SIL Hardware Architecture & Fault Tolerance 1 — Questions and Answers
Question 1: What does Hardware Fault Tolerance (HFT) represent in a safety-related subsystem?
- The number of faults a subsystem can withstand while maintaining its safety function (Correct answer)
- The fraction of failures that result in a safe state
- The required interval between proof tests
- The maximum acceptable probability of failure on demand
Correct answer: The number of faults a subsystem can withstand while maintaining its safety function
HFT = N means the subsystem can tolerate N faults and still perform its required safety function.
Question 2: According to IEC 61508, how is a 'Type A' subsystem characterized?
- A complex subsystem containing components whose failure modes are not fully defined
- A software-only safety controller with no hardware element
- A subsystem in which failure modes of all components are well defined and can be fully enumerated (Correct answer)
- A subsystem requiring HFT ≥ 1 regardless of Safe Failure Fraction
Correct answer: A subsystem in which failure modes of all components are well defined and can be fully enumerated
Type A subsystems have well-defined, fully enumerable failure modes for all components, distinguishing them from the more complex Type B.
Question 3: Which statement correctly describes Safe Failure Fraction (SFF) per IEC 61508?
- The probability a system fails to reach a safe state when a demand occurs
- The fraction of total failures that are either safe failures or dangerous but detected failures (Correct answer)
- The percentage of operating time without any recorded failure
- The ratio of detected dangerous failures to all dangerous failures
Correct answer: The fraction of total failures that are either safe failures or dangerous but detected failures
SFF = (λS + λDD) / (λS + λDD + λDU), combining safe and detected dangerous failure rates as a fraction of all failures.
Question 4: What is the Hardware Fault Tolerance of a 1oo2 (1-out-of-2) voting architecture?
- HFT = 0
- HFT = 1 (Correct answer)
- HFT = 2
- HFT = 3
Correct answer: HFT = 1
In 1oo2, one channel can fail while the other still initiates the safety action, so the system tolerates one fault — HFT = 1.
Question 5: Why is a 2oo3 voting architecture commonly preferred over 1oo2 in process safety applications?
- It requires fewer installed components than a 1oo2 system
- It reduces spurious trips compared to 1oo2 while still maintaining HFT = 1 (Correct answer)
- It completely eliminates common cause failure contributions to PFD
- It removes the need for periodic proof testing once commissioned
Correct answer: It reduces spurious trips compared to 1oo2 while still maintaining HFT = 1
2oo3 requires agreement from two channels for the safety action, reducing spurious trips versus 1oo2, while a single failure still leaves two channels able to act (HFT = 1).
Question 6: What is the primary purpose of architectural constraints defined in IEC 61508-2?
- To specify proof test intervals based on device failure rates
- To define minimum software complexity for each SIL level
- To establish maximum response times for safety instrumented functions
- To set minimum hardware redundancy requirements based on target SIL and subsystem type (Correct answer)
Correct answer: To set minimum hardware redundancy requirements based on target SIL and subsystem type
Architectural constraints in IEC 61508-2 Tables 2 and 3 mandate minimum HFT values based on the target SIL and whether the subsystem is Type A or Type B.
Question 7: In safety architecture notation, what does the 'D' suffix signify in a 1oo1D configuration?
- The system uses dual redundant power supplies
- Direct wiring to the final element without intermediary relays
- Diagnostics are incorporated to detect and respond to dangerous failures (Correct answer)
- Deterministic response timing is guaranteed by the controller
Correct answer: Diagnostics are incorporated to detect and respond to dangerous failures
The 'D' in 1oo1D denotes that diagnostic coverage is used to detect dangerous failures and initiate a safe response, compensating for the lack of hardware redundancy.
What does Hardware Fault Tolerance (HFT) represent in a safety-related subsystem?