Re: [Flightgear-devel] RFD: JSBSim Cessna 172p stability patch
Jon Berndt wrote: Thanks, David. There are a couple of things I can think of to do, here. One is that I sure wish I had time to make a DATCOM model of the C-172 that could give me some aero data for comparison. But with my schedule now I can't make any promises, but I will get to this someday! The second comment is that it ought to be possible to get some real numbers for you pretty quickly from comparable aircraft. I think I can do that tonight or tomorrow. Third, there are some equations that ought to shed some light on things. Fourth, it will be interesting to see if pilot perceptions suggest altering all/most coefficients in the same direction for all aircraft in order to give the perception of proper flight characteristics. That is a tricky issue, because using spring-loaded controllers gives a significantly different feel than fully-loaded aircraft controls, and there is always a danger of altering the flight characteristics to compensate for the control differences. To try to avoid that problem this time, I tried to concentrate on rates (how fast the plane could bank, pitch, and yaw), coupling (how much adverse yaw a particular bank causes and how much bank a particular yaw causes), and recovery (how the plane recovers from a yaw or pitch disturbance). I deliberately ignored control feel. The model I posted seems a lot closer to what I remember of the 172 and what I observe in my PA-28, but unfortunately, I do not currently have measurements to test it against. All the best, David ___ Flightgear-devel mailing list [EMAIL PROTECTED] http://mail.flightgear.org/mailman/listinfo/flightgear-devel
Re: [Flightgear-devel] RFD: JSBSim Cessna 172p stability patch
Jon Berndt wrote: Given these numbers I'd suspect that if there is a problem, perhaps we need to review our MoI's. That makes a lot of sense -- I was worried that the damping numbers were masking a different problem. All the best, David ___ Flightgear-devel mailing list [EMAIL PROTECTED] http://mail.flightgear.org/mailman/listinfo/flightgear-devel
RE: [Flightgear-devel] RFD: JSBSim Cessna 172p stability patch
That is a tricky issue, because using spring-loaded controllers gives a significantly different feel than fully-loaded aircraft controls, and there is always a danger of altering the flight characteristics to compensate for the control differences. FWIW, I sent an email to Cessna customer support last night asking them about the availability of some aero and mass props data for their C-172 - or at least Ixx and Clp. I wonder if I will even hear back. To try to avoid that problem this time, I tried to concentrate on rates (how fast the plane could bank, pitch, and yaw), coupling (how much adverse yaw a particular bank causes and how much bank a particular yaw causes), and recovery (how the plane recovers from a yaw or pitch disturbance). I deliberately ignored control feel. The model I posted seems a lot closer to what I remember of the 172 and what I observe in my PA-28, but unfortunately, I do not currently have measurements to test it against. I have data for the PA-28 somewhere. I ought to validate the PA-28 JSBSim mode we have and see how they all compare. Jon ___ Flightgear-devel mailing list [EMAIL PROTECTED] http://mail.flightgear.org/mailman/listinfo/flightgear-devel
Re: [Flightgear-devel] RFD: JSBSim Cessna 172p stability patch
On Thu, 2004-05-20 at 05:17, David Megginson wrote: Jon Berndt wrote: Thanks, David. There are a couple of things I can think of to do, here. One is that I sure wish I had time to make a DATCOM model of the C-172 that could give me some aero data for comparison. But with my schedule now I can't make any promises, but I will get to this someday! The second comment is that it ought to be possible to get some real numbers for you pretty quickly from comparable aircraft. I think I can do that tonight or tomorrow. Third, there are some equations that ought to shed some light on things. Fourth, it will be interesting to see if pilot perceptions suggest altering all/most coefficients in the same direction for all aircraft in order to give the perception of proper flight characteristics. That is a tricky issue, because using spring-loaded controllers gives a significantly different feel than fully-loaded aircraft controls, and there is always a danger of altering the flight characteristics to compensate for the control differences. To try to avoid that problem this time, I tried to concentrate on rates (how fast the plane could bank, pitch, and yaw), coupling (how much adverse yaw a particular bank causes and how much bank a particular yaw causes), and recovery (how the plane recovers from a yaw or pitch disturbance). I deliberately ignored control feel. The model I posted seems a lot closer to what I remember of the 172 and what I observe in my PA-28, but unfortunately, I do not currently have measurements to test it against. Another good way to take the controls out of it are with the Dutch roll and phugoid characteristics. Once the initial rudder and elevator inputs, respectively, are complete the aircraft response should be about the same. All the best, David ___ Flightgear-devel mailing list [EMAIL PROTECTED] http://mail.flightgear.org/mailman/listinfo/flightgear-devel -- Tony Peden [EMAIL PROTECTED] ___ Flightgear-devel mailing list [EMAIL PROTECTED] http://mail.flightgear.org/mailman/listinfo/flightgear-devel
