Advanced Search

Works here.
Can you access google or youtube or www.linuxcnc.org ?

Nilres wrote:

All Stahlwerk Cutters have a build in THC Connector.

Black ones, maybe, the white ones not all for sure.
They are good and have a few years warranty, but aim for 70A at least, the exact one that Hakan proposed.

Hi everyone,I’m currently using Probe Basic UI on a 3-axis milling machine. Tool setting and probing are working perfectly — kudos to everyone who made that process so smooth.However, I ran into a potential issue. I noticed that the .ngc file generated from Fusion360 only prompts for a tool change (e.g.) without performing any actual tool change or probing routine. That’s risky — I almost started a job without realizing it wouldn’t automatically run the tool-length probing cycle after each change.I’d like to make the workflow safer and more automated. Ideally:

• Whenever a tool change (

  M6

  ) is called, the machine should automatically go to the tool probe and measure the tool length before continuing.
• No manual pause/resume or separate probe command should be required — it should just happen as part of the

  M6

  sequence.

Would this be best handled by:

• A custom M6 remap in the HAL/INI file, or
• A modified Fusion360 post processor (.cps) that inserts probing commands automatically after each tool change?

Also, I noticed another small issue:
The spindle starts cutting immediately after thecommand, without waiting for it to reach full speed. I know I can tweak ramp or lead-in settings in Fusion, but I’d prefer a permanent spindle-speed wait in the HAL or remap (something like a delay or “wait-for-at-speed” check).For testing, I’ve been running with the spindle disabled and using a pencil in place of the endmill — so nothing dangerous, just making sure all the logic works before real cutting.Any guidance or examples of how others have integrated automatic tool probing and spindle speed waits into Probe Basic or LinuxCNC would be much appreciated!Thanks in advance

I started looking into to this and while I haven't been able to pinpoint the cause yet I found these lines of code in vtk_backplot.py

 

The lines highlighted red and yellow seem to be duplicates of each other and the ones marked green don't seem to be used anywhere.


Digging a bit further it seems that the path_offset_start_point dictionary where the KeyError comes from only gets written to in one location (marked orange below).

 

So I assume that the dictonary is supposed to be populated with values via the VTKCanon class.

 

That'd be amazing. I'll be putting some more time into the setup tomorrow so I'll get in touch once I'm back at the machine.

Hello,
yesterday i installed a fresh Linux and Linuxcnc. Today i received a handwheel for my mill. During the installation prozess i have to use the terminal window with the command:git clone github.com/LinuxCNC/linuxcnc linuxcnc-dev I get this error:
Cloning into 'linuxcnc-dev'...
fatal: unable to access 'github.com/LinuxCNC/linuxcnc/': Failed to connect to github.com port 443 after 3 ms: Couldn't connect to server

 I can´t reach many https sites. Any ideas? I don´t have a proxy and (probably) no firewall

Thanks

Thanks, i installed a new linux

I upgraded to the development version and did the required config changes and copied over the new routines etc.

However the bug is still there. The stack trace has changed to another line but it seems to be the same error:

 

[qtpyvcp.widgets.input_widgets.gcode_text_edit][INFO]  File encoding: ascii (gcode_text_edit.py:538)
[qtpyvcp.widgets.display_widgets.vtk_backplot.vtk_backplot][INFO] 

---

Draw time 1.025238275527954 seconds --- (vtk_backplot.py:831)
[qtpyvcp.app.launcher][CRITICAL]  Unhandled exception in vtk_backplot.py line 917 (launcher.py:37)
Traceback (most recent call last):
  File "/usr/lib/python3/dist-packages/qtpyvcp/widgets/display_widgets/vtk_backplot/vtk_backplot.py", line 917, in load_program
    point_01_pos = self.path_offset_start_point[prev_wcs_index]
                   ~~~~~~~~~~~~~~~~~~~~~~~~~~~~^^^^^^^^^^^^^^^^
KeyError: 1

This is the setup I'm probing.

 

I've tried probing the parts in different order etc but as soon as G56 finishes probing the error pop up. Now in the development version pressing the stop button after the error has appeared doesn't do anything instead the GUI mostly frozen for any user input however it still updates if I use my physical button for e-stop etc.

Could there be a typo somewhere relating to G56?

The THC voltage is raw cutting voltage. a THCAD can use this with the right scaling resistors. I would expect there are some resistors between these terminals and the cutting terminals to avoid lethal current! Check with a multimeter.

here the thcad manual is your friend. 200 V full scale will be best but 300v is also Ok

Its here, just add a negative sign, minus-sign, like below
 refClockSyncCycles="-5">
 

 Is synchronisation something that has to be set up on the PI or is it configured within the .xml file ? 
 

I can assist you remotely

Show ethercat-conf.xml, please.
You probably need to add synchronization with DC clock.

Current Setup:

• Raspberry Pi 5
  • 3 × Leadshine EL8 Servo Drives and 400W Motors
    
    Problem:
    I’ve recently set up LinuxCNC on a Raspberry Pi 5 and configured it to use the Probe Basic user interface.I’m experiencing an issue where the servo drives occasionally disconnect. The main symptom is that the motors sometimes make a “clonking” sound while running, as if there’s a mechanical fault and judder under movement . Power-cycling the system (turning it off and back on) sometimes clears the issue, and the motors then run normally.
    Could this problem be caused by poor latency on the Raspberry Pi 5, or possibly by the EtherCAT network going out of sync?Any advice or suggestions on how to resolve this issue would be greatly appreciated.