SimXperience II + JoyRider = Realism??

I haven't finished my SimXperience II yet, but I'm already dreaming of my next project. :lol:

I am considering combining a JoyRider cockpit with an SCN5 powered seat (maybe only 50mm or 100mm). I was thinking that the Joyrider could move as a car actually would (into the corners) and the SCN5's could move in the counter direction to simulate g-forces. I would add a buttkicker to touch it off with engine vibrations and the finer rumble effects.

I have been looking at several commercially available motion controllers because I'm not interested in the time and headaches required to assemble my own. Can anyone give me an approximate idea of how many amps continuous / peak are required for a joyrider type design? Pololu has a motor driver that claims to handle 13 amps continuous / 30A peak for only $99 here: http://www.pololu.com/catalog/product/777 Maybe there are better options? I also found an interesting USB stepper controller but I'm not sure that it would be up to the task. http://www.phidgets.com/documentation/Phidgets/1062.pdf

Any opinions on the best commercially available motors? Has anyone considered using SCN6's for a joyrider style design?

At first glance, this seems like it could be a really cool sim if they were combined properly. Any advice, links to information or threads on this forum, etc.. are greatly appreciated.

What do you think of this selfmade Simulator ?. I like it . It moves left, right and simulate G-Forces. Pedals, screen and wheel move also.

@Michael

I wasn't going to comment on it because I don't want to pick on other peoples simulators, but since you asked.... IMHO, the movements are very unrealistic, too slow and delayed.

This is too much weight for the SCN5's. In order to be able to move that much weight, you would have to slow them down quite a bit. Watch the video and pay close attention to how long it takes the actuators to move forward under heavy braking without even having anyone in the seat. Then, when the audio was depicting that the vehicle was accelerating, the cockpit was still braking which shows the obvious lag of some sort.

There is a way to accomplish this, but it isn't with SCN5's and this particular design IMHO.

hi bernard
you want to fit two dof 2 you ad none to it only more movement
working counterwise with theme you cood get none movement at all in a bad case.
scn5 are not strong you cood get problems going further offbalance
i sugest to google with cnc router,when you see one
take te z axis off and fit your frame where the mill sits
this way you get 4 axis
a hit in your back the frame moves forward ,braking the frame moves backward and chair forwrad
thisway you get double action.
regards Ad

I was thinking of using SCN6's or something like what Tronic is doing except for with a commercially available controller (if I can find one). My hope was that the Joyrider would have alot more movement that the SCN5 based seat and therefore the effects wouldn't cancel each other out.

However, you make a good point. If I'm to go to this much trouble, some different designs would give me additional axis!

:

Dear sirs,

It would be interesting to see a video of seat based simulator, to see only the feet on the pedals. To see how much they move when you are tossed about with longer stroker actuators.

I have been on websites where they sell these 2dof, 6dof simulators and they are so expensive.
And they seem so unrealstic because it seems like they are on a roller coaster.

What feels wrong with the seat based moving simultor that Mr Burns and etc people have built.
Are you guys not satisfied enough. Something not feel right from thigh to feet coardination?

Kind Regards.

I think you wont be able to combine a SFGT with a Joyrider. Just think about how the point of gravity displaces in relation to the rubber of the SFGT. The SCN5 are not strong enough to compensate that.

@Fahim

I am in no way dissapointed with my current 2DOF simulator. In fact, I am building two of them. It is just my nature to contemplate possibities for improvement, and I view this forum as a place to kick around these kinds of ideas.

@ Michael N

Well, the Frex type of sim really depends on the movement of the seat in relation to the steering and pedal. It attempts to recreate the sensation of the body moving about, however little, to the relatively stationary control devices in a racing car.

This design eliminates that function by moving everything in unison. I believe that moving the screen is okay, but the Frex sensation is destroyed. Ultimately, it attempts to be more like a Joyrider without adequate rotation to simulate the g-forces. IMHO it fails in both methods. About the only thing is can believably simulate are the bumps from the z-axis.

R-eng

@bvillersjr

I too have had the opinion for quite some time that a combination of joyrider-frex would be the most realistic simulator for achieving an inexpensive 3-dof.

The seat of the joyrider could be frexed to give a rapid z-axis motion, and the lateral rotation of the joyrider to give the sensation of g-forces. I think a limited range for the seat motion of + 25mm, with a pivot point more towards the front of the seat, might be the ticket. And if at all possible with X-sim2, move the seat with a soft start to the lateral motion, ending with a hard stop. This would then accentuate the correct forces acting on the body when initiating a turn-in manuever, with the capability to maintain lateral force through the turn.

But I won't be able to do this until I get my 100A h-bridge designed (with Thanos' assistance) and built.

R-eng

If you keep in mind, that the SCN5 provide 10kgf each only, you'll have a problem, if you move the whole seat.
As long as the Frex based sim is in default position (no rotational movement on X and Y) the body wheight rests totally on the pivot (rubber or crossjoint).

