Wacky time in Flight Dynamics land

[Guest]

(Read Flight Dynamics Discovery thread way down there somewhere for background)None of this stuff is as straightforward as it looks. Changing the elevator angle of incidence from + to - made my FS2K 172 magically start flying properly. I looked at Ron's SP air file and his wing was set to +1.4 and his tail to +2.0. If flew about the same, go figure.I decided to try and duck the issue as he suggested and set the tail to 0.0. Then I fiddled with the wing angle until the deck angle of the plane looked like the real one in full power cruise. That turned out to be 3.0 degrees. A measurement of a plan of the 172 from the POH shows that the difference between the tail and wing should be 4 to 5 degrees. Anyway...On to glide test. Woah! Power off glide angle now about half what the POH shows. The only thing that should have changed is the angle of attack of the fusalage. Clearly, Wing Angle of Incidence and Tail Angle of Incidence are either mislabled or the flight model makes some very strange assumptions. Thrust line isn't an issue in this test because there isn't any.Tried bumping up Induced drag. No result. Bumped up Parasite drag. Nada. Began to wonder if the variable in the air file was actually read so I put a 2 in front of it to raised it from the 2000 range to the 22000 range. That made a bit of difference. Change Induced drag scalar from 1.0 to 2.0 in the cfg file and the plane suddenly was right on the glide slope. (I'm using one measured on a test of a real plane which shows better performance than the POH. Some thing the lawyers had a hand in that part of the POH.) Wow, I've now bumped up parasite drag about 10% and Induced Drag about 2200% on a flight model that worked really well, all as a result of a couple degrees change in wing angle that shouldn't have made much difference.Fiddled with prop and engine until I got a good match with POH speed at 2000 feed. It's much closer over the range of RPM than I ever had before.I'll fly it a bit tonight to check the handling and feel. It should be ready to pack up and upload soon.I'm having the library clean out a lot of my old files. If you want some of the obscure stuff like the girls for the back of the FS98 182, go get'em soon.

[Geofa 108]

I have found much the same.I had a fm in fs2000 that was perfect except it was 10 knts. too fast. Put it on fs2002 and it was way underpowered. I havetweaked this fm almost just the way I want it now in fs2002 except it climbs at 2000 ft. a minute. If you know a magic way to reduce this without mucking everything else up I'll take the hint-every time I try something it screws something else up ! :-)http://members.telocity.com/~geof43/geofanim2.gif

[Johan_Dees 84]

Indeed wacky time. I found that good models in FS2000 work differently in FS2002. We have to start all over again, and I am not sure where to start somtimes.Thats why Ron has created a good nickname for them.I follow your messages with great interrest, and let us mail again.Might have someting for you.Johan[A HREF=http://www.phoenix-simulation.co.uk]Phoenix Simulation Software[/A]Unofficial PSS website:www.people.zeelandnet.nl/johd

[Guest]

That's an easy one (maybe). Sometimes when an FS2K plane is run for the first time, FS incorrectly writes the wrong prop type into all the air file info it puts in the aircraft.cfg.Find Propeller_type= and make sure it is 1 for fixed pitch or 0 for constant speed.

[Guest Ron Freimuth]

>(Read Flight Dynamics Discovery thread way down there >somewhere for background) >>None of this stuff is as straightforward as it looks. >Changing the elevator angle of incidence from + to - made my >FS2K 172 magically start flying properly The MSFS people simply don't know what they are doing. Everything suggests they don't understand how to model AC and don't understand the flight model they supposedly purchased. Thus, there is little hope MSFS will get better. Based in experience with FS2K2, their supposed plans for FS2K4 to rewrite the flight model can only mess it up more. The CFS people seem to understand things a bit better. And, if MSFS is bad, X-Plane is really full of bugs. AC don't show the effects claimed. I see few if any transonic effects, and the Wing construction dump shows Austin uses a theoretical model for Mach effects that is far from realistic. His 'Blade Element' prop is only half a model. I noted prop efficiency increased without limit as Blade angle increased (RPM was pulled down). I noted the EGT gauge in the X-Plane Cessna seemed to move appropriately. Trouble is, it moved the same with the ignition off! The X-Plane Concorde wouldn't even take off with anywhere near rated fuel. That was improved in V 6.13, but it's still far from realistic. I didn't have any trouble gitting it to FL 700 with a lot of fuel and payload. The real Concorde would only be able to climb to about FL 550 under those conditions. The autopilot was also unstable during part of the Concorde's climb. However, X-Plane is good for some things, and generally runs smoothly. It's just too much for mainly one person to do a good flight simulator. At least Austin keeps working on X-Plane and making UG's available. Fly! I/II don't model Mach effects either, in contrast to X-Plane, it appears to take a lot of random screwing around to get reasonable AC for it. Such as Rob Young did. Fly! has many superior features, but overall, has too many weaknesses. One guy contacted me complaining of a Janes Combat sim. He wanted more than generic AC dynamics. I suggested some ways to use CFS2 AIR file in the Janes cfg file and he said he managed to improve the flight models. Thus, NO ONE in a Flight Sim Company seems to know what they are doing. I bet the guys who set up the professional simulators burn up a Million $ getting a reasonable 'flight model' for each AC variation. And, even they don't model everthing correctly. Me? I'm still learning. Ron PS: I still claim a good flight model in FS2K converts well in FS2K2. Other than the Oswald Efficiency probably should be set to what is assumed in the AIR file. As you noted, the prop type can get set incorrectly. One might need to change the '1800 fpm' set for most or all AC in the [autopilot] section of aircraft.cfg. Of course, the autopilot is now a feeble toy. Wade Chafee said he programmed a new autopilot for PIC's FS2K2 update. While it appears the PSS Airbus has independent EPR coding. The DF archer had to set some load currents in the gauges, since the lines in aircraft.cfg have no effect. Oil Pressure was modified so it didn't fall too low on idle -- DF wanted it to read as the real Archer's some beta testers flew. Small details. It's things like the VC staying where you set it are very nice. MS may have thought of that, but "didn't have the time" to implement it. ;)

[Guest]

oh for #### sake!please allow me a brief interjection of temperment.go buy a C64 and run Fsim that way ... come on! Saying they haven't got it perfect is fine, saying they don't know what they're doing, blah blah, rant, rant -- YOU BOUGHT A $70 GAME. Of course it doesn't have a perfectly real flight model, but come on guys.I'm so sick of having a ball in FS2002 only to come here and read nothing but criticisms...Geez, man alive, am I the only one having fun with this? Am I the only one who used to fly "green sticks" and knows how ####'ing amazing this $70 program is?????michael in rare M$ defense mode...

[Avcomware 1]

Hi. Thank you guys for trying to improve the Flight models, and I hope you can ignore some of the comments of some of the people that seem to be FORCED into reading a valuable post. I