Question Budget 2DOF rig for Microsoft Flight Simulator VR - advice needed

Hi, I fly in Microsoft Flight Simulator 2020 in VR (Reverb G2 - WMR). I can't justify the cost of an off-the-shelf motion simulator, so Iam assessing the feasibility of a DYI build, and found this forum with so much information i'm lost. I have a few very specific noob questions I couldn't find the answer easily, that I would appreciate answers to, and any relevant advice:

1. Has anyone done a successful DIY motion simulator for MSFS and VR?
2. Is motion compensation for WMR OpenXR VR (not SteamVR) possible at all? VR needs to know and disregard motion simulator motion. If yes - how?
3. Is there a software that works with both MSFS and WMR VR for such a project? Can it transmit cockpit position from MSFS to the motion simulator and motion simulator's position to VR or whatever motion compensation solution above?
4. What approach would be the cheapest and easiest to build? I'm not looking for a lot of power, it can be modest, but I'd like about 40 degree roll (20 each side) and whatever I can manage for pitch. Doesn't have to be pretty, I'm in VR anyway. I don't have metalworking equipment or skills, but wood or PVC piping should be good. I also have 3D printer, if that is relevant. Will need to mount yoke, pedals and throttle.

I appreciate all advice.

Thanks for the pointers. MSFS is a bit different than other VR games though. With Reverb G2 WMR it uses OpenXR (not exactly OpenVR?) WMR driver, and though it can use SteamVR OpenXR it't so bad it doesn't really work at all. So I'm not sure the OpenVR solution actually works with WMR in MSFS. I'd love to get a feedback from somebody that has done it already, before committing on expensive and time-consuming build.

I haven't received any replies yet, but from what I read pure WMR OpenXR motion compensation may not be possible. However you were saying what I quoted above, that VR is best with precise small movement, and that with small axis movement I could get by without motion compensation at all. That sounds interesting. I would love to hear from someone that did motion simulator that uses small movements in VR for MSFS without motion compensation, but so far I can't find such a build. I hope someone shows up and tells me if it really is good enough, what the feeling is with rolls larger than the platform can do, and if it's worth the effort to build a system with small axis movements without motion compensation and how realistic does it feel. First-hand experience would be nice. It's an expensive build to just try and see... Also need to figure out what motors and setup would be good tor this.

Interesting. But it looks like most of the builds here, including yours, are racing-sims oriented. I'm a purely flight-sim guy, and not military like DCS but MSFS with mostly calm and gentle flight (well, turbulence happens of course). So I wonder how well would that work with flight sims, where angles are much more than in cars. What intrigues me is small angles that can be used without motion compensation open a way for me to do this on my Reverb G2 WMR. But how small the angle should be to feel realistic in VR and still not use motion compensation? At the same time not too large to be beyond ablity of these motors. That's the question that I'm wondering about. I would put yoke andpedals inside the movable platform as well.

@Ronan Design , I'm using a Reverb G2 myself. Have you considered a g-seat setup of some kind?

You can get away with very small motors, even servos which may save cost. If you do opt for a motion-integrated design, the movement with g-focused setups should be much smaller so you don't need to (and really shouldn't) use motion compensation.

It may also be more appropriate with the "gentler" nature of general aviation aircraft, since you're going to be working from sustained slow movements as opposed to snappier turns and bumps.

That all being said, I haven't used the MSFS plugin yet, so I don't know what the g-force output looks like.

Thanks. I don't even know what a "g-seat setup" is yet, but I will read up on that. Can you please answer a few more questions for me? Did you try any flight simulator on your rig, and soes it feel realistic in VR on Reverb G2? Doesn't matter which simulator, I'd like to find out in principle, if that is a good way to go. Does it feel good and realistic enough to be worth investing time and money into building a small-angle 2DOF platform?

Gentle co-ordinated flight does not require allot of roll in a sim, a bit more pitch and of course heave is a bonus though for realism. Don't have a G2 so won't comment further.

I'm such a noob. I still don't have an understanding of what exactly heave is and how it can work in a 2DOF system, for example...

I (for now) exclusively run DCS on my rig (my performance in MSFS in VR is... abysmal right now). "Realistic" is relative. It's been well over a decade since I gave up pursuing a pilot's license IRL and switched to simming. Nothing compares to actual flying, there's just way too many forces acting on the body, the inner ear, etc. That being said, every little bit adds the "immersion factor" and can help you feel more of what's going on in the software, and consequently help you fly better.

