Response Electronics CL6011B Specifikace Strana 77

  • Stažení
  • Přidat do mých příruček
  • Tisk
  • Strana
    / 320
  • Tabulka s obsahem
  • KNIHY
  • Hodnocené. / 5. Na základě hodnocení zákazníků
Zobrazit stránku 76
3Ć57 Theory of Operation: Redundancy
Original Ć June 1990 CE4.2:CL6211
implemented at the DCU level. The redundant controllers perform simply
as slaves to the RDCU in a redundant switchover decision. The following
paragraph(s) briefly explain how this failure detection mechanism is
implemented.
The Redundant Data Concentrator Unit (RDCU) is designed to
periodically poll the ACTIVE controller for status information and
Operating Data, and at the same time periodically interrogate the
STANDBY controller for status information. Any failure critical to the
control or operator interface, detected at the ACTIVE RDCU, will cause a
switchover to occur. These conditions include:
H Primary controller response message timeout
H Arithmetic Processor error (Interactive controller only)
H Input/Output self-test error (includes calibration out of tolerance)
H Communications error rate excessive
Although multiple failures are not meant to be covered in all of their
potential combinations, there is a limited set of optimization switchover"
decisions that are incorporated into the RDCU decision mechanism. For
instance, if a communications timeout occurs on a primary controller
communications channel, the RDCU will switch even if the secondary
controller system is currently reporting a calibration error. The premise
here is that some control is better than no control at all.
Note however that a switchover would not occur if a communications
timeout occurs on the primary controller system, when another
communications failure is being reported on the secondary controller
system, because the RDCU cannot determine which would be a more
optimum" control system under the failure conditions reported. Also,
optimization cannot be performed when multiple communications fails on
a controller subsystem (example, two controller timeouts occurring on
the primary system, and only one on the secondary system), as there is
no voting mechanism to determine the highest availability of the control
system.
The RDCU will only make one switchover (or switchback) decision, and
then the automatic switch decision algorithm is disabled. Automatic
switch enable/disable status is reported on the RDCU diagnostic
displays. Disabling automatic switch will prevent any possibility that the
RDCU may cycle between redundant units when the system is
experiencing intermittent failures, such as power cycling problems, etc.
Once automatic switch is disabled, the operator must intervene to
correct the error condition, and then enable the automatic switch
decision algorithm. Any integrity error, including disabled automatic
switch, is reported as an integrity problem with the RDCU, requiring
operator attention.
Zobrazit stránku 76
1 2 ... 72 73 74 75 76 77 78 79 80 81 82 ... 319 320

Komentáře k této Příručce

Žádné komentáře