My first build. Another budget flightsim 2DOF rig.

Another budget flightsim rig build in this thread. I don't think there'll be anything radical or ground-breaking here but hopefully someone might find something interesting.

Since I've never built anything like this before the priority will be to keep the cost right down, at least until there's some sign that it might actually work.

Inspired by rigs from @Ronan Design here and SCK Gaming on Youtube. Thanks to both for sharing their knowledge and experience.

So it's going to be a simple 2DOF job with perhaps a yaw axis added later. Made of wood and looking something like this:

And I've made the base! Timber is all 95x45 or 150x45 on a 4mm plywood tray. The motors are high enough for their levers to rotate continuously without hitting anything so hopefully I won't need safety stop switches. The joint in the middle is a drive shaft from a neighbour's scrap car. It's very greasy, hence the plastic bag.

Here's a proof of concept electronics rig. No point in going on to the expensive stuff if I can't get this to work. Fortunately it actually seems to work. DCS feeding Flypt feeding SMC3 feeding a couple of old RC servos.

https://youtube.com/shorts/_aXbukVyYe0

This a re-post. The previous got deleted for apparently being all in capitals.

These are the motors I'm going to use. They're 24V wheelchair motors that can be found on Amazon or Ali-Express. I've seen them used successfully on other builds. I made some 3D printed hall sensors for them. Don't know how well they're going to work yet but they look kinda funky.

My motors have a 17mm shaft with a 6mm keyway, not the easiest thing for a DIY constructor to make arms for. My solution: a hub to take two 6mm grubscrews to bight into the shaft's keyway and a separate bolt-on arm. Hopefully it will be strong enough.

Working on the assumption that it's easier to shorten the CTC than to lengthen it, I've started out at 110mm. I suspect I might end up somewhere around 75mm. The motors seem to have plenty of torque to handle the longer length though.

Seat platform done. I'm pleased with how rigid the pedal extension is with the two metal stays on the bottom.

I'm not having any joy with IBT-2 controllers. I bought two, one was faulty to start with, the other shorted out after half an hour of testing with no load on the motor. So I bought two more and one of these has decided to only work properly in one direction. I know these things are cheap but.......

Time to buy something better or am I just unlucky?

So mechanically I'm close to finished now. Seat and controls all fitted and it's really comfy to sit in.


I've still got some electrical issues to sort out though. It all works with the SMC3 setup utility and the feedback tracks the target well, but the motion is not smooth. The motors seem to pulsate at about 1Hz rather than moving at constant speed. The IBT controllers keep packing up too so I'm going to re-design with a Sabertooth and see how that goes.

So more progress made and I think I'm nearly done with the electronics.

This photo is obviously before tidying of wiring etc but it seems to be working well. The mess on the breadboard in the foreground is mainly just a bit of RC filtering on the motor feedback. It's quite effective at quietening down grumbling and twittering motor sounds. Not so necessary with the new Sabertooth as it was with the IBT2s but still worth having.
The usual Sabertooth seems to be the 2x32 but I couldn't find one those in stock locally so I bought a 2x25 instead. As the Sabertooth amp rating is continuous and the peak current is much more I think the 2x25 is well up to the job. However I made the incorrect assumption that all the logic driver stuff would be the same between the 25 and the 32. It's almost the same, but not quite. The 25's max baud rate is only 38000 which seems too low to use with the SMC3 packet serial version, so I'm limited to the simple serial but that's ok I guess.
Because of the apparent possibility of motor runaway with the simple serial I decided some ultra reliable end stop switches would be a good idea. It's quite hard to design a switch that can be hit by a runaway motor and survive so I went for an optical coupler arrangement.

It works well. The coupler releases a relay if the orange quadrant stops blocking the beam and the relay not only disconnects the power but also shorts the motor winding to form a brake. It stops very quickly.
So onward with tidying wires and making that breadboard into something permanent, and then I'm going flying.

So this is the part of the build thread where I come on and rave about how wonderful my rig is and how it was worth all the effort and how happy I am. Well, sort of... Maybe.
For the last few weeks I've been testing and tweeking and refining, along with a little bit of reinforcing here and there, and lots and lots and lots of mucking around with computer settings. And it works! And it's amazing how the human brain (or at least mine) can be fooled into accepting that a little bit of movement in a chair is the same as wanging around the heavens in a plane or helicopter. Even rolling over upsidedown and doing aerobatics can feel pretty convincing, or at least not unnatural. I have to remind myself that my hair is not hanging upwards and I'm just in a wobbling chair. And I'm actually pretty nervous of crashing because it can get quite violent.
So in terms of what can be achieved for a flight sim with a few hundred pounds/dollars and a chair that tilts with only about 20 degrees of movement, my expectations have been smashed. I am amazed.

