A320 Wind Indication on PFD during Takeoff-Roll

[RvB]

Probabably discussed already...But can anyone supply information and source to the question how wind speed and direction is calculated without drift on takeoff-roll ( A320 ) ?

Thanks RvB

[mcdhu]

Mmmmm........what a good question. I can't come up with the answer. Surely someone can help?

Cheers,
mcdhu

[Field In Sight]

I haven't actually looked to see what the indication is on take-off.

But I suppose it doesn't calculate it properly and constantly uses the relationship between the aircraft heading/IRS ground track and the TAS/IRS groundspeed to calculate the wind.

It will then sort itself out once airbourne when the aircraft crabs into wind.

FIS.

[samusi01]

I don't see why it couldn't generate a wind vector - the system I use does generate a wind vector. As Field In Sight stated, all the inputs that the aircraft normally uses to generate the vector are available. The only problem is artificially imposed constraints and therefore the vector is probably not going to be correct.

[samusi01]

The aircraft should be using track and heading to calculate the wind direction. No difference between them during the takeoff roll so (if I recall correctly) you'll more than likely get a direct headwind or tailwind on the vector until you crab after lifting off.

[mcdhu]

No, you do get a pretty accurate wind arrow and readout which comes in at 100kts on the top left of the ND. I have no idea how it does it - but it does. It's Honeywell stuff in the Airbus if that helps.

Cheers,
mcdhu

[wiggy]

I think you are pretty much correct, Though it's certainly not suppressed on the 744 'cos I've watched this on the N/D (as heavy :) ) of the 744 a few times.
...what seems to happen is that whilst the aircraft is firmly on the runway the wind speed is vaguely credible ( I guess it's just using IAS minus Inertial G/S to derive windspeed), however wind direction is pants, with the wind arrow showing direction from straight ahead, since aircraft heading equals track. However once the aircraft unsticks and weather cocks into wind then often as not Track no longer equals heading and the wind "arrow" slews quite rapidly around to the correct bearing.

[Mäx Reverse]

Most interesting information during T/O for me is windspeed, especially in an ATC-Environment where you ask: 'What's the wind?' and they reply 'What do you need?' And the Head/Tail-Wind-Component is correct, whereas the direction becomes only when approaching liftoff.

If you're using reduced Thrust an calculated with L/V-Wind and then you see 10 knots or more from behind, maybe it's a good idea to go up to TOGA Thrust.

Regards, MAX

[aulglarse]

the wind vector is true wind direction and speed provided by the ADIRS. when departing on a runway equipped with an ILS this freq is auto-tuned. when take-off power is set and no trk/hdg preselected, the aircraft track can be seen on the ND at 30kts as a yaw bar ( the LOC)and 'RWY' on the FMA, so I guess the a/c is tracking magnetically along the runway with a true wind indication courtesy of ADIRS. :ugh:
FLYJETS, not sure what you mean-hope this helps: avionics cooling;
Once takeoff power is selected, the inlet and extract valves close (closed configuration)with avionics cooling coming from the a/c airconditioning.

[yanjing]

A320 can know the wind from head or tail use the diffrence between IAS and GS .
but how to know the drift ?

[Chris Scott]

Unless Airbus has retro-fitted a wind-direction vane on top of their fuselages or tail-fins, there is no way the ADIRS can possibly calculate the wind direction (as opposed to headwind component) while the aircraft track is being dictated completely or even partially by its main wheels on terra-firma.

I'm shocked that any professional pilot posting on this thread could think otherwise. Do you not study the basics of air navigation - including the vectors of heading, track and wind - for your licences nowadays? :ugh:

[767-300ER]

Well Chris Scott....if you understood the question...

Airbus aircraft (A318, 319, 320, 321, 330, 340) display a wind vector on the ND when groundspeed > 100 Kts. And it doesn't just display it as a head or tailwind, it will show a direction as well as a speed.

The original poster was asking how it does this.

It's a great question.

Sounds like your professional piloting is quite limited.:ugh:

[Peter G-W]

Perhaps it uses rudder or rudder pedal displacement? I do look at it after the 100 kts call to check on the hwc/twc.

[peekay4]

I'm guessing it's mainly just sensing the headwind/tailwind component, not actual winds. The exact direction is likely unreliable and fluctuates with even tiny heading changes during the takeoff roll.

