1. For downloading SimTools plugins you need a Download Package. Get it with virtual coins that you receive for forum activity or Buy Download Package - We have a zero Spam tolerance so read our forum rules first.

    Buy Now a Download Plan!
  2. Do not try to cheat our system and do not post an unnecessary amount of useless posts only to earn credits here. We have a zero spam tolerance policy and this will cause a ban of your user account. Otherwise we wish you a pleasant stay here! Read the forum rules
  3. We have a few rules which you need to read and accept before posting anything here! Following these rules will keep the forum clean and your stay pleasant. Do not follow these rules can lead to permanent exclusion from this website: Read the forum rules.
    Are you a company? Read our company rules

Showroom 3DOF Motion G-Seat

Discussion in 'DIY Motion Simulator Projects' started by Sielu, May 22, 2021.

Tags:
  1. Sielu

    Sielu New Member Gold Contributor

    Joined:
    Feb 7, 2021
    Messages:
    13
    Location:
    Georgia, USA
    Balance:
    99Coins
    Ratings:
    +24 / 0 / -0
    My Motion Simulator:
    3DOF, DC motor, JRK, Motion platform
    Update -- She's done(tm)!



    PXL_20210606_172234891.MP.jpg PXL_20210606_172256473.MP.jpg PXL_20210606_172315753.MP.jpg

    --Original Post--

    Ok folks, I've been lurking for some time now while passively working away in the background. I didn't want to start a thread until I was sure my design would work, and that I would finish... two things that are often up in the air with me. I'm at about the 80-90% point now, with only finishing work left to do (which of course will take some time).

    The seat itself is heavily inspired by Bergison's "Motion G-Seat." I really like that seat's focus on G-forces + just enough motion to whip you around a little. Since I exclusively run flight simulators, the ability to simulate negative Gs from his design was very attractive to me. As we know, he's trying to market his design so there wasn't much out there to copy for my DIY version. I've built quite a few static simpits over the years, and have gotten skilled enough in CAD to be able to map out a design so I thought I'd take a stab at doing just that.

    Video of action:


    Bill of Materials (so far):

    - 3x PGSaw Motors (ebay)
    - 3x JRK G2 18v19 (pololu)
    - 3x Rotary hall sensors (digi-key 987-1393-ND)
    - 3x 12v, 15A LED power supplies (ebay)
    - 1x 24" x 48" 3/4" plywood, routed for seat supports
    - 1x 24" x 48" 1/2" plywood, routed for seat panels
    - 2x 8" x 36" 1/4" aluminum sheets, routed for:
    -- motor supports
    -- custom T-slot framing connectors
    -- pusharm yoke connectors
    -- rod-end mounts
    - 1x 4" x 36" 3/4" aluminum sheet, routed for pusharms
    - 1x 36" 12mm steel hex rod, for axels
    - ~24' 1.5" t-slot framing
    - ~36" allthread, 3/8" x 24 for push rods
    - 8x rod ends, with 55 degrees of travel


    2021-05-22 15_28_13-Autodesk Fusion 360 (Personal - Not for Commercial Use).png
    I opted for a 3-motor, 3DOF motion platform as a core, to which I would add just a seat and then secure panels to a frame to act as the G paddles. From what I can tell, Bergison is using a similar design where motors drive primary seat motion and then the paddles are passively moved based on the seat. His, however, is MUCH more compact and probably more efficient. I went with 3x PGSAW motors, which hit all my design criteria (and then some) for a very good price, and readily available here in North America. I suspect power-wise they're a little bit overkill for what I need, but I'm happy to know I can gain a bit of weight and still use this thing!

    The build log:

    First pusharm cut out of aluminum, secured to an aluminum axle as a test. The aluminum didn't end up being strong enough so I swapped it for a steel axle of the same dimensions, those have worked flawlessly:
    PXL_20210317_233343918.MP.jpg

    First motor assembly, custom plate to secure motor to T-slot: PXL_20210324_211800751.MP.jpg

    And the full assembly, with my original sensor mount design. I ended up swapping this design for one that secures directly to the motor. 3D printed on a small resin printer:
    PXL_20210327_213543347.MP.jpg

    The "center triangle", motors on the bottom and the seat platform off to the side:
    PXL_20210423_165244792.MP.jpg

    And everything mounted in the outer-frame, awaiting the completed seat:
    PXL_20210518_204654676.MP.jpg
    PXL_20210518_204703748.MP.jpg

