Universal 6DoF Recessed Simulator

Yes sir, after talking to the 750w AC actuator supplier and hearing about the challenges I had a look at the motor specifications. Which then made me realise that AC isn't an option, the 750W motors are not really 750W motors but even so, it's too much power for a home outlet. I've instead arranged with a motor manufacturer to manufacture 800W BLDC motors and controllers to my specifications, which I will use instead. This also means that the system can be plugged in at a typical home outlet, on the same breaker as a game pc, direct drive steering wheel etc. without tripping when the load spikes.

The second problem with linear travel actuators is inertia, the screw types systems can carry a lot of load, much more than is needed for simulation, but they also have a lot of inertia, you have a speed and acceleration limitation. Crank systems are very responsive and power efficient, but have limited travel. Typical belt drives are fast but a bit awkward and the load capability is a little bit low, also most power hungry of the 3 designs. So I looked at the servo-belt concept and designed an inverted version of that. I also found a local distributor for Tillett seats, so I've asked them to bring in a W1 when they order their next batch of kart seats, so first we'll concentrate on the single seater formula position. For more performance you can add another motor in parallel.

Anyway here is the actuator.

So for the project, the motors and drivers will arrive in mid August, belt probably next week, the alumnium tubing(Decided against carbon for now, about 10KG saving but increase of $800 over the project cost) and plates and bearing material I expect to have by end of next week as well, I haven't yet drawn the jigs for laser cutting but I'll get to that soon. I have the limit switches on back order but they should be here in a couple of weeks, backed Thanos' kickstarter for the controllers which will be ready by mid September. I received the power supply and charge controller today, I haven't bought the batteries yet but that's quick and easy in any case. The parts for the pedals should come in within the next few weeks as well, the direct drive steering motor has to wait a little bit for funds though. I have a G27 that I can use in the meantime in any case.

Project Scorpion

Is there any way to invert the actuators? It would save lots of inertial mass to move.

The piston is about 2.5kg, the lower part of the actuator is about 13kg, so inverting it would be a lot worse. placing the motors at the bottom isn't possible with this kind of belt drive system, only with a "normal" belt drive.

Yeah, many are on the stereotype of the motors being mounted on bottom, while the real heavy part is not the motor it self... great work!

The motor is fairly heavy, I think with the gearbox and brake about 6-7kg, but it doesn't move with the piston. About 30% of the actuators weight should be added to the platform for X and Y linear movement, when the motor is up high it's a little more, probably 40-50% or so, nevertheless the light piston and low rotational inertia should more than make up for it.

Got some questions, because i can't follow you
You say crank systems are power efficient? afaik the wormgears have an Efficiency of about 0,6 and ball screws about 0,9. So, they are more power efficient than what?
I also don't agree with the point of the Maximum load of a linear actuator. Look at the build of @SilentChill, he had Problems with his max. load. The Motors got too hot. After changing his build it got better. I know, you just can take stronger Motors, like you do With 0,8kW the actuators can have a low angle without any Problems...

Also if the linear actuators are to "slow" you can take a ball screw with a pitch of 10 mm and a Motor with 4500 rpm. than you get 750mm/s linear travel

And why do lin. actuators have a lot of inertia? Do you have some values? I cannot imagine, there i much difference with the belt System. But what i think, is that the belt System isn't quite responsive/direct as a ball screw actuator, like with the wheels (DD vs. belt)...

Why are your actuators round on the outside and sqaure on the inside?

Hi

Sorry I should be more clear.

When standing still, there is no back-drive on the crank system, so it doesn't require energy to stay stationary, whenever the actuators are not in use they don't draw any power, with screw and belt there is constant draw to keep the platform stable(although with the screw not really much). I've included brakes on the motors to reduce this power requirement though.

Max load on the shaft not on the motor, a 20mm screw shaft has a much higher loading capacity than a 30mm belt for instance. Long belt drive systems have damping issues, yes, however if you look at the servo belt mechanism only a very short part of the belt is in use at a time(about 20mm), so there is much less stretch compared to a full length belt system and a lot less damping.

