Transforming a Drill into a Force Feedback Sim Wheel

In this video, I convert parts from an inkjet printer and cordless drill into a force feedback steering wheel for driving sim games.
The firmware for the Leonardo can be found here: github.com/ranenbg/Arduino-FF...
STLs for the work so far: jasonwinfield.nz/my-stls/
DIY force feedback
Homepage: jasonwinfield.nz
Instagram: / jasonwinfieldnz
Twitter: / jasonwinfieldnz
TikTok: / jasonwinfieldnz
Cults3D: cults3d.com/en/users/jasonwin...

[@jus4027]

Looking forward to you posting further updates, cracking project, cheers

[@grzeskucin1909]

Any updates?

[@rankovicmilos86]

Very nice build and tnx very much for using my Arduino firmware ❤ For best experience make sure to also use my gui, you left that part out from the video. The firmware is plug and play, but there is some configuration to be done as everyones setup may be done differently.

[@JWNZ]

The GUI was in the video but I wanted to leave this till the next one when I have the pedals etc up and running.

[@bena2.014]

Do you know if it is possible to use one of the analog axis as the steering? At the moment my wheel has a potentiometer for positioning, although I intend to upgrade to a encoder in the near future.

[@rankovicmilos86]

@@bena2.014 yes, it's possible with minimal changes in the code

[@bena2.014]

@@rankovicmilos86 so it just changing the axis that is associated with the steering on the game settings wouldn't work, I'd have to change what the firmware interprets as the steering input?

[@rankovicmilos86]

@@bena2.014 you can do it like that, but then you will not have ffb, because it is tied to X-axis only. If you want to have ffb then you have to do this change in my firmware. There was already someone asking about it on my github and we already had some disscusion what needs to be done.

[@torstonvodesil6709]

Cool stuff man

[@ethanhorner55]

love the packaging of your design so far. I'm making one atm also, can give you my findings if you want

[@UncleNewy1]

I don't know if you've figured it yet, but the loud clicking is the impact system from the drill you used. That's what the notched ring and the heavy spring is. You can take them out completely and just have the motor.

[@JWNZ]

Yes I did in the end. I removed the hammer function and replaced it with a washer. :)

[@seanbradley3706]

Amazing work.

[@JWNZ]

Thank you! Cheers!

[@_hanz73]

Fascinating!!!

[@virtualbrake]

Logitech's got nothing on this man!

[@eiurandelfkhan2299]

GL.. Happy diy

[@eamonrafferty701]

That is beautiful!

[@JWNZ]

Thank you! Cheers!

[@jakehiggins5118]

sounds like once there's enough torque, its spinning the anvil on the end of the impact driver motor and causing that sound. it's the same sound it would make if you were to use the impact driver to drive a screw.

[@JWNZ]

100% correct. In the next video I showed my fix. I removed the spring from the hammer and replaced it with a plastic collar.

[@juanignacioarena3141]

@@JWNZ hi! i was testing my FFB using wheelcheck and my motor made the same noise, im using it with gt2 pulley sistem, how can i fix it? couldn't find in others videos

[@ollieandsally212]

Nice. Ive wanted to do this with cordless drill motors, but was worried about the amount of backlash in the gearbox causing a problem when the feedback changes direction.

[@erick2328]

what motor driver you use? with drill gearing ratio i doubt that motor can produce enough torque for the wheel (the motor looks like RS775/995 or equivalent), also using planetary gear might result in cogging and snapping, also the noise gonna be loud but the ffb strenght gonna be lacking i do build my own wheel with 24v 300w scooter motor with bts7960, and stm blackpill board (emc pro software by EbolzMagi), and even with more powerfull motor i still need 1:8 ratio hdt3m belt and pulley to get around 5nm with 70%% ffb strength on the software

[@JWNZ]

There isn't a model stamped on the motor, I am not sure what it is. So far the torque has been suffcient and I can't say it is overly noisy. My one concern is the gears breaking but we will have to wait and see on that one.

