Yes, this is one of the reasons why I ditched Rufus and now exclusively use and recommend  balena etcher on all platforms

If your 7i92 doesn't have a free input for the Z signal, that's too bad. I'd recommend upgrading so you have more inputs. Alternatively, you can set HOME_USE_INDEX = No, and then LinuxCNC will use the home switch for reference. However, this is less accurate.

automata wrote:

Hi grandixximo,
Congratulations on a viable solution to this long standing gap in Linuxcnc implementation.
I was going through how you have taken into account the kinematics in userspace for planning which is needed for kinematics bounds checking.
I also see that you have made the relevant changes in the kinematics defined in src/emc/kinematics/* files.
There is another set of switchable kinematics defined in the src/hal/components/*.comp files (method made viable by David Muller / aciera on this forum). For example xyzab_tdr_kins.comp, xyzacb_trsrn.comp etc.,
I somehow find the idea of implementing the kinematics as a loadable .comp module compiled by halcompile very fascinating and useful
Is there a guideline on how to transition those kinematics also to this new setup?
-automata

Hi automata,

Good question — the .comp-based kinematics (like xyzab_tdr_kins.comp, xyzacb_trsrn.comp, etc.) were missing planner type 2 support because halcompile doesn't generate the nonrt_attach() entry point or the shared-memory parameter bridge that the userspace planner needs.
I've added a glue header (comp_kins_uspace.h) that makes the conversion straightforward without having to rewrite the .comp as a hand-written .c file. The idea: HAL pin values are mirrored into shared memory by the RT module each servo cycle, and the userspace planner reads a consistent snapshot from there instead of touching HAL pins.
The conversion is 5 mechanical steps:

#include "comp_kins_uspace.h"

Call comp_kins_uspace_setup(comp_id, "mykins", num_joints, "XYZAB") in your setup function after hal_ready()

Replace each *(haldata->pin) with KINS_READ(haldata->pin, N) where N is a unique index per parameter

Bracket kinematicsForward/kinematicsInverse with COMP_KINS_BEGIN(haldata) / COMP_KINS_END()

Add COMP_KINS_NONRT_ATTACH("mykins") at the end of the file

For switchable kins, also sync the switch type with COMP_KINS_SET_SWITCH_TYPE() in kinematicsSwitch().
xyzab_tdr_kins.comp has been converted as a reference example — look for the lines marked with +++ comments. The kinematics docs (docs/src/motion/kinematics.adoc) now have a full section on this with annotated code examples.

Hello
Please see the attached Files.

Thanks for looking through them.

~Takuya

Thank you for taking the time to go through the messages from the last few days. I suspected as much, the AI's "fix" wasn't really anything useful.

Regarding the first slave lacking DC, you are correct. I actually changed the order of slaves in the chain yesterday so that the A6 servos are now defined first. They do have assignActivate defined. However, that change didn't make any difference to the dc-phased pin or the pll-*** pins.

I have attached my ethercat-conf.xml along with HAL and INI files if they are of any help.

ethercat master command after LinuxCNC has started:

Master0
  Phase: Operation
  Active: yes
  Slaves: 6
  Ethernet devices:
    Main: 10:60:4b:5b:58:34 (attached)
      Link: UP
      Tx frames:   146273
      Tx bytes:    9824436
      Rx frames:   146272
      Rx bytes:    9824318
      Tx errors:   0
      Tx frame rate [1/s]:    800    709    308
      Tx rate [KByte/s]:     93.7   84.1   32.2
      Rx frame rate [1/s]:    800    709    308
      Rx rate [KByte/s]:     93.7   84.1   32.2
    Common:
      Tx frames:   146273
      Tx bytes:    9824436
      Rx frames:   146272
      Rx bytes:    9824318
      Lost frames: 0
      Tx frame rate [1/s]:    800    709    308
      Tx rate [KByte/s]:     93.7   84.1   32.2
      Rx frame rate [1/s]:    800    709    308
      Rx rate [KByte/s]:     93.7   84.1   32.2
      Loss rate [1/s]:          0     -0      0
      Frame loss [%]:         0.0   -0.0    0.0
  Distributed clocks:
    Reference clock:   Slave 0
    DC reference time: 826802114159640000
    Application time:  826802133398422723
                       2026-03-14 11:15:33.398422723

