WIP - RacingMat's closed circuit Gseat

Hi guys!

In 2014, I thought about building a pneumatic Gseat with a closed-circuit in order to prevent the use of a compressor, valves, exhausts (leading to noise, oil, large footprint, cost).

1/ I have built a 2DOF full mobile : it was fun but enventually unsatisfactory
- long effects were not sustained
- movements induce 3 phases effects with wrong feelings:
during a braking, the 2 DOF starts to push you in the back (wrong) and then to restrain you with the harness (OK), and nothing special with a slightly tilted plateform (wrong).
- I didn't like being shaked (am I getting old? )

2/ I have owned a Gecko gSeat but having the backrest of an arm chair strapped on my back like a backpack, was not comfortable and the seat itself was not self-centered, so the user has to keep in mind to stay within the limits. If not, it bumps the mechanical stops. These factors are not good for immersion...

3/ Black friday made the GS-5 at 2500$ vs 3000$ but where would have been the DIY fun? And I don't like the position driven gSeat. I only believe in torque (or pressure) driven gSeat.
With a position driven gSeat, if you move away, the wrong force will be applied.

Since that I have spared some money and I'm starting to build my pneumatic closed-circuit Gseat!

The advantages are :
- no precision mechanics (bearing blocks, rollers, cardan shaft...)
- 1 DOF by pressure pad: no mixing needed. Gain of time in setups. Simplier code: 1 arduino per DOF.

Good Wife Acceptance Factors:
- preserve the racing seat esthetics: pressure pad under the seat cover.
- small footprint
- no moving parts around the seat dangerous for bystanders (2DOF ou 6 DOF actuators): pressure devices enclosed in a remote box (under the seat?).

The drawbacks are :
- low reactivity. This will be counteracted by tactile transducers (and by tubes of large diameter if necessary or faster motors).

The principle is similar to blop jumping!
you press on 1 side, the other side is pushed!



Here we go:




Meanwhile @xxpelle did an excellent job with his gSeat! The idea works
https://www.xsimulator.net/communit...imulator-dof-reality.8570/page-38#post-163816

And @volker metzger
https://www.xsimulator.net/com…deal-for-beginners.13089/
video channel https://www.youtube.com/channe…view=0&sort=dd&shelf_id=0

and @Banfy https://www.xsimulator.net/community/threads/a-foldable-diy-sim-rig-with-g-seat.14948/#post-201156

14 décembre 2015
Frex

https://www.facebook.com/video.php?v=1085019251531142

FREX:
"We are working on pricing.Standalone USB device. It will be around US$810, 770 euro plus shipping from Japan.
One set includes 2x2 bags for side sway or surge. 2 sets need for 2DOF."
"1 set for side effect when turn corner is very good.
1 more set on back and under knee for accel/braking.
So 2 sets will be full AirG-Force setup: 1620$"


ACME Worldwide Enterprises, Inc

ACME's Simulator G-Suit Systems – Training for G’s without the Actual G’s
Fighter jets operate in a dynamic world of banks, climbs, dives, rolls, vibrations, shudders. The jet continuously changes accelerations and G-forces. G-loads are an inseparable part of the fighter environment.

Pilots use the cues from the motion, pressure, and vibrations to fly the jet. The flight sensations are part of the pilot’s awareness of the jet’s status and situation. And, G-suit pressure is a companion to the G-loads in flight. Outside of the centrifuge, there is no way to apply flight G-loads to the crew in the simulator. But, G-suit pressure is an excellent way to train sustained G-cues. In flight, G load is either constant, increasing, or decreasing. The pressure in the G-suit starts at specific loads then varies according to G-Load above the pressure onset.



GForces Vest

"3RDSpace Ballistic Impact Vest" since 2009
with a compressor
for FPS games to feel the impacts

https://tngames.com/pages/how-it-works at 139$

Actuators:


The working angle of the servo is 180 degree or 270 degree (PWM: 500us-2500us)
Some remote controls can only output 1000-2000, it is recommended to buy 270 degree.

I choose these servos beacuse I thought the brackets are really convenient in this configuration to squeeze a wedge
Here is what I received.


