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

Trip's G-seat project (primarily for flight)

Discussion in 'DIY Motion Simulator Projects' started by Trip Rodriguez, Nov 19, 2019.

Tags:
  1. BrassEm

    BrassEm G-Seat + SFX100 Builder

    Joined:
    Apr 15, 2015
    Messages:
    119
    Location:
    NE of YMML
    Balance:
    1,055Coins
    Ratings:
    +47 / 0 / -0
    My Motion Simulator:
    DC motor, AC motor, 4DOF, 6DOF
    Just make sure you have got a fair amount of true sway (not roll) on your seat if you can. You will definitely miss it if you don't have it.
  2. Trip Rodriguez

    Trip Rodriguez VR Pilot

    Joined:
    May 8, 2016
    Messages:
    710
    Location:
    Lake Ariel, Pennsylvania
    Balance:
    4,830Coins
    Ratings:
    +330 / 6 / -0
    My Motion Simulator:
    6DOF
    My plan for sway is the ACME DMS style. The bottom will roll, the back will have true sway. Does that sound good to you?

    How much sway do you recommend by the way? I had been planning on more but after seeing what SeatTime and ACME seats are using I figured I'd make my back sway travel similar to those.
  3. Trip Rodriguez

    Trip Rodriguez VR Pilot

    Joined:
    May 8, 2016
    Messages:
    710
    Location:
    Lake Ariel, Pennsylvania
    Balance:
    4,830Coins
    Ratings:
    +330 / 6 / -0
    My Motion Simulator:
    6DOF
    A little more elaboration on my decision to eliminate the surge/sway table.

    I had figured out my initial steel purchase based on the earlier design, and knew I'd need to buy more so I left the trailer hooked up to the truck and ready to go.

    After cutting as many of the parts as possible from what I had, it was time to decide what additional materials I needed to buy to complete the assembly. Well, it was a lot more than I expected! It was going to cost another $200 plus to get the material, and not just that but every dollar of additional steel was also more weight.

    As I said I had some concerns about having added surge and sway, and had more or less figured that if it didn't work out I'd just remove it. Now I found out that it's going to be a lot more expensive for a feature I thought I was fairly likely to only remove after testing.

    Eliminating the surge and sway tables solves a lot of engineering issues.

    I was concerned about the forces from accelerating the G-seat and pilot etc. weight on those axes being too much for the four linear guide rails to tolerate. Adding an additional four would add another $300-400 to the build cost so that wasn't an option. I was also worried about the fact that the CoG would be moving around enough to apply significantly more torsion load to those same linear guides.

    In addition to that, the idea to add the surge and sway tables was based on the idea that I could simply use all the surplus power of two of the G-seat servos, but this meant that the surge/sway table and the related two G-seat axes would be mechanically linked. I thought there was a good chance I could make a decent compromise there, but not at all sure of it.

    Adding these features also increased the size of the sim more than I originally figured. Part of the reason I decided to go with this design was that it would take less room than the Stewart rig. Now it was actually larger than the first sim! It was up to almost 7 feet by 7 feet. That's a big chunk of space to give up in my little home. IIRC I've now got that down to about 4.5 feet by 5 (or 5.5) feet. That's a lot easier to find room for.

    The drawing for the elevator is 100% finished, and I should have that built pretty quickly. After that it's on to the G-seat which still has a few details to be worked out as far as dimensions.

    I will be adding surge back to the seat bottom. I could potentially add some sway to the seat bottom as well, but it's not part of the plan as of this writing. I've probably got a week or two before I commit to the G-seat design so input is welcome!
  4. BrassEm

    BrassEm G-Seat + SFX100 Builder

    Joined:
    Apr 15, 2015
    Messages:
    119
    Location:
    NE of YMML
    Balance:
    1,055Coins
    Ratings:
    +47 / 0 / -0
    My Motion Simulator:
    DC motor, AC motor, 4DOF, 6DOF
    My seat true sway is about 12mm at the moment (Need to add more spacers to increase throw). If I could get 20mm or greater I would instinctively be grabbing for the side rails :eek:
    Roll for sway gets lost in the motion, true sway for both seat and back really does kick in and stands out for what little I've got. (Same for surge).
    If possible I recommend back scrub to really accentuate the higher g loading.
    Ball screws have really got me re-thinking my design.
    • Informative Informative x 1
  5. Trip Rodriguez

