Compact Budget Build that is a bit different

I wanted a triple screen rig but didn't have space for a big one or want to spend a whole lot so I built my own.
The three main things I did to keep the footprint small (for triples):

1. using small screens mounted close
2. having the seat adjustment via telescoping the frame so when I push the seat in, the frame tucks in also
3. a fairly unique motion system which doesn't need extra space for rig movement

I ended up building the frame from 25 x 75mm RHS welded for rigidity. I worked out all my cut lengths so it could all be cut from the one length of material and adapted the original screen mounts to retain the adjustment features an reduce fabrication and materials.
There were quite a few minor revisions along the way but the result is detailed below

Here it is tucked away in time out...


And stretched out for use...
(the hood actually stays on for use but took it off for visibility in the photo)


The seat sliders lock up by levering the tongues against the slots as pictured below. Small packers at one end of the main frame set the slots at a slight incline relative to the ground ensuring my weight in the seat provides initial preload/lockup. Force exerted when braking adds to the levering force locking the sliders even tighter. Doesn't budge at all even under my heaviest braking, but if i take my weight off the seat it is free to slide.


Motion. I couldn't resist giving it a go...

1. keeping it cheap eliminated a lot of the good but costly options (big motors, actuators, power supplies etc)
2. keeping it compact ruled out rig movers and weight neutralising gimbal arrangements.
3. i wanted my seat, controls and screens to maintain their positions relative to each other which typically would require a full rig mover

I felt the key to achieving effective results from a cheap(low power), compact(low clearance) system would be applying force as directly to the driver as possible. My motion system applies force via belts similar to a harness, but rigged in such a way to provide movement force forward, backward, left and right (with heave data thrown in to alter tension so you can feel the bumps a bit too). Targeting the shoulders/back gives good leverage and sensation without having to move the significant heft of my lower half.

It has taken a few trial configurations to find a good way transfer the movement from 2 motors to pull from all the different directions required. Various pulleys, rollers and sleeves and shoulder harnesses were tried and failed.
Findings from trials:

1. No straps near free rotating equipment (as in able to freely rotate beyond 360deg if a motor gets out of control)
2. Only one motor acting on each shoulder. A mix of the two ends up with them working against each other and pulling down on the driver
3. Sideforce from the back was best acheived with the left motor pulling on the right shoulder (and vice versa) to get a good angle.
   
4. Front straps shouldn't cross as this becomes a hassle getting in and out, especially as if they are undone as we have rear straps to get confused with when buckling back in.

To meet these findings I needed to transfer forward pull from the left motor to the right side (and vice versa) without crossing straps. I did so using fairly simple lever mechanism detailed below.



So there is three tubes running inside each other, each welded to their respective levers at either end.The rear levers on each side are then linked to the motors. I was lucky enough to find a broken swingball set and an old shock absorber giving me three tube diameters which fit accordingly. I found some brass scrap, machined up some bearings to make it fit and some steel retention rings to weld into/onto the tubes for the brearings to shoulder up against. Had to use dodgy gasless wire as it was thin material and I had no gas for MIG





Done, few pics in different orientations


It works well, running Arduino Uno clone, SMC3, Moto-Monster motor shield, 2 x 12v 60W motors and a 30A 12V power supply.

The only issue I have to resolve is a clash between my SimHub shakers and Simtools. I can get the port forwarding to work, but Simhub seems to want more detailed telemetry from LFS than Simtools will accept, so I can either get Shakers to work well with poor motion, or really good motion with poor shakers. I really want both, my shakers work really well so I won't go without them. Any help would be greatly appreciated - maybe it is not an issue with plugins for other games?

And other some minor DIY details to round this off:

Potentiometer drive - machined a drive bush corresponding hole thru my gearbox casings...


Machined and welded threaded bush to casing...

Terrible pic of machining radial slot in potentiometer index plate... (i clearly don't take time for photos when im busy building)


Pot mounted in indexing plate... What a godsend... was a nightmare before i added this adjustment


motor mounting and slotted drive bush... (many holes and marks from previous trials...)


Bedspring instead of rod end to deal with angular misalignment.... cable tie to limit stretch (YUCK! it works * shrug*)


test piece for effectiveness of my super cheap bezel-free kit... with small screens I didn't want to be distorting a significant portion of good image with a fresnel lens so i went reflective instead. Fully done now but this test piece at the truck shows the difference well

laid carbonfiber over over a plywood dashpanel, drinks stand and mouse pad. The dash really serves to hide cables holds the suction cup behing my wheel for attaching phone (running SimHub Dash)
Screens still go up and down, handy for when i'm not racing and don't want the wheel obstructing my view

Really interesting concept, I’m a sucker for compact self-contained designs, especially those that get smaller. I have to say though, I’m a little confused as to how the mechanism works. Do you have a video of it moving?

