2DOF with traction loss in aluminum

Like many I got the crazy idea to build a DIY motion rig after watching YouTube and reading these forums. So much great info and so many great individuals.

I decided to build in stages to spread out the cost and allow myself to continue to get some driving in from time to time.

First stage was to build what will ultimately be the top frame. That would allow me to better enjoy sim racing while I learned and planned the rest of the rig. My original plan was to source and build out of steel but after digging around it appeared that many of the high end rigs were using aluminum extrusion builds. I also liked the idea of it being very versatile and “future proof”.

I designed and ordered what I thought would work for the top and mid frame. It turned out to be a good start but will be making some tweaks as the phases continue.

I’m using my Xbox and borrowed the basement workout tv for now. I picked up a Thrustmaster wheel to get me going. I will upgrade this equipment as part of my phased approach but needed something as placeholders.



Next I moved on to building the mid frame. I started collecting parts rapidly. The famed crab pot motors, h bridge motor controllers, Arduino, hall sensors, rod ends, a u joint, various threaded rods and other hardware. I built the mid frame out of extrusion and a mocked up of the top frame out of 2x4’s. I designed and 3d printed some test shafts and ctc arms. Various stages of mock ups, testing and design iterations took place.

The u joint has a little rotational slop. I was a bit surprised by that. I may have to source something different but for now, it will do.




Scheming motor placement while waiting for parts to arrive.



Soldered up leads to the hall and 3d printed some excellent brackets by bixler13 pgsaw pot bracket and coupler (xsimulator.net)



After testing with my mid frame and various mock up parts, I welded up some proper ctc arms to start committing to the build. I weld about once a year with a really old and cheap welder so try not to judge too much!


After being content with my midframe design, I moved on to designing the base frame and traction loss. This brought more challenges than expected as I have been trying to keep the design as light and compact as possible within reason. I have really tried to keep everything inside of the 2’x4’ footprint. I found there were several challenges surrounding where to mount the motor and how to keep the CTC arm for it within the boundaries of the mid frame. Couple that with locating the wheels and pivot, I spent a significant amount of time sitting on my basement floor thinking.

I finally opted to compromise and raise the height of the mid frame a little which allowed me to mount the casters under the u joint mount. I fabricated some plates for the ball bearing casters I purchased and mounted 3 across the back and 2 in the front.



I evaluated many different options for the pivot but finally decided on using a 1” pillow block I had and welding a 1” pipe to a plate that would be mounted to the frame. This setup wouldn’t be ideal for holding any weight but the wheels do all of that anyway. I’ll have pictures of that soon.

I have more aluminum on order to build the base frame and I should be able to start putting the whole thing together. I figured this was a good time to start a build thread.

Finally got the remaining aluminum I needed. Now I just need time! I was able to build out the base frame tonight. It's aluminum extrusion frame and cross members with melamine as a smooth and solid surface for the midframe wheels to roll on. It's 3' wide giving me 6" either direction for traction loss travel for my 2' wide mid and top frame. It has 8 rubber feet affixed to the bottom to protect the floor and stop sliding.





Some carriage bolts in place to affix the pivot and traction loss push rod mount (seen on floor). The end beams sit proud to sit flush with the 3d printed end caps seen in the first picture.

The monitor uprights currently mounted to the top frame will be moved to the base frame to reduce weight on the top frame

Hopefully will get a bunch done this weekend!

I completed the basic form over the weekend.

The mid frame sitting on the base frame. There will be a tray in the open space between the motor area and the pivot mount for all of the electronics. Still have to source metal. Probably cut up an old computer case. The Tnuts are already in place.



This is my pivot I dreamed up. It is very smooth and strong. As mentioned, it consists of a 1" pillow block I had (it's a bit oversize probably), 1" pipe and a plate. I welded the pipe to the plate and mounted the plate to the frame. It's extra-long for allowing height changes in the future if needed. Want to reiterate that I have front wheels supporting the weight. This pivot wouldn't be up to the task of holding the weight as the bearing load direction would be wrong.



And here we have my current state. I still have to mount the monitor mounts and such but that will come after it's tested more probably. All motors and rods hooked up ready to go. I have done SMC3 testing with the front 2 motors only at this point. They work well. I may be moving them back to get more pitch. The roll is plenty.



Put a short video of SMC3 action:


Sadly, you can easily see the slop in the u joint. I contacted the seller and he's ignoring me. I guess I'm going to have to build my own but it's a place holder for now.

I probably will be a bit slow for the next few weeks with my schedule, but I'll get back to it as soon as I can!

Yup. I found a complete flanged joint on eBay a guy sells brand new. I wouldn't have thought for a second they had 100k miles of off-roading on them before he painted and sold them: CARDAN SHAFT FLANGED UNIVERSAL JOINT U JOINT HIGH ANGLE | eBay

I guess I'll source and buy 2 yokes and the joint myself and weld flanges. So much for the $65.

