On 12/20/2008 10:24 AM, Ron Jensen wrote: > Lets look instead at finding out the real reasons why the output > behavior is not as it should be. > > First question, prop_81in2v.xml gives a minimum and maximum propeller > pitch of 12.0 and 31.8. Any idea if these numbers are right or are they > just guesses? > > Second question, prop_81in2v.xml gives a minimum and maximum rpm of 900 > and 2400 rpm. Any idea if these numbers are right or are they just > guesses?
Hmmm ... prop_81in2v.xml is not a Hartzell part number or even close, so I have no idea what RW propeller that file is supposed to represent. Similarly, there is no documentation in the file to tell me. The "locate" command suggests that this file is associated mainly with the SW c182 and c182rg. I have here a RW 1979 Cessna R182 manual that says the propeller has a low pitch of 15.8 and a high pitch of 29.4. Mr. Cessna is kind enough to specify that this is measured at the 30" station (otherwise the numbers would be meaningless). The RPM numbers quoted above are probably about right as to the governor. Full forward on the prop control sets the governor for 2400. I reckon full back might set the governor for somewhere near 900 but I'm not sure. I recall it is impressively low, but I don't recall the exact number. I usually only see this number during emergency glide conditions, and then I usually have other things to look at. Of course the governor numbers are not hard limits; at any prop setting if you pull the throttle back far enough the propeller will drop out of regulation. That's because the aforementioned pitch limits *are* hard limits; the mechanism literally hits the stops. The FGFS model seems to capture this out-of-regulation behavior OK. There is also transient behavior; it is easy to overspeed or underspeed the prop temporarily. ======================== Now let me explain why that's not the right starting point. Two reasons: physics and engineering. You don't need to tell me the propeller and engine interact. I'm pretty sure I knew that already. That's exactly why they should be tested separately. Think about software engineering: We write modules and test them individually. Yeah, they interact, which is why we start by testing them separately, so if there is any funny business we know where to look. We always *end* by testing everything together, but that's not where we start. The same engineering principle applies to hardware. That's why dynamometers and prony brakes were invented. I guarantee you Lycoming tests the engines on a test stand before they go anywhere near a propeller; I've seen the data. (I don't have a copy; sorry.) As to the physics: In the steady state, if we know the torque and the revs, we don't need to know *anything* about the propeller to ascertain engine performance. It doesn't matter whether the engine is connected to a dynamometer or to a propeller or to bunch of pom-poms on a broomstick. You can formalize this in terms of Sturm-Liouville theory if you want. The fundamental equations of physics are low-order differential equations, and they need only a small number of initial conditions and/or boundary conditions. Since it has been "stipulated" that there is misbehavior in the engine, basic engineering principles suggest debugging the engine before allowing it to interact with other subsystems. Otherwise there is jeopardy of ending up with two bugs: in particular, unrealistic prop behavior that compensates and masks some part of the unrealistic engine behavior. ------------------------------------------------------------------------------ _______________________________________________ Flightgear-devel mailing list [email protected] https://lists.sourceforge.net/lists/listinfo/flightgear-devel
