Talk:Flight dynamics

From Wikipedia, the free encyclopedia

	This article is within the scope of WikiProject Physics, which collaborates on articles related to physics.	Physics Portal
???	This article has not yet received a rating on the assessment scale. [FAQ]
???	This article has not yet received an importance rating within physics.
Help with this template Please rate this article, and then leave comments here to explain the ratings and/or to identify the strengths and weaknesses of the article.

OK we can definitely do better than this stubby article. It's not just orientation and control -- it's orientation, change of orientation due to the forces acting on the body, and then the control to maintain a specific orientation or another desired condition. Lets hear about the state space model, the perturbation equations, the stability derivatives!

I don't think the usual mathematical handle-cranking approach would be that widely accessible. Still, I've made a start - what is missing are pretty pictures showing the contribution of the aircraft geometry to each stability derivative. As I see it, the aim of the game is to try to impart an intuitive understanding of the relationship between the aircraft geometry and its behaviour to as wide a readership as possible. Any fool can make an easy subject difficult, let's try to make this 'difficult' subject easy. Gordon Vigurs 09:04, 5 July 2006 (UTC)

Contents 1 Signs missing from article and wrong in picture 2 Confusing explanation 3 overlap between articles 4 Flight Dynamic by Robert Stengel

[edit] Signs missing from article and wrong in picture

The article needs to explain that yaw increases with clockwise rotation as seen from above. Pitch increases as we tilt upwards. Roll increases as the right wing dips. The picture is wrong on all three counts.

Sofixit Dan100 (Talk) 10:28, 29 October 2005 (UTC)

Well the roll is right; right wing dipping down would be an increase in roll. Same with pitch.

And yaw. Amended. --Ross UK 00:04, 5 July 2006 (UTC)

This is still wrong for aircraft. The rotation directions are now correct, but are now counter-clockwise in pitch and roll. The aeronautical set is with Y positive to Starboard, Z positive Down. The 'positive to port' convention is used on ships and land vehicles. With spacecraft, which don't have a clearly defined front and rear and are oriented so as to point antennas or solar arrays, rather than in the direction of motion, the axis set definition is completely meaningless. I have changed the axis set for consistency with Babister, and most aeronautical texts on stability and control. The picture is now incorrect. Gordon Vigurs 08:14, 26 July 2006 (UTC)

[edit] Confusing explanation

I find the introductory definitions of yaw, pitch and roll quite confusing. Do these angles fix the orientation of the aircraft in absolute terms (based on fixed north-south, east-west and up-down axes), or do they only describe *changes* in attitude relative to axes based on the plane's current orientation?

For example, suppose an aircraft has a pitch of 10 degrees and a roll of 20 degrees. I imagine this to mean that the nose-to-tail axis is first pitched up 10 degrees to the horizontal, and the aircraft is then rotated 20 degrees about its nose-to-tail axis. If the aircraft now pitches up a further 5 degrees then is that 5 degrees a rotation about the wingtip-to-wingtip axis, or about a horizontal axis?

Similarly with yaw. Does it make sense to say an aircraft has a yaw of 30 degrees, and if so 30 degrees relative to what? Or does a yaw of 30 degrees just mean the aircraft has *changed* heading by 30 degrees and could actually be pointing in any direction? And is the yaw axis always vertical, or is it perpendicular to the nose-to-tail and wingtip-to-wingtip axes, and therefore varies depending on the aircraft's current attitude?

The "Coordinate systems" section, which I hoped might clarify, actually does nothing of the sort. It says that "the pose of an object" is described as follows:

"The positive X axis goes out the nose of the airplane The positive Y axis goes out the left wing of the airplane The positive Z axis goes out the top of the airplane

Roll, pitch and and yaw constitute rotation around X, Y, and Z, respectively. The directions of all three elements are depicted in the picture above."

This makes no sense, because if the axes are relative to the object then there is, at any time, never any rotation about any of the axes, so these angles can at best describe only changes in the "pose" of the object, not the "pose" itself.

I could go on, but I'll just conclude by saying I think this stuff need a rewrite by someone who fully understands it. 00:36, 23 July 2006 (UTC)

You are correct - we are only interested in small angle changes about a nominal flight condition. However, you are also correct that the axis definition is far from clear - time to correct it. Thank you for your observation Gordon Vigurs 09:38, 23 July 2006 (UTC)

Roll, pitch and yaw, usually refer to angular velocity components, moments or incremental angles. Large angles tend to adopt a different nomenclature in aeronautics, such as 'heading' for yaw and 'bank' for roll. However, there does not appear a universal nomenclature as to whether we are concerned about perturbation motions about axes, or specification of orientation, so I will not make an issue of it. In most contexts where large angles are used, the attitude would usually be defined as a quaternion. Gordon Vigurs 12:23, 23 July 2006 (UTC)

I made a change to the coordinate systems section to try to clarify this. I think the main source of my confusion is that yaw, pitch and roll do sometimes, in some contexts, seem to refer to angles measured relative to a fixed coordinate system. So, in some contexts (though perhaps not aeronautics), an object's pitch, roll and yaw angles completely specify its orientation in space. That is what the coordinate systems section originally implied, in contradiction to other parts of the explanation. If you can find any way to further improve this, perhaps weaving in some of your explanation above, then please go ahead! 13:34, 23 July 2006 (UTC).

