6DOF Costs?

I've been searching around all night trying to see what some people have ended up spending on some of these 6DOF setups? I've done tons of DIY work, non in motion simulation but probably quite a bit of otherstuff that would probably apply to a lot of what goes into building a motion simulator.

Basically I'm wondering what it'll cost to build a 6DOF machine? Mainly looking for racing simulators and what not but very possible that I'll get into flight Sims as well in the near future. I have a wife and 2 young kids and she's not going to let me build more then one machine so I'd have to get what I want the first time

Are these anywhere near fast enough or tough enough weight wise when 6 of them are used? I'm guessing the payload is going to be something like what 2 of them can hold up by them selves?

http://cgi.ebay.com/ws/eBayISAPI.dll?Vi ... 1575wt_905

I'm guessing the actuators or motors are going to be one of the most expensive parts.

There is a world of difference between a 6DOF platform and an other 6DOF platform.

Your cost will greatly depend on basically everything. Size, weight, payload, performance, the amount of work you can do vs. parts you have to get off the shelves, electronics for interfacing, actuators used, etc, etc.

Your price right now is anywhere between ~$4.000 - $the starlit sky. As I said, depends on what you want.

I don't think you can stop after your first platform, so you better sort that out with your wife!
If you build a 6DOF platform, it will technically never be done... There will always be improvements you will want to do to it...


edit: Those actuators will be way too slow (36/min unloaded) and IMO way too week for fast movements. Although 6 of them will be quiet strong.

sirnoname is right, the actuators will be the least of your worries. They are only a very small (but very important) parts of the system.

So there aren't any DIY controller designs out there for a 6DOF setup?

And yes tweaks and mods are ok...but if it looks like I scrapped and built a completely new machine she may get upset...lol


Edit: More needs,

I don't think I'd need to support more then say 500lbs or so? I've been trying to search for actuators and what not... it seems you can get a lot of air cylinders for pretty cheap. Of course if there isn't a cost effective or DIY controller to be built for such a beast then really it wouldn't matter.

You need very expensive air cylinders with very expensive valve control system to achieve fast and precise motion quality.
Hydraulics are the best, but they will run you into the $10.000 range only for the actuator system.

Your best bet is electric DC motor based actuators, since they are very cheap and there are a lot of drivers (h-bridges) available.
At the moment there is no ready to use DIY electronics available to control a hex platform, but you can put together more controllers for more actuators.

The problem comes when you have to control the positions of your 6 actuators in sync with each other, depending on the 6DOF position.
You can run the stewart algorithm (has to be custom tuned for your own geometry) in software, then output the calculated positions to each of your actuators.

For a quick example, here is a closed loop (positioning) DC motor controller: http://www.pololu.com/catalog/product/1393
You could use 6 of these, with 6 DC motor actuators like in your ebay link.
If you have the capabilities, you could consider building custom actuators that accept a variety of DC motors and screw shafts for flexible gearing.

After the actuators and interfacing is done, you will need a math plugin for the Profiler to take care of the geometrical calculations and make sure the actuators are not working against each other.

You mean like DC stepper motors or even servo motors like used in CNC machines?

And I didn't mean ready made DIY controllers but a PCB layout that I can etch and build. I've built lots of electronics.

http://www.drillspot.com/products/41331 ... r_Cylinder

These are some of the pneumatic cylinders I was looking at. At 75 bucks that's not horrible but the cylinder is probably going to be one of teh cheaper parts of the pneumatic setup.

You are right about the cylinders, they are the probably the most inexpensive parts of the system. In the end a pneumatic system will have a much higher price/performance ratio than an electronic system. A $2000 electric will outperform a $10000 pneumatic setup.

Servo motors used in CNC machines are DC stepper motors in most cases. You could use those with stepper controllers, no problem.
You could even use CNC controllers if you manage to interface it with X-Sim. They are in fact a motion platforms, only on a smaller scale.

I was referring to brushed DC motors such as wiper motors, or basically any other...
If you look in the hardware section you will find the AMC and dMSHB if you want to make it on your own.

In my opinion it is not worth the time and money to build, for a little bit more you can get a professionally designed and manufactured full PID controller with current limiting, thermal / voltage / reverse battery protection, easy to use software package along with support and warranty.

Wiper motors are strong enough to lift several hundred pounds?

Are you talking about doing it like this?

http://www.youtube.com/watch?v=MxXEOcYY1M0

Also what would a 2DOF on top of a 6DOF look like? I see this mentioned quite a few times and some people seem to think it would be the best for racing sims?

Yes, something like that. The motors don't necessarily have to drive a worm gear with a servo arm attached to it, they can also have linear movement with a ballscrew for example. Same idea as SCN5 actuators, but using a brushed DC motor with PID control instead.

You would be surprised how strong some wipers are. You have to make the mechanical design as efficient as possible. For example, you can reduce the static load (gravity pulling down even when you are not moving) with springs pushing up, and also with a wormgear/screw shaft where the gearing holds the platform even if you have the motor completely removed. I probably don't have to mention, but you also want to make top moving half as light as possible.

If you manage all the above, the only type of load you will worry about is dynamic loading (inertia), whenever you change the position. This is where power comes in the equation. More power = faster and more accurate response.

