Effect of crosswind on fixed wing field performance
Moderator
Thread Starter
Effect of crosswind on fixed wing field performance
Does anybody happen to have any papers / books / reliable rules of thumb / etc. that covers the effects of crosswind component (as opposed to head/tailwind component) on the landing and (especially) take-off performance of a fixed wing aeroplane?
G
G
Guest
Posts: n/a
Oooohhh Genghis - you've hit sweet spot of mine. Unfortunately not shared by CAA or some worthy aeronauts of my acquaintance.
The short answer is no - haven't found anything in the literature. I've had honours student projects on the topic for the last 3 years and we've scoured the literature. We've also measured take-off performance (run and distance) using a hand held GPS (that's a story in its own right) for a range of aircraft
- Grob Tutor (thanks to the local UAS)
- Bulldog
- PA38
- Beagle Pup 1
- Bolkow 208c
We used 10 take-offs in each case, to get a reasonable stab at calculating 95% probability/confidence bounds on the results. With nil wind all aeroplanes meet the relevant FM/POH figures. In a x-wind (15 kts) , the 208 requires a factor of 1.2 to be applied to get the book figure. The others haven't been tested in a decent x-wind so the results there are inconclusive.
One might think that x-wind is an issue - after all the aeroplane is sideslipping during the take-off run so one might expect the drag to be higher. With this in mind an aircraft like the 208 might be amongst the worst, being boxy and slab-sided, whereas the Tutor might be the best with its slender tailboom.
I'm in the process of putting together a paper for the Aeronautical Journal on the subject, but it'll be a while yet - I need to get max x-wind days - believe it or not, quite difficult to get up here!
If anyone has any reference material on the subject, I'd be most keen to see it. BTW CAA view seemed to be that x-wind didn't need its own factor - that's what the 'recommended' PT 1.3 factor was there for.
Sorry for ramble....
The short answer is no - haven't found anything in the literature. I've had honours student projects on the topic for the last 3 years and we've scoured the literature. We've also measured take-off performance (run and distance) using a hand held GPS (that's a story in its own right) for a range of aircraft
- Grob Tutor (thanks to the local UAS)
- Bulldog
- PA38
- Beagle Pup 1
- Bolkow 208c
We used 10 take-offs in each case, to get a reasonable stab at calculating 95% probability/confidence bounds on the results. With nil wind all aeroplanes meet the relevant FM/POH figures. In a x-wind (15 kts) , the 208 requires a factor of 1.2 to be applied to get the book figure. The others haven't been tested in a decent x-wind so the results there are inconclusive.
One might think that x-wind is an issue - after all the aeroplane is sideslipping during the take-off run so one might expect the drag to be higher. With this in mind an aircraft like the 208 might be amongst the worst, being boxy and slab-sided, whereas the Tutor might be the best with its slender tailboom.
I'm in the process of putting together a paper for the Aeronautical Journal on the subject, but it'll be a while yet - I need to get max x-wind days - believe it or not, quite difficult to get up here!
If anyone has any reference material on the subject, I'd be most keen to see it. BTW CAA view seemed to be that x-wind didn't need its own factor - that's what the 'recommended' PT 1.3 factor was there for.
Sorry for ramble....
Moderator
Thread Starter
Out of interest, is your own attack on this purely experimental or have you had a stab at modelling the (performance) flight mechanics of a crosswind take-off?
G
G
Guest
Posts: n/a
Genghis
Majored on the experimental, but tried to model using the equation for 'Sg' in Eshelby's book. X-wind effect modelled purely by using an increased Cl/Cd obtained from 'power-off' glides, with and without sideslip. Grossly overestimates take-off run.
Used parameter estimation to get acceleration parameters for model as well - we get uniform accelerations of the order of 1.4m/s^2 for the Pup. This is needed for Eshelby's model but prediction still poor - overestimate of the order of 40% with or without x-wind, so the model needs to be addressed.
PS emailed Kenji - all is fine now. Sorry for delay. Well done!
Majored on the experimental, but tried to model using the equation for 'Sg' in Eshelby's book. X-wind effect modelled purely by using an increased Cl/Cd obtained from 'power-off' glides, with and without sideslip. Grossly overestimates take-off run.
