August 24, 2025
Description
This is a basic set of 3D printed sim racing pedals using hall sensors for accelerator and clutch, and a load cell for the brake. I use the accelerator and brake almost daily for a little over a month as of the original posting - the clutch is (at this point) untested. The mechanics of the clutch should work, but I'm not sure how the pedal feel is.
https://www.printables.com/model/1394243-sim-racing-pedals-with-load-cell-v2/
A fair amount of the hardware is shared, but V2 uses a different loadcell and M6 instead of M8 bolts and rod ends for the spring systems.
I've only tested PETG, but I believe ABS and PLA should work.
Parts will print fine at 0.2mm or 0.3mm layer heights
Depending on your printer's bridging ability, the mounting slots on the Pedals-SidebaseA/B may need support
6 perimeters (at least) and 95-99% rectilinear infill for the ModuleSide parts (these take the vast majority of the loads)
6 perimeters and 20% infill for other pieces
McMaster-Carr part numbers have been provided where applicable, since they give detailed specifications. Most parts can be sourced more cheaply from AliExpress or eBay, but can be harder to find. Springs are shared between on the accelerator and clutch, so if you're buying locally make sure to get as many as you actually need. The McMaster springs come in 6 packs, so you'll have extra if you get them from there.
The springs on the brake are about as stiff as I think the PETG parts I use can handle, but the accelerator spring will probably be a bit light by the standards of sim racers - I based my spring selection to give a similar throttle and brake feel to my street car.
The main pedal pivot bearings work fine with the M5x70 bolts, but there is a slight wiggle since the major diameter of threads are slightly under nominal size. The M5x70 on each pedal can be swapped out for precision ground rod that you cut to length and thread on each end for a nut - this basically eliminates wiggle. The same goes for the spring arm pivots - the M8 bolts work fine, but getting an 8mm x 35mm x M6 shoulder bolt eliminates the play entirely.
I used some o-rings I had around the house to put under the head of the M8x140 bolt that runs through the center of the spring arm to dampen the clack from letting off the pedals. This isn't specced in the BOM, but makes them a little friendlier to housemates. I also put a bit of plastic-safe lubricant on the spring arm bolt where it passes through the plastic to avoid squeaking.
The software is a quick and dirty hack for a SparkFun Pro Micro (and should work on other 32u4 based boards) that I intended to replace, but so far it works well enough that I haven't bothered. A short press of the button initiates a 10 second calibration sequence, where you should move the accelerator through it's full range of motion and push the brake to a comfortable 80% (BRAKE_CALIBRATION_POINT as float) feel. After calibration, the firmware lops off the bottom and top 3% (DEADZONE_PERCENTAGE as float) and adds the extra 20% brake to the range, and saves to eeprom. A long press of the button resets the eeprom in case there's something stored that's causing an issue, though I've never had to use that over just recalibrating.
There are debugMode and debugSpam constants for getting serial output of what's going on under the hood if needed.
Additional details can be found in each subfolder which will contain a BOM-x.txt
This is UNTESTED as of this posting
After another month of use, I was finding the steel bolt heads on the load cell top biting into and grinding away the aluminum spacers - obvious in retrospect, given the cycle loading and shifting.
Very minor changes to the position of the load cell for additional clearance between the bottom of the loadcell and the aluminum spacer in the bottom rear of the brake. With more flexible plastics, the bottom of the load cell would touch the spacer and give an inverse output response near the maximum deflection. The cutout on the brake pedal arm was also moved to accommodate the slightly shifted load cell bracing.
There were also some minor changes to the codebase.
Added PedalSet.stp as a complete STEP export of all 3 pedals.
Corrected the bearings in the BOMs from F605 (5x14x4) to F605 (5x14x5). Either size should work.
Corrected a software bug with the brake loadcell reading where it would overflow when crossing the zero point in raw reporting from the HX711
License:
Creative Commons — Attribution
Andrew B
