Designing and Building a Full Motion Cockpit

Discussion in 'Hardware Building/Buying/Usage Advice' started by vegaguy5555, Feb 22, 2017.

  1. vegaguy5555

    vegaguy5555 Registered

    Joined:
    Jan 28, 2017
    Messages:
    960
    Likes Received:
    505
    Hi guy's. And ladies??

    I design and build sim cockpits and am now working on something specific for VR. Because of the small almost none existent space of VR I can make the whole rig much smaller and still be NASA complicated, which should cover all physical movements of more then one 'G', where all the fun really begins.

    I am trying to make it as physically accurate as possible. I can feel the finer braking points of my tires more easily in a real car and would like that same precise, physical feeling in a simulated environment.

    My first build is coming in around $20 000 CNA. But the production model's should come down quite a bit and have a foot print of about six to seven foot radius of my smaller design. And the lager one is coming in around the eight foot area of size and has more "G's" and accuracy.

    There are some questions for all of you.
    Should I do it?
    What would you be willing to pay for something that is a lot more involved then just a shaker seat, which can be tens of thousands?
    Can anyone help with programming some movement that may not exist?
    Would Studio 397 be interested and or help with programming?
    Does the programming already exist?
    And of course what do you generally think?

    Thank you!!
     
  2. 64r

    64r Registered

    Joined:
    Jan 16, 2012
    Messages:
    191
    Likes Received:
    35
    Hi,

    I think that perhaps you are oversimplifying (or underestimating) what is needed to reproduce accurate physical forces, especially once you get above 1g. There exist several ways to reproduce real world forces, but unless you start building something significantly bigger you will start to introduce unwanted forces and motion as you try to change direction or re-centre the rig between breaking/corning moments. Maybe you can describe your concept further and I can help identify the issue areas specific to your concept.

    To answer your questions:
    1) Do it only if you have the free time and money to do it with no expectation of sale or success. Or if you can find somebody to bankroll the process for you.
    2) The motion and cueing algos are more complicated then the hardware and will be specific to the hardware/concept you implement, there are some smart people on this and other forums so help will always be available.
    3) I suspect Studio 397 would need to see a commercial upside, however they may be willing to implement any required base plugin functionality required, although at the moment I think everything you require is already available using the existing plugin code.
    4) Have you looked at x-sim
    5) Good luck, see point 1)

    I would suggest that you start with something significantly easier but just as relevant, a yaw only based platform. Real racing drivers feel g, but respond to yaw. If you were to build a turntable based platform that was driven by yaw only you would easily be able to reproduce real world yaw forces that will give you almost everything that you or a professional driver would want. The next evolution of that concept is to mount actuators/hexapod onto that turntable to reproduce real world ride heights. With that combination you would have very accurate real world yaw, pitch and roll 1:1 as felt by the driver. As far as I know, there exists no consumer level simulator that does this, but it would cost significantly less then 200k and would only be missing the g aspect, which actually is not really that important.

    Hope this helps.
     
    vegaguy5555 likes this.
  3. vegaguy5555

    vegaguy5555 Registered

    Joined:
    Jan 28, 2017
    Messages:
    960
    Likes Received:
    505
    I have seen x sim. I don't mean any disrespect to someones work but the bottom line is I'm not impressed because in the end it's more or less still a shaker seat.

    Good response though!! Well thought out. Thank you!

    My design includes pitch, roll, yaw and 30 cm suspension travel.
    "G force" Yes. The robotics engineering team I have accuse to have brought this point up, which is why my smaller design is more attractive at the moment.
    I have given this some thought in the past and taken this into consideration and worked out an angular momentum formula.

    I appreciate all reactions. This really helps me picture the movement.
     
    Last edited: Feb 22, 2017
  4. vegaguy5555

    vegaguy5555 Registered

    Joined:
    Jan 28, 2017
    Messages:
    960
    Likes Received:
    505
    I just re-read your post and you are talking about negative G's on a neutral axis, which is something I am concerned about.
    I think it won't be too bad because it doesn't need to stap roll or pich. I suppose if you were rolling at a slow speed in a 70's muscle car and snapping the fuel, which I would try for sure. This will produce a negative G scenario.
    It will just need to have a speed limit.
    The physical cockpit will only be half I suppose. The other part will be the programming of how fast it will move and react.
     
  5. Adrianstealth

    Adrianstealth Registered

    Joined:
    May 28, 2012
    Messages:
    4,578
    Likes Received:
    1,071
    I built my own after a few prototypes,

    My final one I had welded and powder coated so its a professional finish, I also had to have some parts engineered
    its truly brilliant , I opted for the simexperience software for the motion its about the best available imo

    platform pitch, rear skid, rocker seat on top ( using x4 SCN6 actuators )
    approx. total cost including attached items £13k (excludes prototypes)

    dont tackle this unless you are tech minded & have the time , its a great project if you have though I know my setup inside out
     
    vegaguy5555 likes this.
  6. vegaguy5555

    vegaguy5555 Registered

    Joined:
    Jan 28, 2017
    Messages:
    960
    Likes Received:
    505
    Cool!!