    Trip Rodriguez VR Pilot

    Joined:
    May 8, 2016
    Messages:
    710
    Location:
    Lake Ariel, Pennsylvania
    Balance:
    4,830Coins
    Ratings:
    +330 / 6 / -0
    My Motion Simulator:
    6DOF
    That kind of reaction from under an inch of sway!? I was originally planning for +/- 50mm or even a little more.

    Back scrub you mean the friction of moving across the seat surface I assume? Yes I've given that some thought. Have you experimented with it?

    I was trying to decide if it would be better at the lower or upper part of the back. Upper seems logical, but in my experiments I noted that it seemed like I actually felt the friction mostly in the lower half of my back. It's hard to say though.
  6. BrassEm

    BrassEm G-Seat + SFX100 Builder

    Joined:
    Apr 15, 2015
    Messages:
    119
    Location:
    NE of YMML
    Balance:
    1,055Coins
    Ratings:
    +47 / 0 / -0
    My Motion Simulator:
    DC motor, AC motor, 4DOF, 6DOF
    More is better!!!

    The back of my seat is from the hip to just under the shoulder blades so I can't comment higher. The whole back of the seat moving up and down does accentuate the g loading, especially high +g pinching at the lower back and butt as you mentioned.
  7. SeatTime

    SeatTime Well-Known Member

    Joined:
    Dec 27, 2013
    Messages:
    2,584
    Location:
    Newcastle Australia
    Balance:
    27,737Coins
    Ratings:
    +2,844 / 37 / -0
    My Motion Simulator:
    DC motor, AC motor, Arduino, Motion platform, 6DOF
    IMO It really depends on how well fitting your seat is and how much travel it takes to push you into the seats side bolsters (mine are adjustable). That's were all the 'G' realism is felt for me. Without the force into the bolster/pressure pads it would end up like any other motion rig with no real continuous pressure/Gs. I even have pressure plates for my Shoulders.

    Having owned a Stewart Platform, the G's being simulated in my sim are way beyond anything a small Stewart Platform can produce.
    May not look it, but my little sim can generate some bone crushing quite realistic Gs :) (still to add back my head system as it is being updated to better suit my new setup).




    Have fun :thumbs.
    • Informative Informative x 1
  8. Trip Rodriguez

    Trip Rodriguez VR Pilot

    Joined:
    May 8, 2016
    Messages:
    710
    Location:
    Lake Ariel, Pennsylvania
    Balance:
    4,830Coins
    Ratings:
    +330 / 6 / -0
    My Motion Simulator:
    6DOF
    Thanks guys. =)

    @SeatTime Looking great! About the side bolsters thing; My sim is primarily for flying. In most of the aircraft I'll be flying there isn't anything (in the real aircraft) to the side of the seats as you had mentioned in your own thread. How would you recommend dealing with this?

    At present what I've got planned is just the back sway like you had before the adjustable bolsters were put in. What would SeatTime do? =D

    I'll certainly add removable bolsters when appropriate (i.e. racing sims), so my question only pertains to when I'm flying aircraft where there aren't any.

    Progress update: Today I hit a landmark. Welding has begun! All of the elevator parts have been cut, and I've welded up the four half-frames for the perimeter frame that holds the towers in place. That's all I can weld for now without causing problems later.

    Next steps are drilling the elevator lift frame for the bearing blocks at the four corners, drilling the corner towers to attach the guide rails, and drilling the lift towers to attach the actuator parts.

    I'm a bit tied up now waiting for a new drill press to arrive. My old one wasn't very good to begin with and after 25+ years of use it's got issues, not the least of which is run-out. The drill press will arrive between Wednesday and Monday, so in the meantime I'm going to work on my drawings for the G-seat itself and the details of how to mount the servo motors to the lift tower actuators.