[Chris Scott]

Quotes from 767-300ER:
(1) "Airbus aircraft (A318, 319, 320, 321, 330, 340) display a wind vector on the ND when groundspeed > 100 Kts. And it doesn't just display it as a head or tailwind, it will show a direction as well as a speed."
Yes, and - as wiggy has explained - once you get airborne it'll give you an accurate wind direction and speed. Prior to that, forget it!

(2) "The original poster was asking how it does this.
It's a great question."
I'd say it was a very fair question, perhaps with a hint of sceptical surprise?

(3) "Sounds like your professional piloting is quite limited. :ugh: "
Love it! :rolleyes:

Now, if you are convinced the displayed wind direction is accurate before the a/c becomes airborne, can you answer the OP's question?

[downwindabeam]

I don't know anything Airbus specific but generally speaking groundspeed displayed on MFDs (and some PFDs such as 777) is calculated by the navigation system utilizing both heading information as well as TAS information generated by the ADC (or whatever your flavor name for ADC is, be it ADIRU, ADIRS etc).

To solve for TAS, you need a good measurement of either TAT or OAT. In transport category jets these days TAT is what's used by means of the TAT probe in concert with the data generated by the static and pitot ports.

Usually TAT isn't very reliable on the ground and def not until it's heated up by some real friction and not by the metal just sitting there being either cold by the outside temperature or being heated by the sun in warm places because of its metallic color.

I believe and I could be entirely off base here, that because the TAT isn't reliable on the ground and at very low airspeeds that some of the data generated by its mean is suppressed purposefully because of that. Wind drift/velocity would be a good example of that.

Now if you ask why do they still display TAS next to GS showing them equal... I have no idea. One would think they'd suppress that too but I know some airplanes display that along with SAT which is obviously wrong too since it's a derived calculation based on TAT.

In particular as an answer to your drift question. It's based on heading course and TAS. Usually provided to you seamlessly by the FMC or whatever is doing the navigation portion of your overall system/fabric that is in the end displayed to you as one picture on both MFD+PFD.

[Meikleour]

Chris: your critical poster should be aware that ANY heading fluctuations such as are typical for a cross wind takeoff will allow the FMS to try to compute a wind direction - however it will be in error!
This feature was often seen during base training when the trainee had insufficient/too much rudder in for asymetric flying. You could watch the W/V readout change continually with the rudder changes! This often led the trainee to suspect, when in the sim. that "sir" had tweeked up the wind without telling. Not so, just poor rudder use. I occasionally used to demo this on the downwind leg at low speeds.

[friendly flyer]

It's a great question and I've asked training captains many times how it does it but nobody seems sure. One suggested some info was derived from rudder inputs during take off roll but I'm not convinced. The arrow at 100knots is pretty accurate though........

[lurkio]

As far as pilot FCOMs are concerned there are only a couple of F.Ctl inputs to the ADIRS system. ELACs and the BSCU. Given that it should not (it is Airbus after all) be ELAC it could well be the BSCU. Agreed it would be a very coarse way of measuring the input required to maintain track but it would appear to be all we have. Logically by the time we reach 100 knots we should have the thing vaguely under control so a measurement about then would be roughly correct.

Logically that is and as I said it is Airbus.

[Chris Scott]

Thanks Meikleour,
I think rudder movements on the T/O run may well have something to do with the displayed wind-direction fluctuations you and others describe. **
(BTW, I'm not sure the sim is always representative. I seem to remember a F/O once telling me that in the sim the secret of tracking the ILS localiser accurately on the approach was to use the nosewheel steering...)

Quote from the OP, RvB (my emphasis):
"...can anyone supply information and source to the question how wind speed and direction is calculated without drift on takeoff-roll ( A320 ) ? "

Was hoping not to resort to didactics, but... back to basics. The way an ADIRU calculates W/V is the same as a navigator would, isn't it? Except it does it instantaneously. The relevant triangle of vectors includes HDG(T)/TAS and TRK(T)/GS, with W/V (wind bearing and true speed) completing the triangle.

The IRU measures HDG(T), TRK(T) and GS. The AD unit calculates the TAS from IAS, TAT and pressure altitude.

