2 DOF

Hello everyone. This will be my take at building a 2dof motion simulator. Because i strive for getting the most of what i already got, i will not build a traditional sim mounted on a universal joint. Instead, i am going for a joyrider-type, which will maximise efficiency, by shifting the center of rotation as close to the center of weight as possible. Also, i plan on making the lever mounts below the seat adjustable, to increase the range of rotation for flight sims, or decrease it for racing.
For power, i use two monster moto mated to 24v dc motors. I power them at 14v and they draw about 15 amps at stall, so should be enough power. I already tested the smc3 code and the motors work. I have not mounted the pots yet.

I sketched a rough design of what i envision(not exactly to scale, final design will maximise space efficiency).

a) I do not know mechanics well, would anyone care to comment, does everything look ok, safe? The frame will be made from 60/40/3 rectangular hollow steel, bearings are 20mm inner diameter, two on each side for pitch axis, and two on the back one after another for roll axis, the bearings connect with a M20 semi-threaded bolt to the frame, and the connecting levers from motors to seat (not shown in photos) are regular helm joints with m10 threaded rod between them.

b) Also, another question regarding the placement of the lever mounts from motors to the seat...what is the difference between: 1) both levers connected to the back working together on both axes; 2)one lever at the back, and one lever on the side, working separately on each axis?

This is the type of mount i have in mind(not to scale), which is widely used on the u-joint design. Would it work in my case?

I'm not sure I understand. You mean to say that in this configuration, because there is little leverage, the torque of the motors would not be enough? Or what would be the problem in this design?

I see this type of design, where the motors are mounted on separate axis, but that means they are doing work separately on the x and y axis.
https://www.xsimulator.net/community/threads/2dof-joyrider-flight-race-sim-build.7498/

I would make an educated guess, that in the design I posted, the motors work together for any movement, and that would increase the efficiency of the whole system. And leverage can easily be increased by mounting the motors farther from the centre of rotation, and lowering the CTC lever.

Last couple of days i finally found some time to continue the project. There were some design changes: the roll axis was tilted 45 degrees and mounted in front, to make the whole frame waste less space unnecessarily. The pitch axis is mounted on the roll axis. The frame is solid as far as i can tell, and those bearings do ok with the load (100kg person+at least 50kg pitch axis and chair). There are some squeaks and rattles but i will deal with those later.
The COG when seated is very balanced: there is very little force needed to move the chair, you can move it with a finger if there are no motors with worm drive connected. This was the main reason i complicated the design, and from this point of view, the project is a success. Unfortunately, this was my first time dabbling with welding, drilling, etc, and the fit and finish reflects that. You live and learn.

front view:


side view:



bearings for pitch and roll:

It turned out that the wood boards weren't up to the task. In my short testing, one of the the motor mount simply ripped out of the wood. So, i got rid of it completely and replaced it with metal parts. The motors were mounted on some metal plates that are press fit under the support frame, and the seat was bolted directly to the pitch axis frame. Lever mounts were also bolted to the pitch axis frame. I also made some potentiometer mounts directly tied to the motor frame, and for pot contact to motor, i used heat shrink tube. After shrinking it is rigid, but weak enough that it will not rip the pot apart in case the motor somehow runs amok.

I did a short test,all is ok, except some imperfect contacts on a monster moto which causes one of the motors to not run unless i lightly move some wires on the board. Of course, this is because i used a glue gun and not proper soldering - very bad practice.

Random technical info:
-There is one monster moto per motor, and it currently runs at 9.5V from a chinese power supply (it has 9.5-13.8 range). The motors themselves are rated 24V, so they run very cool even with load, and have a lot of juice left in them. From preliminary testing, because the frame is balanced, they seem to work great even at the minimum voltage of the power supply, 9.5V. They draw about 8 amps each in locked rotor state, so that's about 80 watts maximum power draw per motor. At 24V i estimate this should increase to about 500-600 watts, if someday i will replace the monster motos with a sabertooth, they would make a great team together.
-the ctc arm is 8 cm and the distance from the mounting point to the roll axis pivot is 28 cm, and to the pitch axis pivot is 32 cm, this should give generous movement of about 15 degrees per axis. The lever angle to both pivots is almost perpendicular, maybe 10 degrees off at most, but this will be dealt with later, because i am thinking of experimenting with longer ctc anyway, and in my particular case, if i would increase the ctc lever a couple of cm, would make the applied force perfectly perpendicular.
After i redo the wiring, i will post some videos with load testing.
For now, here are some pictures of the various mounts/levers/pots.

