On Wednesday 14 November 2001 09:06 am, you wrote:
> Hi,
>
> I already whined on fgfs-user about ^U not working since about two weeks.
> Now I've looked further into that problem, still without solution, but at
> least with some details. (Of course, I have updated plib, SimGear,
> FlightGear and FlightGear-Data from cvs before, this time even with "cvs up
> -A"! My copy should therefore be an exact copy of the CVS data. Furthermore
> I did all tests with and without using my local settings in ~/.fgfsrc,
> without any differences.)
>    I found out, that it's in fact not a problem with the keyboard settings,
> as ^U does still work with "magic carpet" and "larcsim". Furthermore,
> not only changing the altitude in JSBSim via ^U, but also via the new
> property-picker and via telnet yields the same results:
>    When started from KSFO every external manipulation of the altitude
> places the aircraft in the sea at 5ft AGL. The following snippet from
> the logs shows something strange: Although I am at 89 feet altitude
> and the altitude is correctly passed to FGJSBSim::setAltitude, pressing
> ^U triggers a =Ground=(!) trim, which does of course not have a
> desirable effect on the position:
>
>
>   find_props: allocation of node
>   files->2 longitude-deg = '-122.363764' (double)
>   files->3 latitude-deg = '37.617349' (double)
>   files->4 altitude-ft = '89.868794' (double)
>   files pointer=1159395616
>   FGJSBsim::set_Altitude: 1089.87
>     lat (deg) = 37.6173
>     Ground Trim
>       Initial Theta: 2.535
>
>     Trim successful
>
>     JSBSim State
>       Weight:    2036 lbs.  CG:  42.2,   0.0,  40.3 inches
>       Flaps:  Up  Gear: Down
>       Speed:    0 KCAS  Mach:  0.00
>       Altitude:       5 ft.  AGL Altitude:       5 ft.
>       Angle of Attack:   0.00 deg  Pitch Angle:   3.11 deg
>       Flight Path Angle:   0.00 deg  Climb Rate:    -0 ft/min
>       Normal Load Factor: 0.00 g's  Pitch Rate:  0.46 deg/s
>       Heading: 298 deg true  Sideslip:  0.00 deg
>       Bank Angle:  0.00 deg
>       Elevator:  0.00 deg  Left Aileron:  0.00 deg  Rudder:  0.00 deg
>       Throttle: 100.00%
>       Wind Components:  0.00 kts head wind,  0.00 kts cross wind
>       Ground Speed:    0 knots , Ground Track:   0 deg true
>       Trim complete
>   find_props: allocation of node
>   files->2 longitude-deg = '-122.357197' (double)
>   files->3 latitude-deg = '37.617382' (double)
>   files->4 altitude-ft = '5.103165' (double)
>
>
>
>
> I had a look at JSBSim.cxx and put a SG_LOG line into FGJSBsim::do_trim to
> show if fgic->GetVcalibratedKtsIC() is involved. It is! This value was
> always reported to be zero, so it's no wonder that I always get a "Ground
> Trim".

The fgic object always remembers that last state jsbsim was initialized to.
Since the last state you initialized to was on the ground at KSFO with 0 
speed, that's how it reinitializes JSBSim when Ctrl-U is pressed.  

What needs to happen is the code that handles Ctrl-U should reinitialize
the sim at the last flying state but with the new altitude.
 

>
>
>
>   SG_LOG(SG_FLIGHT,SG_INFO,"FGJSBsim::do_trim: " <<
> fgic->GetVcalibratedKtsIC() ); if(fgic->GetVcalibratedKtsIC() < 10 ) {
>       fgic->SetVcalibratedKtsIC(0.0);
>       fgtrim=new FGTrim(fdmex,fgic,tGround);
>   } else {
>       fgtrim=new FGTrim(fdmex,fgic,tLongitudinal);
>   }
>
>
> Why does a Ground Trim also change the longitude?
> Why is fgic->GetVcalibratedKtsIC() always zero?
> Can someone familiar with JSBSim imagine what's wrong (and why it could
> just happen on my installation, although this is without any doubt a 1:1
> copy from CVS)?
> What else can I do to find the problem? (I'm not really familiar with
> JSBSim.)
>
> m.   :-)
>
>
>
> PS: I'm running fgfs (CVS HEAD from today) under a Linux 2.4.14 kernel.
> Everything is compiled with a (known as non-buggy) gcc 2.95.2 on a SuSE
> 7.1.
>
>
> _______________________________________________
> Flightgear-devel mailing list
> [EMAIL PROTECTED]
> http://mail.flightgear.org/mailman/listinfo/flightgear-devel

-- 
Tony Peden
[EMAIL PROTECTED]
We all know Linux is great ... it does infinite loops in 5 seconds. 
-- attributed to Linus Torvalds

_______________________________________________
Flightgear-devel mailing list
[EMAIL PROTECTED]
http://mail.flightgear.org/mailman/listinfo/flightgear-devel

Reply via email to