Aerodynamic Flight Model... Huge Kudos To The Devs

[Buzz313th]

Huge Kudos to the devs on this.  I played around with making a simple jet fighter and balanced the  COM, COT  and COP (COL) in the VAB.  I also turned off the reaction wheel, turned off SAS and locked the engine gimble to explore where you all went with the flight model.  Low and behold, it is phenomenal.    The plane exhibits both power on and power off stalls, with an early asymmetric wing drop, that can progress into the beginning of a spin.  Thrust imposes a nice and balanced positive stability at the limit and high AOA feels like it should dynamically, both with and without power regarding an offset thrust vector.  I see where you all are going and I love it.   If your attention to detail follows through with the rest of the game, we will have an aeronautical physics study sim with all the bells and whistles for tremendous longevity.

 

A couple things I noticed that may help others at this point...

1. Seems like the COP needs to be just a hair aft of the COM.

2. Turn down control surface authority to explore stable flight since we don't have the ability yet to control the flight surfaces linearly with flight control hardware.

3. First balance the aircraft with all flight aids off..  SAS, Reaction wheels and thrust vectoring before turning them back on.  Use your trim.  

 

A couple requests moving forward....

1. Please give us hardware support soon, (Joystick, rudder and throttle).  It's a shame to have such a good flight model without the ability to to really experiment with it.

2. Please give us First person cockpit view, and if the reason we don't already have it is because the virtual cockpits aren't complete, then please just give us an unobstructed front view from the cockpit POV.

3. When building aircraft in both KSP1 and KSP2 it always seems like the COM is inherently too far forward, like the cockpits and nosecones weigh too much, or the engines don't weigh enough and it becomes a virtual challenge to build a conventional aircraft with wings close to the COM  and rear horizontal stabilizers.  I always find I have to push the wings too far forward in reference to the fuselage to get a good balance.  It seems like both games favor the rear delta and forward canard type aircraft.   I'm curious why that is.

4. And please give us the ability to select airspeed as an additional speed indicator.  

 

Thanks you

And

Cheers

Bz

Edited February 28, 2023 by Buzz313th

[Caspet]

Hey I have big problems with building a plane in ksp 2. The planes always yaw when I want to roll. They can't even fly straight but have to lean to one side and fly a little bit sideways. My ruder has roll disabled and my col is behind Com. I tried a few airplanes and only one haven't had this issue. There is no chance to fly a plane in ksp 2 for me. In ksp 1 I build a ton of flyable airplanes. Do you know what I am doing wrong. You seem to know how it works

[Buzz313th]

4 hours ago, Caspet said:

Hey I have big problems with building a plane in ksp 2. The planes always yaw when I want to roll. They can't even fly straight but have to lean to one side and fly a little bit sideways. My ruder has roll disabled and my col is behind Com. I tried a few airplanes and only one haven't had this issue. There is no chance to fly a plane in ksp 2 for me. In ksp 1 I build a ton of flyable airplanes. Do you know what I am doing wrong. You seem to know how it works

Does the yaw only happen during aileron deflection and then stop when you stop the roll?  

 

Post a screenshot of your plane build in the VAB with the COM, COL and COT in an orthogonal side and front view.

Edited February 28, 2023 by Buzz313th

[JoeSchmuckatelli]

5 hours ago, Buzz313th said:

Huge Kudos to the devs on this.  I played around with making a simple jet fighter and balanced the  COM, COT  and COP (COL) in the VAB.  I also turned off the reaction wheel, turned off SAS and locked the engine gimble to explore where you all went with the flight model.  Low and behold, it is phenomenal.    The plane exhibits both power on and power off stalls, with an early wing drop.  Thrust imposes a nice and balanced positive stability at the limit and high AOA feels like it should dynamically, both with and without power regarding an offset thrust vector.  I see where you all are going and I love it.   If your attention to detail follows through with the rest of the game, we will have an aeronautical physics study sim with all the bells and whistles for tremendous longevity.

 

A couple things I noticed that may help others at this point...

1. Seems like the COP needs to be just a hair aft of the COM.

2. Turn down control surface authority to explore stable flight since we don't have the ability yet to control the flight surfaces linearly with flight control hardware.

3. First balance the aircraft with all flight aids off..  SAS, Reaction wheels and thrust vectoring before turning them back on.  Use your trim.  

 

A couple requests moving forward....

1. Please give us hardware support soon, (Joystick, rudder and throttle).  It's a shame to have such a good flight model without the ability to to really experiment with it.

2. Please give us First person cockpit view, and if the reason we don't already have it is because the virtual cockpits aren't complete, then please just give us an unobstructed front view from the cockpit POV.