Your theory is correct until you try and achieve that speed and find out that you need a bigger motor than you thought. With a screw you have to rotate a long metal rod, energy required is squared with the increase in speed and mass being accelerated. With the belt system you accelerate the gearbox and pulley, so if you look at the amount of material you use vs the amount of material in use, with a long screw you have to spin a lot of material at high speed compared to belt and crank. And you realize a 500w motor can barely get the screw turning to 200mm/s. I think 800W is okay on the screw, 1200W would be much nicer though. I'm not against the screw drive in any case, I just think the belt is better, screw is easier to make and cheaper as well, but nobody has done this specific belt design yet. I might end up with useless gearboxes and a healthy helping of humble pie, but no risk, no reward. Let's see

Actuators need to rotate as the platform moves, I have a sliding contact bearing in the bottom strut, so the top section is fixed to the platform and the bottom can rotate. Otherwise it binds, I could have done a square bottom with a round insert at the bottom, but square sizes are not as abundant as round.

Cannot say I agree with 'everything' you have stated here (and I did actually test lots of different drive systems before settling on whats fitted to my current sim) - but you sound like a man who likes to find out for himself, which is OK, so will just look on - all the best in your project.

I know that My question was about the geometry of the actuator. It is square on the inside (a). And round on the outside (b).


Is there a special reason fot that?

In your first design both parts were round.

Another question:

Which motor controllers will you use for your BLDC motors?

I also look forwards to seeing how your actuators work out.
I don't necessarily agree with all of you reasoning, specifically regarding the importance of the low inertia in the actuator mechanism, it's only really relevant when there is no significant load on the actuator.
What's interesting to me is the low RPM requirement of the actuator, it makes Steppers, or hybrid servos interesting options, since they provide massive amounts of low RPM torque, you'd likely still need to close the control loop or oversize the stepper.
My back of the envelope calculations says you get about 60-70N of force for every 1NM of torque (based on a roughly 30mm diameter drive pulley), steppers producing 10+NM of torque, which is I'd guess where you would want to be, put you at the upper end of the Nema-34 sized steppers, so you'd be looking at $250-$300 for motor + driver + PSU.

Yes, you need to stop it one way or another from rotating. It made sense to go this route, if the piston is square you can keep it from rotating with a square bearing, instead of having a guide to stop it from rotating as well as a round bearing structure.

Thanos controller with AEAT-6012-A06 Encoders and drive amplifiers by the manufacturer of the motors.

If you look at the inertia, assuming a platform weight of about 175KG, having a 750mm x 20mm ballscrew your motor will have the same inertia load at 500mm/s as a servo belt at 750mm/s. So the belt will accelerate 50% quicker and reach a speed 50% higher than a comparable ballscrew. Of course if your platform weight is 350KG this is no longer true so it's not about the actuator, you'll need to reduce the speed of the belt. Looking at it another way, a 360KG platform on a 500mm/s screw would load the motor the same as a 580Kg platform on a 500mm/s belt.

With the gearbox on the motors they are 25Nm rated and 75Nm peak and at the lower end of your price point.

Hi

Just thought I should show an update, so this is the new one using vive audio band and 3-in-1 cable.

Hi there

As it turns out the local importers for Tillett seats are in the karting business and it doesn't appear I'd be able to get a seat from them. Importing one directly isn't worth the cost due to the volume of the package. So I'm revisiting the idea of making a seat. I'll do this using the same technique as I use for the headset design. Just with a different surface finish. So I've modeled this (the inside) based on a F1 seat, same position etc, the outside is to match up with the simulator.

woow, i have loocking for something like this on internet, please document your process .

how would you make the carbon fiber stuff... you said in other post that you usually send the job to a company, but that company just would make the seat from your drawings, or you will 3d print the seat so they can use it as cast (or casting or base, i dont get the word )???

I'll machine the seat out of foam layers then join them and have it skinned after. With the headset I glued the layers together first then machined it in one piece after, but the seat is too big to go that route, so I'll make it in 75mm thick sheets then join them. I'm not sure if I will have it finished with carbon fiber or just with a coating and then painting and upholster, we'll see later.

in that case maybe a layer or two of fiber glass, before uphulstery may come just fine

Look to take up space, but it's beautiful. This is a very good design.

The footprint is now a bit smaller than originally thought.

We're about 35% complete.

woww this crazy project, I hate to see its finished