Ethercat random jitter fix

Better yet, enable synchronization with syncToRefClock="true"
github.com/linuxcnc-ethercat/linuxcnc-et...s.md#master-settings

The refClockSyncCycles is how often to synchronize the clocks between the slaves
which is something different altogether, and doesn't help this problem.

I currently have refClockSyncCycles="-1" which according to the docs you linked above, means its enabled for backwards compatibility. Sadly, I have these mentioned issues where the axis starts clunking randomly (on a random axis) after some time. Its a fresh install from end of january. Does SyncToRefClock still need to be set?

There was a bug with "-1", fixed now but it sounds like you have it.
Use a positive value in refClockSyncCycles and set syncToRefClock="true"
and that should hopefully fix it.

Now as in just recently or in the patch from last year? or reaaally recently
Also, what does the refClockSyncCycles value do and is there any baseline on what to set it at?

Dec 27 it looks like from github. Actually, the bug shouldn't affect you.
A negative refClockSyncCycles value should switch on the PLL and synchronize servo loop time with DC time.
You can always check the lcec.pll-* hal variables. If some of them move, it's on.
refClockSyncCycles is how often the master sends a synchronization signal to the slaves to compensate for clock drift.
I use 1000 because the slaves' DC clock run pretty accurate as it is. Many people use 1 as value.
I have not seen that any number is better than the other, maybe there is, but feel free to use any number you prefer.



 

I could share my experience with this issue. Attached is a link to a video showing how the jitter sounds on StepperOnline A6 servos. This is exaggerated by disabling the synchronization. Once synchronization is enabled (syncToRefClock="true" or refClockSyncCycles="-1"), it still happens but much more rarely, just like grandixximo explained at the start of this post. Last week, I upgraded to the latest version of linuxcnc-ethercat, and am currently testing how the changes committed at the end of 2025 affect my machine configuration.

​​​https://imgur.com/a/FXSuvOS

After activation of PLL there shouldn't be any random clicks anymore.
Check dmesg if you get any errors there, it can be the network adapter too.

andrax wrote:

Okay, thank you.

I had already thought that the issue had something to do with packet loss when starting LinuxCNC.
However, this seems to occur during operation, which I have not yet experienced. If this work results in ethercat running more stably, I will simply wait until Sascha announces an official patch, which will likely be available via update.
Until then, I will simply stick with refClockSyncCycles="1">.
 

An official release is still a few months out, the details are been worked out, still discussing and will soon get to testing on real hardware, the objective is indeed running ethercat more reliably, the current release is non-deterministic, there is nothing you can do as a user to completely get rid of the jitter noise if you have non zero CPU latency, we will hopefully find an agreeable solution for all parties, some XML settings may require changing, will try to document it all, the problems are known, now is just about finding a solution from which all can benefit.

TAKUYA wrote:

Now as in just recently or in the patch from last year? or reaaally recently
Also, what does the refClockSyncCycles value do and is there any baseline on what to set it at?
 

github.com/linuxcnc-ethercat/linuxcnc-et...84ef36b53f0b734eaedf
the above is my patch, which I tested on real hardware, both 
refClockSyncCycles="5"
or any other positive number, this will have linuxcnc ask the slaves to sync to it every 5 cycles, put in whatever number you need to get phased in the time frame you desire, there is a dc-phased pin which tells you when you are phased, you can link that with iocontrol.0.enable-out so that you won't be able to turn on the machine till the distributed clock is properly phased.
also working
refClockSyncCycles="-1" 
or
syncToRefClock="true"
This is working with linuxcnc drifting its own phase to sync to the slaves, this is done every cycle, it is slightly more CPU intensive, as it is happening every cycle, but will phase pretty quick and works just as well in my testing.

github.com/sittner/linuxcnc-ethercat
if you are willing to do some testing of what we are building towards, you can test Sasha's current release

github.com/sittner/linuxcnc-ethercat/tre...2ea38d79aa47a58b89f7

This I believe only works well with
refClockSyncCycles="-1" 
at the moment, more testers would be appreciated, I will do my own testing of this soon.
 

Atsu wrote:

I could share my experience with this issue. Attached is a link to a video showing how the jitter sounds on StepperOnline A6 servos. This is exaggerated by disabling the synchronization. Once synchronization is enabled (syncToRefClock="true" or refClockSyncCycles="-1"), it still happens but much more rarely, just like grandixximo explained at the start of this post. Last week, I upgraded to the latest version of linuxcnc-ethercat, and am currently testing how the changes committed at the end of 2025 affect my machine configuration.

​​​https://imgur.com/a/FXSuvOS
 

Thank you for the independent testing, if you want a better version

github.com/linuxcnc-ethercat/linuxcnc-et...84ef36b53f0b734eaedf

this I have tested and works well, unfortunately Scott has not been around to merge it, so it is not in the main linuxcnc-ethercat repo yet, but it is a better version