Risk Software S.A. de C.V. Determinación de PFDavg (SIL) de un Sistema Instrumentado de Seguridad (SIS) Preparado para: Curso en Análisis de Riesgos y Seguridad Funcional Preparado por: Victor Machiavelo Salinas Risk Software SA de CV www.risksoftware.com.mx Risk Software S.A. de C.V. 1. Introducción El valor de PFDavg (Probabilidad de Fallas Sobre Demanda Promedio) es utilizado en la Seguridad Funcional para determinar el Nivel de Integridad de Seguridad -NIL- (Safety Integrity Level- SIL) que un Sistema Instrumentado de Seguridad -SIS- tiene para una Función Instrumentada de Seguridad -FIS- dada. La figura #1 nos muestra la relación que guarda un Sistema Instrumentado de Seguridad entre la relación (frecuencia) de demandas (eventos/año) en que el SIS es requerido por el proceso dada una condición insegura y la relación (frecuencia) de eventos indeseados finales (eventos/año) ocurridos dados la ineficiencia/falla/incapacidad, del SIS. Relación de Demandas (D) SIS Relación de Eventos (H) Figura #1 PFDavg = H/D = 1/(Factor de Reducción de Riesgos) El nivel NIL/SIL, es una relación del valor numérico calculado de PFDavg para un SIS, donde incluimos a los elementos sensores (presión, temperatura, Flujo, etc), al controlador lógico programable y a los elementos finales de control (válvulas, motores, actuadores, etc). El valor de la PFDavg Total para un SIS es la suma algebraica de la probabilidad de fallas sobre demanda promedio del sensor mas la del controlador lógico mas la del elemento final de control como se muestra en la figura #2 Controlador Logico Sensor Elementos Finales Figura #2 PFDavg Total = PFDS + PFDL + PFDEF para realizar el calculo de la PFDavg de un sistema SIS, el estándar ANSI/ISA 84.01-2004 recomienda tres métodos: 1. Ecuaciones Simplificadas (Diagramas de Bloques de Confiabilidad) 2. Análisis de Arboles de Falla (FTA) 3. Modelos de Markov. El presente informe técnico se centra en el calculo de la PFDavg, utilizando los dos primeros métodos, los cuales son los mas utilizados en la seguridad funcional, aclarando que los modelos de Markov son mas precisos y pueden modelar sistemas en el tiempo, con secuencias y reparables. Determinación de la PFDavg 1 Risk Software S.A. de C.V. 2. Falla de los Sistemas Es necesario comprender la forma en que los sistemas y equipos fallan, debido a que las ecuaciones utilizadas para determinar el valor de PFDavg depende directamente del mecanismo de falla de los sensores, controlador lógico y elementos finales. La figura #3 muestra los modos de falla que pueden tener los componentes de un SIS. Modos de Falla Fallas Descubiertas Relación de Paros en Falso Fallas Cubiertas Relación de Paros Peligrosos λS = 1/MTBFsp λD = 1/MTTF Se debe vivir con perdida de la producción Paro de Planta o Permanecer en Riesgo Mientras se Repara Detectadas Por Diagnosticos No Detectadas Por Pruebas manuales El SIS esta Fuera Durante las Pruebas Figura #3 Modos de Falla MTBF = Mean Time Between Failures (Tiempo Medio Entre Fallas) MTTF = Mean Time To Fail (Tiempo medio Para Fallar) Modos de Falla Descubiertas: Son conocidas también como fallas “Reveladas” debido a que estas fallas son conocidas en cuanto suceden, como ejemplo tenemos la falla de la señal de un sensor cuando los cables que conducen la señal son cortados o bien la falla de la bobina de una válvula solenoide. Las fallas descubiertas normalmente generan una respuesta del sistema conocida como “Falla Segura” la consecuencia mas común es una parada por emergencia del proceso. A esto se le conoce como “Relación de Disparos en Falso” en muchos procesos esta condición es indeseada debido a que afecta directamente a la producción o a los tiempos de producción, en procesos continuos como en la industria química o petrolera esta condición es muy costosa debido a que volver a iniciar los procesos no es una tarea fácil ni rápida, en ciertos procesos esta condición también puede ser muy peligrosa, ya que parar proceso inherentemente peligrosos donde se manejan grandes cantidades de materia y energía puede ocasionar condiciones riesgosas para el personal, medio ambiente y bienes de las empresas. La forma en que podemos evitar que esto ocurra es incrementando la tolerancia a falla en los sistemas y equipos (redundancia). La norma IEC-61511 en el punto 11.4 nos indica los mecanismos y niveles de tolerancia a falla para los sistemas SIS. Determinación de la PFDavg 2 Risk Software S.A. de C.V. Modos de Falla Cubiertas: Las fallas cubiertas, son fallas peligrosas hasta que son detectadas y corregidas. El calculo de la PFDavg se basa en este tipo de fallas. Típicamente las fallas cubiertas se manifiestas en dispositivos que tienen la función de generar o conducir al evento final, como pueden ser los dispositivos de salida de las tarjetas del PLC, la bobina del relevador, el actuador de la válvula o bien la lógica del controlador. El problema principal de estas fallas se presenta en dispositivos que no han sido operados por periodos lagos de tiempo, tres tipos de condiciones se presentan en las fallas cubiertas: 1. Fallas que pueden ser detectadas por auto diagnósticos. 2. Fallas que pueden ser encontradas en un periodo de pruebas. 3. Fallas que permanecen ocultas sin ser detectadas en el sistema hasta que se presenta una falla en demanda. Cada una de estas fallas contribuyen al valor de PFDavg del SIS. Cada falla requiere un tratamiento diferente de calculo de confiabilidad. Las formulas para el calculo de sistemas basados en Auto diagnósticos, están generalmente referidas a controladores lógicos programables ya que estos sistemas utilizan técnicas avanzadas de diagnósticos, en la mayoría de los sistemas cuando nos referimos a “diagnósticos” no estamos refiriendo a la capacidad del sistema a realizar pruebas sin necesidad de intervención del ser humano, estos diagnósticos que también son referidos como “activos” son pruebas funcionales del estado del sistema, como por ejemplo seria cambiar de estado la posición de las salidas de las tarjetas del controlador abrir/cerrar (On/Off) para poder probar que el sistema tiene la capacidad de llevar al proceso a condición segura. Estas pruebas se realizan de forma muy rápida generalmente en milisegundos, evitando que las pruebas sean en si mismas una condición peligrosa para el proceso. Cálculos: El calculo de las fallas reveladas (llamadas también fallas seguras) es importante desde el punto de vista de la operación de los procesos, la instalación de un sistema de seguridad es un proceso complicado y costoso, lo que menos deseamos es que este sistema sea en si mismo quien genere una condición potencialmente inseguro o binen sea quien ocasiona perdidas de producción o económicas. La selección de un sistema de seguridad sin tolerancia a fallas deberá ser cuidadosamente evaluada desde el punto de vista de la seguridad y de la operación de los procesos, el diseño del sistema bajo el concepto de ciclo de vida deberá incluir los costos de disparos en falso y los costos asociados a la tolerancia a fallas. las fallas relevadas también tienen dos componentes, fallas seguras detectables y fallas seguras no detectables. El echo de que ambas conduzcan a un paro seguro del proceso minimiza la necesidad de detallar cada una en una ecuación diferente. Las fallas cubiertas (llamadas también peligrosas) como se muestra en la figura # 3 tienen dos componentes, Determinación de la PFDavg 3 Risk Software S.A. de C.V. 1) Fallas peligrosas detectadas por auto diagnósticos, las cuales realizan el proceso de prueba y detección de errores y fallas de forma automática, asociamos a estas fallas a las provocadas por los sistemas complejos como los controladores lógicos, sin embargo en los últimos años algunos dispositivos de campo como sensores y actuadores de válvulas, han incorporado altos niveles de auto diagnostico en su electrónica. Típicamente el tiempo de las pruebas con auto diagnósticos fluctúa entre 1 y 10 segundos. 2) Fallas peligrosas detectadas por pruebas manuales, son pruebas que no pueden ser realizadas por diagnósticos y es necesario que manualmente se realice la prueba y el diagnostico, típicamente el tiempo de estas pruebas es mucho menor que el MTBF, este tipo de pruebas esta asociada a dispositivos de campo y elementos finales de control. La figura #4 muestra la diferencia de pruebas requeridas para los diferentes dispositivos, existe una gran diferencia entre las ecuaciones utilizadas para modelar el valor de PFDavg para sensores y elementos finales de control y las ecuaciones para modelar a los controladores lógicos, no solo por que estos realizan sus pruebas de auto diagnostico, también debido a que cada sistema puede contener diferentes dispositivos en diferentes configuraciones y numero (módulos de entradas y salidas, fuentes de poder, procesadores, comunicaciones, etc). Relación de Demandas (D) Sensor Controlador Logico Elementos Finales Pruebas Manuales Pruebas Auto Diagnosticos Pruebas Manuales Relación de Eventos (H) Figura #4 Requerimientos de Pruebas para Dispositivos Las ecuaciones para modelar a los controladores lógicos programables han sido definidas a detalle en la norma IEC 61508-6.Edición 2.0 2010-04. También se cuentan con ecuaciones simplificadas para los controladores lógicos programables, que hacen mas fácil pero menos exacta la determinación del de la PFDavg. Determinación de la PFDavg 4 " 27 " ISA-TR84.00.02-2002 - Part 2 ISA-TR84.00.02-2002 - Part 2 " 28 " ISA-TR84.00.02-2002 - Part 2 " 27 " (Eq. No. 9) $S = ISA-TR84.00.02-2002 - Part 2 1 MTTF spurious " 28 " ISA-TR84.00.02-2002 - Part 22 ISA-TR84.00.02-2002 Part 28 the " third term is the systematic er The second term is the- common cause term, "and ISA-TR84.00.02-2002 - Part 2 The second -term is2the term, Part " 28 " and the third term is the systema ISA-TR84.00.02-2002 - Part 2 common cause " 27 ISA-TR84.00.02-2002 " S The above equations apply to elements with the same failure rates. If elements with different fa S DD SNOTE (Eq. (Eq.No. No.9)10) $ = STR = $ + $ + $ F ISA-TR84.00.02-2002 - Part 2 (See 28the " with "thethird MTTF spurious The second term is the common cause and is theIfsystematic erro appropriate adjustments must beequations made ISA-TR84.00.02-2002, Partterm 5 for rates. method). NOTE The above apply toterm, elements same failure elements with diffe 1oo1 1 S S S 1 appropriate adjustments must be made (See ISA-TR84.00.02-2002, Part 5 for method). Where $ =is the safe or spurious failure rate for the component, (Eq. No. 9) (Eq. No. 9) $ $ = 1oo1 spurious The second term is the common causewith term, and the third term is the systematic er MTTF spurious NOTE The process above equations apply the taken same failure rates. If elements with different MTTF SIS industry typically must be of service to make repairsfail w S S in the DD - Part 2to elements " 28 "outPart (Eq. No. 10) +DD $+DD +ISA-TR84.00.02-2002 $ dangerous detected rate for theadjustments component, and $ is theSTR F appropriate must be made (See ISA-TR84.00.02-2002, 5 for method). (Eq. No. 10) STR ==$ S$+ $ $ SF failure Risktaken Software S.A. de C.V. SIS the process industry typically must be out of is service toadditional make repa The second terminredundancy is the common cause term, is and the third term the for systematic er detected unless of components provided. Accounting 1oo1 NOTE The above equations apply to elements with the same failure rates. If elements with different fa 1oo1 detected unless redundancy of components provided. short Accounting additio S are being made is typically not considered due to theisrelatively repairfor time. C is the systematic failure rate for the component. S $S$isF the Where safesafe or spurious failure rate forrate the component, adjustments must be made (See ISA-TR84.00.02-2002, Part 5 for method). S DD failure S appropriate Where is the safe or spurious for the component, $ are being made istypically typically notwith considered due of torates. the relatively short repair tim SIS in The the above process industry must be taken out service to make repairs wh (Eq. No. 10) STR = $ + $ + $ F NOTE equations apply to elements the same failure If elements with different fa systematic error are handled as described in 5.1.5. Therefore, the equations above S DD S DD second term isandthe common causeas term, the third term is thefor systematic systematic error areofishandled described in 5.1.5. the equationser a the equation dangerous the component, and $ inis the (Eq. No.The 10)second term STR =is$detected $ failure +The $rate the+ dangerous detected failure rate term the third term the detected unless redundancy components isand provided. Accounting additional fa F for appropriate adjustments must be made (See ISA-TR84.00.02-2002, Part 5Therefore, for method). DDS the following: SIS in the typically must be taken out of service to make repairs w systematic $ error rate term. The dangerous detected failure term process isthe included inindustry the spurious trip calculation is the dangerous detected failure rate for the component, and Where is the safe or spurious failure rate for the component, $ following: are being made is typically not considered due to the relatively short repair time. Co 3. Determinación de lafailure Relación de Disparos en Falso STR S dangerous when the detected puts that channel (ofcomponent. a redundant system) or system (if it is nonfor the S $ F is the safe systematic failure rate detected unless redundancy components provided. Accounting forwith additional NOTE The above equations apply of to elements with theissame failure rates. If elements different faf DD systematic error are handled Where redundant) $in S$ais the safe or spurious failure rate the component, safe (de-energized) state. This can be done either automatically or by humanas described in 5.1.5. Therefore, the equations above is the dangerous detected failure rate forfor the component, and SIS in the process industry typically must be taken outrelatively of to make repairs 1oo1 appropriate adjustments must be made (See ISA-TR84.00.02-2002, Partservice 5 for method). EcuacionesThe para la determinación de la Relación de Disparos en Falso (Spurious Trip Rate -STR). 1oo1 are being made is typically not considered due to the short repair time. wC intervention. If dangerous detected failure does not place the channel or system into a safe state, this $ is the safe systematic failure rate for the component. second term failure rate term and the third term is the F in the equation is the dangerous detected the following: term is not included in Equations 10 through 15. detected unless redundancy of components is provided. Accounting for additional DD S systematic error term. Thesystematic dangerous detected failure term is included the spurious tripand calculation systematic error are handled as described in 5.1.5. Therefore, the equations abovef is isthe detected failure rate theincomponent, $ $rate thedangerous safe failure rate for"the component. 27 ISA-TR84.00.02-2002 - Part 2 " for when the detectedF dangerous failure puts that channel (ofbeing a redundant system) ortypically system (if it is non-considered S S are made is not due to the relatively short repairrepairs time. w C SIS in the process industry typically must be taken out of service make ComoThe comentamos conveniente conocer la(Eq. relación de disparos un sistema tendrá, estotonos following: second the(de-energized) equationes isstate. the dangerous detected failure rate term and the en third termque is $ the 1oo2 term No. STR = redundant) inanteriormente ainsafe This canthe be done either10a) automatically or10a) by human (Eq. No. STR = $falso 1oo1 The second term in term. the equation is the dangerous detected failure rate and the term is the systematic error are handled as described in 5.1.5. Therefore, the equations above intervention. IfSdangerous detected failure does detected notdetected place thefailure channel or term system into a third safein state, this systematic error rate The dangerous term is included the spurious trip calculation unless redundancy of components is provided. Accounting for additional f $sistemas isinterm. the safe systematic failure rateterm for isthe permitirá seleccionar basados los asociados a disparar/parar un procesos por la falla de alguno de los Frate S detected DD systematic The dangerous included the spurious trip calculation term isNo. noterror included Equations 10 15. [2puts ]channel [,(of ]component. (Eq. 11) STR = through %puts ($en + $DDare )costos +failure % ($S (of + $made )1oo2 + $SFisin typically when when the detected dangerous failure a redundant system) (if it is due non- to the relatively short repair time. C the following: being considered 1oo1 the detected dangerous failure that that channel a redundant system) or system (ifnot itor is system nonS 1oo2 componentes instrumentado de seguridad: 1 isstate. (Eq. No. 10a) STR redundant) in sistema ainsafe state. This can beeither done either automatically or = by$human redundant) aSsafe(de-energized) (de-energized) This can be done automatically or by human 1oo2del systematic are as The term indangerous the equation the term dangerous detected failure ratehandled term and the described third term is in the5.1.5. Therefore, the equations above term is the common cause and the third term is error the systematic error rate term. (Eq.intervention. No.second 9)The second = $ intervention. If detected failure does not place the channel or system into a safe state, this If dangerous detected failure does not place the channel or system into a safe state, this spurious systematic error rate term. The dangerous term is included in the spurious the following: S trip calculation S S DD detected S failure DD S 1oo1 term is not included in Equations 10 through 15. MTTF [channel No. 11) STR = [210 % ($through + $ )] + , % ($No. + $ 10a) )]a+(Eq. $ F No. 11a) term is (Eq. not included in Equations 15. STR S = 2%$ (Eq. STR == $ 1oo3 when the detected dangerous failure puts that redundant system) or system (if$it is non-/ISA TR (Eq. STR 2Simplificada % 1oo2No.(of11a) Arquitectura Ecuación Compleja/ISA TR 8402p2 Ecuación 1oo2 redundant) in a safe done automatically or by human The second term is(de-energized) the common causestate. termS andDD the third systematic error rate term. Sis theDD S 1oo1 1oo2 [,notterm (Eq. No. 12) STR = [3 %failure ($ + $This )]+ can %No. (be $place +10a) $ the )]either +channel $SF 1oo1 8402p2 (Eq. STR = a$safe 1oo3 intervention. If dangerous detected does or system into state, this 1oo2 S DD S DD S 1oo3 ]+1oo3 [, % ($No. (Eq. No.included 11) STR = [210 % (through $ + $ )15. + $ 11a) )]+ $F (Eq. STR = 2 % $S term is not in Equations The second term is the common cause term the third term is the error S and DD S systematic DD S rate term. (Eq. No. 11) STR =[3 %2($%S S+$$DD+)]$+DD %10a) $]+(Eq. +S $ No. +12a) $ F rate term. S , (Eq. STR = $ SSTRS = 3 % $S [,third (Eq. No.1oo1 12)term is the common STR =cause %+ ($+No. +S$DD $systematic second is )the error (Eq.1oo2 No.The 10) STR = $term +and$the $term 1oo2 F 11a)F (Eq. No. STR 2oo2 1oo3No. 12a) (Eq. STR == 23%% $$S The second is thecommon common cause termterm and the third term is theterm systematic error rate term. error rate term. 1oo3 The second termterm is the cause and the third is the systematic S DD (Eq. No. 13)S STR = [2 % $S ($ % ($S DD + $DD )] +S$SF DD)% MTTR ] + [,S2oo2 S (Eq. No. 11) 1oo2 STR = spurious 2 %S $S DD ++$$1oo2 +No. ,rate %11a) $for + $the + $F Where or failure component, $ is the safe 2oo2 S (Eq. STR == 23 % $ 1oo3 S DD S [ ] [ ] (Eq. No. 12) STR = 3 % ( $ + $ ) + , % ( $ + $ ) + $ (Eq. No. 12a) STR % $ 2oo2 F 1oo3 The second term is the common cause term and the third term is the systematic error rate term. This (Eq. No.1oo1 9) $S = " 27 " 1 MTTF1 spurious " 27 " [ ( )] [ ( )] [ ( )] [ ( )] [cause ]+ [,isfailures (Eq. No.term 13)asDD STR =14 2 %and $ (15, $term + $ and )% third MTTR %term ($(Eq. +can $ No. $13a) equation, well as Equations assumes that safe be)] + detected on-line. If safe F The second is the the third is the systematic error term. (Eq. No. 11a) STR = rate 2 and % $ S = 2 % $S % MTTR STR The second is thecommon common cause term and the term therate systematic error rate term. is dangerous detected failure for the component, $term failures1oo3 can only bethe detected through testing the12a) testing (orDD inspection) TI should be $ S or inspection, DD S S interval 1oo3 (Eq. No. STR = 3 % 2 (Eq. No. 12) STR = 3 % $ + $ 2oo2 + , is%the$systematic + $ error + $rate F term.STR substituted for MTTR. The second term is the common cause term and the(Eq. third term This = 2 % $ S No. 13a) % MTTR 1oo3 2oo2 equation, as well as Equations 14 and 15, assumes that safe failures can be detected on-line. If safe S S DD S [ ( DD )] [ ( ( ) 2 S S )] ( ) $ STR [ ( ] [$ (+ $ )]+)] $ )(Eq. STR = = 32%% $($ ) % MTTR (Eq. No. 12) STR = [3 % ($ + $ 2oo3 )]+ [No. , % (13a) STR = [6 % ($ )% ($ + $ )% MTTR ]+ [, % ($ + $ )]+ $ (Eq. No. 14a) STR = 3 %STR = 6 % ($ ) % MTTR (Eq. No. detected 12a) The2oo2 second term in the equation is the dangerous failure rate term and$the third term is the 2oo2 (Eq. No. 13a) STR = 2 % ( $ )) %% MTTR The second term is the common [cause term and the third term is the systematic error rate term. 2oo3 [,failure ($ ) ($ $ (Eq. ) No.] 14a) ($ $term )] $ is included systematic error rate term. The dangerous detected trip