Andy:
One thing to consider is whether the autopilot is optimizing for
the correct solution to the equations. There are always two solutions,
one slower with higher drag and the other faster with lower drag.
Depending on altitude and power, one of these can be below stall speed.
Near the service ceiling, the two solutions may be only a couple of
knots apart and difficult to distinguish by airspeed (have to use pitch).
A popular mistake that both auto- and human- pilots make is to accidentally
select the wrong one, which forces the aircraft into a trim oscillation.
For humans, this is a problem in cruise, where the frequency is slow enough
that the human doesn't figure out what's going on. For autos, it is a
problem at slow speeds, where the frequency is high enough for the
circuit never to notice the second solution and jump over to it.
> Andy Ross <[EMAIL PROTECTED]> said:
>
> > In other news, I think I've got a 2.95.2 build that fails the same way
> > yours does. I'm still at the "yup, it don't work" stage, though. No
> > analysis yet.
> >
> Interesting. Also a few months ago we discussed a power problem at altitude
> with the 747. Around 25000 feet or so the airspeed starts to fall when the
> autopilot is engaged.
>
> But now I'm thinking that the real issue has to do with the lift calculation.
> The pitch angle seems to be way too high when at altitude (not climbing),
> which of course would cause a decrease of airspeed and power. The altitude
> starts to decrease smoothly as IAS passes below 300knots...not in a "stall"
> fashion. This is running with the default half full tanks. Just wondering if
> you're seeing this as well on your 2.95.2 build.
>
> Best,
>
> Jim
>
> _______________________________________________
> Flightgear-devel mailing list
> [EMAIL PROTECTED]
> http://mail.flightgear.org/mailman/listinfo/flightgear-devel
>
>
_______________________________________________
Flightgear-devel mailing list
[EMAIL PROTECTED]
http://mail.flightgear.org/mailman/listinfo/flightgear-devel