Don't forget, in stewart platform all of the actuators work at the same time. To lift it vertically (max power needed) all 6 actuators have to push up, so theoretically you can get away with 1/6th the power you would need if only 1 motor would do the work. Good practice to x2 everything.

A 2DOF on top of a 6DOF looks like just as you imagine it. Picture a regular 6DOF platform, large enough to have a 2DOF seat mover on top of it. The 2DOF part takes care of the faster effects and produce muscle tension, the 6DOF will produce the actual g-forces on your body along with pitch/yaw/roll angular movements. Haven't seen it built yet so I cannot really show you.

So like the standard 2DOF designs where it has 2 actuators on the back of the seat and the seat pivots some? Or like the joyrider?

I'm not sure how you'd incorporate support springs into something like a stewart platform. I've seen a few 3DOF where they had a spring in the center but that seems like it would limit movement a bit. The DC motor route (or even AC as I've seen a few people do) seems like a good option. I'm not sure what kind of wiper motors I should be looking for? Or if there is another options that's not too much more expensive. I know plenty of places to get stepper motors and servo motors for CNC style stuff though. No idea how I'd get the controllers for a CNC machine to work with X-Sim.... I know they are somewhat a midi based design so not sure how X-Sim works exactly.

Of course, I said seat mover. A joyrider would be a larger structure than the stewart platform itself. Besides, why would anyone simulate pitch and yaw twice?!

In case of a servo arm, a separate spring shock under each leg would do the job. I have seen bungee cords used for a similar mechanical pre-load as well.
In case of a linear actuator, you can just put springs on each leg in the exact same manner as a spring shock.

It is very hard to say what kind of wiper motor... and you won't find it out until you try it. I say wiper motors because they are easy to find at a junkyard for very cheap. Problem is, you never really know what you are getting until you try it. I picked up 2 exact same motors before, but their speed is very different. You will need 6 of the same motor, preferably new. Search for worm gear DC motor.

I'm not sure how you would interface stepper motors, but you can try to find a stepper controller/driver that takes serial commands on a COM port.

http://www.northerntool.com/shop/tools/ ... sku=161795

Something like this? But cheaper I hope...lol I'm hoping to find something around 100 a piece.

Is there a particular model/year car that a wiper motor is good on? Maybe I can just find 6 of those?

Can't help you there. Even if I knew all specs of your desired platform, there are just too many variables and very limited information about individual wiper motors. And if I gave you an exact model number, you would probably have a hard time finding the same one.

The type of wiper motor will greatly depend on:
-Voltage and current capabilities of the driver. ex: high voltage/low current, or low voltage/high current? High turn motors need high voltages, low turn motors will draw more current.
-Speed of the platform, movement range, torque. You can trade off any of these for an other by changing the gearing and servo arm.

The best advice I can give you is to start off with only 1 actuator and do some testing with weights. Most wiper motors are 12V, and draw about 20A peak under moderate loading.

The pololu controller I have been recommending will handle up to 16V and 30Apk, for $100 a piece for the complete controller/driver and great software, you cannot go wrong. They can be interfaced to X-Sim fairly easily, and soon we will have a plugin when sirnoname receives some samples. You can pick up wipers from the junkyard anywhere from $10-$50 and see how they react.

I would also advise you to look around the forum and see what others have done. You will find a lot of information and great ideas.

Yeah I've been looking around the forum. Not a ton of 6DOF machines though. Most of them appear to use hydraulic actuators, or electric actuators. A couple use DC motors and one uses an AC motor setup.

AC motors are very nice and powerful, but cost a lot for a whole setup.

This is what you have to do to keep the peace in the family:
-Build a simple 2DOF seat mover and get familiar with the concept and gather experience.
-Tell your wife it is not done, because you will need the 6DOF underneath, which is true.
-By that time you will have a pretty good idea of what actuators to use, and perhaps more hardware/plugin options will be available for you.
-In the end you will have an 8DOF simulator that you can use put together, or have 2 separate platforms.

Yeah going to have to research some more. I won't be getting this project finished for a while and won't really start buying parts until feb or so of next year. I think DC worm gear motors or something like that might be better in the long run and easier to find without buying a bunch of them to test. Even if I found a good wiper motor from a car there is no guarantee that it would be the same from the same make and year car, companies tend to switch brands all the time.

What kind of speed is needed? I see some actuators with 200lb's of force and speed of around .94 a sec? I'm not sure if these would work well but they aren't too expensive?

http://progressiveautomations.com/linea ... od_70.html

http://progressiveautomations.com/linea ... od_71.html

24 of travel is the same cost? I'm thinking these may not be strong enough though?

Stroke Size: 24
Load Capacity: 200 lbs
Speed: 0.94”/sec (200 lbs)
Type of duty: 10%

They don't give enough information. 1/sec is very slow, but that is when you have 200lb on it. What about only 100lb? We don't know.
And you don't want anything with 10% duty... Unless you use it in a Barbi doll house of course.

As I said before, It is going to be very hard to spec out an actuator without trying it and without knowing the platform's specs.

I would first design the mechanics and geometry. You don't want the platform to be compromised in the long term because of a bad decision on actuators.
Don't forget, the actuators you will have to change sooner or later anyways...