calculation 2oo3 STR = in 6 %the ($spurious MTTR 2oo3 S 1oo3 failures through testing or inspection, the testing (or interval TI should be 2oo3 can only be S 2 isdetected the safe systematic failure rate forSinspection) S term S DD DD component. The second is the common cause and the third term isthe systematic error rate term. 2oo2 F S (Eq. No. 12a) substituted for MTTR. (Eq. No. term 13) STR = 2 % $ $ + $ % MTTR + , % $the + $SSF S DD S $ +DD 2oo2 S S DD S DD S F (Eq. No. 14) F 2oo3 The second term is the common cause term and the third term is the systematic error rate term. This equation, as well as Equations 14 and 15, assumes that safe failures can be detected on-line. If safe S term, DD S The second term is the common cause the third term is the systematic rate term. (Eq. No. 14) STR = 6 % testing % orS and +inspection, % MTTR + % S + DD + error failures can only be detected through be S S DD the testing (or inspection) S Finterval DD TI should 2oo4 No. 13) for MTTR. STR = 2 % + + % + + SF substituted 2oo2 % MTTR 2oo4 [ ( ) ] [ ( S 2 S S 2 S 2 )] (Eq. $ that $ channel (of ,a redundant $ $ system) $ when the The detected dangerous failure$ puts or system (if it is nonsecond term is the common cause term, and the third term is the systematic error rate term. 2oo2 2 2oo3 SS 2by human redundant) This can be done either automatically or 2oo3in a safe (de-energized) state. (Eq. No. 13a) STR = 2 % $ % MTTR 3 (Eq. No. 14a) S is the STR 6 term. MTTR2 DD and 2 S 3 This 2oo4 The second term the third error rate 2oo4 (Eq. No. 15) is the common STR = cause 12 % ($S term MTTR + [,term %(Eq. ($ +No. $DD )]systematic + S$SF DDSTR 2oo4 = S+ $ S ) %does DD S 12 % $into intervention. Ifwell dangerous detected failure not place the(15a) channel or system safe state, this S 2 %aMTTR (Eq. No. 13) STR = 2 % ( $ $ ) % MTTR + $ + $ ) + $ S$ S + DD S, % DD S detected equation, as as Equations 14 and 15, assumes that safe failures can be on-line. If safe F (Eq. No. 13a) STR = 2 % $ % MTTR (Eq. No. 14) STR = [6 % ($S )%DD ($ 3 + $ 15. )% 2MTTR ]+ [S, % (DD$ + $ S )]+ $F S 2 termfailures is not included Equations [ ] (Eq. No. 15) be in STR through = [12 % (10 $ testing +through $ ) %or MTTR ] + , % ( $ + $ ) + $ can only detected inspection, the testing (or inspection) interval TI should be (Eq. STRS =3 6 % $ %2 MTTR F 2oo4No. 14a) 2oo3 spurio 5.2.6 Combining spurious trip rates for components to obtain SIS MTTF (Eq. No. STR = 12 $ % MTTR substituted for MTTR. The second term is thecommon common cause term, and and the third term is15a) the systematic error rate % term. The second term is the cause term the third term is the systematic error rate term. This S λ1oo2 equation, es la as relación fallas seguras en assumes falso parathat cada 2oo3 well asdeEquations 14 ando15, safecomponente. failures can be detected3 on-line. If safe 2 2 solver, and power supply portions are eva 2oo4 2oo4 S logic Once the (or sensor, 2oo3 failures can only be detected through testing or inspection, the testing interval TI should be spurious (Eq. No. 15a) STR =inspection) 12 STR %final $S =element, (Eq. No. 14a) 6%%MTTR $for % MTTR spurious 5.2.6 Combining spurious trip rates components to obtain SIS MTTF DD for the SIS being evaluated is obtained as follows: MTTF substituted for MTTR. λ es la relación de fallas peligrosas detectadas componente. S 2 S DD 3S 2para cada S DD S DD DD S S [14a) (Eq. No. 15) STR = 12 %=($ 2+%$S $) %(Eq. , %, ($ %+ $$ )]++ $$F S STR S $ No. DD+ + (Eq.(Eq. No.No. 11)14) STR +MTTR +$DD $ FS =+3$6S% $ %2 MTTR spurious STR = 6 % $ % $(Eq. + $No. %Combining MTTR + ,spurious %=$12+% 15a) STR $ logic %F MTTR 5.2.6 trip rates for components to supply obtain SIS MTTF Once the sensor, final element, solver, and power portions are evaluat 2oo3 2oo4 S STR = STR + STR + STR + STR λ F es la relación de fallas sistemáticas seguras para cada componente. spurious SIS Si Li PSi + $ (Eq. No. SIS 16) being evaluated for the is obtained as Ai follows: MTTF 2oo4 second termisisthe the common term, and the third term is the systematic error rate term. TheThe second term commoncause cause term and the third term is the systematic error rate term. S S S solver, and power supply portions are spurious Once the sensor, trip rates for components to obtain SIS MTTF evaluate (Eq. No. 14) STR = 6 % ($S )% ($5.2.6 + $DDspurious )Combining % MTTR + final ,spurious % ($element, + $DD )S +logic 3$ F 2 El valor final de la relación de disparos en falso(Eq. del sistema SIS (utilizando las% in ecuaciones simplificadas) es follows: laterm, suma de used cadawhen systematic e NOTE The last term the equation, theobtained systematic failure is only for the SIS being evaluated is as MTTF No. 15a) STR = 12 $ % MTTR 2oo4 3 the user individual component STRSTR desires toAiinclude an overall value for the entire STR = + STR + STR + STR $FS Sand 2 1oo3 SIS Si Li PSi +system. (Eq. No. 16) elemento del sistema: No. 15a) =element, 12 % $error % MTTR The second term is the common cause term, (Eq. and the third term isSTR the systematic rate term. and power supply portions are Once the sensor, final logic solver, evaluat spurious S spurious 3 SIS being evaluated obtained follows: MTTF 2 for the S DD S rates+foriscomponents = +trip STR STR Ai + as STR STR 5.2.6 Combining spurious to obtain MTTF SIS Si Li + SIS PSi + $F (Eq. No. 15) STR = 12 % $S + $S DD %DDMTTR + term , %SSTR $ +DD $ $ S the (Eq. No. 16) S F 1 2oo4 s p u r io u s spurious NOTE The last in the equation, systematic failure term, is only used when systematic error ha STR = 12) ∑STRSensor + ∑STRCLP (Eq.SISNo. STR+ ∑STR = 3 %EF$+ λ5.2.6 +F$ + , % $ +spurious $ +Mtrip $TF Trates Combining to obtain SIS MTTF F for components = anS T individual component STR and the user desires to include overall value for the entire system. R S IS (Eq. No. 17)element, 3 Once the sensor, final solver,STR andAiterm, power supply portions evaluate STR = ) + $STR + is only STR STRare $FS has NOTE The in Sthe equation, the systematic failure used when systematic S DD 2last term DD Slogic SIS Si + Li + PSi +error (Eq. No. 16) (Eq. No. 15) (Tiempo Medio STR = 12 % ( $ + $ ) % MTTR + , % ( $ + $ spurious ElThe valor de MTTF Para Fallar) esta dado por: F Once the sensor, final element, logic solver, and power supply portions are evaluate second term is the common cause term and the third term isand the error rate individual component STR thesystematic user desires to include anterm. overall value for the entire system. for the SIS being evaluated is obtained as follows: MTTF spurious The result is the MTTFspurious for the SIS. 1 s p u evaluated r io u s for the SIS being is obtained as follows: MTTF T Tequation, F NOTE The last termM in the the systematic failure term, is only used when systematic error ha = M2oo2 TTF En Falso = 1/STRSIS T1R S an IS overall value for the entire system. S individual STR and=the toSinclude (Eq. No.component 17) s p user u r STR io udesires s STR + STR STR Li + STR PSi + $FS Si= Ai + (Eq. No. 16) M TSIST F STR = STR + STR STR Li + STR PSi + $F SIS Si Ai + S T R S IS (Eq. No. 16) (Eq. No. 17) spurious S S the MTTF (Eq. No. 13) STR = 2 % $S $The + result $DD %isMTTR + , % $for +the $DDSIS. + $SF [[ (( ] [ ] (( ( ) ( ) ( ])]) [ (] [ ( )] )] ( ( ) ) )) ) ] [ [ [ ( () ( [ ] [ [ [( ( [ [ ) (NOTE (# ] ) ( ) ) ) # # )] [ ] [( ( ] )] )# ] # ] # # # # # # # ) [ ] [ ( )] # # # # # # # # # # # # 1 s p u r iothe u ssystematic failure The last term in the equation, term, is only used when systematic error ha M Tand Tequation, Fthe user spurious = toSinclude NOTE The last termMTTF in the systematic failure term, is only when systematic The result is the forthe the SIS. individual component STR desires an overall valueused for the entire system. error ha T R S IS (Eq. No.component 17) individual STR is and thesystematic user desires to include anterm. overall This value for the entire system. term and the third term the error rate The second term is the common cause equation, as well as Equations 14 and 15, assumes that safe failures can be detected on-line. If safe spurious The result is the MTTF s p ufor r io inspection) uthe s SIS. 1interval TI should be failures can only be detected through testing or inspection, M theTtesting 1 s p u(or r io u s = T F M TTF = SS T substituteddefor MTTR. Determinación la PFDavg (Eq. No. 17) TR R SS IS IS (Eq. No. 17) 2oo3 spurious The result result is is the the MTTF MTTFspurious for The for the the SIS. SIS. S S DD S DD S 5 Risk Software S.A. de C.V. ISA-TR84.00.02-2002 Part2 2 ISA-TR84.00.02-2002 - -Part " 24 " 24 " " 4. Determinación de la Probabilidad de Falla Sobre Demanda IfIfISA-TR84.00.02-2002 systematic errors(functional (functional failures) are included in the calculations, systematic are to to be included the calculations, sep - Part 2failures) 24 " be " coninprueEcuaciones para la determinación de la Probabilidad de Fallas Sobreerrors Demanda PFDavg para Sistemas sub-system, available,may maybebeused used equations above. An alternate a sub-system, ififavailable, in in thethe equations above. An alternate appro bas manuales. value for for functional functionalfailure failurefor forthe theentire entire SIF and add term as shown in Eq SIF and add thisthis term as shown in Equati If systematic errors (functional failures) are to be included in the calculations, se NOTE Systematic failures are rarely modeled for SIF Verification calculations due the difficu Systematic failures are rarely modeled SIF Verification calculations due thetodifficulty in - Part 2 para sistemas " 22 " La Probabilidad de Fallas ISA-TR84.00.02-2002 Sobre Demanda con pruebas manuales, esta relacionada generalmente aabove. los elesub-system, if available, may be usedforin the equations Anto alternate app effects and data forfor various types of systematic failure. However, thesethe fai and effects andthe thelack lackoffailure offailure failurerate rate data various types of systematic failure. However, value for functional for the entire SIF and add this term as shown in Equa ISA-TR84.00.02-2002 Part 2 24 " " mentos de campo, como son sensores y elementos finales de control. can the SIF performance. ForFor thisthis reason, ANSI/ISA-84.01-1996, IE and can result resultininsignificant significantimpact the SIF performance. reason, ANSI/ISA-84.01-19 Equations for typical configurations: ISA-TR84.00.02-2002 - impact Part 2toto " 24 " provide aa lifecycle design andand installation concepts, validation and testing provide lifecycleprocess processthat thatincorporates incorporates design installation concepts, validation and tesc NOTE Systematic failures are rarely modeled for SIF Verification duefailures. to the difficulty TI ) & D change. TI & ) ISA-TR84.00.02-2002 - Part 2intended 24 " the " This process isis intended to to support the reduction incalculations the SIL Ve This lifecycle lifecycle process support in systematic the systematic failures. S PFD avg = ($ DUentre % + + pruebas $ % change. tiempo o intervalo manuales (TI), que tiene como objetivo lareduction identificación yfailures. errors failures) are to betypes included in the calculations, separate va and effects and the lack(functional of failure rate data for various of systematic failure. However, these 2 * '( F predominantly 2 If *+ systematic ' concerned thethe SIS performance related to random " 22 " predominantly concernedwith withassessing assessing SIS performance related to random failures. La base de estas ecuaciones es el (Eq. No. 3) 1oo1 ISA-TR84.00.02-2002 - Part 2 If systematic errors (functional failures) are to be included in the calculations, separate sub-system, available,impact may be inperformance. the equationsFor above. An alternate approach is t and can result inif significant to used the SIF this reason, ANSI/ISA-84.01-1996, localización de fallas peligrosas en el$DUsistema o elementos del sistema. sub-system, if available, may be used in the equations above. An alternate approach i where is the undetected dangerous failure rate Equations for typical Las ecuaciones queconfigurations: describen los (Eq. No. 3) value afor functional failure the entiredesign SIF and this term as shown in Equation 1a in provide lifecycle process thatfor incorporates andadd installation concepts, validation and testing Ifvalue systematic errors (functional failures) are to beadd included in the calculations, separate The simplified equations the terms for multiple failures during com for functional failurewithout for the entire SIF and this term assystematic shown inrepair, Equation 1aV The simplified equations the terms for multiple failures during repair, change. This lifecycle process iswithout intended to support the reduction in the failures. SIL D sub-system, if available, may be usedforinSIF thefor equations above. Antoalternate approach $ F is the dangerous systematic failure rate, and systematic errors reduce toPeligrosas following use incalculations the procedures outlined in 5.1 NOTE Systematic failures are rarely modeled Verification due the difficulty in assessing predominantly concerned with assessing the SIS performance related to random failures. systematic errors reduce tothe the following for use in the procedures outlined in sistemas utilizan el componente de Relación de Fallas Sistemáticas. value for functional failure for the entire SIF and add this term as shown in Equation 1a and effects and the lack of failure rate data for various types of systematic failure. due However, these failures are NOTE Systematic failures are rarely modeled for SIF Verification calculations to the difficulty in assess TI is the time interval between manual functional tests of the component. result significant impact rate to the SIFfor performance. For reason, failure. ANSI/ISA-84.01-1996, 61508, and can effects andinthe lack of failure data various types of this systematic However, theseIEC failures a 1oo1 The simplified equations without the terms forFor multiple failures during repair, com TI failure & DU TI- Part) 1 model & theDsystematic )1oo1 provide aSystematic lifecycle process that design and concepts, and criteria, and NOTE failures areincorporates rarely modeled for SIF installation Verification calculations due to thetesting difficulty in asses NOTE The equations in ISA-TR84.00.02-2002 as an error occurred the and canthat result in during significant impact to the SIF performance. this reason, validation ANSI/ISA-84.01-1996, IEC 615 PFD = $ % + $ % specification, design, in change. the SIF component being susceptible a avgimplementation, errors reduce to following for use in the procedures outlined in 5isa This lifecycle process is intended support thetypes reduction in the systematic SIL Verification ( commissioning,+ or maintenance ( F that 2resulted +butsystematic and effects and the lack oftofailure ratethe datato for various of systematic failure. failures. However, these failures provide lifecycle process that random failure. Some systematic failures do not2manifest themselves randomly, exist at time 0aand remain failed throughout the incorporates design and installation concepts, validation and testing criteria, 1oo1 ' * ' * TI predominantly concerned with assessing performance related random failures. mission time of the SIF. For example, if the valve actuator is specified improperly, leading to the inability close the valve under and can result in significant impact to thethe SIF performance. For thisinto reason, ANSI/ISA-84.01-1996, IEC 615 change. Thistolifecycle process is intended toSIS support the reduction the systematic failures. SIL Verificatio DU the process pressure that occurs during the hazardous event, then the average value as shown in the above equation is not provide a lifecycle processwith thatassessing incorporates and installation concepts, validation predominantly the design SIS performance related to random failures.and testing criteria, applicable. In this event, the systematic failure would be modeled using $ % TI . When modeling systematicconcerned failures, the reader DU avg change. This lifecycle process is intended to support the reduction in the systematic failures. SIL Verificatio must determine which model is more appropriate for the type of failure being assessed. (Eq. No. 3a) selección, PFD =implementación $ % TI Esta relación representa las fallas sistemáticas introducidas durante diseño, y mantenimiento de 1oo1el (Eq. 3a) equations PFDavg = $ the terms %2 for multiple failures during repair, common cau TheNo. simplified without where $DU is the undetected dangerous failure rate 2 predominantly concerned with assessing the SIS performance related to random failures. los elementos de campo del Sistema Instrumentado de Seguridad. systematic errors reduce to the following for use in the procedures outlined in 5.1.1 throuc The simplified equations without the terms for multiple failures during repair, common 1oo2 $ DF TI for use in the procedures outlined in 5.1.1 thro systematic errors reduce to the DU following 1oo2 (Eq. No. 3a) equations PFDavgwithout = $ the%terms for multiple failures during repair, common The simplified 1oo1 1oo2 2 for /ISA systematic errors reduce to following use inTR the procedures outlined in 5.1.1 thr Arquitectura Ecuación Compleja/ISA TR 8402p2 EcuacióntheSimplificada 1oo1 & & TI 2 ) TI ) & D TI ) DU 2 2 DU 2 DU DD DU ] " , ) % $ ) %manual + [(1 "functional % MTTRof % TI + (component. , ) % $ % $ tests , % $ % + + ( $F % + TI % TI $ avg = (((1between + TI is the timePFD interval the DU 2 3 * 2 * = $ 8402p2 ' ' ' 1oo2 (Eq. No. 3a) 2 * PFD %$ DU % TI 2 1oo1 Equations for typical configurations: (Eq. No. 4a) PFDavg 2TI avg = DU ISA-TR84.00.02-2002 - Part 2 " 23 " (Eq. No. 3a) PFD = $ % 3 NOTE The equations in ISA-TR84.00.02-2002 - Part 1 model failure as(Eq. an error that occurred during the = No. 4a) PFD ISA-TR84.00.02-2002 -Consequently, Part 2 avgavg " 23 assumed "the systematic For simplification, 1-, is generally to be one, which yields conservative results. 22 TI TI & ) & ) resulted in the SIF component being susceptible to specification, design, implementation, commissioning, a DUTI DU toor maintenance 2 (Eq. No. 5) = reduces the equation 1oo2No. 3a) PFD % + + ($ DF % that (Eq. No. 3) 1oo1 1oo1 (Eq. PFD = $ DU$%the 3% TI ($manifest +* random failure. Some systematic failures avg do not but exist at time 0 and remain failed throughout avg 2 themselves 2randomly, ' * ' 1oo3 2 & ) (Eq. No. 4B) & leading ) &No. TI TI1oo2 TI4a) 2 /to ) PFD =under mission time the5)SIF. For example,PFD if the=valve actuator is specified improperly, to close the avg valve % (Eq. $ %inability -+ +the (Eq.ofNo. (($ ) % + + [($ ) % $ % "MTTR 23 " % TI ]+ ( , % 0 $ISA-TR84.00.02-2002 - Part 2 4 *event, then the 2shown 2 +* above equation is not DU 2 1 as1oo3 .* (' in the ' DU the process during the hazardous average 'value ( $ DU) 2%3 3TI 2 3 where pressure$ that is occurs the undetected dangerous failure rate 1oo2 applicable. In this event, the systematic failure would be modeled using $ % TI . When modeling systematic failures, the reader (Eq. 4a) = $$DU %%TITI 3 No. 5) & (Eq. )multipleDUfailures TIfor2 type TInegligible TI ) TI PFD &2 isDU ) /No. DU & DU is3 more )second=term 2for &TIThis] +factor )&$DF &for TI TI The during repair. typically ( DU )3 3 2 3 ) avg must determine which model appropriate the being DU 2 %assessed. ($DUaccounts )2%%term [MTTR PFD $ of %failure $DD% % MTTR % short 1oo3 (' +error +*% ($the )dangerous ) % repair +avgsystematic %+rate, + the %& ('0,term %%TIsystematic $&('The $3DD ,fourth $% $is(Eq. $TI2DFterm. PFDavg1oo2 $=DF(is failure and third is +the common causeTI term and the -+No. 2/)+*No. 5a) PFD = + times. ( + ( + )* 2 TI (Eq. 4a) PFD = 2 & ) avg % TI $ 42oo2 avg 23 " % TI' ]+ ( ,1%ISA-TR84.00.02-2002 2 2 * PFD 0 $ % 2-+ .+*($ %' - Part ' * = (($ ) % 4 + + [($ ) % $ %" MTTR $ DU ) 34% TI 2 ( 2 .* ' 2 *+ 1 ' ' * 1oo2 1oo3 (Eq. No. 5a) PFD = 3 TI is the time interval between manual functional tests of the component. (Eq. No. 4a) PFDavgavg= where MTTR is the mean time to repair TI ) & (Eq. No. 5) $ DU3 4% TI 3 (Eq. No. 6) PFD = [$ % TI ] + [, % $ % TI ] + ($ % + 1oo3 2oo2 The second term accounts for multiple failures during repair. This factor2 is typically negligible for short ' * 3 NOTE The equations in ISA-TR84.00.02-2002 Part 1 model the systematic failure as an error that occurred during the DD (Eq. No. 4A) The second term accounts fortimes. multiple during repair. factor is(Eq. typically short = DU repair term common causefailure termThis and theand fourth term is the systematic error term.for No.negligible 5a) PFD $failures is dangerous detected rate, &Theorthird )is the that & avg ( $ TI TI 2 beingTIsusceptible /) & ) % 4TI 3 specification, design, implementation, commissioning, resulted in the SIF component to a) )the = term % common + [(cause % $ and%and %term % %error $ ismaintenance $ ) term ,systematic +systematic PFD MTTR TIfourth 0 $ is - term. repair times. The third term is the the term the The term third is]+the ((common +cause 1oo3 ('$ % 2 +* the error term. random failure. Some systematic failures dosecond not manifest themselves randomly, buttheexist at time 0('and failed throughout 2oo2 1oo3 2oo2 4 * 2 .