Re: Rules of thumb...

Lee Elliott wrote:
> I found that by lowering it from the typical cruise speed of 431kt @
> 36000ft to 300kt @ 1000ft has improved the take-offs no end

Yeah, 431 knots with a prop this size is a *really* high pitch (most
aircraft would call this "feathered").  Placing the peak efficiency
there means that lower pitches (i.e. takeoffs) are going to land on a
much less favorable part of the efficiency curve.

With a constant speed prop, you have a lot of latitude here.  The
governor will find the best pitch for you in most cases.  The code is
written from an energy balance principle anyway, so you can't get
non-physical behavior.  The worst* failure mode of a bad propeller
definition is one that is slightly more efficient (or much *less*
efficient) than it should be.

* Well, no.  If you lie about the power numbers, you can end up
  bolting a Cessna prop to a giant turbine (or vice versa); the engine
  will spin to ridiculous RPMs and the solver will freak out.  But the
  power value is a really easy number to get right.

> and it's also allowed me to reduce the camber on the wing which was
> causing sudden pitch-downs at low altitude/low weights/high-speeds.

This has been a long-standing bug.  I *think* it's in Surface.cpp, and
happens due to a discontinuity in the lift curve at zero (?) AoA.
That theory matches all the reports, anyway.  But I've stared at the
code repeatedly and can't find it.  If anyone gets a simple
configuration that shows this effect really well, let me know.

> However, I've not been able to get a config working if I specify a
> condition lever anywhere and I imagine that this points to a serious
> problem in the config.

This feature isn't mine. :) But reading it, it looks pretty simple; it
just modulates the idle speed of the engine.  It shouldn't make any
difference to the output at full throttle.

I'm equally unsure of exactly what the condition lever is supposed to
do, and I suspect it's very engine-dependent; this implementation
looks safe, though.

Andy

_______________________________________________
Flightgear-flightmodel mailing list
[email protected]
http://mail.flightgear.org/mailman/listinfo/flightgear-flightmodel
2f585eeea02e2c79d7b1d8c4963bae2d