DIY Virtuix Omni

[colocolo]

nice, will you add it to an Omni?

[blazespinnaker]

Regarding the Original Omni - I believe by Jan.2014 there'll be so many adaptations of the Omni that the original will be obsolete.

If the energy keeps up, I think you're right.

[blazespinnaker]

How about doing it like you say then putting a layer of hardboard over it? You could even wet it to get it to take the shape better?

Yeah, I'm running into some problems though with the MDF. It's not as thick as I'd like it to be and the sheets I can get it in can only reasonably be split into to 2 with lots of wastage. I can still do it, but I want to create a solution which is less than $50

[blazespinnaker]

Today, i had another idea about a ODT.
Mainly like the Omni, but flat and no grooves.
Your shoes would have cleats like football shoes that can extrude out if no pressure is applied.
So, if your shoe glides over the low friction surface the low friction cleats retract and would glide backwards until they drift over the center of the device. There the platform has many holes like a grid. The pins extrude out and give you halt when stepping again forward.
I had this idea also with steel pins that would be attracted by a electromagnetic shoe sole, but think
this way its much simpler.
I think retractable cleats at the shoes and small holes in the center of the Omni could eventually improve it.

I like the out of box (literally) thinking approach here.

I'm starting to think that the dish might not be the right way to go. But rather say roller skates which stop rolling when you apply vertical pressure.

[Shannonb1]

Today, i had another idea about a ODT.
Mainly like the Omni, but flat and no grooves.
Your shoes would have cleats like football shoes that can extrude out if no pressure is applied.
So, if your shoe glides over the low friction surface the low friction cleats retract and would glide backwards until they drift over the center of the device. There the platform has many holes like a grid. The pins extrude out and give you halt when stepping again forward.
I had this idea also with steel pins that would be attracted by a electromagnetic shoe sole, but think
this way its much simpler.
I think retractable cleats at the shoes and small holes in the center of the Omni could eventually improve it.

I like the out of box (literally) thinking approach here.

I'm starting to think that the dish might not be the right way to go. But rather say roller skates which stop rolling when you apply vertical pressure.

I think at that point its just a wizdish, as being flat. the point of the omni is to allow running and walking, the slope gives you that foot forward natural motion. I would imagine you would need a ton of holes to not get caught up and stay on top, then you would need to have as few pins as possible and I cant imagine that changing direction would go well.

i could see using 1/4 lumber, 15$ -20$ sheet, cut slices, like a u but more bowl shape, make 2 and corss them. then go across on an angle, hooking them together in the center. Then take your mdf into 1/4's, and flex it down. You could layer this if need be. Then spray glue down a piece of material for the slippage.

If you used a lubricant on the slope, you could then use a circle in the center with sand in the glue to give you that sticking point. somewhat like sand paper. you could change the texture to be more rigid as you reached the center to give you a gradual slowing affect.

[blazespinnaker]

Yeah, I think the key is going to be experimenting with different approaches. Build up rough prototypes, try it out, learn from it, and move onto the next.

There are certain fundamentals I think are necessary though:

- Harness (mine is from the ceiling)
- Wireless OR (laptop in backpack seems to be the easiest)
- Motion cap (iPod strapped to shin is how I go)

[bobv5]

A slightly silly thought I had for testing. Instead of building full size prototype that might not work, you could make very small prototypes, and walk on them with your fingers.

[blazespinnaker]

Not a bad idea! I'm using that

Interestingly, for fun, I did some research on injection molding. The cost of the molds will be in the thousands, but once that's done each unit will be pretty cheap (I estimate around < $50, before shipping).

The problem, though, is weighing the base down. If it's not heavy enough, it's going to be slipping and sliding all over the place. It's going to need significant weight for that.

My thought is that Jan should have managed expectations better on that. I wouldn't be surprised if he recommends people buy sandbags.

[JanVR]

Hi Blazespinnaker, keep in mind that there are no upward forces on the Omni, so weight at the bottom is not necessary. All you need is high-friction rubber feet at the bottom to prevent lateral slide. Think of a regular treadmill, which does not require sandbags either. Hope this helps!

[geekmaster]

A slightly silly thought I had for testing. Instead of building full size prototype that might not work, you could make very small prototypes, and walk on them with your fingers.

It looks like some reward tiers in the Virtuix Omni kickstarter come with a "finger walker" version of the device.

