Emerson SI-EtherCAT, Руководство пользователя

Страница: 78 / 106

78
SI-EtherCAT User Guide
Issue Number: 2
8 Advanced features
8.1 Distributed clocks
SI-EtherCAT supports Distributed clocks. This is the scheme used by EtherCAT to accurately time
synchronize slave devices. Position, speed and current control loops can all be synchronized.
When the option module is connected to a drive which can take a time synchronization signal (e.g.
a Unidrive M600 or above), the EtherCAT Distributed Clocks facility can be used to provide this
signal so the drive speed and current tasks are synchronized to the network. The position
controller, and appropriate motion features will also be synchronized to the drive speed task.
8.1.1 Time synchronization support
When the option module is connected to a drive which can accept a time synchronisation signal
(e.g. on Unidrive M600 and above), the EtherCAT distributed clocks facility can be used to provide
this signal so that the drive tasks (including the motion, speed and current tasks) are synchronised
to the network. On Unidrive M600 and above, the option synchronous task will also be
synchronized to the drive OPT_SYNC signal; on Unidrive M200 to M400 it will be executed every 5
ms, asynchronous with the drive control loops (i.e. there is no synchronization).
If possible, the option will provide a synchronization signal suitable for the longest interval drive
cyclic task (this will also, of course, synchronize the higher rate tasks, if they are phase locked to
the lowest rate task).
The drive motion and speed loops are executed every 250 µs on Unidrive M600 and above, which
coincides with OPT_SYNC edges. This will be referred to as the control loop cycle. Synchronization
must be enabled if the Cyclic Sync or Interpolated Position modes are used; an error will be
indicated if this is not the case. These modes have an operating mode cycle time of the
interpolation cycle time; other modes will have an operating mode cycle time matching the
synchronous task (250 µs or 5 ms, depending on drive type)
Any operating mode cycle will be restarted every operating mode cycle time, in phase with the
synchronisation events; if synchronization is enabled, the operating mode will not start execution
until the first synchronization event occurs. If synchronization is lost, an error will be indicated, and
the standard EtherCAT action for this event will occur.
Command and feedback values which are handled cyclically will be read at defined times in the
cycle. Command values handled/used every cycle (operating mode or control loop) will be cached
from the object dictionary in the task immediately before the drive critical update period. Any
feedback values read during a cycle will be scaled as appropriate in that cycle, cached, and then
written to the object dictionary in the task occurring immediately after the drive critical update
period. Feedback values that change internally between control loop cycles (but whose objects are
only updated every profile cycle) will be read from the last control loop cycle in the operating mode
cycle. PDO data will be copied to and from the object dictionary (from and to the sync manager
memory areas) in the drive critical update period at the beginning of every operating mode cycle.
PDO data mapped to drive parameters (but not parameters accessed using Inter-Option
Communications, or eCMP), will be written to those parameters in the critical update period at the
beginning of every control loop cycle. This behaviour can be modified by the advanced cyclic data
configuration objects.
In CoE interpolated position mode the position command provided by the master every
interpolation cycle time is used to generate a position command for the drive every 250
µs.
NOTE