>>>>Perhaps people have tired of waiting for ACES to address and communicate about some of the technical issues with the air files?<<<

>>>>It's 'Cm_dT'. 'T' for Thrust. Cm_dt is pitch moment due to trim. Had to check an AIR file to verify my nomenclature.<<<

Hi Ron,Thanx for all the info.Keep posting!!! We need MORE expert knowledge around here...Bert

>Hi Ron,>Thanx for all the info.>Keep posting!!! We need MORE expert knowledge around here...>Bert I usually post on flight dynamics in the AC and Panel forum, but sometimes more technical threads come up in the general MSFS forum(s). I've even posted in the sci.aerodynamics newsgroup, but not for some time. Too much to keep up with. I had commented on the backwards wind arrow and wrong TAS indication in a couple of the FSX jet panels. I now see they appear to be in XML, so should be easy to fix. Ron

This is extremely important info... While most will not fully understand it. What it does is raise several flags that tell the non-pilots that the reason they bought the FLIGHT SIMULATOR may not as faithful to reality as marketing tells them. For those who are eye-candy oriented the program is more than adequate.My problem is that one must have faith in the flight simulation aspect to be able to truly suspend themselves in the imaginary aircraft. Once they understand that the imaginary craft is somewhat removed from reality it becomes difficult to convince yourself that you are actually flying and not just "moving" above a quite good rendition of the surface of the Earth.I recall similar posts early in FS2004/FS9. Eventually those who focus on the flight dynamics of the program sorted things out to a level that made "some" of the aircraft closer to reality. It is a shame that while Microsoft takes great pride in the visuals they seem to continually ingore the essence of the program - FLIGHT.One other note - write down the names of people who are posting knowledgeable info here. Then ONLY "fly" airplanes that they have worked on IF you want a more true-to-life simulation.Dick Boley

>>>>>I see the Beaver is set at -0.0512. It's a FP variable, so>there is no question of scaling. >Some day I'll try to check out large values.<<<<>>You absolutely should look into it, since using a larger value>for Cm_dT (I remembered the shift key that time :))>will let you raise the nose on the Skyhawk or Skylane early in>the takeoff roll. That same capability was there in the FS2000>182, but disappeared in FS2002 unless higher numbers were>used. While "Prop Gyro effect" will yaw the nose laterally way when you pull up rapidly, the other way when you push down. A turn should move the nose up in one turn direction, down in the other. I've seen that in SEL's. People often move the 'realism' sliders down to get what they think is are realistic effects. Full Right should give just the right amount. However, that may be too much when the Prop MoI, etc. are set too high. Similarly for 'JS sensitivity'. I see no reason for registered FSUIPC to adjust JS effects. If the JS is calibrated (normally, that's done automatically) the controls surfaces and effects should be just right for the AC. Assuming they are set appropriately. ;) Control sensitivity is about double a reasonable amount in FSX. Users might reduce the sensitivities in the aircraft.cfg files. I set elevator, etc. moments so I can pull +3 G at normal air speeds. More at high speeds. That's enough for most AC. The Extra needs more. That makes it twitchy. A good JS would help, but I've never seen one that felt good to me.>>>>>I'd thought 'prop torque' had disappeared in more recent>MSFS versions, but someone (you?) said it still exists. One>has to fly pretty slow to see the effect, but I did find it,>at least in FS9.<<<<>>Yes, it can be set in aircraft,cfg; see

> It is a shame that while Microsoft takes great pride>in the visuals they seem to continually ingore the essence of>the program - FLIGHT.For starters, you can take any desktop flight simulation and pick apart specific issues of flight dynamics. The same happens with X-Plane, which is a program that's hyped marketing wise, as being the ultimate, yet has a large laundry list of problems, and as with MSFS, probably always will. As a program, MS has succeeded quite well in providing the "feel" of flight, and FSX is even better than previous versions for creating a simulated changing airmass. At the moment, out of all desktop programs that I own, it's FSX that tops the list for the "feel" and movement of actual flight. Therefore, "flight" is far from being ignored.What's being talked about here, are specific issues such as torque effects, braking distances, etc, which by the way can have many variables between the same types of aircraft, and doesn't take away with the overall rendition of flight. For instance, you can take a simple tailwheel aircraft, and have varied reactions with rudder stearing depending on the brand of tailwheel & it's spring linkages. Just changing props on the same airplane can make drastic differences in performance and feel. Some airplanes require quite a bit of rudder in turns to correct for adverse yaw, while others require very little. Same goes for varying amounts of up elevator while turning. These are examples of a specific issues, for a specific airplane. You could even fly a real airplane, thats out of somewhat out of rig, compared to the same model sitting next to it, and requires some constant rudder pedal in flight. Just the bends in a hand adjustable rudder trim plate can make a difference between the two.Since it wouldn't be good practice, to think that you could learn total control of a tailwheel aircraft from use of a desktop simulation, it's not critical that the simulated program is exactly precise. As another example, I have yet to see a simulator that depicts the braking action of a constant speed prop as well as in real life. Yet within the simulation I deal with the problem, and treat descents more along the lines of a fixed pitch. It's once again a specific issue and doesn't detract from the overall feel. >>One other note - write down the names of people who are>posting knowledgeable info here. Then ONLY "fly" airplanes>that they have worked on IF you want a more true-to-life>simulation.>These and a few others...L.Adamson

>>>>People often move the 'realism' sliders down to get what they think is are realistic effects. Full Right should give just the right amount. However, that may be too much when the Prop MoI, etc. are set too high.<<<

Our comments are getting too long! I'll reply to only a part right now. >>>>>People often move the 'realism' sliders down to get what>they think is are realistic effects. Full Right should give>just the right amount. However, that may be too much when the>Prop MoI, etc. are set too high.<<<<>> ............>those conditions, so that

>>Personally, having the viewpoint moving around on me makes me>feel like I

>For starters, you can take any desktop flight simulation and>pick apart specific issues of flight dynamics. The same>happens with X-Plane, which is a program that's hyped>marketing wise, as being the ultimate, yet has a large laundry>list of problems, and as with MSFS, probably always will. .............. There is a note about different kinds of flight models in some JSBSim location. JSBSim is used in FlightGear. I think the note was by Jon Bernet, an aeronautical engineer in Houston. He said most of the engineering simulators use lookup tables, rather than try to calculate aerodynamics from a few basic equations. JSBSim and MSFS both use the 'engineering approach'. That means one has to understand a lot of details to get them realistic. However, one can get the basic things pretty good. Simulators such as X-Plane may give a fairly good simulation from setting a few parameters, but can't be adjusted to get close aircraft performance over a range of conditions.>As another example, I have yet to see a simulator that depicts>the braking action of a constant speed prop as well as in real>life. Yet within the simulation I deal with the problem, and>treat descents more along the lines of a fixed pitch. It's>once again a specific issue and doesn't detract from the>overall feel. >L.Adamson That lack of braking is due to a 'bug' in the FS prop model. Prop efficiency is wrong in the windmilling region, and the engine friction doesn't load the prop, thus increase the drag, as much as it should. I think one or two lines of code could fix the bug, but I doubt Aces is even aware of it. I've adjusted the Prop Efficiency table to increase the drag, but there aren't enough points available to do it as I'd like. Ron

> That lack of braking is due to a 'bug' in the FS prop model.>Prop efficiency is wrong in the windmilling region, and the>engine friction doesn't load the prop, thus increase the drag,>as much as it should.>> I think one or two lines of code could fix the bug, but I>doubt Aces is even aware of it.>> I've adjusted the Prop Efficiency table to increase the>drag, but there aren't enough points available to do it as I'd>like. >I've known that it's been a pain, and learned to live with it. :)'cause those real RV's can slow down pretty darn fast in the pattern with a C/S! :-hah And I agree.....Long live CFS2's flight dynamics! Not perfect, but the best yet! :D L.Adamson

Ron, Would you be kind enough to let us know how to make that fix? Craig

> Ron,>> Would you be kind enough to let us know how to make that fix?> Craig Well, the small amount of code to fix the low prop drag would have to be in the sim1.dll source. Only MS can change that. The problem is that prop efficiency is always defined as Pair/Pshaft in the prop code. Efficiency runs about 80% in normal flight and is set by the Prop Efficiency table (TBL511)in the AIR file. When windmilling, Pair/Pshaft should be greater than 1.0, since there is more power absorbed from the air than gets transmitted to the shaft. Engine friction absorbs power from the shaft, thus from the air. The prop table shows prop efficiency dropping as the windmilling condition is approached, then continuing to drop when it's in the windmilling condition. Again, that's Pair/Pshaft; which should be greater than 1.0 in this condition. Since Pshaft/Pair has to be less than one or the prop would act as a windmill with over 100% efficiency. The other prop table (TBL512) sets Cp, the Coefficient of Power. When Cp goes through zero (which it does in the default tables), Power flows into the engine shaft from the prop, rather than the normal direction. I've tried to change the prop efficiency table by making Ceta (efficiency coefficient) about 1/Ceta before the change. That is, if Ceta were 0.4 in the windmilling region, I'd make it 1/0.4 = 2.5. Though, I've found I shouldn't make it increase too much. I've also bunched the data points for the prop tables together in this region, since I don't want prop efficiency increasing before Cp goes negative. Engine friction torque also has to be reasonable, since it absorbs the power. But, too much slows down the prop in the windmilling mode. I think default AIR files have way too much friction torque. So, it's kind of tricky, and one needs a test gauge that shows prop coefficients, Shaft HP, and Prop output power. Herve' Sors' AFSD does this. In fact, I used AFSD to figure out what was happening. I shows more power delivered to the shaft than absorbed from the air flowing past the prop with normal prop tables. One might best cut the engine ignition for initial testing, otherwise the idle power might be too high. When the approach works I see approximately what I want, and with the engine idling. So, not such an easy task. I've mentioned it to a few others but don't think anyone tried to implement it. ;) BTW, AFSD's powerplant window also shows how prop efficiency drops with icing. And, if prop deice is working. 'Boot deice' breaks the ice away right away, but has to be applied again as ice collects again. Heating the prop takes some time to melt the ice, but keeps it off, you don't recycle that kind of deice. I'd never understand what was happening without something like AFSD. Ron

Well it's a bit over my head on the first read. I just downloaded afsd_244.zip and will take a look when wide awake. Hopefully after reading your post a few dozen times I'll be able to sort it out. I was hoping it would be something a bit easier such as drag scalars in the cfg file. At any rate thank you. I'm sure it will give me a challenge and provide for a lot of headscratching. Craig