Liking your sim,

Can you explain how the 'bezel free' works, is it a mirror tape?

Yeah no problem - it was a super cheap idea that worked really well!

It is aluminium angle, polished, clearcoated (so it doesn't oxidise and lose its reflective properties) and stuck to the black edge of my screen.
I had seen someone do something similar with mirrors, but on old screens with big bezels which i couldn't replicate on modern screens with thin edges (plus I had no idea where to get custom mirrors made?) I had a think about it and clicked that aluminium is cheap, light and easily polished so gave it a go.

Do some measuring to figure out what will work for you. In my case i needed aluminium rightangles 20mm x 7mm to fit my screen - my screen edges are 7mm, and at 7mm in from the edge, the 2 right angles would intersect at less than 20mm out from the display (16mm was enough, but my screen mounts can rotate so a few mm gap between stops adjacent screens crashing into reflectors if i shift them)
Find the cheapest, lightest aluminium section you can find to do the job at the local bunnings or similar, some double-sided tape, a cheap can of clear and some polishing compund (the stuff which is used with a buff in a battery drill makes life easy)

I already had everything except the aluminium, I bought 1m of 20 x 20 x 1mm channel for about $12NZD, cut to length then split down the back as in pics below:



polish 'em up, clear coat, stick 'em on - job done!

A tip I would add is best results are achieved by sticking them direct to the screen surface if possible and make sure the reflector covers all the black portion next to the pixels, as any remaining will be reflected and it interrupts the illusion somewhat. Pay particular attention on the side screens, parallax becomes a factor as these screened are viewed from an angle and have a clear layer between the pixels and the surface. Simple solution is to check from the normal driving position as you fit the reflectors, ensuring all the black is covered

Yeah good point I didn't showcase that very well. I tried to do a vid tonight but a video struggles to show as this is more compression/movement against the occupant than tossing a seat around - the rear belts are obscured by the driver, but with out the driver supporting them it they don't sit in the operating position so it is a poor display either way. I'll have another go when I get time, need a friendly ghost, but failing that I'll try suspending the straps somehow

Ahhh I see, this makes sense. Please don't stress too much about it, I can work a little harder to understand!

your bezel free solution is really neat !

i have a friend who has a glass company and may be able to cut me some mirror !

do i want a mirror on each bezel frame or one mirror between the 2 bezels ?

I always wondered if I could get hold of a triangular prism and see if I could get that to work

has anyone tried a prism ?

(discussion about reflective wrap here https://www.isrtv.com/forums/topic/11646-mod-hidden-bezels-on-triple-screen-setup/)

I have 2 reflectors back to back where screens meet, so one on each bezel, reflective surface at 90degrees to the display. 4 reflectors in total - 2 on the middle screen and one one the inner edge of each side screen if that makes sense. Your friend could maybe do something in one piece but it would need 2 reflective surfaces, the angles and fit would have to be perfect too - the result would be good though if you could get it right.
I thought about prisms but going down this path would likely lead to linear fresnel lenses instead, as this is what the commercial options appear to have settled on.
I did investigate Linear fresnel lenses somewhat and it would be interesting to try, they are plentiful on Aliexpress etc in a good range of options, you would need to do a bit of R+D to find figure out the right width and focal length.
I had not seen the reflective wrap post, but my first trial actually was aluminium tape. I quickly realised i could polish a piece of aluminium instead and it could be a lot smaller, more rigid and durable which was the next trial piece i show in my first post. My final solution turned out to be a lot easier to handle and arrange than tape/cardboard, would definitely recommend going straight to aluminium section rather than tapes/wraps

I'll try to explain here by focusing on one shoulder and one motor. This photo still distorts the situation as the straps wrap around your body rather than running straight but the following points should locate everything:

• The red buckles on the end of the black straps are sitting over your shoulders
• The rivet where all the straps meet is on your back right between your shoulder blades
• the blue vee of straps joined at the rivet run from center of your back - the left one we are talking about in this example wrapss diagonally down across your lat towards kidney then out the front to left front lever.
• The left front lever is rigidly connected by a tube/axle under the seat to the right rear lever as per the diagram posted earlier (ignore the straps in the previous diagram, that is a slightly different configuration also possible with this setup). When one lever pulls the other does the opposite and loosens. This double acting lever is driven both ways only by the right motor.
• When the left front lever pulls forward (yellow arrows) you feel a push in your back as it pulls you forward and to the left from the center of your back where all the straps meet. At the same time, the rear right lever loosens the strap over your left shoulder, allowing the movement forward and left
• When the right rear lever pulls back (red), it pulls your left shoulder firmly back and to the right and the front left lever loosens it's strap behind you.
• Note that the black strap over your left should runs through the right seat belt hole in the seat. This allows significant force to the side

