Advanced Search

Search Results (Searched for: )

  • RotarySMP
  • RotarySMP's Avatar
02 Apr 2026 20:16
Replied by RotarySMP on topic Schaublin 125-CNC retrofit.

Schaublin 125-CNC retrofit.

Category: Turning

Thanks for that, I was not able to get into the forum for the last couple of days.

I had similar advice from AI, and tried both the solution yours and mine gave and it has not corrected it.

I even used visual code studio, and rewrote the whole program be hand, to ensure there were no secret character codes in there. (See attached file)But the result is still the same.
hal:turret:1: character unknown '  '

Any other suggestions much appreciated. 
Cheers,
Mark
  • motionmasterupgrade
  • motionmasterupgrade
02 Apr 2026 19:41
Replied by motionmasterupgrade on topic MotionMaster Upgrade

MotionMaster Upgrade

Category: General LinuxCNC Questions

Before you make any decisions, you've got to sort out power
 

THat's what I did today:
The VFD takes 480V directly from source (I am going to replace VFD and spindle, so it doesn't matter).
A 480V single phase line powers the DC bus which in turn supplies power to all 4 drives.
The transformer ties two of the lines into a neutral-ground, and outputs 120V to power the controller and peripherals. I replaced this with 120V AC to test if the controllers are working. Was told by seller that it takes a while to boot. It's abootin as I write this.

The servos are all 140V 65amp max. I do not need the maximum capacity (I am carving wood, and I am doing low throughput. I was thinking to sell the drives (they are out of production and I imagine someone who is running this machine in a production setting would want parts) and replace them with step/dir ones as someone else had recommended. The VFD takes 3 phase power at 360 to 480. I was wondering if I could just supply 220V power 3 phase to the VFD and have it power a 220V 3 phase spindle. This would make the work a lot easier, or just replace it with a single phase 220V VFD to power a 4ish HP spindle. That would work too. Either way, I really don't see why I should stick to 480V single phase since the part that is drawing power is the spindle which I really do not need.
  • motionmasterupgrade
  • motionmasterupgrade
02 Apr 2026 19:39 - 02 Apr 2026 19:40
Replied by motionmasterupgrade on topic MotionMaster Upgrade

MotionMaster Upgrade

Category: General LinuxCNC Questions

placeholder
  • spumco
  • spumco
02 Apr 2026 19:26
Replied by spumco on topic CSS/G96 for other than X-axis?

CSS/G96 for other than X-axis?

Category: Advanced Configuration

What would be needed here is to have the _interpreter_ reject the axis word which was disabled. 



So the NML description in the LCNC manual isn't accurate? Or perhaps not exactly clear what happens?
8.10. DEACTIVATE_JOINTThe DEACTIVATE_JOINT command turns off all the calculations associated with a single joint, but does not change the joint’s amp enable output pin.

I would have thought 'turns off all the calculations..." would have removed that joint/axis from the TP calcs entirely.  
  • spumco
  • spumco
02 Apr 2026 18:20
Replied by spumco on topic CSS/G96 for other than X-axis?

CSS/G96 for other than X-axis?

Category: Advanced Configuration

On second thought it probably wont work like that  I can't have "v" defined in case-0 as well as an 'unused coordinate.'

But maybe there's a way to define the motion math of case 1  pos->v= j[0]; so that no motion results regardless of input.
 
  • spumco
  • spumco
02 Apr 2026 18:14
Replied by spumco on topic CSS/G96 for other than X-axis?

CSS/G96 for other than X-axis?

Category: Advanced Configuration

I wonder if this could be done through the switchkins component.

Do you think LCNC would throw an error if I just left joint-0 out while operating under Case 1? 

Instead of this:
case 0:
            pos->tran.x = j[0];
            pos->tran.z = j[1];
            pos->c      = j[2];
            pos->v      = j[3];
            break;
        case 1:
            pos->tran.x = j[3];
            pos->tran.z = j[1];
            pos->c      = j[2];
            pos->v      = j[0];
            break;
    }
    // unused coordinates:
    pos->a = 0;
    pos->b = 0;
    pos->u = 0;
    pos->w = 0;

Maybe this:
case 0:
            pos->tran.x = j[0];
            pos->tran.z = j[1];
            pos->c      = j[2];
            pos->v      = j[3];
            break;
        case 1:
            pos->tran.x = j[3];
            pos->tran.z = j[1];
            pos->c      = j[2];
           break;
    }
    // unused coordinates:
    pos->a = 0;
    pos->b = 0;
    pos->v = 0;
    pos->u = 0;
    pos->w = 0;

And do something similar for the inverse later in the comp.

​​​​​​​
  • Aciera
  • Aciera's Avatar
02 Apr 2026 17:53
Replied by Aciera on topic CSS/G96 for other than X-axis?

CSS/G96 for other than X-axis?

Category: Advanced Configuration

What would be needed here is to have the _interpreter_ reject the axis word which was disabled. So just like if the operator used an axis letter that was not configured in the ini file. Any MDI or gcode program line containing the disable axis letter would create an error 'Bad character x used'.

