Discussion X-Y plane aerofoil?

Urwumpe

Not funny anymore
Addon Developer
Donator
Joined
Feb 6, 2008
Messages
37,588
Reaction score
2,312
Points
203
Location
Wolfsburg
Preferred Pronouns
Sire
Couldn't you define two airfoils in the same plane, with their centers of pressures on opposite sides of the vessel and with equal and opposite lift coefficients at 0 angle of attack? The center of pressure of the airfoil does not need to be on the plane of symmetry of the vessel; it is set with a vector based on vessel coordinates.

This is how things like stabilizers and rudders are established, but those are usually symmetric airfoils with lift coefficients of 0 at 0 angle of attack.

Would need to try how this works out, but its a pretty uncommon solution for models of real air- and spacecraft. We can't measure them independently and they interact.
 

Thunder Chicken

Fine Threads since 2008
Donator
Joined
Mar 22, 2008
Messages
4,327
Reaction score
3,248
Points
138
Location
Massachusetts
Would need to try how this works out, but its a pretty uncommon solution for models of real air- and spacecraft. We can't measure them independently and they interact.
I used this years ago to model asymmetric stall in a Cessna to get it to enter spins. Left and right wings were independent airfoil definitions. Sadly I don't have the code anymore and OH seems to have eaten the add-on, but it did work. https://www.orbiter-forum.com/threads/cessna172-enhancement-module-v-2-0.12257/
 

Urwumpe

Not funny anymore
Addon Developer
Donator
Joined
Feb 6, 2008
Messages
37,588
Reaction score
2,312
Points
203
Location
Wolfsburg
Preferred Pronouns
Sire

BrianJ

Addon Developer
Addon Developer
Joined
Apr 19, 2008
Messages
1,676
Reaction score
900
Points
128
Location
Code 347
Hi again!
Sorry but....I still have problems with trying to reproduce the behaviour of an XY plane aerofoil by using the current Vertical and Horizontal aerofoil lift/drag functions. The problem is one of geometry this time.
Maybe someone can see where I'm going wrong(?).

I'm just going to consider profile drag for this purpose.

Lets assume that my "XY plane" aerofoil (I'll call it XYP) is symmetrical around the Z-axis, and instead of using angle-of-attack (aoa) or slipangle (beta) to set the Cd, we use the angle between the airflow vector and the normal to the aerofoil surface (vessel -Z axis). We'll call that angle "angle of incidence" (aoi).

So for a capsule flying with blunt end (-Z) directly into the airflow:
aoa = 180
beta = 180
aoi = 180

if we pitched the capsule blunt end down 10deg we get
aoa = 170
beta = 180
aoi = 170

or if we yaw it 10deg we get
aoa = 180
beta = 170
aoi = 170

The drag for XYP is calculated in the same way as for Vertical and Horizontal aerofoils by Orbiter:
Drag = Freestream-dynamic-pressure * wing-area * Cd
For this discussion lets set Freestream-dynamic-pressure = 1, so:
Drag = wing-area * Cd

For XYP lets say wing area is 10m^2 and we set the Cd so that:
at 180 aoi, Cd = 1
at 170 aoi, Cd = 0.9

Lets say we use two identical (for symmetry around Z axis) Vertical and Horizontal aerofoils to mimic the behaviour of the single XYP. They each have area 5m^2, and we can set the Cd according to aoa or beta thus
Horizontal Airfoil:
at 180 aoa, Cd =1
at 170 aoa, Cd = 0.8
Vertical Airfoil:
at 180 beta, Cd =1
at 170 beta, Cd = 0.8

Let's see how this works out for some different orientations......

For aoa = 180, beta = 180, aoi = 180:
For XYP, Drag = 10 x 1 = 10
For Vertical + Horizontal aerofoils, Drag = (5 x 1) + (5 x 1) = 10
That works ok!

For aoa = 170, beta = 180, aoi = 170:
For XYP, Drag = 10 x 0.9 = 9
For Vertical + Horizontal aerofoils, Drag = (5 x 1) + (5 x 0.8) = 9
That works ok! And similarly ok for aoa = 180, beta = 170, aoi = 170.

BUT.... what happens if we have both aoa and beta angles??
If we assume aoa and beta are equal, and aoi is 170, a little work with vectors(which I can show if required ;-)
shows that aoa and beta must equal 172.89

So, assuming linear interpolation between Cd points:
Cd for the Vertical and Horizontal aerofoils is:
Cd = 1 - { (180 - 172.89)/(180 - 170) } * (1 - 0.8) = 1 - (0.711) * (0.2) = 1 - 0.1422 = 0.8578

So for aoa = 172.89, beta = 172.89, aoi = 170:
For XYP, Drag = 10 * 0.9 = 9
For Vertical and Horizontal aerofoils, Drag = (5 * 0.8578) + (5 * 0.8578) = 8.578
Drag is not equivalent!!!

Or you could take a simpler case where say the Cd for XYP is 0.4 at aoi = 120
and you have the situation where: aoa = 120, beta = 180, aoi = 120
For XYP, Drag = 10 * 0.4 = 4
In this case, Horizontal aerofoil would require negative Cd!
i.e. For Vertical and Horzontal aerofoils, Drag = (5 x 1) + (5 x -0.2) = 4

I conclude that in situations where neither aoa or beta are 0, there is no exact equivalence of behaviour between XYP and
Vertical+Horizontal aerofoils. Its down to the geometry of using two angles vs. just one.

Maybe one can use a transformation of aoa and beta in the lift/drag functions to compensate.
Or use some kind of non-linear interpolation, but...ugh!
And to have negative Cd seems weird.

I still think AIRFOIL_CAPSULE concept has value

Phew! Had to get that out....it was making my brain itch!

If you bothered to read all the above...thanks for your interest!!

Cheers,
BrianJ
 

Thunder Chicken

Fine Threads since 2008
Donator
Joined
Mar 22, 2008
Messages
4,327
Reaction score
3,248
Points
138
Location
Massachusetts
Hi again!
Sorry but....I still have problems with trying to reproduce the behaviour of an XY plane aerofoil by using the current Vertical and Horizontal aerofoil lift/drag functions. The problem is one of geometry this time.
Maybe someone can see where I'm going wrong(?).

I'm just going to consider profile drag for this purpose.

Lets assume that my "XY plane" aerofoil (I'll call it XYP) is symmetrical around the Z-axis, and instead of using angle-of-attack (aoa) or slipangle (beta) to set the Cd, we use the angle between the airflow vector and the normal to the aerofoil surface (vessel -Z axis). We'll call that angle "angle of incidence" (aoi).

So for a capsule flying with blunt end (-Z) directly into the airflow:
aoa = 180
beta = 180
aoi = 180

if we pitched the capsule blunt end down 10deg we get
aoa = 170
beta = 180
aoi = 170

or if we yaw it 10deg we get
aoa = 180
beta = 170
aoi = 170

The drag for XYP is calculated in the same way as for Vertical and Horizontal aerofoils by Orbiter:
Drag = Freestream-dynamic-pressure * wing-area * Cd
For this discussion lets set Freestream-dynamic-pressure = 1, so:
Drag = wing-area * Cd

For XYP lets say wing area is 10m^2 and we set the Cd so that:
at 180 aoi, Cd = 1
at 170 aoi, Cd = 0.9

Lets say we use two identical (for symmetry around Z axis) Vertical and Horizontal aerofoils to mimic the behaviour of the single XYP. They each have area 5m^2, and we can set the Cd according to aoa or beta thus
Horizontal Airfoil:
at 180 aoa, Cd =1
at 170 aoa, Cd = 0.8
Vertical Airfoil:
at 180 beta, Cd =1
at 170 beta, Cd = 0.8

Let's see how this works out for some different orientations......

For aoa = 180, beta = 180, aoi = 180:
For XYP, Drag = 10 x 1 = 10
For Vertical + Horizontal aerofoils, Drag = (5 x 1) + (5 x 1) = 10
That works ok!

For aoa = 170, beta = 180, aoi = 170:
For XYP, Drag = 10 x 0.9 = 9
For Vertical + Horizontal aerofoils, Drag = (5 x 1) + (5 x 0.8) = 9
That works ok! And similarly ok for aoa = 180, beta = 170, aoi = 170.

BUT.... what happens if we have both aoa and beta angles??
If we assume aoa and beta are equal, and aoi is 170, a little work with vectors(which I can show if required ;-)
shows that aoa and beta must equal 172.89

So, assuming linear interpolation between Cd points:
Cd for the Vertical and Horizontal aerofoils is:
Cd = 1 - { (180 - 172.89)/(180 - 170) } * (1 - 0.8) = 1 - (0.711) * (0.2) = 1 - 0.1422 = 0.8578

So for aoa = 172.89, beta = 172.89, aoi = 170:
For XYP, Drag = 10 * 0.9 = 9
For Vertical and Horizontal aerofoils, Drag = (5 * 0.8578) + (5 * 0.8578) = 8.578
Drag is not equivalent!!!

Or you could take a simpler case where say the Cd for XYP is 0.4 at aoi = 120
and you have the situation where: aoa = 120, beta = 180, aoi = 120
For XYP, Drag = 10 * 0.4 = 4
In this case, Horizontal aerofoil would require negative Cd!
i.e. For Vertical and Horzontal aerofoils, Drag = (5 x 1) + (5 x -0.2) = 4

I conclude that in situations where neither aoa or beta are 0, there is no exact equivalence of behaviour between XYP and
Vertical+Horizontal aerofoils. Its down to the geometry of using two angles vs. just one.

Maybe one can use a transformation of aoa and beta in the lift/drag functions to compensate.
Or use some kind of non-linear interpolation, but...ugh!
And to have negative Cd seems weird.

I still think AIRFOIL_CAPSULE concept has value

Phew! Had to get that out....it was making my brain itch!

If you bothered to read all the above...thanks for your interest!!

