Yobuddy's 6DOF seat concept

Re: 1000Watt DC motors for platforms?

@yobuddy
nice to read something from you again

Thats a really great idea imho! Are you familiar with this CAD application? Would it be possible for you to draw the entire platform? I really like that design, and it would be great to have it as a completed CAD file. I'd really like to see your CAD construction in our CAD section.

Theres a question left, that Evan mentioned. The freewheel issue. I cannot say if the motors will hold the platform wheigt even with 60:1 reduction ratio and alloted wheigt to the 6 motors by itself without current. That means it should have to be ensured perhaps, that either the motors get current continiously to hold position, or the controller has to go into braking mode.
A wormgear drive is the better but more expensive way.
Would like to know Evans opinion.

Regards
Christian

Re: 1000Watt DC motors for platforms?

Hi all,

Yes I agree that is a start to a good simple design - even I should be able to weld those corners up! :happy:
I am not sure if belts are the way to go for this design though unless they are toothed.

Freewheel would most certainly be an issue and if gearheads were to be used electromechanical braking would probably need to be addressed. Fortunately these are supplied with motors salvaged from wheelchairs and mobility scooters, but they are slow to release braking.

I have not had much joy locking my motors to a static point using PWM control without unwelcome jitter. I don't think even with higher resolution angle encoders I could remove all jitter to a smooth ripple. The mechanics of the rotor and brushes means the motor must move slightly to switch to the next coil. You would need amazing feedback to balance current on one coil and then also cater for different weights too... way beyond my DIY skills, and hence the assist central spring on my 3DOF platform.

I would love to use my gearheads as they are quick but yet I still perservere with my linear actuator design. Simply because they don't have a freewheel issue.
My gearheads are excellent for FPS because in FPS the movements aren't really position critical like they are in carsim or flightsim. The freewheel in FPS case actually assists in slow return washout.

EvanF

Re: 1000Watt DC motors for platforms?

@Evan

What about integrating DIY wormgears drive into the construction?

Perhaps If we could organise such threaded shafts and and suitable toothwheels.

Though I dont have a clue either how much such parts cost nor if they are standard parts.

Regards
Christian

Re: 1000Watt DC motors for platforms?

@yobuddy
Permanent magnet motors still require electromagnetic force to propel them otherwise you would have perpetual motion! :yes:

I thought the same on my 3DOF and certainly with the spring assist there is some holding force, until a 200KG person sat on it and the spring sunk more. (A nut to manually adjust the spring tension is handy.) But the issue comes in when not all 3 cranks are of equal angle, ie: when the chair is angled one way more than the other, balance is out and weight is not evenly distributed. Without enough power to hold them the motors with more weight on them then spin backwards due to the extra weight and gravity. You have to consider power flowing back into your H-Bridges at this point due to the motors acting like generators.

@christian
Yes including worm gears in the design would be cool but would slow the impressive speed of the crank motor concept. If I were to go down this track and make a 6DOF I would mod more linear actuators for it as they seem ideal for simming.

EvanF

Re: 1000Watt DC motors for platforms?

Hi Evan,

I will not mention here again my thoughts about PID driving of motors, but I can tell that the feedback is certainly takes a part in this too. If the position feedback is too low, then the steps in the positioning will produce jitter effects! When I drive my motors I leave a small area in the middle position to be used as brake area in witch I short together both motors leads to have it hold it self into position. This brake area is small around 10 values (in a 0-1023 range). But the most important is that the feedback resolution is 4 times bigger than the motor control resolution to avoid as much the stepping of the motor!!!

I wish I could explain it better!

Regards, Thanos

Re: 1000Watt DC motors for platforms?

Hi Yobuddy,

Hehe, Nice concept! Now its a 6DOF force dynamics platform!! :highfive:

It can work, but I don't know if there is enough space for the legs of the rider

Regards, Thanos

@yobuddy

I like its simlicity much! I hope we can get rid of the freewheel issue in your concept.