BUT

There's a problem with my rig which I noticed when I first sat in it and has resisted all attempts to fix. It's great when it's moving around with a bit of oomph. The motors provide plenty of grunt and the motion is smooth and accurate. However when moving slowly I can feel the rig stepping. It's as if there is insufficient resolution in the position control. Rather than a slow, continuous movement I feel a hundred little steps and it totally spoils the illusion. Maybe in a car racing sim it would be fine as there's presumably always plenty of road bumps and vibration effects to mask the steps, but in a flight sim it's rubbish.
I have spent many hours wondering if my home made position sensors are to blame, or is 10 bit resolution from the computer enough, or maybe it's the 8 bit link to the Sabertooth, or mechanical stiction in the CV joint, etc etc. Gradually I've eliminated all of these and I'm now pretty sure the problem is the cogging effect of the armature passing the magnets in the motors. It seems my wheelchair motors might not be a suitable choice after all, but they have apparently been used successfully on other home builds. Maybe mine just look the same.

So I've got a bit of a love/hate feeling towards it all at the moment. In the right scenarios the rig is fantastic, way better than I could have hoped for and I absolutely love it, but I hate the stepping and hate that I cannot find a fix for it that doesn't involve another round of searching for motors and butchering the rig to fit them. And summer is arriving, time to get the boat out and go sailing. Maybe come back to this in the Autumn. Unless of course anyone has any suggestions....

Hello Noorbeast. Thanks for replying.
I started out at 110mm. I've since moved in to 90mm hoping to increase the motor movement and reduce the stepping. It has helped a bit. I've got another set of holes at 65mm and presumably using them would help a bit more but I'm reluctant to do this because the range of movement will be so much less. People seem to say that with VR you only need a tiny hint of movement but I find I rather like being flung about a bit. I don't fly airliners but mainly WW2 fighters and bush planes.
I have diagnosed the cogging motors by putting my Sabertooth into RC mode so I can drive it with a stick on an RC transmitter. No Arduino, no feedback, or anything like that, and the motion is still stepping, so it seems pretty conclusive. I seem to remember reading once that you can de-cog a motor by energising the coils in some way but I can't find anything about that now.
When I started on this project I had no idea how it would turn out and whether I could build anything that would actually work and interface to a computer. I was therefore keen to keep the cost down and not end up with lots of expensive components that I couldn't get to work. This seemed reasonable at the time but it has ended up costing me more than if I had just bought good stuff to start with. For example I bought no less than five IBT2s before I finally gave up on them and bought a Sabertooth. Now that I know my rig can work, and just how good the experience can potentially be I'm quite prepared to throw any reasonable budget at getting it perfected. It would be a bit of a heartbreak to scrap all the stuff I built around the current motors but I'd get over it. Maybe some linear actuators are the answer. I'll have to read up. Perhaps I should click on the ad I can see right on this page for something called a PRS200 from Prosimu.
Another possible solution might be some sort of Buttkicker sound/haptic addition which might be able to mask the stepping. I have tried adding noise to the rig pose and it does help but feels, well, noisy.

Yes, running SMC3, the version that does simple serial to my Sabertooth 2x25. I did try the packet serial version too as it has better speed resolution but it made no difference. The preset waveforms generally run fine with the feedback and setpoint lines matching well on the PC graph if the P gain is high enough. I tend to keep the P gain rather low at about 120 because it softens the response a bit a protects the rig if I accidently tune some huge noise effect. I keep the max PWM down around 110 too. I've been at all the settings in SMC3 and none have helped the cogging issue. It's difficult to provoke the cogging with the SMC3 app. The sine and triangle waves are fast enough to cover it and manually moving the position slider isn't subtle enough.

I just spent an hour setting up Microsoft Flight Sim with the rig and does work fantastically well. Great fun bouncing around in the turbulence over St Barts and doing deliberately bad landings just to get a good bump on the runway. A slow, gentle roll in clean air though.... not so good.

Hey there
I too have the same issues,in dcs,slw pitch and roll movements are not pleasant in my 2 dof rig.
I feel the same stepping effect you speak of,I am also battling backlash issues as well.

I tried the tramp spring idea which worked but eventually overheated my motors so just ordered the dampers from aliexpress as supplied by dof reality,so will see how that goes.

Be keen to see if anyone here has a solution.

@noorbeast Tuning Center? Would that be a Simtools thing? I'm using FlyPTMover. I did install the Simtools engine at one point and had the test output sliders working but as I remember the cogging issue was still present.

The motors I have, which are these, and weren't on special offer when I bought them, doh,



are not worm geared and backdrive fairly easily. If I turn off the power and let the rig sink under my weight I can feel the motors cogging a little. I'm pretty convinced this is the problem.

I'm going to transfer my good pedals and force feedback stick from my old pit onto the rig and go and try some online dogfighting in DCS and maybe IL2. I think the rig will be awesome for this as it's mainly 'spirited' movement. I wonder if the sense of motion will improve my kill ratio. I wonder if the plane will shake when the guns fire. That would be cool. I could forgive some stepping for that.

@Richardamo1 I'm sorry you're having the same issue, but I'm also a little relieved that it's not just me. I've tried searching round this forum with keywords like cogging and stepping but nothing much comes up except in relation to force feedback steering. What are the tramp spring and dampers you mention? I've got a fair bit of gear backlash but I got around it be adjusting the rig balance so my weight is pressing down on the motors and keeping the backlash tight even at quick direction changes. The motors have to work harder to lift me but they don't seem to mind. I assume this spring you mention does something similar and hence your motor overheat?