wind an instantaneous calculation? Or is there a slight delay?

Doesn't make much difference in cruise, but on approach it may be showing what the wind was.

I am not sure about any delay, could not find the information in the books.
What is in the book, and many pilots overlook, is that the wind information in progress 2/4 (B777) is in reference to the airplane heading and not track, or runway. This is relevant as you may be crabbing into the wind. So when you de-crab, your crosswind component will increase. Ok, but who will look at that when de-crabbing......impossible in practice....good to know, but in practical terms, irrelevant!

Crosswind (X/WIND)
Displays left (L) or right (R) crosswind component relative to airplane heading.

The ATC wind is taken from a device that directly measures the wind velocity and designed specifically for that purpose. The FMC wind comes from a calculation involving ground speed, heading, track, and true air speed and is not a direct measurement of the wind. Further, two FMCs in the same aircraft may differ by a few knots in calculated wind. Given the above, I would take the reported wind as gospel and the FMC wind as of academic interest only provided it is relatively close to the reported wind. Additionally the flight manual for the type I'm familiar with states that winds are factored such that a tail wind accounts for not less than 150% of the tail wind, head wind accounts for not more than 50% of the head wind and "reported winds may therefore be used directly".

Processing occurs at both ends of the chain - FMC and Tower,

Mathematical algorithm called a Kalman filter is used and takes several seconds to produce the 'live' readout. A simplified definition of the algorithm would be the combination of information in the presence of uncertainty. It's particularly effective in resolving vector calculations but can take 7 or 8 seconds to produce the result.

You've probably been asked at some point for a 'spot wind' by ATC and the same request from your end can often produce an straight readout without any of the data massaging and inherent delays caused by the kalman filtering.

Rob

1. We use the reported wind from the tower. Then, if we consider that we can land safely, we do so.

2. No. That would be very unwise. The data is too untrustworthy to use. It also wasn't designed to be used for this purpose.

But isn't that report interesting. I'm surprised that it did not recommend taking the crew out and shooting them to prevent reoccurence. It's just a shame they didn't cover their own arses by telling us what the sampling rates and margins for error were in the FMS computed values. They also might have been generous and given us the effect of the unreported water patches. Nor did they tell you the effect of passing values in kph to crews used to operating in knots. etc...

And no, I'm not worried about losing my job. That is the action taken by naive, small minded, feeble, ineffectual managers who believe that removing rotten apples will sort problems like this out once and for all. My company are above that.

PM

It is cultural, but not written SOP, in my airline for the PM to read out wind components from the FMC - when relevant, or when requested by the PF.

Examples of when relevant could be:

- near performance limits
- near airframe limits
- changing rapidly
- expected to change rapidly

Is there a scripted mandatory go-around based on FMC wind only? No.

However, if I bent an aircraft and the FMC wind was outside limits approaching the flare, I would be seeking a good lawyer.

Interesting topic.

As noted by others, Winds from the tower are based on a 10m height above the ground, at one location at the airport. The AWOS winds are based on a 2 minute average, including gusts on a 10 minute average.
ATIS winds are updated every 10 minutes, using the same equipment, this is why one must use the tower reported winds for landing.
FMS winds are calculated using the pitot tubes (at least one set front/back) and the distance the INS says it has travelled. Winds are only 2d horizontal headwind and tailwind, as the system can only measure winds in the direction of travel. Depending on the system, the winds are updated every 4 to 5 seconds. Due to latency, it tells you the winds where you were...

When an aircraft is descending, the measurement being only 2D horizontal headwind or tailwind, doesnt really accurately represent the wind because of the 3D vertical component of descent.
Other issues such as turns, once again, winds may not be accurately represented.

Many FMS use a 'blended' wind calculation. The FMS allows for wind input, but only at 3 or 4 locations. There were the forecasted winds in the system from flight ops at departure, the real time winds, and the input winds. Some systems will only update if the winds are outside parameters set from the forecast.
Winds input will be updated on the specific location or waypoint only, but may be blended as an average.