Regards
Christian

The shafts could be fixed with UCF bearings to the construction. Those bearings are able to fix the shaft against linear motion as well.

Hi Thanos

Yes I think you have explained it well enough. Currently on my 3DOF the 10K pots are mounted on the output end of the 40:1 gearhead. The coupling to the pot equates to 1:1 ratio. I have intended to place the pots to the rear of the motor so the motor will have to spin 40x before it equates to 1 revolution on the gearhead output. This should give me enough 'sensitive' feedback don't you think? But it also means more fiddling around with reduction ratios to connect the 270 degree single turn Pot. I think I will have to get free turning optical quadrature encoders for this application. Do you think I would remove jitter this way?

By going down this path do you think we will have solved the freewheel jitter issue concerning reduction gearheads?

But can your current AVR board cater for 3 optical encoders? Perhaps I will need to try the gearheads with my 2DOF, but I am currently working on linear actuators with that. I wish I had more time to try all concepts at once!!

EvanF
EvanF

Hi Yobuddy,

I don't think that such open worm heads will work. At least not for long before they start to wear as they will lack lubrication!!!

As I said to Evan, I will develop an analog decoder for optical encoders. Its easy to add an ADC input on those to calibrate the encoder in center position when its powered up. But Evan's idea to use limit switches for the calibration is also attractive idea...

Keep up this great CAD drawings! They are cool!

Regards, Thanos

Hi Evan,

I think that a 40:1 gear down will remove much of the jitter of your gearheads typemotors.

My current AVR board has only ADC inputs. But thanks to your idea for an analog Decoder I'll probably have it work with optical encoders very soon!!! I know Its not easy to work in more that one concept at time but you could just use a demo construction with a dummy load to test it if it keeps producing jitter effects.

Regards, Thanos

Hi Thanos,
In a previous video, possibly as far back as 16 months ago now!!, I showed a demo of an H-Bridge brake using something a bit like
if < 254 then pin 1 high pin 2 low if > 256 then pin 1 low pin 2 high (in picaxe code) and on the test bench it worked well enough to stop the motor solid without jitter, but in practice with more body weight being used, instead of just my hand trying to move it, the jitter appeared. Mind you the pots are 1:1 with the crank and I allowed a 1 point difference 254/256 between. I will try again with 40:1 and try 10 increments like you suggested, but will need to use an optical encoder instead to avoid the 40x multiple turn issue. Thus a digital/analogue convertor chipset would be cool thing to have handy.
EvanF

Hi Yobuddy,

No your thinking is wrong. Same applies to the schematic you have draw...

Just take for example the simple pitch - roll motions... To produce a roll motion all motors must cooperate together to lean on one side without resulting changing the pitch of the platform. Same thing happens to the pitch... When you want to pitch forward, all motors must cooperate to produce the pitch action without resulting in unwanted roll motion...!

Same applies to the heave effect! All motors must also cooperate to uplift on down lift the platform while keeping it level! Now consider this... Say that you have the platform rolled to 45 degrees to one side, and you want to perform heave motion... Guess what! The half motors are in up position and other half in down position with no space left for heave motion. So in this case the only thing you can do is to limit the pitch - roll motion to, say, 25degrees and have the rest motor travel, free for the heave motion that must be done in cooperated motion of the motors to keep the original 25 degree bank!!!

The motion is getting more complex when you want to perform also yaw motion since it will also need some of motor travel reserved, so the final pitch-roll motion limitation can go down to 15-20degrees!!!!

And there are also other motions that can be performed in certain situations that can end up extending only one of the motors!!! I that case you can have strange positioning effect that can be used for example when you want to simulate the drifting of the rear end of a car!

The best you can do is to make a small (hand moved) platform to see that effect have various positions of the platform to the motors-actuators extending. This is called inverse - kinematics!

Regards, Thanos