Re: [Flightgear-devel] basic flight dynamics
On Sat, 2009-12-12 at 07:29 +, Ron Jensen wrote: On Fri, 2009-12-11 at 10:44 -0700, John Denker wrote: But there may also be issues with the prop efficiency at ultra-low airspeed (high blade angle of attack). Low blade angle of attack? Increasing airspeed increases blade AoA assuming RPM is held constant. Sorry, I was thinking about the angle of the relative wind to the prop disk not the individual blade AoA. You are correct, the individual blade AoA is highest with no forward velocity. I've been too deep in propeller theory lately :) Ron -- Return on Information: Google Enterprise Search pays you back Get the facts. http://p.sf.net/sfu/google-dev2dev ___ Flightgear-devel mailing list Flightgear-devel@lists.sourceforge.net https://lists.sourceforge.net/lists/listinfo/flightgear-devel
Re: [Flightgear-devel] basic flight dynamics
On Fri, 11 Dec 2009, John Denker wrote: Anybody who is tempted to look at this is reminded of the output ... /output section at the end of c172p.xml I have sometimes found it useful. Does anybody know if/where this feature is documented? It is documented in section 3.1.11 (starting at page 50) in the JSBSim reference manual: http://jsbsim.sourceforge.net/JSBSimReferenceManual.pdf Cheers, Anders -- --- Anders Gidenstam WWW: http://www.gidenstam.org/FlightGear/ -- Return on Information: Google Enterprise Search pays you back Get the facts. http://p.sf.net/sfu/google-dev2dev ___ Flightgear-devel mailing list Flightgear-devel@lists.sourceforge.net https://lists.sourceforge.net/lists/listinfo/flightgear-devel
Re: [Flightgear-devel] basic flight dynamics
This reduces by an order of magnitude the amount of adverse yaw _in cruising flight_ in the c172p. This is much more realistic. There is still a ton of adverse yaw during slow flight. commit 74e59d6c9fb1eca08fb446c26c7b5d873c45b0ea Author: John Denker j...@av8n.com Date: Fri Dec 11 11:12:08 2009 -0700 Vastly less adverse yaw during cruising flight. (Still plety during slow flight.) diff --git a/Aircraft/c172p/c172p.xml b/Aircraft/c172p/c172p.xml index 0eae36b..d9304b3 100644 --- a/Aircraft/c172p/c172p.xml +++ b/Aircraft/c172p/c172p.xml @@ -879,7 +879,8 @@ propertymetrics/Sw-sqft/property propertymetrics/bw-ft/property propertyfcs/left-aileron-pos-rad/property -value-0.0053/value +value-0.00030/value +propertyaero/alpha-deg/property /product /function function name=aero/coefficient/Cndr -- Return on Information: Google Enterprise Search pays you back Get the facts. http://p.sf.net/sfu/google-dev2dev ___ Flightgear-devel mailing list Flightgear-devel@lists.sourceforge.net https://lists.sourceforge.net/lists/listinfo/flightgear-devel
Re: [Flightgear-devel] basic flight dynamics
On Fri, 2009-12-11 at 10:44 -0700, John Denker wrote: On 12/03/2009 10:18 AM, Stuart Buchanan wrote: I took the opportunity to check the PoH against the simulator experience. While I didn't go as far as getting the OAT exactly right, the errors I came across were fairly signficant (using a HUD to get accurate altitude/TAS etc.) The current c172p is better but still not great. One thing I noticed: With the throttle all the way back: -- In flight at 60 kias : tachometer = 1000 rpm -- Parked, 0 kias : 900 rpm J=V/nD so, In flight J = 60 knot/ ((1000/min)*75 in) ~=0.97 Parked J = 0.0 Cp @ 0 = 0.058, Cp @ 1 = 0.020 E.g., it takes about 3 times more power to turn the prop parked at 900 rpm than in-flight at 60 kcas and 1000 rpm. This tells me the effect of the prop loading the engine is not being modeled correctly. So, given a Cp and RPM we can estimate the required horsepower... P = Cp*rho*n^3*d^5 ((0.00238 slug/ft3) * ( 900/min)^3 * (74 in)^5) * 0.058 ~= 7.6 hp ((0.00238 slug/ft3) * ( 700/min)^3 * (74 in)^5) * 0.058 ~= 3.6 hp ((0.00238 slug/ft3) * (2700/min)^3 * (74 in)^5) * 0.058 ~= 204 hp Since our engine only reaches 160 hp@ 2700 rpm we can't spin 2700 static RPM, we can only reach 2300-2400 rpm. This is in-line with my understanding of the performance of this aircraft. The parked tach reading should be quite a bit lower. Please define quite a bit