On Wednesday 19 Jan 2005 02:24, Curtis L. Olson wrote: > Yasim has a "magic" solver that is sometimes sensitive to specific > inputs. In the back of my head I imagine a little robot trying to climb > to the highest point on the map by always going up ... but then coming > to the top of a smaller hill and getting stuck. > > The solver tunes the lift and drag coefficients to make the aircraft hit > the numbers you specify ... so if you provide engines that are too weak, > you will end up with a super "slick" model which an incredibly efficient > wing ... thus it can still hit the numbers but has really slow > acceleration and climb. On the other end of the spectrum, if you > provide too much power, you end up with a high drag, low lift model (so > you don't blow past the provide performance numbers.) This will give > you great ground acceleration and probably great climb, but will still > top out at whatever numbers you specify. > > So once you have your basic YAsim model flying, you can tune things like > rate of climb by adjusting actual engine output. You can tune > roll/pitch rates by adjusting the size or effectiveness of the control > surfaces. > > I'm not convinced you could get a YASim model close enough in every area > to get FAA level 3 certification or higher, but you can get a really > fine flying model in most regimes with a bit of tweaking and > understanding (at least at a simple level) how various configuration > options relate to each other. > > The other thing that confused me early on was how YAsim handles weight. > I don't remember the rules well enough off the top of my head to > summarize them here, but the solver solves at 80% fuel load I believe. > This means that unless you are very careful with your fuel load and the > weight the solver uses, you won't hit your performance numbers exactly > ... those number only are for one particular aircraft weight. Once you > figure out how to control the weight the solver uses and figure out how > to configure the aircraft at that exact same weight, you do hit the > performance numbers dead on. > > For someone like me with zero aeroengineering background, YAsim is a > *really* fun tool to play around with. After a few hours with it, I > almost feel like I understand it enough to build pretty plausible > numbers. When it comes to stability derivatives and aero coefficients, > I'm still pretty much as clueless as the day I was born. > > Curt.
Thanks for the advice; the B1900D FDM is really coming on now. I've got her flying the 'envelope' and I'm managing to balance out the flight characteristics nicely. Something I noticed early on is that the mass needed distributing for things like Engine+Gearbox sets and Maingear etc as Yasim just evenly places the dry mass otherwise. I do agree that Yasim is great fun to work with - feels like I'm learning a lot. A bit more flight testing and then I will show what I have got and you can all 'shoot me down' ;-) Dave Martin Footnote: It appears that the B1900D props *do not* counter-rotate after all. _______________________________________________ Flightgear-devel mailing list [email protected] http://mail.flightgear.org/mailman/listinfo/flightgear-devel 2f585eeea02e2c79d7b1d8c4963bae2d