If you move the sim 30°, the force vector of the body weight does not go through the pivot, and therefor the body weight (approx 60-100kg) is applied to the actuators, which actually just are able to take 10kgf max (increasing the speed even decreases the kgf, as the SCN5 are stepper driven). Even with the stronger SCN6 model I doubt that it would work well.

For a combination of large movement range and massive rumble effects I'd suggest to use the some linear motors like te carmikaze simulator uses, in conjunction with a system das provides large movement range like joyrider or others.

Regards
Christian

Hi
Ego you are absolutly richt
scn5 ore 6 are fare to weak fore it.
frex use 100mm ones, only to avoid fore going to far off balance
dont forget when your wicht is 100kg and you go 10 degrade offbalance you need around 15kg force :thbup: only to hold.
gegards Ad

Yes I agree also about using the Frex system including SCN5's like this. The loads are all wrong. I only meant that applying similar movement to the seat itself, would probably make for a more defined sense of heave. Even if the joyrider were at 30' laterally, a perpendicular movement to the seat bottom would appear to the mind as z-axis, because it would not be moving the body sideways within the seat. And most importantly, in relation to the screen, thereby maintaining correct visual cues.

Then imagine lateral forces being applied to the upper torso, which even at 30', would appear to the mind as true lateral-G. The movement doesn't need to be excessiive; + 25mm either way I think would be more than adequate. We are only trying to introduce mind reinforcing cues. The Joyrider would provide the sustaining forces.

I would think wiper motors would be very suitable for this purpose, one for each force direction with quick and short motions.

R-eng

Hi,

IMHO, R-eng is right... I cannot add to the joyrider z-axis motion as it is, because the only way is to add it to the seat. But this will provide wrong motion cues in relation to the screen.

The only solution could be to use a pyramid like method, that goes like that:

Outer rotation frame: x axis
Middle rotation frame: y axis
Inner up-down frame: z-axis

With this order will avoid possible balance issues, but the up-down frame should be supported with at least two high power motors (this is where the 100A h-bridges fit!), even if the motion they provide is just a few inches!

The only other effective way is to use a 3DOF coupled motion cockpit like Force Dynamics! :brows:


Regards, Thanos

Actually, I think in theory a Joyrider could provide the heave motion with just the 2-DOF. Let's break down the anatomy of a bump (heave). It is the front wheels which make first contact, riding up and over the bump. Consider this a positive half sine wave. A few milliseconds later the rear wheels ride over the same bump, so consider this another half sine wave. But if we view the rear sine wave in relation to it's rotation around the front axis, it is a negative sine wave. Combined, this is a complete sine wave. To simulate this, the y-axis of the Joyrider would need to rise at the nose, return to neutral, then rise at the rear (or inversely, dip at the front) before finally returning to neutral again. This motion is a full sine wave. If it is a curb, then the x-axis is also invoked.

The problem is, are the motors and drive system of the simulator up to the rapid forces needed? We need big balls to do this with virtually no hysteresis. Current Joyrider implimentations IMHO do not!

But that does not make this theory invalid. And this is where we can have afrexed Joyrider to fool our minds into a 3-DOF with smaller/less expensive balls.

R-eng

Well, I approached to what you suggest already, using the flexibility of the PVC pipes to create a small up down motion, on bumps, curbs, even when entering a slope dive!

http://www.youtube.com/watch?v=d1BpKMXlvOw

Take a close look on the 3:20!!!

Somehow the cues was felt right, but more travel or less after effect bouncing would be better!

I haven't tried it with AMC controller yet, but in ADW show I inserted a fraction of that motion you see on the above video...

Regards, Thanos

The problem lies in the flexibility of the Joyrider pvc frame. This is nothing but hysteresis, and it would dilute or cancel, or even reinforce arbitrarily the correct motions desired.

This is extremely obvious in your video, throughout the whole video!

R-eng

Hi
Thanos and E-eng you,re making my day with your new h-bridge.
It means whe are able to get in fd range,and whe are able to make one fore around 500,00 euro
I,am working on that ,going to safe money so i can donate a actuater project and blow every thing open.
regards Ad

You guys are missing the whole point..
http://www.youtube.com/watch?v=nBVOnT36Qw4

The next step would really get you into this realm:

http://www.youtube.com/watch?v=dyRleobz7hY

Watch at the end when he crashes the car. To move that much weight in that much time, you'll be hard pressed to run motors unless you run 75volt system and large servo motors. You can get these things at OrientalMotors. I deal with them throughout many projects but they cost quite a bit.

Once you throw that seat past 10-15 deg in any axis, especially at a high rate of speed, to bring it back to center would require even more force especially if your quite heavy. What the FREX system does seems to work great as a package. Placement of the SCN5s and 100mm is probably reaching the limits of their capacity. I would think extending the force moment by extending the bar behind the chair would ease the forces (think of a long torque wrench). It would look ugly, so there is a price to pay.

To increase the movement on the FREX system would become an issue since now your moving at drastic angles towards/away from the screen, steering wheel and pedals, it would become extremely unnatural. So, you add the screen, pedals and wheel to the movement = much more weight (especially large flat screens), going to be tough with SCN5s in my opinion.