Many of these issues are coming up as a result of the trajectory and time based flow management programs being researched.

I wonder about this. Airbus has the automated wind shear alert system, that senses winds, looking for the probability of wind shear. From what I understand, if this system alerts, it is a mandatory GA. I am not certain of the parameters, but from what I understood, it does look at headwind/tailwind, and horizontal/vertical displacement.
Perhaps A drivers can add something about this.

There are also systems that send an ACARS message with data like this for flightdeck use...
http://i60.tinypic.com/vcudd2.jpg

Does anyone use this type of information?

Same here. I think you’d be foolish not to take them into consideration.

Would I initiate a go-around based solely on FMS data? Yes, especially on a tighter/contaminated runway. A GA is generally a non-jeopardy manoeuvre - the same can’t be said of ending up on the stopway or grass, or setting the brakes alight.

I threw an approach away at 50’ in the Caribbean some time ago as it just didn’t seem right with a “calm” wind reported. FDR later showed 23kts tail!

This is an interesting question. I would add a couple more issues. My company has excessive tailwind on touchdown as a Foqua item. If you bust it by enough you will get a polite enquiry from our safety pilot, who by the way is a good guy. But you still need to explain yourself.

It is tricky providing evidence that the tower gave you a different wind then the one the plane recorded on landing. You can ask for the tapes but unless you crashed (in which case you have bigger problems) good luck getting these from some military tower in southern somewhere.

Also the issue of gusts and wind varying complicates matters. In theory the gusts may not be limiting. In practice a ten knot crosswind with twenty knot gusts that at the the moment of touchdown are mainly tailwind can be difficult to spot.

The simple rule if in doubt is to go around. The example from Full Wings demonstrates why this is a good idea. We do not have to explain ourselves if we go around, but we do if we bust a tailwind limit, which I think is the right way around.

If it's close to your limits just don't ask ;)

I threw an approach away at 50’ in the Caribbean some time ago as it just didn’t seem right with a “calm” wind reported. FDR later showed 23kts tail!

Excellent that some of us still have feel and the nouce to act on it; and the courage. I hope RHS learnt a very salient lesson.

However, to those who advocate a GA as first choice: there are many these days flying around on minimum fuel. So you make an unexpected GA. Then what if you are now at RESV fuel? Try again, wait a while or bgr off to said ALTN. CP might not be easily convinced. Surely the length of LDA has to be a major factor and should have been considered in advance. I don't mean that you expected a major wind shift, just that the crew is aware of the LDA & LDR ratio should anything happen.

All good advice and clearly you way up all the factors. A recent example I had was the tower giving wind just in limits, shortish runway, checked distance required was ok and then experienced increasing tailwind to nearly double what the tower was saying at 50 feet. Maybe it changed at the last minute, maybe they just wanted to delay changing the runway. It does not really matter. It was obviously way out of limits and with no obvious good reason to push our luck I carried out a missed approach. There is no one size fits all, you exercise your best judgement.

Although the tower reported wind is controlling, as mentioned above, there are plenty of places where one should be suspicious of tower reported wind. If the answer to a request for a wind check is "what do you need?" One should be very suspicious of reported winds.

Several airfields in Italy and Spain are notorious for this sort of creative wind reporting. The decision for runway to request is a matter of airmanship. If it's a one way place like Parma, Lamezia, Pescara &c it is wise to have briefed the circling approach so you have it if required. Especially in places where people are being pushy trying to get expeditious runways, don't budge on the runway you want for landing.