Used parameter estimation to get acceleration parameters for model as well - we get uniform accelerations of the order of 1.4m/s^2 for the Pup. This is needed for Eshelby's model but prediction still poor - overestimate of the order of 40% with or without x-wind, so the model needs to be addressed.
PS emailed Kenji - all is fine now. Sorry for delay. Well done!
Moderator
Thread Starter
PS emailed Kenji - all is fine now. Sorry for delay. Well done!
=============================================
Just pulled Eshelby off the shelf, presumably the basic simplicity of the model is letting you down - you need something much more complicated to allow tweaking of rthe various parameters, viz..
- Variation in rolling friction due sideforce
- Variation in profile drag due sideslip (and due rudder and aileron deflection?)
- Variation in propeller efficiency due sideslip
- Variation in lift due aileron deflection (which will cant the lift vector sideways and cause an increase in download on the tyres for any given airspeed).
?
I tried Hale (ISBN 0-471-07885-9), which is slightly more friendly, but a basically similar model to Eshelby's - but what about Lowry? (ISBN 1-56347-330-5) It's a much more complex model, but might be more amenable to tweaking for variations in these factors during take-off (That's for the ground roll, the air segment should be just trig I'd venture).
Aint easy though, I confess the more I think about it, the more I like the idea of starting experimentally and trying to generate some reasonable experimental factors.
G
Last edited by Genghis the Engineer; 31st Mar 2005 at 05:57.
Ghengis. Slightly off the subject, I know - but talked to a colleague recently who was F/O on a crosswind take off in a B737-800, where the crosswind component was 25 knots and the captain wound in full aileron at the beginning of the take off run and kept it there till rotation. My friend said that around 110 knots IAS the acceleration slowed markedly and V1 became meaningless.
I recall in the early years of 737 flying the flight crew training manual for the 737 recommended a certain amount of pre-set aileron as the take off run commenced. This has been removed I think from current FCTM. I could never understand why it was there in the first place. All it says now is that excessive control wheel movement will cause spoiler actuation and markedly increase the TOD. The point is there is no guidance as to what is excessive ccontrol wheel movement.
My experience is that even with 35 knots of crosswind component, there is no need for wheel into wind unless you feel the wing is lifting. Similarly on landing after touch down, with all spoilers up, is there any real need to apply any control wheel deflection during the landing run? Maybe on a light GA aircraft (Cessna, Piper types for example) yes - but seems a overkill on a big jet?
With full aileron deflected on a 737 type during the take off run in a strong (35 knots plus) crosswind, is there any way of estimating how much the planned TOD would be exceeded due to spoiler drag?
I recall in the early years of 737 flying the flight crew training manual for the 737 recommended a certain amount of pre-set aileron as the take off run commenced. This has been removed I think from current FCTM. I could never understand why it was there in the first place. All it says now is that excessive control wheel movement will cause spoiler actuation and markedly increase the TOD. The point is there is no guidance as to what is excessive ccontrol wheel movement.
My experience is that even with 35 knots of crosswind component, there is no need for wheel into wind unless you feel the wing is lifting. Similarly on landing after touch down, with all spoilers up, is there any real need to apply any control wheel deflection during the landing run? Maybe on a light GA aircraft (Cessna, Piper types for example) yes - but seems a overkill on a big jet?
With full aileron deflected on a 737 type during the take off run in a strong (35 knots plus) crosswind, is there any way of estimating how much the planned TOD would be exceeded due to spoiler drag?
Join Date: Sep 2002
Location: La Belle Province
Posts: 2,179
Likes: 0
Received 0 Likes
on
0 Posts
If you wanted to have a really crude estimate.
Look at the distance penalties for the corresponding spoiler failure cases on landing, and apply them. It's very crude, but if loss of the OB pair of spoilers causes a 10% distance penalty on landing, you might guess that having one of the pair stuck up as a spoileron might cause a 5% TOD impact.
