Advanced Search

Aciera wrote:

So, the .pts files contain a path on the surface of a 3d model as consecutive 3d points and you are trying to post process this to create 4axis (xyzc) gcode where the c axis keeps the tool perpendicular to the model surface.

Yes.

Aciera wrote:

Seems you are basically trying to use a python filter script to do what a CAM and post prosessor would usually do.

Yes.

Aciera wrote:

You could use dx and dy from the pts file to calculate the tool orientation angle in the xy-plane (ie the c axis value) that keeps the tool perpendicular to the path segment.

How?
I used some Gemini generated scripts but all had issues with ... everything.

Aciera wrote:

In order to keep the tool positioned on the work piece while the c-axis moves you need TCP kinematics.

Yes of course, already set and configured ( i think), but a bit fuzzy on G53 vs G54 values, as i have G54 X0Y0 at the center of the C axis, but G53 are far from it.

Aciera wrote:

The 1mm tool path offset could possibly be achieved by simply using a 2mm bull nose bit (depends on the required detail resolution of course).

I do not think this is important much, the material is soft, both of them, so has to just pierce the thin one (0.8-1mm) and does not matter if it goes deeper, as long as that "deeper" is under 4-5mm.

Aciera wrote:

Of course, the proper and easy way to do this would be to use a CAM

Yup, know of any good open source one?
reminds me, have to check FreeCad if it can do anything about it, but this adds another layer for the client to deal with, while the script/filter one is neat, open the file, home the machine (has switches), run.

Hi, I bought a used Rat Rig StrongHold that ran fine for the previous owner. I re-flashed a fresh Flexi-Pi image and now I’m stuck. Setup: • Rat Rig StrongHold, Flexi-HAL (STM32F4, SPI), Raspberry Pi 4• LinuxCNC 2.9.8, QtDragon_hd 1.5• remora-flexi config (~/linuxcnc/configs/flexi-hal/)• Gantry: TWO Y motors (joint.1 + joint.3), but only ONE Y limit sensor• Sensors: Heschen SN04-N2 (inductive, NPN, NC) Problem: • Status shows “hard limits tripped”, machine won’t turn ON• HOME X / HOME Y / HOME Z read as triggered (red) permanently in the INPUTS panel, machine sitting free in the middle• The separate “LIMIT” input does NOT light up Key finding — this is the important part: • If I DISCONNECT a limit switch, NOTHING changes — the home input stays triggered.• So the input reads as permanently active regardless of whether a sensor is connected or not.• Toggling .not on the home-sw-in lines in the HAL does nothing either. Commenting the lines out completely also changes nothing.• The sensors themselves work fine (LED on sensor and breakout board switches with metal, they have power). I did NOT change any wiring — it’s exactly as it ran for the previous owner. So it seems the inversion/logic is not on the LinuxCNC side at all. My question: Where does the limit/home input polarity get set for the Flexi-HAL? Is it a grblHAL firmware setting on the board itself (like $5 invert limit pins, or an NC/NO setting), separate from the LinuxCNC HAL/INI? If so, how do I read and change it — through LinuxCNC, or do I need to connect to the board another way? Since disconnecting the switch doesn’t change the input at all, I suspect the board is configured to expect the opposite switch type (or the pins are inverted at the firmware level). Also (probably separate): MODBUS ERRORS counter keeps climbing (was ~174, rising), FAULT 0x0 — that’s the VFD/RS485 side (vfdmod, set for Durapulse GS10 by default). Happy to post photos of the board, breakout PCB, INPUTS panel, and my full HAL/INI. Thanks a lot!

Re: forum.linuxcnc.org/30-cnc-machines/42689...0-ap?start=80#346618

1. These two errors are connected. The parameter #<_tool_in_spindle> is not defined because 'change_prolog' cannot be found.

2. Looking at your ini file it seems that your paths are messed up. TOPLEVEL should point to the file 'toplevel.py' not just some folder:

Since we don't have your complete config folder there is no telling where you have the relevant files but if you compare your setting with the built in 'sim/axis/remap/extend-builtins/' you might get some ideas.

(You will find the [PYTHON] section at the very bottom of the ini file below):
github.com/LinuxCNC/linuxcnc/blob/2.9/co.../extend-builtins.ini

Sorry, been away for a while. I will get them posted soon.

Hi everyone,
I’ve tried a few more times to solve my problem, but unfortunately, I’m not getting anywhere. Perhaps someone can tell me how to fix the issue so I can get the tool changer up and running.

