LF364001_MNL - Overfill Protection SIS Design Manual

50 / 443

Local Fuel – The Shoreham Oil Terminal
Gasoline Import - Safety Instrument System
P & I Design Ltd DOCUMENT NO: LF364006_RPT
2 Reed Street, Thornaby, UK, TS17 7AF ISSUE: C DATE: 13.05.16
Tel: + 44 (0)1642 617444 PAGE 10 OF 23
Fax: + 44 (0)1642 616447
6.2 Logic Solver
6.2.1 Logic Solver Function
The Drexelbrook RF Admittance level switch monitors tank level and on detection of the
floating roof reaching a set high level the milliamp signal reduces below 8 mA. On a fault the
milliamp signal increases to 22mA. The signal is wired into a trip amplifier and isolator which
opens the input circuit of the safety relay on high level or a fault condition. The safety relay
continually monitors the status of the input. Upon open circuit the safety relay de-energises its
safety contacts, hence removing the signal to the final element.
Once the level has fallen below the set high level, the Drexelbrook switch will automatically
give out its 16mA signal. However, in order to reset the safety relay it is necessary that the
operation of the manual reset facility will cause the safety relay to re-energise its outputs,
providing a signal to the valve.
The Drexelbrook level switches are intrinsically safe devices and as such have combined IS
isolators/trip amplifiers installed in the safe area.
The trip amplifier will be set to switch at 15mA falling and 17 mA rising. The 4-20 mA
signal from the level switch will also be monitored via a PLC (future).
The Exida report specifies 3 setpoints to be monitored. The design only uses a single setpoint
(as above) as anything outside this range will cause the system to activate.
The inputs and outputs have been detailed in Section 5.
There are no Input or Output overrides to the system and reset is manual.
6.2.2 Trip Amplifiers/IS Isolators
The combined trip amplifiers and IS supply isolators are manufactured by Pepperl and Fuchs.
Tag Numbers LS01, LS02, LS03, LS04, LS05, LS06, LS07, LS08, LS09, LS10, LS11, LS18,
LS18A, LS21 & LS22
The manufacturer has provided failure rate data for this device. The data supplied by Pepperl
and Fuchs is contained in the Exida document Report Number FS-0013PF-20C Version V2
dated January 17th 2011.
The date is summarised as follows
λSD = 9 x 10 -9 hours
λSU = 347 x 10-9 hours
λDU = 90 x 10 -9 hours
λDD = 89 x 10 -9 hours
PFd calculated is 3.98 x 10-4 based on a mean time to repair of 24 hours and a proof test interval
of 1 year.
For supporting documentation see Appendix 2