Of course there is an emergency switch that kills the signals to all motors within 1/1000 of a second.

But even if everything goes wrong there is no big danger. We had a situation that because of longer arms we used for testing, the cockpit positioned itself in a angle larger that 45 degrees and the two front motor arms were locked in place as they were parallel to the rods. I would expect something catastrophic to happen but no. The motors were just kept whining trying to reach the position. Didn't overheated, didn't shutdown the inverters from overload with fault message, didn't even dropped the mains fuses...!!!

If the same happened with DC motors you would be lucky not to burn the h-bridges or the motor coil!!!

Best Regards,
Thanos

Hi dawesdust_12,

Its not dangerous if you take some precautions. Worst case scenario is if one of the position sensors breaks down, the motor might start rotating continuously. It happened once, as the camera guy was walking by the motors and stepped on the cable of the sensor while it was working and pulled it of its connection, but did not affect anything else of break anything on the platform. The other 5 motors kept the platform in place...

The cost of materials is six times the following:
AC motor 0.75hp = $112
Gearhead 60:1 = $241
AC VFD drive 1HP = $224
Magnetic sensorless position encoder= $15
Supports, rod, arm = $55-$70

So much inexpensive to built it yourself instead of getting ready built 6DOF platforms (Ckas for example)

Best Regards,
Thanos

Last edited by tronicgr, Tue 28 May 2013, 19:08.

Hi,

Just want to drop in and present you my 6DOF stewart platform using AC motors and VFD inverters:

http://motionsim.blogspot.com/ ... sim3-using-beta-6dof.html

I use LFS and the new 6DOF plugin for x-sim3!!


Thanks!
Thanos

Quote from Hollywood :

You should put together a parts list and a basic "how-to" guide.. I'm sure others would love to try it themselves after seeing your awesome results.

Hi to all!

I just wanted to pass by, and show you my Home motion simulator site. Instructions to make my simulators are available, as well the electronics and the source code of the motion module... Although I present it mostly for use with flight sims it can work with LFS too just by switching the motion software to x-sim and attaching on it the momo wheel and the pedals.

Thanos home motion simulator

Regards, Thanos

Yes, I have done connection of LFS with AVR microcontroller for speed and tach (the second one is still in progress)...

You can read more info here:

DIY analog servo car instruments

Regards, Thanos

tristancliffe is right. The motion is that takes first role in simulating acceleration and true vertical force of gravity then... Sometimes there is no time to settle in a stable position to actually feel the vertical gravity force.

I spend 800 euro making it (all PVC pipes, metal dexion parts, motors and electronics, bucket seat). Now I said 30 times assuming the cost of the 301 is around 30k, If I'm wrong correct me...

Regards, Thanos

The latency is very small (below 8ms) and the delay you may see in some cases is from the servo ramping code I implemeted to protect the motors from wear during fast forward-backward action.

After all it performs very well for the junkyard compoments I used. And it costs 30 times less that the professional 301 force dynamics...

Regards, Thanos

This is my full motion simulator that I recently converted it to car racing simulator with the use of x-sim software. And with the help of this forum I fixed the data output acceleration data.

Its based on the old joyrider design but I motorized it with the use of car windshield wiper motors.

For the motion, I used a Basic Stamp2 microcontroler to read the data from the USO interface of the x-sim. And the motors are driver with the use of Parallax's HB-25 motor controllers that converts the wiper motors to big servos!

http://www.x-simulator.de/forum/uploader/pool/PICT0576.JPG

http://www.x-simulator.de/forum/uploader/pool/PICT0578.JPG

http://www.x-simulator.de/forum/uploader/pool/PICT0579.JPG

http://www.x-simulator.de/forum/uploader/pool/PICT0580.JPG



Full Motion Sim on LiveForSpeed using X-sim #1
http://www.youtube.com/watch?v=QWzkWVQ1SMQ

Full Motion Sim on LiveForSpeed using X-sim #2
http://www.youtube.com/watch?v=qffTsdikHlY

Full Motion Sim on LiveForSpeed using X-sim (closeup)
http://www.youtube.com/watch?v=3PiZFq6b3ko


Enjoy, Thanos

Quote from the_angry_angel :

Out of interest, do you find that you're hitting the limits of the serial port (in terms of bandwidth) for your motion platform?

No, not at all. Even if I used only 2400bps for the communication it can handle even more data (i.e. more axis). All is depending on how fast you process the data and drive the motors so to be ready to receive the next set of data in time. The data it receives anyway is just a few bytes (an sychonization character and x,y in the form of HEX(0-255) each.

X-sim software indicates 4800bps as the minimum baud it can work without data loss, but it works for me fine, with the half speed (2400)!

Regards, Thanos

Thats, right. It is indid S2 licensed. But i'm not using it online since I moved to bigger house to have room for my motion simulator. (no internet, no phone yet... )

Anyway I think I found what I needed to make the math formula so the acceleration data will be car-oriented...

I'll have soon ready a page that gives details on how to construct this motion simulator (for free) on your own here: http://www.etherealsounds.com/


Regards, Thanos

Hello,

I'm Thanos from Greece, Athens.

I made a motion simulator cockpit that can be used for either flight or driving simulators using electric motors. See it how it works on flight simulators:

http://www.youtube.com/watch?v=TBD9Juqx7PI
http://www.youtube.com/watch?v=dYOmygb5XtQ

Recently I had connected x-sim software with Live4Speed and used the X,Y acceleration data to drive my motion cockpit. It works really well and you can feel even the smaller acceleration effect but...

...While the data are comming right when starting the race, they seem to shift around i.e. from x to y while driving around the circuit. Its like the acceleration data aren't car oriented but circuit oriented...

See the attached photos where I accelerate in various places in the circuit and receive different values for x,y...

http://www.x-simulator.de/foru ... r/pool/LFS_acc_data_1.gif
http://www.x-simulator.de/foru ... r/pool/LFS_acc_data_2.gif
http://www.x-simulator.de/foru ... r/pool/LFS_acc_data_3.gif
http://www.x-simulator.de/foru ... r/pool/LFS_acc_data_4.gif

As you may notice on the photos I gave above, the acceleration direction is indeed relevant to the orientation of the car so you need to use this output to calculate the correct x,y acceletation values. It maybe be a simple calculation like shift the x data to y by the amount indicated on the orientation data. But its beyond my mathematical knowledges.

I know that many commercial motion simulators (i.e. Force Dynamics 301) using these data LFS and somehow calculating the acceleration data to be car oriented.

So if there is a ready math formula for this I would really appreciate it! I could give it then to the xsim programmers to fix it for everyone that use x-sim to play LFS on a motion simulator.


Regards, Thanos

-------------------------------------
Home made motion simulator:
www.motionflightsim.eu.tp

Quote from madprof :

Thanks, I got it all sorted today, I translated the German x-sim web site and figured out I was using the wrong outputs, x-sim seems to be a very under rated program and I would recommend anyone who wanted to make a motion sim take a look. Now this program is working and looking promising I am going to start the build process.

Hello, Calvin

I have the same problem with acceleration values shifting from x to y. How did you fix it?

Regards, Thanos

Last edited by tronicgr, Mon 9 Jul 2007, 14:01.