(Yes, I know there are braking efficiency issues and different speeds and what-not; I did say 'crude')
Look at the distance penalties for the corresponding spoiler failure cases on landing, and apply them. It's very crude, but if loss of the OB pair of spoilers causes a 10% distance penalty on landing, you might guess that having one of the pair stuck up as a spoileron might cause a 5% TOD impact.
(Yes, I know there are braking efficiency issues and different speeds and what-not; I did say 'crude')
Join Date: Jul 2000
Location: Smogsville
Posts: 1,424
Likes: 0
Received 0 Likes
on
0 Posts
Having just left the A330/340 and now started on the 747, I wanted to bring this to the top to see if there was any more information available.
The reason being during crosswind takeoffs the Airbus required virtually no aileron input into wind, but the 747 requires considerable into wind aileron especially during rotation and landing.
I'm particularly interested in finding out information about spoiler lift and takeoff performance.
I came across two schools of thought regarding spoiler lift on the Airbus.
1. Put what's required in as spoiler lift is accounted for during flight testing.
2. Keep the roll input to a minimum to avoid spoiler lift.
Also two on the 747 which is to put 1 unit of roll in per 10kts of crosswind & don't exceed about 1 unit as the spoilers lift.
Note: The spoilers lift at 8 degrees which is about 1 unit of aileron trim.
http://www.airliners.net/open.file/841191/L/
Both flight manuals mentioned when the spoilers lift but neither seem to commit to a take off technique, other than keep the wings level.
Any further information would be greatly appreciated.
Regards SMOC.
Video of crosswind landings (just for interest)
http://www.linhadafrente.net/bin/Pousos.wmv
The reason being during crosswind takeoffs the Airbus required virtually no aileron input into wind, but the 747 requires considerable into wind aileron especially during rotation and landing.
I'm particularly interested in finding out information about spoiler lift and takeoff performance.
I came across two schools of thought regarding spoiler lift on the Airbus.
1. Put what's required in as spoiler lift is accounted for during flight testing.
2. Keep the roll input to a minimum to avoid spoiler lift.
Also two on the 747 which is to put 1 unit of roll in per 10kts of crosswind & don't exceed about 1 unit as the spoilers lift.
Note: The spoilers lift at 8 degrees which is about 1 unit of aileron trim.
http://www.airliners.net/open.file/841191/L/
Both flight manuals mentioned when the spoilers lift but neither seem to commit to a take off technique, other than keep the wings level.
Any further information would be greatly appreciated.
Regards SMOC.
Video of crosswind landings (just for interest)
http://www.linhadafrente.net/bin/Pousos.wmv
Join Date: Feb 2005
Location: flyover country USA
Age: 82
Posts: 4,579
Likes: 0
Received 0 Likes
on
0 Posts
Crosswind - critical direction?
On a prop-powered craft, a bit of rudder (sometimes a LOT of rudder) is required to hold the runway centerline on takeoff - the old "torque" effect (rotating slipstream against the fin, I'm told).
For a craft here in the Colonies with a clockwise-turning prop, it's right rudder that's required. (You backwards folks can do it your own way!)
If it's a left crosswind of any strength, you may well run out of rudder authority unless power is applied with caution. A lovely old taildragger met an unkind fate near here due to this effect.
Thus reduced takeoff performance is due to the reduced / delayed power used.
However, an equal but opposite crosswind would pose no such problem.
Are you considering the direction of crosswind in this study?
For a craft here in the Colonies with a clockwise-turning prop, it's right rudder that's required. (You backwards folks can do it your own way!)
If it's a left crosswind of any strength, you may well run out of rudder authority unless power is applied with caution. A lovely old taildragger met an unkind fate near here due to this effect.
Thus reduced takeoff performance is due to the reduced / delayed power used.
However, an equal but opposite crosswind would pose no such problem.
Are you considering the direction of crosswind in this study?
Flying BN2s in Shetland I frequently needed to use asymmetric power* to control yaw during the take-off roll. I'd be surprised if the lack of power on one side didn't adversely affect the TODR. Problem was that it had to be masked by the HWC of whatever wind was present *and*, conversely, by me keeping the a/c on the ground for bit longer than usual to ensure a fast, clean transition from wheel-borne to wing-borne.