    If anyone has built ballscrew actuators on steel angle iron, feel free to drop me a link here. =) These will be my first ballscrew actuators. The motors will be mounted beyond the end of the 2" x 2.5" x 3/16" steel angle at the top end. The bearing blocks will be mounted to the wider side of the angle.

    I believe I've got the mechanical design of the seat all figured out, but I've changed what materials I'll be using to build it so I need to rework the structure.

    Here are a couple very incomplete images. I don't want to waste time finishing the drawing, too much to do! The yellow part is the angle iron frame that goes up and down. The G-seat and the entire cockpit will be built on that.

    The black frame on the floor is the bottom "perimeter" frame. There will be another identical one three feet or so up from that one, and these are stationary. They hold the towers in position. I want to weld them to the towers rather than bolt them, so I split them in half to make two assemblies (left and right halves) that can be moved through doorways. Those frames are what I welded up today.

    About the towers (red):

    The tower at the corner is one of the ball bearing guide rail towers. There will be one at each corner (four total), attached at the side of the elevator platform.

    The one that isn't at a corner will be one of the two lift actuators. There will be one to match on the opposite side. The black "tabs" on the yellow frame will be welded steel angle, and are what will be attached to the ballscrew nuts. They too are 3/16 steel angle, but I think I should probably add some extra support structure to them as well since the dynamic load of the entire platform will be supported on them alone.

    elevator 1 - Copy.png elevator 2.png
  9. SeatTime

    SeatTime Well-Known Member

    Joined:
    Dec 27, 2013
    Messages:
    2,584
    Location:
    Newcastle Australia
    Balance:
    27,737Coins
    Ratings:
    +2,844 / 37 / -0
    My Motion Simulator:
    DC motor, AC motor, Arduino, Motion platform, 6DOF
    In any high speed aircraft/car etc there is always something holding you in place and often the cockpit is not that big. So you are always pushing up onto something, even if it is only the harnesses. Really your choice on how you do it , but IMO you really need to do it.

    jet cockpit.jpg

    In my type of sim the more touch/pressure points on the body the better.
    • Informative Informative x 1
  10. Trip Rodriguez

    Trip Rodriguez VR Pilot

    Joined:
    May 8, 2016
    Messages:
    710
    Location:
    Lake Ariel, Pennsylvania
    Balance:
    4,830Coins
    Ratings:
    +330 / 6 / -0
    My Motion Simulator:
    6DOF
    Where I don't see any side support in many aircraft (especially ejection seats) is the backrest. Seems like the harness would be the only thing. I'll have to experiment with adding bolsters even if they shouldn't actually be there.

    The G-seat final schematic is mostly done, and it's bed time. =) It's missing lots of detail because in this case all I'm figuring out is the structure. I'm using 1" x 1.5" 11 gauge rectangle steel tube for most of the construction. Not too heavy and easy to work with.

    I'm also making the seat bottom and seat back completely separate assemblies bolted to a pair of angle iron rails at the bottom. This will allow me to adjust the angle of each independently and also how far forward/back the backrest portion is relative to the bottom cushion. This way if I decide I don't like the way it feels I can adjust it fairly easily.

    I also want to be able to tilt the whole thing back significantly for the F-16, and also potentially F-1 cars though I've never done the F-1 thing before.

    final G-seat assy.png
    Last edited: Feb 11, 2020
  11. SeatTime

    SeatTime Well-Known Member

    Joined:
    Dec 27, 2013
    Messages:
    2,584
    Location:
    Newcastle Australia
    Balance:
    27,737Coins
    Ratings:
    +2,844 / 37 / -0
    My Motion Simulator:
    DC motor, AC motor, Arduino, Motion platform, 6DOF
    For me the paddles only work up to a point and then the pressure pads are the most noticeable thing as the Gs build up. To increase the touch points I am going to also have my foot plate able to move in heave which will be driven by my current heave system. Not allot of movement and it will be in opposition to the seat base.
  12. Trip Rodriguez

    Trip Rodriguez VR Pilot

    Joined:
    May 8, 2016
    Messages:
    710
    Location:
    Lake Ariel, Pennsylvania
    Balance:
    4,830Coins
    Ratings:
    +330 / 6 / -0
    My Motion Simulator:
    6DOF
    The main thing I was uncertain about was what to do with the upper body in aircraft where there are no bolsters or whatever.