    And the seat!
    PXL_20210522_132650222.MP.jpg
    with harness attached to the frame:
    PXL_20210520_224408595.MP.jpg
    PXL_20210520_224415276.MP.jpg

    Still have to come up with a seat cushion, and then complete some paneling on the side to lock everything in (and keep curious cats from killing themselves).
    • Like Like x 5
    Last edited: Jun 6, 2021
  2. Ads Master

    Ads Master

    Balance:
    Coins
    Ratings:
    +0 / 0 / -0
  3. cfischer

    cfischer Active Member Gold Contributor

    Joined:
    Sep 7, 2015
    Messages:
    253
    Location:
    Colorado
    Balance:
    1,774Coins
    Ratings:
    +175 / 0 / -0
    Cool design and nice execution! Would love to try this style of gseat.
  4. MarkusB

    MarkusB Active Member Gold Contributor

    Joined:
    Jun 30, 2015
    Messages:
    536
    Location:
    Germany
    Balance:
    3,995Coins
    Ratings:
    +541 / 2 / -0
    My Motion Simulator:
    2DOF, 3DOF, DC motor, Arduino, Motion platform
    Very nice work, @Sielu .
    I am looking forward to reading about your experiences on how the motion feels once your build is complete.

    My rig is also inspired by Bergison, though I did it with 6 motors instead of 3.
  5. Sielu

    Sielu New Member Gold Contributor

    Joined:
    Feb 7, 2021
    Messages:
    13
    Location:
    Georgia, USA
    Balance:
    99Coins
    Ratings:
    +24 / 0 / -0
    My Motion Simulator:
    3DOF, DC motor, JRK, Motion platform
    Thank you @cfischer, I really enjoyed reading through your build... the conforming fiberglass frame looks incredible. Mine’s definitely missing a bit of that “squeeze” factor... ideas for the future!

    @MarkusB, your build and the “roller coaster” video helped convince me to start the project, knowing I wasn’t completely nuts trying to do this! I originally planned to have a central spring as well, but after trying the motors realized I could get away with out it. I opted not to go with a ‘live back’ system just to keep the complexity down. Do you find it adds significantly?

    As for my rig right now, I’ve been testing and tweaking constantly over the last few days as I build out some padding. I’m surprised to find the “positive” Gs for heave and surge (acceleration with afterburner, pulling high-G turns, etc) are so far relatively underwhelming. I think I’m missing a bit of body squeeze that a build like @cfischer’s can generate.

    Similarly, sway forces are just... completely meh. I’m not too upset by this as you really don’t get much lateral acceleration while flying if you’re doing things right, but I would like it to be a little more present. Again, I think improving the “wrap” of the panels around the thighs and sides of the back could improve this, but it’s not a focus for me just yet.

    “Negative” Gs though... oh my. Hitting the air brakes and feeling the jet decelerate, or getting caught in the wires on the carrier deck, flying upside-down or cresting a hill and getting pushed ‘up’, feeling your body get pushed into the harness straps... all just feels incredible.

    Overall I’m very happy so far, even with the underwhelming bits. Being able to feel the virtual plane in so many new dimensions definitely adds to the experience!
    • Like Like x 1
  6. Banfy

    Banfy FR/AU

    Joined:
    Jun 11, 2020
    Messages:
    116
    Location:
    Australia
    Balance:
    173Coins
    Ratings:
    +46 / 0 / -0
    My Motion Simulator:
    DC motor, Arduino
    Awesome rig! Very nice and pure design, like it
  7. Sielu

    Sielu New Member Gold Contributor

    Joined:
    Feb 7, 2021
    Messages:
    13
    Location:
    Georgia, USA
    Balance:
    99Coins
    Ratings:
    +24 / 0 / -0
    My Motion Simulator:
    3DOF, DC motor, JRK, Motion platform
    Update! I've spent the last couple of weeks discovering and subsequently working out issues before I move on to the finishing touches.

    First, my original push arm design left much to be desired... such as a greater range of motion! So I redesigned the arms to 'fit' the T-slot framing better. I've now got more motion than I know what to do with, which has made the entire experience better.