*As an aside, I sometimes had to use asymmetric power during landing to get sufficient yaw rate to be able to touchdown without drift in particularly strong x-wind components ie >40kt. Bloody Islander was soooo sluggish otherwise... Wonder how much difference this made to LDR?
*As an aside, I sometimes had to use asymmetric power during landing to get sufficient yaw rate to be able to touchdown without drift in particularly strong x-wind components ie >40kt. Bloody Islander was soooo sluggish otherwise... Wonder how much difference this made to LDR?
Join Date: Oct 2000
Location: South West UK
Posts: 367
Likes: 0
Received 0 Likes
on
0 Posts
Another factor to throw into the melting pot (only relevant to certain old aircraft, eg Chipmunk, Hunter, JP in fact anything British with no nose/tailwheel steering!) is the use of brakes to keep straight early in the T/O run. The manuals for some of these aircraft state that the take off run may be increased by this but make no attempt to quantify the effect! I often wondered what the penalty would be but never found any answers.
I guess there would also be a decrement in thrust caused by the air entering the jet intake (or the propeller) at the sideslip angle and also an increment of intake momentum drag!
3 Point
I guess there would also be a decrement in thrust caused by the air entering the jet intake (or the propeller) at the sideslip angle and also an increment of intake momentum drag!
3 Point
During the take off roll I fail to see how holding even large control wheel deflection angles into the wind (with resultant upwind spoiler deflection) would increase drag affecting ground roll, as long as most of the correction is removed prior to rotation, some into wind aileron must be held through rotation however.
My point is that, up until rotation this is, essentially a jet propelled ground vehicle (albeit a lift producing one) and that during acceleration spoilers merely kill lift, but that should not affect acceleration, UNTIL the aircraft is airborne.
My point is that, up until rotation this is, essentially a jet propelled ground vehicle (albeit a lift producing one) and that during acceleration spoilers merely kill lift, but that should not affect acceleration, UNTIL the aircraft is airborne.
Join Date: Jan 2004
Location: A Barren Featureless Wasteland
Posts: 106
Likes: 0
Received 0 Likes
on
0 Posts
stilton,
As I'm sure you know, it'll have drag even when it's not producing lift. If you stick a spoiler up on take-off it will increase the drag during the T/O roll.
Up until rotation (I would say) that it is not a 'lift producing jet propelled ground vehicle' (albeit some lift may be generated depending upon the ground attitude lifting surface incidence angle). I'd say it was pretty much all form drag on the pre-rotation ground roll.
Mobius :-)
As I'm sure you know, it'll have drag even when it's not producing lift. If you stick a spoiler up on take-off it will increase the drag during the T/O roll.
Up until rotation (I would say) that it is not a 'lift producing jet propelled ground vehicle' (albeit some lift may be generated depending upon the ground attitude lifting surface incidence angle). I'd say it was pretty much all form drag on the pre-rotation ground roll.
Mobius :-)
Your point on form drag is well taken, however since ground acceleration would be affected by total aerodynamic drag amongst other things, specifically in the crosswind case, let's say you apply no input into wind causing the upwind wing to rise as you accelerate surely you are creating more drag in this case.
I would even suggest that the correct amount of upwind control yoke input with resultant spoiler deflection allows the aircraft to accelerate more 'easily' up to the rotation point.
I would even suggest that the correct amount of upwind control yoke input with resultant spoiler deflection allows the aircraft to accelerate more 'easily' up to the rotation point.
Join Date: Jan 2004
Location: A Barren Featureless Wasteland
Posts: 106
Likes: 0
Received 0 Likes
on
0 Posts
Ummm, ok - I don't know about that - what's your reasoning behind the AC having more drag if the wing is lifted by the crosswind?
Perhaps it is suitable to think about looking at the AC head on as it accelerates toward Vr. Generally speaking, the more of the AC you 'see' the more drag there is (form). I can't see why a wing being lifted up by the crosswind would give more form, but I can see how a deflected spoiler would?
We might be barking up the wrong tree entirely here - perhaps the dominant factor (for the AC under consideration) is in fact the requirement to apply asymmetric braking (with consequent effect on accel) due to asymmetric main wheel load/drag vice the effect of spoiler drag?
