|
Author
|
Topic: More Questions!
|
Sv
Pilot
|
posted 12-06- 07:17 AM
   
First,Michael, how did you do the dihedral in your Camel's wings? All, how should dihedral be done? My guess is a flat wing, rotated in OPS. That way the dihedral will be modeled in the FM, right? Second, Can anyone explain the basic properties of the airfoil objects that help define the FM? Stuff like airAdvantage, etc - (I know that is the macanical force needed to overcome the control surface) Also, how fo you edit the lift/drag stuff, etc? Thanks as always! ------------------
-Sv =FC=
WWI in SDOE!
IP: Logged |
Hawk
JAG
|
posted 12-06- 08:16 AM
Sv when are you going to make that "flying razer" you picture at the end of your post and in that color scheme? I am looking forward to the more colorful german models.IP: Logged |
Sv
Pilot
|
posted 12-06- 08:42 AM
When my questions are answered   -Sv =FC= IP: Logged |
Michael Harrison
General
|
posted 12-06- 09:50 AM
For dihedral, build the wing with the angle built-in. The behavior will naturally fall out.Have you looked through the OpenPlane doc? It's not the easiest read, but you need to know the basics in there. If I remember correctly, there are some friendlier docs available at the OPP site. IP: Logged |
Sv
Pilot
|
posted 12-06- 10:21 AM
Yes Michael, I read the OpenPlane documentation all of the time. When you do not understand the big picture things can seem very criptic; once you have a general idea of what's going on the document comes to life.I infered from the doc that objects having airfoil duty will vector the forces in relation to how the LOD is orientated. In other words it does not matter at all what the LOD looks like, it is how it is rotated. I could make a wing with 90 degrees AOA in the LOD and the plane would still fly just fine so long as the orientation of the LOD remains unchanged. So if I have the dihedral built into the LOD, this dihedral would not be part of the FM. If I make the wing flat and rotate the LOD into the correct dihedral angel then the dihedral will be modeled in the FM. This is partialy confirmed to me by turning force vector dbug on and seeing that the lift force of the wing is perpendicular to the wing's surface. Not being an expert I figured I would double check - I still was not 100% sure. More importantly I wondered how well dihedral works within the OpenPlane engine - is it over/under sensative, etc. Also I just wanted to bring up issues around "designing with the FM in mind" here to get some debate going I do find the OpenPlane documentation makes perfect sense once I know the answer ------------------
-Sv =FC=
WWI in SDOE!
IP: Logged |
Michael
Pilot
|
posted 12-06- 01:36 PM
I did the dihedral by rotating the wings in AC3d. I believe the dihedral "works" simply because when the plane is banked, the "airpoint" will move outboard on the low wing and inboard on the high wing. (The airpoint is the point at which the game engine applies the aerodynamich forces generated by the airfoil object.)In the latest version of OpenPlane Studio (in which it is unfortunately difficult to save anything) you can see the position of the airpoint for each airfoil object, which should be in the middle of the wing, 25% of the chord back from the leading edge. To edit lift and drag you need the airfoil extractor program which I think you can get from Tailslide's site. Has anyone managed to rename an airfoil file? An obvious problem arises if you can't increase the number of files available. IP: Logged |
Sv
Pilot
|
posted 12-06- 02:13 PM
You are explaining bank. Without rotating the object in OPS I think the FM is working on the basis of zero dihedral. This may cause the plane to be too maneuverable, and less stable.Some squadrons reduced the dihedral of their SE5a's to make them more maneuverable. This might not alter the FM much though... ------------------
-Sv =FC=
WWI in SDOE!
IP: Logged |
Michael
Pilot
|
posted 12-06- 03:42 PM
Well, dihedral tends to restore a plane to level when banked. Thats why a plane with dihedral is more stable. So long as the airpoint doesn't get left behind when you rotate the wing, there shouldn't be a problem. Or am I not understanding your question?IP: Logged |
Sv
Pilot
|
posted 12-06- 04:03 PM
You said you did NOT rotate the wing in OPS, right? You built the dihedral in the 3D LOD object - right?Maybe I just need more sleep ------------------
-Sv =FC=
WWI in SDOE!
