Open Source Joystick FFB / DIY FFB Joystick

Have any interfacing solutions been worked out yet or is the plan to make the hardware first and then try to figure that out?

 

Also to anyone stumped on how to create -smooth- power transition, read or reread the post I made about shaft winding... https://forums.eagle.ru/showpost.php?p=3274336&postcount=255 this is a viable and accessible option and also used in RL systems for its performance. It's feasible/cheap enough to diy, yet good enough for the Air Force One Blackhawk sim and many others. No belts or gears in this shoestring budget will be able to provide anything near a smooth output.

As this project is really a community effort, I believe the Interfacing solution is being worked on by others. I have no skills in programming so my contribution is what you see above.

We are aware of shaft winding and also "driven pot" solution that you designed (don't take me wrong but one can not miss that in this threat) and as much as I'm sure that it provides the smoothest transmission, I kind of don't like the idea of assembling, tuning and maintaining such device.

I still appreciate your input though:thumbup:

And a brushless motor:

https://www.ebay.com/itm/High-torque-3-8N-m-12V-DC-motor-DC-high-power-600w-resolver-brushless-servomotor/222596428686?hash=item33d3c7f38e:g:-NkAAOSwJLJZfAnF

Well that is a beast in deed.. BUT where you gonna get 80A of current for each one of those?

Money vs Torque..

we can use combustion engine if you want lot of torque, cheap:D

As this project is really a community effort, I believe the Interfacing solution is being worked on by others. I have no skills in programming so my contribution is what you see above.

 

We are aware of shaft winding and also "driven pot" solution that you designed (don't take me wrong but one can not miss that in this threat) and as much as I'm sure that it provides the smoothest transmission, I kind of don't like the idea of assembling, tuning and maintaining such device.

 

I still appreciate your input though:thumbup:

Actually I don't recommend using the driven pot circuit at all, or any other diy interface because it represents an astronomical amount of work and very limited usability. I recommend instead using a MSFFII modified to output 400% the stock torque and connect that to real motors using your own psu in the form of simply 24 power bricks. No idea what you mean by 'threat' lol.

As to shaft winding, as far as mechanical systems go you are not going to find much simpler an arrangement to construct, especially one with no backlash or cogging which are (or should be considered) deal breakers for what you are trying to do. Steel cable is subject to creep and stretch and would eventually need tightening however kevlar (Spectra/Dyneema etc) is not. Such a system would req little if any tuning and it would be pretty straight forward to incorporate a cable tension mechanism where it mounts on the bellcrank, or even little idler pulleys.

Steel cable is like a rubber band made of stabby needles though, where it elongates some 35% at break vs kevlar which is more like 3.5%. Kevlar also doesn't poke holes in your fingers and instead feels like a waxy shoelace. There are rl aircraft that use it over steel now in control circuits.

Best of luck whatever you guys decide...

I mean thread not threat:)

Well two things come to mind reading your post..

First: There is only so much MSFFB2 sticks in the world though, we will have to come up with alternative eventually.. Bigger companies gave up because of the stupid law suits but we are not dependent on them any more (as we were decade or so ago).

With so much tech available to buy, resourceful open minded community of enthusiasts and professionals, we can pull this of and develop something even better that MSFBB2.

Second: The winding technique has it's limits in number of revolutions on the winding axis thus transmission ratio.

I know you'll suggest the transmission ratio should not allow for to much spin on the motor it self because of the inertial feel.

However moving the stick in the real plane introduces some inertia as well.. You are moving lot of mass around.

Have any interfacing solutions been worked out yet or is the plan to make the hardware first and then try to figure that out?

I for one am keen to start working on the software but to be honest I'm kind of a bit stumped on how to get started.

I've installed "Atmel Studio 7" as suggested by MetalGear_Honk and have attempted to write a simple "heart beat" program (the embedded equivalent of a "Hello World" program) but unfortunately Amtel don't have a 'ATMega32u4' simulator available so I'm not really able to run anything.

From what I understand the 'ATMega32u4' is part of what is known as the "AVR 8-bit family" and is the main processor on the 'Teensy' board:

adafruit have kindly made a "Breakout Board" featuring the ATMega32u4 available which, from what I understand of this page, is basically a slightly more developer friendly variant of the Teensy:

Dispite the page saying:

This board doesn't come with any 'learn to program' tutorials! If this is your first time with a microcontroller, we suggest going with an Arduino which is easier.