FMS wind

The FMS wind is computed as a vectorial difference between the airspeed aligned to the aircraft heading and the ground speed aligned to the ground track. The FMS-calculated wind vector is normally displayed on the Navigation Display (ND) or on the Control Display Unit (CDU). Some FMS installations provide a decomposition of the wind vector in cross- and tailwind components. The FMS calculates the wind for the altitude the aircraft is actually flying. Note that tailwind limits and the tailwind used for field performance refer to the wind measured at a 10-meter height. The FMS wind is therefore of little value to the pilot when he makes his decision to land, i.e. at top of descent, during descent and upon initiation of the final approach. Nevertheless many pilots tend to monitor the FMS for exceedance of the maximum tailwind.
Uncertainties exist in the determination of derived inertial quantities (like ground speed and ground track) that will influence the accurate determination of the FMS wind vector. Especially the calculation of the drift angle should be treated with suspect in a dynamic environment like an approach. Secondly, the airspeed is assumed to be aligned with the heading, sideslip is not measured nor incorporated in the calculation of the FMS wind, yielding questionable results once the aircraft has commenced decrabbing in crosswind conditions that might be present as well. Finally, FMS computations are filtered, resulting in a typical time delay of 3-5 seconds. A second relevant effect of this filtering process is that gust values will not be displayed to the flight crew. For these reasons the use of FMS wind is normally accurate only in the cruise phase of the flight. However, it can be shown that, although the crosswind component determined by the FMS can be highly inaccurate in the final phase of the flight, the tailwind component is relatively insensitive to FMS errors in the determination of the drift angle. This is a direct result of the geometry of the speed vectors involved.

Eh? I don't decide to land at TOPD, or for that matter on initiation of final approach.

If the FMS is showing a 15kt tailwind at 200ft, then you'd better have a good look at the windsock before you flare.

This job has and will always have a requirement for common sense and self preservation.

As a Captain you should take all information into account. That's CRM.
So to suggest that ATC is more or less important than the FMS is totally missing the point. There all relevant to getting the job done. To also suggest TRE/management should be able to give a clear answer to every possible scenario is not realistic.

I have had exactly this scenario on the 777. ATC reporting 12 kt tailwind (limit 15kt) and FMS 16kt tail all the way down the approach. My decision was at minimum if FMS still showing more than 15 we go-around simple. It went to 13 and we landed. I don't need to ask management because management have appointed me to make the decision.
At that moment its my aeroplane, my career and my life. If asked as an instructor I can give this as an "opinion/technique" not company policy. Its not written down and never will be. Its called getting the job done.

If ATC states something is within limits but you have another source of information to the contrary surely you take the sensible safe decision and use that information !

RAT 5....I have never heard of an "unexpected Go-around". Shouldn't every approach be a fully planned Go-around and the landing is the bonus. Its interesting to see guys pre-selecting ground frequency when handed over to tower. You know a Go-around has gone from the mind

RAT 5....I have never heard of an "unexpected Go-around". Shouldn't every approach be a fully planned Go-around and the landing is the bonus. Its interesting to see guys pre-selecting ground frequency when handed over to tower. You know a Go-around has gone from the mind

8che: Totally agree. I must have been unclear about my true meaning. I've spent years drumming into students that every takeoff is an RTO and every approach is a G/A. Only when past the PNR for either phase is it your lucky day. Your last comment about freq' use always gave me the eebyjeebies and was a no no. Usually the trick of smart asses, or so they thought.
We are in violent agreement.

8che;I don't need to ask management because management have appointed me to make the decision.
At that moment its my aeroplane, my career and my life. If asked as an instructor I can give this as an "opinion/technique" not company policy. Its not written down and never will be. Its called getting the job done.

Exactly, very well said and a perfect summary of what captaincy is all about.

Should be emblazoned above the back board in every command course class room and recited before prayers every day.

Which minimum?cat1?cat2?cat3?
If you go around at 200 ft because your FMS shows TW of 16kts,it is your prerogative of course,however i rather rely on tower and if at 50 ft it is still way above then initiate a go around otherwise when all land with threshold TW of 10kts and go home while you will safely go to your alternate.
I agree with all the rest,it is the captain decision.

Should an additional 3kt of tailwind alone result in an overrun? Hopefully landing performance was calculated with the maximum tailwind of perhaps 10kts. If there was an additional 3kts during the landing, whilst this is outside of the a/c performance limitation, any excess energy would be contained by the performance safety factor wouldn't it?

