6. Redundancy with NX3030 CPU
271
Redundancy Control Panel
PX2612
- Through buttons, allows commands of switchover or redundancy
states transition for maintenance;
- LEDs signalize the redundancy state in each half-cluster;
- NO relay allows a half-cluster to switch off the other in extreme
situations. A button allows the other half-cluster reactivating;
- Embedded resources for buttons, LEDs and relays tests.
- Indicate failures in the PLCA and in the PLCB, independent of
their states (Active or Inactive);
- Prevent “obscure failures”;
- Allow quick maintenance, essential for high availability.
- Allow the execution of the same PX2612 control panel actions,
among other commands (e.g. switchover command).
- Can be executed in the local CPU, or transmitted to the other
CPU (remote) via synchronism channels NETA/NETB;
- Can be received through MasterTool or a SCADA system;
- Can be executed through user application.
- Register diagnostics and redundancy commands changes, with
timestamp, allowing an investigation of the switchover causes.
SNTP (Simple Network Time
Protocol)
Allow the events to have a precise timestamp adjusted to the
world hour. It also synchronizes the CPU real time clock for other
applications.
Commands and diagnostics
synchronization
Each MainTask cycle, PLCA and PLCB exchange diagnostics and
commands through synchronism channels NETA or NETB. This
way, a CPU knows the other diagnostics and commands.
Redundant data
synchronization
Each MainTask cycle, the Active CPU copies redundant data to
the Inactive CPU through the synchronism channels NETA and
NETB. Non-redundant data are not synchronized.
Redundant forcing list
synchronization
Each MainTask cycle, the Active CPU copies the redundant
forcing list to the Inactive CPU through the synchronism channels
NETA and NETB. This list includes only forced redundant
variables, this way PLCA and PLCB can have different non-
redundant data groups forced, as these variables are not
synchronized.
Single project for PLCA and
PLCB
There’s a single project for the PLCA and PLCB, generated by
MasterTool. The project is composed by the applicative project
(executable code) and the archive project (source code).
Through MasterTool, a NX3030 CPU identifies itself as PLCA,
PLCB or non-redundant CPU. This identification isn’t part of the
applicative project generated by MasterTool, even though is
written in a CPU using MasterTool. The CPU identification allows
the feature of a single project for PLCA and PLCB.
Automatic project
synchronization
If the Active CPU project becomes different from the Inactive
CPU, it is copied from the first to the second. This synchronization
can take several MainTask cycles. One must remember the
project is composed by the applicative project (executable code)
and the archive project (source code), and both are synchronized.
This synchronization can be disabled in special cases in order to
allow visualization of project modifications which can only be
downloaded off-line in non-redundant CPUs.
On-line expansion of modules
and PROFIBUS remotes
There are project modifications that can’t be done on-line in a
non-redundant CPU, such as the inclusion of new modules or
PROFIBUS remotes.
However, using the CPU and the PROFIBUS network
redundancy, it was defined a procedure to accomplish this goal,
very important for systems which need high availability.
Private IP addresses for PLCA
and PLCB
It’s possible to connect to a specific CPU (PLCA or PLCB) using a
private IP address, to obtain half-cluster specific diagnostics, for
instance. The PLCA IP address will always be associated to the
PLCA NET(i) interface, while the PLCB IP address will always be
associated to the PLCB NET(i) interface..
Name of a strategy that allows the Ethernet client connect to a
server from the redundant CPU using always the same IP
address. This prevents the necessity of complex scripts to change
the IP address when switchovers occur due to redundancy. The
Active IP address will always be associated to the NET(i) interface
from the Active CPU.
Name of the strategy which allows two Ethernet interfaces from a
half-cluster to form a redundant pair sharing a same IP address.
To Next Page
Loading...
General