I haven't ever completed (or even seen the results of) a trial with/without into-wind controls on take-off performance - anyone else done/see this? I have, however, always applied into wind stick/yoke in every AC I've flown on take-off (as far as I can recall) - because that's what I was taught to do from day one.
I'm not going to get to sleep now stilton!!
MT :-)
edited because I suddenly realised what you meant! I suppose it would be a toss-up between the drag from the 'lifted' wing and that from a spoiler to keep it down, perhaps? However, the wing is being lifted (in some respects) in a different way to 'normal' wing lift because, amongst other things, the side fin force is 'tilting' the AC. As the AC accelerates, the 'lifting' force becomes more conventional (due to the RAF).
Perhaps it is suitable to think about looking at the AC head on as it accelerates toward Vr. Generally speaking, the more of the AC you 'see' the more drag there is (form). I can't see why a wing being lifted up by the crosswind would give more form, but I can see how a deflected spoiler would?
We might be barking up the wrong tree entirely here - perhaps the dominant factor (for the AC under consideration) is in fact the requirement to apply asymmetric braking (with consequent effect on accel) due to asymmetric main wheel load/drag vice the effect of spoiler drag?
I haven't ever completed (or even seen the results of) a trial with/without into-wind controls on take-off performance - anyone else done/see this? I have, however, always applied into wind stick/yoke in every AC I've flown on take-off (as far as I can recall) - because that's what I was taught to do from day one.
I'm not going to get to sleep now stilton!!
MT :-)
edited because I suddenly realised what you meant! I suppose it would be a toss-up between the drag from the 'lifted' wing and that from a spoiler to keep it down, perhaps? However, the wing is being lifted (in some respects) in a different way to 'normal' wing lift because, amongst other things, the side fin force is 'tilting' the AC. As the AC accelerates, the 'lifting' force becomes more conventional (due to the RAF).
Last edited by MobiusTrip; 17th Nov 2005 at 20:09.
I had not considered the effect of asymetric braking and would not expect to have to use that in this case.
It does bring up another factor , however, that of the aircraft, not only wanting to weathervane into the wind, but the 'heeling' affect that the crosswind produces increasing the frictional drag of the downwind main gear.
I think, that, once again though, this affect would be minimized by the appropriate amount of upwind spoiler deflection, 'more equally 'evening the loads on the main gear, and lessening drag prior to rotation.
I think the only way you could conclusively prove anything is to try , in a simulator using varying degrees (and zero) spoiler deflection and noting the effect on ground roll.
It does bring up another factor , however, that of the aircraft, not only wanting to weathervane into the wind, but the 'heeling' affect that the crosswind produces increasing the frictional drag of the downwind main gear.
I think, that, once again though, this affect would be minimized by the appropriate amount of upwind spoiler deflection, 'more equally 'evening the loads on the main gear, and lessening drag prior to rotation.
I think the only way you could conclusively prove anything is to try , in a simulator using varying degrees (and zero) spoiler deflection and noting the effect on ground roll.
Join Date: Jan 2004
Location: A Barren Featureless Wasteland
Posts: 106
Likes: 0
Received 0 Likes
on
0 Posts
Indeed, if you have the luxury of NWS then asymm braking isn't going to be a factor.
Your 'other factor' is what I meant when I talked about asymm mainwheel drag (perhaps I expressed it poorly). It certainly would be minimised by appropriate spoiler/aileron. You could (to some extent) steer the F4 on the runway using aerodynamics to load/drag up the appropriate main gear.
I'll give it a go some time when we have suitable crosswinds and I have a spoilered AC to fly :-) If you get there first, please let me know what happened!
Cheers,
MT
Your 'other factor' is what I meant when I talked about asymm mainwheel drag (perhaps I expressed it poorly). It certainly would be minimised by appropriate spoiler/aileron. You could (to some extent) steer the F4 on the runway using aerodynamics to load/drag up the appropriate main gear.
I'll give it a go some time when we have suitable crosswinds and I have a spoilered AC to fly :-) If you get there first, please let me know what happened!
Cheers,
MT