MINI OMNI PLUS T-SHIRT PLUS SIGNED POSTER SET: Get all rewards above and the Omni... in mini format (5"x4")! Put the Omni on your desk or make it a gift. Throw pens or paper clips in the Omni... or an action figure.

[blazespinnaker]

Hi Blazespinnaker, keep in mind that there are no upward forces on the Omni, so weight at the bottom is not necessary. All you need is high-friction rubber feet at the bottom to prevent lateral slide. Think of a regular treadmill, which does not require sandbags either. Hope this helps!

That's very surprising Jan, thanks for the tip. (no pun intended)

As for comparing with a regular flat treadmill (which is pretty heavy, btw!), you're not creating any lateral force vectors that'd you be creating when your feet hit the side of a dish treadmill.

[D3n]

What about using lots of ball bearings on the sides of the dish, all held in place by magnets underneath the surface (so they don't lose their place, but each can still rotate). Depending on the strength of the magnets (and the size of the ball bearings*) the friction level can be adjusted. And since the steel balls rotate, you don't need special shoes, any would do, i think. Also, the wear and tear would be inexistent.

*small bearings are faster, bigger are slower (maybe use the bigger ones around the center of the dish, to further slow you down?), at least that's how i imagine them to work.

[Applemung]

What about using lots of ball bearings on the sides of the dish, all held in place by magnets underneath the surface (so they don't lose their place, but each can still rotate). Depending on the strength of the magnets (and the size of the ball bearings*) the friction level can be adjusted. And since the steel balls rotate, you don't need special shoes, any would do, i think. Also, the wear and tear would be inexistent.

*small bearings are faster, bigger are slower (maybe use the bigger ones around the center of the dish, to further slow you down?), at least that's how i imagine them to work.

How will the bearings still rotate?

[D3n]

They rotate in place. Each ball bearing sits next to another ball bearing, while all of them sit on a solid base (dish). In the center of the dish is a high friction surface, to help you stop. The ball bearings sit in their place with the help of gravity and magnets. I'm not even sure if the magnets are necessary, but they could help customizing the friction level of the slope, for example using stronger magnets around the base, to help you stop easier as you reach the base of the slope (this is a more elegant solution than using different sized ball bearings, mentioned in my initial post).

Here's a rough drawing:

[shole]

balls are a bit unnecessary since there's almost never any circular movement
most of the leg movement is in center-edge directions
so if not a slippery surface, how about a wheel conveyor arrangement?

obviously much tighter arrangement than this
there could easily be some pressure sensors under each one to do the actual tracking too

[D3n]

balls are a bit unnecessary since there's almost never any circular movement
most of the leg movement is in center-edge directions

Yes, but i think my design (if it works) it's much easier to construct. You need lots of small ball bearings (smaller than in my drawing) and you just fill the sides of the dish with them. Magnets are optional, but it would be really cool to test this setup with some strategically placed neodymium magnets, though.

[shole]

just came across this
http://www.youtube.com/watch?v=QbanKNuK-bA
at the VR wikinet
looks like there's a fairly comprehensive collection of existing ideas

[JDuncan]

D3n idea had me thinking how the balls would be friction-less.

What I thought up was the balls are inbetween a square grid, and a flat surface.
The metal grid is walked on by the feet, and a ball pops out from one hollow square and the feet use the balls to roll the feet over the square grid.

The balls and flat surface below the metal grid are coated in teflon PTFE spray.
The metal grid over the balls doesn't clamp down on the balls but lets the balls jiggle in the square socket it's held in.

The metal grid is able to be lifted up so the balls and flat surface can be recoated with teflon for maintenance.
A tray slides out from the side of the metal grid after the balls and grid are locked is.

[Applemung]

They rotate in place. Each ball bearing sits next to another ball bearing, while all of them sit on a solid base (dish). In the center of the dish is a high friction surface, to help you stop. The ball bearings sit in their place with the help of gravity and magnets. I'm not even sure if the magnets are necessary, but they could help customizing the friction level of the slope, for example using stronger magnets around the base, to help you stop easier as you reach the base of the slope (this is a more elegant solution than using different sized ball bearings, mentioned in my initial post).

Here's a rough drawing:

I still don't see how the balls will rotate. Once you step on them they will be a solid mass. Even with a teflon base the bearings will not rotate. Lets just say one bearing starts turning, then the second one behind it tries to turn but cannot because the turning direction of the bearing before it is in the exact opposite direction. Picture two inter meshing gears, they can not turn in the same direction.
It's the same thing with bearing but only surface friction, that friction would overcome the force of the magnets and you would get bearings all over the floor.
You would need individual bearing holders or bearing cups separated slightly to allow free movement.