lower. Also, from a standing start, with the throttle all the way back, the model will accelerate to 30 knots in a few thousand feet. This is wildly unrealistic. I did not observe this in JSBSim stand-alone. I let the sim run for 30 seconds at 900 rpm and only reached 9 knots, after 5 minutes the speed was only 26 knots. Perhaps you had a tail wind or were headed down hill? That said, reducing idle engine power to 4.4 hp yielded a propeller RPM of 730 and a 300 second speed of 0.3 knots. Changing the throttle detent so that the throttle can be more fully closed would help with this, and maybe also with the static tach reading (above). Since thrust is directly related to RPM, yes, reducing RPM will reduce thrust here, as noted above. But there may also be issues with the prop efficiency at ultra-low airspeed (high blade angle of attack). Low blade angle of attack? Increasing airspeed increases blade AoA assuming RPM is held constant. In RL the prop efficiency is markedly reduced under such conditions. Depends on the propeller, actually. Ron -- Return on Information: Google Enterprise Search pays you back Get the facts. http://p.sf.net/sfu/google-dev2dev ___ Flightgear-devel mailing list Flightgear-devel@lists.sourceforge.net https://lists.sourceforge.net/lists/listinfo/flightgear-devel
[Flightgear-devel] basic flight dynamics
On 12/03/2009 10:18 AM, Stuart Buchanan wrote: I took the opportunity to check the PoH against the simulator experience. While I didn't go as far as getting the OAT exactly right, the errors I came across were fairly signficant (using a HUD to get accurate altitude/TAS etc.) As I think you've noted before, the climb rate is too high - I consistently see 1000ft/min up to 8000ft ASL, instead of approx 800ft/min at sea level, and 400ft/m at 8000ft ASL. In contrast, the cruise speed is a bit too low - I don't recall what I saw at sea-level, but at 8000ft ASL, I saw 107 KTAS rather than 120 KTAS (though as that was very close to the IAS, it may be that the environment was not quite right). I'm really glad you're looking into this. I will happily help as much as I can, but I've spent about 200 hours in hospitals in the last three weeks, which cuts into my flying and virtual flying time I'm not sure what to make of this. Being a test pilot is very demanding work. Nothing easy about it. Perhaps the drag and power should be reduced, or possibly the alpha drag needs to increase? A distinct possibility. Suggestion: Try plotting the power curve. Start by measuring the power-off vertical speed as a function of airspeed. The top of the curve should correspond to the POH power-off best-glide speed. If the induced drag (alpha drag) is off, the curve will be wildly off. Finding the exact top of the curve is not easy, but a little curve-fitting helps a lot. The next step is to measure the power-on power curve. In the real airplane, this is a pain in the neck because engine power depends so strongly on altitude. In the sim the pain is slightly less, because you can easily record _thrust_ horsepower (thrust times TAS) as you go along, and take that into account in the analysis. I suggest turning on auto-coordination. The yaw axis behavior in the C172p sim is quite unrealistic, and we don't want that to pollute the power and drag measurements. We can deal with the yaw axis some other time. I suspect I need to use JSBSim directly to tune these parameters better. There's nothing easy about that, either. At one point I started writing a wind tunnel based on JSBSim, i.e. something that would systematically map out the aerodynamic coefficients. But I got a negative amount of cooperation, so I moved on to other things. If there is sufficient interest maybe that effort could be revived. -- Join us December 9, 2009 for the Red Hat Virtual Experience, a free event focused on virtualization and cloud computing. Attend in-depth sessions from your desk. Your couch. Anywhere. http://p.sf.net/sfu/redhat-sfdev2dev ___ Flightgear-devel mailing list Flightgear-devel@lists.sourceforge.net https://lists.sourceforge.net/lists/listinfo/flightgear-devel
