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 Download Package Now!
  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 here. Do not following these rules will lead to permanent exclusion from this website: Read the forum rules.

Showroom Full frame 2DOF G-Seat with Wind & Vibe

Discussion in 'DIY Motion Simulator Projects' started by MarkusB, Jul 12, 2018.

  1. MarkusB

    MarkusB Active Member Gold Contributor

    Joined:
    Jun 30, 2015
    Messages:
    315
    Location:
    Germany
    Balance:
    2,256Coins
    Ratings:
    +273 / 1 / -0
    My Motion Simulator:
    2DOF, DC motor, Arduino, Motion platform
    Hi folks,

    after finishing my first rig (see my first build thread), I took it apart again and started from scratch. Too many things had to change, so that it was not possible to go on with what I had.

    My second build is again a 2DOF full frame simulator with integrated G-Seat, but with quite some improvements compared to the first one. The most important aspect is that the 2DOF and the G-Seat motors are much stronger than before.

    This was my old rig:
    MyOldRig.jpg

    And that’s the new one:
    NewRig_small.jpg NewRig_small2.jpg

    Only the seat is still the same as on my first rig. Well, at least its steel frame, because I ripped off all the fabric.

    You see two kind of boxes, one on the left and one on the right side of the rig.
    The right box contains all the power supplies, which can be separately switched on and off:
    05 TrafoBox mounted_s.jpg 01 TrafoBox_s.jpg
    • 24 V: needed for the wheel chair motors and the Super500 servos
    • 19 V: needed for the force feedback wheel
    • 12 V: needed for the Seaflo blowers, the cooling fans, and the bowden cable motor
    • 9 V: needed for the Arduinos
    • 6 V: needed for the 33kgcm servos
    • 5 V: needed for the signal lines of the Super500 servos

    And the left box contains the main parts of the electonics: The sabertooth 2x32 for the wheel chair motors, a JRK12v12 (had this one from my first rig) for the bowden cable motor, 2 Arduinos, and a 10 port USB3 hub:
    ElectronicsBoxMounted_s.jpg ElectronicsBox_Front_s.jpg ElectronicsBox_withFans_s.jpg ElectronicsBox_Back_s.jpg

    Here are the 2DOF motors. I am using second hand wheel chair motors, which really have plenty of power.
    They are controlled via an Arduino and a Sabertooth 2x32.
    2DOF Motors.JPG


    This is my G-Seat.
    09 G-Seat_s.jpg

    Each surge flap is powered by three 33kgcm servos, ...
    06 Lift mechanism 02_s.jpg

    ... and each heave flap has one Super500 servo connected:
    08 Heave Flaps 02_s.jpg

    One special thing about my G-Seat is that the backrest including the surge servos moves up and down. This shall increase the impression that you are pressed into the seat (when the backrest moves up) or lifted out of the seat (when the backrest moves down). I got this idea during a real flight, when I observed in which way I sensed the different forces. Here are some pictores of the lift mechanism, which is powered by two Super500 servos.
    05 Lift mechanism 01_s.jpg 06 Lift mechanism 02_s.jpg

    All 10 servos are controlled by a single Arduino board:
    10 Electronics 01_s.jpg 11 Electronics 02_s.jpg
    The Super500 servos need 24 V for the motor and 5 V for the signal, the 33kgcm servos run with 6 V, and the Arduino gets 9 V. So this circuit board is connected to 4 different power sources.
    By the way: The software I use is based on the one from @Spit40. The backrest flaps only move on positive surge and sway values, while the heave motors run in both directions.