Your modification is correct. It's just that in this context we are trying to linearise equations to study dynamics, rather than solve the equations of motion explicitly. Gordon Vigurs 19:28, 23 July 2006 (UTC)

I consulted the article in hopes of clearing up precisely this point, so was somewhat disappointed. The above discussion helps, and it would be good to add something of the sort to the article. E.g. the yaw axes in particular is fixed with respect to the platform, and is used to describe things (angular velocity components, moments of inertia, torques, or incremental angles) that do not depend on the the current attitude of the platform. If the pilot looks forward and sees the world generally moving from right to left, then he would say he has a positive yaw rate, and could use the rudder to correct it. That applies regardless whether he's flying north, east, or upside-down. In the context of flight dynamics, I think in can make the same statement about pitch and roll. --Jrvz 13:31, 27 July 2006 (UTC)

Correct: the absolute alignment of the inertial axes (Earth axes) would only be important if the motion of the Earth contributed significantly to the total motion. Body axes are fixed with respect to the body, and move with respect to Earth axes. Wind axes are fixed with respect to the velocity vector and also move with respect to Earth axes. We are considering straight and level flight where the wind axes are initially aligned with Earth axes. In other flight conditions, there would be an initial large angle orientation to take into account in the equations of motion. I think the fact that for this particular design case, the two axes sets are initially aligned, is the source of the confusion. Perhaps analysis of the dynamics in a steady dive might help clear this up. Gordon Vigurs 18:47, 27 July 2006 (UTC)

I fixed two problems in the coordinate section: You cannot calculate orientation from the angular velocity, and inertia is not a vector (it's a tensor). --Jrvz 14:16, 27 July 2006 (UTC)

Incorrect; orientation is calculated from angular velocity by integration of the quaternion rates of change, each of which is a linear combination of the angular velocity components. Alternatively, and not to be recommended, the Euler angle rates of change may be calculated from the angular velocity, and integrated with respect to time. Finally, the direction cosines rates of change are also linear combinations of the angulatr velocity components, these also may be integrated to generate the rotation matrix, provided measures are observed to retain orthogonality. These are the methods used most frequently both in simulations of atmospheric flight vehicles and in inertial navigation. Not only can the rotation matrix be calculated from the angular velocity, this is in practice the preferred approach. Gordon Vigurs 20:28, 27 July 2006 (UTC)

Nobody ever claimed inertia was a vector, or even implied it. Gordon Vigurs 18:16, 27 July 2006 (UTC)

[edit] overlap between articles

This article, Spiral divergence, Phugoid, Dutch roll, and Instability modes of an aircraft overlap a lot and should probably be brought into agreement with one another. -68.59.121.204 03:30, 1 September 2006 (UTC)

The cited articles give good qualitative descriptions of the phenomena, whilst we are trying to quantify them in terms of aircraft geometry. The articles are directed at different audiences. This article is already very long, and the presence of mathematical formulae would almost certainly deter readers who would benefit from these other articles. Perhaps this article should be placed in the 'Engineering' category, whilst those cited should be in the 'Aeronautics' category? Gordon Vigurs

[edit] Flight Dynamic by Robert Stengel

a much better reference is the model presented by Robert Stengel in his book on Flight Dynamics.

A mass as the inertia formally exists in his model wheres inertia is removed in this presentation. A pilot has a mass and a weight in Stengel's theory of flight.

This is a very importent thing to get correct for the FAA needs to correctly determine the flight theory. This aspect or weight of the aircraft confounds pilot training and good reference in flight dynamics is hard to find.

on page 49 of "Flight Dynamics" the whole evelope is state in a single Hamiltonian function. And the equation 2-3 states weight!!!!!!!!!!!!!!IN euler-angle representation.

And the absolute elegence of the Hamiltonian presentation far outweighs the other aspect.

Maybe another wiki section on Stengel's Hamiltonian method can be added?

--207.69.139.156 15:40, 28 October 2006 (UTC)

Incorrect. The only place weight is introduced in this presentation is as a force in considering the equilibrium lift, everywhere where motion is being quantified, inertia is present.

Our objective is not to impress academics with the elegance of the method, and in so doing present the general reader with yet more Emperor's new clothes, it is the extremely difficult task of relating aircraft geometry to its behaviour, in a form which is accessible to as wide an audience as possible. Hamiltonian methods by their nature provide a means of writing down the equations of motion literally without thinking about them, indeed omitting the very understanding which we are trying to impart.

The article clearly states that we exploit our qualitative understanding to solving the equations of motion, in identifying which states are known to be relevant to which modes. It does not start from the most general possible solution and derive the answer by formal handle-cranking with absolutely no understanding of what the solutions mean in terms of causing air sickness. This illustrates the difference between an engineer's and a mathematician's thought processes.

The easiest thing in the world is to make a simple subject difficult, any fool can do that. Our objective is to inform the uninformed, and not to impress our peers. Gordon Vigurs 10:57, 4 November 2006 (UTC)

It was a suggestion idea. You are doing the hard part and making a very good resourse available.

Thanks for reading the suggestion though.

--Eaglesondouglas 00:01, 13 November 2006 (UTC)