#helipopper

For your first question, I think this is simpler than you’re making it out to be.

Watch the yellow debug ray on top of the rotor. That’s the position and direction at which the rotor is applying force on the helicopter. The length of the ray shows how much force is being applied. 

I use AddForceAtPosition to add some force at the rotor’s position and in the rotor’s up direction. When I tilt the rotor, this force follows its tilt. You can see this at the beginning of the video where I tilt the rotor around and ray changes.

That’s all I do. With that force physically lifting the helicopter from the rotor’s position, as well as vectoring that thrust, the helicopter can naturally pitch and roll as a consequence of the forces being applied. I don’t ever need to manually apply torques to rotate the helicopter because physics is doing it for me.

Welcome to the wonderful world of aircraft design. Designing your aircraft, be it a plane or helicopter or whatever, means making it easy to fly and reducing those kinds of forces. It’s why helicopters and planes look the way they do. If they didn’t, they wouldn’t fly very well. I learned a lot about aircraft design while trying to make something flyable using SimpleWings.

You could do AddRelativeTorque (9 times out of 10 it’s what I do) if it’s good enough for your purposes. It’s less realistic, but you should also ask yourself if that actually matters and if it making it more realistic actually adds anything of value to the game. 

For the record, I did use AddForceAtPosition for the tail rotor on the more realistic version of the helicopter, but I also was very careful to make sure that it was in line with the center of mass of the helicopter. If it’s not, then it’ll cause the helicopter to bank slightly because it’s applying force off-center. Sometimes that’s not completely possible though and it just means you will have to apply cyclic to counter that motion and that’s just how that helicopter is gonna fly.

Hello! I wrote an old post about this on my blog. It gives a high level description of what I did. Here’s what it looked like in action:

The wings aren’t used for the rotor. The rotor’s lift is applied through an AddForceAtPosition on the rotor hub. This is the simplest way and I think should be plenty realistic for whatever you might be working on. If you’re looking for a more accurate simulation than that, then I might not be able to help.

The only thing I use wings for is to represent the body lift/drag and the stabilizing surfaces on the tail.

Though, that does make me curious what would happen if you tried to do rotor lift by just attaching wings to the spinning hub. It probably wouldn’t work without some modifications to the wing code. Having just tried it, it indeed does not work due to the way Rigidbody velocity is used in wings to calculate angle of attack.

I’m not entirely sure what you mean by “the correct way 2 engines in both wings for an aircraft”. 

If you’re talking about having more than one engine on a plane, I’d say unless you have a specific need for asymmetric thrust, only do the typical AddRelativeForce on the whole aircraft and don’t even bother with trying to simulate multiple engines. If you absolutely must simulate multiple engines applying thrust accurately, then you would use AddForceAtPosition on the each engine. 

Keep in mind though that you will likely end up creating torque on your plane which will have to be trimmed out. Again though, using the AddForceAtPosition method is not something I’d recommend unless you’re going for a real flight simulation approach, because it will significantly complicate your control systems and requiring a lot of tweaking to get the desired effect.

In a video, MagzTV accidentally described how a helicopter flies in an amazingly intuitive way that I've never seen before. Watching how the War Thunder helicopter flies I noticed that my helicopter was wrong in EXACTLY the same ways. After seeing the explanation of why it was wrong and how to make it right, I was very curious to see what might happen with some simple changes to an old helicopter of mine.

There were two main changes I made. The first was that that all forces were now applied at the positions they would be in real life. This means that the rotor applies lift above the center of mass, pitching and rolling is induced simply by tilting the rotor blades, while yaw is induced through a tail rotor. The second was that I brought in the SimpleWings from a previous project and added wings in reasonable places to represent body lift and the small wings on the tail.

The results are pretty amazing. It's a night and day difference and the helicopter now flies similarly to how is described that a realistic one should. With just these two changes, a great many properties of helicopter flight just emerged naturally as a consequence of the forces involved.

The main example he used was how rolling onto the side did weird and artificial things in the fake physics, but with realistic-ish physics, once you pass a critical threshold you just kind of lose control because of the way the forces are being applied. You can watch my inputs and it's just doing it all on its own once I passed like a 45 degree bank angle. It was super neat.

Another example is how the helicopter is kind of self-stabilizing to an extent. With the center of mass below the rotors, it generates torques on its own to put the helicopter into an almost hover. It won't hover precisely or anything, nor would it in real life, but there is still a sense of the helicopter always trying to correct itself just due to how its forces are applied.