Tutorial G-Seat where to start

Well you asked for it...
Now you got to watch it

This is a short video of the servo movement with No Limits2 Crystal Beach.
The servos are unloaded so it would be a bit more dampened with me in the seat.
However at some points it does bat you around quite a bit.
The 2dof system is turned off for the video.
Between 14 and 40 seconds the paddles are deployed about 50% as the coaster is being lifted up the hill.....

And the front view of the paddles movement.
Again at about 16 seconds the paddles are about 50% deployed as you climb the hill....

That's quite a lot of travel - will shove you out of your chair!

Good work though I'm actually quite impressed with the speed of those motors, I was under the impression they were quite slow.

If I recall correctly the swing out on the edge is about 50mm total travel.
I did position the camera to maybe exaggerate this degree of movement.
The original GS4 Gseat used 4 Hitec HS 805's which have a stall torque of 19.8kg.cm and a rotational speed of 0.19sec/60° at 5v dc. In the YouTube video Barry shows the power supply to be a 5 volt Meanwell.
So each heave paddle only has about 80kg cm of power.
The ASME-4a servos are faster and have a lot more power than 80 kg/cm
I have to assume that with the big wide paddles and such low torque the servos in the GS4's heave section are always stalled out .
I have been playing around with a different paddle mounting system. Instead of the typical pin hinge I thought I would try ball bearings and 8mm rods.
It might be too difficult to get everything aligned especially with the wooden based seat.
here is a picture of the 8mm 3d printer rod bearing parts I am going to try..

happy to report the G seat is done...
And it works.
I am using 2 of the 260kgcm servos per side on the heave paddles
I did try using just one and it actually worked fine.
Adding the second one just reduces the strain on the servos.
When paralleling the servos it is important that they have the same throw and centerpoint. I was able to use a trim resistor to accomplish this.
I soldered a small 50k ohm pot between the spot the green arrow is pointing to and then one side or the other where the pink arrows point.
With this extra trim pot I was able to match both servos exactly.
The doubled up heave servos have no problem pushing my 114+kg with authority.

Update:
The Gseat was performing well ..... Until I burned out a motor.
Maybe I learned something from this minor disaster... I had one of the slower Servos that is rated @.5sec/60deg. And it appears to use the exact same RS555 brushed motor. So I swapped it out and the Gseat is running again..
I did run some tests to try and figure out exactly what motor they use.
The only thing I could really test was RPM/Volt.
So this is what I measured.
12vdc no load is 4900 rpm.
18vdc no load is 7260 rpm.
24vdc no load is 9600rpm.
The rs 555 motor is a 5 pole design which gives it greater torque than the same sized RS550 motors.
From what I can tell the motor is a Mabuchi RS-555ph-3255 Which is size 32wire with 55turns.
The problem is finding this motor with a solid shaft and not cut with a flat.


The original motor also has a pressed on pinion.
13 teeth...

The windings on the defective motor looked fine. As a matter of fact I completely unwound it and there was no discolored wire.
I am guessing it was a wire attachment to the commutator that broke.

The motor failed as I was enjoying an intense few laps at the Daytona Speedway oval.
The motor that failed was the right side heave sway .
So It was working really hard in the nearly constant left turn.

The second servo on the right heave/sway side was not even warm.
I am assuming there were some problems in my mechanical adjustment/pushrod geometry...

A minor glitch that was easy to solve (because I had a spare motor)....
No More Daytona !

When reading your post, I also immediately thought of adjustment problems.

When multiple servos are used in parallel with a rigid connection, there is always the risk that they are working against each other. Even with an accurate adjustment it can be that manufacturing differences between the servos lead to minor movement deviations.
I used two servos for moving my seat backrest up and down, and the picture below shows how I mounted the rods in a way that movement deviations are compensated.

Yes it is a problem I am going to have to take care of. It is complicated by the fact that the aluminum paddle is so long (front to back). There is lot of flex in 3mm aluminum plate.
I found a new picture of the real Gs-4 seat and the heave paddles are much shorter than the surge. Mine are the same length about 350mm from the front to the back of the chair.
I believe another problem is the pushrod attachment points are spaced about 150mm apart. There maybe a lot more body weight on the front attachment point than the rear.
If I reverse the output position on one of the servos I can move the attachment point much closer together. But that would take several hours of work.
Right now I am going to try and preload the rear servo a bit more than the front and hope it even things out.
I really need to measure the current draw of the motors in real time to get a better feel for what is going on.
It would also be nice to have a source of replacement motors handy.
From what I can tell these are about $3.00 motors so they are cheaply made.
Like I said the motor that failed, failed at the commutator.
I did email the servo manufacturer Wingxine and asked for help with spare motors.

I see how your pushrod arrangement can compensate for any unevenness.

Thanks for posting.
Tom

It's not the first time I have seen these motors/servos fail. Good luck in sorting it out.

I have been looking for a source of replacement motors and pinions.
There is a pretty low cost solution.
A replacement pinion is the GoolRC 2Pcs 48DP 3.175mm 13T Motor Pinion Gear for RC Car Brushed Motor
A replacement motor is the Mabuchi Model: RS-555PH-3255
The pinions are 2 for $3
And the motors run $6- $10 each.
https://www.ebay.com/itm/GoolRC-2Pc...e=STRK:MEBIDX:IT&_trksid=p2057872.m2749.l2649

https://www.ebay.com/itm/Mabuchi-55...h=item3f44169b0c:g:-iMAAMXQ9u1REgiv:rk:3:pf:0

Would you change something in the size and position of your paddles if you where to do it again?

I understand there is consensus for hinging the paddles at the edge/almost at the edge. The first g-seat I saw was Bergisons "classic" build. I think that he stressed the importance of hinging the seat paddles some 30(40?)% towards the center. This was to have both the cues of "sink" an pressure. He had the paddles very tight coupled with the lower harness tensioner, as they where driven by the same motor: pressure from the paddle and imediate release of tension in the harness and vice versa.

I would not change anything except making the the heave paddles 30-50mm shorter (front to back).
In my opinion the heave paddles are optional.
The heave effect for me did not work..
However the bottom paddles are very effective for a sway effect.
I think the hinge line should stay nearer the inner edge of the paddle.
I was concerned about pushrod leverage and placement.
But no real scientific reasoning
I am building a new rig and I have a new seat for it.
I plan on installing the back paddles from the old seat but less likely I will do the heave paddles.
The new rig is 3dof with heave being the 3rd dof.