[@jaiden270274]

Looking good, have you removed the hammer assembly from the gearbox? The other thing is that your encoder is before the gearbox, as such you need to calculate the gear reduction and adjust it in software to get the game and wheel in sync. To get a decent result you may need a better motor controller that has more development time. I moved to OpenFFBoard, if you dive into the documentation, you'll find there are cheap ways to go make it work with development stm32 boards and basic motor drivers

[@JWNZ]

I have now bypassed the hammer assembly causing the noise. Just hoping I don't strip any gears. But it's all in the next video. I did look at OpenFFBoard but as far as quickly getting up and running quickly (where I live) you can't really go past a Leonardo. This was just supposed to be a handbrake project that just spiraled a bit.

[@dumdidumdumification]

​@@JWNZ I wouldn't worry about the gears, those things can take a beating. Great project btw

[@atomgaming7170]

Hello @The Making of Jason Windfield! Great work and thank you for this guide! I plan on building one and I am new in the OSW field. As I understand from your video, the parts I need are the following: 1. Cordless drill motor 2. Cordless drill battery 3. 3D printed body 4. An H-bridge 5. A random printer encoder 6. Arduino LEONARDO. Please correct me if I am wrong and thanks in advance!

[@JWNZ]

Those are the core components yes. You would also need the ball bearings, steering wheel along with some screws but these should be scavanged from the donor printer. I have also made some paddle shifters that require the springs from the print head so if you can get the same model printer that would help. I haven't done it yet but there will be a full write up on my site jasonwinfield.nz/ soon.

[@atomgaming7170]

@@JWNZ Thank you so much! I was wondering another thing too: how to connect the arduino to the drill motor and the encoder? I am a newbie in these things, I suppose it is really simple, but just asking in case it isn't. Thanks in advance again and will follow the progress of your project with interest!

[@JWNZ]

@atomgaming7170 If you checkout the github link in the show notes the developer has done a great job of laying out the wiring and where everything should go.

[@atomgaming7170]

@@JWNZ thanks!

[@guest238]

Friend, can you show how you did the connections of the encoder with the arduino? I have an HP desjet 2374 printer. I wanted to know how to connect it?

[@TheLucasmalla]

good job bro i like your work saludos desde argentina 😉

[@JWNZ]

Thank you so much 😀

[@tronpham8920]

What types of motors can your project support? If I use a 24v motor, can it run?

[@JWNZ]

The h-bridge controller I used is rated to 22v so 24 would be too high. You might be able to find a higher voltage h-bridge

[@said_lastardunet9471]

Hello it is possible too make a 3nm or 5 NM wheel with juste un arduino and a big motor or you need something in between too give power or transform information too arduino?

[@JWNZ]

Cordless drills will have at least a modest 100NM at the chuck. I expect there will be well over 5NM at a wheel. It is whether the gears in the motor could withstand the abuse.

[@oldestries]

Your car needs alignment 9:51

[@gparadisept8976]

Maybe a nice PETG material for some parts

[@mayby1]

hello, I followed your video to the end but once uploaded the .hex file on Arduino Leonardo I go to Game Device Manager and I appear as joystick, the problem is that connected the encoder, as indicated by the diagram, does not move any axis and the buttons from 1 to 12 indicate them as pressed, I also tried to use the software contained in the folder as DXTweak2, When I turn the encoder the program increases the value by 2 and then returns to 0. Do you know how to help me?

[@JWNZ]

What type of encoder? From a printer?

[@mayby1]

thanks for responding, the problem was encoder which was faulty, I changed and now it works perfectly.@@JWNZ

[@Factsfun-kg4xc]

FFBeast have a perfect guide on how to use the motors from a Segboard / Hoverboard. and turn it into a Direct drive. there is even a parts list and everything. i have ordred everything i need to build it and are just awaiting the parts to arrive. the total including a Used Segboard ran up to $120 for direct drive 15Nm Torque wheel. you cant beat that. and its even compatible with CLS pedals.

[@JWNZ]

That's an option but TBH there is less than $60 in parts in this project.

[@letmeinletmeinnnn]

I have a question, so you don't need an encoder (potential meter) for this project?

[@JWNZ]

You do, this uses an encoder out of an inkjet printer.