@TAKUYA
@Atsu

Please share your full XML
and the output of ethercat master command after you have started linuxcnc

Atsu wrote:

Experienced some jitter as well today while running grandixximo's fork. I doubt there is any synchronization being done while the dc-phased pin is false and all pll-*** pins are 0. Ran sudo dmesg to see what is happening in the background. Maybe someone can notice something clear in the log:

During startup:
[  801.691210] EtherCAT: Requesting master 0...
[  801.691215] EtherCAT: Successfully requested master 0.
[  801.691915] EtherCAT 0: Domain0: Logical address 0x00000000, 74 byte, expected working counter 12.
[  801.691918] EtherCAT 0:   Datagram domain0-0-main: Logical offset 0x00000000, 74 byte, type LRW.
[  801.691932] EtherCAT 0: Master thread exited.
[  801.691933] EtherCAT 0: Starting EtherCAT-OP thread.
[  801.691970] EtherCAT WARNING 0: 1 datagram UNMATCHED!

During operation:
[  807.424674] EtherCAT WARNING 0-1: Slave did not sync after 5000 ms.
[  808.944250] EtherCAT 0: Slave states on main device: OP.

During shutdown:
[  841.183382] EtherCAT 0: Master thread exited.
[  841.183401] EtherCAT 0: Starting EtherCAT-IDLE thread.
[  841.183449] EtherCAT ERROR 0-0: Failed to receive AL state datagram: Datagram initialized.
[  841.183462] EtherCAT 0: Releasing master...
[  841.183464] EtherCAT 0: Released.
[  841.188247] EtherCAT 0: Slave states on main device: SAFEOP, OP + ERROR.
[  841.196539] EtherCAT ERROR 0-0: AL status message 0x001A: "Synchronization error".
[  841.197317] EtherCAT 0-0: Acknowledged state SAFEOP.
[  841.208612] EtherCAT ERROR 0-1: AL status message 0x001A: "Synchronization error".
[  841.209116] EtherCAT 0-1: Acknowledged state SAFEOP.
[  841.220587] EtherCAT ERROR 0-2: AL status message 0x001A: "Synchronization error".
[  841.221590] EtherCAT 0-2: Acknowledged state SAFEOP.
[  841.256274] EtherCAT 0: Slave states on main device: PREOP.
[  841.828318] EtherCAT WARNING: Datagram 0000000031def7fe (master-fsm) was SKIPPED 1 time.
 

This seems a similar issue, please share your XML file
also send this command 

ethercat master

and share the output
the output should look something like

Master0
  Phase: Operation
  Active: yes
  Slaves: 12
  Ethernet devices:
    Main: 88:03:8f:28:ca:33 (attached)
      Link: UP
      Tx frames:   78788870
      Tx bytes:    65756113151
      Rx frames:   78788869
      Rx bytes:    65756112300
      Tx errors:   0
      Tx frame rate [1/s]:   1000   1000   1000
      Tx rate [KByte/s]:    822.0  822.0  821.9
      Rx frame rate [1/s]:   1000   1000   1000
      Rx rate [KByte/s]:    822.0  822.0  821.9
    Common:
      Tx frames:   78788870
      Tx bytes:    65756113151
      Rx frames:   78788869
      Rx bytes:    65756112300
      Lost frames: 0
      Tx frame rate [1/s]:   1000   1000   1000
      Tx rate [KByte/s]:    822.0  822.0  821.9
      Rx frame rate [1/s]:   1000   1000   1000
      Rx rate [KByte/s]:    822.0  822.0  821.9
      Loss rate [1/s]:          0     -0      0
      Frame loss [%]:         0.0   -0.0    0.0
  Distributed clocks:
    Reference clock:   Slave 0
    DC reference time: 826799805203739108
    Application time:  826800839056809695
                       2026-03-14 10:53:59.056809695

 