With that friction plate you are going to want to have zero friction for about 50% of that plate and only have friction in the very centre otherwise you are going to be running on an awkward angle as though you are climbing the side of the omni.

I like that wheel idea though, I could see it working but the cost of bearings would set you back over $500 alone.

[D3n]

It's the same thing with bearing but only surface friction, that friction would overcome the force of the magnets and you would get bearings all over the floor.

Yup, this was the thing that worried me. I thought that with carefully chosen magnets, the force will be enough to keep them in place, while still allowing for rotation (albeit with friction, as you said).

But....what if you add teflon tape around each row of ball bearings? If a ball's diameter is 10mm, then the teflon tape height would be a bit less, 6-8mm. Each line of tape should be glued to the dish base. Like this:

With that friction plate you are going to want to have zero friction for about 50% of that plate and only have friction in the very centre otherwise you are going to be running on an awkward angle as though you are climbing the side of the omni.

Good point.

You would need individual bearing holders or bearing cups separated slightly to allow free movement.

Something like JDuncan's suggestion a few posts above, but it's hard to implement, imo.

[Applemung]

Yeah that would work with the teflon separators. You would have to reinforce the separators with a metal core otherwise the bearings might still escape because teflon is quite flexible.
You would have to use rear earth magnets of a cylindrical shape, you wouldn't want the magnets too close to each other. The only thing now would be the over the top cost required to build this unless you could buy bearings for 1 cent each. Each bearing also needs a magnet and good magnets are not cheap. And then with all the magnets you would get a sizeable magnetic field and stuff any sensors that you would have on your body.

I think that the idea in general is good because it would require little maintenance to keep the unit in good shape but the cost to make it would be huge. I would be close to the cost of my own omni!

[D3n]

Yeah that would work with the teflon separators. You would have to reinforce the separators with a metal core otherwise the bearings might still escape because teflon is quite flexible.

That or stronger plastic of the same material that you use to put under furniture, in order to slide it on the floor. I don't know what's the name of those, but i think it's the same plastic that's used on virtuix omni's feet pads.

You would have to use rare earth magnets of a cylindrical shape, you wouldn't want the magnets too close to each other.

I think you could hold them in place by other means and use any shape. But this needs testing, to find the right magnets...

The only thing now would be the over the top cost required to build this unless you could buy bearings for 1 cent each. Each bearing also needs a magnet and good magnets are not cheap. And then with all the magnets you would get a sizeable magnetic field and stuff any sensors that you would have on your body.

You can find steel balls on ebay quite cheap, for example, without much time spent searching, i've found these: http://goo.gl/wmrmuK
Magnets are more expensive, especially rare neodymium magnets, and i'm not sure how many you would need. The magnetic field is another potential problem, as you said, but depends on how many magnets you use. Again, this is hard to plan in theory, it needs testing, imho.

[android78]

The problem with the ball bearing design, I think, is that you will still need a reasonable amount of friction to prevent the user from constantly feeling like they are slipping. Also, having the middle with a lot more friction then the rest will make it worse... I think you should at least have fairly consistent friction.
I'm not sure the benefit of having the balls on the sides as opposed to just low friction materials. Also, wouldn't it be cheaper to just have the dish and an array of balls on the bottom of the shoes?

[D3n]

The problem with the ball bearing design, I think, is that you will still need a reasonable amount of friction to prevent the user from constantly feeling like they are slipping.

The force of the underneath magnets should add a bit of friction. Or maybe just adjust the material the balls are sitting on.

Also, having the middle with a lot more friction then the rest will make it worse... I think you should at least have fairly consistent friction.

Yes, frankly i was under the impression that the Virtuix had a similar high friction centre, hence my design. Probably i was wrong....

I'm not sure the benefit of having the balls on the sides as opposed to just low friction materials. Also, wouldn't it be cheaper to just have the dish and an array of balls on the bottom of the shoes?

As i see it, one advantage was that you didn't need special shoes for it, maybe you could even use a pair of socks and have a little foot massage. The shoes with ball bearings would be hard to make and difficult to control, imo. But who knows, maybe it could be a better solution if you design them properly, so you don't slip everywhere.

[android78]

for an easy one, you could just have a sock that goes over your shoes and a polished dish. I just have trouble seeing benefit to the balls.

