LinuxCNC S-Curve Accelerations

Grotius wrote:

Hi,

Can we test the cia-402 and shortcut linuxcnc pos-cmd to pos-feedback so
there is no ferror in linuxcnc that can trigger a motion stop? 
Or test it by hooking the cia-402 to a stepgen or pid without closed loop feedback. 
 

The only time this is necessary is during internal homing (if used). During homing, motion is happening that Linuxcnc does not know about so it generates a following error if this is not done.
In the drive, motion is measured in encoder pulses and in Linuxcnc uses device units
One of the other  things the cia402 component does is scale the encoder pulses to Linuxcnc device units (mm or inch)

It depends on the drive but many of them will have a SDO that tells the drive to use open loop or closed loop. There are some reasonably cost effective open loop Ethercat cia402 drives from Rtelligent and Leadshine. Both offer drives that can control two open loop stepper motors. The Rtelligent ECR60X2A is USD $60 each ex works (for 2 steppers). 

wrote:

 
hi
in my last test there was no following error on gcode running, the problem was the very low acceleration.
following errors appears on homing procedure

cia402 defines pins to set velocity and acceleration while homing. 0x6099 and 0x609a

Tried this out but got following errors with large acceleration violations

Running the LinuxCNC logo:

 

Hi Pcw,

One last question, if you run the scurve simulation config :  axis_scurve_mm.ini   using the startup menu.
Run's this ok for you?

I can not test that as it requires EtherCat hardware.

Hi Pcw,

The referred config file is a simulation for use with no hardware. 
I thought if this runs ok on your pc, maybe you have a idea why it will not run ok on real hardware.

i have it running with huge ferror values.