Hello everyone. I haven't been to the forum for a long time.
Need help and complete clarity in using HOME_INDEX = YES and STEP/DIR
When I upgraded old machines that used analog control and encoder feedback, this issue was not there. In the topic Homing index andypugh writes and gives a link to the source file homing.c. I asked chatGPT to analyze this file. And the chat says that indeed to reset the HOME_INDEX search flag you need a signal from encoder.n.index-enable. That is, linuxcnc does not support the function of searching for 0 labels in step/dir mode.
Most of the new hobby machines that I and my friends are currently assembling use servos in step/dir mode and without a closed loop. Since for a hobby, the driver <-> motor with encoder connection is enough. But home_switch gives a small but noticeable error. Therefore, such a request will appear more and more often.
On my 7i92 there are no free 9 pins for 3 x A B Z encoders. And almost all servo drivers have a simple transistor output that duplicates the 0 encoder mark.
Are there any "macros" or "tricks" to get around this problem?
Is it possible to add this functionality to the homing.c source?
Thx for help.

Atsu wrote:

I have the same issue as TAKUYA, the dc-phased pin never goes high. I managed to fix this using Claude AI and now the dc-phased pin goes high. I have attached a detailed notes of what was changed in the code as well as some screenshots from before and after.

NOTE: this was done entirely with the help of AI and I am not qualified enough to say whether it is correct or not! Grandixximo, could you please review the notes and share your thoughts? Opening the attached HTML file in a browser, you can view the notes which were generated. It is too long to copy-paste here.

EDIT: After testing the code more thoroughly, I'm getting jitter again. It's very likely that the AI's "fix" only addressed the dc-phased pin but broke something else... I've reverted back to the master branch ( github.com/grandixximo/linuxcnc-ethercat ), and everything is working again except for the dc-phased pin.
 
 

The issues the AI found are not issues,

1. Same-cycle read — No, 
   Code:

   reference_clock_time
    reads from the previous cycle's response, already received in 
   Code:

   lcec_read_master
2. Missing 
   Code:

   sync_monitor_queue
    — Not needed; DC ref clock time comes from the normal sync datagrams
3. Code:

   sync_reference_clock
    guard — Intentionally skipped in M2R mode (correct behavior)

And not something related to my changes, the behaviors were already present from upstream, can you share your xml file? Your app time seems good, but the pll-err is suspiciously to 0
which tells me you probably do not have DC, is assignActivate set to 300 on your DC capable devices?
Your first device lacks DC, I would advise to keep the first device on your network to be DC active, or you may need to set which slave does the DC timing reference in the XML, I always keep first device DC, have not tested differently, can't say for sure.

• EK1100 — coupler, no DC
• EL1008 — digital input, no DC
• EL2008 — digital output, no DC
• 3× StepperOnline A6 — these should have DC, but only if 
  Code:

  assignActivate
   is set in the XML config

In your XML you will need something like

    <master idx="0" appTimePeriod="1000000" refClockSyncCycles="-1">
        <slave idx="xx" type="devicewith DC" >
        <dcConf assignActivate="300" sync0Cycle="*1" sync0Shift="250000"/>
        </slave>Or you share your XML here, I will take a look, I think you do not have DC at all at the moment, you never even used any of my code changes at all.

endian wrote:

Hello, 

I can see merge of your progress to master? Am I right ?

Regards e.
 

No, we will make another PR for merging into master, this is more of a testing version, it was shared as a draft, it is good to get feedback on it, as it will be the base for the future work.

 

/eugenius

Unfortunately I am new to alll this hal stuff. Is there a good place to learn more about it?
I attach the hal and ini files.
Thanks!

the same for .hal

it should (must) be solved one day.

MR-J3-A (as well as MR-J,J2,J2S,J4,J5-A) drives have generic pulse train position or +-10V analog speed and/or torque control

MR-J,J2,J2S-B - copper SSCNET I/II control interface
MR-J3,J4,J5-B - optical SSCNET III or III/H
MR-J5-G - CC-Link IE TSN over gigabit Ethernet
MR-J5-G-N1 - EtherCAT

Sounds like a HAL file setup error of some kind rather than a hardware related error.

Posting the hal/ini files would be helpful