Heyo anyone that reads this and attempts to pass me some knowlege.

I have been tasked with researching how to Retrofit our VMC Bridgeport 760 with a Heidenhain 370 Control to use Linux CNC. I will continue checking any post  related to the topic already on this website along with other websites and stuff. However I figured making a post on a forum for the software we are planning to use would be a great start because I could learn some users here that may be able to help point me in the right direction to start or where they may be good sources of knowledge for a project like this. 

We are looking to retrofit after years of using the original electronics because some of those electronics are starting to fail and we'd like to switch over to an easier system for maintenance versus sending these components in for repair or buying refurbished parts. So any information on where to look for knowledge on retrofitting this machine or a machine closely similar to ours would be much appreciated. 

Thank you for any information that someone is able to provide.

I tried your suggestion of changing the "Display = gmoccapy", with no success.
Then I decided to update to v2.9.10 .
That did the trick, I changed "Display = gmoccapy" in axis. Now i can jog and home all axis in gmoccapy.
Thank you for your patience..

If you have a long string of conditional pins/signals, the logic component is a little less messy in HAL than a bunch of nested/sequential and2 components.

linuxcnc.org/docs/stable/html/man/man9/logic.9.html

If you just want some interlocks to prohibit starting the spindle while chuck is open, or opening the chuck while spindle is turning, and2 or logic work fine.

However, I prefer using Classic Ladder for my spindle(s) and drawbar(s).  My lathe has two spindles, each with a pneumatic drawbar.  I control the drawbars with either physical buttons OR via M-codes.  M-code control is necessary for automated bar-feed operations.

I also want to be able to open the collet closer while the spindle is running, although at a low maximum speed.  Also need ability to orient spindle while the closer is open (hex/square stock).

Given the complexity of my particular config, I found ladder be be more intuitive and easier to configure (and test).

Yes, same here. I was just searching the forum if others have the same problem or if it is just me "holding it wrong".

In my usual workflow, I load a toolprobe using a manual toolchange with automatic tool length measurement. This happend in MDI. Then I usually forget to hit the MAN button and wonder why my ShuttleXpress does not work anymore.

Same with using the toolprobe in the Probe tab. After probing we are again back to MDI mode.

I was wondering if there is a reason for being in MDI mode. From a user standpoint I was manually operating the machine, expecting to be back in manual mode after pressing a few buttons, even when those trigger MDI mode.

Guido

Try this image:
www.linuxcnc.org/iso/linuxcnc_2.9.10-amd64.hybrid.iso

Hi everyone,I'd like to add some safety conditions for enabling the spindle on my lathe.The spindle should only be allowed to start if all required conditions are met, for example:

• The auxiliary drives are enabled.
• The chuck is clamped (clamping pressure is present).
• (I may add more conditions later.)

My first idea was to simply combine these signals with one or morecomponents and use the result as the spindle enable signal.Something along the lines of:Is this the usual way to implement this in LinuxCNC, or is there a better or more recommended approach?How have you implemented spindle interlocks on your machines?Thanks!

rodw wrote:

2.9.10 has been released. Maybe try it.
 

We don't have a 2.9.10 .iso yet. I will try to generate one today. 
 

I've not seen any macros or subroutines ever posted for C-axis milling.

There have been a couple of polar interpolation posts, but nothing I'd call 'universal.'

You might look at this:

forum.linuxcnc.org/show-your-stuff/53476...rical-mapping#308478

Note that Fusion360 appears to have (somewhat recently) added a feature for polar interpolation mode.  Rather than using a machine's built-in polar interpolation F360 outputs XC moves directly instead of XY.

I've not tried it yet, but worth a look.

Have been considering buying a stepper motor and some parts and seeing if i can make them 'function' before i invest the time into the steel to build the machine

I think that's a good plan. Then maybe work on using that motor to operate a CNC backstop, for example, if you have a manual press-brake already.

Hi all! 
I came having problems when trying to run a new probe routine after running a program in QtDragon, a "cannot switch pages while in AUTO mode" message appears but only affects to te PROBE tab all the rest seems to work fine.

I try to solve ending programs with M30, pushing STOP button, switching to MDI before MANUAL mode but no way.

I´m running linux uspace 2.9.0~pre1, kernel 6.1.0-28-rt-amd64 in LMDE 6 Faye (Cinnamon 6.4.8) and using BasicProbe

Can it been fixed easyly or it would be better to upgrade to a new fresh realese?