dave perry wrote: > Laurence Vanek wrote: > >> For my money, the Seneca is one of the best done aircraft in FG. Have >> been flying it a good deal of late. >> >> >> > Agreed! > > <snip> > >> [2] On localizer during an ILS approach the A/P seems unstable >> (increasing roll oscillations) when within ~1 mile of approach end of >> runway. Perhaps this requires tweak of A/P controller constants. On a >> CAT II or III approach one would still not be able to see the runway at >> that distance. I usually disable the A/P at DH point. >> >> > I have commented before about the difference between yasim and jsbsim > when it comes to two autopilot issues. > 1. jsbsim becomes almost uncontrolable with turbulence. Even at 0.1 or > certainly at 0.2 turbulence, the c172p will not complete an approach > with the kap140. So, make sure you have turbulence at 0. > 2. I spent many many hours tweaking the SenecaII autopilot config > file. Since jsbsim (in my opinion) has a very exagerated adverse > aileron yaw response and most autopilots have aileron but not rudder > roll outputs, this creates the oscillations. That is why I put in the > menu options for the SenecaII a yaw-damper that may help some. It is a > controller that applies rudder to try and keep the ball centered. > > To see the fdm difference, fly the Senecal II and then the pa24-250 > (which has the same autopilot nasal file and very similar autopilot > config file) down the same ILS. The pa24-250 with the Century III > autopilot will be rock solid all the way to the runway center line even > with light to moderate turbulence. The pa24 uses yasim. You can see > the same comparison between the jsbsim c172p and the yasim pa28-161 > which both have the kap140. > > If you try to tweak the SenecaII/Systems/CENTURYIII.xml parameters, you > must remember that there are cascaded controllers in this file. Note > the input and output for each stage: > > Stage > Input > Output Reference > Approach hold heading-needle-deflection > target-intercept-angle 0.0 > heading-bug-filter heading-bug-error-deg > filtered-heading-bug-error-deg > Heading Select filtered-heading-bug-error-deg > target-roll-deg target-intercept-angle > target-roll-filter target-roll-deg > filtered-target-roll-deg > Wing Leveler indicated-roll-deg > aileron filtered-target-roll-deg > > Does anyone know of a jsbsim model with autopilot that will fly an ILS > with no oscillation with even very light turbulence? > > -Dave Perry > Thanks Dave for the advice/comments.
It may be that my approach is a tad slow also which may be contributing to the problem (Im a bit below Vref). I have turb set to zero in my ~/.fgfsrc file. Jsbsim seems to fly with very lite damping. Im not a real pilot so have no real world ref. point to base that on for the Seneca. I try the yaw damper. Any input on the heading bug disappearance when its in the top portion of the HSI? ------------------------------------------------------------------------- SF.Net email is sponsored by: The Future of Linux Business White Paper from Novell. From the desktop to the data center, Linux is going mainstream. Let it simplify your IT future. http://altfarm.mediaplex.com/ad/ck/8857-50307-18918-4 _______________________________________________ Flightgear-users mailing list Flightgear-users@lists.sourceforge.net https://lists.sourceforge.net/lists/listinfo/flightgear-users