I ordered one anyway and am hoping I'll be able to figure out what to do with it when it arrives :thumbup:

Hello,

i've been thinking about building my own ffb joystick too. But i want to go in another direction. Instead of building something super strong and powerfull, I would like to 3d print the mechanics and use widely available drive components (exact specifications are yet to be determined). This has the huge advantage, that everybody with a 3D printer can reproduce it cheaply. I've got quite some experience with control theory and embedded system design so i think i could do the electronics but i newer used one of this PLA and ABS extruding 3D printers and don't know if it's possible to print the mechanics. Please tell me what you think about this idea. I will post any further updates on the project in this forum.

Hello,

 

i've been thinking about building my own ffb joystick too. But i want to go in another direction. Instead of building something super strong and powerfull, I would like to 3d print the mechanics and use widely available drive components (exact specifications are yet to be determined). This has the huge advantage, that everybody with a 3D printer can reproduce it cheaply. I've got quite some experience with control theory and embedded system design so i think i could do the electronics but i newer used one of this PLA and ABS extruding 3D printers and don't know if it's possible to print the mechanics. Please tell me what you think about this idea. I will post any further updates on the project in this forum.

Hi

Have a look at my web to see what you can achieve with 3D printer.. Basically it is possible to print quite strong components (if well designed) but every surface that is to be precisely fitted to another or is moving against another one, should be really replaced or supported by metal part.

That's the conclusion from my endeavor:smilewink:

My web - https://vrflightsim.wixsite.com/mysite/in-progress

Finally as promised, here are some pics (still rendering more). I call him "BadBoy - Zero" :D

Enjoy...

Finally as promised, here are some pics (still rendering more). I call him "BadBoy - Zero" :D

 

Enjoy...

That's realy impressive. Are you going to machine all the parts by your self? I have to admit, i would like to do this too but i don't have a milling machine and i think it would be really expensive to order custom aluminium parts online. Correct me if im wrong.

That's realy impressive. Are you going to machine all the parts by your self? I have to admit, i would like to do this too but i don't have a milling machine and i think it would be really expensive to order custom aluminium parts online. Correct me if im wrong.

I gave up my workshop when we moved house so I'm not able to do the machining my self at the moment. I want to get few quotes on these parts being manufactured by "professionals" but first I'm gonna knock up simple prototype of the geared part to ensure my ratio calculation are spot on.

as mentioned earlier, anybody in UK with decent metal shop? :helpsmilie:

Jay

I gave up my workshop when we moved house so I'm not able to do the machining my self at the moment. I want to get few quotes on these parts being manufactured by "professionals" but first I'm gonna knock up simple prototype of the geared part to ensure my ratio calculation are spot on.

 

as mentioned earlier, anybody in UK with decent metal shop? :helpsmilie:

VR, I'm in San Jose, CA, USA, but maybe I can still help with making parts. Your design is similar to a design I've been working on, so maybe if I make parts for you, I can also copy your parts to test out my own plans (and hence help each other out).

I own/run a professional CNC shop, so this isn't hobby equipment. I can easily make and remake parts with changes as needed to get things right. But I"m here as a hobbyist like the rest of you and I would like to help out.

I'm also an engineer (I design digital and analog electronics for a living), so I hope to do the motor drive circuits, custom hall effect sensor arrangement, and software for my own FFB stick eventually, but for now making a working gimbal/gearing/motor mount design is the thing I am focused on, and passing what I learn onto everyone else. I've done a lot of tuning in PID loops for my CNC equipment, and a FFB stick is similar. Each stick (given it's motors, gearing, stick weight and length, etc.) requires custom tuning and everyone will eventually need some help with that too.

BTW, you are totally right about thinking of torque vs. cost. This is why I am focused on the mechanical/motor design first. The software and electronics is a big deal, but figuring out the cost vs. torque to handle a stick and move it correctly is the bigger deal to me right now. I was going to start by mocking up the software and electronics to an existing FFB stick (an old Logitech Wingman wire driven FFB stick that has big motors) to learn how to get it all working, but I think I want to do all that with a stick that is fairly close to my final stick rather than going through all the FFB tuning parameter effort twice. So building something and testing it to find that sweet spot of cheaper motors and gearing that still achieve a quality FFB stick is my focus.

Let me know if I can help.