    I just did some youtube research and it looks like at least in some of these modern jets, in this case the F-16, the pilot's shoulders and/or arms will hit the sides of the cockpit in hard maneuvers. Here is a particularly good example: This should be easy to recreate and very effective!

    It looks like the F-14 and F-18 aren't quite so tight, but it's still plausible I think so I'll just use this for all of them.

    I just found this one, where you see the pilot hit the side even in the "roomy" A-10, so I think I'll call this a good solution. =)


    Thanks!
    • Like Like x 1
    Last edited: Feb 12, 2020
  13. SeatTime

    SeatTime Well-Known Member

    Joined:
    Dec 27, 2013
    Messages:
    2,584
    Location:
    Newcastle Australia
    Balance:
    27,737Coins
    Ratings:
    +2,844 / 37 / -0
    My Motion Simulator:
    DC motor, AC motor, Arduino, Motion platform, 6DOF
    I am currently spending allot of time designing/adding a heave system for my head, so important for aircraft simulation. Really looking forward to getting this back in my sim.
  14. Trip Rodriguez

    Trip Rodriguez VR Pilot

    Joined:
    May 8, 2016
    Messages:
    710
    Location:
    Lake Ariel, Pennsylvania
    Balance:
    4,830Coins
    Ratings:
    +330 / 6 / -0
    My Motion Simulator:
    6DOF
    Really? I guess I'll have to add heave to the plans for head loading as well then. =)

    My only concern with the head loading is that I'm not sure using a helmet will be practical. I want to use a helmet, but I also want to be able to use my HMD (it will be the 8K X) for regular VR games without it being a huge project to swap from helmet to head strap. If it's easy to swap the Pimax MAS and soft strap that should solve the problem, if not I'm not sure what I'll do. I probably won't have a good idea of whether or not the helmet will be practical until the 8K X actually arrives which is likely to be a two or three months at best.
  15. Trip Rodriguez

    Trip Rodriguez VR Pilot

    Joined:
    May 8, 2016
    Messages:
    710
    Location:
    Lake Ariel, Pennsylvania
    Balance:
    4,830Coins
    Ratings:
    +330 / 6 / -0
    My Motion Simulator:
    6DOF
    I had a eureka moment last night just before bed.

    It's relative to my plans for creating Euler angle pitch and roll cues when on the ground. For me this is primarily important for lifting a helicopter off the ground and wanting to feel it start to tip. It will also be good for racing sims and such I think. For helicopters this will likely only been used with an "on ground" trigger.

    Here is the plan:

    Pitch nose down: Seat bottom heave up, seat back/bottom surge forward. This will tilt the bottom and back forward, almost like tipping the whole seat. Pitch up simply opposite this.

    Roll RWD: Seat bottom roll right edge down, seat back sway right. Again this will in effect tilt the whole seat to the side. Just for clarification, my sway cue will have the seat back sway opposite the bottom tilt.

    The one thing I was really concerned might be an issue was that when I had a Euler angle roll cue I'd be pressed laterally against the shoulder harness. If it was true roll the harness would mostly move with me since the whole aircraft was tilting. If I had the harness move with the seat back sway to prevent this, then in the much more important sway cues the lateral harness pressure cue would be partially or even entirely missing which is totally unacceptable.

    Well, it wasn't until I wrote a note to myself last night that I realized there is an obvious solution. I'm going to have the shoulder harness sway left and right, but it will be actuated by the seat bottom roll actuator rather than the seat back sway actuator!

    This way when the seat back and bottom move together (for Euler angle on the ground or whatever) the harness moves with you, but during sway cues when they move opposite the harness will move opposite my body and should actually improve the sway cue! Winning!

