Advanced Search

yourmachine.ini  file,  in  [TRAJ]  section  POSITION_FILE = posfile.txt   stores machine positions at shutdown. If this is not set, the machine starts with all axes set to zero

 

Can you elaborate on what are these architectural issues?

andrax wrote:

Have you ever had the problem where one of the Y-axes doesn't start moving during homing and the gantry is crooked?
That happens to me a lot. I have to restart the machine and then it works again.

That should not happen, ever, and it did not happen as i have probably over 15 machines with tandem joints, some industrial ones with over 400KG heavy gantries.
You should check the wiring first, then test the failing side drive.
Are you using the same drives mentioned here?

I've checked all the reported spikes, and the issue is not the new s-curve in itself, but the trajectory planner architecture, we will have to do a much more involved refactoring, will look at Tormach work, maybe talk with Rob if that happens, and take it from there.
The current scurve is best that can be done within the limitations of the current tp architecture.

Thanks for the reply
I tried corexykins, didn´t work because i use uspace?
so i went with KINEMATICS = corexy, which then caused problems with loadrt [KINS]KINEMATICS and loadrt [EMCMOT]EMCMOT...
Why are home switches necessary? I dont have the machine connected at the moment since I´m only trying to get the software to start.

Is this combination of software even the right choice?
I´m building a small portable plasma cutter while using some parts of an old 3d printer and i would really like to use corexy.
Are there other GUIs that work better with this?

...

Edit: Launched linuxcnc on terminal, getting "bad spi payload"
(Bad SPI payload = 32393a38)

The value of the payload is strange; it's almost half value of what you are suppose to get:

#define PRU_DATA            0x64617461     // "data" SPI payload
#define PRU_READ              0x72656164  // "read" SPI payload
#define PRU_WRITE             0x77726974  // "writ" SPI payload
#define PRU_ESTOP           0x65737470  // "estp" SPI payload

I'd try with a lower SPI frequency.

The Maho is back up and running again, but I still dont really know what the FError and MinFError "should" be for such a machine.


Thanks for your help.

Simple question that I cannot seem to locate the answer:
No matter what offset I was using when I shutdown LinuxCNC I want my DROs to read 0 after I restart and then home. I can do this with the Touch-Off button in Axis but how do I set my INI/HAL file to clear offsets when LinuxCNC restarts?

Of is there a mdi input that would have the same outcome?

Thanks,
HH

Hi, that looks really good.
How stable is your system?
Have you ever had the problem where one of the Y-axes doesn't start moving during homing and the gantry is crooked?
That happens to me a lot. I have to restart the machine and then it works again.

It's been almost a year since I originally started this thread and thought it might be of interest to give a status update on my machine.It's been humming along without issue powered by the A6 400watt AC ethercat servos.It is a 4x4 size router direct driven with ballscrews on all axis (XYYZ). I highly recommend the A6 servos for simple set-ups. Mine is not a demanding setup but for the cost, in my opinion they are an unbeatable value. Thank-you to those that gave me pointers when I was making the transition from step drive to ethercat!

 

That will most probably require a bit more hand editing as you also need to have some of the contents of 3joint_xyz.hal, but i do not recall exactly what or how, it was a long time ago since i used this.
Did you try changing the kinematics in the in file to
KINEMATICS = corexykins
And did that produce any result?
But first, does the machine have home/limit switches on X and Y ? That is a must.

Hi everyone, i´m new here.I’m having major issues getting my LinuxCNC setup to work with QtPlasmaC and CoreXY kinematics.My setup:

• LinuxCNC 2.9.6
• Mesa 7i96s Ethernet + 3 steppers (X, Y, Z)
• QtPlasmaC GUI
• CoreXY kinematics (tried corexykins and corexy_by_hal)

QtPlasmaC seems to try to access pins and signals that don’t exist with CoreXY kinematics.
I had previously tried running the machine with SCARA kinematics, but I naively thought CoreXY would be simpler.

Is a setup like this possible for a beginner or should i try a different approach?
Why are kinematics that complicated and troublesome anyways?

Thanks in advance for any help!

Do those files need to be executable?

tommylight post=341235 userid=17274

andypugh post=341234 userid=723... No momento, não consigo visualizá-los. 
 
+1
 
 
Peço desculpas, é que eu não gerencio repositórios e não percebi que os links estavam bloqueados.

1 drive.google.com/file/d/1xICFF-2mq_rz-vi...0W8/view?usp=sharing 

2 drive.google.com/file/d/1WV1pdWmNyH8BmgD...qg-/view?usp=sharing

3 drive.google.com/file/d/14Z10IOZI5bvgmf_...Rn6/view?usp=sharing

4  drive.google.com/file/d/1fwGb5Qo9pw8lGLp...LOE/view?usp=sharing 

Yes, you are right — I was able to update the AX5112 to firmware version 2.16. I can now configure it using Drive Manager 2. Thanks.

Now the next problem occurs: “PreOp to SAFEOP failed.”
Do any additional settings need to be made here?

10.01.2026 17:17:32 814 ms | "Antrieb 1 (AX5112-0000-0216)' (1001); 'PREOP to SAFEOP' failed! Error 'check device state for SAFEOP', AL Status Ox0012' read and Dx0004' expected, AL Status Code 0x0045 - MBX SoE'10.01.2026 17:17:32 814 ms | 'Antrieb 1 (AX5112-0000-0216) (1001); state change aborted (requested 'SAFEOP', back to 'PREOP).

An AX5801 industrial safety card is installed in the AX5112. The two safety inputs are currently supplied with 24 V. Could this be causing the problem?

The connected servo motor and resolver are not listed in the database, so we have to identify manually which type of resolver it is (Eco&Logic Syncrotech SY150…). From the old servo drive we know that it is a resolver.