Should an additional 3kt of tailwind alone result in an overrun? Hopefully landing performance was calculated with the maximum tailwind of perhaps 10kts. If there was an additional 3kts during the landing, whilst this is outside of the a/c performance limitation, any excess energy would be contained by the performance safety factor wouldn't it?

Just to throw it on he table for a bar-room chat. Purely the over-run question.

A/C has -10kts AFM limit. Company has some airfields with -15kts approval. So a/c is capable of landing in -15kts; aerodynamically.
B737 landing F40 LDR < LDA by a good margin.
SE F15 landing. Vref15 = Vref40 +20.
TE flaps up landing. Vref = Vref40 + 40.
All flaps up landing Vref = Vref40+55.

The LDR is then factored = 1.3. The LDA is long enough for all these scenarios with good BA. Thus the runway is long enough to have a touchdown GS similar to a very strong tailwind.

Can you land F40 A/C in normal condition.

1. practically,
2. safely,
3. legally?

I'm only asking it against the idea that a 3-5 kts extra tailwind requires GA or even diversion. It is the written limitation that casts doubt and that addresses No. 3, but what about No. 1?

Limits can be very limiting. Just a light discussion.

I'm amazed that this is getting so much replies. The topic could have been closed after the first reply ( mine :O )

a) If it's legal for you to land and there's no doubt on your mind that a safe landing can be made, then go ahead and land.
b) If there's any doubt on your mind that a safe landing can be made, whatever the twr reported wind may be, then go-auround and find a better option.

I recently made a go-around in Rhodos Rwy 25 (3305 m length) in an A320 at 58T landing weight, because the twr reported 12 kts tailwind. Landing on R25 would have been safer then what we ended up doing: a go-around and vectors for a VOR app on Rwy 07 with crosswind and moderate turbulence on short final. Was it unsafe to go around? Maybe not. But the go-around and landing on 07 added nothing to safety and it took extra time, extra fuel and it would needlessly scare some passengers. I knew that before I initiated the go around!

I made a go around just because I feared that someone might file a report that I landed with a reported 12 kts tailwind and not because that was the safest thing to do. If you think about it, that's nothing to be proud of! (Sadly some people will say that thinking or using common sense is not done according to SOP)

It's a pity that it has come to the point that sometimes you have to be afraid of choosing the safest and best decision...

(Standing by now to get flamed because I dared to say that landing with 12 kts tailwind under the given circumstances would have been safe. Of course it's perfectly safe to land with 10 kts tailwind, but reckless to land with 12 kts tailwind. :} )

Here where I fly, the company uses the CDU'S PROGRESS page as the source data for a Continue/Go Around call at the DA.

Nevertheless I've already seen quite a couple landings with a tail wind slightly above the limit (1, 2kt) and none of us ever got a call from that.

It's interesting, however, that we are not following down here what looks like most of the world's operators do (tower wind) as the official data for backing up the decision process. Probably they just don't trust the ATC, and I can understand that (South America).

Great discussion guys,

All the best.

In the UK, reported ATC winds are usually a 2 min average with 'instant' wind available on request. ATIS are usually updated every 30 min and the wind on this and the METAR is a 10 min average at the time of the ob and hence could be up to 20 min out of date.
Some units may not have the facility for a 2 min average report and will say 'instant wind is......'

In isolation, you could probably stop most aircraft quite safely within the LDA with a 30kt tailwind, assuming you landed in the right place at the right speed. You could also take off 10% overweight and would be fine as long as nothing else went wrong. Busting minima on a precision approach will get you in 99.9% of the time.

It’s like Stabilised Approach Criteria. Most operators have a hard limit below which THOU SHALT NOT GO if you do not satisfy your SAC and also instructions to throw the approach away should you deviate from them later on. Much of this came from post-accident analysis where there was found to be a high correlation between unstable approaches and runway excursions. A good way to reduce the incidence of UA to landing? Make it simple to reject an UA and back that up with FOQA. You have to draw a line in the sand somewhere to stop people “having a go” believing they are doing their best for passengers and company.