Where did you find the motors? Looking good so far!!!

Thanks! They are the crab pot motor on eBay.

I also want to point out that the seller for the u joint finally got back to me and is sending me a new joint. Hopefully it has no play!

While I understand why adding a third motor to the processing queue will slow the Arduino's response to each motor, is there somewhere where this was discussed in greater detail? I haven't found it in my searching.

I don't really care about the cost. I just would rather not add the added area for system breakdown. Specifically, Windows and com ports

As a project update, the replacement u joint had the same issue. I ordered a drive axle for a BMW I found for cheap and will modify that. Should arrive tomorrow.

I got the rig fully setup and wired up but I'm having some issues with configuration. I configured smc3 and it follows sine waves well. I configured simtools 3 and tests for roll, pitch, yaw and traction loss work as expected. I feed it telemetry from forza and it really doesn't do much. The occasional wiggle or if I turn the intensity up to 100 it will often vibrate uncontrollably. I've attached my settings in hopes somebody that knows what they are doing can tell me what I'm doing wrong.

I definitely have some backlash in the motors. They are the pgsaw with about 3.5" ctc arms except for the traction loss which is about 7" iirc.

I have motion! There must be something wrong with the Forza Motorsports 8 plugin. I switched to Horizon 5 and the Horizon 5 plugin and it worked immediately. I was even able to use the Horizon 5 plugin with Forza Motorsports 8 but anytime I go back to the FM8 console plugin, the motors all vibrate and act crazy.

Next up, dialing in motor settings and buttoning up the rig. I'm waiting for some gear for the safety stops and such I need to add.

I ended up getting a front drive shaft from a mid 2000's BMW X3 for cheap on eBay. I separated the two yokes and combined them into a single unit to give me the ujoint I needed.



I had a setback with the traction loss motor. It had a couple of stiff jolts and then sheared the pin right off and actually cracked the shaft. I was trying to avoid permanently welding my arms to the motors but no longer have a choice.

I put a cut off nail through the lot and welded it in place. I also welded around the full shaft to the nut on my arm so it shouldn't be able to break again.

Also started putting the electronics close to where they are going to be. This gives me time to dial stuff in without committing to wiring.


I have been pleased with the cheap 100a AliExpress motor drivers so far. They have given me no issues and have stayed pretty cool. The LED power supply has had no cutouts. It's pretty oversized at 125a but I figured overhead was safer.

I have circuit breakers I will be wiring in line with the motors which should pop if the motors go to stall for whatever reason. I also have a 20a relay and emergency stop button on the way to allow me to cut power to the LED power supply from the chair.

nice clip showing your motion status.

I have the rig basically done at this point. I have plenty of tweaks to do over time but the unit is usable as is. I added a floor with a piano hinge panel to allow access to the electronics below:





I'll likely add anti skid tape or something. I may replace it with diamond plate some day but wanted to try the idea with junk wood I had.

Also wired up a first pass at a control panel. For now, an emergency shut off that cuts power to the motor power. I also added a reset button for the Arduino as I find myself fighting with the motors being off due to the rig constantly leaking down when unpowered.



This brings me to my frustrations. I was aware that there was inherently backlash in the gear motors but I find it incredibly distracting and frankly ruins the whole experience. Any s-curves or a little quick zig to avoid idiots braking when they shouldn't and it clanks sloppily back and forth against the backlash. Maybe I'm unique but it doesn't trick my brain in any way that I'm going around a corner. I feel like I'm just on a bucking bronco. Any suggestions would be greatly appreciated.

I have put many hours to date into messing with parameters. I certainly started one axis at a time like in the FAQ. Over time pretty much everything has gotten all mixed up.

The rig is definitely biased forward. With or without somebody on the rig, it falls down. I'm surprised at how easy it is to move the 50:1 motors. I have to put blocks before shutting it off or it sinks down.

I'll give it another clean try tomorrow.

I have the same problem with my rig 24v 250watt wheel chair motors with a 21 newton meter out put. Unfortunately this is only when the motors are powered, as soon as the power is turned off they just free wheel .

Here is the sine.

Just rig:


Me sitting in chair:


I set sway to 30% and everything else off. min/max at +-25 as set by capture method of a lap around the track with no anomalies. It still rocks back and forth.

I think I just plain don't like the motors. Maybe mine are abnormally sloppy.

The bottom frame has been fixed up. There were several issues that hadn't been addressed yet in the early videos. The aluminum does flex a little here and there but for the most part it's pretty good now.

The play is rocking back and forth between the backlash. So if you lean left then right then back, the left motor goes down and the right motor up and then the left goes up and the right goes down. click clack.

I don't believe my play is any worse than yours. I've added some new videos: