David Megginson wrote: > The YASim C172 flies fairly well. Take-off works, and p-factor is > about the same as in the FLY! C172 -- i.e. you need a lot of right > rudder to keep the nose straight during a power climb. Levelling off > at 3000' ASL, however, I had trouble establishing a cruise above 100kt > at about 85% throttle -- I'd expect 110kt - 125kt, depending on the > engine.
This part is easy. Right now, the Cessna with full throttle at 8000' MSL will get 125 KTAS exactly. I got this number from somewhere, but I can't remember where. Change those parameters (in the <cruise> section of the XML) to 0.85, 3000 and 120-125. What would be best, of course, is finding a POH that lists exact power settings at some reasonable cruise. I didn't do this (does anyone have an online source for POH performance numbers?), and instead copied a bunch of numbers in from different sources. So, in fact, you get about 110 KIAS with full throttle at 3000 feet. > Where YASim differed most was the stall. [...] the left wing > suddenly dropped out from under me and the ground started spinning > towards me at an alarming rate. OK, I got a chance to verify this. The wing dropping is a "feature", but I'm stumped as to why it's happening here. The Cessna is a straight-wing plane (no way to get the different sides at different AoAs), and really shouldn't do this. Well, at least not until asymmetric stuff like prop wash gets modelled. I hadn't noticed it on the cessna before; I'll take a look. > Given that YASim models airflow over surfaces, I expected a more fluid > feel to the controls (i.e. more like a boat than a car), but it felt > much the same as the JSBSim C172 -- the plane responds instantaneously > to all control inputs, as if it had virtual tires firmly in the air. The plane's _orientation_ responds more or less instantly. The actual velocity of course take a while to change. Turn on the hud and watch the velocity vector. It's also worth pointing out that the 172 is a VERY nimble plane. At the airspeeds it travels at, it takes very little force to change the velocity. The other planes, especially the A-4, are significatnly less stable, travel at higher airspeeds, and should feel more fluid, if I understand what you mean. > YASim's gear code, like JSBSim's, has trouble handing wind while on > the ground. While sitting on the ground with a 20kt crosswind and > full brakes, YASim will still weathervane into the wind, while > JSBSim will bounce around with the nose hitting the runway. Yup, verified. This has me a little confused too -- the main wheels certainly should be holding the aircraft steady, even if the nose wheel lacks authority (which it does on the 172 -- there's very little weight over the nose in the current model). > I didn't land the C172 during this review, so I do not know whether > YASim supports ground effect yet. It does, but I'd be curious to see what people think. The 172 doesn't show much effect, being a high wing plane. The 747 definitely sees it, though. One thing you'll notice is that the Cessna has "sticky" landing gear. Currently, the gear don't have tunable bounce/stiffness parameters, so the Cessna gets the same feel as the big jets. Real spring steel is a lot less damped. > The greatest weakness in YASim right now is the powerplant. Unlike > JSBSim, LaRCsim, and FLY!, YASim does not allow you to start the > engine, does not allow you to lean it in flight to get extra power > (leaning just cuts the RPM proportionally to the mixture), and doesn't > give a useful EGT or fuel-flow display. All true, except for the mixture stuff. Leaning cuts fuel flow on an absolute scale. This is linear with setting only at sea level, etc... It's basically correct, except for the fact that it doesn't model power loss at very rich settings. And fuel flow should be working, but isn't. Bug noted. Andy -- Andrew J. Ross NextBus Information Systems Senior Software Engineer Emeryville, CA [EMAIL PROTECTED] http://www.nextbus.com "Men go crazy in conflagrations. They only get better one by one." - Sting (misquoted) _______________________________________________ Flightgear-devel mailing list [EMAIL PROTECTED] http://mail.flightgear.org/mailman/listinfo/flightgear-devel