With tailwinds, there are SOP/FCOM limitations. If you knowingly and deliberately (as judged by your company/authority) operate beyond these, as opposed to “accidentally” ;) then you are placing yourself in a rather precarious position should anything happen, even if it is not directly related.

In an emergency, of course, I would take a serious out-of-limits tailwind fully expecting to get away with it but why turn a routine flight into an emergency? I’m paid to avoid that.

De facto,


Deciding on a Go-around at 50ft in a B777-3 when you had the choice to do it earlier puts you at very high risk of striking the tail and therefore very poor judgement ! At that height even with a perfectly executed G/A the gear will likely strike the runway leaving as little as 8 degrees pitch to tail strike. You have to be incredibly careful but that comes with knowing your aeroplane. Delaying the decision until 50ft is not on with aircraft of this size. They will punish "push-on-itis" severely. I'll leave that kind of thinking to guys flying puddle jumpers !

If you can take off from 0' without hitting the tail, I'm sure you can go-around from 50' without incident. :confused:

You can go around from a floating 1 ft if you have to ( I've done it, I'm a rubbish pilot)just be aware that you can't haul the nose up until you're away and flying.

Yes you can do it guys.....but why when you have a choice to do it earlier than 1ft ? Capt Fathom you are at much greater risk of striking the tail on Go-around rather than take off, the fact you can do both isn't the point. Its risk management. Rotating with landing flap is different to take off and affected by inertia, rotating with the oleos compressed on Go-around reduces tail clearance, the semi levered gear system is for take off only..... I could bore you with the technical however there are a number of carriers including more recently our Japanese friends who have scrapped the tail doing just that. Lets not forget a "normal" go-around is the most screwed up manoeuvre in the industry. How many times do pilots practice take offs compared to Go-arounds and how many practice it from a very low Go-around. That's why Boeing specifically caution it. I am in a fortunate position to see the fun on a weekly basis in the sim.

Ok. Simple question. So you can land safely on a given runway at Vref40 + 40 with a flap problem, but you are not allowed to do a normal F40 landing with a 20kts tailwind. Why? It can't be the LDR because that issue is covered.

I agree with the comment about Rhodes, having been there is some nasty Wx. However, it is always the commander's decision to deviate from an SOP in the interests of safety. Is an ILS on 25 with an extra 2kts better than an NPA onto RW07. No question.
What did guys do before OFDM's & FMC's. Has our decision making been made easier or the opposite? Have these written limitations, in combination, inhibited sound judgement and decision making?
If you have a choice between a straight in landing ILS with 25kts tail on a long runway, or a scratch around on a circle in 3km's vis OVC500 at night to the opposite runway in a place you've never been before. Which would you consider safer? The other option is divert to somewhere with no crew or much chance of pax overnighting etc. etc.
The runway can be calculated as long enough via NNC QRH LDR's. It's the >10kts tailwind that is the gotcha, not LDA > LDR.
Remember this is a light hearted discussion. I doubt we are going to change any rules over night.

Aerodynamic and handling characteristics are different in this case. Please review the basics!:E

Also a flap problem with Vref40+40 landing would have been thoroughly briefed whereas a last minute tail wind excursion beyond limits might probably not.

Don't take it too seriously. All I'm asking is; are you really jeopardising the a/c landing with a 'suspected' 13kts tailwind instead of a supposed limit of 10kts on a sufficiently long runway?
Remember this extra 3 kts people are talking about is derived from the FMC while still above flare height. If ATC said it was 13kts then the approach would become a discussion time even before it commenced.
This is really about the perceived safety v practical argument of GA or land decisions being made at very low level.
I suspect it will become an eternal circular discussion with no clear answer and up to the individual on the day. However, there seem to be some CP's who have devised an SOP to consult the magic box at DA; and some who have made it up themselves.