So the difference between HDG and TRK (i.e., drift), is an essential datum to calculate W/V. On T/O, If the aircraft was mounted on rails like a V1 (doodlebug) from WW2, there would be zero drift, even in a strong crosswind. In that case the ADIRU would show the wind direction as being either the same as the a/c HDG or the reciprocal, depending on the comparison between TAS and GS.

In practice, the tread grip and profile of the (radial) tyres can only do so much in a crosswind to try and achieve that. A small discrepancy between HDG and TRK is going to develop, particularly on a wet runway. This will constitute drift. Also, the PF will be making frequent changes of rudder angle, and tiny changes of HDG, to stay near the runway centreline. The resulting lateral accelerations will lead to momentary fluctuations of the drift, which probably explains the rapid changes in the wind direction displayed on the ND.

BTW, the title of this thread refers to "wind indication on PFD." Am I correct in assuming that the OP meant the indications on the ND? If not, things have changed since my day.

**
As you suggest, [B]Meikleour, in flight any asymmetry-induced sideslip (including bad flying) will presumably be interpreted by the ADIRU as drift (difference between HDG and TRK). downwindabeam also raises an interesting point about the possible difficulty in calculating TAS using the TAT from a heated probe (fully heated for T/O?) at low speeds.

[ACMS]

Oh come on fellas...:ugh:

It's on the ND but the direction CANNOT be correct, it will jump around depending on the sensed track and heading changes on the roll.....

Half the time the left and right ND wind displays are different on the Airbus anyway!! Not just a little either.....:sad:

[mcdhu]

Careful study of this in my twighlignt years of flying (now retired but active in th sim) revealed that, as CS says above, there is no way the ADIRSs can work out the S/W between 100kts and getting airborne. All you get is a head or tail component which is influenced by 'drifting' left or right of the C/L.
The subject is further confused by th fact that the Thomson/Thales/L3 sim that I sometimes use does display the S/W while still on the RW while the CAE sims only display the head/tail component.
Careful monitoring of the ND wind indication during asymm flying will reveal all sorts of 'strange' winds brought about by unbalanced flight which the IRSs see as drift.
I hope this helps.
mcdhu

[RAT 5]

One poster says this is displayed >100kts. Some questions: I can understand the curiosity of how this vector is calculated, but why are you looking at it? ATC have given you the wind in the takeoff clearance. During takeoff roll you PF is concentrating on the centreline and far end of the runway and speed. Rudder input will tell you which side the wind is coming from.
So my question is, what useful information, at 100kts, is this vector giving you? I also ask the question as to why AB created it rather than inhibit it until wheels up?

[ZeBedie]

I also ask the question as to why AB created it rather than inhibit it until wheels up?

If you knew there was a risk of a microburst and you started picking up a massive tailwind, it may make you think twice about continuing the T/O?

[Derfred]

I would pick that up as a stagnation in IAS. Not by looking at the wind vector.

Read the manuals. Does anyone's manual suggest looking at the wind vector during takeoff? Mine doesn't, but it does tell me what to do if the IAS stagnates.

Don't make **** up.

[deefer dog]

If you knew there was a risk of a microburst and you started picking up a massive tailwind, it may make you think twice about continuing the T/O?

Great theory, but useless in practice. Apart from the distraction of looking at a tiny wind vector arrow while barreling down the runway, trying to keep straight and monitor engine parameters with a "massive" tailwind, how could one determine whether one has already used up the stop distance? Any V1 based on indicated airspeed would be useless if the tailwind was so massive.

As for the aircraft displaying wind direction with any degree of accuracy while on the ground - it's poppycock!

[Uplinker]

Interesting OP.

The only wind data available to the aircraft on the runway is pitot pressure and air density. So, it can only calculate a head or tail wind component. As for wind direction, my guess is that it must take an average measurement of the rudder angle trend which would enable it to give an estimation of wind direction left or right. Combine that with the head/tail component calculation, and you can give an estimate of the wind vector.

I notice the A380 has AoA style vanes mounted vertically on the nose - presumably to actually measure the crosswind?

@RAT5 On take-off, when looking in to confirm and crosscheck your airspeed reading at 100 kts, it's always worth a quick glance to the wind vector to confirm that the tower wind was reasonably correct. I had a very interesting departure from the Caribbean once because the wind they gave us was incorrect. :ooh:

A tailwind greater than 10 kts would clearly be a cause for concern, but when spotted at 100 kts, the T/O can be abandoned. (Well ANY unexpected tailwind not allowed for is a potential problem of course, as we almost found out that day)

[NSEU]

The A380 has beta vanes. Does the A320 have these or can it detect sidewinds by comparing left/right AOA angles as mentioned in this PPRuNe thread?