[colocolo]

a ball carpet in a dish will rotate but it depends mainly on the incline angle.
it also depends on the size of the balls. i use 16mm marbles. my first attempt wasnt that sucessful.
i use a satellite dish and and at the lower incline , but this is almost flat and in the middle of the dish, nothing moves. even with silicon spray. But as i said its in the middle. next i will expand my spiral to the outer incline levels.
Remember those ones?
EDIT: I have changed the inclination angle of the dish. I have put one side more upwards. Now it works.
Definitely depends on friction and inclinination angle.
I hope that it will work also with the outer part of the dish.

[colocolo]

@ 3Dn you cant have a high friction base in the middle. Your shoe wouldnt glide backwards anymore and you would be stucked on the balls because of the transistion moving balls/high friction. If one half of your foot doesnt move anymore because of the high friction at the back than the other front part would stop the balls.
The middle has to be low friction.

[oma]

Hi,

I've been thinking kind of different aproach to this realistic 'walking/runnin' virtual reality. Im no where expert of building anything nor i have knowledge of being the superior in physics. So i made a little pciture also to illustrate what i have had in my mind, but it didnt go that well as my photoshop is buggin.

I named my idea as every idea needs a name even tho if its a lost case or a perfect idea, so i named this Moonwalk (MJ anyone!! ).

Moonwalk idea:
Instead of being on the ground you would be floating in the air, yes i know it sounds stupid (how do u get in there, you no get no friction thats space science maaan). You could propably build somekind of framework to keep u in the air safely that your feet wouldn't hit the ground, so how would you get in there propably somekind of harness would needed so you could do it alone.

Friction ideas:
A Ball:
so how would you get friction, i've thought about couple ideas 'A ball' which sounds funniest shi* ever (but a ball on a disc and somehow stopping it to spin too much->giving friction) and somehow make that your feet would barely hit the ball (cuz its round, so you couldnt stand on it) and only when you take those speedy steps would it spin and give friction.

A Disc (similar to a omni):
You would have similar icy disc where to run but then this kind of build would only change the belt system to this 'floaty' thingy. (personally i'm not that hyped for the omni, it looks pretty awkard and in tests on a youtube it looks difficult and honestly very very slippery.

Flat area:
If you really want you can actually "run" in your home floor by putting socks on and try to run without moving, it looks silly but its pretty tough on your muscles. But then would you need the whole framework, propably not. This flat area tho could be changed with somekind of bearings or something to give your legs a feeling that you are actually running forward instead of being stable.

Turning?:
I never said ideas are perfect, mostly you will be running forward, no strafing but you could propably invent somekind of upperbody movement so the part of the frame would turn (meaning you would use ur shoulders to turn)..

EDIT: forgot about crouching and jumpin
Croaching and Jumping
In theory you could have somekind of suspension on the framework so you could do a "small" hop in the air to have "jump" effect, croaching would be much harder to do as you are floating in air so i dont know how would you do it but maybe with the suspension setup, this would be possible at least in some small level of croaching..

Anyway just wanted to empty my mind even tho if its a stupid idea. Would love to read more about your aproach to this 'design' or your own.
btw im finnish and my english isn't perfect so hopefully you got the idea anyway.

[colocolo]

it would work, but the sphere would have to be very big in order to make normal sized steps.
Look.
http://www.youtube.com/watch?v=cflWk74iOYg

[cadcoke5]

First, a comment about fabrication. If you can make a section from urethane foam, you can make a vacuum formed copies fairly cheaply. Once I got a quote of $4 per pull, plus the cost of the plastic. This was for a machine that handled a 4ft square piece of plastic. Their are design consideration that must be considered, like undercuts. E-mail me if you want ideas on how to handle that.

I really have doubts about the whole idea of making the system frictionless for the user. You really need friction to keep yourself upright. The Whiz disk works because it has some friction, allowing the user to keep upright. It is just slippery enough to allow shifting your feet back and forth to simulate walking.

In my mind, for a better simulation, you to settle for something less mechanically complicated than controlling two feet. One idea is to ride a 3-wheeled all terrain vehicle. They have a good turning radius,and perhaps a simulator can have a bounce actuator to allow the user to feel the ground they are going over.

In an alternative idea that keeps the user's hands free, I am picturing a dish, that hosts a smaller round platform on casters. In the virtual world, the user rides on a round hoverboard. In the real world, the round platform on casters is pulled up the slopes of the dish to simulate the forces of acceleration and stopping. Wenches with cables are what pull the hoverboard around. See my attached drawing below.