    The one I'm building is based on an equation for angular movement I have come up with.

    The frame and arms are all designed and will be scratch built. I am still looking at actuators. I would like to use electric servos but air rams are much cheaper. It won't change the final design much.

    Although I didn't get a degree in mechanical engendering I have a close family background of robotics engineering to pull from. We have been kicking this idea around for a couple of years now.
    VR makes it possible now.
     
  7. Adrianstealth

    Adrianstealth Registered

    Joined:
    May 28, 2012
    Messages:
    4,578
    Likes Received:
    1,071
    Pneumatics (air) can make a noise & increases complexity

    Stick to SCN6 -these are neat & compact & will give you more speed than you'll need - I have to turn down the acceleration and travel speed

    Be careful on movement types
    The "feel" you may be imagining could be much different in real actual use , in fact little actual movements are needed for best immersion , some things are the opposite to what is imagined to be the case

    Only from my experience of course good luck which ever way you go
    -it is worth prototyping -I used 80/20 adjustable tubing 1st just to get everything spot on before my frames were fabricated , flex can also be a problem which is why I choosesteel box tube/weld etc
     
    vegaguy5555 likes this.
  8. vegaguy5555

    vegaguy5555 Registered

    Joined:
    Jan 28, 2017
    Messages:
    960
    Likes Received:
    505
    Thanks for the tips. Will look into those actuators for sure. I hope they work. I'm not looking forward to using air.
     
  9. vegaguy5555

    vegaguy5555 Registered

    Joined:
    Jan 28, 2017
    Messages:
    960
    Likes Received:
    505
    I looked into those actuators. Wow!! They are cheap compared to the air which are about 4x that. Plus the control panels are cheaper as well. They are a bit slow but they have a lot of torque. I think I can make it work?

    Thanks again. I'll keep you guys posted!
     
  10. Alex72

    Alex72 Registered

    Joined:
    Dec 27, 2016
    Messages:
    1,317
    Likes Received:
    992
    Always wanted to test those sim-seatbelts with motors that constrict your body as you go on the brakes. The one i saw cost some 8,000€ i think though LOL! :D I think it was the one they used in F1 simulators hehe. But there is cheaper ones however i never looked into it atm as im out of a job. As soon as i get back in business again im going to get AccuForce wheel (or another good DD wheel) and some great pedals. What follows after that we will have to wait and see. But definitely getting more gear than wheel, pedals and stick. Might come to you for pointers then.
     
    vegaguy5555 likes this.
  11. 64r

    64r Registered

    Joined:
    Jan 16, 2012
    Messages:
    191
    Likes Received:
    35
    A hexapod style design will give you pitch, roll and yaw. The pitch and roll will be 1:1 with the real world but the yaw will be faked as the base cannot continuously turn. Commercial hexapod simulators handle yaw by yawing with a relative speed and direction, then returning back to centre (washing back) at a slower speed that the user cannot notice, this helps allow the hexapod to be ready again for the next yaw movement. I once wrote a paper on this issue (sorry not publically available) and as an example, after 1 lap of Monza in a current F1 car, a simulator that handles yaw 1:1 with the real world would complete the lap with the base platform turned 87 degrees compared to the start of the lap, this is due to the accumulation of the real world yaw.

    Whilst a lot of people want a simulator to reproduce g, what you really want is to reproduce yaw. Every steering input and correction that a driver makes is driven by a response to yaw (not g), the g is just a side product of the motion, hence my suggestion to start with a turntable and mount a platform on top of that.

    I'm still not sure how you plan to handle g force, there exists no simulator in the world to reproduce real world g forces so any detail that you have on that would be welcome. If you are planning on using a hexapod or something similar you need to really consider carefully the pivot points of the platform, will you be pivoting the platform below the drivers seat, or at the drivers head level, there is a big difference between the two. You probably should start reading up on the Vestibular system to get a bit more of an understanding of what you are up against.

    In terms of hardware, what you will need is something that can move very quickly but also smoothly and in a repeatable way. I know that a common simulator test is to put a sine wave through the control system to test the speed and accuracy of the movement, if you can move your whole platform (and driver etc) up and down producing 5g at 50hz you will be getting close to commercial levels of performance.
     
    vegaguy5555 likes this.
  12. Adrianstealth

    Adrianstealth Registered

    Joined:
    May 28, 2012
    Messages:
    4,578
    Likes Received:
    1,071
    The SCN6 actuators are more than fast enough

    For example mine are set to only approx 30% acceleration & speed
    -turned up they could shake your eye balls out of your sockets ha ha

    Remember though design of the setup has to relieve the actuators of loading as much as totally possible to make easy work for them (leverage reduces load)
    Also damping (actual mechanical ) will help actuators ( I designed some levers that contain rubber -this also gives a great feeling )

    Re.hexapods -there's always trade offs , the bigger the setup the fine movement detail reduces ,
    The bigger is not necessary better

