On 11/22/2007 07:28 AM, LeeE wrote: > FWIW I also spotted that the coded hard-coded hud has problems around > headings of 180 deg so that if you are heading slightly to one side of > 180 deg and there's enough of a cross-wind to swing the direction > indicator to the other side of 180 deg it disappears, probably due to a > wrap-around type error.
Sounds like a bug to me. It could be a bug in the FDM (which seems kinda unlikely) or it could be a bug in how the HUD uses the information. Bugs like this are par for the course when such things are quantified in the Euler-angle representation or the axis-angle representation. For a lumbering bomber you can get away with using a lame representation, but for something more maneuverable you would be much better off using the geometric algebra representation aka Clifford algebra aka multivectors aka quaternions. If you think you've got problems when the heading angle goes to 180 degrees, think about what happens when the pitch angle goes to ±90 degrees. > Yaw, to me means the difference between the direction that the aircraft > is pointing and the direction that it is moving. Moving? From what follows I gather that means moving relative to the airmass (not moving relative to the ground). It is important to think clearly and speak clearly, so as to not blur this distinction. > Flying directly into, > or with, any wind would produce no yaw but flying in a cross-wind would > produce yaw. When sitting stationary on the ground, any cross-winds > would result in an effective airspeed, so yes, I would expect there to > be a large yaw element under those circumstances. That's the slip angle. It is almost universally denoted by β (beta). It's in the tree as /orientation/side-slip-deg /orientation/side-slip-rad which are computed by the FDM. If that variable doesn't show a large slip angle when parked in a crosswind, that's a nasty bug. You can easily enough check whether the FDM is calculating this correctly by recomputing it yourself. It's just the angle between the relative wind vector (projected onto the XY plane) and the heading vector. It's trivial to calculate: u = /fdm/jsbsim/velocities/u-aero-fps v = /fdm/jsbsim/velocities/v-aero-fps beta = atan2(v, u) convert from radians to degrees if you wish ... although (as mentioned above) for most purposes you're better off using the (u,v) vector as a vector, rather than converting it to an angle. If the FDM is not calculating u-aero and v-aero properly and/or not calculating β properly and/or not putting β in /orientation/side-slip-rad then it is a very serious bug. Please don't call it the yaw angle. Please call it either slip angle or β. Note that the thing some pilots call a yaw string really should be called a /slip string/. http://www.av8n.com/how/htm/multi.html#sec-slip-string > As to what I would do with it - I assume you are asking out of > curiosity - it's needed to set the steering in tandam/quadracyle > landing gear aircraft so that they can take-off and land in > cross-winds. For that you need the motion of the airplane relative to the *ground* not relative to the airmass. That's different. That's why I asked what it would be used for. I'm not sure what that quantity should be called. It's not the yaw angle. I reckon it's just the difference between the slip angle and the wind-correction angle. This item you need to calculate for yourself, since it's not in the tree AFAICT. It's trivial to calculate u = /fdm/jsbsim/velocities/u-fps v = /fdm/jsbsim/velocities/v-fps angle = atan2(v, u) ... although (as mentioned above) for most purposes you're better off using the (u,v) vector as a vector, rather than converting it to an angle. ------------------------------------------------------------------------- This SF.net email is sponsored by: Microsoft Defy all challenges. Microsoft(R) Visual Studio 2005. http://clk.atdmt.com/MRT/go/vse0120000070mrt/direct/01/ _______________________________________________ Flightgear-devel mailing list Flightgear-devel@lists.sourceforge.net https://lists.sourceforge.net/lists/listinfo/flightgear-devel
