I have an ongoing issue where if I adjust any line on my AR4 software, the nano board quits talkin. I have a servo gripper and a sensor running on the nano. In windows device manager the port is still showing. I can open Arduino IDE and talk to the nano, but to get AR4 to talk again I have to reboot windows or unplug and re-plug the usb to the nano.

I have had this issue for a while, but I'm now implementing the robot and having to make small adjustments. A simple 1 min adjustment is taking 15 mins.

I have the latest version on both the teensy and nano. Have replaced both teensy and nano, all usb cables (using good cables). Using AR4 6.3

Any help here will be greatly appreciated.

Hello!

I'm following the ROS2 setup tutorial for AR4 but run into issue at step 7. Build the Workspace.

I'm running Ubuntu 24.04 on VMWare. The previous steps were successful.
However, at step 7, the terminal shows stderr: ar4_moveit_cpp_demo
CMake Error: File /home/.../example_files/package.xml does not exist
Summary showed
Failed ar4_moveit_cpp_demo package, exited with code 1
Aborted annin_ar4driver

Please let me know if there was something I could do to address this.
Thank you!

I've uploaded software version 6.6 beta. I've made some improvements to the kinematic solver to help get through wrist singularity more cleanly and I've updated the wrist configuration to solve issues with J4 or J6 wrapping around during movements. Give it a try and let me know if you spot any issues. Don't forget to upload both the 6.6 sketch file and the 6.6 HMI program.

Thanks,
Chris

Hi Chris and the community,

I am using an AR4 (Teensy 4.1, Software v6.3) for a high-speed pick-and-place application. My current cycle time is limited by the physical speed of the robot, so I need to drastically increase the speed of specific axes.

Application Context (Payload): The target workpieces are very lightweight, ranging from 1g to 12g. Due to this low payload, I believe the motors have enough torque headroom to handle much higher speeds and accelerations than the default settings.

My Goal:

1. J1 (Base): Increase speed by 2x (Steps: 800 -> 400).
2. J6 (Wrist): Increase speed by 4x (Steps: 800 -> 400 + Software Speed Increase).

The Issue (Implementation Block): I attempted to change the J6 microstepping from 800 to 400 (SW4=ON) and updated J6 Step/Deg to 22.222 in the AR4 software (Kinematics tab). However, immediately after saving and calibrating, I encountered "Collision/Motor Errors" even when jogging unrelated axes like J5.

My Diagnosis: I noticed that float J6StepDeg = 44.444; is hardcoded in the "AR4_teensy41_sketch_v6.3.ino" file. I suspect the collision detection logic in the firmware is using this hardcoded value, causing a mismatch with the new encoder feedback ratio.

Questions:

1. Procedure: To successfully change the gear ratio/microstepping, is it required to manually update the J6StepDeg value in the .ino sketch and re-upload the firmware to the Teensy?
2. Feasibility: Given the lightweight payload (1g - 12g), is a 4x speed increase for J6 physically realistic for the standard NEMA 14 motor? Or are there other limiting factors I should be aware of?

Any advice on properly configuring the robot for higher speeds would be appreciated.

I was soldering J4 brown wire to the cat5 and gave it a gentle tug to make sure the connection was tight after I heated the heatshrink, the brown wire pulled out of the crimp terminal without the crimped end on it. Since it is microscopic I failed to repair it, need a new pigtail to fix this.

I can't find any details on exactly what the connector is, it seems like a JST SH 9 pin with 1mm spacing, and 1 pin unused. Can anyone confirm?

I am assemling the AR4 arm I just got. I do not have a soldering kit, heat gun and tubing. Looking for recommendations of what worked for others. The ones at my local store seem a bit overkill.

Thank you!

Just posting my experience with j1/j2 since I only found 1 other instance of this in the search.

J1 key on the motor was not fitting in the keyway even though it appeared to be all the way in. Not recommending this, but I wrapped it in wire harness tape to protect the shaft, then squeezed the key in with needle nose pliers evenly. it went in fine after that

Fast forward to J2, I ran into the same issue.. this motor is much bigger, and it looked like it was all the way in.. problem is you don't really know, but there's a 90% chance if it's not going in the keyway that the key is just not pressed in far enough.

I tried using a clamp, did not work. tried freezing it, tried tapping it in with the side of my pliers, ultimately I had to go buy digital calipers and confirmed like another poster that it was sticking out 1mm too far. I then bought a vise, wrapped it in electrical tape as the guide mentioned, and pressed it in pretty hard. after this it went in fine.

Unfortunately I was so focused on trying to get the motor in that I installed it slightly off to where the screw holes don't align, and now need to find a way to tap the motor back out and reinstall it. So word of warning be careful on that step!

Edit: Got the motor back out by having my son pull/wiggle the motor as I tapped in a screwdriver that had a socket slightly smaller than the diameter of the motor shaft, I wrapped it in electrical tape then tapped it with a rubber mallet. Took a lot of force, but it seems fine afterwards.

There is a phenomenon where coordinates disappear. In what cases do coordinates disappear? The operation is normal, but the coordinates disappear.