    I tried various setups & although my wheel & pedals do move with the base platform -my body moves separate due to the chair also moving separately on top giving the g force messages of how ones body moves in relation to the steering wheel. ( going fast in a car one uses arms to keep body pushed back in seat under braking & body slightly moves to sides under g-force during corners etc )

    I'll have to take & post pics of mine when I get a mo as can't really see much in my name pic

    Prototyping testing experimenting is the key to get a really brill setup
    If not then best simply try a few & then base your design on the same principals as your preferred choice

    Software setup is critical too
    As one can have best motion hardware but terrible settings many I've tried have been setup terribly like fair ground rides

    Hope this is of help

    Ps a good motion setup / hardware / VR is heaven & still a big saving on doing real life
     
    vegaguy5555 likes this.
  13. vegaguy5555

    vegaguy5555 Registered

    Joined:
    Jan 28, 2017
    Messages:
    960
    Likes Received:
    505
    Very good!! Full marks!!

    Yes, the 'vestibular system' or inner ear where fluid travels through a tube with little hairs to sense movement. The human bodies gyro.

    I have been looking into the dizziness/vertigo you can experience in VR, which I think is caused when you spin what you are seeing but not your inner ear. I refer to as reverse dizziness.

    My angular momentum formula solves this problem.

    I am looking into another more complicated formula I haven't finished yet. Instead of physically moving the whole body and inner ear I just want to move the fluid in the tube with a subatomic particle known as the phonon. But my math is not good enough yet.

    I think a lot can be achieved with subatomic sounds.

    I read that they can amplify sound by stimulating its admittance of radiation even though it is so slow. I think it was a team in France had solved the math and had a successful test. Very interesting.
     
  14. vegaguy5555

    vegaguy5555 Registered

    Joined:
    Jan 28, 2017
    Messages:
    960
    Likes Received:
    505
    Yes! This is big help. Thank you!

    Correct information is the only thing that matters. If my ego gets in the way please remind me of this.

    I think I will build a simplified version first with these actuators. Rubber mounts is something I haven't given thought to thank you.

    Unfortunately I can't share exact details because of the people funding the project. As a designer it's hard not talking about it to anyone who will listen especially others who have built anything simulor. But at the same time it's these conversations that help me see the project better.

    Thank guys.
     
  15. vittorio

    vittorio Registered

    Joined:
    Jan 11, 2012
    Messages:
    1,118
    Likes Received:
    540
    The last years i always wanted to buy a motion simulator. I had a SimXperience stage 3 in mind. But then VR came and i thought it would be better to wait so there is enough time for engineers to solve any motion sim problems when used with VR headset.
    Today, after few days vacation and some time to read, i think i will build me a DIY 6DOF motion simulator, like this one:
    https://www.xsimulator.net/communit...ooth-2x60s-was-mms-and-arduinos-osw-wip.6948/

     
    bluet and vegaguy5555 like this.
  16. vegaguy5555

    vegaguy5555 Registered

    Joined:
    Jan 28, 2017
    Messages:
    960
    Likes Received:
    505
    Thanks for the link and video. Although my design is long way from this idea it is nice to look over and see the actuators they are using. I do, however, like the suspension, which mine will have lots of.
     
  17. 64r

    64r Registered

    Joined:
    Jan 16, 2012
    Messages:
    191
    Likes Received:
    35
    Care to explain the concept more? Are you considering a spinning platform to reproduce force, if so this introduces another set of issues.

    I think that if you are asking for help on programming etc you will need to explain what you are trying to build in a more detailed way. At the moment I have no idea on what you are trying to do, only on what you want to achieve, and there appears a great divide between the two, especially when you start talking about sub atomic particles.
     
    vegaguy5555 likes this.
  18. Adrianstealth

    Adrianstealth Registered

    Joined:
    May 28, 2012
    Messages:
    4,578
    Likes Received:
    1,071
    Re the vid

    -reading through his log it seems like an ongoing project

    They are not servos but I think window wiper motors -the response time (only based on watching the vid) actually looks pretty ok

    Be great if similar could be done using scn6's would also make a compact setup
     
    vegaguy5555 likes this.
  19. vegaguy5555

    vegaguy5555 Registered

    Joined:
    Jan 28, 2017
    Messages:
    960
    Likes Received:
    505
    Sorry. Two separate projects. The subatomic particle one is the one I have now. It's based on a lot of base in a confined area. It works pretty good but I would like to improve it but still doing research one this one.

    In the meantime I'm designing a physical moving cockpit.

    Is there a way to have a private conversation with only so many people and with someone from rFactor 2 or Studio-397?
    This is new and I can't put it on a public forum because I will lose my funding. The way it moves is something I have not seen anywhere.
    This is untested and unproven but I think I am on the right track.
    Sorry for being elusive.
     
  20. vegaguy5555

    vegaguy5555 Registered

    Joined:
    Jan 28, 2017
    Messages:
    960
    Likes Received:
    505
    Is that what those motors are?
     

Share This Page