Advanced Search

Unfortunately, I got the lovely task of coming in behind somebody else and I need to figure it out. I knew nothing of Linux prior to a month ago, but I'm a fast learner. This has had me stumped for quite some time. If you could point me towards some documentation or if you could give some guidance on how to accomplish what is needed, that would be greatly appreciated.

If you look at the hal file, none of the important spindle connections
actually connect to anything, that is, the hi-lighted signal names are
not connected to any driving source.

EDIT: can you post the other hal files? I suspect the missing
connections are there.

HALFILE = Knuth.hal
HALFILE = 6i24.hal
HALFILE = LadderIO.hal
HALFILE = custom.hal
POSTGUI_HALFILE = custom_postgui.hal


Also using halshow to watch hm2_5i24.0.7i83.0.2.analogout0
when running the spindle in each direction would help debug this

They were posted earlier in this thread. Not much has changed since then.

Can you post your hal/ini files?

Upon closer inspection, it is 7i83. I'll get a flashlight next time. How can I tell if the version I have is too old? (LinuxCNC 3.4.6) Furthermore, can I update it?

You probably mean a 7I83 or 7I77 (the 7I87 is an analog _input_ card)

My guess is that there is a missing abs function in the hal file.

The 7I83 and 7I77 have bipolar +-10V analog outputs, but most VFDs
only take a unipolar ( typically 0 to +10V) input range, so you either need to
use the absolute spindle speed pin from motion or add an abs function
if you have an old enough LinuxCNC version that does not have the absolute
spindle speed pin.

Alright. Finally back on this project. The control appears to come from computer to a mesa 7187 to the VFD. The forward and reverse commands from the board go into STF and STR to command direction. Pins 2 (input) 5 (analog common). In forward the voltage ramps up and down as commanded through the gui and shows rpm in the bar. But in reverse the voltage to the VFD stays the same. If I switch forward and reverse command wires in the VFD they simply act in opposite. This eliminates the VFD as the issue. I need to find out why Linux isn't controlling the voltage to the VFD in reverse and also why it's not displaying reverse RPM in the GUI.

Glad it's working, I just automatically applied the patch.

The patch docs don't mention any project file
incompatibilities but that's what seems to have
been the  issue.
 

Hello PCW,

Your advice was right. I applied the 2025.1 patch an everything went as expected.
Efinity 2025.1.110.3.16 works fine on ubuntu 24.04.
I can edit the files, change them and flash my 7i96s wit the expected result.

Thank you very much

I've got linuxcnc 2.9.3 installed on Debian 12 Bookworm with prempt RT kernel. I followed these instructions to install Probe Basic via apt.

kcjengr.github.io/probe_basic/stable_quick_start_apt.html
(Actually, I've tried both stable and dev versions).

I'd like to try Z-compensation via external offsets and was hoping to use surfacemap to automatically probe an XY grid of heights. (This is for PCB engraving on a mini mill).

Surfacemap is apparently implemented as a Probe Basic user tab:

github.com/zmrdko/01_surfacemap?tab=readme-ov-file

In the probe basic probing tabs I've had a few troubles with the entry boxes in the UI not accepting values. I think I managed to fix this problem by adding missing variables into the parameters files.

The surfacemap user tab seems to be giving me similar problems but I wasn't able to fix that just by adding parameters.

Has the surfacemap code perhaps been broken by the latest version of probe basic? Or had it perhaps never reached a fully working state?

Any other options for automated Z-level probing and compensation?