+* No. 1remain ' (Eq. 5a) PFD = DU DU 3 , fraction of mission time of the SIF. For example, if the improperly, leading tomore the inability to(Eq. close the under system failures that impact than one channel ofvalve a redundant No. 6a) PFDavg %)4 TI 2oo3valve actuator isisspecified % TI 3 ( $TI avg = $ 2 the process during the hazardous event, then the average value as shown in the& above is not TIequation (common cause). ) 2oo2&pressure that occursDU ) 2 (Eq. TI TI & ) & ) [, %modeling No. 6) PFD = [DU $%failures % TI ]during +DD $ % TI ]systematic + (factor $ % 2oo2 DUshortPFD D 3 (Eq. No. 5a) = + (Eq. No.would 7) termbe The second accounts forusing multiple repair. This is typically negligible for applicable. In this event, the systematic failure modeled $ TI . When failures, the reader DU avg $ % DU 2* + , %$ ' % TI PFD = ( (1which " ,model ) % is$more appropriate %Thetimes. +the (type 1represents "term %cause % MTTR %forPFD ,is)the%multiple $ assessed. $ during avg + The TI 3 repair third common term and the fourth is the systematic error term. (Eq. No. term failures repair. Thisterm factor is typically negligible short (' 6a) + +avg ( F=( $$ 2 )+*4%%TI must determine for of failure being 2oo2 3second repair times (typically less than 8 hours). The common term. term*is ' TI)) cause TI ) The fourth 2 ' * 2oo3 &No. & the The second term is theDU common cause term and thethird thirdterm term is is the systematic error term. TI & (Eq. 5a) PFD = DU D ( ) ( ) + % PFD = $ % TI + 3 $ % $ % MTTR % TI + , % $ % $ [ ] [ ] 2oo2 the systematic error term. PFD % TI ] + [, % $ % TI ] + ('( $ F % (Eq. 2+*+ '(No. 26a) *+ 2oo2 avg = [$ 1oo2(Eq. No. 6) PFDavg = $ DU %4 TI 2oo3 2TI2oo2 ' & 2oo3 *) No. 6a) (Eq. PFDavgavg= $ DU % TI2 1oo3 (Eq. No. 4A) systematic failure rate, and is the dangerous ISA-TR84.00.02-2002 - Part 2 " 22 " 3 DU 3 DU 2 DD 2 DU D F avg [( 3 DU 3 DU 2 ] DD 2 DU D F avg DU DU D F avg 3 DU 3 DU 2 DD 2 DU D F avg ( [ ) DU DU DU 2 DU ] D F avg 2 DD DU D F avg PFD (Eq. No. 6) [ [([( [( [( [[ ] [ ] ]] ] ] ]] ) ) ) [ [( ) ] ] [( ) ] [ ] [ ] [ ] ( = $DU % TI + , % $DU % TI + $D % ) DU ) 2 (Eq. 7) term in the equation avg (' F This +* No. TheNo. second represents multiple failures during repair. is typically (Eq. No. 4A) 2 factor (Eq. 7a) $ % TI 2oo2 avg = DU negligible assumed for short repair times. The third term is the common term. The fourth term is the For simplification, 1-, is generally be one, which conservative results.PFD Consequently, The second term is the common cause termtoand the third term yields is cause the systematic error term. (Eq. No. 6a) PFD % TI 2oo3 systematic error term. TI TI ) avg = $ ) & term. & 2oo3 ThePFD second=term third%term is the systematic error the equation reduces to ($ is )the%common (TI ) ] +cause + % 3$ term % $ and% the MTTR TI ] + , % $ % $ [ [ ( 2 DU 2 2 2 TI *+ ) PFD *+ '(& 7a) & 2oo3 DU 2 TI2oo4 ) DU$ TI2 )No. (Eq. & ' D avg= = 2oo3 + [(1 " , ) % $DU % $DD % MTTR % TI ] + , % $DU2oo4 (Eq. No. 6a) PFD $ % TI2 % TI ( ) PFD2oo3 = ( 1 " ) % % % + % , $ $ avg avg F ( + ( + ( + 2 2oo3 2 DU 2 DU No. 8) 3(Eq.second 2factor ' *No. 'is typically (Eq. No. 4B)' The * This 7a) PFD = =$ $ % TI% TI (Eq. No. 7a) 2 * PFD (Eq. No. 7) term in the equation represents multiple failures during repair.(Eq. DU 2 2 DU DD DU avg D F ( ) avgavg ( ( ) ) (Eq. No. 7) TI ) & TI ) & DU 23 2oo4 2oo3 PFD = [( $ ) % ( TI ) ] + [4( $ ) % $ % MTTR % (TI ) ] +&(, % $ % TI+)+ (&$ % TI+ ) 2oo4 ) %%TI(TI2 )3 PFD = (($$DU *+No. +'($8a) %2 * + PFD = [($ ) % (TI ) ] + [3$ % $ % MTTR % TI ] + ', % $(Eq. % 2 No. avg = (Eq. 7a) PFDavg ) 2oo4 2 2 For simplification, 1-, is generally assumed to be one, which yields conservative ('results. Consequently, * * ' 2oo4TI ) & TI ) & the equation reduces to % + PFD % + + ($ TI 2 & DU= 2[($ TI)2 )% (TI )DU] + [3$DD % $ % MTTR % &TI ] + (',DU% $ (Eq. TI ) 2No. &* Combining )2PFD (Eq. No. 7a) ($DU(DU$DU ) 2 )%3 TI D avg = = * components’ ' 3 PFDs )3 SIF PFDavg 8a) PFD % TIobtain 3(to 5.1.6 2oo4 ( ) [ ] PFD = % $ + $ % $ % % + , % $ % + $ % MTTR TI avg 3 avg F ( + ) ( ) (Eq. No. 4B) (Eq. No. % TI TI 3 ) &8a) TI ) +PFDavg = ($ & ( + ( DU MTTR es + ($ % 2+ * PFD = [( $ reparación %2 $ * %No. ) % (TI ) ] + [4( $ ) % $ % MTTR ' % (TI ) ] + (', (Eq. ) % (TI ) ' el tiempo3medio * PFDpara 2'+* 8a) 2* avg = ($ ' negligible for short repair times. The third term is the common cause term. The fourth term is the 3 3 2 DU 2 DD DU D systematic errorDUterm. avg 2oo4 DU 2 DU 2 2 DU DD DU FD F avg 2 DU D F The equationrepresents representsmultiple multiple failures during repair. is typically Thesecond secondterm term in in the equation failures during repair. ThisThis factorfactor is typically negligiblefor forshort short The third is is thethe common cause The The fourth term term is theDis the negligible times. TheDU third term common cause term. 2 term 3 2 term. DU 3repair times. DD DU fourth systematic error term. term. avg F systematic error (Eq. No. 8) avg DD DU 2oo4 Once the sensor, final element, DU logic solver, and power supply (if applicable) port For configurations other than those indicated above, see Reference 3 or ISA-TR84.00.02-2002 - Part 5. 2oo4 The second term in the equation represents multiple failures during repair.5.1.6 This is typically components’ to obtain SIF PFDavg (Eq. No.Combining 8a) PFD %PFDs TI 5.1.6factor Combining components’ to obtain SIF PFD avg = $ avg the SIF beingPFDs evaluated is obtained by summing the individual overall PFD The terms inThe the equations representing common causeTI (Beta factor term) and systematic failuresterm are for avg times. third term is&the common cause term. The fourth is the negligible es& laDUfor relación fallas peligrosas detectadas No. 8) ) de (Eq. 2 short TI 2 repair TI ) & ) 5.1.6 Combining components’ DU DD DU D The second term in the equation represents multiple failures during repair. This factor is typically 3 PFDs3 to obtain SIF PFDavg typically not included in calculations performed in the%process + industries. These factors are usually DU PFD = % $ + $ % $ % % + , % $ $ % MTTR TI systematic error term. for the SIF for the event being protected against. is the PFD where avg ( MTTR is meanforfortime to design repair Fexperience. + thenegligible short repair times. The third term is thebased common cause term. (Eq. The fourth term is the avg ( + ( + accounted during the by using components on plant No. 8a) PFD = $ % TI Once the sensor, final element, logic solver, and power supply (if applicable) portions are avgelement, 3 * 2* ' 2 *5.1.6 ' systematic error TI TI ) final the logictosolver, andPFD power supply (if applicable) ' ) Combining &sensor, components’ PFDs obtain SIF avg + (events PFD % overall $ DU %external $ DF % for ( $DUterm. ) 3 % (TI ) 3 + 4( $DU ) 2 % $DD % MTTR % (TI ) 2 + &(,Once avg =cause +*PFD SIF being evaluated is obtained by summing the(if individual compon Common includes environmental factors, e.g., temperature, humidity, vibration, avg 2 +* the 2 ' ' Once the sensor, final element, logic solver, and power supply applicable) po β 2oo4es la fracción de fallas que impacta en uno o mas canales de los sistemas redundantes (Factor de falla Común). For configurations other than indicated above, seecalibration Reference 3 or overall ISA-TR84.00.02-2002 - Partfor 5. the SIF being evaluated is obtained by summing the individ avg such as lightning strikes, etc. those Systematic failures include errors, design errors, PFD programming DD λDD [ ( ) ] ] [ [ ] ( ) ( ) ( ) ( ) 3 3 thethe SIFSIF for the event being protected against. is the PFD & D TIa is dangerous failure andrefer to ISA-TR84.00.02-2002 errors, etc.detected If there is concern related to rate, these factors, - Part 1Combining forsensor, aavg for 5.1.6 components’ PFDs to being obtain SIF for being evaluated is Aiobtained byagainst. summing individua overall PFD Once the final element, logic solver, and power supply (if applicable) portions avg avgfor PFDthe PFD PFD PFD = for + PFD (Eq. No. 1a) the SIF the event protected is the PFD The terms equations common cause (Beta factor term) and systematic failures are PFD avg discussion their impact onrepresenting the PFD calculations. PSi + ($ F % SIS Si + Li + where MTTR is the mean timeinoftothe repair typically not included in calculations performed in the process industries. is These factors are usuallyfor (Eq. No. 8) for the SIF being evaluated is obtained by summing the individual overall PFD ' comp2 the SIF for the event being protected against. the PFD avg avg The secondfor term in the represents multiple failures during This factor is typically accounted during theequation design by using components based on plantrepair. experience. TI ) & the sensor, finalSIF element, logic and supply (if applicable) negligible for short repair times. The third term is the common cause term.Once The fourth term is theforPFD the event being against. is the PFD avg PFD PFD PFDprotected PFD =for the + solver, + power + + ($ DF % portions (Eq. No. 1a) PSi SIS Si Ai Li $,DDis is dangerous detected failure rate, and + & D fraction of failures that impact more than one channel of a redundant system systematic error includes term. Common cause environmental factors, e.g., temperature, humidity, vibration, external events Determinación de la PFDavg 'PFD 2 * +comp the evaluated overall PFD 3 2 etc. Systematic TI TI SIF )SIS being ) avg& for &errors, 2 such3 as lightning DU strikes, failures include calibration design errors, programming PFDSi + is obtained PFD Ai +by summing PFD6Li +the individual = (Eq. 1a) DD DUNo. D PFD & D($ F PSi ) ( ) + % PFDavg =(common $ DU % (cause). TI + 4 $ % $ % MTTR % TI + , % $ % $ errors, etc. If there is concern related to these factors, refer to ISA-TR84.00.02-2002 Part 1 for a configurations other than those indicated above, see Reference(3 or (Eq. ISA-TR84.00.02-2002 Part 5. eventSibeing protected is theNo. PFD F the PFD PFD PFD PFDLi + = the PFD + +*SISfor +*avg ('-for & PSi , is fraction ofFor Ai + against. D +TI ($)F' % discussion theirimpact impact onmore the PFD 21a) 2SIF ' failures ofthat thancalculations. one channel of a redundant system PFD PFD PFD PFD PFD $ = + + + + % (Eq. No. 1a) PSi SIS Si Ai Li ( F 2' + The terms in the equations representing common cause (Beta factor term) and systematic failures are (common cause). ' * The second term representstypically multiple failures during repair. This industries. factor is typically negligible for short not included in calculations performed in the process These factors are usually & D TI ) accounted for during the design by using components based on plant experience. PFD PFD PFD PFD PFD $ = + + + + % (Eq. No. 1a) repair times (typically less than 8 hours). The third term is the common cause term. The fourth term is PSi F SIS Si Ai Li ( The second term represents multiple failures during repair. This factor is typically negligible for short 2 +* ' the systematic error term. Common cause includes environmental factors, e.g., temperature, humidity, vibration, external events $ # avg [( ) ][ ( ) avg ] # # # # # # ## # # # # # # # ## # # # # # # Risk Software S.A. de C.V. Para sistemas redundantes el segundo termino en las ecuaciones complejas representa las múltiples fallas presentadas durante la reparación y el tercer termino representa la causa de falla común (CCF). En las ecuaciones simplificadas se considera que el segundo termino es despreciable debido a que el valor es muy pequeño cuando el tiempo de reparaciones es menor a 8 hr. El tercer termino es despreciable debido a que se considera que el diseño de los sistemas en los procesos industriales esta diseñado considerando las fallas de causa común, y el cuarto termino las fallas sistemáticas son despreciables si se utiliza una metodología para el diseño de los SIS como puede ser seguir los requerimientos y consideraciones en el diseño basado en el Ciclo de Vida de Seguridad de la IEC 61511. El valor final de la PFDavg es representada como: PFDSIS = ∑PFDSensor + ∑PFDCLP + ∑PFDEF + λSF En términos generales es aceptado el uso de las ecuaciones simplificadas para sistemas con pruebas manuales como son los sensores y elementos finales, si bien es común el uso de estas ecuaciones para los controladores lógicos programables, la norma IEC 61508 Edición 2.0 2010-04. Ha desarrollado ecuaciones mas exactas para describir a los sistemas que cuentan con pruebas basadas en auto diagnósticos. 