Flight Sim Rudder Pedals

by Taumelscheibe Mar 8, 2019
Download All Files

Thing Apps Enabled

Please Login to Comment

This looks great!
Is there a list how often i need to print each stl file?

Edit: never mind. found it in the bom spreadsheets

Wow! This is going to take a little bit to get used to! Still need to get some beefier shock springs, but the stock ones aren't too horrible.

Quick note for the springs. You can get spring steel wire at you local big box hardware store. In the insulation section they have wires to hold insulation up between floor joists before the sheetrock is attached. Waaay more than needed, but they'll probably come in handy for other stuff.

It was hard to find them nearby, so I had that idea of using some bike spokes...

what other model of 109mm shock absorbers, you can use in your project, several difficulties to find, some materials !!
Do not change the design!

You can find the dimensions of the shock absorbers in the instruction, those are from the front axle of a 1/8 scale RC buggy.
I guess any 1/8 scale shockabsorber will work just fine as long as the diameter is 20mm or less.

Even if they don't fit into the platic parts you could always file them down to make them fit.
You could also use longer shockabsorber if you extend the 150mm aluminium profiles.

Hi, thanks for answering!
A bad news, here in Brazil does not have these aluminum profiles(Profile 20 I-Type slot 5 and 20x40 I-Typ Nut 5), I'm upset about it, I want to run your project.
Taumelscheibe, how to use V-Slot profiles, and how to modify the slider part.
Please help, how will these changes be, in the parts printed in 3D?
Sorry to bother you !!!
Thank you !!

If you want to to modify the parts you will need Fusion360 (it's free for hobbyists: https://www.autodesk.com/campaigns/fusion-360-for-hobbyists )
Open Pedal Slider v1.f3d and look for the design history at the bottom of the screen, double click sketch3 in the design history to open it.

You have to change the size of the rails so that the slider can still move along the rail but doesn't wobble around, it might take a few test prints to get it right.

In the attached files you can also find a version of the slider with chamfered rails, i suspect that the original slider might wear out faster on V-Slot Rails.
You'll just have to try it out which one works best, with a little bit of luck the attached file will even work without modifications.

Don't forget to add silicone lube when testing the slider.

I forgot to ask about the aluminum profiles.
Aluminum profile are V-slot, have any more nomeclatura, for them ??
Sorry for this question !!
This project is very good !!
Thank you !
Take Care!

The profiles are 20mmx20mm I-Type with 5mm slot, they are not V-slot profiles.
You could try to use V-Slot profiles but you'll have to modify the slider part.

Hello Taumelscheibe,
I can not identify, where the Hall Sensors are located on the Pedals, you could give some tips !!!
Bumpers, what measures to use ??
Could the magnet be N45 ??
Thank you !!!

You can find the installation of the sensors in the instructions (chapter 8 Electronic Installation).
The exact dimensions of the Shockabsorbers can be found in chapter 2.
N45 magnets will work just fine, but it is possible that you'll get a lower resolution if the magnets are weaker.

It looks absolutely amazing, thanks for sharing it!
This is definetly the thing which makes me to buy 3D-printer (already ordered it).

If i'll lucky enough, I'll do the pedals in next month or two. Again, thanks a lot for sharing!

Fantastic design! I just finished building my set, and I am thoroughly impressed with the detail in all the parts, the fit and the finish! It shows how much effort you've put into this and the knowledge you possess to design it - thanks for sharing such an awesome project!!

A few observations I jotted down as I built mine:

-2020 slot cover (printed part for final assembly) is not included in the Thingiverse files (not a big deal, I used zipties to hold my toebrake harnesses to the 2020 extrusion)

-Some parts (e.g. bearing holders, the heel of the pedal) should be printed with supports - I didn't realize it as I hit "Print", but the prints were still fine (just a bit of cleanup for the failed bridging). Might be good to note that in the print instructions.

-I used M8 Ny-Loc nuts in place of the standard hex nuts (on all corners and at the 2040 bearing holders). I noticed the standard hex nuts were loosening pretty quickly after moving the pedals. With the Ny-Locs, I get much better rigidity and sensor stability. Loc-Tite (thread locker), or lock washers might also be a good alternative.

-I would increase the recommended harness length for the toe brake harness from 300mm to 320mm in the instructions. Since the "toebrake strain relief" part is not mirrored, the sensor really only fits in the left hole (looking at the front of the pedal block), and therefore, the right toebrake harness is stretched a bit tight. I can't put my Arduino cover on with the recommended harness length.

-In DIView, I'm getting an effective resolution of about 1% for the Yaw sensor, and 0.5% for the toebrake sensors. Does this sound about right? I assume it has to do with the fact the Yaw sensor moves through a full rotation, while toebrake sensors only move through half of the sensor range?

Again, thank you for sharing this awesome design! After calibration, I plugged it into my sim computer, and XPlane 11 picked it right up! I've never had proper rudder pedals, so I have nothing to compare these to, but it's such a great feel, and the pedals behave perfectly!

