Hi All,
A few quick questions regarding flight dynamics...
I'm trying to shorten the ground roll of Marcel Kuhnt's Islander. Every other aspect of the Flight Dynamics seems fine, but the ground roll at max weight is around 40% longer than the real spec. (The manufacturer's specs are also for max weight).
She'll fly off happily at 50ish knots if you do a touch and go and keep the nose off the ground, but with the nosegear on the ground she won't rotate till over 65.
1) Rotation speed: The tailplane incidence angle is set to 0.0. Would changing the tailplane incidence affect rotation speed noticably?
2) Trim: The rotation speed with full up elevator is also affected by the trim. This may be the case on airliners where the trim is actually varying the tailplane incidence, but with simple trim tabs there should be virtually no effect. Can this be adjusted?
Thanks for any assistance you can provide.
Cheers,
--Graeme
Flight Dynamics Question
Moderators: Guru's, The Ministry
Hi Graeme,
I can't agree that "every other aspect of the Flight Dynamics seems fine" with MK's Islander: many of the aerodynamic derivatives are significantly out and the pitch trim goes the wrong way. Also, the centre of gravity is misplaced by about 1 ft, an astonishingly large error which degrades the handling considerably! These comments are based on the FDEs I downloaded (from the MK site) in Aug 2004, and further changes may have been made since then.
If you want, I could send you a copy of the amended (although not yet fully-optimised to my own satisfaction) FDEs so you could try them out versus the set you have, to try to isolate the problem.
In answer to your specific questions:
1. Yes, but I'd advise not doing this, as it will have several other (unwanted) effects too. I think that, as Felix suggested, cg may be the issue.
2. Elevator trim in Microsoft FS is a sore point: in the real world, elevator trim is an aerodynamic effect, moving the net aerodynamic centre of the aircraft in order to reduce the load held by the pilot. It can be implemented by tabs or tailplane incidence, but the effect from the pilot's point of view is the same, regardless. What MSFS has implemented instead (why??) is to have the trim move the CG, not the aerodynamic centre. This materially affects the aircraft's handling and is hard to compensate for when writing FDEs: however, from the user's point of view you shouldn't see much effect unless you're making end-to-end sweeps of the trim range.
You'll see from (2) above that if you already have full up-elevator on a real aircraft, trimming will only change the stick loads, not the amount of pitch authority available (I except certain complex/rare/unusual aircraft from this). In fsim, if you continue to trim aft, the aircraft will continue to pitch up; this is wrong, but is an inherent characteristic in FS9, not specific to the MK BN-2.
I hope this helps
Cheers,
Kevin
I can't agree that "every other aspect of the Flight Dynamics seems fine" with MK's Islander: many of the aerodynamic derivatives are significantly out and the pitch trim goes the wrong way. Also, the centre of gravity is misplaced by about 1 ft, an astonishingly large error which degrades the handling considerably! These comments are based on the FDEs I downloaded (from the MK site) in Aug 2004, and further changes may have been made since then.
If you want, I could send you a copy of the amended (although not yet fully-optimised to my own satisfaction) FDEs so you could try them out versus the set you have, to try to isolate the problem.
In answer to your specific questions:
1. Yes, but I'd advise not doing this, as it will have several other (unwanted) effects too. I think that, as Felix suggested, cg may be the issue.
2. Elevator trim in Microsoft FS is a sore point: in the real world, elevator trim is an aerodynamic effect, moving the net aerodynamic centre of the aircraft in order to reduce the load held by the pilot. It can be implemented by tabs or tailplane incidence, but the effect from the pilot's point of view is the same, regardless. What MSFS has implemented instead (why??) is to have the trim move the CG, not the aerodynamic centre. This materially affects the aircraft's handling and is hard to compensate for when writing FDEs: however, from the user's point of view you shouldn't see much effect unless you're making end-to-end sweeps of the trim range.
