fast and accurate motion sim, AC or DC ?, which controller and h-bridge

Sorry this is so late but the kangaroo can only work with the sabertooth or syren motor controllers from DE. The kangaroo communicates with the sabertooth using serial commands that are proprietary. And actually the Kangaroo has the PID logic on it. The sabertooth is just an h-bridge with a micro controller on it allowing different methods of access to control the h-bridge.

I like the design...

You may be interested in these pages - similar designs using home made linear actuators with big wormgears:

http://bffsimulation.com/linear-act.php

@RufusDufus ,I am in the process of building a pair of them currently when time allows. Mine will be in industrial plastic as aluminium is to expensive.
Ian specifies his design in wood and ideally not to be constantly loaded long term, obviously stronger material would allow more load.
Short story long, it is feasible to use these in a 301 sort of design.

No Kangaroo is specially designed for the Sabertooth and Sysgen family.
Nice design , right placement of motors and not using extreme angles , 12v wont be an issue, if using ballscrews, as very little torque is required compared to sim running on levers. The beauty of ballscrews. Another note how does this design not fall over, on a roll it will try and push side ways , the actuators need to lean in to support each other ie push against each other plus a lever or spring tension is needed to keep them pulling towards the cg of the sim , or it becomes unstable.
But you have a really nice idea.
A couple of things to consider should be a wild ride, for sure.
I like.

The actuator on the back can't lean to the sides, it can only lean forward and backwards. The left and right actuator can only move left and right and not forward and backwards. So the actuators limit themselves in the possible movements. It's the same principle like in the force dynamics design. Of course the structure of the actuators has to be strong enough and there have to be strong angle-bearings which can hold strong axial/pushing movements in the foot of every actuator.

I wanted to build it as low as possible that's why I chose this actuator design.
I checked the lean angles in sketchup and I think I have to choose the telescope actuator-design because, like youbuddy said, when tilting the actuators lean towards the driver and the seat leans towards the actuator on the oppsite side. I thought that this effect wouldn't be so strong with 20 degree platform angle. I could move the actuators more to the outside but it that would make everything much bigger and more expensive.

@rufus: thanks. I already have an idea in my mind of how I can make the telescopic actuator with a ballscrew and a linear bearing.

The force of the inertia will cause the internal frame to twist this from my experience wont work any other way, Its how a 301 is designed.
But welcome to try ignore the help.
Good Luck
We are only trying to help, alot of us have built many different styles of sims, myself in the past 4 year has been up to 11 sims not one was the same style.
Please take care.

Good, please share what you intend to build

Here is my contribution: please find the caracteristics I check when I'm looking for linear actuators on the web (I do that regularly, ha; ha !)
- dynamic load
- stroke
these two caracteristics are binded as a weaker actuator can be took away from pivot but will require more stroke to give same angulation to the sim.
- speed under load (it's impressive how quickly the speed reduces under load)
- duty cycle (our sim are continuously moving: it's important that atuator doesn't overheat)
- static load and irreversibility: it's much better that the actuator doesn't move when supply is cut off.
- sensor (encoder or potentiometer). If nothing, you will have to add a pot somewhere on the sim's structure to know its position.
- voltage and current (for Hbridge and PSU pairing)
do you wish to add something?

And I found this nice and simple locomotive's design (here is a picture: drag and drop ):

Sorry, maybe my answers sounds like I'm a smartass sometimes. I don't understand exactly what you mean. I know that the frex-style setup needs some struntting because the platform wants to twist but I can't see where it is needed with the force dynamics design. I also viewed some videos of the force dynamics. Here is a picture of what I think:

The middle pole can't tilt to the sides, so to left and right pole will tilt a little bit, because if the seat tilts, the distance between the left and right poles is getting smaller (in this picture they nearly don't tilt at all compared to my first design). In my opinion there can't be more side movement because the side poles, which are connected to the middle pole, would have to bend the middle pole.

