Advanced Search

Which GUI are you using? I seem to recall that Axis needs to be on the manual tab for jogging to work.

Can you share your HAL file?

Crumbs I was sure that the angle used in M19, was the angle after an index pulse, but can see now its after the enable edge. Guess I cannot get the index to trigger the enable. Pulling oneself up by the bootlaces I expect.
I've been wondering whey I never really followed the logic behind how those on the forum have got tool changing to work.
So what is orient intended to do as I cannot picture what it can do if not related to a fixed point on the spindle.
A index relatable orient sounds like what I'd like, but not really got a clue how to start looking at that. Perhaps a bolt on python bit, a language I'm completely unfamiliar with. I'll do some research.
Thanks Andy.

1) Why Bookworm and not Trixie?
2) Why not just install from the ISO image?

It looks like the automatic updater has not updated the manual page to reflect anything since 2.9.4.

You can see what the page is meant to currently say here:
github.com/LinuxCNC/linuxcnc/blob/master...etting-linuxcnc.adoc

Thanks a lot.
I am attaching my ldscript which I made based on existing ldscripts included in repository. And attaching my platformio.ini.

It sounds like the servo amps are being enabled without the PIDs being enabled.

Start LinuxCNC from the command-line and make sure that there are no error messages in there.

I don't think that Orient uses index.
(It probably should, and adding spindles to the homing sequence is something that I meant to do and seem to have forgotten).
If you were to set the spindle-index-enable in the HAL file then the encoder would be aligned to index as soon as it does a revolution in spindle mode.
The "orient" component exists specifically to account for the many thousands of accumulated turns accumulated in the spindle encoder counter.
If you wanted to re-index on every orient then one way would be to create a modified version of "orient" that outputs an index-enable and that also commands an initial position command morfe than one turn from the starting postition until the index reset is detected. 

Hello,
I've been trying unsuccessfully all weekend to get the latest Debian Bookworm with the latest Linuxcnc preempt-rt up and running. Since I have little experience with Linux, I'm reaching my limits. 
I downloaded and installed Debian Bookworm 12.13. Now I wanted to follow these instructions linuxcnc.org/docs/html/getting-started/g...th_preempt_rt_kernel , point 7 (with preempt-RT kernel). Would that be correct so far? 
At point 3, Open Applications Menu > System > Synaptic Package Manager, search for linux-image, right-click on the original non-rt, and select Mark for Complete Removal, I don't know which original non-rt I have to select. I have attached a screenshot.
Otherwise, please attach Linux terminal commands so I can quickly achieve my goal.

Greetings from Germany and many thanks.

Translated with DeepL.com (free version)
 

Hello everyone, we've now integrated several pushbuttons and switches. The signals are being received and are functioning correctly. However, a new problem has arisen: as soon as I switch on the machine, the X and Y axes start moving immediately, and Linux CNC goes into error mode. I don't recall ever having this problem before. Nothing has been changed in the wiring. Perhaps someone has a crucial suggestion.

You might try increasing the interval (it's in microseconds so 1000 is just one ms)
And that may be too fast for the Omron

I will look into getting a MX2 to try to have some way of duplicating the issue.
The GS10 setup was straightforword
 

OK 
I have given up on making it work after messing up several versions of my hals and inis
Instead i now want to make new ones by using mesaCT like i did from the beginning (the way that so far let me get the most successfull result)

I get an "OK" when trying to test the configuration
But when building the configuration i get this message from meseCT:

File "../../usr/lib/python3/dist-packages/libmesact/buildconfig.py", line 61, in build
    parent.updateini.update(parent, iniFile)
  File "../../usr/lib/python3/dist-packages/libmesact/updateini.py", line 135, in update
    self.update_key(item[0], item[1], item[2])
  File "../../usr/lib/python3/dist-packages/libmesact/updateini.py", line 765, in update_key
    start = self.sections[f'[{section}]'][0]
            ~~~~~~~~~~~~~^^^^^^^^^^^^^^^^

KeyError
'[MESA]'

First download the deb and reinstall
to be sure you have the latest build date

Then please file an issue at
github.com/jethornton/mesact/issues

 

Hello everyone. Who has ever worked with socket raw on linux cnc? raw socket requires a set cap, does it break Linux Cnc? I wrote a test driver, on raw socket, receiving data via mmap. As a controller, I use STM32F7 (4 PWM encoders). Maybe I'm doing something wrong, I don't understand the specifics of Mesa, but the first tests show some kind of prohibitive bandwidth, the jitter is lower than on mesa.

Ill try the other release and see how it shakes out. Its killing me because I see the T470/T480 in use in that thread and this is just the newer version of that laptop. Also I figured I was safe because this one can be bought from Lenovo with Debian preinstalled.

Sorry to clarify the bios is not password locked its just extremely limited by the manufacturer. Locked down in terms of restricted to basics.

ip a:

 

It seems that the connection is ok
mesact finds and recognizes board 
ip a return
BROADCAST, MULTICAST, UP, LOWER_UP at 10.10.10.10

@andypugh I relocated the net commands after the motmod .......

IT WORKS!

I cannot tell you how grateful I am -- I've been going round in circles for so long and now cannot stop smiling.