Think about it like this:

3 Screen adds way more immersion than a single screen
VR adds WAY more immersion than any flat screens (not an opinion shared by all, but your a VR user as well so I suspect you and I are on the same page here)
Tactile Transduction (seat shakers, bass shakers, etc.) will add vibration to your rig to simulate engine RPM, and can be tuned further to simulate the vibratory forces from things like ground bumps, stall buffeting, afterburners, weapon releases, etc.
G-seats use moving panels, air bladders, etc. to increase pressure on parts of your body to simulate your body being pulled by g-forces in various directions
Belt tensioners apply tension to a belt or harness to simulate "negative" Gs - your body being pushed up into your seatbelt while diving, suddenly applying air brakes and decelerating, etc.
Seat movers will apply small motion to just you sitting in the seat, which can add fidelity to turbulence, acceleration, deceleration, etc.
A 2DOF rig will give you roll, pitch, and some acceleration, deceleration, and some 'heave' bumps
A 3DOF rig will give you roll, pitch, some accel/decel, and more "true" heave OR some yaw, depending on how you set it up
A 6DOF rig will give you more of all of the above, plus some more lateral motion in all directions

All of these are additive, and many can be combined 'a la carte'. Each little bit you add will give you a little bit more fidelity, but your mileage will vary with each one compared to others. For me, the sweet spot right now is a 3DOF seat mover + g-seat components + belt tensioning + tactile transducer + VR. As good as "real" flying was? Nowhere close. But MILES better than just running joysticks on a flat screen!

OK, I see. thanks for that. I actually made a DIY dual vibration transducers connected to MSFS via mixing specific software that feeds it "events" like touchdown, ground roll, gear drag etc. and general MSFS audio through a high-pass filter (to add engine rumble, and other feedback where there is a low-frequency component). DIY transducers added a lot to the immersion, much more than I anticipated. I hope in combination with 2DOF it would make another large immersion step.

I now want to add some movement but YAW VR2 is way too expensive for a toy, and they don't even know if motion compensation will work on WMR/MSFS. But I can do a simple sub $500 DIY build, wood/plywood based. So I'm trying to figure out the best bang for the time/money invested.

I think the lack of WMR OpenXR motion compensation rules out any large-degree platform (similar to YawVR/VR2). But small movements that do not require motion compensation give me hope. Not sure about G-Seat, as with general aviation it's not as important as military or aerobatic. But 2DOF should be doable.

I'm now thinking about hoverboard hub motors, if I can find a broken one, free or cheap. They are 36V and have a lot of torque, and have an arm is unnecessary or rather "built in" a wheel - just need to drill close to the edge and connect a push rod. Maybe if I can figure out the correct motor drivers and PSU it would be workable. Are there special requirements for Arduino board? I have a few Arduino Nano lying around, are they compatible with Simtoolkit?

I found that and it looks good, but it seems to only support up to 27V I think. I wonder if I can use it with a hoverboards hub motors. I may salvage 2 of those free, and they are 36v. Maybe they can work on 27V and generate enough torque... Generally they seem like a good idea as they are quite powerful and should be good enough to move a seat. And they don't need an arm - just connecting a push rod to the drilled hole at the edge of the metal wheel should give enough leverage. Worth trying, I think. What do you think? Maybe I'm missing something, as I don't see people using those much...

Thanks again, I will ask specific questions there.
A very noob general question then: as much as I want to, the seat frame/platform can't be perfectly balanced. Let's say it's in neutral position, without any forces applied to the motors (I'm flying straight and level) So what is preventing the pressure from the whole badly balanced rig to rotate the motors to whatever position? Or are the motors rigid when energized and are constantly applying power to keep them from rotating? Or would it rotate hall sensors and make the Arduino/Driver board to apply counter pressure? Just trying to understand the basic principles. I think this question is too basic to be explained in FAQs, but it's not self-evident for me. I just tried my daughter's hoverboard (not the one I'm going to use ) and when energized, wheels can be rotated by hand, unless you apply power by rotating the platform, then it's very powerfully drives one way or another.

Thanks. I understand those points. I just wasn't clear what's happening if there is no signal to rotate the platform, but the platform itself pushes back on the motor by its own weight. so, as far as I can guess what will happen is the encoder would start rotating back and Arduino/driver would push back with the motor for it to get back to the desired location. Of course motor should be able to easily overcome the weight and push back. I just wasn't sure I understood what I read and saw correctly. Will read and watch some more...