VR Round Hoverboard.png

The user's control their hoverboard by leaning the direction they want to go. For example, if they want to start moving forward, they put some weight on the front of the hoverboard, and then the system pulls the hoverboard up the slope of the dish. The user feels the shift of the board forward, and so they must stay leaning forward on the hoverboard as long as they are accelerating in the virtual world. Then, as they reach coasting speed, they lean back a bit, and the system moves the hoverboard back to center.

-Joe Dunfee

[blazespinnaker]

Someone mentioned a turning unicycle in another thread as well, which was a good idea.

You're quite right about the frictionless surface, though. I've tried it.. it's .... unpleasant. You really need the supports, which I fear will chaff over time.

[cadcoke5]

The user's control their hoverboard by leaning the direction they want to go. .... and then the system pulls the hoverboard up the slope of the dish.

I just realized that the cables don't need to pull the hoverboard up the hill. The weight of the user putting weight on one side, will cause that side to seek the lowest level. So, the cable system may not even have to be actuated. Perhaps it only has a spring to keep things wound up, and applies controlled friction. That just reduce the cost and complexity of the system immensely. Perhaps the sensors I described as being on the board, are only in the spools, tracking its rotation.

The type of cable system I now envision is sometimes called a tool balancer. On E-bay you can cheap Chinese ones with 10lbs lift, for about $20 including shipping. 5lbs, for about $15. I imagine more expensive ones may incorporate some system to prevent the reel from spinning too fast when it is suddenly without a load, and perhaps a braking system.

-Joe Dunfee

[colocolo]

The user's control their hoverboard by leaning the direction they want to go. .... and then the system pulls the hoverboard up the slope of the dish.

I just realized that the cables don't need to pull the hoverboard up the hill. The weight of the user putting weight on one side, will cause that side to seek the lowest level. So, the cable system may not even have to be actuated. Perhaps it only has a spring to keep things wound up, and applies controlled friction. That just reduce the cost and complexity of the system immensely. Perhaps the sensors I described as being on the board, are only in the spools, tracking its rotation.

The type of cable system I now envision is sometimes called a tool balancer. On E-bay you can cheap Chinese ones with 10lbs lift, for about $20 including shipping. 5lbs, for about $15. I imagine more expensive ones may incorporate some system to prevent the reel from spinning too fast when it is suddenly without a load, and perhaps a braking system.

-Joe Dunfee

i think this concept is really cool. there are certain games that would certainly need such a thing.
snowboarding or surfing, that would be awesome.

[colocolo]

here is a new basic idea i have thought of concerning flat ODTs.
the basic idea behind this is to split a spherish underground into many pieces so it would be free to move around a inner core. Much like the crust of the earth does.
i dont know if this would work and if those pucks or any other shape would be better to shift around.....
although i have not drawn every puck it should be understood as a all-around conveyor belt.
Small motors mounted at the edges of the upper side could move the belt in any direction
or from the lower side. therefore smaller holes have to be made into the lower surface so that the motor would have a contact point to the floating pucks.

[JohnyX]

Hello guys.
Im working on odt prototype and recently had very interesting consultation with specialist in locomotion rehabilitation...
After watching the way people move and position their body in current rigs he suspects users will develop lower back problems due to unnatural vectors of force being applied to areas of the body.

[android78]

Hello guys.
Im working on odt prototype and recently had very interesting consultation with specialist in locomotion rehabilitation...
After watching the way people move and position their body in current rigs he suspects users will develop lower back problems due to unnatural vectors of force being applied to areas of the body.

Interesting. I wonder if it would be better or worse then sitting in a chair (or bean bag) for hours at a time though.

[colocolo]

Hello guys.
Im working on odt prototype and recently had very interesting consultation with specialist in locomotion rehabilitation...
After watching the way people move and position their body in current rigs he suspects users will develop lower back problems due to unnatural vectors of force being applied to areas of the body.

Interesting. I wonder if it would be better or worse then sitting in a chair (or bean bag) for hours at a time though.

yeah,good argument. sitting remains still the most unnatural position what i can only confirm.

[QUAKE]

I'm gonna try to simplify a bit here:

Physical bowl: old satellite dish, or firebowl, possible skatepark bowl call up a local skatepark ramp making place?

Tracking/Sensor: DDR dance pad, some kind of compass in pocket (smartphone, wiimote, etc) to track which direction legs facing.

Top layer: plastic floor protector, or other slipper surface, wool socks?

Result: omnidirectional treadmill that's 100% compatible with all WASD games.