Batty's 6DOF ODrive BLDC Build

Hi guys, been lurking for a while trying to digest as much as possible from @Peacemaker105, @Dirty, @SilentChill and decided to start building a rig. I've had luck in the past with odrive building a hoverboard FFB wheel so I decided to jump into the deep end.

I have no experience welding so I'm using aluminum extrusion and adapting @DepartedReality's actuator for the BLDC motors. I re-modeled some of the 3d printed parts and have a single actuator put together for testing. Trying to get the electronics to behave now...

Parts so far:

Turnigy 6364 190kv motors
Sequre Odesc 3.6 56v controllers / 100w 8ohm braking resistors / USB isolator / @SilentChill's firmware
AMT 102 encoders / cat8 shielded cable for connection to odrive
Cisco AA23340 Power supply - 220v input / 42v 2900w 63A output
600mm 1610 Ballscrews

I'll upload some pics but I want to get some advice on testing out the actuator.

I have @SilentChill's firmware loaded and I can move the actuator with FlyPT Mover with a single serial output and a simple direct input (Axis1a, 18 bit output). If I move the slider too fast or too often, the motor stops responding.

I have it calibrating the motor and encoder on startup.

I've monitored the input_pos and pos_setpoint and the inputs are still coming over ascii but the motor just stops spinning. If I move the motor by hand, it seems like it's trying to hold position but it will respond erratically if I push too hard. Seems like it's hit some limit but I didn't move it far enough to hit the travel limits set in the config.

Iq_measured seems to go to the limit and just stay there. I have the motor's current_lim set to 46 and the Iq_measured just sits there with no response. No errors from dump_errors(). On occasion, the odrive will hang and I have to power cycle it.

It doesn't look like the encoder is slipping but I just marked it and I'm validating by eye. Is there a better way to determine if the encoder is slipping?

In my day job, I'm a software developer so I'd love to understand the firmware we're running. Does anyone have the custom firmware source or is anyone updating the firmware to work with the latest odrive version?

Or, is my testing methodology just bad? Is using a simple direct input in FlyPT Mover an accurate way to test a single actuator?

Can anyone point me to some odrive experts to help diagnose?

Thanks!
Batty

Here's my FlyPT Mover setup:

Hi,
Welcome to odrive family
I think you must to verify all your setting into odrive .... I'm not sure, but i think your current limit is too low... That's why odrive is switching to a safe mode when there's a peak of current...
You must to learn how odrive is working and all the setting that it countains to use it correctly....

Will

Thanks for the reply Will. I tried pushing the current limits up but I didn't have any luck.

I've been over the odrive config and it's tough trying to figure out what's going on without knowing what the firmware is doing. In the latest firmware, there is no "x" ascii command so I'm not sure what the numbers mean.

It seems like it's going into some safety mode but there is no error reported by the odrive.

I'll keep researching. Thanks again!

I will propose you a huge solution for odrive soon....I will keep you and other odrive users informed ;-)

Hi Batty!
I run my 6DOF-Odrive-Sim with the official Odrive firmware. In the past, I also used the firmware that was shared by silentchill (formerly modified by insectatech), but It was kinda blackbox for me and sometimes I had some strange problems.
So I experimented with the official firmware and I can say, its running now since aprox 1 year without any problems!
But you will need a modified flyPT mover Version (I got it from FlyPT for testing - 3.6 or so ?).
When this mover version goes official, I can post a documentation about how to configure the official odrive to this mover...
Really a great combination!!
Wickie

That's awesome! I'd love to work with the latest firmware so I can see what's going on in the code. Also, it'd be nice to use the latest tools to monitor the odrives while testing.

Curious about what changes were required in flyPT. Is it to handle the rotation per second vs. the CPR for the position targets?

I've continued working on my build and I have an actuator working off the rig with the older modified firmware. I'm at a point where I'm wiring up the rest of the actuators to do some testing of all six before hooking them up to the rig.

Any chance I can help beta test the latest flyPT with your changes?

Thanks,
Batty