background image

4

Controller

A single-board controller is the heart of the system. It includes the main processor and three Ethernet drivers for
communications with networked I/O and two additional Ethernet drivers for the control network. One, two, or three
controllers are provided as required for redundancy.

Some suppliers support redundant processors and communication boards in a common rack with a split backplane for
improved fault tolerance. Another form of redundancy has two controllers with two processors in each of the two controllers
in a quad configuration. If one of the processors has a partial failure, there will be a discrepancy between the data from the
two processors on one board, and the other board takes control.

A key evaluation point for any redundant control system is the failover time from one controller to the other. If one controller
normally drives the control valve and the other is on standby in a hot-backup configuration, then there is a finite time for the
backup to determine that the main controller has failed so that it can take over. This failover time is critical to the process.

Another method is to have both controllers continuously reading inputs, running application software, and providing outputs
to the control valves and relays, so there is no failover time between controllers. Somewhere between the controllers and the
control valves / relays a decision is made to follow the commands from one controller or the other. This decision point is a
critical item in determining failover time, failure mode, and overall system fault tolerance.

5

I/O Network Redundancy

All control systems have internal communications between the main processor(s) and the I/O regardless of whether the I/O is
separate from the controller board rack, or mounted inside the rack with communication on the backplane. The I/O network
consists of active electronics at both ends and multiple failure modes, so its redundancy is just as important as the main
processors and the I/O electronics that interface with field devices.

In a Mark VIe control system, IONET provides communication between the main processor(s) in the controller(s) and the
local processors in the I/O packs that are located on the I/O modules. This communication architecture is a star configuration
with the network switch(s) in the middle. Switches manage communication traffic to eliminate data collisions and increase
network determinism. Networks conform to IEEE 802.3 for 100Base-Tx and 100Base-Fx (fiber).

Typically, simplex controls have one IONet, dual controls have two IONets, and triple redundant controls have three IONets.
However, even in simplex controls a second IONet is often provided.

Switches send their input data to all controllers, which are continuously online. Each dual redundant controller uses the data
from its designated switch, but uses the data from the other switch, with no delay, if the data from the first switch is not
received or if the data has a bad checksum. Each triple redundant controller receives data from all three switches, individually
votes the data from contact inputs, and selects the median value of analog inputs. In addition, diagnostics identify any
discrepancy between the three inputs. This is important to minimize MTTR and enable online repair.

Output data is sent from each controller to its designated switch and then to the output electronics. The methodology for
selection of output data from the redundant controllers for driving control valves, solenoids, and other components varies
widely between control systems and is significant for determining the system reliability. These details are discussed in the
next section.

Instruction Guide

GEI-100728A

7

Public Information

Summary of Contents for Mark VIe

Page 1: ...nt contains non sensitive information approved for public disclosure GE may have patents or pending patent applications covering subject matter in this document The furnishing of this document does no...

Page 2: ...Controller 7 5 I O Network Redundancy 7 6 I O Redundancy 8 6 1 Dual Redundancy 8 6 2 Triple Modular Redundancy 11 7 Tripping Reliability 15 8 Digital Bus Reliability 17 8 1 FOUNDATION Fieldbus 17 8 2...

Page 3: ...ting the process Field components for example sensors actuators and wiring cause over half of forced outages Therefore redundancy of field components is an important consideration in the overall contr...

Page 4: ...one component is out voted by the other two In the case of three lube oil pressure switches the protective system performs a simple logical vote with no need to predict in advance a probable failure m...

Page 5: ...ing Blocks Controller 3 IONET Ports 2 Control Network Ports 1 COM Port IONET Switch I O Module I O Pack Local Processor Data Acquisition Terminal Block Typical Mark VIe Architecture Instruction Guide...

Page 6: ...ther the controllers are redundant or non redundant But they can be supplied in redundant pairs too if required Additional redundancy options are available Internal Power Converters Create 28V dc for...

Page 7: ...board rack or mounted inside the rack with communication on the backplane The I O network consists of active electronics at both ends and multiple failure modes so its redundancy is just as important...

Page 8: ...at the I O pack level Therefore replacement of I O has minimal impact on monitoring and control of the overall control system An extension of this is to add a third sensor whose value can be voted in...

Page 9: ...ovides recovery data 4 Receives external commands 5 Creates process alarms Data outputs from dual redundant controllers are normally implemented with each controller sending its signal to its switch a...

Page 10: ...ollers IONET switches and Ethernet ports on a common I O pack which controls a relay driver and a relay Level 2 also provides dual redundant controllers and IONET switches but extends the redundancy t...

Page 11: ...ed or transmitted individually Fanned inputs are transmitted on the IONET with three I O packs on the I O module Therefore a failure of an I O pack does not inhibit any controller from seeing all of t...

Page 12: ...les 3 Less IONET switches Non critical data that is being used for non essential monitoring is usually implemented without redundancy Redundant and non redundant I O coexists in most control systems T...

Page 13: ...nt sharing circuit Analog Voting occurs with a3 coil servo Extended Voting at Field Device Example 3 Coil Servo Valve Actuator Typical nuclear configuration Valve Regulator Valve Ref Valve Regulator V...

Page 14: ...Requirements for no single point failures must be evaluated on a case by case basis to determine the best way to approach this from the system level The preceding figure displays a variation of outpu...

Page 15: ...responding I O control protect and monitor the turbine Primary protection includes a full set of all trip functions and backup protection includes a small subset of the protection functions to backup...

Page 16: ...control valves and de energize the trip solenoids from the primary side Another example the backup protection monitors communications from each controller so it can be configured to initiate a trip o...

Page 17: ...the designated controller and a secondary controller Therefore the primary linking device is the one connected to the primary controller Less common are applications with a single controller and redun...

Page 18: ...al Controllers Supplies Switches Simplex Non redundant Dual with Redundant I O Full Triple Redundant Dual Redundant Triple Relative Contributions to Reliability and MTBFO Adding redundancy to the elec...

Page 19: ...Public Information...

Reviews: