Lift/Drag attack point, flipping ships, and other SC3 issues!

[Epsilon]

New to the forums, relatively new to Orbiter here, but certainly not new to the world of modding.

So here's my issue. Firstly, I can't figure out what, exactly, a lift/drag attack point is, with reference to the orbiter APIs and Spacecraft3.dll - I've been over the API docs as well as the SC3 docs, and it goes over them all relatively vaguely. It almost looks like it's a balance point for lift, but the craft acts stupid when I end up setting it at (0,0,0). I imagine this would aid me in getting this thing to fly more like a plane and less like a thrown brick with a -very- good arm.

Secondly, I need more lift! Once again, the resorces I'm looking at aren't helping me, and I'm wondering if I'm asking the wrong question at this point. The situation is as such:

I've got a craft with a weight of about 725000kg. After fuel, etc. it's about 820k. I've got engines powerful enough, but the takeoff speed is roughly mach .5, and this seems fast. Once I get it into the air, about 45km up, it stops generating enough lift to pitch the craft up, even leaning on the elevators, and until that point (because it's going so bloody fast), if you want to come down for some reason, you can't pitch down enough to get your vertical accelleration to go negative. I'd like two things: First, I'd like to be able to generate enough lift to not have that brief levelling off at 45km (which sorts itself at about 55km - there's enough residual vertical movement that it can pitch up at that point, so I -can- get it into orbit), and secondly, I'd like the elevators to be a bit more responsive without increasing the lift change -again-, since it's at 1.75 right now already.

I can fudge on the wing size, elevator size, aileron size, etc, and fudge on the weight as well, but changing the weight didn't seem to make it anything other than a bit less stable - the takeoff speed is -still- about mach .5 (using the SurfaceMFD measurement)

Any suggestions? :mellow:

http://www.etmoonshade.net/powell.jpg - this is the ship, crappily textured.

As an afterthought, I guess that what I'm asking on the second one is what change is likely to cause the largest change in lift with the smallest change in numbers, and therefore the smallest change in the model.

 

[RisingFury]

The lift attack point, from what I've been able to figure out, is the point on the wing where the force is acting on...

Sure, force acts on the entire wing, but you can represent that as being one single point. The location of the point depends on the shape of the wing... If I had to guess, it'd be somewhere between 2/3's to 3/4's back... but if you give me some time, I can do some calculations, if you give me the shape of the wing... the three points that define the triangle.

The lift of the wing is a product of many things. Surface area and velocity squared are two of the things that play a large role. Also I have to mention the density of the air - which at high altitudes is very low.

The problem you're having is that at 45 km, the air becomes so thin that you need massive velocity to get enough lift. One thing you can do is increase the surface area of the wing... and for the AF control surfaces, you should increase the surface area as well... but the main thing you should do is turn on RCS. At high altitudes, no matter what you do, your wings will lose lift. You'll need to use the RCS to pitch your ship back and divert some of the engines' thrust down - to aim some of the engine force up. That will compensate for the lost lift.

Again, if you give me the three points where your wings form a triangle, I should be able to calculate where the lift attack point should be.

 

[Epsilon]

Quick response for the win. Thanks!
So, at 45km, RCS should be doing more of the work of rotating the ship than the control surfaces. I can see that, and I can make that work - I think. And I can twiddle with the wing area and see if I can make that do something. I assume that by "dependent on the shape of the wing", the lift/drag attack point is going to be somewhere around the center of mass of the wing?

Another vaguely related question - which one is the vertical airfoil? Does "vertical" refer to the general direction of the lift, meaning the wings, or does it refer to the orientation of the airfoil, which would be the tailfin?

-Eppy!

 

[RisingFury]

Hmmm...

Well, if you assumed that the wing was the same mass per square cm at every point of the wing, you could say that the attack point is the center of mass for the wing.

In fact, I was gonna calculate it assuming that every part of the wing gives the same amount of lift force - probably not entirely correct, but it would still be a good estimate.

The horizontal / vertical is described in the API guide (not reference) file. You got me seriously confused right now... I think it refers to the direction of the wing - the wings would be horizontal, and the fins would be vertical, but not 100% sure here cos I don't have the guide here...

 

[Epsilon]

I'm glad I'm not the only one!

See, the problem is, when I define the horizontal airfoil like wings, I don't get lift worth a crap. I wanted to know if I was imagining things, I hosed something up, or if it's a difference how it's defined from logic. Maybe it has something to do with the mesh orientation... *wanders off to look*

Yet another side question - is it possible to define multiple airfoils for the horizontal and vertical directions with Spacecraft3.dll? I know it's possible through actually doing the code, but I'm not cool enough to do code for a ship.

 

[Linguofreak]

The problem with takeoff speed being so high and not varying with weight may be an indication that you have your touchdown points placed in a bad position relative to your center of gravity. The further the main gear are from the CG, the more torque it takes to bring the nose off the ground.

The issue with not being able to get the nose down when in the air at low altitudes may have to do with your center of lift being too far forward relative to your CG, in other words, you're tail-heavy. Tweaking that lift-drag attack point might help here.

