Ethercat retrofit question

to get kernel version type "uname -a" in Terminal.

Here's the output of uname :
After configuring the servos as per the pdos command output they all seem to work . Today I'll be working on the operator panel, for which I found an excellent manual . And if all goes well continue with the spindle .

db1981 wrote:

So i created a way to get etherlab master working without the readout content of the SII EEprom. Now you can load a bin file (that has been created from esi xml) at bus startup, i think this will help a lot in the future.

Could you explain a bit more on how to do that ?
I just wrote to the manufacturer about an xml for the spindle drive .

for the moment, coarse explanation:

-modify the ec-master

fast method: i've putted deb files for 4.19.0-17 rt there : github.com/dbraun1981/lcec_ec_master/tre...preempt_4.19.0-17-rt
remove the installed version before.

-check if the master works as before 
-create an ethercat folder in your os default firmware dir.  (/lib/firmware/ethercat)
-if you have an working beckhoff configuration, you can create an bin file for SII Rewrite . 
  Ethercat Tab of the device, extended ->  ESC -> Hex Editor (shows the virtual SII Data, depending on the configuration) -> write to file

-rename the file to  ec_xxxxxxxx_yyyyyyy.bin  x= 8 digits Vendor ID, y = 8 digits Product ID  (example: ec_000001dd_10400200.bin = Delta MS300 Vfd)  
-place this file in the firmware ethercat folder.
-reboot or restart ethercat.

Now the master checks at bus restart if a bin file for the device is present and uses the bin content, if no file is present the SII Content is readed like without the patch.

I'll have a chance to try that saturday .
From what I see in the machine xml it sends a few commands like these :
  And if I'm right this corresponds to the line :
ethercat download -p8 0x1A00 10004160 That just seems like 6041 register with 16 bit length .
So if things don't go as planned can't I just write a simple script with all these lines and launch linuxcnc after them ?

This are initcmds they will be sent via mailbox communication at preop state, before you will change the file state to safeop -> op.
Them are for configuring the slave like pdo mapping or init values.

You can sent these commands with lcecconf. One is to use the sdoconfig command in the ethercat_conf or by putting them to an extra xml file and load this as initcmd file in the ethercat_conf file. there are examples on linuxcnc-ethercat github.

If this is the slave with the faulty eeprom read from the first post. I think this will not help without repairing the eeprom or loading a bin file.

I am sure the drive is fine . I had it running yesterday with the old cnc . After a bit of research I think uses an acontis EC master .
I also checked out the examples you mentioned . Is there a manual for this driver ?

OK so still no luck with the spindle .
At least we have the drive software now . And it really is programmed to accept the pdos . Bad thing is the software cannot generate a bin file .  still haven't figured out a few bits of the original ENI xml from the old master . This is just a part of it .

Some of these lines really look like the answer . Although what should I put for idx and subidx ?
Any advice is welcomed .

the commands you have posted adp, ado, are all working at the lowest layer of the ethercat communication pre Mailbox communication. Them are for setting up the base things like syncmanager, mailbox com etc.

This base config stuff the etherlabmaster only reads from the eeprom, exept you go the way with the bin file as descripted above or rewrite the eeprom (risk) .
Other Master like Beckhoff or Arcontis can us them by reading the from the xml.

If you can post an ESI XML that belongs to the drive, i can check if the device can get opend with beckhoff.

Is the Spindle driver actually EtherCAT or like in the manual, CANopen?