You'll see from (2) above that if you already have full up-elevator on a real aircraft, trimming will only change the stick loads, not the amount of pitch authority available (I except certain complex/rare/unusual aircraft from this). In fsim, if you continue to trim aft, the aircraft will continue to pitch up; this is wrong, but is an inherent characteristic in FS9, not specific to the MK BN-2.
I hope this helps
Cheers,
Kevin
Thanks for the information guys.
The trim does operate in the correct sense Kevin, so I may have a different version that the one you have.
I've made some minor changes to the cfg file and seem to have made some progress.
1) Increased wing incidence angle to 3 degrees. It seems to cruise in a slightly more level attitude now (more like the pictures of the real ones). Also needs less nose up attitude on take off.
2) Reduced vertical placement of both engines. Now when you close the throttle you don't need full pitch down. Also, rotation is much easier as the engines are not producing as much of a pitch down force.
3) Changed thrust scalar to 1.1. Ground roll closer to what is required now, at the expense of about 80fpm better climb than spec.
FS9 trim sounds like a bit of a nasty hack. What you are describing is what I've been seeing with the Islander. (Probably others too, but more noticable with the initial Islander issues).
I've added a working trim position indicator to the panel now, and the takeoff and climbout is much better when the trim is in a consistent position :roll:
It would be good to compare FDEs Kevin, I'll PM you my email address.
Cheers,
--Graeme
The trim does operate in the correct sense Kevin, so I may have a different version that the one you have.
I've made some minor changes to the cfg file and seem to have made some progress.
1) Increased wing incidence angle to 3 degrees. It seems to cruise in a slightly more level attitude now (more like the pictures of the real ones). Also needs less nose up attitude on take off.
2) Reduced vertical placement of both engines. Now when you close the throttle you don't need full pitch down. Also, rotation is much easier as the engines are not producing as much of a pitch down force.
3) Changed thrust scalar to 1.1. Ground roll closer to what is required now, at the expense of about 80fpm better climb than spec.
FS9 trim sounds like a bit of a nasty hack. What you are describing is what I've been seeing with the Islander. (Probably others too, but more noticable with the initial Islander issues).
I've added a working trim position indicator to the panel now, and the takeoff and climbout is much better when the trim is in a consistent position :roll:
It would be good to compare FDEs Kevin, I'll PM you my email address.
Cheers,
--Graeme
After a bit more thought, I realised the trim was actually operating incorrectly. When the airspeed was increasing, I was actually trimming a bit more nose up :roll:
However, moving the Empty CofG forward one foot fixes the problem.
Finally, moving the main gear contact point forward one foot restores the rotation speed.
Cheers,
--Graeme
However, moving the Empty CofG forward one foot fixes the problem.
Finally, moving the main gear contact point forward one foot restores the rotation speed.
Cheers,
--Graeme
Graeme,
Yes, from looking at the aircraft.cfg file, I wonder if MK confused himself with the cg position: from some of the numbers, I suspect he started with the cg in one position and then moved it. The numbers in the .cfg and .air files are not an entirely consistent set, and the pitch trim direction is definitely wrong.
Glad it works for you now.
Cheers,
Kevin
Yes, from looking at the aircraft.cfg file, I wonder if MK confused himself with the cg position: from some of the numbers, I suspect he started with the cg in one position and then moved it. The numbers in the .cfg and .air files are not an entirely consistent set, and the pitch trim direction is definitely wrong.
Glad it works for you now.
Cheers,
Kevin
I had a go at the Islander last September/October and swapped a couple of mods with Mark 'Dark Moments', but can't remember now exactly what changes we made. I think they will be in amongst the following sections:
[WEIGHT_AND_BALANCE]
reference_datum_position=0.000, 0.000, 0.000
empty_weight_CG_position=2.000, 0.000, 0.000
//station_load.0 = "176.000, 5.900, -1.000, 0.000, Pilot"
//station_load.1 = "660.000, 1.8, 0, 0, Reihen 1 u. 2"
//station_load.2 = "660.000, -3.0, 0, 0, Reihen 3 u. 4"
station_load.0=160.000, 5.500, -1.100, -1.000 //pilot
station_load.1=160.000, 5.500, 1.100, -1.000 //copilot
station_load.2= 0.000, 4.200, -1.100, -1.000
station_load.3= 0.000, 4.200, 1.100, -1.000
station_load.4=170.000, 0.000, -1.100, -1.000 //rear passengers
station_load.5=170.000, 0.000, 1.100, -1.000 //rear passengers
station_load.6= 75.000,-1.500, -1.100, -0.000 //cargo
station_load.7= 75.000,-1.500, 1.100, -0.000 //cargo
[fuel]
LeftMain=0.000, -9.800, 2.300, 68.000, 0.000
RightMain=0.000, 9.800, 2.300, 68.000, 0.000
[airplane_geometry]
wing_area=322.000
wing_span=49.000
wing_root_chord=6.660
wing_pos_apex_lon=2.100 // 1.665 //
[piston_engine]
normalized_starter_torque=0.6 //0.900
idle_rpm_friction_scalar=1.0 //1.150
[flight_tuning]
parasite_drag_scalar=1.400 //1.000
[Flaps.0]
pitch_scalar=-1.000 //1.000
Obviously the bits that work will be by Mark and the ones that don't will be mine! If you haven't applied his changes to the default DC3, you should. Some of the changed lines above are probably in the original model, but I don't feel like going through the install procedure to check. I suspect the main areas played with will be the cg pos, payload distribution - and I definitely can't remember where I dug that up from, wing pos apex(but I don't remember playing with that) and the flap pitch scalar.
Naturally if I'd kept a backup of the original before carving it up, I'd know what I'd changed! I'm sure you all do.
Mark also made changes to the .air file. I can send a copy if anyone is interested.
One setting that I didn't change but which has made a difference in other aircraft is to change the horizontal tail incidence, confusingly called htail_incidence.
Anyway the changes made and Mark's .air file mods at least allow the brute to cruise hands off without bucket loads of trim. I recall it headed heavenwards before with the enthusiasm of the original UK Fighters Buccaneer.
One thing Mark noticed is the way the elevators trail at full down until speed builds up, but that is also to be seen in pictures on airliners.net, so maybe it is a bit of cunningly accurate modelling!
[WEIGHT_AND_BALANCE]
reference_datum_position=0.000, 0.000, 0.000
empty_weight_CG_position=2.000, 0.000, 0.000
//station_load.0 = "176.000, 5.900, -1.000, 0.000, Pilot"
//station_load.1 = "660.000, 1.8, 0, 0, Reihen 1 u. 2"
//station_load.2 = "660.000, -3.0, 0, 0, Reihen 3 u. 4"
station_load.0=160.000, 5.500, -1.100, -1.000 //pilot
station_load.1=160.000, 5.500, 1.100, -1.000 //copilot
station_load.2= 0.000, 4.200, -1.100, -1.000
station_load.3= 0.000, 4.200, 1.100, -1.000
station_load.4=170.000, 0.000, -1.100, -1.000 //rear passengers
station_load.5=170.000, 0.000, 1.100, -1.000 //rear passengers
station_load.6= 75.000,-1.500, -1.100, -0.000 //cargo
station_load.7= 75.000,-1.500, 1.100, -0.000 //cargo
[fuel]
LeftMain=0.000, -9.800, 2.300, 68.000, 0.000
RightMain=0.000, 9.800, 2.300, 68.000, 0.000
[airplane_geometry]
wing_area=322.000
wing_span=49.000
wing_root_chord=6.660
wing_pos_apex_lon=2.100 // 1.665 //
[piston_engine]
normalized_starter_torque=0.6 //0.900
idle_rpm_friction_scalar=1.0 //1.150
[flight_tuning]
parasite_drag_scalar=1.400 //1.000
[Flaps.0]
pitch_scalar=-1.000 //1.000
Obviously the bits that work will be by Mark and the ones that don't will be mine! If you haven't applied his changes to the default DC3, you should. Some of the changed lines above are probably in the original model, but I don't feel like going through the install procedure to check. I suspect the main areas played with will be the cg pos, payload distribution - and I definitely can't remember where I dug that up from, wing pos apex(but I don't remember playing with that) and the flap pitch scalar.