I've recorded a build video which I'll be posting to my YT channel soon, just in case it helps others to have a step-by-step guide to follow (despite the fact your instructions were very thorough and easy to follow!).

Thanks and cheers!!

-BK Hobby

Thanks for your input, i have updated the instruction and added the missing slot cover stl file.

About the DIView question, you are right with the sensor range of the toebrake sensors, the toebrakes basically have only half of the maximum possible resolution.
However your values seem a little bit odd to me.
I get about 0.12% for the yaw axis and 0.24% for the toebrakes, maybe your magnets are a little bit weak are you using N42 or N35 magnets?
What sensor output voltages do you get on full deflection? It should be close to 0V-1V for left rudder input, 4V-5V for right rudder and 4V-5V for max toebrake.

Thanks for the response!

I think you're absolutely right - I uncommented the test function in your code and played with different magnets. I definitely get a better response to some than others. I was able improve to 0.6% for the yaw, and 0.3% for the brake sensors, so I'm very happy with that resolution (and I can probably find a better set of magnets to improve that some more).

i cant seem to find magnets of that specific size, where did you get them?

I found mine on AliExpress, here: http://s.click.aliexpress.com/e/cf7zygwG (Affiliate link) - they took a while (~3 weeks) to deliver (and tracking # was useless), but they did arrive and work very well!

You can find suitable magnets on ebay, just search neodymium magnets.
It seems that 6mmx3 magnets are more common than 6mmx5, you can use those as a substitute just stack a few of them together.

Have an old pair of thrustmaster rcs pedals and thinking about making this to get toebrakes, hall sensors, and usb. Taumelscheibe, your prints look great, what settings and material did you use?

Also, questions about the sensors. The way you used the allegro sensor for the toe brakes seemed to be more in a linear than a rotation mode like the yaw. Did you have to do anything special to do this. I was thinking about using either a mlx90333 in analog for all three or a tle5010 in digital mode for the yaw to get better resolution with the promicro or maybe just adding an external adc.

Thanks for any help.

The foot-pedals are printed in PETG and the other parts PLA with 0.28mm Layer height, i used 90% infill for the bearing holders and 30% for everything else.

The hall effect sensors are actually linear sensors, if there is no magnetic field perpendicular to the hall element, the sensor will output 50% of the supply voltage, if there is a field the output voltage will either decrease or increase depending on the polarity and intesity of the field.

If the yaw axis is centered, the field produced by the magnets is parallel to the sensor, this means the perpendicular component of the magnetic field is 0 and so the output will be 2.5V, if you apply rudder the magnitude of the perpendicular component will increase, the polarity of the field depends on the direction of the rudder input, for left rudder input the voltage should decrease, for right input increase.

It shouldn't be too hard to modify the yaw axis for a mlx90333 or tle5010 but i don't know if they can be used for the toebrakes aswell.

I'm not sure if the better resolution would be much of a benefit, just for comparison Saitek Pro Flight pedals have 9 Bit resolution for the yaw axis and only 7 Bit for the brakes, my pedals have 10 Bit yaw and 9 Bit toebrake resolution (9 Bit because the toebrake output voltage can only be between 2.5V and 5V).

However the programmable gain of the mlx90333 could be very usefull.

FYI, 2020 extrusion for this can be had on eBay very cheaply. I also found some 109mm shock absorbers there - we'll see how well the work. :) Taumelscheibe, you did a great job on the design! I look forward to building a set! I may add some support under the pedals to avoid stressing the center mechanism though. If I do, I'll send you the files.

"I may add some support under the pedals to avoid stressing the center mechanism though"

I've already tested supports with ballbearings under the pedals, but they created a lot of friction and didn't really help with stability,
i even bought a few ball casters, which should work better than bearings, unfortunately metal ball casters are just too noisy.

The center mechanism should be rigid enough to function without additional support, but you could increase it's rigidity if you want to.
The easiest option would be to create taller bearing holders and replace the 2040 extrusion with a 2060 extrusion (or 3 stacked 2020s).

I'm not going to get terribly worried about it until I've built it as-designed. If have an actual problem (or it looks like I'll end up breaking it), I'll try my tweaks. No sense in fixing something that's not broken. :)

Again, for those concerned about 2020, I bought 10 meters (10, 1M "sticks") of the stuff for $79 USD, so it can be had pretty inexpensively.

Do you think you could substitute v-slot rail instead? Metric 2020 can be tricky to get affordably in north america


Yes v-slot rails should work, but you'll have to modify the linear-slider.

would be nice to maybe make the hinge narrower and then have F-16 style pedals come out the side

They only thing more impressive than the actual design is the thorough documentation you provided with it. Well done.

Really impressed with the return spring mechanism.

Amazing design, thank you for your work!

I'm glad you like it.