RE: [Flightgear-devel] RFD: JSBSim Cessna 172p stability patch
Coming back to the (default) JSBSim Cessna 172p after spending a couple of months flying the YASim pa28-161, I find the 172 extremely slippery, far too much for a trainer. While the Cherokee tends to feel more stable in flight than a 172 (i.e. it has more roll, pitch, and yaw damping), the 172 still shouldn't be nearly so light in roll, pitch, or especially yaw as it currently is. I've made a patched-up file that allows the 172p to handle much more realistically, but I'm not willing to upload it to CVS yet, because I'm not sure that I've done the right thing. To keep the 172 from wallowing, I increased the roll damping coefficient (Clp) from -0.484 to -1.2, the pitch damping coefficient (Cmq) from -12.4 to -15, and the yaw damping coefficient from -0.0937 to -0.2. The patched file ($FG_ROOT/Aircraft/c172p/c172p.xml) is available at http://www.megginson.com/Aviation/c172p.xml I don't want to commit this file to CVS yet for a couple of reasons: 1. Increasing the damping may simply mask a different problem, such as a unit mismatch or mistaken value for another coefficient (the biggest problem comes around yaw/roll coupling) -- someone who understands aerodynamics better needs to look at the actual moments generated by each coefficient at runtime to see if they are out of whack. 2. I have over 200 hours in a Piper since I last sat in a Cessna 172, so I don't know how accurate my memory is of its handling -- I remember that it felt a bit less stable than a Cherokee, but as a trainer, it still had a *lot* of damping compared to a high-performance aircraft. My revised file seems to fly much better to me, but I'd like to hear opinions from other people with real experience in 172's, as well as the aerodynamic engineering types. I'd also be interested in feedback from non-pilots about the handling difference. Thanks, David. There are a couple of things I can think of to do, here. One is that I sure wish I had time to make a DATCOM model of the C-172 that could give me some aero data for comparison. But with my schedule now I can't make any promises, but I will get to this someday! The second comment is that it ought to be possible to get some real numbers for you pretty quickly from comparable aircraft. I think I can do that tonight or tomorrow. Third, there are some equations that ought to shed some light on things. Fourth, it will be interesting to see if pilot perceptions suggest altering all/most coefficients in the same direction for all aircraft in order to give the perception of proper flight characteristics. Jon ___ Flightgear-devel mailing list [EMAIL PROTECTED] http://mail.flightgear.org/mailman/listinfo/flightgear-devel
RE: [Flightgear-devel] RFD: JSBSim Cessna 172p stability patch
For comparison: I've made a patched-up file that allows the 172p to handle much more realistically, but I'm not willing to upload it to CVS yet, because I'm not sure that I've done the right thing. To keep the 172 from wallowing, I increased the roll damping coefficient (Clp) from -0.484 to -1.2, the pitch damping coefficient (Cmq) from -12.4 to -15, and the yaw damping coefficient from -0.0937 to -0.2. The patched file ($FG_ROOT/Aircraft/c172p/c172p.xml) is available at The Navion has Clp = -0.410 Cmq = -9.960 Cnr = -0.125 The Cherokee has Clp = -0.429 Cmq = ?? Cnr = -0.0873 Equations from Nicolai (Royal Aeronautical Society) has: Clp = -0.350 My thinking is that given this, it appears to be unlikely that Clp should go from -0.484 (a seemingly reasonable number) to over twice that. Further, the other numbers seem OK, too. Jon ___ Flightgear-devel mailing list [EMAIL PROTECTED] http://mail.flightgear.org/mailman/listinfo/flightgear-devel
RE: [Flightgear-devel] RFD: JSBSim Cessna 172p stability patch
The Navion has Clp = -0.410 Cmq = -9.960 Cnr = -0.125 ... Given these numbers I'd suspect that if there is a problem, perhaps we need to review our MoI's. Jon ___ Flightgear-devel mailing list [EMAIL PROTECTED] http://mail.flightgear.org/mailman/listinfo/flightgear-devel