IP: Logged |
Michael
Pilot
|
posted 12-06- 04:32 PM
I built all the wings flat on the xy-plane in AC3d and then rotated the lower wings to give the dihedral, also in AC3d and not OPS. I can't see that it would make a difference which program is used. Some time ago I also searched the openplane document for dihedral and found nothing. As far as I can tell the only effect of the rotation that the game engine would act on is the new position of the airpoints relative to the airplane's centre of mass. The only way I can think of to test whether the engine is giving the desired results is to give something a deranged dihedral like 30 degrees and then see how the roll stabilty feels. (I will absolutely not look at this forum again before leaving work - I swear!) IP: Logged |
Bryan Russell
Pilot
|
posted 12-06- 05:46 PM
To change the terminology a bit, you need to create the wing LOD so that the wing tip and the centre of the wing root are aligned along the X-axis on the 0.0 co-ordinate. Then you need to change the Matrix associated with the Model so that the wing is bent up the correct amount. LOD = the polygons etc. Model = the built up object that SDOE displays that includes up to 5 LODs, positional/rotational matrix and a proplist. I don't know if the rotating Michael is talking about in AC3D is modifying the LOD mesh or the matrix, OPS definitely modifies the matrix. To check it you could load a AC3D saved plane in OPS and check the wings position relative to the selection box. I’m not sure if mposis used AC3D or not but the wings in both the Zero and BF109 originally had the mesh rotated rather than the matrix. This definitely has a difference, I bent a P51’s wings up to about 45deg and the performance changed considerably.
Bryan
IP: Logged |
Sv
Pilot
|
posted 12-06- 06:58 PM
Thank you Bryan! That is what I was hoping for!How could SDOE possible know your LOD is angled? FM DOES NOT use any LOD vertecies! This is the key: LODS have nothing to do with FM, only the airFoil properties working with the matrix position of that LOD. Now I can sleep easy Thanks! Heheh I do need the sleep you know. ------------------
-Sv =FC=
WWI in SDOE!
IP: Logged |
roadtoad
Pilot
|
posted 12-06- 07:00 PM
as a sidebar that relates to the FM: all the best fighters of ww1 exhibited what was then called 'lateral instability', for unlike ww2, the ability to turn was paramount in ww1, and constant use of the rudder birds like the camel, dr7, neiuport, etc were the best - like a dirtbike, you steered with the rear (the neiuport didn't even have a stab for stability, just rudder). Given OP's hyper-rudder, this may be a problem if you code wizards are to make them more than just slow ww2 birds - aercoupes with guns.
It's amazing how much info there is on the ww1 stuff compared to ww2. In just an hour on the net I found a huge amount of data on nearly all of them - stuff like rudder area, weights, climb, stall, even the lengths and diameters of all the wires for the se5.IP: Logged |
Michael
Pilot
|
posted 12-06- 08:51 PM
Good thing you raised this Sv. It seems I've been working with some shaky assumptions. I looked at the selection boxes on the Camel's wings in OPS and they are not rotated. So I guess I will have to rotate them back in AC3D and then back again in OPS.Bryan: Can you tell me how many degrees of rotation one key click gives in OPS? IP: Logged |
Bryan Russell
Pilot
|
posted 12-07- 05:22 PM
The units are actually in Rads, but it works out to approx:
Unmodified = 5.7 deg.
With Shift Key = 57.3 deg
With Control Key = 0.05 deg
I have also defined a Alt combination (.57 deg ) but I forgot about it being the menu key so it doesn't work.  I should probably change this so that the steps = 0.1 deg, 1 deg and 45 deg maybe. IP: Logged |
mposis
Pilot
|
posted 12-07- 05:58 PM
Bryan,Thanks for fixing the 109. IP: Logged |
Sv
Pilot
|
posted 12-08- 05:54 AM
Thanks Bryan,I GREAT addition to ops would be offering the abilty to move or rotate an object a given amount - open up a dialog box and let us enter in something like 4 inches, or 8 feet - and for rotating, 3.4 degrees clock-wise. This would help us build planes to spec easier. Also this would kick ass: let "ctrl" modify the move and rotate tool in camera mode as well - to slow down the movement. That way when we are zoomed close to the origin of the model we can still rotate - right now it is so touchy in close that it is close to impossible to navigate into position to move/postion cockpit stuff. I know you're in bug-fix mode now, so I'm just adding to the pile ------------------
-Sv =FC=
WWI in SDOE!
IP: Logged |
Bryan Russell
Pilot
|
posted 12-08- 06:07 AM
What I was planning to do was add some status bar stuff that showed amount of rotation, movement etc, a manual entry box could also be a good idea. It is difficult to show or specify absolute rotations, so it would only be the additional rotations. I notice that 3DS MAX only shows changes in rotations as well so it might not just be me (note the sly reference to an established company to hide my own inadequecies) I tried to be too smart for my own good with the view movement, and ended up screwing it up, you'll notice at some zoom distances things don't move at all.
mposis,
Glad to help, actually, I was really interested to see what effect the "correct" dihedral would have with Zurs expert FM work behind it.
Bryan
[This message has been edited by Bryan Russell (edited 12-08-1999).] IP: Logged |
Sv
Pilot
|
posted 12-08- 08:12 AM
Cool Bryan,Absolute rotations is plenty! Especialy if you add a function to "snap object to axis" (notice the way I cover up my request for new functionality by making it see m like part of an existing request). -Sv -Sv IP: Logged |
roadtoad
Pilot
|
posted 12-08- 08:59 AM
haha! you guys are great!IP: Logged |
FSIC Messageboard
UBBFriend: Email This Page to Someone! 