5. Calculo de la Probabilidad de Fallas Sobre Demanda PFDavg Ecuaciones para la determinación de la Probabilidad de Fallas Sobre Demanda PFDavg para Sistemas con pruebas basadas en Auto Diagnósticos, tomadas de la norma IEC 61508-6 Edición 2.0, 2010-04. La Probabilidad de Fallas Sobre Demanda para sistemas complejos con auto diagnósticos considera las relación de fallas peligrosas totales, dadas por la suma de la relación de fallas peligrosas detectadas y no detectadas. λTot = λDU + λDD Ecuación para sistema con arquitectura 1oo1: La arquitectura consiste en canales sencillos, donde la cualquier falla peligrosa genera una falla de la función de seguridad cuando se genera una demanda: Canal Diagnosticos Figura #5 Diagrama de Bloques Fisico Determinación de la PFDavg 7 61508-6 IEC:2010 - 31 - λ Dλ D λD λ DU tc1 λ DU λ DU λ DU = T MRT c1= =T _1T _+1 +MRT _1t+c1tMRT t = T 2c12 _1 + MRT 2 2 t CE λλD DD λ DD λ DD λ DD tc2 tc2 = MTTR tc2==MTTR MTTR tc2 = MTTR tCE tCE tCE IEC IEC 325/2000 325/2000 IEC 325/2000 IEC 325/2000 Risk Software S.A. de C.V. Figure B.5++reliability 1oo1reliability reliability blockdiagram diagram FigureFigure B.5 + 1oo1 block block diagram B.5 1oo1 Figure B.5 + 1oo1 reliability block diagram λD Figures B.4 and B.5contain contain therelevant relevant blockdiagrams. diagrams. Thedangerous dangerous failure rate for the the Figures B.4 and B.5 contain the relevant block diagrams. The dangerous failure failure rate forrate thefor Figures B.4 and B.5 the block The channel is given by Figures B.4 and B.5 contain the relevant block diagrams. The dangerous failure rate for the λ λ channel is given DU DD channel is by given by T + MRT tC2 = MTTR channel is given by tC1 = 1 2 λD=+=λλDDDU++λλDD λD = λλDU D λDDU= λtDUDD+ λDD CE Figure B.5shows shows that thechannel channel can beconsidered considered comprise of two two components, components, one FigureFigure B.5 shows that the channel can becan considered to comprise of two of components, one one B.5 that the be totocomprise Figura #6 λ resulting from undetected failures and the other with with a dangerous failure rate Figure B.5 shows that the channel can be considered to comprise of two components, DU from de undetected failuresfailures and the other with awith dangerous failurefailure rate λ DU λ DU resulting from and the with othera with aaone a dangerous rateresulting Diagrama Bloquesundetected de Confiabilidad λ resulting from detected failures. It is possible to calculate the dangerous failure rate λ resulting from undetected failures and the other the with a with a dangerous failure rate DD from DU detected failures. It is possible to calculate the dangerous failurefailure rate λrate λ DD resulting from detected failures. It is possible to calculate dangerous DD resulting t , adding the individual down times from both channel equivalent mean down time λ resulting from detected failures. It is possible to calculate the dangerous failure rate CE t , adding the individual down times from both channel equivalent mean down time DD La configuración sencilla mean se ve comprometida resultante por la relación fallas peligrosas no detectables t CEla, falla adding the tanto individual downdetimes from both channel equivalent down time CE por t and t , in direct proportion to each componentNs contribution to the components, t , adding the individual down times from both channel equivalent mean down time c1 c2 t and t , in direct proportion to each componentNs contribution to the components, DU, y la relación t c1 and t c2 , in direct proportion to each componentNs contribution to theMedio Abajo components, c1 λprobability de c2 fallas peligrosas detectables λDD.CE Es posible la equivalencia del sistema para el Tiempo of of failure ofand the channel: t c1of t c2 , in direct proportion to each componentNs contribution to the components, probability of failure the channel: probability of failure the channel: (MDT) para los dosofcomponentes tC1 ychannel: tC2: probability failure of the λ tDUCE =Tλ1λDUDU TT1 1 + MRT λ DD +λλDDDD MTTR + t CE = t CE MRT + MTTR = + λ D MTTR λ 2 +TMRT λ 1 λ D t CE2λλD=D DU +λ MRT 2 λ D+ DD MTTR D 2 λD λD For every architecture, the detected dangerous failure rate and the undetected dangerous For every architecture, thedeldetected dangerous rate and undetected dangerous For every architecture, the ladetected failure ratethe and undetected dangerous Para cadarate componente canal relación dedangerous fallasfailure peligrosas no detectables y the detectables esta dada por: failure arearchitecture, given by every the detected dangerous failure rate and the undetected dangerous failurefailure rateFor are given rate are by given by failure rate are given by λDU = λD (1 − DC ) ; λDD = λD DC λDU = λλDDU (1 −=DC λDD) ;= λD DC λ (1) ;− DC = λ DC λDUD = λD (1 − DC )DD ; λDDD= λD DC For a channel with down time t CE resulting from dangerous failures For a channel with con down time resulting from dangerous failuresfailures For channel with downt CE time Paraaun canal un tiempo abajo ttCE resulta en una dangerous falla peligrosa: resulting from CEque For a channel with down time t CE resulting from dangerous failures PFD 1 − e − λD tCE − λ D=t CE PFD = 1 − e PFD =≈1 λ− et − λD tCE − λsince PFD D=CE 1 − e D tCE λD tCE << 1 ≈ λDtCE ≈ λ tsince λsince D tCE << λD1tCE << 1 D CE ≈ λDtCE since λD tCE << 1 Hence, for a 1oo1 architecture, the average probability of failure on demand is La de fallas sobrethe demanda una arquitectura 1oo1 queda establecida Hence,Hence, forprobabilidad a 1oo1 average probability of failure on demand is como: for a architecture, 1oo1 architecture, thepara average probability of failure on demand is Hence, for a 1oo1 architecture, the average probability of failure on demand is PFDG = (λDU + λDD )tCE PFDG =PFD (λDUG +=λ(DD )tCE PFDλDU= +(λλDD+)tCE λ )t G DU DD CE B.3.2.2.2 1oo2 B.3.2.2.2 1oo2 B.3.2.2.2 1oo21oo2 Ecuación para sistema con arquitectura 1oo2: connected in parallel, such that either channel can ThisB.3.2.2.2 architecture consists of two channels This architecture consists of two of channels connected in parallel, that either channel can process the safety function. Thuschannels there would have toinbe asuch dangerous failure in both channels This architecture consists two connected parallel, such that either channel can can This architecture consists of would two channels connected in parallel, such that either channel process the safety function. Thus there have to be a dangerous failure in both channels La arquitectura 1oo2function consiste en dos canales conectados paralelo, en los any cuales cada uno puede realizar la función de before a safety failed on demand. It isenassumed that diagnostic testing would process the safety function. Thus there would have to be a dangerous failure in both channels process the safety function. Thus there would have to any be adiagnostic dangeroustesting failure would in both channels beforebefore a safety function failed on demand. Itdeberán isnot assumed that only report the function faults found and would change anyforma output states orque change thedewould output a safety failed on demand. It is assumed that any diagnostic testing seguridad. En esta arquitectura ambos canales de fallar de peligrosa para la función seguridad before a safety failednot onchange demand. It output is assumed any diagnostic testing wouldfalle en only report the faults foundfunction and would any statesthat or change the output voting. only report the faults found and would not change any output states or change the output demanda. asume quefaults cualquier diagnostico deberánot ser change reportado any y la falla encontrada un cambio en el estado only Se report the found and would output statesy no or habrá change the output voting.voting. voting. final de la votación de salidas. Las figuras # 7 y 8 muestran los diagramas de bloques para la arquitectura 1oo2, tCE es calculado de la misma manera que como calculamos 1oo1, pero ahora debemos calcular tGE que esta dado por la ecuación: Customer: jose angel alvarado - No. of User(s): 1 - Company: CSIPA CONSULTORIA EN SEGURIDAD INDUSTRILA Y PROTECCION AL AMBIENTE SA DE C No.: WS-2010-007542 - IMPORTANT: This file is copyright of IEC, Geneva, Switzerland.INDUSTRILA All rights reserved. Customer:Order josefile angel alvarado - No. of User(s): 1 - Company: CSIPA CONSULTORIA EN-SEGURIDAD Y PROTECCION AL AMBIENTE SA DE CV ThisCustomer: isjose subject toalvarado a licence [email protected] Tel.: EN +41SEGURIDAD 22 02 11INDUSTRILA Customer: angel - agreement. No. of User(s): 1 - Company: CSIPA CONSULTORIA Y PROTECCION AL AMBIENTE SA DE C alvarado - No. ofEnquiries User(s): 1to- Email: Company: CSIPA CONSULTORIA EN919 SEGURIDAD INDUSTRILA Y PROTECCION Order No.: WS-2010-007542 IMPORTANT: This file is copyright of IEC, Geneva, Switzerland. All rights reserved. Determinaciónjose de -laangel PFDavg 8 AL AMBIENTE SA Order No.: WS-2010-007542 IMPORTANT: This file is copyright of IEC, Geneva, Switzerland. All rights reserved. Order WS-2010-007542 - IMPORTANT: This file is copyright- of IEC, Geneva, All rights reserved. This file is subject to No.: a licence agreement. Enquiries to Email: [email protected] Tel.: +41 22 919 Switzerland. 