    Note: I probably won't actually use Euler angle data to drive this, but the lateral and longitudinal acceleration of gravity that results. This is the proper method as per @Dirty 's cueing software thread.

    @SeatTime May I ask where you source your bowden wires? I need to order some very soon.
  16. Trip Rodriguez

    Trip Rodriguez VR Pilot

    Joined:
    May 8, 2016
    Messages:
    710
    Location:
    Lake Ariel, Pennsylvania
    Balance:
    4,830Coins
    Ratings:
    +330 / 6 / -0
    My Motion Simulator:
    6DOF
    It's been a long road with an unbelievable number of iterations drawn up in Fusion 360 but the G-seat design is nearly finalized! =)

    Still have to work out the details of the harness tighteners, head loading, and G-vest but most of that is Bowden cable stuff and should be very easy.

    G-seat.png
  17. MarkusB

    MarkusB Active Member Gold Contributor

    Joined:
    Jun 30, 2015
    Messages:
    473
    Location:
    Germany
    Balance:
    3,534Coins
    Ratings:
    +474 / 1 / -0
    My Motion Simulator:
    2DOF, 3DOF, DC motor, Arduino, Motion platform
    The following may be self-evident for you, but it may still be worth mentioning.

    When coming to the bowden cables, the most important thing (according to my own experiences) is making sure that the cables are always tightened. They don't need to be under tension, but it is essential that they cannot build any loops on their way.

    My approach after producing some hardware damage and after strapping my fingers in a cable loop (of course in a way that the emergency stop button was out of reach) is to always use two cables for each moving part, i. e. one for each direction.

    For example, one cable pulls a backrest paddle forward, and a second cable pulls it backwards, with both cables connected to the same motor.

    An alternative would be to use bungee cords or springs for keeping a cable tightened or just rely on gravity or the weight of your body, but this did not work so well for me.
    • Informative Informative x 2
    • Like Like x 1
    Last edited: Feb 13, 2020
  18. Trip Rodriguez

    Trip Rodriguez VR Pilot

    Joined:
    May 8, 2016
    Messages:
    710
    Location:
    Lake Ariel, Pennsylvania
    Balance:
    4,830Coins
    Ratings:
    +330 / 6 / -0
    My Motion Simulator:
    6DOF
    As you seemed to suspect, I came to the same conclusion myself. =) I'd always much prefer that someone tell me something I know then fail to tell me something I don't!

    The only places this isn't dead simple are surge on the vest and surge on the head loading. For those I have two possible designs in mind that I'm going to look at.

    At this point the head loading is the only thing I haven't given much attention. I'm a bit worried about how to attach to the head because I won't be able to wear a helmet with the 8K X unless I remove the nice head strap. That would require swapping the strap on and off any time I was going to switch from sim to regular VR games which might be impractical.

    PS- Oh yeah, @MarkusB may I ask where you purchased your Bowden wires? I haven't found a confidence inspiring supplier with decent prices yet.
    • Friendly Friendly x 1
  19. MarkusB

    MarkusB Active Member Gold Contributor

    Joined:
    Jun 30, 2015
    Messages:
    473
    Location:
    Germany
    Balance:
    3,534Coins
    Ratings:
    +474 / 1 / -0
    My Motion Simulator:
    2DOF, 3DOF, DC motor, Arduino, Motion platform
    I never had a G system for my head, but I had a G-Vest. And for this vest I also had 2 cables for surge (1 for positive and 1 for negative/braking). You can find some details here:
    https://www.xsimulator.net/communit...-seat-with-wind-vibe.12194/page-2#post-166670

    About my bowden cable supplier: My cables are bicycle brake cables and I bought them in a usual bicycle shop.
    • Informative Informative x 1
  20. Trip Rodriguez

    Trip Rodriguez VR Pilot

    Joined:
    May 8, 2016
    Messages:
    710
    Location:
    Lake Ariel, Pennsylvania
    Balance:
    4,830Coins
    Ratings:
    +330 / 6 / -0
    My Motion Simulator:
    6DOF
    Thanks Marcus, I'll take a look.