A monster moto was dud from factory. The other one worked fine, but i ordered 2 ibt-2 just in case. The sim worked fine in lfs for a while. I increased the lever arm to 15 cm, tested a bit, the power and response was good.
But....out of nowhere, a motor broke. Specifically, the worm gearbox broke. It slips and cannot hold any notable load. The other motor also developed about 5 degrees of slop in the gearbox, so i presume it will also break.
I'm not sure if it can be repaired, or to buy just a new gearbox, it seems like too much headache.
What options for worm gear dc motors of 12 or 24 v are in europe? I would like sturdy durable ones even if they are expensive(but they still have to be worth it, cheaper than an ac motor with driver for example)

I opened the gearbox just to take a quick look. It seems that no teeth are damaged, and it is a bronze gear, not plastic. But i do not know mechanical enginnering and cannot seem to open it more than this to take a deeper look, as the ctc levers are welded on the shaft. What do you think?

I tried but it was not this.
I think i found the problem. The output shaft is slipping from the brass gearing. They are connected with 2 set screws. Maybe those are the problem, they seem a little undersized for the job.

I confirm it's the shaft slipping, the gears are fine. It also slips from one side to another if i push it firmly.

Normal position shaft:



Slipped to one side shaft:


There seems to be 2 opposed set screws, but look undersized:


I'm not sure how to proceed. I can try to tighen those set screws but i'm afraid this will be useless in the long term, considering they failed in the first place.
Is it possible to weld the shaft to the gearing? Can steel be welded to bronze?
Or make a keyhole in the bronze gear, and weld a key on the steel shaft.

Or drill through those set screws into the shaft, and insert a bolt. This seems like the easiest solution

Is there anything i'm missing and there is a better solution?

I have no idea what kind of screw is that, nothing fits


I drilled through the screw into the shaft. Broken 3 drill bits and gave up. I'l go to a machinist later and see what can be done.

I managed to repair the problem.
The shaft was held in position with two tiny set screws. There was no way to extract them, so i went to a machinist and he barely drilled a 3mm hole through the shafts of both gearboxes. Then sent me away. The hole turned out to be 3mm on one side, and about 4 the other side. I stick some m3 and m4 bolts with epoxy and started to reassemble the gearboxes. One was good, but another did not reassemble as before, because a hole was drilled off center of the shaft by about 1 cm. This means that now the worm gear is not perfectly centered on the spur gear. The motor still turns, but i can feel an increased resistance compared to the other. At least it works, so i guess it is what it is.






In the meantime i mounted 6 bass shakers under the chair.






I also tidied up the electrical part and found a better solution for the potentiometer connection to the shaft, a part rubber part metal clamp for pipes. This solution is easier to work with when making adjustments or replacing a pot, which by the way happens more often than i would want; i really should order those hall effect pots from overseas.

I have good and bad news.

The bad news is that the current project is dead. Too many problems encountered for how big, heavy and ugly it was. The pseudo-joyrider design was not worth it.

The good news is that i could reclaim most of the useful materials back, and i can reuse the other parts.

I decided to go for a simpler project: 2dof seatmover on a universal joint. Basic design will be similar to the other 2 dof compact seatmovers posted here.

Also, as i was pondering the idea wether it would make sense to put spring assists on the corners to reduce the load on the motors, i had another idea.

What if instead of spring assists, you would use inflatable bags as those used in the pneumatic g seats posted here? Put 4 inflatable bags, front, back, left, right and couple them to other 4 inflatable bags in a "anti-g suit" that you can wear? This anti-g suit would be simply worn on the torso, pretty tight fit. You would have the sensation of g force coupled to the movement of the simulator. As the simulator pitches forward when braking, it exerts pressure on the front bag, which exerts pressure on a bag on the chest. And so forth for the other bags.
What do you guys think?

Thanks for the reply and useful info.
I'l look into it, refine the idea and post more about it when there is some progress with the build.

Thanks for reply and info. Already checked simcalc, the angles and forces look OK.

Regarding the implementation of the anti g suit, as i will refine the idea, i'l post updates when i have some results.

For now, i'l start the new build. Tommorow the u joint i ordered should arrive. I could not find a classic u joint without bearing play or rust at the local scrapyard, so screw that and i ordered a new part i found on some agricultural parts website.
It's a u joint used in tractor transmission. In the description it said it's rated for 55hp and weighs 2kg, while being very compact and using m10 metric bolts as mounts. I assume it will do the job just fine. By the way, it's hard to find u joints on europe ebay, compared to american ebay. I guess it's because most european cars are fwd with economic engines,