02 11 This file is subject to a licence agreement. Enquiries to Email: [email protected] - Tel.: -+41 9192202919 11 02 11 This file is subject to a licence agreement. Enquiries to Email: [email protected] Tel.:22+41 61508-6 IEC:2010 61508-6 IEC:2010 " 32 " " 32 " Channel Channel Risk Software S.A. de C.V. Diagnostics Canal Diagnostics 1oo2 1oo2 Channel Diagnosticos Channel 1oo2 IEC 326/2000 IEC 326/2000 Figure B.6 7 1oo2 Canal physical block diagram Figure B.6 7 1oo2 physical block diagram Figura #7 Diagrama de Bloques Fisico 1oo2 λDU λD λDU tCE λDU λD λDD λD λDD tCE λDD tCE Common Common cause failure cause failure Falla de causa Comun tGE IEC 327/2000 IEC 327/2000 tGE Figure B.7 7 1oo2 reliability block diagram tGE Figure B.7 7 1oo2 reliability block diagram Figura #8 Figures B.6 and B.7 contain the Diagrama relevant block diagrams. The value of t CE is as given in de Bloques de Confiabilidad 1oo2 Figures B.6 and B.7 contain the relevant block diagrams. The value of t CE is as given in B.3.2.2.1, but now it is necessary to also calculate the system equivalent down time t GE , B.3.2.2.1, but now it is necessary to also calculate the system equivalent down time t GE , which is given by which is given by tGE = λDU T1 λ λDU+ MRT λDD T1 + DD MTTR tGE = + MRT MTTR λD 3 λD + λD 3 λD The probability of failure on para demand for the architecture is Laaverage probabilidad de fallas sobre demanda arquitectura entonces The average probability of failure onlademand for1oo2 the queda architecture isdada por: T 2 T1 PFDG = 2 ((1 − βD )λDD + (1 − β )λDU ) tCE tGE2 + βD λDDMTTR + βλDU 1 + MRT PFDG = 2 ((1 − βD )λDD + (1 − β )λDU ) tCE tGE + βD λDDMTTR + βλ + MRT 2 DU 2 B.3.2.2.3 2oo2 B.3.2.2.3 2oo2 This architecture consists twoofchannels connected in parallel so that both channels need to Ecuación para sistema conofarquitectura 2oo2: This architecture consists two channels connected in parallel so that both channels need to demand the safety function before it can take place. It is assumed that any testing demand the safety function before it can take place. It is assumed thatdiagnostic any diagnostic testing would only report the faults found and would not change any output states or change the the La arquitectura consiste dos canales de not formachange paralelo,any ambos canales deben or de demandar a la función would only2oo2 report the en faults found conectados and would output states change output voting. output voting. de seguridad para que esta se ejecute. Se asume que cualquier diagnostico deberá ser reportado y la falla encontrada y no habrá un cambio en el estado final de la votación de salidas. Determinación de la PFDavg 9 61508-6 IEC:2010 - 33 - Channel Risk Software S.A. de C.V. Canal Diagnostics 2oo2 2oo2 Diagnosticos Channel IEC 328/200 Canal Figura #9 Diagrama de Bloques Fisico 2oo2 Figure B.8 6 2oo2 physical block diagram λD λ DU λ D λDU tCE λD λ DD λDD λ DU λDU λD λDD λ DD tCE tCE tCE IEC 329/2000 Figura #10 Diagrama de Bloques de Confiabilidad 2002 Figure B.9 6 2oo2 reliability block diagram Figures B.8 and B.9 contain the relevant block diagrams. The value of t CE is as given in La probabilidad de average fallas sobreprobability demanda queda establecida por: B.3.2.2.1, and the of failure on demand for the architecture is PFDG = 2 λD tCE B.3.2.2.4 1oo2D Ecuación para sistema con arquitectura 1oo2D: This architecture consists of two channels connected in parallel. During normal operation, both demand safety function before it canDurante take place. In addition, if the canales deben La channels arquitecturaneed 1oo2Dtoconsiste en the dos canales conectados en paralelo. la operación normal, ambos diagnostic tests in either channel detect a fault then the output voting is adapted so that the de demandar la función seguridad para que esta se ejecute. adición, If si los cada find canal detectan una overall output astate thendefollows that given by the other En channel. thediagnósticos diagnosticentests falla,inentonces la votaciónor deasalida es adaptada de tal manerabe que la operación con el canal que the se encuentra opefaults both channels discrepancy that cannot allocated to continúe either channel, then output goes to the safe state. In order to detect a discrepancy between the channels, either rando sin fallas. Si los diagnósticos encuentran una falla en ambos canales o existe una discrepancia que no es posible locachannel can determine the state of the other channel via a means independent of the other lizar en algún canal, entonces las salidas se sitúan en una posición segura. Para poder detectar una discrepancia entre los channel. The channel comparison / switch over mechanism may not be 100 % efficient canales, Kambos canales the deberán poder elofestado del otro canal decomparison forma independiente. comparación oi.e. el mecanismo de therefore represents efficiency this inter-channel / switchLamechanism, thetransferencia output maypuede remain on the 2oo2 voting even with one channel detected as faulty. que no sea 100% eficiente, por lo tanto K representa la eficiencia de los mecanismos de comparación o mecanismo de transferencia. NOTE The parameter K will need to be determined byCanal an FMEA. Diagnosticos Channel Diagnostics Diagnostics Channel 1oo2 D Diagnosticos 1oo2D Canal Figura #11 Diagrama de Bloques Fisico 1oo2D Determinación de la PFDavg IEC 330/2000 Figure B.10 6 1oo2D physical block diagram 10 Risk Software S.A. de C.V. 61508-6 IEC:2010 61508-6 IEC:2010 " 34" "34 " λDU tGE′ tGE′ λDU 61508-6 IEC:2010 λDU GE Falla de Causa Common ComunCommon cause failure cause failure λDU λDU λ tGE′ DU tCE′ t ′ CE " 34 " λDUt λDD λ DD λDU λDD λSD λ SD λDD tCE Common cause failure IEC 331/2000 λSD λSD IEC 331/2000 tCE′ Figure B.11B.11 4 1oo2D reliability block diagram Figure 4 1oo2D reliability block diagram Figura #12 IEC 331/2000 Diagrama de Bloques de Confiabilidad 1oo2D Figure 4 1oo2D reliability block diagram TheThe detected safesafe failure rateB.11 for every channel is given by by detected failure rate for every channel is given The detected safedefailure rate for every channel given bydada por: λSDcada =λis λScanal DC La relación fallas seguras detectadas para esta = λ DC SD S λSD = λblock S DC Figures B.10B.10 and and B.11B.11 contain the the relevant diagrams. The The values of the equivalent Figures contain relevant block diagrams. values of the equivalent mean down times differ from those given for the other architectures in B.3.2.2 and hence are are mean down times differ from those given for the other architectures in B.3.2.2 and hence Aquí los valores de equivalencia de are de Tiempo Medio Abajo están dados por : labelled t ′ and t ′ . Their values given by Figures B.10 B.11 relevant block CE t CE ′ and GE t GEcontain labelled ′ . Their the values are given by diagrams. The values of the equivalent mean down times differ from those given for the other architectures in B.3.2.2 and hence are labelled t CE ′ and t GE ′ . Their values are T given by 1 1 λDU λDU + (λDD (+ λSD )MTTR +TMRT ) 2 2 + MRT + λDD + λSD MTTR tCE ' =t ' =T CE )λ)SD λDU +λDU ) λDU 1 + MRT +(λ(DD λ+DD(+λDD +λSD λ+SD MTTR 2 tCE ' = λDU + (λDD + λSD ) T 1 tGE ' =tGE1' =+TMRT + MRT 3 3 T tGE ' = 1 + MRT 3 for the TheThe average probability of failure on demand is is average probability of failure on demand for architecture the architecture La probabilidad de fallas bajo demanda para la arquitectura 1oo2D queda dada por: The average probability of failure on demand for the architecture is T1 T1 (1 − β )λ β ()(λ1DU− (β(1)λ−DUβ )+λDU(1 −+ β(1D−)λβDDD )+λDDλSD+)tλCESD' )ttGECE''+t2GE(1' +−2K(1)−λDDKt)CEλDD' +tβλ PFDPFD + βλDU + MRT + MRT G = 2 G = 2 (1 −DU CE 'DU 2 2 T PFDG = 2 (1 − β )λDU ((1 − β )λDU + (1 − β D )λDD + λSD )tCE ' tGE ' +2 (1 − K )λDDtCE ' + βλDU 1 + MRT 2 B.3.2.2.5 2oo3 B.3.2.2.5 2oo3 Ecuación para sistema con arquitectura 2oo3: ThisThis architecture of three channels connected in parallel withwith a majority voting architecture consists of tres three channels connected a majority voting B.3.2.2.5 2oo3consists La arquitectura 2oo3 consiste en canales conectados en paraleloinconparallel un arreglo de votación a la salida, aquí el estado de arrangement for for the the output signals, suchsuch that that the the output statestate is not changed if only one one arrangement output signals, output is not changed if only lasgives salidasano difiere siresult solo unwhich canalchannels muestra discrepancia otroschannels. dos canales. Se asumevoting que cualquier diagnostico channel different disagrees withwith the con other two This architecture consists of result three connected in los parallel a majority channel gives a different which disagrees the other two with channels. deberá for ser reportado y la falla encontrada no habrá cambiostate en el estado de la votación de salidas. arrangement the output signals, suchy that theunoutput is notfinal changed if only one channel gives a different result which disagrees with the other two channels. It isItassumed thatthat any any diagnostic testing would onlyonly report the the faults found and and would not not is assumed diagnostic testing would report faults found would Canal change any any output states or change the the output voting. change output states or change output voting. Diagnosticos It is assumed that any diagnostic testing would only report the faults found and would not change any output states or change the output voting. Canal 2oo3 Channel Channel Channel Diagnostics Diagnostics Canal Channel Channel Diagnostics Figura #13 Determinación de la PFDavg 2oo32oo3 Diagrama de Bloques Fisico 2oo3 Channel Channel Channel 11 2oo3 IEC 332/2000 IEC 332/2000 61508-6 IEC:2010 61508-6 IEC:2010 - 35 λD λ DU λ DU tCE λ DD λD - 35 - λ DD tCE 2oo3 λD λ DU λ DD tCE Common cause failure Common 2oo3 cause failure Risk Software S.A. de C.V. tGE tGE 2oo3 Common cause failure IEC 333/2000 tGE IEC 333/2000 Falla de causa Comun Figure B.13 6 2oo3 reliability block diagram Figure B.13 reliability block diagram tGE 6 2oo3 λD IEC 333/2000 Figures B.12 and B.13 contain the relevant block diagrams. The value of t CE is as given in λDD λDU is as the given in B.3.2.2.2. The average of failure on in B.3.2.2.1 andB.12 the and valueB.13 of t GE tCE Figures contain relevant block diagrams. The probability value of t CE is as given demand for the architecture is B.3.2.2.1 and the value of t GE is as given in B.3.2.2.2. The average probability of failure on Figure B.13 reliability block diagram Figura #14 demand for the architecture is 6 2oo3 Diagrama de Bloques de Confiabilidad 2oo3 T1 2 ( ( ) PFD = 6 1 − β λ + (1 −relevant β )λDU ) tCE tGE + diagrams. βD λDDMTTRThe + βλvalue + MRT D