Cheers,
BrianJ
I really think you are really confusing yourself with the third airfoil definition and the third angle aoi. It is not necessary at all. I think you are struggling with the image of an "airfoil" being this streamlined object that can't possibly mimic the drag of a blunt object. But, as I said, in computational simulation you can give an "airfoil" the drag properties of a potato if you want. It's just math and numbers.

What you want to do is, for any combination of alpha and beta, have the total sum of the drag calculated by those two airfoils to be the correct total drag on the vehicle. You can make a table with alpha on one axis, beta on the other axis, and total Cd as a function of both. You then need to decide how to allocate what fraction is a function of alpha and what fraction is due to beta. You can do this any way you like, so long as the total is the correct value. I believe it is possible to pass the yaw and pitch angles into a single drag coefficient definition so you could potentially just implement drag as a function of both alpha and beta in one of the two airfoil definitions and just set the drag on the other to zero.

The sign of Cd is due to the convention of positive drag being in the -Z direction. If you are flying backwards, drag is accelerating the vehicle in the +Z direction. There is nothing wrong with this.
 
Last edited:

BrianJ

Addon Developer
Addon Developer
Joined
Apr 19, 2008
Messages
1,676
Reaction score
900
Points
128
Location
Code 347
Thanks for the swift reply!
I really think you are really confusing yourself with the third airfoil definition and the third angle aoi. It is not necessary at all. I think you are struggling with the image of an "airfoil" being this streamlined object that can't possibly mimic the drag of a blunt object. But, as I said, in computational simulation you can give an "airfoil" the drag properties of a potato if you want. It's just math and numbers.
Not really. I'm fine with just looking at the numbers :)
I think any confusion is down to my ignorance about how the lift/drag coefficient functions can be constructed.

What you want to do is, for any combination of alpha and beta, have the total sum of the drag calculated by those two airfoils to be the correct total drag on the vehicle.
Yes, thats it.

You can make a table with alpha on one axis, beta on the other axis, and total Cd as a function of both. You then need to decide how to allocate what fraction is a function of alpha and what fraction is due to beta. You can do this any way you like, so long as the total is the correct value. I believe it is possible to pass the yaw and pitch angles into a single drag coefficient definition so you could potentially just implement drag as a function of both alpha and beta in one of the two airfoil definitions and just set the drag on the other to zero.
Ah, yes! If one can pass both alpha and beta angles to the same lift/drag coefficient function for either Vertical or Horizontal aerofoil, then we're ok. But then, effectively you just end up calculating my "aoi" angle from alpha and beta. Which is not really a problem. In fact easier to calculate my aoi angle from vectors elsewhere in the code and pass that over - I hadn't thought of that :)
The sign of Cd is due to the convention of positive drag being in the -Z direction. If you are flying backwards, drag is accelerating the vehicle in the +Z direction. There is nothing wrong with this.
I would have said the convention is that postive drag acts in the direction of the airstream, regardless of vessel orientation.
Negative drag would act in the opposite direction to the airstream.

Anyway, I think you've solved my problem by suggesting passing both alpha and beta to the same lift/drag coefficient function (or calculating my aoi angle, which is what I need, elsewhere).

Yes, its my tendency to look at a function in the code and think "well that's what I have to work with" rather than seeing how to modify it - thats my problem!

Next, on to the Lift - which is a slightly different case, and I need to look at the API guide, etc.
I hope I don't have to bother you again!

But, yes, I STILL think it would be nice to have a AIRFOIL_CAPSULE option, where the "aoi" angle is provided directly, and the lift vector put into the correction orientation (perpendicular to airstream, in the plane of airstream and z-axis).
Then you need only have one aerofoil, and no need for calculating angles.

As you say, now Orbiter is open source, I could make it myself - but that's probably beyond my very limited coding and compiling skills for now.

Many, many thanks for your help.
BrianJ
 

Thunder Chicken

Fine Threads since 2008
Donator
Joined
Mar 22, 2008
Messages
4,327
Reaction score
3,248
Points
138
Location
Massachusetts
Ah, yes! If one can pass both alpha and beta angles to the same lift/drag coefficient function for either Vertical or Horizontal aerofoil, then we're ok. But then, effectively you just end up calculating my "aoi" angle from alpha and beta. Which is not really a problem. In fact easier to calculate my aoi angle from vectors elsewhere in the code and pass that over - I hadn't thought of that :)
I don't see why you need to calculate aoi at all. Make Cd a function of pitch and yaw and call it a day. No sense making the computer do math that isn't needed. It already handles the vector math for you.
I would have said the convention is that postive drag acts in the direction of the airstream, regardless of vessel orientation.
Negative drag would act in the opposite direction to the airstream.
That's another way to say the same thing, but the drag force must be applied to the vessel in Orbiter in the vessel coordinate system, not relative to the wind direction, hence the sign convention. If the relative wind comes from the vessel -Z direction, your Cd must be negative to accelerate the vessel in the vessel +Z direction.
 
Top