-
Notifications
You must be signed in to change notification settings - Fork 1
Troubleshooting
Make sure all the cables are properly connected and keep rebooting until you hear the second fan turning on - the yellow monitor light will then turn to blue.
It's most likely an ethernet configuration issue. Make sure you can ping the robot ip and that you are using the right ethernet port.
Each robot joint as a maximum limit. When exceeding it, you'll hear a click sound from the robot, lights will turn red on the control panel and you might see something like "hardware limit exceeded..." in the logs.
- Switch to T1 control mode (turn key-switch)
- Click on the "LBR_iiwa_14_R..." tab from the control panel and go to "Joint positions"
- Press the gray button half. If the axis names on the right side are highlighted in white, you can control the joint positions and move them away from the joint limits using the "-/+" buttons on the right side of the panel.
- If you cannot move manually the joints as described above, you need to re-master the axis (go to mastering and follow the instructions for mastering)
The axis need to be mastered.
- Select RKF mode (use the key switch of the control panel)
- Select the iiwa robot tab (upper left) in the main menu
- Select Mastering . Keep the enabling switch (grey) pressed while mastering an axis Sometimes, mastering fails in which case, just reboot the robot and try again
Reboot the cabinet, un-master and re-master all axes starting from axis 7.
The "position sensor not referenced" warning does not seem to prevent mastering nor to have all green lights & operate the robot.
If you are unable to remaster the axis after exceeding torque limits with the following error
And if you see something like that in the robot device status :
It means that a fuse is burnt in the low power supply (grey box inside the sunrise cabinet) . First identify which fuse is dead
- Get a multimeter and a screwdriver
- Switch off power, unplug the cabinet and open it (see p. 65 of the sunrise cabinet documentation)
- Remove the fuses (one pink F1 4A, one brown F2 7.5A, cf. p. 74 section 12.3 of the same document)
- Test them with the multimeter (continuity or ohms mode) The fuse looks like this (for instance the pink one / F1 4A)
Get a new fuse. Spare fuses are in the KUKA box on the shelf .
- Replace the fuse
- Close / replug everyhing and try to master again To order new spare fuses, use this reference .
CK_COMPOUND_RETURN_ERROR with "Wrong FRI session state in "Monitoring (Wait)" in setup phase of FRI motion"
It's a latency issue . You may also see in the DGM logs : permission denied on "dev/cpu_dma_latency"`
The fix is simply to give permission sudo chown {username} /dev/cpu_dma_latency
If you see mlockall failed: Cannot allocate memory
when running the DGM, try to run with sudo -s
.
If you see something like :
[foonathan::memory] Allocator foonathan::memory::new_allocator (at (nil)) leaked 400 bytes.
[foonathan::memory] Allocator foonathan::memory::heap_allocator (at (nil)) leaked 33384 bytes.
make sure that you have a clean ros install (see ubuntu installation scripts)
Find process PID running on that port
lsof -i :30200
Output :
COMMAND PID USER FD TYPE DEVICE SIZE/OFF NODE NAME
dgm_iiwa_ 72687 root 5u IPv4 434170 0t0 UDP *:30200
Kill it
sudo kill -9 72687
Description: The robot has hit a joint limit. It is impossible to jog the axis manually in T1 mode, nor to re-master the axes (pressing grey button on the smartPAD does not highlight the axes). Hard reboot of the sunrise cabinet and re-mastering the axes solved the issue.
All our open source software are licensed against the BSD 3-clause license.
Copyright 2016-2020, New York University and Max Planck Gesellschaft.
- Home Page
- Contribute to the wiki
- Logo
- Introduction
- Our Codebase
- Build Our Codebase
- Build tools introduction
- Build chain tutorials
- Dependencies
- Building our software stack
- Troubleshooting
- Robot Tutorials
- Programming
- ODRI Robots
- MicroDrivers
- Solo12
- Bolt
- NYUFinger
- Kuka
- Debugging