DD the DU Figures B.12 andG B.13 contain block 2 ofT1t CE is as given in 2 given in B.3.2.2.2. The average probability of failure on B.3.2.2.1 and thePFD value of t − β Dis)λas + MRT G = 6 ((1 GE DD + (1 − β )λDU ) tCE tGE + β D λDD MTTR + βλDU 2 demand for the architecture is La probabilidad de fallas sobre demanda para la arquitectura 2oo3 se establece como: B.3.2.2.6 1oo3 B.3.2.2.6 1oo3 This architecture three channels2 connected in parallel withT1a +voting PFDGconsists = 6 ((1 − βof MRT arrangement D )λDD + (1 − β )λDU ) tCE tGE + β D λDD MTTR + βλDU 2 for the output signals, such that the output state follows 1oo3 voting. This architecture consists of three channels connected in parallel with a voting arrangement for the output signals, such that the output state follows 1oo3 voting. It is assumed that any B.3.2.2.6 1oo3 Ecuación para sistema con diagnostic arquitecturatesting 1oo3: would only report the faults found and would not change output states or change the testing output voting. It isany assumed that any diagnostic would only report the faults found and would not This architecture consists of three channels connected in parallel with a voting arrangement La arquitectura 1oo3output consistestates en tres or canales conectados en paralelo change any change the output voting.con un arreglo de votación de salida de 1oo3, cualquier for the output signals, such that the output state follows 1oo3 The reliability diagram will be the same as for the 2oo3 case voting. but with voting 1oo3. The value falla detectada por diagnósticos ocasionara que el sistema se posicione en falla segura. Se asume que cualquier diagnostico is reportado as giveny diagram B.3.2.2.1 and value of as case given in laB.3.2.2.2. average of t CE deberá ser lainfalla encontrada ythe nothe habrá unas cambio en is el estado finalbut de votación deThe salidas. The reliability will be same for t GE the 2oo3 with voting 1oo3. The value It is assumed that any diagnostic testing would only report the faults found and would not probability failure on demand for the architecture isoutput as given in B.3.2.2.1 and the value is of t is as given in B.3.2.2.2. The average of t anyof change CE states or change the output voting. GE La probabilidad de fallas sobre on demanda parafor la arquitectura 1oo3 seisestablece como: probability of failure demand the architecture T1 )3 tCE PFD β D )λbe + (1 same − β )λDUas tG 2 E2oo3 + βD λcase βλDUvoting + MRT The reliability diagram fortGEthe but+with 1oo3. The value G = 6 ((1 −will DD the DD MTTR 2 T1 The average 3 of t in B.3.2.2.1 and the value is as given in B.3.2.2.2. of t CE is as given GE PFD = 6 ((1 − β D )λDD + (1 − β )λDU ) tCE tGE tG 2 E + βD λDDMTTR + βλDU + MRT probability of failure Gon demand for the architecture is 2 Where IEC 332/2000 Donde: T1 Where PFD = 6 ((1 − β )λ + (1 − β )λ )3 t t t + β λ MTTR + βλ MRT G D DD DU CE GE G 2 E D DD DU IEC +332/2000 2 λDU T1 λDD tG 2 E = + MRT + MTTR λD λ4DU T1 λD λDD tG 2 E = + MRT + MTTR Where λD 4 λD IEC tG 2 E = 332/2000 λDU T1 λ + MRT + DD MTTR λD 4 λD Determinación de la alvarado PFDavg - No. of User(s): 1 - Company: CSIPA CONSULTORIA EN SEGURIDAD INDUSTRILA Y PROTECCION 12 Customer: jose angel AL AMBIENTE SA DE C Order No.: WS-2010-007542 - IMPORTANT: This file is copyright of IEC, Geneva, Switzerland. All rights reserved. This file is subjectjose to aangel licence agreement. [email protected] - Tel.: +41 22EN 919SEGURIDAD 02 11 Customer: alvarado - No. Enquiries of User(s):to1Email: - Company: CSIPA CONSULTORIA INDUSTRILA Y PROTECCION AL AMBIENTE SA Order No.: WS-2010-007542 - IMPORTANT: This file is copyright of IEC, Geneva, Switzerland. All rights reserved. This file is subject to a licence agreement. Enquiries to Email: [email protected] - Tel.: +41 22 919 02 11 Risk Software S.A. de C.V. Cuantificación del Efecto de las Fallas de Causa Común: Los cálculos de PFDavg deberán incorporar el efecto que causan las fallas de causa común en los sistemas redundantes, en la seguridad funcional es común utilizar la metodología de factor Beta (β) para determinar la falla de causa común. en un articulo técnico posterior describiremos como se determina este factor. El efecto final en la ecuación de PFDavg del factor de causa común se representa con la siguiente ecuación: PFDFCC = ( PFDa x PFDb x..... PFDn ) + (β x PFDPeor) Donde: PFD a.....n representa la probabilidad de falla sobre demanda del dispositivo a al n. PFDPeor representa la probabilidad de fallas sobre demanda del dispositivo mas débil o peor. Beta (β) representa el factor de falla común. 6. Arquitecturas Redundantes Arquitecturas de sistemas redundantes para Diagramas de Bloques. E A FALLA DE CAUSA COMUN B S Figura #15 2oo2 A FALLA DE CAUSA COMUN E S B Figura #16 1oo2 Determinación de la PFDavg 13 Risk Software S.A. de C.V. A A B FALLA DE CAUSA COMUN E B C S C Figura #17 2oo3 E A B C FALLA DE CAUSA COMUN S Figura #18 1oo3 Determinación de la PFDavg 14 Risk Software S.A. de C.V. Arquitecturas de sistemas redundantes para Arboles de Falla. Bloques OR (Se Suman). Bloques AND (Se Multiplican) Salida A OR AND OR OR Salida FCC B A FCC B Figura #20 1oo2 Figura #19 2oo2 Salida OR OR OR Salida AND A AND AND AND FCC B FCC C Figura #22 1oo3 A B A C B C Figura #21 2oo3 Determinación de la PFDavg 15 Risk Software S.A. de C.V. 7. Ejemplos de Determinación de PFDavg. Podemos modelar la PFDavg de un sistema utilizando diagramas de bloques utilizando en las siguientes simplificaciones: ✓ Cadenas en paralelo se Multiplican. ✓ Cadenas en serie se Suman. Ejemplo: Considere el siguiente sistema de protección de presión a la entrada de una plataforma marina que maneja grandes volúmenes de gas natural, una sobre presión podría generar un gran impacto ocasionando ruptura de la tubería y generando una fuga mayor que podría incluso generar un gran fuego o explosión: ENTRADAS LOGICA PT-9002A D PT-9002B PT-9002C SALIDAS SVA FALLA DE CAUSA COMUN FALLA DE CAUSA COMUN TMR ESDV H SVB Considere Arquitectura 2oo3 Determinación de la PFDavg 16 Risk Software S.A. de C.V. Se cuenta con los siguientes datos: Valores PT (FIT) λsd λsu λdd ISA-TR84.00.02-2002 - Part 2 " 24 " λdu TMR (FIT) Solenoide (FIT) Válvula Corte (FIT) 396 71 0 440 0 1401 52 99 0 69 1 765 ---- ---- 1 año 1 año 8 hr 8 hr 5% ---- If systematic errors (functional failures) are to be included in the calculations, separate values for each sub-system, if available, may be usedSFF in the equations above. An alternate approach is to use a single 92.8% value for functional failure for the entire SIF and add this term as shown in Equation 1a in 5.1.6. TI 1 año NOTE Systematic failures are rarely modeled for SIF Verification calculations due to the difficulty in assessing the failure modes and effects and the lack of failure rate data for various types of systematic failure. However, these failures are extremely important 8 ANSI/ISA-84.01-1996, hr and can result in significant impact to the SIF MTTR performance. For this reason, IEC 61508, and IEC 61511 provide a lifecycle process that incorporates design and installation concepts, validation and testing criteria, and management of change. This lifecycle process is intended to support the reduction in the systematic failures. SIL Verification is therefore β 5% predominantly concerned with assessing the SIS performance related to random failures. 2.5 x cause 10- and The simplified equations without thePFDavg terms for multiple failures during repair, common systematic errors reduce to the following for use in the procedures outlined in 5.1.1 through 5.1.4. 4 1oo1 Problema: Dibujar el diagrama de bloques para el sistema y calcular el valor de PFDavg para el sistema: (Eq. No. 3a) PFDavg = $ DU % 1oo2 Solución con Diagramas de Bloques: Lo primero que debemos realizar es calcular los valores de PFDavg para cada TI 2 bloque, para$ DU esto2 % utilizamos la formula: TI 2 (Eq. No. 4a) 1oo3 [( ] ) PFDavg = 3 1) Para los transmisores tenemos: PFDavg = (69 x10-9 x 8760)/2 = 3.02 x10-6 (Eq. No. 5a) [( $ DU ) 3 % TI 3 ] -6 -6 -12 PFDavg PFD =(A x B) = 3.02 x10 x 3.02 x10 = 9.13 x 10 avg 4 PFDavg (A x C) = 3.02 x10-6 x 3.02 x10-6 = 9.13 x 10-12 2oo2 (Eq. No. 6a) -6 -6 -12 PFDavg PFD =(B$xDUC)%=TI3.02 x10 x 3.02 x10 = 9.13 x 10 2oo3 PFDFCC = (3.02 x10-6 x 3.02 x10-6 x 3.02 x10-6) + (0.05 x 3.02 x10-6 ) = 1.51 x 10-07 (Eq. No. 7a) -6 -6 -6 -6 2 PFDavg PFD == (3.02 $ DU ) x10 % TI 2+ 3.02 x10 + 3.02 x10 = 9.07 x 10 2oo4 PFDavg tot = 9.07 x 10-6 + 1.51 x 10-07 = 9.21 x 10-06 (Eq. No. 8a) 3 4 3 2) Para el PFD = controlador $DU % (TI ) lógico tenemos PFDavg = 2.5 x 10- 5.1.6 avg avg avg ( ) Determinación dePFDs la PFDavg Combining components’ to obtain SIF PFDavg Once the sensor, final element, logic solver, and power supply (if applicable) portions are evaluated, the overall PFDavg for the SIF being evaluated is obtained by summing the individual components. The result is the PFDavg for the SIF for the event being protected against. 17 Risk Software S.A. de C.V. 3) Para las Válvulas Solenoides Tenemos: PFDavg = (1 x10-9 x 8760)/2 = 4.38 x10-6 PFDavg = (4.38 x10-6 x 4.38 x10-6) = 1.91 x 10-11 PFDFCC = (4.38 x10-6 x 4.38 x10-6 ) + (0.05 x 4.38 x10-6 ) = 2.19 x 10-7 PFDavg tot = 1.91 x 10-11 + 2.19 x 10-7 = 2.19 x 10-7 4) Para la válvula de corte tenemos PFDavg = (765 x10-9 x 8760)/2 = 3.35 x10-3 El valor de PFDavg para el SIS será: PFDavg SIS = 9.21 x 10-06 + 2.5 x 10-4 + 2.19 x 10-7 + 3.35 x10-3 = 3.61 x10-3 FRR = 277 SIL2 Determinación de la PFDavg 18 Risk Software S.A. de C.V. Solución con Arboles de Falla: Falla SIS OR OR OR 3.61 x10-3 PT OR OR 9.21 x 10-06 SV 2.19 x 10-7 SV 2.19 x 10-7 OR 1.51 x 10-07 OR 9.07 x 10-6 FCC FCC A B A C 3.02 x10-6 B CLP C A 2.5 x 10-4 2.19 x 10-7 AND AND AND AND 1.91 x 10-11 FCC SCV B 4.38 x10-6 3.35 x10-3 Los valores mostrados en los eventos iniciales están dados en PFDavg Determinación de la PFDavg 19 Risk Software S.A. de C.V. Ejemplo: Cálculos utilizando FTA-Pro de Dyadem Determinación de la PFDavg 20 Risk Software S.A. de C.V. Resultados al Tiempo: 8760 Falta de disponibilidad 0.007206 Frecuencia: N/A Tiempo Falta de disponibilidad 0.00000 0.000000 796.36364 0.000657 1592.72727 0.001314 2389.09091 0.001970 3185.45455 0.002626 3981.81818 0.003282 4778.18182 0.003937 5574.54545 0.004592 6370.90909 0.005246 7167.27273 0.005900 7963.63636 0.006553 8760.00000 0.007206 Total de Tiempo Sistema Parado 30.972005 PFDavg: 0.003536 FRR = 282 SIL=2 Determinación de la PFDavg 21 Risk Software S.A. de C.V. Los comentarios de este documento expresan el punto de vista de: Victor Machiavelo Salinas TUV FS Expert ID-141/09 Risk Software SA de CV [email protected] www.risksoftware.com,mx Agradeceremos cualquier comentario. Determinación de la PFDavg 22