Speed: 0.17sec/60 degree at(6v)
0.15sec/60 degree at(7.4v)
0.13sec/60 degree at(8.4v) -----------> 1,3 tours/sec
Torque: 58kg.cm.at(6v)
65kg.cm.at(7.4v)
70kg.cm.at(8.4v)
Voltage: 6v-8.4v
////////////////////////////////////////////////////////////////////
Speed: 0.13sec/60 degree at(5v)
0.12sec/60 degree at(6v)
0.11sec/60 degree at(7.4v) -----------> 1,5 tours/sec
Torque: 29kg.cm.at(5v) -1.9A
32kg.cm.at(6v) -2.1A
35kg.cm.at(7.4v) -2.3A
Voltage: 5v-7.4v
///////////////////////////////////////////////////////////////////
Speed: 0.19sec/60 degree at(5v)
0.17sec/60 degree at(6.8v)
Torque: 18.5kg.cm.at(5v)
21kg.cm.at(6.8v)
Voltage: 4.8v-6.8v
//////////////////////////////////////////////////////////////////
Speed: 0.16sec/60 degree at(5v)
0.14sec/60 degree at(6.8v)
Torque: 15kg.cm.at(5v)
17kg.cm.at(6.8v)
Voltage: 4.8v-6.8v

Max current
60 kgcm : 6,2A stall current at 8,4V
35 kgcm : 1,5A stall current at 7,2V

Beware that these servos need particular voltage supply...
I had to buy 2 power supplies
a 8DCV adjustable +/-3% -> between 7.77V and 8.24 V
and a 7DCV -> between 6,80V and 7,21V

R&D needs some sacrifices...

reserved

Code
Simtools 2.4 setup to prevent negative vales and movement
To prevent negative vales and movement take @yobuddy's advice and use SimTools v2.4, then allocate a '0' for the min value in Sway and Roll.

reserved

Some videos:

@xxpelle videos of his great project






with SFX100 actuators

Based on the number of bags their 2dof system has I should be charging over $3000 for a Jiffy G-seat. And I include a way of mounting everything on your seat too . Also I didn't mention it but an option is you can move the brake bladders on a Jiffy G-seat to the seat bottom instead and have the same braking functionality as the GS-5.

I'm not sure Frex sold many of them...

In my opinion, it's a really bad design:
- With a single double-action actuator for 2 cushions, the cushions are at half pressure at minimum and you cannot actuate them simultaneously for a bump...
- Plus they didn't take the opportunity to use both tubes for quicker response. They monopolise 1 tube only for occasionnal refill...
- And final, it's not pressure driven...

here is my fiber bucket seat being stripped!

seat clamps that are screwed, it's fine


but the clamps that are clipsed... I broke it...





I'll now try to determine how to attach tactile transducers (back, left and right), throne thumper below and brackets for a harness tensionner maybe.

I received the 7V PSU (for the 25kg servo not received yet) and I'm waiting for a 8V for the 60 kg servo
and pressure sensor
and ampmeter
I got work to do

that's really a small sensor (see arduino pins in comparison)! hard to solder it nicely...

When things get small you can use a flux pen, solder paste + hot air for a way better finish.

Hi!

first trials with this "pacman" configuration : servo at the center
the torque is too weak
it's sad because it was compact and mechanically fine


so I went the @xxpelle way for pressing the cushion: servo at the end

In these trials, I'm using the pressure sensor (under the white tape) to monitor the system.
If the pilot is moving away, the servo will press more in order to keep the right pressure (send from Simtools)
if the pilot gets closer, the servo will release its pressure.

I am interested in making an air operated G-Seat also.
I have purchased this Air bladdered car seat in hopes of adapting it/hacking it.. lol
I am going to try using air valves and a light weight compressor.
Here are some link.s to the stuff I have ordered.

Valves
Air Bladder Seat
amazon.com/ENGREPO-Massager-Electric-Adjustable-Especially/dp/B07NSHL76N

Here is the only video I could find about the air seat. Badly done but better than nothing!

that's interesting!

I don't know why URL links are disabled... but with some searchings, I found them.
I didn't go this way because of the noise of compressor and exhausts:
you forgot exhausts like this

MEDIA=amazonitem
@noorbeast: do you know about this (new?) tag? and how to use it?
it made visually disappear the links I posted but there are in the code view...

Code:

[MEDIA=amazonitem]amazon.com/dp/B01LYHYHEA[/MEDIA]
[ATTACH=full]68142[/ATTACH]

[MEDIA=amazonitem]amazon.com/dp/B07WR2ZQHY[/MEDIA]
[ATTACH=full]68143[/ATTACH]

In order to improve the control, I modded my servos to read the internal pot: it gives me the actual position of the arm.

Then I'll be in closed loop feedback.
I wrote an instructable to show how to modd the servo: https://www.instructables.com/id/How-to-Mod-a-Servo-to-Get-Closed-Loop-Feedback/