My 6 DOF with ODrive

Hello, I want to share with you my 6 dof project, I'm verry bussy building it. I'm building this as a graduation project.

Here are some pictures:

If you have any questions please send me a message, if you have advice, send it to me ;-)

Carelsbergh Stijn

hi @Carelsbergh_Stijn welcome to the forum and have to say your actuators have the makings of being quality units , do you have access to machinery or are these parts custom made by sub contracted machinist ?
great to see another 6DOF project on the go , so like all the other followers , were looking forward to the development and progress of this version ..Lovely workmanship already and this is early days for you .. great stuff

I'm making the parts myself, I have a lathe and a milling machine.

Hi @Carelsbergh_Stijn then you Sir must have a great man cave, I have a lathe Myford 7 but needs 3 phase motor and inverter upgrade and the my CNC router for lightweight stuff ( so many bits so little time ) a Mill would be icing on the cake, But 1st I'd need to add floor space to my garden workshop man cave ( the neighbour's might object to that dammit ) Nice work and a taster of whats to come ... keep up the Good work

Nice project wish I had a lathe have you got a plan where to mount the pots yet?

I don't know the lengths yet but I can just chance them and everythings addapts to that, so no problem

thanks you and I will

I guess it's not a huge deal especially with a lathe but just seems something people tend to have difficulty with and might be worth planning earlier rather than later

Using ODrive? Curious to see the result.
This drive/motors is on my radar for some time and if it works, it's a great and simple alternative.
Be careful with the size in the feet area.
Also, you have to allow rotation on the top or bottom or inside the actuator.
In your drawing it's not visible, but with such a simple mounting, looks like there's no rotation allowed:

Hi,

I don't know if this is correct but I've not seen a 6 dof simulator with universal joints that allows roation of the actuators, but what I have seen is when using rod ends instead of universal joints, they do alow it, maybe you can tell me if this is correct?

Do you have any suggestions on what to change on my design? for example get the tops of the actuators further appart? I already knew about the feet area, It was just to give an idea, this needs to be changed. my botom radius is 750mm, is this ok?

many thanks

In my rig, I allow the top joints to rotate.
It's no the best design, in my case. It has some play: https://www.xsimulator.net/communit...shless-diy-actuators.10799/page-6#post-163059
You can do something similar or allow the actuator to rotate removing the internal guides. Rotation will be limited by the ujoints, so it still works.
Removing the internal guides makes the actuator simpler, but if you are using positioning by rotation encoder, it will introduce an error, because one rotation on the motor might not be an exact rotation, since the actuator is compensating the fact of the ujoint being "solid".
You could correct this by software, but it's not easy. You need inverse kinematics in real time for each actuator.
I would go with first solution, like I have done, unless you where using a belt system with encoder/pot to detect position.
But try to put a roller bearing or a good pivot. It's something I need to change.

I think 750mm radius is enough, I'm using less than that. I wanted it to be compact.

The actuator doesn't turn much so if I program my system to not use the last 5mm of the actuator in the case that the actuator is rotated a little bit, could this solve the problem? I don't have a problem with the disseret angles not being perfect. (my actuator can turn inside the guiding block)

For example, if you make Yaw, your joints must allow rotation.
Or you can use a solution like @JAD (https://www.xsimulator.net/community/threads/jads-6dof-with-ballscrews.11686/). Where the actuator can freelly rotate. There's no guide, the guide is the joint that doesn't rotate.
What I mean in the last post is that in this case (JAD), one rotation of the motor is not always the same travel, because you are compensating rotation on the ball screw.

Hi

as you can see in the pictures, my guide block can rotate freely (it is held onto the metal with 2 circlipses), so there is no problem of the guiding block blocking the rotation:


so the top part of the actuator is free to turn from the botom part, is there still another problem?

Kind regards

Carelsbergh Stijn

Ah, OK,

I don't see other problems.
Waiting to see it moving!

Hi,
Thank you for the quick response, so the only disadvantage of doing what I do is that the encoder reading is not always 100% correct with the real position?

Kind regards

Carelsbergh Stijn

I would recommend using track rod ends rather than universal joints they can get notchy at some angles and How's is your encoder setup any pics?

Hi,

small update on my project, all the parts for the actuators are done, now I'm starting on the frame.


Cheers

Carelsbergh Stijn

Hi

my frame is finished



And my drawings too ofcource:



Now all that's left is programming the thing, I have already asked for a DIY licence twice and I don't get a reaction, does anyone know why?

Thanks

Cheers

Carelsbergh Stijn