Along with the left shoulder control described there is of course an opposite setup driving your right shoulder/flank:

• when these systems work in opposing directions, they can provide strong sideways forces/body movement into your fixed seat, and best of all they can but sustained as long as you continue cornering
• when they work in the same direction, the will either apply pressure from the back (accel) or tighten down and back on your shoulders, also sustainable.
• Interestingly, this is a combination of reflecting true G-force (when cornering/sway/roll) and the not so-true Reaction force (when brake/accel/surge/heave/pitch). As I mentioned other configurations are easily achieved, I originally set this this up with as all true G-force, but i found i preferred the reaction force for front/back
• heave is also surprisingly effective in this config, pitch and roll less so (they give false cue of surge/sway if too strong, a shame because they feel great in a big crash)

GENERAL PROGRESS UPDATE:

got my license and plugin for Rfactor2 - this setup is working awesome.

Once I got the tuning center figured out, i was in for a surprise or two. I was expecting good sustained cornering and braking, but i was blown away how it felt skipping over road undulations and totally surprised the first time i hit the rev limiter as a decent speed - i was so used to just the noise and lack of acceleration, it had never crossed my mind that i was missing the feel! Not that it is particularly nice, but of course a very real sensation which was totally absent before. Response is very quick, loving it.

@JBung , thank you! I have a much better understanding of your design now. I have to say it's quite inventive, I would not think to run straps in that complex a configuration, and it's neat how you've been able to effectively get "3 DOF" with only two motors, and nothing but tension!

Sustainable forces are the best. Especially (in my opinion) for flying, since most forces you encounter are sustained for some number of seconds or minutes depending on what you're doing. It's a little sad how many commercial rigs seem to miss this concept.

I was also blown away the first time I fired up my rig a couple of weeks ago. I was fully expecting the kick from the afterburners, the positive (and negative!) Gs from high speed flight, etc. What I was not expecting was the reverse kick from just backing OFF the throttle, suddenly myself lurch forward against my harness as the air resistance overpowered the plane without the AB kicked in. So many little nuances missed without movement.

Really innovative design! I'd love to see it in action if you ever post a video.

You mentioned a clash between Simhub shakers and Simtools. By any chance did you figure out to resolve the clash? I just added rumble motors to my pedals via Simhub and now I'm having conflicts in Live for Speed. Any recommendations?

You are quite correct regarding plugins for other games - no problems with other games once I got the DIY license and plugins for them. No such luck for me with Live for Speed.

Regarding a demo, it is hard to show as unless you are the one in the seat as there is not a lot of dramatic external movement (other than the levers moving) because the straps are already pulled up tight, and they just lean on you and apply more pressure.
It is even less obvious now as aside from all changing to all black belts, I have also adjusted the lever ratio for more force/less movement and run tighter belts - the increase in detail and feel was dramatic.
I haven't tried any other motion rigs, but I am really happy with how this works - to the point where i'm not really seeking anything better.
It did take a bit of tuning/learning to get it right though. Surprisingly slight ergonomic changes make all the difference so the ideal setup is very specific to a persons height/build - for example the height of the point where the belts meet on your back can totally change the balance of brake feel vs acceleration, trading either direction from too much one way to nothing in the other.
I have finally traded out the springs and cable ties for proper rodends which became necessary with increased force of the lever ratio I run now.

other incidental changes since last post:
- i have parts on the way for DIY load cell brake conversion
- did a super cheap speed DIY proportional wind system
- tried (and abandoned) a super cheap dynamic lighting system. FPS tradeoff and distraction wasn't good.
- reconfigured shakers, now have a shaker for each key touchpoint - one in seat, one at feet, one in wheelbase. Great improvement have all key touchpoints active. Wheel shaker effect is biased toward the steered wheels, whereas feet and seat shake in sync like they are part of the same chassis and express more rear traction loss. (previously i had feet front wheel biased, seat rear wheelbiased and no wheel shaker. It was good, but new setup is a big improvement.) Also changed to my own custom traction loss effects in SimHub. I can only vouch for them working great in rfactor2, they do work in other games but I have barely tested
These effects do front lockups, oversteer/understeer, ignore rear wheelspin/lockups because it was just annoying, and are surface dependent so you don't feel lockups on the grass etc, also wet track kills this feedback just like real life.

