Jonota - 3 DOF (or is it 2 DOF w/Traction loss?)

Hello all!

I posted in the newbie section. I know myself well enough to recognize that the only way I can get a project off the ground is to jump in... so I've done just that. Feel free to attempt course-correction - I'm kinda stubborn though, and have knack for making things "work" vs. doing "the best".

I'm struggling to remember the 2 forum members that got me started down this path, but I think it's Dr. ADHD and DMax, along with perusing other build threads and seeing input from Avenga and Noorbeast. I was initially just going to get a DOFReality H3 - but you guys have convinced me to give it a shot myself.

I've sourced ~60' of 1.5" 14ga steel, along with some 2" that loosely fits around it for some adjustability for the foot pedal mount and steering wheel. I've obtained a large full-swivel caster and 3 large fixed casters for my traction loss, 6 heim joint/rod rends, 3 IBT-2's, an UNO replica, a 12v kill switch, have a large amount of miscellaneous bolting and various metals (sheet in 2 gauges and square tube in several sizes), an entire Chevrolet driveshaft for the universal joint (had it sitting around for years). I'm borrowing a "racing seat" from a project car of mine - so I'm trying to keep in mind I need flexibility there. And lastly, I went "full cheap" and will be attempting to use the 12v motors from the Am@zon car winches (15a max, intending to use car battery and trickle charger). I'm open to changing motors later, or if they don't work out. Already eyeballing the distressing lack of mounting options and 1-piece shaft.

I already have a GT Lite setup with Moza direct-drive wheel w/F1 wheel adapter, Moza pedals w/clutch, and TH8 shifter (I'm "meh" on it, thinking of getting the Moza version) and a pretty well-decked gaming PC. While I have multiple screens I'm thinking the majority or the "full immersion" will be with VR - currently just a Quest 2, but I'm a pre-order on the Omni One setup so may end up with another headset (if it's higher quality).

I have relatively limited room - significant length available but a maximum of about 36" in width (it's 40" from desk side to wall). I can possible do some moving around, but I'd rather keep a very compact footprint. Keeping in mind that I'm currently a "no-motion" racer of relatively limited-to-zero talent at actually racing, even a small amount of motion is going to really increase my fun factor. I already feel really immersed with VR - I just want that extra.

While I don't intend on agonizing for long, I'm trying to nail down my intended parameters. I'm currently looking at ~4' long, 2' wide, front-mounted motors for accel/decel/lateral motion, and rear-mounted for traction loss (similar to the good Dr's setup). I do want to avoid it being extremely tall, but I recognize there's only so much that can be done.

Things I know I'll need help on: Any terrible outcomes from the above-listed setup, or key tips/learnings from it to incorporate. Also, while I understand PID controllers (I'm in the nuclear industry and we use 1970's analog versions), I have little to no experience with anything like Simtools. Also, understanding the lever arm calcs for relatively realistic movements vs. strength of the motors is a weakness.

I'll post some pics and initial drawings later tonight. I fully intend to start welding metal together starting tomorrow.

That's one of the specific questions I had. How much movement is "adequate", and is that movement different for the different axis? I really only have 6" of movement allowed on either side in the space my rig is to sit for traction loss - so the "felt" movement in seat will be less than that. That's roughly 300mm total movement, which "sounds" adequate. I also imagine that with less movement you can gain response speed/accuracy.

Your rig design looks very similar to how I've done mine, I inverted my pedals as that's what I drive IRL and that also doubled to add extra leverage for the motors. My rig is compact for a full motion, being only 700mm wide on the base which the T/loss moves side to side on. The motion feels pretty sharp and snappy with plenty of power.

300mm should be plenty for T/loss, I think mine may be around 250 and it feels plenty, the gap between the cockpit and middle frame for 2 DOF movements is 70mm all round, this also feels plenty for surge and sway.

2 decent days to fabricate, coming along on the main apparatus.

For finding the center of gravity for the upper frame - I have a sliding seat, and the foot petals and steering assembly can/will be able to move fore and aft. Is the balance important enough to need pinpoint accuracy? I can add little brackets to put weights on to help balance if/as the configuration is changed.

Nah doesn't need to be pinpoint accurate, just get comfortable with yourself and all peripherals, find the centre, then if you decide to adjust the pedals, wheel or seat a bit later for comfort it will be fine.

I'd consider lowering the height of your top frame, I'm assuming that horizontal bar at the top of the u-joint is the start of the cockpit?

You want the cockpit as close to the u joint movement as possible to avoid weird movement / extra strain on motors.

Finished up main assembly. Ready to start rigging up the motors. I discovered a few potential benefits of the jack motors: they have 2 motor cutout travel limit switched, and the block for the opposite end of the jack should make for a good sturdy mount bracket point.

How do we go about getting the SimTools? Hopefully will be needing it in the next few weeks.

Haven't had much time, but got the front motors mounted, arms setup. I wasn't sure what to shoot for in terms of travel, so I tried ~3" CTC lever arms and they backdrive bad even without me in it. Surprises me considering these would be holding a car!

What's a reasonable low-end travel? I suppose it all depends on the setup. I did get in it and full travel felt like a LOT of movement. It WAS able to easily tilt me but without active motor drive it was very precarious.

Welp - I cut the CTC down to 3/4" and still had terrible back-drive. I was able to get in it but any motion and it wobbled all over the place.

I have 2 600w tarp motors on the way. I'd still like to salvage use of the 2 smaller motors. I was thinking that traction loss doesn't have to have any holding power, so they might function there. If 1 doesn't work, I was thinking I can run them both in tandem driving the traction loss point. Thoughts?

Looking at your photo, the motors could be mounted a lot better which would significantly reduce the effect of backlash.

You have them mounted very narrow, allowing a lot of side to side "roll" felt due to backlash.

I think if you mount them further out, towards the limit of your rig, the effect may be greatly reduced.

Could you elaborate? They're not permanently mounted, I'll grant you, but both ends of the shaft are supported and didn't give when testing. Are you saying that if I move the motors to the outside of the rig they would be more stable? I'm not challenging that, it just doesn't make sense to me why that would make a difference. Or perhaps I'm misunderstanding what you mean.

Think of it like a seesaw, your pivot point is in the middle. If you grab the seesaw at the end and move it up and down, the movement is slower and more controlled, grab it closer to the pivot, and the movement is more exaggerated. Its difficult to explain but I believe it will magnify the feeling of backlash.

Another example where you may be able to see this, is your rig will more likely have excess movement on the side to side (roll / sway) than the front to back (pitch / surge).

See the photo below for what I think will help.

I think I understand. Do you mean simply to move the motors OUT, or to flip them around to the other side of the metal they're on so the shafts face outward vs. inward?

Yeah if you turn them around, so that the lever arm is attached on the external side rather than inside of your frame.

Tarp motors installed, I swapped from internal to external connections, have one of the scissor jack motors handling traction loss, seems to do just fine moving the rig side to side. Next up to rig up the potentiometers. I planned on simply angle brackets, and a glob of hot glue from motor center to pot shaft.

After that is wiring everything up. I've seen a lot of enclosed boxes - is that really necessary (and isn't it worse for heat/airflow)? I was thinking just a section of plywood at the front with all of the electronics mounted on it, and then a fan pointed at the entire rig.

All wired up - sort of. Working on getting arduino programmed, and noticed that I miswired slightly in that I used pinout 8/9/10 for PWM motor 1/2/3 respectively. Is there a way I can adjust the coding to allow for using these 3 instead of having to unsolder/resolder those 3 connections? Seems like that would be simple...