Unfortunately I do not know how to dynamically pass the disabled axes to the interpreter during runtime.
  • Aciera
  • Aciera's Avatar
02 Apr 2026 17:44
Replied by Aciera on topic CSS/G96 for other than X-axis?

CSS/G96 for other than X-axis?

Category: Advanced Configuration

Not sure if there is an easier way of trying this but I inserted the the actions of 'EMCMOT_JOINT_ACTIVATE' / 'EMCMOT_JOINT_DEACTIVATE' (ie set/reset 'SET_JOINT_ACTIVE_FLAG' for a specific joint :
 

into 'EMCMOT_SPINDLE_ON' / 'EMCMOT_SPINDLE_OFF':

 
 


Which lets me clear 'joint.1.active' using the spindle control but it does basically the same as disabling the drive as 'joint.1.pos-cmd' is still updated as the axis is jogged or commanded through the interpreter:

 

The problem with the following error when reactivating the joint remains the same as if it was done in hal or the drive disabled. 
 
  • Hakan
  • Hakan
02 Apr 2026 17:21
Replied by Hakan on topic Ethercat SDOS to XML

Ethercat SDOS to XML

Category: EtherCAT

After you set it the first time, does it disappear from the sdos listing or does it just become read-only?
  • andrax
  • andrax's Avatar
02 Apr 2026 16:56
Replied by andrax on topic Ethercat Setup help needed for beginner

Ethercat Setup help needed for beginner

Category: EtherCAT

A servo of this size is not a toy.
The “Servo On” or “Enable” signal comes from the controller (PLC / lcnc).
The emergency stop and limit switches are hardwired.
  • andrax
  • andrax's Avatar
02 Apr 2026 16:51
Replied by andrax on topic Ethercat SDOS to XML

Ethercat SDOS to XML

Category: EtherCAT

Have you tested the module in Twincat yet?
If it runs without any issues there, you could use Wireshark to monitor the communication. Maybe that will turn something up.
  • Finngineering
  • Finngineering
02 Apr 2026 16:02
Replied by Finngineering on topic XHC WHB04B development?

XHC WHB04B development?

Category: General LinuxCNC Questions

I imagine that you will be satisfied with the updated xhc-whb04b-6 component by hdiethelm, which allows the pendant to reconnect in case it disconnects (as well as several other improvements). But you need to compile it from source, same as building LinuxCNC from source (which is not all that difficult). He made a backport to LinuxCNC 2.9.8:
github.com/hdiethelm/linuxcnc-fork/tree/...x-v2_backport-v2.9.8
So basically git clone that repository, checkout xhc-whb04b-6-fix-v2_backport-v2.9.8 and then compile. I don't know about exact version compatibility, but I don't think you actually need to use v2.9.8 to run the component. You can just give the full path to the newly built compnonent binary in the loadrt command.

I made a custom firmware for the USB dongle that I use myself, but I can't really recommend that for others. There is a decent risk that it would irreversibly break your dongle in case the hardware is slightly different than what I have.

I don't know your proficiency with git and compiling LinuxCNC from source. Just ask if my somewhat terse instructions are not clear to you.
  • jefsaro
  • jefsaro's Avatar
02 Apr 2026 15:49
Replied by jefsaro on topic new probe

new probe

Category: General LinuxCNC Questions

desoler , je ne comprend pas ,car j 'ai bien le signal dans la config de hal
  • spumco
  • spumco
02 Apr 2026 13:23
Replied by spumco on topic CSS/G96 for other than X-axis?

CSS/G96 for other than X-axis?

Category: Advanced Configuration

I'm not aware of a way to block the operator from commanding a move of a particular axis. Would be a handy feature to have.


Appreciate you having a think about it.

I'd like to experiment with that NML message I found (DEACTIVATE_JOINT).  If I can deactivate joint-0 near the beginning of the sub-spindle transfer process I can simplify the parting-slide jogging disable (which happens later in the overall sequence).

Any idea how I can send an NML message to LCNC?

Maybe it's done via python interface, but I don't see anything obvious in the LCNC manual section on python-interface that includes that specific command.
 
  • Aciera
  • Aciera's Avatar
02 Apr 2026 12:47
Replied by Aciera on topic CSS/G96 for other than X-axis?

CSS/G96 for other than X-axis?

Category: Advanced Configuration

Any ideas how to no-kidding stop an axis from being moved without disabling the drive? like DEACTIVATE_JOINT or something?


I'm not aware of a way to block the operator from commanding a move of a particular axis. Would be a handy feature to have.

- I use Gmoccapy where the onscreen jog buttons can be hidden and only physical jogging through hal pins needs to be handled.
- MDI entries are more involved as one would need to parse all commands that an operator enters, still possible without too much effort.
- Rehoming can be handled using 'motion.homing-inhibit'.

Still leaves program execution and any mdi commands that are not coming from the gui entry.
Displaying 76 - 90 out of 17464 results.
Time to create page: 0.232 seconds
Powered by Kunena Forum