VR, I'm in San Jose, CA, USA, but maybe I can still help with making parts. Your design is similar to a design I've been working on, so maybe if I make parts for you, I can also copy your parts to test out my own plans (and hence help each other out).

Thanks Drakoz

Prototyping the parts would be enormous help, however before we start cutting precious aluminium, need to make sure the gear ratio vs torque vs inertial factor are correct. Also cogging from belts and the motor it self needs to be as low as possible.

I'm gonna test all this on a simple timber frame (one axis only), by powering up the motor in one direction and feel the result at the end of the stick.:joystick:

I have ordered the pulleys and DC motor (same as in my cad model) for this purpose but it will take up to four weeks to deliver.

Are you familiar with Fusion360?

Regarding electronics, OS-CL (open source-control loading) dedicated motor drive circuit would be THE BEST possible solution but it is a huge challenge though.

Jay

Thanks Drakoz

Prototyping the parts would be enormous help, however before we start cutting precious aluminium, need to make sure the gear ratio vs torque vs inertial factor are correct. Also cogging from belts and the motor it self needs to be as low as possible.

Understood. Though aluminum is cheap. Time is the cost factor, which is why I'm offering to help. It's not a big deal to prototype something in aluminum as a test. Once I have the CAM work done, I can spit out a new part in probably 30 minutes for most pieces, and it's not a big deal to make minor changes to parts. Shipping to you (USA to UK) might be the expensive part. But regardless, let me know how I can help. Of course my goal is to play with your design coincidently, offer another perspective maybe, and then report back to this forum to help others if that is OK with you. I'm planning a shorter, stool mounted stick, but since your design is so close to what I was drawing up, I'm hoping to try yours out as a learning platform and use what I learn to reduce the size.

Are you familiar with Fusion360?

I use Solidworks for CAD, but for the CAM work, I just need a STEP or IGES file.

Regards,

Michael

Hey Michael

My testing motor just arrived:D

It is a weaker and cheaper brother of the 150W version intended for my design but even this one (120W) has one hell of a torque:thumbup:

Now just waiting for the pulleyes..

Well I really appreciate your enthusiasm so here's something for you.

This link http://a360.co/2zwBePp will take you to the project site where you can inspect the latest version of the model, take it apart, measure components, etc.

Please be aware that the model hierarchy is total mess. It's a work in progress though.

I don't know if you saw this but I've done simple model for short throw FFB gimball already. (smaller device for side pedestal or desktop)

You can also download this model via the link in "description" section of the video here:

Don't worry about the shipping cost, I would obviously cover that plus something for your work and material too.

Jay

@ MetalGear Honk

Hey mate, long time no see:)

Have you made any progress with your arduino?

Jay

When it comes to reasonable gearing it is important to find out what is the minimum output RPM which is responsive enough. Maybe the much bigger DIY FFB steering wheel community has some numbers on this.

I do have an OSW and I'm not sure if it's the best option here.

It's based on the Granite Devices servo drives and those already cost around 400-700€, depening on your exact configuration.

And then you need to add the servo.

Regarding the software, MMos is unfortunately strictly for racing simulators and we don't have the source code for it available.

GD are implementing their own software currently, but in contrast to the original plans, they don't do it in open source anymore, because of some chinese hardware copies (they don't want to make it too easy for the chinese, which is understandable).

Regarding the RPM … it completely depends on your servo spec.

The servos we use for simracing usually have a rated speed of 1500-3000 rpm, but these were selected for torque (usually 20-30 Nm peak torque) rather than speed.

IMO, for a flight simulator, torque is secondary, so I would definitely select different servos or drives.

I don't know if you saw this but I've done simple model for short throw FFB gimball already. (smaller device for side pedestal or desktop)

You can also download this model via the link in "description" section of the video here:

Is there a necessity for the gearing? Wouldn't it make the whole construction more compact if you could mount the motors directly?

Hi

Not much News in Software.

I am working on my mechanical test Setup.

For the OSW we had that discussionalready. The Software is not compatible for two axis.

Another Point is the third axis (pedals) wewere talking about.

That is not possible as well. Only twoaxis are supported by DirectX.

Here are some Info’s about ffb:

https://msdn.microsoft.com/en-us/library/windows/desktop/bb153254(v=vs.85).aspx

http://www.usb.org/developers/hidpage/pid1_01.pdf

And again: We are not integrating with anyflightsim! The Developers are offering all Forces when they offer FFB support. Sowe do not need to read any data from the sim itself.

