Question DIY Dual-Purpose Sim Rig (Racing + Flight) – 2DOF vs 3DOF Motion Advice Needed

I recently built a 15 Nm wheelbase using a hoverboard motor. It works, but I’ve realized I now need a proper sim rig to really test it in good conditions.
My goal is to build a dual-purpose rig for both sim racing and flight simulation. Later on, I’d like to add a joystick and be able to quickly swap between rudder pedals and racing pedals depending on whether I’m flying or driving.
I’m also interested in building a motion system, but I’m not sure where to start. I’m currently hesitating between a 2DOF and a 3DOF setup, and I’d like to better understand how cost, complexity, and performance scale between the two.
From what I’ve seen, 2DOF systems seem simpler and more affordable, while 3DOF adds vertical motion, which looks like a significant upgrade in immersion. I’m trying to figure out which one offers the best value overall for a DIY build.
I came across MikeyEarly’s project, which I really like:

What would you recommend in my case?
Is there a specific type of motion system or architecture already made here that would work best for what I’m trying to achieve?
I’m comfortable building with aluminum profiles and working with motors, so a DIY solution is definitely within my capabilities.

Thank you, that’s exactly what I was looking for.
Regarding the budget, what kind of cost should I expect for the main components such as aluminum extrusions, motors, and controllers?

I have three hoverboard motors rated at about 15 Nm at 600 RPM. I plan to attach magnetic encoders, like I did on my hoverboard wheelbase, and use worm gears to increase torque. With a 1:5 ratio I expect around 75 Nm at 120+ RPM, and with 1:10 around 150 Nm at 60 RPM, all controlled by an ODrive. The components are inexpensive, so the total per axis would be around 65–70 euros. Do you think this setup is better than using second-hand wheelchair motors?

Also, for a 3-DOF setup, are you using a two-axis ODrive plus one single-axis, or three separate single-axis ODrives?

Hi, 2DOF is the simplest and cheapest way to go and to start. Also is the option with less risk of mistakes that make the whole design/build unusable and if it happens is relatively easy to redesign/rebuild it with fixes. 3DOF with heave is a GREAT step up specially with flight sim BUT you need powerful motors and very good in performance reduction gear. Normally worm gears are very inefficient, large part of the energy is lost in them so try to avoid it. The design, materials and building process needs to be much more accurate thant with the 2DOF.

Here are my three motion rigs I have built, a 2DOF and two 3DOF (one with wormgears and one with ballscrew):

2DOF: https://www.xsimulator.net/community/threads/2-dof-flight-sim.18232/#post-242487

3DOF (wormgear): https://www.xsimulator.net/community/threads/from-2-dof-to-3-dof-flight-sim-motion-rig.18430/

3DOF (balscrew): https://www.xsimulator.net/communit...ild-with-diy-sfu-actuators.19455/#post-256023

With the ballscrew version (my final version) I have to update some little changes I've done to maje the arms more rigid, basically I used 10mm plate instead of 5mm for the arms.

Ballscrews seem more efficient than worm gears, but the challenge is converting rotational force to linear motion. I've seen several approaches, and yours looks particularly well-designed. Do you think a 350W, 48V, 16Nm, 800 RPM BLDC motor would be sufficient for this? I notice you’re using a 700W motor, but it’s less efficient than the one I have. Excellent build, by the way!

I think that motors with 800 rpm are too slow. Think that a ballscrew like mine (2005) have a diameter of 20mm and advances 5mm per revolution. Pairing it with your motor you will get an output speed of: (800 / 60) x 0,05 = 6,6 cm per second.

My motor output force is 3 Nm and is more than enough. With 16 Nm you have plenty of strength. If you do a gear that increase the speed by 3 or 4 (and at the same time decrease the force by the same scalar) you will still have force enough and a optimal speed. Keep in mind to check the performance loss and calculate all twice taking in count the performance loss before buy the gearboxes.

Another way to go is using 2010 ballscrew instead of 2005 and a gear that invreases only by 2 the speed. If it is well designed theoretically you'll achieve ~ 6,6 x 4 = 26,4 cm per second that is very decent