Naturally if I'd kept a backup of the original before carving it up, I'd know what I'd changed! I'm sure you all do.
Mark also made changes to the .air file. I can send a copy if anyone is interested.
One setting that I didn't change but which has made a difference in other aircraft is to change the horizontal tail incidence, confusingly called htail_incidence.
Anyway the changes made and Mark's .air file mods at least allow the brute to cruise hands off without bucket loads of trim. I recall it headed heavenwards before with the enthusiasm of the original UK Fighters Buccaneer.
One thing Mark noticed is the way the elevators trail at full down until speed builds up, but that is also to be seen in pictures on airliners.net, so maybe it is a bit of cunningly accurate modelling!
Hi guys,
If there is any interest in this, I could take a further look at the overall Islander FDEs (.cfg and .air files) with a view to uploading a properly refined version of what I've already done for my own use. However, if Allan and/or Mark B feel like publishing theirs, I'll desist: I have great respect for Mark's work!
It would be useful to have revised FDEs available to the whole community: after all the Islander is truly Classic British and the MK visual model is very good indeed.
Really, it isn't all that much of a change in the FDE coding to get it right - as I said above, I think that MK made a relatively straightforward error which threw off virtually the entire pitch axis.
Cheers,
Kevin
If there is any interest in this, I could take a further look at the overall Islander FDEs (.cfg and .air files) with a view to uploading a properly refined version of what I've already done for my own use. However, if Allan and/or Mark B feel like publishing theirs, I'll desist: I have great respect for Mark's work!
It would be useful to have revised FDEs available to the whole community: after all the Islander is truly Classic British and the MK visual model is very good indeed.
Really, it isn't all that much of a change in the FDE coding to get it right - as I said above, I think that MK made a relatively straightforward error which threw off virtually the entire pitch axis.
Cheers,
Kevin
The mods that Mark and I made were really just to get a flyable plane, if you'd care to give it a full going over I can send you Mark's .air file and our cfg file as a starter.
It's possible that we both left it for the other to publish (!) , but other things got in the way, so if you would care to run with it I'd be delighted.
It seems the original is fs2002 or possibly earlier and the FDs lost something in the translation to fs2004.
It would be good to have a sorted set of dynamics as the basic model is fine and as you say it is one of the few success stories of the recent past. It is now under Omani ownership (perhaps they could bid for BAe Systems?) and recently awarded a contract to re-assemble Cirrus aircraft headed for the European market. Gunter Kraemer's SR-20 is another fine model, but I digress!
Curiously enough the changes we made seemed to be just the tiniest bit more stable with the fan model than the standard one. Might just be my imagination.
I just got it fixed so that it could potter round the Orkneys using Loganair's Island schedule. Retractable undercarriage is fairly redundant between Westray and Papa Westray. Last I heard Mark was using it round the Carribean - warmer rain and fewer sheep. But enough of my hobbies.
It's possible that we both left it for the other to publish (!) , but other things got in the way, so if you would care to run with it I'd be delighted.
It seems the original is fs2002 or possibly earlier and the FDs lost something in the translation to fs2004.
It would be good to have a sorted set of dynamics as the basic model is fine and as you say it is one of the few success stories of the recent past. It is now under Omani ownership (perhaps they could bid for BAe Systems?) and recently awarded a contract to re-assemble Cirrus aircraft headed for the European market. Gunter Kraemer's SR-20 is another fine model, but I digress!
Curiously enough the changes we made seemed to be just the tiniest bit more stable with the fan model than the standard one. Might just be my imagination.
I just got it fixed so that it could potter round the Orkneys using Loganair's Island schedule. Retractable undercarriage is fairly redundant between Westray and Papa Westray. Last I heard Mark was using it round the Carribean - warmer rain and fewer sheep. But enough of my hobbies.