    Original pusharms: PXL_20210529_182853652.MP.jpg

    Updated:
    PXL_20210529_182832893.MP.jpg

    That socket cut around the axle-hole partially addresses another issue with my original design. With the PGSaw motors, the little axle nub that sticks out on one side was causing each push assembly to be slightly off-center, and when multiplied by 3 that caused the whole seat to sit slightly ajar.
    PXL_20210529_181454057.MP.jpg

    The other issue I was having was a slight bit of wobble between the hex shaft and the pgsaw axle:
    PXL_20210529_182531381._exported_stabilized_1622372763281.gif

    That wobble was causing a whole HOST of issues, especially when combined with the seat being ajar. So, I 3-D printed a part I nicknamed the "axle squid" which gets inserted over the axle and stabilizes it, and also adds the exact amount of space that the PGSaw 'nub' does on the other side of the motor:
    PXL_20210529_174325903.MP.jpg

    The little squid "arms" squeeze in between the PGSaw shaft and the hex flats, and they taper thicker as you get closer to the "head", so the whole thing gets nice and tight once you squeeze it on. Throw on the pusharm and lock it down and your axle is nice and solid:
    PXL_20210529_182622736.MP.jpg

    With that, the whole motor assembly is even:
    PXL_20210530_120230355.MP.jpg

    But! I was still having an issue with the right-hand side of the platform striking the outer frame and locking up occasionally.
    PXL_20210530_122109094.MP.jpg

    I thought centering my axles would fix it, but it did not. Took me a while to find but it turns out the whole bottom assembly was 1/4" off center, causing even off-kilter-ness in the seat. Loosened some bolts and took a clamp and slowly re-positioned the whole bottom assembly. PXL_20210529_213129287.MP.jpg

    And with that, the whole thing's been running beautifully! Started adding in the creature comforts now. Managed to make a decent-looking foam pad out of some seat foam and a bolt of suit fabric:
    PXL_20210530_122202624.MP.jpg

    And then today I cut out the mounts for the Joystick and Throttle. Outside of all the wires hanging out and about, it's now a fully functioning simulator!
    PXL_20210530_184753638.MP.jpg PXL_20210530_184758354.MP.jpg

    Still don't have a great keyboard solution. I'm hoping the streamdeck can fulfill 90% of my keyboarding needs, but that's something I can work on long-term.

    Attached Files:

    • Like Like x 6
  8. hideki

    hideki Member

    Joined:
    Feb 3, 2014
    Messages:
    101
    Location:
    Spain
    Balance:
    44Coins
    Ratings:
    +25 / 1 / -0
    My Motion Simulator:
    2DOF, SCN5
    That is insane! Wow, congrats what a set up
  9. Sielu

    Sielu New Member Gold Contributor

    Joined:
    Feb 7, 2021
    Messages:
    13
    Location:
    Georgia, USA
    Balance:
    99Coins
    Ratings:
    +24 / 0 / -0
    My Motion Simulator:
    3DOF, DC motor, JRK, Motion platform
    Thank you @hideki !

    Small progress update: Started in on re-wiring and getting my external panels installed. Had to run almost all new wires all over the place. The cable management started off ok (and was going to be great... in my head) and then of course devolved into a rats' nest closer to the JRKs. Looks like I kinda bunched things up a bit too much in the name of space efficiency. I'll keep an eye on temperatures to make sure I'm not creating a fire hazard here.
    PXL_20210601_232929082.MP.jpg PXL_20210601_232933811.MP.jpg

    Had a bit of a setback when I went to fire her up again after the rewire. My #2 motor immediately went into a spin and sheared my retaining bolt. This.. has happened a few times over the course of the build. Fortunately the retaining bolt is by far the weakest link in the system so if anything goes haywire it shears quickly and no further damage is done. Took me a minute, but tracked down the issue to my hall sensor. One of my solder joints for the comm cable came off.
    PXL_20210602_005831669.MP.jpg

    I've also discovered, now that the wiring is all packed up, that I can pretty easily maintain the beast just by flipping her on her back. Easy access to all the wiring and motors:
    PXL_20210602_005854277.MP.jpg

    Figured this was as good an excuse as any to finally implement the strain relief for these hall sensors. I had definitely thought through this issue in the design stage and then promptly forgot to include that during initial install:
    PXL_20210602_010049478.MP.jpg

    All of this happened pretty late last night, enough time for me to finish the teardown and diagnosis, but figured I had a good stopping point with that. Plan to re-solder the sensor after work and put the #2 motor assembly back together... then actually get back to flying!