    Although my focus is on flight simulations, I also want to try some racing. For this purpose, I can easily exchange my flight pedals with racing pedals. (In contrast to the flight controlers, my racing equipment is rather low-end. I got it marked as 'defect' from ebay, though it still works.)
    01 Pedal Mount_small.jpg 02 Flight Pedals_small.jpg 03 Racing Pedals_small.jpg

    Oh, and then - inspired by @SeatTime - I added a bowden cable for simulating negative surge forces (i. e. braking). For this purpose, I re-used one worm gear motor from my first rig, added a wheel, and connected a bowden cable that goes to the front of the rig.
    BowdenCableMotor_1s.jpg BowdenCableMotor_2s.jpg BowdenCableOutput_1s.jpg BowdenCableOutput_2s.jpg

    My current problem with this component is that I don't have a suitable harness. I tested a shoulder belt that is meant for carrying a saxophone, but this did not work. I feel the belt tighten around my shoulders instead of my body being pulled forward.
    If I don't find something suitable, I will use the bowden cable for a usual seat belt tensioner. For this purpose, I just need to lead it to the top of the backrest instead of to the front of the rig and connect it to my seat belt. (So luckily I am not finished yet. :))

    With my previous rig I used rumble motors for simulating engine vibrations. Now that @yobuddy has relased the ingenious Game Vibe software, I changed to this one. For now I just have a very basic setup with one amplifyer and two bass shakers that are mounted to the backrest:
    01 Amplifier_s.jpg 02 Shaker_s.jpg 03 Shaker_s.jpg

    And finally, I added two Seaflo blowers for getting an airstream, based on @SilentChill's phantastic thread:
    Seaflo_s.jpg

    Last but not least: Here's the probably most important feature, which is meant to increase the WAF (Woman Acceptance Factor): The dimensions of the rig make it fit exactly into a corner of the room, so that it becomes nearly invisible.
    WAF_s.jpg

    And finally, here you see the new rig in motion.

    The first video shows how the single axis move. For making the bowden cable movements visible, I have connected it to the backrest with a rubber band. You will see that the Super500 servos are rather slow, but since I am focusing on flight simulations, this is ok for me.


    The second video shows a rollercoaster ride. For making the wind visible, I have mounted some tissue paper to the blowers.

    Attached Files:

    • Like Like x 8
    • Winner Winner x 3
    Last edited: Jul 29, 2018
  2. Spit40

    Spit40 VR Rift Flyer

    Joined:
    Nov 3, 2016
    Messages:
    271
    Location:
    United Kingdom
    Balance:
    2,180Coins
    Ratings:
    +143 / 1 / -0
    My Motion Simulator:
    3DOF
    Wow, that is some piece of kit. 'Nearly invisible' did you say !!
  3. MarkusB

    MarkusB Active Member Gold Contributor

    Joined:
    Jun 30, 2015
    Messages:
    315
    Location:
    Germany
    Balance:
    2,256Coins
    Ratings:
    +273 / 1 / -0
    My Motion Simulator:
    2DOF, DC motor, Arduino, Motion platform
    Only if it hides in the corner. Otherwise it is far from being a secret 007 simulator. Guess it‘s more the Rambo style.
    • Like Like x 1
    Last edited: Jul 14, 2018
  4. mariano68

    mariano68 Active Member

    Joined:
    Nov 15, 2014
    Messages:
    584
    Occupation:
    builder contractor
    Location:
    Argentina
    Balance:
    3,628Coins
    Ratings:
    +380 / 2 / -0
    My Motion Simulator:
    2DOF, Arduino, 4DOF
    Lots of work, congratulations!
    • Like Like x 1
  5. MarkusB

    MarkusB Active Member Gold Contributor

    Joined:
    Jun 30, 2015
    Messages:
    315
    Location:
    Germany
    Balance:
    2,256Coins
    Ratings:
    +273 / 1 / -0
    My Motion Simulator:
    2DOF, DC motor, Arduino, Motion platform
    Heave Flaps

    From time to time I will add some more details about my rig. Today it's about the 'heave' flaps of my G-Seat.

    In order to make it easy for the servo motors, I mounted the hinges of the flaps 22.5 cm away from each other, as shown in the picture below.
    12 HeaveFlapsWithMeasurements.jpg
    The distance of 22.5 cm is a bit more than the distance 'd' in the picture below, which I measured on my own body.
    PelvicBone.jpg
    In this way, the flaps don't need to lift me, but instead only squeeze my body. For me the effect on flight simulators as well as No Limits 2 is quite convincing, so that I will keep it this way.