3. When building aircraft in both KSP1 and KSP2 it always seems like the COM is inherently too far forward, like the cockpits and nosecones weigh too much, or the engines don't weigh enough and it becomes a virtual challenge to build a conventional aircraft with wings close to the COM  and rear horizontal stabilizers.  I always find I have to push the wings too far forward in reference to the fuselage to get a good balance.  It seems like both games favor the rear delta and forward canard type aircraft.   I'm curious why that is.

 

Thanks you

And

Cheers

Bz

I'd appreciate a longer post from you (if you're willing) to help some of us figure out how to build effective planes and understand how to use the procedural parts effectively.   Plus any other observations you might want to share.  Do you mind writing up a tutorial?

 

If you do, please mention me so I can find it!

 

Thanks!

[Buzz313th]

1 hour ago, JoeSchmuckatelli said:

I'd appreciate a longer post from you (if you're willing) to help some of us figure out how to build effective planes and understand how to use the procedural parts effectively.   Plus any other observations you might want to share.  Do you mind writing up a tutorial?

 

If you do, please mention me so I can find it!

 

Thanks!

Hey Joe,

I'm still experimenting with the procedural flight surfaces and have come to the conclusion that not all adjustable parameters affect the aerodynamic properties.  I believe that increasing the wing surface area (Span and length) will increase lift and increase induced drag, but I haven't experimented enough to see if chord height, and surface sweep do anything.  In the VAB the COP does not move when you sweep the wing for or aft, which makes me believe that the wing angle is not doing anything yet.   But for what exists, the flight model is behaving predictably when the aircraft is set and balanced.

 

Regarding a tutorial.  I probably won't go there, but I can try to help by giving a few tips.   

1.  For a nicely balanced aircraft, one that flies predictably while under thrust and not.  Make sure your Center Of Aero Pressure (COP) is just a hair aft of the Center Of Mass and on the same horizontal plane.  Try to get your Center of thrust (COT) also on the same horizontal plane as the COM.  If the COT is offset from the COM, then you will notice a pitching tendency when you change thrust.

2. Try to place your main lifting surface (Wings) close to the COM.  This way your Ailerons that control roll, will be  rotating around the crafts center of mass.  This will keep the craft from pitching when you roll.

3. Start with one set of control surfaces per control axis. Ailerons for roll, horizontal stabilizers or elevators for pitch and rudder for yaw.  Don't combine two or three axis with one control surface, that is, until you understand why you might get a strange control reaction.

4. The further away a control surface is from the COM, the more effective it will be and the less deflection you will need from it to pitch, roll or yaw the aircraft.  This tip can come in handy when trying to decrease control surface drag to increase speed, but also maintaining the same control authority.  You can increase the length of the craft (Increasing the pitch and yaw arm) and decrease the surface area of the  horizontal and vertical stabilizers, thus decreasing drag. 

5. To balance and tune the basic airframe flight handling, make sure you do it with all other force vectors turned off.  Lock the engine gimbals.  Turn off the Reaction wheels.  Disable SAS.  Use trim.  This way, when you flight test the airframe, you are only testing the airframe.

6. Check power off flight first.  Is it stable, is it predictable?

7. Then test Power on Flight.  If  under a power change the airframe pitches, rolls or yaws, then your thrust vector is no longer in line with your COM.    At higher velocities (Higher air pressure) a thrust change can induce a pitching moment if your COT is not lined up with your COP.  

8. Until the Devs give us flight control hardware support or just analog axis control, then all of this is pretty much worthless, because right now we can only send all or none for control surface deflection.

9. AND AN AIRSPEED INDICATOR WOULD BE FOUNDATIONAL.

 

I hope this helps

Cheers.

 

Bz

Edited February 28, 2023 by Buzz313th

[JoeSchmuckatelli]

7 hours ago, Buzz313th said:

In the VAB the COP does not move when you sweep

I'll have to go back and look.  I know it doesn't do anything while you are adjusting the sliders - but I thought it updates after you close the window. 

 

7 hours ago, Buzz313th said:

Ailerons that control roll, will be  rotating around the crafts center of mass

So make sure the control surface is behind the COM?  This alone may be a source of some of the problems I've had! 

 

7 hours ago, Buzz313th said:

Don't combine two or three axis with one control surface

Can you break this down Barney style for me?  Neanderthal gamer recognized the words but the meaning is too big brained for me atm!

 

7 hours ago, Buzz313th said:

Use trim

Where is this control? 

 

7 hours ago, Buzz313th said:

Until the Devs give us flight control hardware support or just analog axis control, then all of this is pretty much worthless, because right now we can only send all or none for control surface deflection.

9. AND AN AIRSPEED INDICATOR WOULD BE FOUNDATIONAL

Yeah one of my 'Big Asks' for the team is HOTAS support.  Like the airspeed, I'd also like to see speed presented in KM/H because M/S isn't intuitive for me in a plane.  (It's also just a number I watch with rockets, but I have flown enough in KSP to have an idea of what it means while I'm flying those 

 

Thanks! 

[Guest]

26 minutes ago, JoeSchmuckatelli said:

Can you break this down Barney style for me?  Neanderthal gamer recognized the words but the meaning is too big brained for me atm!

Each control surface should only control one of pitch, yaw, roll, not two or more:

• Wingtip ailerons control roll
• Tailplane or canards control pitch
• Rudder controls yaw

Neanderthal troubleshooting of control surfaces (assumes that the plane is close to aerodynamically neutral, with CoP only very slightly aft of CoM):

• Plane won't roll? --> you need bigger ailerons or more control authority on them (this is very rare, they almost always roll)
• Plane won't pitch? --> you need to move the tailplane back or canards forward, or have a bigger tailplane/canards, or give them more control authority
• Plane won't stay straight? --> you need to move the rudder back, or you need a bigger rudder 

25 minutes ago, JoeSchmuckatelli said:

So make sure the control surface is behind the COM?  This alone may be a source of some of the problems I've had! 

Doesn't matter for roll. The further they are to the left and right, the more effective they will be. I.e. the tailplane or canards aren't very good at inducing roll, use wingtip ailerons instead.

27 minutes ago, JoeSchmuckatelli said:

Where is this control? 

Alt+WASD 

[Buzz313th]

21 minutes ago, JoeSchmuckatelli said:

I'll have to go back and look.  I know it doesn't do anything while you are adjusting the sliders - but I thought it updates after you close the window. 

 

So make sure the control surface is behind the COM?  This alone may be a source of some of the problems I've had! 

 

Can you break this down Barney style for me?  Neanderthal gamer recognized the words but the meaning is too big brained for me atm!

 

Where is this control? 

 

Yeah one of my 'Big Asks' for the team is HOTAS support.  Like the airspeed, I'd also like to see speed presented in KM/H because M/S isn't intuitive for me in a plane.  (It's also just a number I watch with rockets, but I have flown enough in KSP to have an idea of what it means while I'm flying those 

 

Thanks! 

1. I had a quick look and the regardless of wing sweep or wing angle, the position of the wing root sets the position of the COP.  It did not change for me atleast.

 

2.  The control surfaces that control roll should be lined up with the COM.  These surfaces are called ailerons.

 

3.  Barney style..  sure.     Each set of control surfaces should only be allowed to control one control axis.  The control axis are Pitch, Roll and Yaw.  In the parts manager under "Advanced something" for a given control surface you can filter which control axis it reacts to. 

4. Trim is set by using ALT and the control keys WASD.   ALT X resets to neutral trim.

5.  KMH?  LOL, I would prefer knots.   Nice polite troll.

 

Cheers.

 

[JoeSchmuckatelli]

2 hours ago, Buzz313th said:

OL, I would prefer knots

Grin, me too.  Just calling back to some pre-EA shenanigans where we made up names for units of measure and agreed that the devs were unlikely to introduce an analog to Standard. 

 

 

 

 

(I was taught that the US uses 'Standard' the Brits 'Imperial' and the silly French persons 'metric') 

 

[psunavy03]

10 hours ago, Buzz313th said:

9. AND AN AIRSPEED INDICATOR WOULD BE FOUNDATIONAL.

In knots and Mach.

[whatsEJstandfor]

On 2/28/2023 at 7:21 AM, psunavy03 said:

In knots and Mach.

Would nautical miles be 1/10 the size of our own 

[JoeSchmuckatelli]

10 hours ago, whatsEJstandfor said:

Would nautical miles be 1/10 the size of our own 

Big brain stuff, right there! 

[pmoffitt]

I have noticed that the Center of Lift does update once you exit the Procedural Parts Dialog box after changing the sweep of the wings.  (I swear it didn't at first, did something change?) It would be better to get that feed back while moving those sliders.

(That would take a change in code for when the slider changes to call the update Center of Lift function.  Then there are some concerns on how often it updates.  I can crash simple code I have written using WinUI3 and onsize change calls by moving the edge of a Window back and forth as the calls come too fast for the system to process them and then the stack will overflow. Real Time updates can be a pain.)

[Steve7893]

I also have had great success with planes in KSP2.   No issues.   Built one pretty easily with 2 whiplash and one rapier that made it to orbit.