    I plan to get the other two panels on in the next couple of days, then just need to figure out how to guard the top. So far the cats have not been interested, but as soon as they see that A: this is now a box and B: it's temperature is anything above ambient they will get curious. I'm not sure my shear-bolts would lose to kitty paws, and don't really want to find out.
    • Like Like x 2
  10. Sielu

    Sielu New Member Gold Contributor

    Joined:
    Feb 7, 2021
    Messages:
    13
    Location:
    Georgia, USA
    Balance:
    99Coins
    Ratings:
    +24 / 0 / -0
    My Motion Simulator:
    3DOF, DC motor, JRK, Motion platform
    Updates!

    Sensor and wiring repaired, and got 'er moving again just fine. Since then I've just been putting the finishing touches on the motion profiles. I've got things dialed in pretty well to my liking at this point, although I'm sure I'll be tweaking for the next eternity. This may be controversial, but I'm intentionally allowing for significant clipping with the axis profile percentages. After doing a LOT of trial and error this is my reasoning:
    • First and foremost, if I cut the clipping down it just doesn't feel as good! So I turned it back up.
    • The focus of this sim is on the g-forces, not so much on the motion, as such the dominant G-force is going to overpower the senses anyway, and a clipped axis I don't think is as noticeable
    I made sure to leave a little "room" at the end of each axis so you could still feel changes. The only exception to this is the 'roll' axis currently which I may cut down a tad. Right now it's tuned via the tuning center to require more angular velocity than I can give it, so it hasn't maxed out yet. Speaking of, the 'roll' axis was added in even though this isn't a G-force. I noticed when rolling the 'sway' axis would kick out to the opposite side of what I would expect--e.g. when rolling to the right, your body should be 'pushed' to the left and left-hand pressure should increase on the body. I found with just sway my seat was "pushing" me on the right-hand side, which felt backwards. Sway was still correct in side-slips and when skidding along the ground, so I added the roll axis back in and intentionally made sure it would overpower sway when rolling the plane. I'm quite happy with the results at the moment!
    2021-06-06 13_41_35-Game Engine.png 2021-06-06 13_42_58-Clipboard.png 2021-06-06 13_43_22-Game Engine.png 2021-06-06 13_44_04-Game Engine.png

    On other matters:

    I didn't like the ergonomics of the mouse on the joystick base, so I 3D printed a big wedge for the mouse to sit on, which vastly improves the feel of the thing in the hand. It looks a bit goofy, but is far more comfortable to mouse around while flying. I also got the streamdeck up and running, which has been able to supplant all of my usual keyboard commands. With that combined with WMR's desktop preview, I've been able to run the whole sim 'headless', which is really convenient.
    PXL_20210606_173558289.MP.jpg PXL_20210606_173611467.MP.jpg PXL_20210606_173628300.MP.jpg

    Also got my old transducers hooked up to the back of the seat. Until I can get simvibe configured, I'm just using a rudimentary system that I've actually had on my static sim for years: I mirror audio out of steamVR to an optical DAC, which then gets tossed through a low-pass filter and into an amplifier to run a left-side and right-side transducer. It's not sophisticated, but it gives a really nice bass effect and 90% of the time works fantastic for getting game 'events' to vibrate as needed. Main advantage here is there is minimal configuration, as long as the steamVR audio mirroring is working. Downside is that the bass tones shift based on which way your head is turned, which is not entirely realistic but typically not noticeable.
    PXL_20210606_180822956.MP.jpg PXL_20210606_180828296.MP.jpg

    I've opted to nix the final two panels, my wife and I agreed that our cats love boxes way too much to risk it until I can guard the top. At this point they've shown absolutely no interest in the thing we're going to monitor the situation while I evaluate long-term solutions. I'm thinking of installing some stretchy fabric between the seat and frame to prevent any intrusions, cat or otherwise. In the meantime, accent lighting!
    PXL_20210606_172246966.MP.jpg

    I've also moved her into what should be her final home, and cleaned up all the external wiring. At this point I'm actually ready to call it "Finished" - understanding full well that I'll be making tweaks to it for the next... eternity. I've got some more glamour shots, as well as a video of the whole thing running which I'll append to the main post, above.
    • Like Like x 3
    Last edited: Jun 6, 2021
  11. cfischer

    cfischer Active Member Gold Contributor

    Joined:
    Sep 7, 2015
    Messages:
    253
    Location:
    Colorado
    Balance:
    1,774Coins
    Ratings:
    +175 / 0 / -0
    Nice work. Will be curious to see your stretchy fabric solution.