[@letmeinletmeinnnn]

@@JWNZ I'm a bit confused about your 3d design that the encoder is not very obvious for me to understand how they work. My last question is where to find the or any other encoder that I could use for this project? Thank you very much!

[@ca5ualm3dia]

what makes the DIY wheel device MOST ACCURATE, is it the encoder ONLY or the FFB feature that allows you to be more accurate and smooth in inputs. I dont need realism, I need most accuracy!

[@JWNZ]

The encoder is from a printer that has 1200 lines per inch. It's pretty accurate.

[@ca5ualm3dia]

ok but does any of the force feedback feature add accuracy to the build , or are they just there ti simulate the forces. id be happy with thumb wheel, like an rc controller,but the market is barren for any type of thumb wheel.Only the e swap has one and its 160 or more usd. IM looking to build my own xbox wheel simulator, not a sim racing wheel if that makes sense !@@JWNZ

[@dogcreek-customs5168]

No I am not here for that but I do have a 3d printer and was thinking of a way to make power steering for a gocart using this and servos or drill motors. lol

[@FranciscoAngelini]

about tork ? how many nm ?

[@JWNZ]

Pretty hard to calculate as this was a whitelabel motor without any specs. I would say it is at least 100nm at the shaft so even at 15cm it is going to be as high as anything you buy from the store.

[@FranciscoAngelini]

Work with odesc 4.2 ?

[@JWNZ]

@FranciscoAngelini I don't think so. The signal from the arduino is a simple direction and PWM.

[@kiwicombat9629]

I couldn't get the patreon stuff to work to download the STLs.

[@JWNZ]

You clicked on join for free?

[@kiwicombat9629]

@@JWNZ I've joined for free on the patreon, but when I click refresh to refresh the page it sends me to a patreon checkout to upgrade. I would like to donate though.

[@JWNZ]

@@kiwicombat9629 you are probably on the advanced stls? The others should be fine. What are the first 4 letters of your email in patreon?

[@kiwicombat9629]

​@@JWNZthe first four letters of my email are noob

[@kiwicombat9629]

@@JWNZ I'm pretty sure I'm not in the advanced stls but when I got to the patreon access page nothing shows up and it says I need to join the patreon

[@VityaYALTA]

О LFS))

[@chipie_guy]

ig you should use metal cuz the plastic isnt strong enough

[@coridefeugarte8106]

Federico Ugarte sanz

[@Theo_Lxill]

its loud

[@SianaGearz]

Is torque the most important? I doubt that. Haven't you seen a million items advertised with pointless specs before, while the important ones are hidden? I have been thinking about building one, but not knowing how i want it to behave and what was important in a wheel, so i simply bought one. Driving Force GT (DFGT) can be had for around $50US used, this is what i got. Momo Force is worth maybe half that. A lot of Logitech wheels be had cheaply with a broken encoder which is an easy fix for you. Another advantage of a bought wheel is broad software support not only in the simulators where you can set everything up point by point, but out of the box in simcade titles. Torque could indeed be higher, i think double wouldn't hurt. But i doubt it's that important, i mean, i know somebody who built a wheel which will break both of his arms, and he runs it at like 5% of its nominal torque; so then what was the point of building all that torque? The motor alone was $500 or something, it's gargantuan, he can barely even lift it alone! What else is important is max turning speed. When you let it go, you don't want it to take 5 seconds to centre out! But well speed is low on DFGT as well. Double wouldn't hurt at all. Then there's noise, and DFGT is among the loudest wheels ever made, but it's still fine, the noise isn't too objectionable but much louder would be bad. But this already tells me that i probably wouldn't want to build one out of a drill gearbox Then there's smoothness. I can feel the teeth of the gearing mechanism. That's not ideal. And then there's backlash. Another point against drill gearbox. My personal approach is to try to avoid building from unique things i have found cheaply. Because i'm not the most reliable worker and tend to kill things dead, maybe i drop a screw, forget to turn off the power, maybe i short something, maybe i dial in wrong voltage on my lab PSU, maybe i have bad wiring, maybe i make a programming mistake. So it's OK if i can just buy a replacement, but not OK if i kill a one-off unobtainium item, well that's the end of the journey then. Let me continue in a reply.

