Don't use EtherCAT if you want to use a step/dir driver.
There are advantages to skip the intermediate step with
step/dir for steppers and analog +-10V for servos.
The right way is to have a EtherCAT drive that controls the
motor directly, skipping that middle step.

I have built an EtherCAT board that gives step/dir signals.
But it's not the right way to go.

Hi, Rodw! Glad to see you in this topic)

Jog before homing.. hmmm.. for example, edm machine , some emergency, estop pressed or power loss.. and mechanical of the machine stucked. I have to move before homing, because homing may broke everything on the work table

Yes, it will work. But only after homing.
Before homing its work as 3rd number (all axus numbers are 0-1-2-3 )

Thanks for your answer.
I will try it

An what about External power button to switch ServoOn? I need separately button to send command the same as power_pb ib flexgui

TS

Find the culprit?

here is some info on a cheap ethercat stepper drive. But it is not as cheap as your A4988.
4 axis stepper drive

You could try  setpci and verify if the [disabled] goes away from the lspci listing

the options in my BIOS PCI Subsystem Settings are:
Above 4G Decoding (enable/disable)
SR-10V Support (enable/disable)

and in BIOS Onboard Device Configuration
PCIEX1_1 Mode (auto) <--but i think this only refers to speed. I have this selection/option for all the PCI slots on the board

I tinkered with all of these earlier today to no result. I do have an ASUS A520M motherboard...BIOS options seem pretty generic.

I do have access to some BIOS PCI settings, but not sure if I can screenshot them. I am afraid the best I can do is take a picture. 

I did some searching...will this command do anything for me?

The problem is this:

Region 0: Memory at fce00000 (32-bit, non-prefetchable) [disabled] [size=64K]

Which is likely something the BIOS did so may be fixable if you have access to BIOS PCI settings
I will take a look into what is needed to get mesaflash to re-enable the memory region (lLinuxCNC does this)

BryceJ wrote:

Ran into an issue with IRQ balance after updating to trixie
 

Late to the party but currently best practice is not to use irqbalance. the balancing is now managed internally in the kernel so adding a second balancer can confuse things.... but it worked here....

Here we are! I super appreciate it...so does my wife image missing /media/kunena/emoticons/laughing.png Ive been trying to make this work all day

I once set a digital input in my probing sequence that was and2'd to the probe signal to stop this.
I think a better way would be to use motion.motion-type = 5 to signify probing was in use and disable the signal if this is false.

andrax wrote:

and even professionals sometimes find it frustrating.

Sometimes????? More often than that! image missing /media/kunena/emoticons/w00t.png