Front custom effect (suits rfactor2, RWD.):
Front Left:
if([GameRawData.CurrentPlayerTelemetry.mWheels01.mSurfaceType]<>0,0,([GameRawData.CurrentPlayerTelemetry.mWheels01.mLongitudinalPatchVel]*[GameRawData.CurrentPlayerTelemetry.mWheels01.mLongitudinalPatchVel])+(([GameRawData.CurrentPlayerTelemetry.mWheels01.mLateralPatchVel]*[GameRawData.CurrentPlayerTelemetry.mWheels01.mLateralPatchVel])*[SpeedKmh]/100))
Front Right:
if([GameRawData.CurrentPlayerTelemetry.mWheels02.mSurfaceType]<>0,0,([GameRawData.CurrentPlayerTelemetry.mWheels02.mLongitudinalPatchVel]*[GameRawData.CurrentPlayerTelemetry.mWheels02.mLongitudinalPatchVel])+(([GameRawData.CurrentPlayerTelemetry.mWheels02.mLateralPatchVel]*[GameRawData.CurrentPlayerTelemetry.mWheels02.mLateralPatchVel])*[SpeedKmh]/100))

Rear Custom Effect(again rfactor2, RWD):
Rear Left:
if([GameRawData.CurrentPlayerTelemetry.mWheels03.mSurfaceType]<>0,0,(([GameRawData.CurrentPlayerTelemetry.mWheels03.mLateralPatchVel]*[GameRawData.CurrentPlayerTelemetry.mWheels03.mLateralPatchVel])*[SpeedKmh]/100))
Rear Right:
if([GameRawData.CurrentPlayerTelemetry.mWheels04.mSurfaceType]<>0,0,(([GameRawData.CurrentPlayerTelemetry.mWheels04.mLateralPatchVel]*[GameRawData.CurrentPlayerTelemetry.mWheels04.mLateralPatchVel])*[SpeedKmh]/100))

Hi @JBung gotta say i loved the Bezel idea ,so crazy thought , would this work with say 2 x 32" monitors , Thinking mainly for F1 Sim with HALO .. Cheers GYI im going shoping for monitors tomorrow and might need side trip to DIY store now lol , Thanks for posting

It can work with 2 monitors as long as you bring them in on enough of an angle that the edges of the reflectors meet nicely in the middle, and the sim will need to be able to correct for screen angles (angle correction is often built around triples but it might work?). Getting the edges to meet nicely will be important (with triples you hide hide the edge of the inner screen reflector just behind the edge of the outer reflectors, not sure if the same can be done with 2 screens as the join is directly in front of you) .
You may also need to set your in game viewpoint quite close to the halo so that the reflections don't include the sky/scenery (would be more distracting than black space to have reflected scenery in the middle of the halo. Moving closer makes halo wider, until there is enough halo either side of the reflectors nothig else shows in the reflection.
The in game dash could be messy with reflected text, but if you run an external dash (ie simhub on a phone)in front of the game dash that won't matter

Cheers @JBung appreciate the feedback , def helps with my monitor selection research

Good to know, thanks again @JBung!

Curiosity got the better of me again. I bought a single linear fresnel lens from ali-x (300x100xf40) around $25USD +shipping/tax.

My setup is likely a little different as i run small screens up close and wanted to minimise distortion/obstruction of my center screen. My lenses mainly take the image from the side screens and stretch it to hide the bezels rather than blending from (distorting) both. Plus I am cheap and wanted to only pay for 1 lens. To achieve this I split it down the middle to make 2 opposite lenses, the cut edges (formerly the center of the lens) sit against the outer screens.

The trickiest part of the whole thing was mounting the lenses. Best solution ended up being stands the lenses clip into which i made by cutting and folding some aluminium angle as pictured (with some soft foam feet to allow them to be squeezed under the screens and held firmly in place)
Still need a coat of black paint at some point...


Results are pretty good (pics below). Probably not as good as the more expensive off the shelf options with slight distortion of the sidescreen image at the merge point but I achieved my goal of leaving the main center screen almost untouched. In theory a slight performance bonus as i am no longer rendering extra resolution to compensate for GPU bezel correction.
I do prefer this over my previous cheaper reflector mod, purely because the the reflections could produce slightly distracting images particularly with shapes which ran at an angle over the meeting of the screens, as the reflections would be at opposite angles. BUT the reflectors were really effective for the minimal cost.

oh yeah and I converted my G29 pedals to 100kg Loadcell.
This worked great, using INA122PA (so direct pot replacement, no need for arduino or extra software etc) based on a combination of info from @RacingMat and this tutorial https://imgur.com/gallery/gOjAf#SXX5zrY