I'm trying to model a spitfire as closely as I can (within certain approximations of course).
I have gotten a stable aerodynamic model by following the tutorial series, using the spitfire's wing span (~11m) and shapes, however I had to set the mass of the plane rigidbody to 300kg for it to become aerodynamically stable. When I set the mass to it's actual value (~2500kg) it becomes greatly unstable. I'm wondering if it's worth pursuing the actual mass or if I should be focusing my attention elsewhere. The main reason I'm doing so is because right now it's almost "too stable" and flies more like an acrobatic airplane instead of a heavier spitfire. Thanks! Really loving AO so far.
You have discovered the effect of advance ratio on propeller performance! Advance ratio is defined as the ratio of the forward speed of the aircraft to the tip speed of the propeller. As advance ratio increases (high forward speed or low prop rpm) the mean angle of attack on the blades decreases and at some point it will become negative, producing a net drag from the prop instead of thrust. At this point you have effectively a wind turbine on the front of the aircraft absorbing power.
The way past this is to increase the pitch angle of the propeller blades and/or increase the rpm of the propeller.
Note that at 3000rpm that is 50 revs per second and at the unity default fixed time step of 0.02 seconds that is one rev per time step so you would definitely have to use a rigid body (non kinematic) propeller to avoid interpolation errors. At the AO default fixed time step of 0.002 that would give 10 steps per rev.
One other thing is that you may wish to initially test your aircraft with an AO 'Simple Thruster' component (just add a SimpleThruster.cs script to )that just provides a force directly on the rigid body. This would allow you to sort out all the other challenges without worrying about propeller performance characteristics.
I just recently did some work on speedboat propellers that goes into a bit of propeller theory written up as a case study, see https://www.aerodynamicobjects.com/_files/ugd/28492b_86278d4cee3944e992afde104323a793.pdf Propellers are a bit of a dark art and its not as obvious as you would hope trying to choose the right one for an aircraft (or boat)
Enjoy experimenting!
Cool that you are making a spitfire, got to be one of the most elegant shapes in the sky. Regarding your mass issue, there could be a number of things going on here. Firstly, the actual mass of an aircraft should not affect stability in the first instance - it is the location of the centre of mass that matters. But increasing mass will increase the moment of inertia, depending on how you have the moment of inertia set up on your rigid body (auto from colliders or manual). Have you had your spitfire on a gimbal to start with as in the tutorial development process? This is the best way to debug stability issues.
The other issue with increasing mass relative to a given wing area (wing loading) is that the plane will need to fly a lot faster to get lift = weight. This may lead to numerical issues if the Unity fixed time step is not small enough (plane may move a long way in one time step ). It will also affect visualisation tools like flow particles if you are using them.
Also if you are going fast with a propeller aircraft, the propeller rotation rate has to be really fast and you have to use a non kinematic propeller for this rather than the kinematic one used in the tutorials. The transport plane in the demos uses non kinematic propellers so you could try those.
So in summary, increasing mass should not be a problem in itself, but it may be leading to other probles to do with numerics as the flight speed goes up. There also could be other numerical things to do with Unity that makes things looks unstable. It is often hard separating what is genuine phsyical instability and what is numerical instability in Unity.
The other advice (apart from always starting with tests on a gimbal before trying to fly) is that try increasing the mass incrementally to understand where things are going wrong. If things just explode immediately you dont get a chance to learn.
Happy to look at your model if you cant progress - would be nice to fly a spitfire!