6DOF motion platform using ODrive

Out of interest, why do you need Hall sensors when you have encoders fitted?
Nice project, btw :smiley:

1 Like

Hi Towen, I may be wrong, “as I often am” but I think Wickie may mean “position” sensors for safety as the encoder is not absolute.

Jerry :slightly_smiling_face:

1 Like

Hi towen,
Jerry is right - I meant the magnetic Reed-Switches I use for the endstop detection :grinning:

2 Likes

I’m seriously considering going down this rabbit hole.

The only motor-control project I’ve ever done was using a floppy-drive stepper to dip/etch electrodes in a physiology lab. This is a totally different level of complexity, but it seems like there’s a ton of support in the community and by the time I get to that stage you’ll have one more working example I hope!

My metalworking/machining abilities are probably my weak point. Were any parts of the physical build especially difficult? How hard is it to cut threads, make the custom pieces for your U joints etc? I don’t even own a lathe yet so I’m basically starting from scratch. I’m not in a rush to complete and the journey is kind of the point but I wonder if I’m biting off more than I can chew.

Hi

I finished my 6 dof racesimulator using a manual lathe and milling machine, most of the parts I bought came from AliExpress. If you need links to any of the parts let me know.

Cheers

Carelsbergh

Hi Gambledog!
Welcome to my thread!!
Building a motion simulator is a great experience but also a mega challenge with (sometimes) an unsure result. If you love tinkering, solving one problem after another and pushing your possibilities to the limit you should build a simulator.
But if you are not skilled or have good mechanical support around where you are living it would be better to reduce the amount of DIY parts and maybe buy a ready actuator. You can find a lot of successful simulator builds on the www.xsimulator.net forum. Let you being inspired and rate your possibilities realistically.
My motion platform in the current design is difficult to build without a lathe, but when I am finished and everything is running perfect I maybe will make a design that is doable with 3D-printed-parts and some parts made by a workshop.
I think my universal joints are easy to make with help of a 3D-printer for the black center parts. The metal work can be done relative cheap by an web online workshop - I think I wrote something about this in my thread.
For me, the most difficult part was to make all the threads on the cylindric parts as precise as necessary. Also all cylinder-pipes have to be absolutely flat - difficult to make this without a lathe.
There are some DIY-sims on the web that are easier to make- maybe that one from Zennix - have you seen his thread?
I hope I could help you and didnt destroy your dreams too much.
warm regards
Wickie

Hi Wicke,

Thanks for getting back to me! I just found out we have a local makerspace with access to a lathe, a manual mill, PLA 3d printing and some CNC gear, plus a few people doing similar projects so there’s a lot of support. I’m also near an auto wrecker and a decent source for scrap profile/angle. I think the 2DOF on a U joint setup will be my starting point. Cheap wipers first but then progressing to something Odrive compatible to learn the system since I want to possibly use it for wheels/yokes/rudders/bigger actuators later on.

You definitely haven’t killed my dreams. It’s the learning/building process I’m excited about.

Hi all!

I have a new problem with my Reed-type endstop switches. See link:

Maybe one of the professionals has a tip?
Thank you very much!
Wickie

Edit: Wetmelon will try to fix the problem and include the endstop branch to the main branch. See: Please Help! - Problem with Reed endswitches

Hi Wickie,

Looks like you have been on a year-long adventure with this project :slight_smile: Wonderful and ambitious project indeed.

I am working on a project to motorize a wheelchair - aiming to provide a low cost solution. Can you please provide a link to your magnetic reed switches? I am thinking of using traditional limit switches but dont think they will be easy to configure within the space.

Thanks,

Anil

Hi Anil,

I am using this Reed-switches: https://www.conrad.de/de/p/hamlin-mdrr-dt-reed-kontakt-1-wechsler-175-v-dc-0-25-a-5-w-glaskolbenlaenge-14-73-mm-503642.html.

And this magnets: https://www.maqna.de/?ActionCall=WebActionArticleSearch&BranchId=0&Params%5BSearchParam%5D=4x7-N45-N&Params%5BSearchInDescription%5D=1

The distance from the magnet to the Reed switch is aprox. 8 to 10mm in my application.

The orientation of the magnet (cylinder) must be parallel to the cylinder of the reed switch - otherwise the reed switches will not switch as planned.
The reed switches are mechanically extremely sensitive, so it is always better to solder them into a small circuit board.

Wickie

Thanks, Wickie.

I think this may be a really good solution for my project.

Anil

Hi all!

after a while of silence regarding the progress of my project, here are some news…

Good news - and bad news !

First the good news:
after some really bad evenings/ nights of endless searching for bugs I finally managed to make my 6 actuators running and connecting them with the interface software “mover interface”.
So this big milestone is reached!!!
My son immediately wanted to try riding my motion platform. The speed and the strokes are set extremely slow/short at this test…so dont wonder.

Now the bad news:
I wasnt able to solve some problems that I had with the endstop/RazorsEdge firmware. So I had to switch to the main branch firmware. But this doesnt have the endstops and homing funktion, that I mandatory need for the interface software and a safe use.
The next weeks I will build the upper platform…
and…
wait for Wetmelon having included the endstops and homing funktion in the main branch firmware…

Wickie

2 Likes

PeterW alias Wickie

Hi all,
some little updade…
The upper platform is nearly finished and so we have been able to check the dimensions and the comfort…we love it :grinning:

Wickie

2 Likes

Hi

Glad to hear you again, thought something happened haha

Cheers

Carelsbergh Stijn

Hi Wickie, great job again, you must be pleased.

Jerry. :smiley:

Hi all!
Today we had a breakthrough in our long sim building journey. The last days my son and me were struggling with an error code in Odrive so that it was not possible to set for maximum actuator speed. Always one actuator was switching off. BUT today we tried a different pre setting for dirt rallye in the in the mover-interface (many thanks to FlyPT !!!) and now all problems are gone!!! Everything, the speed, the drive feeling and the precise response is just mindblowing now!! We dont know why, but the the Odrive is stable now. The force feedback from the rig is so strong, that we dont want to increase it. A crash feels sometimes as you would die the next moment. So we are super happy now and finally got the reward for our hard work over one year :D
Regarding the noise of the actuators I can report that it is no problem at all. When driving with headset and without a crash somebody could even watch TV in the same room ;). But a crash is a crash!
What I dont understand is, that in videos taken with a mobile phone cam, the noise seems to be horrible - but believe me, in reality its not loud nor annoying.
https://www.youtube.com/watch?v=tJWsRy_g0bo

4 Likes

Wow, this is so cool!!! Really glad you got this working, congrats! :grin: :metal:

Hi

I’m glad it’s working, about the noice, in my videos it’s also way louder than in real life, don’t know why.

Cheers

Carelsbergh Stijn

Looks really awesome. Finally you made it working. Congratulations

Hey that looks great! I know it was a long road to get there but I’m glad everything’s working now :smiley: