Andy Ross said: > Vivian Meazza wrote: > > Here are some calculations on propeller rpm. > > [...] > > We can see that 2850 is unlikely to be the rpm of a 10.75 diameter > > propeller > > Yeah, you're right. This is a real bug. I was playing with it this > morning, and we're hitting an edge case in the propeller solver. > > The propeller as defined is actually fine. It sinks the right amount > of power and generates appropriate thrust at the specified cruise RPM. > The problem is that (due to a deficiency in the model) the torque > required to turn the propeller are *lower* RPMs increases faster than > the engine torque does*. So while the engine and propeller are > matched at cruise; the combination can't get there because it can't > accelerate the prop at low speeds. > > The end result is that this breaks the "stabilize" step in the solver, > which tries to iteratively solve for the steady state RPM for an > engine/prop before running the aero FDM. Mathematically, the current > propeller model has two minima, and it's picking the wrong one. > > The spitfire is hitting the condition because of the high gear ratio, > recent changes in the engine code which reduce available power at low > speeds (to get idle speeds right), and a miscalbration quirk in the > manual pitch handling (setting "0.5" for manual pitch doesn't produce > the same results as a non-variable propeller). > > I'm not quite sure what the right thing to do here is. One trick > would be to jigger the stabilize routine so it starts from an RPM > within the right range, but that's going to be really hard to > maintain over time. Let me think about it... >
Any ideas on this yet? I caught a little bit of this thread before heading to NY and have been sitting on the edge of my seat ever since :-) It is an issue with the p51-d as well, of course. It'd really be nice to finally get that one right. Best, Jim _______________________________________________ Flightgear-devel mailing list [EMAIL PROTECTED] http://mail.flightgear.org/mailman/listinfo/flightgear-devel