[@SianaGearz]

Inside the DFGT is found a big gear comprising the steering wheel shaft, a transmission gear, an RS-555 size motor (same mounting size as 540, 550) driving the transmission gear, and an optical encoder coupled to the steering wheel gear on the opposite side. The gears are straight cut. The total transmission ratio from motor to wheel shaft has been measured by someone else to be around 30:1 (actually 27.65:1). There's also a wire harness going to the wheel buttons inside the wheel shaft and a rack with stops across the bottom. The rack limits the total rotation to 900° and prevents the wires from progressively twisting themselves to death - you can skip it if you use slipring coupling, wireless buttons, or skip the buttons entirely. On startup, the motor turns the wheel both ways until the encoder stops emitting steps, thus finding the centre - you can instead demand that the wheel be started in top centre and operate relative to that. A higher end Logitech wheel from the era, such as G27, has helical cut gears for more linearity and smoothness, no intermediate gear, and the motors are doubled up. The lack of intermediate gear ramps up the speed, and the doubled up motors improve backlash, smoothness and restore the lost torque. The encoder is moved to the back of one motor, where it's heated by the motor shaft, which causes the polyamide plastic to shrink and be split by the shaft, requiring replacement, this is an engineering flaw. Mind that a nicer wheel is not distinguished by more torque! For now i'll get a different motor and see how well it does for itself, i suspect it should be twice the unloaded speed and stall torque, as well as twice peak power, the one i'm buying for now is Mitsumi M36N-5E. I will also get a new PSU because the one i got is sus and i'll want more power headroom for the new motor anyway. My next step might be a belted conversion or a printed herringbone gear rebuild, to try to dial in backlash and smoothness. Some people perform a low torque, high speed conversion of the DFGT. They move the motor to the encoder position and couple it with the gear that the encoder used to use; then they move the encoder to the back of the motor shaft. So the encoder ratio to the wheel stays the same, but the motor to wheel ratio is reduced to roughly half the original. This also runs much quieter, smoother and with less backlash than original construction. I don't think i want to do that but i'll consider yet. If i was building a wheel from scratch, i would likely go for an RS-755 type motor, just a random one from Amazon or Ali, 24V-capable, then belted couple it to the steering wheel shaft. I would probably design for a 19V laptop PSU.

[@Schmoking]

I think the cheapest direct drive wheel is around 300 dollars... i do not remember the nm, but lower than 5. But in review it was mentioned as fast. Direct drive is the best possible option for sim wheel and that price is hard to beat, even in diy! If racing, especially in cars like f1 then torque will help "feel" the car, but in a bit of an unrealistic way as it uses to much force. For driving, road car feeling, drifting... then less torque and more speed is better.

[@SianaGearz]

@@Schmoking That be Cammus C5. Yes that's a strong proposition. Yes at 5nm rating it's double the Logi G-series and DFGT, which is probably all one realistically needs on a smaller size wheel. But it's still a chunk of money, and people may be uncertain whether they really want a device like that. Building something from scratch or modifying something is always fun and educational, but in turn you don't know whether it'll turn out usable so you want to not overspend there either, and a belted or geared build can be cheap. That being said a hoverboard motor is less than 100 currency-units, and maybe you can find much cheaper, people got a piece of fire trash on their hands. So maybe you can beat Cammus at their game?

[@Schmoking]

@@SianaGearz I already have wheel, the t300rs. Its belted, as upgrade from logitech it is much smoother! If I upgrade then i must have direct drive. But i am thinking about handlebars for a motorcycle... For Ride 4 and MX bikes in VR. Load cell on the brakes, and force feedback on steering. Not sure on motor choice though, kinda tricky... don't wanna waste a bunch on testing different motors. A lot less steering input on a bike, a road car has 900 degrees rotation, not sure what different bikes has but probably not even 90... Been wondering if several smaller motors might be an option?

[@Schmoking]

@@SianaGearz Do you know around what rpm, volt and stall torque is on those hoverboard motors?