Ver 2.9.4 iso clean install Trouble when disabling EOE

Those latency values should work, perfectly normal for linuxcnc.

What I see is in your syslogThe slaves were not able to synchronize their time in 5 sec.
I have had that problem myself and didn't solve it, it just went away one day.
The ethercat master sends a synchronize request every 1 msec, so 5000 requests are sent.
What I understand is that this initial synchronization is a slow process and sometimes 5000 requests is not enough.
What I read is that TwinCAT can send tens of thousands of synchronization requests when needed.

As an experiment you can increase this value to 50000, 10 times more, why hold back when experimenting?
It is this value in ethercat/master/fsm_slave_config.cChange toCompile the ethercat master, install it and make sure the new version runs.
And try again.
Interesting to hear if that works.
 

Thanks! ( I haven't done it yet. )

My progress is...

I found ( in "ethercat slaves" command response ) that one servo drive version was 1.07, and anotehr was 2.09 in 3axis. But I have orderd  them recently at the same time.

Yaskawa support person said, this version is ESI version ( in servo driver ? ), but I haven't send ESI file to slaves as far as I remember.
Their suggestion is to use the same ESI file, and sent me the link of the latest ESI file for Sigma X.

How do I apply the latest ESI file to IgH Master and slaves?
I googled it and ask AI how to do it, but I couldn't get the answer for IgH master.

That's right. Linuxcnc doesn't use an ESI file.
The ethercat master reads out the info it needs from the slave directly.

The safest way to update the SII eeprom contents of the drive is to use TwinCAT.
There are a number of steps needed, but in principle not difficult.
Download and install TwinCAT from Beckhoff it you don't already have it,
Place the ESI file you got from support in C:\TwinCAT\3.1\Config\Io\EtherCAT
or where your twincat installation is.
In this video I update the firmware on a device, the principle is the same for your drive.
Go to 5:20 for the relevant part.

 

I think that for Yaskawa, the safest approach is to use Yaskawa’s own tool (SigmaWin+), since it knows the hardware and the drive’s constraints, and it’s also the approach their support can reproduce and help with if something goes wrong.

Thank you so much,  This video is helpful.

And I counld not find ESI upload function in SigmaWin+.
I agree that using SigmaWin would be better if I could use it to do that.

And I think .. sometimes updating new ESI to older software version device is a little bit risky..
I'm considering a little bit about it.

Anyways, thank you so much you guys.

tar_san wrote:

And I think .. sometimes updating new ESI to older software version device is a little bit risky..
I'm considering a little bit about it.

Yes, the ESI file must agree with the firmware in the drive.
I thought about warning for that, but since you were in contact
with Yaskawa support I thought they must know, they should know.

Thanks a lot !

They also mentioned, "if you put this newest ESI , some of this problem will fixed."

( But LinuxCNC (IgH master) desn't have this function. And Yaskawa tools doesn't have upload ESI function to drive as default.)

They said also, this ESI file contains fixture for all versions of firmware.

In this case, do you think It is possible for TwinCAT to fix default ESI in ServoDrive , to make LinuxCNC work ?

I know I have to do it with my own risk and I may not need to do it if I found another solutions.

But I'm happy if I could get some hints for it...

You can always test the new version. It is actually very easy to make a backup
of the eeprom contents and later put it back if needed.We need TwinCAT to do the translation from ESI file to eeprom binary contents.
While there are some open-source tools for that too, the safest way is to use TwinCAT.