    The flaps have a width of 15 cm. This is the maximum width that makes them still fit into the seat frame, but it is also large enough, because less than these 15 cm get in contact with my body.

    The next image shows the flaps with their coating, which I cut from a camping mat.
    13 HeaveFlapsWithCoating.jpg
    • Like Like x 2
    • Informative Informative x 2
    Last edited: Jul 29, 2018
  6. MarkusB

    MarkusB Active Member Gold Contributor

    Joined:
    Jun 30, 2015
    Messages:
    315
    Location:
    Germany
    Balance:
    2,256Coins
    Ratings:
    +273 / 1 / -0
    My Motion Simulator:
    2DOF, DC motor, Arduino, Motion platform
    The "Super500" Servos

    I am using high torque servos called "Super500" for the 'heave' flaps as well as for the movable backrest of my G-Seat.
    You can get them via ebay or amazon, where amazon (at least in Germany) has the advantage that toll is already included.

    Since these servos are kind of special, here is some information about them.

    First thing: They are really big:
    08 Super500_small.jpg

    Second: They are the strongest servos with usual RC connectors that I found: The manufacturer claims they provide a torque of 500 kgcm. I did not check this though.

    Third: They are quite slow, which means they need about 0.5 sec per 60 degrees. But as I mentioned in my first post, this is ok for me because I am focusing on flight simulations, and the pressure effect during take-off is really good.

    Now about the setup. At first I tested them with the Arduino sketch provided by @Spit40 and found out the following:
    • They only move in a value range between 50 and 150. Thus you need to adjust the value mapping accordingly in the Arduino code.
    • Initially, the rotation angle was really large (something between 270 and 360 degrees if I remember correctly), which is too much. For adjusting it, you need to open the servo cover (see below).
    • Main drawback: The target angle differs depending on the move direction. So when the servo is at position 50 and you set the angle up to 100, the final position is different from when you move the servo from position 150 down to 100. The difference can be 5-10 degrees, which in my opinion is really essential. I don't know how these servos can be used for robotics, since a robot arm will probably miss its target. However, for my rig also this drawback is not critical.
    • Another drawback: The axis has one side flattened, and the position of this flattened side differs between the servos. So you only have one way to mount the servo wheel to which you connect the lever, resulting in different lever positions. I solved this by drilling multiple holes at different position on each lever.
      Note: I also tried to open the gear part and manually turn the axis, but this did not help. After closing the gear box and powering the servo, the axis moves into is original position. The reason for this is that the position detection is done via a little magnet that is glued to the inner end of the axis.
    The gear box:
    05 inside Super500_small.jpg

    Here is a picture of the levers I cut from 5mm aluminum:
    02 Servo Levers.jpg

    Adjusting the rotation angle
    For adjusting the rotation angle, you need to open the servo (6 screws need to be removed) in order to get to the board. There you find 2 potentiometers: one for the speed (already set to max speed) and one for the rotation angle (see picture below).
    06 inside Super500_small.jpg
    Adjusting the angle is done by turning the pot, which was a bit cumbersome. In order to get the same rotation angle on both servos, I made a circle with a scale from 0 to 359 degrees and another circle with a pointer (see the picture below). Then I adjusted the pot until I got the desired moving range.
    01 Scale.jpg

    For mounting the servos, I build some brackets from 5mm aluminum:
    01 Building servo mounts_small.jpg 02 Servo mounts_small.jpg 03 Mounted Super500 Servo_small.jpg

    Until now all 4 servos do their job, and I really hope that they are durable.
    • Like Like x 4
    • Informative Informative x 1
    Last edited: Jul 30, 2018
  7. Spit40

    Spit40 VR Rift Flyer

    Joined:
    Nov 3, 2016
    Messages:
    271
    Location:
    United Kingdom
    Balance:
    2,180Coins
    Ratings:
    +143 / 1 / -0
    My Motion Simulator:
    3DOF
    You've inspired me. I started working on my gSeat v2 this weekend. Well actually project one was to build a tool to cut mild steel accurately so i made a guide bracket to attach to my angle grinder for use with 1mm cutting discs.

    image.jpg image.jpg
    • Like Like x 2
    • Creative Creative x 1
  8. MAYAman

    MAYAman New Member

    Joined:
    Jun 2, 2010
    Messages:
    18
    Balance:
    58Coins
    Ratings:
    +2 / 0 / -0
    Thank you for the post MarkusB, very informative.

    I’m looking to replace the servos of my GS4 seat with something that is more robust and will last longer. Only got 3 years out of my servos.

    Thanks
  9. MarkusB

    MarkusB Active Member Gold Contributor

    Joined:
    Jun 30, 2015
    Messages:
    315
    Location:
    Germany
    Balance:
    2,256Coins
    Ratings:
    +273 / 1 / -0
    My Motion Simulator:
    2DOF, DC motor, Arduino, Motion platform
    Thank you.

    Unfortunately I cannot tell you anything about the robustness of my servos. They just run a couple of hours until now.

    In any case the Super500 servos will not be suitable as a replacement for you.
    Your replacement servo needs to match with the other GS4 servos regarding size, voltage, and speed.
    The Super500 is probably too large, and for sure it is too slow.
  10. RiftFlyer

    RiftFlyer Active Member Gold Contributor

    Joined:
    Apr 15, 2014
    Messages:
    164
    Balance:
    1,269Coins
    Ratings:
    +97 / 0 / -0
    My Motion Simulator:
    DC motor
    Great work @MarkusB. It’s funny we both came up with the heave on the backrest idea separately. Have you measured the load on the heave component of the backrest? I’m curious how much effort it takes to move those flaps while they are being pressed against your back. I think the servos I got may be overkill and may also be too slow. I’m afraid now that the speed of the backrest heave component won’t match the speed of my wiper motor heave flaps in the base and will make for a disjointed sensation. The fact you have super 500s on both means you don’t have a speed difference. Curious what you’ve found.
  11. sberns

    sberns Member Gold Contributor

    Joined:
    Jun 8, 2016
    Messages:
    52
    Balance:
    228Coins
    Ratings:
    +30 / 0 / -0
    My Motion Simulator:
    3DOF, SCN5, Arduino
    You may wish to try these for the GS-4:
    https://m.banggood.com/CYS-S0650-La...Car-Boat-Airplane-p-1338528.html?rmmds=search

    CYS S0650. They are the right size and specs (strength and speed) seriously outperform the original Hitec servos used in the GS-4. Only potential issues that I see are 1. Longevity (though they only cost half of the hitec) 2. They should be run at 6V and have a high stall Amp draw. The meanwell power supply of the GS-4 I believe is a S-320-5 which can only reach a max of 5.5V and they may not have enough power - you could probably get by running them at 5.5V given how strong these servos are but you could also consider replacing the meanwell PS with something more powerful
  12. MarkusB

    MarkusB Active Member Gold Contributor

    Joined:
    Jun 30, 2015
    Messages:
    315
    Location:
    Germany
    Balance:
    2,256Coins
    Ratings:
    +273 / 1 / -0
    My Motion Simulator:
    2DOF, DC motor, Arduino, Motion platform
    Yes, that's true. :) I got the idea during a real flight when I tried to find out in which way I perceive the different forces. During take-off I was pressed downwards into the seat, which I also sensed with my back, caused by the friction between the fabric of the backrest and my clothes.
    I did it right now with a suitcase scale. The moving backrest itself weighs 5.5 kg, which is already quite heavy. With my back pressed against it, I measured up to 8 kg.
    Yes, the different speeds may be a problem. Have you ever thought of using a wiper motor also for the up/down movement? Then you would have the same speed. Or is the needed space an issue?
    • Like Like x 1
  13. RiftFlyer