SIDE SLIP ANGLE [Archive] - PPRuNe Forums (http://www.pprune.org/archive/index.php/t-492341.html)

[FlightDetent]

Contrary to some above (most actually :)) I do sometimes look at the wind readout on ND and find it useful with reasonable values. How is it calculated - no idea, I wondered myself. Best I could come up with was differentail readings between static ports L/R, but...

[Uplinker]

Best I could come up with was differential readings between static ports L/R, but...

Yer, that could work, and/or dynamic pressure from the pitot probes perhaps? - just to give an estimation.

Which in turn might explain why it is suppressed until 100 kts - when the airflow over the fuselage has stabilised?


Edit:
Checked my old paper FCOM and it just says "wind speed and direction arrow [on the ND] is provided by the ADIRS", but it doesn't say which elements of air data are used.

[RAT 5]

Next question. Can you trust it? Really trust? If not then using it as a decision to RTO is an interesting one. In the normal takeoff brief about reasons to RTO it mentions a/c unfit to fly. Are you linking this scenario under that heading? Remember most takeoffs are at reduced power; all takeoffs assume loss of 1 engine. So, an increase in tailwind, or loss of headwind should be coped with by an increase in thrust on ALL engines. It would really be a very bad hair day to have a sudden shift in wind AND an engine failure.
I remember at Mombasa B757. Naturally a warm day and a gentle breeze from somewhere. Off we went at reduced thrust. I thought I saw the far end of the runway getting closer than expected and we were still <V1. Push up to full power and we rolled through the mirage and lifted off in half the runway. A pilot colleague in the cabin, during our long missed en-route cockpit visit/chat, asked about the sudden power bump during the roll. I explained and he understood, fully.
Like many have said, IAS readout plus looking at the far end might give more an accurate picture than a tiny wind arrow display. What does AB say about using it and its authority to make RTO decisions? Is it a nice toy put there without any particular guidance how to use it. The same is true for FPV on B737, but that's another story. If anyone wants to go down that route please one a new topic.

[compressor stall]

For those suggesting that the wind ector could be used as information for an RTO, consider Airbus' advice for windshear:

The takeoff should only be rejected if significant air speed variations occur below indicated V1 and the pilot decides that there is sufficient runway remaining to stop.

I'd suggest tea and bickies if you'd aborted for an indicated wind vector change with no airspeed variation.

[Uplinker]

Well perhaps, but how significant an airspeed variation would you chose to abort at and how much time would you want to spend watching the speed tape and trend to assimilate it while taking off? Have you ever briefed how much trend variation and in which direction it would have to be for you to stop or go? I am happy to have the ADIRS calculate it and give me a wind vector. Any unbriefed tail wind (or more than 10 kts tail if allowed for), would be a no brainer - either abort, or TOGA, depending on where you were on the runway.

A couple on here have spoken about the 'small size of the wind readout', but the '320/12' takes up more space than the speeds on the speed tape, and together with the arrow indication is not that small really, and you only need to glance at the arrow.

I don't think I would get tea and biscuits for being careful.

[ZeBedie]

Next question. Can you trust it? Really trust? If not then using it as a decision to RTO is an interesting one.

Agreed, but remember, the decision to display this information was made by engineers, who may see things in a different way to you and I.

[Uplinker]

Can you give a citation for that claim ZeBedie?

If, as it seems, some of you don't trust a head/tail wind display, what makes you trust an IAS, N1 or EPR display? (also provided by engineers).

[compressor stall]

So Uplinker,

You're on a Runway where perf figures suggest a 500m margin for takeoff.

Figures predicated on 5 kt headwind. Vr is 140.

Passing 110kts your eagle eyed PM takes his eyes off the engine display and notices a 7 kt tailwind on the wind readout. Neither of you had noticed anything untoward on the speed trend vector or tape.

What would you (or your captain expect you to) do?

How much time would you want to spend watching the speed tape and trend to assimilate it while taking off?