This is my design:

The ballnut is connected to a hollow shaft, which will be later attached to the moving platform. The ballscrew is inside of the hollow shaft.
There is a lathe neccessary to turn the bearing housing, the ballnut-hollowshaft connector and a reduction socket which allows the linear bearing to sit tight in the outer tube. Some details aren't 100% accurate as I'm a bit lazy with sketchup.

The charecteristics are up to the parts you use. If you use 10mm ballscrew-pitch it's faster but it isn't as irreversible as with 5mm pitch. 5mm pitch is slower on the other hand.
You can have pretty long stroke if you use 1 meter long ballscrews.
The rest is up to the motor, which will maybe a ZYT90 @ 24V.

thats a nice setup , but it is still a lever on a motor pushing a rod end, so not an actuator, looks like one, but not one, just a smaller version of what most of us do with the rod arm setup now.
And yes @RacingMat , you have covered the right details to consider.
static load an irreversibility, just leave the power running till you get off the sim, them power it off.
Duty cycle, dont go less than 100%
For feedback on an actuator , heres what I did on my converted brick spreader actuators " Feedback was a tricky one, but a combination of gears from a hobby shop and some leverage allowed me to use almost one full turn of a single turn pot. The extra lever on the large gear was to reduce overall turn, as it was about 380 deg, too little for a multiturn, but too much for a single turn."
http://www.xsimulator.net/community/attachments/2014-01-21-11-33-37-jpg.11627/
and if you watch the video you can sort of see how it works
Heres the photo I was looking for

Nice work.
The feedback is very interesting for me, too. I also like this solution, it could be directly connected to the ballscrew:
http://www.x-sim.de/forum/viewtopic.php?t=368

You are right. I simulated it in Sketchup and it hepls very much to see problems. It gets better if the distance between the head-section is a bit bigger than the distance of the lower section of the poles, like a V-shape but I think I better put it on top of the poles.

Thanks yes that's what I was trying to point out without spending time doing a sketch.
As it rolls it will pull the actuators from left to right and will cause the frame to twist if they don't lean. Another design I helped with had a slip joint on the connection of the actuator to the frame allowing for this. Making it feasable to lock the actuators in a solid position making the whole rig nice and stable. Not sure if he ever finished the wooden 3DOF sim he was building, sorry wont link to it as its on the other site, don't want them earning clicks of our postings. I don't support their site at all.

Thats a nice idea but you would need a 100 turn pot or some form of circuit on the end of a 360 deg pot to compensate. Like a volume control, these days you can turn the knob forever and a circuit handles the actual volume. You could do the same I suppose for the ballscrew.
Heres what I did with my actuators, it looks a bit rough but works very well.




I made a lever setup so it pushes a gear on the bracket to turn a pot. It turns about 250 deg, so the feedback is pretty well spot on.
And thanks to @eaorobbie , I use these hall effects on it http://au.element14.com/bi-technolo...ll-0-2v-20v-to-10v-pin/dp/2319662?Ntt=2319662 , considering I had elcheapo pots originally, the difference is night and day.
Also on the back of a motor shaft, I set up in this case a single turn 360 deg pot so if the lever goes over, the pot doesnt break. Granted the pots cost $30 each, but the feedback should be very good

Thanks, maybe I'll do it this way.

I thought about my design and changed it to this:

I will use a 25mm ballscrew with 5mm pitch. This would be 250 mm/s @ max. upm of 3000.
The motors I consider are zyt90-155 or maybe zyt90-199 (<- click to see specifications).
I'm not sure if the sabertooth 2x25 can handle the stronger zyt? Maybe somebody can give me a estimation to this?

I guess you mean this desgin?:

I wanted to make my poles leaning with no slip joints. After I tried the platform angle of 20° I found out that the actuators barely lean, it is maybe 3°. It's nearly not visible in the picture above. Now I'm thinking about using slip joints as they only would have a travel of maybe 20mm to both sides and as you said it would make the whole thing more stable.

I now use hall effects pots on my feedback setup for the actuators , but I still need my lever system.
I had though of a pot directly on the bottom of the shaft, but it would have to be 100 turns at least, so there went that idea.

That's looks nice!