MetalGear_Honk

Regarding the RPM … it completely depends on your servo spec.

The servos we use for simracing usually have a rated speed of 1500-3000 rpm, but these were selected for torque (usually 20-30 Nm peak torque) rather than speed.

 

IMO, for a flight simulator, torque is secondary, so I would definitely select different servos or drives.

Actually torque is everything. Especially when we are considering full length control columns with somewhat realistic loads. An ffb wheel has 15-20cm lever arm, while a control column has 40-50cm, so you need 2-3x more torque for the same force on the grip.

Some gearing is necessary because a direct mount woud require some crazy huge and expensive motors with monster consumption and footprint.

I think in this case motors with high 5-6000RPM should be preferred so even with high gear ratio the output rpm is acceptable.

Is there a necessity for the gearing? Wouldn't it make the whole construction more compact if you could mount the motors directly?

Torque vs Money

You can surely use direct drive instead of gears, the only problem is the budget.

If you are to spend a grand for two BLDC motors, drivers, and PSU, you can have it mate:smilewink:

But I don't think it would get any more compact though.

We are not integrating with anyflightsim! The Developers are offering all Forces when they offer FFB support. Sowe do not need to read any data from the sim itself.

Actually, I do need to read data from the sims but that's completely different story and additional software for such purpose is going to be developed independently.

The main benefit of OSFFB for me, is to have the arduino recognized by directX as generic USB ffb joystick and that's exactly what I think your intentions are. Correct me if I'm wrong.

Two axis are fine, one duino per device. Good enough for me:smilewink:

Did you solve the center part of your gimbal or does it still jamm?

Jay

Torque vs Money

You can surely use direct drive instead of gears, the only problem is the budget.

If you are to spend a grand for two BLDC motors, drivers, and PSU, you can have it mate:smilewink:

But I don't think it would get any more compact though.

Budget and space... motors big enough to direct drive at the appropriate torque/frequency capacities are HUGE. You also can only use one of them 'direct' and the other has to be coupled 1:1 via linkage so the assembly gets even bigger... it's way too much mass to move along with the gimbals so needs to be isolated. High end systems like this that I've seen use 3ph AC servomechanisms vs BLDC though I'm not sure if that is out of performance or practicality considerations.

After looking closer at the drawings I am left wondering how belt tension is handled? I saw no provisions for this, which would make installing timing belts like this difficult at best and no way to compensate for stretch. Not difficult to accommodate, but I don't think it can be ignored. My apologies if I just missed it..

Budget and space... motors big enough to direct drive at the appropriate torque/frequency capacities are HUGE. You also can only use one of them 'direct' and the other has to be coupled 1:1 via linkage so the assembly gets even bigger... it's way too much mass to move along with the gimbals so needs to be isolated. High end systems like this that I've seen use 3ph AC servomechanisms vs BLDC though I'm not sure if that is out of performance or practicality considerations.

 

After looking closer at the drawings I am left wondering how belt tension is handled? I saw no provisions for this, which would make installing timing belts like this difficult at best and no way to compensate for stretch. Not difficult to accommodate, but I don't think it can be ignored. My apologies if I just missed it..

I wasn't seriously suggesting use of direct drive, it is obviously no go direction.:doh:

As for my big stick model, you are right. I didn't care to much about such detail as belt tensioning or even bearings as it was mainly focused on component layout.

I need to mock up simple rig for one axis first, including identical gearing, motor, and 500mm stick shaft, to see how it really feels and perhaps after that I can figure out the other details.

Jay

I wasn't seriously suggesting use of direct drive, it is obviously no go direction.:doh:

 

As for my big stick model, you are right. I didn't care to much about such detail as belt tensioning or even bearings as it was mainly focused on component layout.

I need to mock up simple rig for one axis first, including identical gearing, motor, and 500mm stick shaft, to see how it really feels and perhaps after that I can figure out the other details.

 

Jay

I didn't think you were considering it, I was offering more reasons in addition to cost as to why it's not a good option here, for the person that was wondering that you were responding to.

Makes sense on your drawing now, it's not what you are using to machine parts from just yet. I thought since you were looking for people to cut metal it was your final design. Testing out single axis the most sensible approach, it will let you evaluate the feel of the motor/drive and also the tactile reporting of the timing belt(s). Carry on..