I can't believe I'm answering this. You have a split second to look at the size of the arrow and the movement of the tape before back outside or to the engine parameters. You don't need to process a value or compute anything. Just long enough to get a feel as to how long until the next speed (100 or v1).

You only need to glance at the arrow.
Unless your ND arrow increases in size with the wind direction you need to read the speed too. If youve seen the arrow going all over the place when the outside wind is vrb at 1-2 kts at altitude you'll understand. How do you know if that sudden tailwind is 1 kt or 10?

To look at the wind vector (which is of dubious reliability anyway) gives you no extra information, takes your eyes away from other more important items, and distracts with the cognitive awareness for other decisions.

The second of time that the PM spends trying to work out if the wind has backed or veered, and what the subsequent headwind component has done would be much better spent looking at the engine parameters.


And regarding unexpected tailwinds on departure, thats what the windsocks are for. If it's saying something you dont like on a marginal runway when you're lined up exit and recalculate with conservative estimated figures.

[Uplinker]

So Uplinker,

You're on a Runway where perf figures suggest a 500m margin for takeoff.

Figures predicated on 5 kt headwind. Vr is 140.

Passing 110kts your eagle eyed PM takes his eyes off the engine display and notices a 7 kt tailwind on the wind readout. Neither of you had noticed anything untoward on the speed trend vector or tape.

What would you (or your captain expect you to) do?


Ensure a very positive rotation and/or select TOGA.


Quote:
How much time would you want to spend watching the speed tape and trend to assimilate it while taking off?
I can't believe I'm answering this. You have a split second to look at the size of the arrow and the movement of the tape before back outside or to the engine parameters. You don't need to process a value or compute anything. Just long enough to get a feel as to how long until the next speed (100 or v1).


If, as you claim, you have such little time to process a parameter, then by your own logic does it not make sense to look at a readout that has already been measured and processed by a computer and therefore only needs a glance rather than a longer look to assess its trend?

Quote:
You only need to glance at the arrow.
Unless your ND arrow increases in size with the wind direction you need to read the speed too. If youve seen the arrow going all over the place when the outside wind is vrb at 1-2 kts at altitude you'll understand. How do you know if that sudden tailwind is 1 kt or 10?

If the wind arrow is in the general direction you are expecting, then there's no need to look at it again. If you have allowed for 10 kts tailwind, you just need to glance at the windspeed readout to confirm it is no more than 10kts.

To look at the wind vector (which is of dubious reliability anyway)...

Why have folk got such a hard on about the 'reliability' of the wind readout? Do you trust the autobrake to keep you straight on landing?, because it does, and it uses processed data from the ADIRS to modulate the brakes to do just that.

... gives you no extra information, takes your eyes away from other more important items, and distracts with the cognitive awareness for other decisions.

Well it does give you extra information: the head or tailwind. Is that not an important parameter to check on the takeoff roll?

The second of time that the PM spends trying to work out if the wind has backed or veered, and what the subsequent headwind component has done would be much better spent looking at the engine parameters.

No working out is needed, that's the point - the wind arrow and speed readout tells you in the fraction of a second it takes to glance at it ! As for looking at the engine parameters, PM has just watched the speed tape on the PFD to announce 100kts, then while moving their eyes back to the engine instruments, glances at the wind readout on the way. Easy!

As for the size of the wind readout; the arrow is thicker than the V/S arrow, and bigger than the beta target indicator; surely you can see those? The numerical element of the wind readout is a similar size to ALT CRZ on the FMA - you can see that surely?

Windsocks. Hmmmm. So you are complaining about glancing at a processed readout which tells you the actual value of the wind, yet you advocate looking outside the cockpit, finding the windsock (at night - is it lit?), and assessing a head or tail wind from that. Is it more than 10 kts? And won't the windsock be behind you by the time you are into the take off roll?

And in what state are the windsocks in some second/third world airports?

I think I would rather rely on the on-board multi-thousand dollar measuring and processing equipment and computation power fitted to the plane.:ok:

(Remember, I am talking about the wind not being what it was after the tower passed it to you and you lined up, or the tower passing an incorrect wind. This actually happened to me and getting a heavy A330 airborne with an unannounced tailwind, was a moment of extreme concentration, I can tell you! - for a second or two I thought one or both elevators had failed. I now always glance at the wind readout on take-off)