 

[Epsilon]

Heh. I think I've got that sorted finally. The next issue I'm having is that on re-entry, the slightest twiddle of any of the control surfaces sends me into ten different axes of spinning (or so it seems). >_>

Sorry for peppering y'all with questions here. I very well may be trying to do the impossible, which is to say that I want good control, but not too sensitive, but I'm more inclined to think that there's some sort of happy medium that I'm missing.

 

[RisingFury]

It's better that you ask and come up with a higher quality add-on then to not ask and have a lower quality add-on.


Remember: "There are no stupid questions, just stupid answers."

 

[Epsilon]

I thought it was "There are no stupid questions, just stupid people."

 

[martins]

It almost looks like it's a balance point for lift, but the craft acts stupid when I end up setting it at (0,0,0). I imagine this would aid me in getting this thing to fly more like a plane and less like a thrown brick with a -very- good arm.

Let me just chip in here, since this point is probably of general interest to vessel modders:

Placing the airfoil attack point (the centre of pressure, CoP) at the origin (the centre of gravity, CoG) is generally not a good idea. It makes the flight dynamics inherently unstable. (In principle there is nothing wrong with an unstable design - I think most modern fighter planes use it, since it improves agility. But you probably need to implement a good flight computer along with it to make it flyable).

Traditionally, the CoP is behind the CoG. This means that the lift force will induce a torque that tries to lift the tail of the aircraft. This torque is counterbalanced by a down force of the rear stabilisers (or more rarely by an up-force from forward canards). Having the CoP behind the CoG is a self-stabilising design:

Starting from straight and level flight, if you lose lift (e.g. as a result of reducing airspeed), the lift-induced torque will also drop, causing the plane to pitch up, i.e. increase AoA, and therefore increase lift. So you got a nice negative feedback system. Same for lift perturbation in the other direction.

The wider the distance between CoG and CoP, the "stiffer" the handling of the aircraft (i.e. the aircraft has a stronger tendency to turn into the flight path, doesn't allow much AoA).

(Disclaimer: This is as far as I understand it. All corrections welcome!)

 

[Epsilon]

I've attracted the attention of the Good Doctor himself!

Anyway. That's very much the kind of answer I was looking for, Martin - much appreciated. I think that working from that answer, it'll make it quite a bit easier to balance this thing out.

The next question is a re-iteration of a previous one - is it possible to specify multiple airfoils with Spacecraft3.dll, or will later definitions overwrite previous ones in the config file? I imagine it depends on how the dll is built. I'll bet I could figure it from the source code...

 

[martins]

The next question is a re-iteration of a previous one - is it possible to specify multiple airfoils with Spacecraft3.dll, or will later definitions overwrite previous ones in the config file? I imagine it depends on how the dll is built. I'll bet I could figure it from the source code...

I'm not familiar with Spacecraft3, so somebody else would have to answer this. Or just figure out by experiment.

 

[Epsilon]

I plan on experimenting if someone doesn't come up with it before I get done with my radio show. An hour and a half 'til I can get back to Orbiter! *nod self*

Of course, reading forums while broadcasting is distracting...

 

[joeybigO]

Just want to add something here...

The CG should be within 25% of the MAC or the CoP. If this is exceeded you will find your results will roll like what you have, high v1.

However if you roll your attack angle point at the most aft point you can, say at the aft stabilizer, then you can have a good lift point. However the aft stabilizer has a lifting MAC too, 25% aft of the Aft CoP.


Edit: I really like the station, and the program. It plays a lot of everything that I like, my listening tastes run about the same as yours, a bit of everything. Nice intro into the station too, Do Ya from ELO is nice...

 

[Epsilon]

*chortles* Glad you like the show. I'm running out of steam for the night.

Question - what do you mean, exactly, by "MAC"? I can sorta guess from the context, but I'd rather ask than screw it up. ;p

 

[joeybigO]

*chortles* Glad you like the show. I'm running out of steam for the night.

Question - what do you mean, exactly, by "MAC"? I can sorta guess from the context, but I'd rather ask than screw it up. ;p

Duh, sorry...
Mean Air Chord, or the lifting point of the wing. Some mentally challenged (me) call it the fat part of the wing.

 

[Epsilon]

Glad I asked. :blink:

Experimentation will commence on SC3 now, to the sounds of Chuck Berry. We'll find out if we've got capabilities for multiple airfoils or not.

 

[joeybigO]

Riding along in my automobile. With no particular place to go.
Does seem fitting there.

 

[Epsilon]

I'm now playing "Spin, Spin Sugar", by Orbital (amusingly enough) after attempting re-entry.
It was going quite well until I got low, and tried to pitch down from 90 degrees (which seems to be the only way to avoid having to go around the Earth in orbit twice more while trying to slow down...) and started spinning around all over the place. ^_^

That being said, it -looks- like you can define more than one airfoil. I'm not positive, but I'm -fairly- certain the secondary vertical airfoil (the fins you see on the side of the mesh) aren't enough to lift this beast. I'm gonna do one more test, by turning the aerodynamics to the equivalent of a couple of pegs on the side, just to make sure.

Would a tailfin that is too large cause wobbling left and right while in "straight" flight? *ponders*

 

[joeybigO]

After consultation with the "minds" the X axis will be very sensitive. Pending the airfoil size. However too large and too little foil means almost no control.