Advanced Search

Good day! Tell me, is a gigabit port absolutely necessary or is 100Mb enough? How important are the jitter readings in the test?

I read your post and it took a while to sink in how well running the native RaspianOS with the 6.12 kernel for what I want to accomplish.  I'm guessing that 6.12+ might be a real game changer across the board but I remember discussions about certain run-time flags being used to determine if PREEMPT_RT was part of the build and people commented that some don't match up yet.

Do you think that might be a hurdle in the short / mid  term?

Do you think someone familiar with the run-time testing code might already be working on a fix to any of those problems?

Ok, I decided to complete a complete re-installation and work from clean. However getting the following:

abcabc@192-168-1-???:~$ apt --fix-broken install
E: Could not open lock file /var/lib/dpkg/lock-frontend - open (13: Permission denied)
E: Unable to acquire the dpkg frontend lock (/var/lib/dpkg/lock-frontend), are you root?
I definitely did not put a root password in. So ideas?

Ok, sorted ran sudo apt --fix-broken install

That's awesome - as always!

I've already starting to dig through it.

Interesting, I will check carefully all the encoder connections to see if anything looks suspicious. If I find nothing I will try to swap two motors to see if the error follows the motor.
Thanks for the idea!

OK, then encoder failure, most probably.
Check the encoder connectors, they tend to cause issues on those drives, some even come with wrong encoder wiring sometimes as i read several times here on the forum.
It can also be from interference, but that would also affect other drives/motors.

I think that there is actually a warning about this in the docs.
(Actually it warns of something slightly different, linuxcnc.org/docs/stable/html/gcode/g-code.html#gcode:g76 )

At the beginning of the pass the Z axis is stationary. It sees the index pulse, and then has to "chase" the spindle-synch point.
If the axis acceleration is limited, then I can see that it might have trouble catching up.

What is your Z accel limit? Given that, what distance does the spondle need to catch up to a 100rpm spindle? (v^2 = 2as )

I am running linuxcnc 2.9.3 on a relatively recent pc that i added a diyeeni PCIe parallel port. I am using a nema34 closed loop stepper motor with an HBS860H drive on the Y axis. The breakout board and cable came from amazon nothing crazy there. If you need other more specific information let me know and thanks for the reply.

I can reproduce the problem.

After some fiddling about to persist the scarkins.ver file (which is normally automatically deleted by the tempfile functions) I found that the problem seems to be exactly this:

unix.stackexchange.com/questions/771815/...hen-building-gettext

Do you have a linuxcnc-dev directory? What does "which halcompile" say?

(I suspect that the "make clean" fix isn't going to help here)

The ETC86 is an Ethercat drive. This is the config I use

github.com/rodw-au/linuxcnc-cia402
you will need to modify it to work for you.

There will probably be a LinuxCNC meetup near Oslo in June, but that's not much help to you right now.

If you show us your current HAL file and tell us what is (or isn't) happening you might well get help here on the forum for free.

#include "Eth.h"
That is Wez's version, somewhere in the files it says the above but the file in question is named eth.h , notice the lower case, edit the src/Config.h with the lover case eth.h, it works then.
Let us know if the Wez's version works for you, it did not work for me, namely the networking on the W5500 does not work, to test in the console type ipconfig, if it reboots, well....
The version Michael linked does work right up until you enable LinuxCNC, then it keeps rebooting for me.

Im using the easyprobe pyvcp thing. its pretty easy for probe xyz, etc
but there is no toolsetter macro

You need a more recent build then the 2.9.3 release.
Not sure if the LinuxCNC build-bot is still working, if not you may
have to build from source (which is pretty fast on a RPI5)

Hallo Gemeinde,
der Postprozessor der bei Autodesk Fusion dabei ist, der ist nicht gerade das gelbe vom Ei.
Normale Bohrzyklen wie zB. G83 etc. werden komplett ausgeschrieben (Z+Z-Z+Z- usw.) ausgegeben und das obwohl ich
"Use canned cycles" angehakt habe. G76 wird allerdings richtig ausgegeben.
Nun suche ich einen PP der dieses Manko nicht hat.

Danke und Gruß
Olli