    RiftFlyer Active Member Gold Contributor

    Joined:
    Apr 15, 2014
    Messages:
    164
    Balance:
    1,269Coins
    Ratings:
    +97 / 0 / -0
    My Motion Simulator:
    DC motor
    Thanks for the numbers. The load shouldn’t be an issue. I’ll test the speed properly once the required standoffs arrive. I could use a wiper motor if it comes to it but it will mean designing a different mechanical arrangement.
    • Like Like x 1
  14. Grigory

    Grigory Active Member Gold Contributor

    Joined:
    Jul 6, 2016
    Messages:
    169
    Location:
    Moscow
    Balance:
    778Coins
    Ratings:
    +258 / 0 / -0
    My Motion Simulator:
    3DOF, 6DOF
    Hi @MarcusB!
    I am now close to finishing my version of a gseat, but I am doubting about the placement of the shakers. Do the hinged flaps still allow you to distinguish between left and right side vibrations?
  15. MarkusB

    MarkusB Active Member Gold Contributor

    Joined:
    Jun 30, 2015
    Messages:
    315
    Location:
    Germany
    Balance:
    2,256Coins
    Ratings:
    +273 / 1 / -0
    My Motion Simulator:
    2DOF, DC motor, Arduino, Motion platform
    To be honest, my current shaker setup is very basic. I did not even configure the shakers differently for left & right. However, I don‘t think that the flaps have any negative impact on the possibilities you have with Game Vibe.
  16. Grigory

    Grigory Active Member Gold Contributor

    Joined:
    Jul 6, 2016
    Messages:
    169
    Location:
    Moscow
    Balance:
    778Coins
    Ratings:
    +258 / 0 / -0
    My Motion Simulator:
    3DOF, 6DOF
    Thanks! I just suspect that it may be difficult to sense the direction from which vibration is coming, if in my seat I can only get these vibrations through the flaps with hinges located close to the seat's centerline.

    I am wondering if you only run a single shaker, will you be able to feel the vibrating side or the whole seat will be shaking.
  17. BrassEm

    BrassEm G-Seat Builder

    Joined:
    Apr 15, 2015
    Messages:
    84
    Location:
    NE of YMML
    Balance:
    615Coins
    Ratings:
    +31 / 0 / -0
    My Motion Simulator:
    DC motor, Arduino
    Very clever idea on the back scrubbing up and down. What was the max range of movement for best effect?
  18. MarkusB

    MarkusB Active Member Gold Contributor

    Joined:
    Jun 30, 2015
    Messages:
    315
    Location:
    Germany
    Balance:
    2,256Coins
    Ratings:
    +273 / 1 / -0
    My Motion Simulator:
    2DOF, DC motor, Arduino, Motion platform
    Thanks
    The maximum movement range from neutral position is about 3 cm up and 3 cm down, which feels good to me.
  19. MarkusB

    MarkusB Active Member Gold Contributor

    Joined:
    Jun 30, 2015
    Messages:
    315
    Location:
    Germany
    Balance:
    2,256Coins
    Ratings:
    +273 / 1 / -0
    My Motion Simulator:
    2DOF, DC motor, Arduino, Motion platform
    This is indeed a valid point.
    I have 2 shakers mounted to the upper backrest, and the main reason for this location was that it was the easiest way to mount them.
    As I mentioned before I did not yet configure them differently for the left and right side. So at the moment the whole seat is shaking. This is ok for me, because my focus is on basic engine vibrations.
  20. SeatTime

    SeatTime Well-Known Member SimTools 2.0 Beta Tester

    Joined:
    Dec 27, 2013
    Messages:
    2,001
    Location:
    Newcastle Australia
    Balance:
    25,134Coins
    Ratings:
    +2,336 / 20 / -0
    My Motion Simulator:
    DC motor, Motion platform, 6DOF
    I love the backrest heave - great idea :thumbs.
    • Friendly Friendly x 1