Simulator rig with ODrive


#1

Hi all,
Now I want to tell you something about my project. Inspired from many others in the xsimulator community I decide to build my own simulator rig. But all of the samples are build with DC motors and expensive motor drivers. That’s not what I want to use. One of my hoobies is flying rc planes and therefore I prefer to use brushless motors.
After some invesigation in the net, I found ODrive and I start to build my own rig.

The drawing of the actuator prototype:

Here is the original one:

The end switches are already in place.

tbc

Greetings Zennix


#2

Hi Zen, fantastic work :star_struck:

Yes I too am waiting for my ODrive and Motors (on backorder) to pursue a similar result, with 4 actuators belt geared down 5:1 Ball screw 20:05.

Your setup looks professional, but mine will be a little rough looking just for the test process.

With the build process with Rig and all parts I’ve no concerns; however its the code that’s been bugging me, :worried: I hope the code deliberation is easier for you.

Great work
Jerry


#3

Hi Jerry,

thank you. Seems we have the same goal.
I’m using simtools. Have a look to this: ASCII commands makes ODrive unreachable
Hopefully I can have a look to the code this week and try to find a solution for that issue.
Greetings / Zennix


#4

Hi Zennix, any further action on your actuators?

My stuff if slow to get here, and every time something arrives, I’m missing something else: hope all arrives soon :unamused:

Regards Jerry :smiley:


#5

Hi Jerry,
I killed one ballscrew nut by making a video. So I have to build the second actuator first and waiting for two new ball screws.
The response of the motor with signals from simtools is working so far. Tests are pending.
I tried to implement the end switch to the software. But no luck till now. I think that feature is very important because I saw the balls screw nut running with full speed into the hard stop :scream: only with a 150W power supply. Don`t know what happend with the final 1000W supply.

greetings Zennix


#6

Hi,
here is a little video to my progress. It shows the actuator moving by simtools. Normally the way of of the actuator is from 0 to 700.000. In this test simtools only uses max 65.000, the velocity is also limtited to 150.000 of 650.000, because my powersupply.

greetings / Zennix


#7

Hi Zen, thanks for the update.

Sorry about the ball screws :open_mouth:

I have the 20.05 ball screws and 1200w*48v PSU’s, waiting on carbon fibre, and the thing that is taking the longest, it M2.5/3/4, I think they must have been lost :worried:

At the moment I’m going blind trying (operative word trying)to produce CAD parts, but the motor mounts work enough for the test, and if all goes well “Aluminium and Carbon Fibre”
but that’s a way off.

Talk soon
Regards Jerry.


#8

Oh man that’s awesome. But you gotta fix that screw alignment! haha


#9

hello,

I’m making exactly the same thing as you are, what type of pulley are you using? I’m using sfu2005 what are you using? I will post it on here when mine is done.

Why aren’t you using a linear bearing on the top?

How much force does it have?

What voltage are you using

What is the gearing ratio?

What motor are you using? (diameter? length? link?)

Sorry about all the questions

Keep us updated!


#10

Hi Stijn,

fine, so we are fighting at the same thing.

To your questions:

  1. Pulley
    I’m using 40 Teeth GT2 pulleys,
    https://www.aliexpress.com/item/POWGE-40-Teeth-2GT-Timing-Pulley-Bore-5-6-35-8-10mm-Fit-Width-9mm-2GT/32790327676.html

  2. Linear bearing


    Is a 3D printing part.

  3. Voltage
    36Volts with 40Amps

  4. Gear ratio
    actually 1:1, 40 teeth on both sides.

  5. motor
    KEDA motor
    https://hobbyking.com/de_de/kd-53-30-high-voltage-brushless-outrunner-190kv.html

  6. ball screw
    I’m using 1605 ball screws
    https://www.aliexpress.com/item/Machined-Cost-for-ballscrew-SFU1204-SFU1605-linear-shaft-and-other-parts-extra-cost/32805398489.html?spm=a2g0s.9042311.0.0.12224c4dJ5vNAX

Hopefully we can see some pictures of your actuator design soon.

Greetings / Zennix


#11

hi,

where did you get 36v 40A?

Do you think the 3d printed part is strong enough?
I’m working on it, when there is progress I will post it

greetings


#12

Hi,

I have a 12V and a 24V PSU and put it together.

About the 3d parts, I dont know if its strong enough. But starting with it is cheap and easy.

Greetings / Zennix


#13

hi,

smps power supply you mean? I didn’t know you could use them in series

any idea on the current draw?

I have the same pulleys but I don’t know if they will slip, do you?

greetings


#14

Hi,

the power supplys has both 40Amps so it shouldn’t be a problem.
My tests without load are working with 30Volts and less than 3 Amps.

In my opinion the max current of one actuator should not be over 20 Amps. Normaly, there is no or minimum load at the actuator, because the center of the gravity sits on the rig.

To prevent the pulleys for slipping, I sanded the machined end with two flat sides in an angle of 90 degrees.

Greetings Zennix


#15

hi,

I’ve bought a DPS5020 wich is a lab bench power supply that can deliver up to 50V 20A for testing, afterwords I will buy the smps that’s best for me.

That’s right, but you can’t have the center of gravity perfect. Also even if the center of gravity is perfect you still need some power to hold it on the right height and also for turning, you can find more information on this link
https://www.researchgate.net/post/How_can_we_calculate_the_required_torque_to_rotate_1200_kg_of_mass

Also why are you using 2 limit switches? You can count the number of rotations so you only need one?

greetings


#16

HI,
thanks for the info.

Two limit switches in case of an error might be better. If something goes wrong the limit switch can do an emergency stop.

Greetings Zennix


#17

Hi,

Is it a hardware emergency limit switch or software?

Greetings stijn


#18

Hi,
this are physcal switches they can be used for both. Depending what Wetmelon will integrate in the firmware.

Greetings Zennix


#19

Hi,

Yes I know but dit you hook them up to an Arduino/raspberry pi/… or does it just cut the power to the board?

Greetings Stijn


#20

Hi,

the limit switch at the zero position is connected to a GPIO at the ODrive board. The upper position is actually open.

After initialization and finding zero position, the switches are only for emergency detection. I will try to use an GPIO of ODrive to show that initialization is finished and the control loop is running. This signal is used to connect both switches to a relais that can cut of the power.

Better ideas are wellcome.

Greetings Zennix