Synchronization Using Distributed Clocks - WAGO I/O-SYSTEM 750 Manual

WAGO-I/O-SYSTEM 750
750-354/000-001 EtherCAT® Fieldbus Coupler, ID Switch
10.1.3.5

Synchronization Using Distributed Clocks

For spatially distributed processes, in which actions are temporally calibrated with
each other or have to be simultaneously executed, such as for the coordinated
movements of servo axles, it is extremely important that the synchronization
works. However, a fully synchronous communication in a network always has the
disadvantage that the quality of the synchronization is immediately degraded as
soon as disruptions occur in the communication.
In contrast, EtherCAT
to a main clock for synchronization.
The time on the main clock is transmitted to the distributed clocks via EtherCAT
for this purpose. This main clock is located in an EtherCAT
special hardware is required in the master.
Based on the logical ring structure for EtherCAT
can detect the run time offset in the slave clocks easily and exactly, and readjust
the resulting correction values to a run time compensation that corresponds to
this.
By this means, a highly exact network-wide time basis is available with a jitter
that is significantly less than a microsecond. Thus, for example, a deviation of
± 20 ns can be achieved in the case of 300 participants and a cable length of
120 m.
By using distributed clocks, the communication system is not sensitive to possible
delays caused by disruptions.
This type of synchronization is described in the new IEEE Standard 1588.
There are, in addition, more advantages for high-resolution clocks, namely, the
supply of exact information for the local time of data acquisition. Thus, controls
from consecutively measured positions can, for example, calculate speeds.
EtherCAT
addition to the timestamps transmitted in addition to the reference data. Using the
measured value, the local time is connected to a resolution of up to 10 ns. Thus,
there is no longer a dependence on the communication system's jitter.
Another expanded data type is the Oversampling Data Type.
The Oversampling Data Type allows multiple sampling of a process datum within
a communication cycle. The oversampling factor indicates the number of
samplings within a communication cycle. Therefore, sampling rates of 200 kHz
are easily possible.
The data are transmitted in an array (outputs from the previous transmission,
inputs from the subsequent transmission).
The sampling is triggered by the local clock and/or the system time.

uses distributed clocks, which can be exactly compared

has an expanded Timestamp data type for these timestamps in
Fieldbus Communication 147

slave. Therefore, no

communication, the main clock
Version 1.2.1

Manual