Fixing a Thrustmaster 458 steering wheel with Lego

Using a linear slide potentiometer and Lego to fix faulty steering on a Thrustmaster 458 steering wheel.

Public Chat
Similar projects worth following
I hardly ever buy something new regardless if I need it or not. What I did not need was a racing wheel for the Xbox 360, but I found one while thrifting and bought it anyway! For that price it was probably broken and in need of cleaning or a repair. This turned out to be true and the solution was Lego!

Don't start screaming at your screen just yet... Version 2 will be a direct shaft connection!!!

The wheel I found wasn’t high end to begin with. A Thrustmaster 458 black. A very plastic experience, but never the less, I tried it and it somewhat worked until it didn’t. After 10 minutes of flaky steering, the steering stopped altogether and extreme left and right made the wheels get stuck. This should be an easy fix!… Or so I thought.

This wheel uses a rotary potentiometer 10K Ohm’s with about 270 degrees of effective rotation or somewhere in that area. Surely I would have a suitable part in my bin for a quick fix. It turned out the potentiometer had some peculiar dimensions. The shaft was sub 4mm with an even smaller flat bit.

I had no drop in replacement for both the correct shaft and the rotation angle. The only semi-official source for a replacement was defunct and the only parts I could find that would potentially do the job meant I had to mod the wheel. I did find some overseas parts, but with added shipping, it would cost more than the bloody wheel itself!

What I also had was a collection of (stereo) 10K linear slide potentiometers. I tried various 10K parts and most seemed to work just fine, but how to mod the wheel or the part to fit. I really liked the idea of using the slide potentiometers, but for that I needed an assembly that transformed rotation into linear motion.

I have Lego!

My goal was to make the hack on the wheel reversible and for that I needed to build a proof of concept. I needed that anyway, because it it drove like a bucket, anything I build would be useless.

The motion needed to be linear, so the easiest option here is a rack gear. Of course I could only find one four-piece and I needed at least 2 and preferably 3. Here is where the first hack came in. I have a collection of RC parts which contained a linear gear that just about matched up with the teeth of the Lego part. A bit of cyanoacrylate and I had the length I needed glued to a Lego bar that would attach to the potentiometer.

The shaft was quite simple. Take a standard lego shaft, Dremel off one of the flutes and the fit is tight and perfect in the existing hole. As a happy accident, the “guide” for the original potentiometer shaft fit a Lego shaft bushing perfectly so it could continue to do it’s job.

Modded shaft
Modded shaft

Finding out what rotary gear to use was just trial and error. I tried various gears until the linear gear moved to it’s extremes in junction with the steering wheels extremes.

Then I needed to fix it all in place compactly and in a way where I could fit it to the wheel assembly. I preferred a tight fit over glue here, but screws were also an option if it came to that. As big as the device is, there is always not enough room. I could fit the assembly on an angle with a universal joint connecting the wheel shaft to the gear shaft. The whole assembly then fit almost near perfectly between the PCB and the support structure for the wheel.

It was time to test it. I took the device to the gaming room and hooked it up. I knew the potentiometer worked and I assumed that dead weight centre was straight forward. After playing around for a bit with settings and making sure the assembly was doing fine, I tested a rally circuit and I came in first! Me thinks this is proof of concept and a more robust and refined version is called for, but really this midday hack is quite playable. Perhaps for testing further iterations, I'll use hot-snot to fix the entire assembly somewhat in place.

For a definite version, I’d like to have a more uniform linear gear, preferably 3 pieces of 4 from Lego. As you can see the whole assembly still moves around quite a lot and there is a lot of stiffness that can be introduced by fixing the gear shaft bits and securing the potentiometer a bit better, but that is all fine tuning really!

The potentiometers I have come from a digital mixer that crashed and burned several...

Read more »

  • 1 × Lego 24 tooth gear
  • 3 × Lego linear rack gears Ideal scenario
  • 1 × Lego universal joint
  • 1 × 10K 60mm sliding linear potentiometer.
  • 1 × 3 wire connection cable

View all 6 components

  • Result and final thoughts

    Timescale01/02/2024 at 14:56 0 comments

    My end product is this. A directly driven rack and pinion connected to the sliding potentiometer. It all fits very well, more than enough space, but the unit can be a lot more compact.

    I already confessed to my little screw up, but luckily, the misalignment was about 1 to 1.5 gears and does not impact the performance very much it seems. The part I messed up is this one :

    See how the Lego bushing fits into the original axle guide for the rotary potentiometer. It is the back plane of this part that receives the rack and pinion assembly. Without the universal joint and the stiffness of the bolted assembly, the whole experience is a lot more controlled and reactive. Of course if you want and the game allows it, you can make it as smoochy as you like.

    In the end, this is a very effective and simple fix. You can fabricate this from about any material you have laying around I think. The Lego is nice for finding tolerances and trying new configurations and gear ratios, but the simplest method would be a wood or metal enclosure for the potentiometer and the sliding rack, bolted to this part which has the right ratio gear at the end of the shaft.

    If you loosely use the configuration I have, the results should match up. This is a Lego pitch rack with a 40toothed gear attacked to a 60mm travel sliding potentiometer.

    I have now played quite a few aggressive rounds of Dirt 3 and it still holds up!

  • Not pretty, but it works.

    Timescale12/26/2023 at 21:00 0 comments

    I got the new "direct drive" installed. It is quite crude and I did F*&k things up, but it works without much of a dead zone. Somehow I managed to misplace the mounting holes for which I used a Lego piece as a guide. Make sure it is level and centred and nothing can go wrong!!!

    Well, that did go wrong. The assembly is now a bit tilted and off-centre. I had to adjust the position of the pod to get the zero position. What ya gonna do? At least there is far less play in the shaft, it seems sturdy and it is responsive. Good enough for me at this point in time.

    Does it need improvements? yes it does, but not right now! In the end, it does remain a hack!

  • Tweak, break and redesign...

    Timescale12/26/2023 at 13:40 0 comments


    I made the assembly a whole lot sturdy by throwing more bricks at it which helped with rigidity a bit. The potentiometer is now relatively fixed in place and the pieces do not wiggle free any more (I think). How ever, due to the fit, there still is much play left to right and thanks to the universal joint, there Is a mechanical dead zone in the middle. These are not very problematic, but I would like to try and solve those issues.

    Flexible shaft test

    This was a very bad idea. One weak point and a part that introduces play is the universal joint. My idea was to use a rubber Lego shaft and use that as a universal joint, stiffening it at places where it needed no flex. Well, it flexed everywhere and the the joint area alone was enough to absorb just about the entire rotary movement.

    Fixing in place
    For making the assembly a bit more rigid, I used foam to secure it. This still gives some play, but mostly it fixes it in place quite securely. Still not ideal and at this point thought for version 2 started to creep in.

    Version 2

    So I broke the universal joint. That is why prototyping with Lego is so nice, it shows you where the weak spots are. I fixed it with glue and tie wraps, but there are only so many bricks you can throw at the problem before a rethink is needed.

    Luckily there is a simpler solution. Version 2 is going to be a direct from shaft connection that will require slight modification of some of the parts of the wheel itself.

    You might have been wondering why it took this long to realise this was the more sturdy and elegant solution, but version 1 had one great advantage. It was easy to tinker with in place and it proved that concept was sound.

    More details on version 2 in the next log.

View all 3 project logs

Enjoy this project?



Similar Projects

Does this project spark your interest?

Become a member to follow this project and never miss any updates