The process safety time is the time from the moment a failure starts the process towards the hazard until the hazard occurs if nothing acts. The SIF must detect, decide and bring the process to the safe state within that time, with margin.
Establishing it
- From dynamic simulation or a hand calculation of the transient (pressure rise rate in a blocked vessel, level rise rate at maximum inflow, temperature rise in a runaway).
- From the trip point, not from normal operation: the time available to the SIF starts when the measured variable crosses the setpoint.
- Using conservative assumptions: maximum inflow, worst case composition.
The response time budget
Response time = sensor response + logic scan and processing + time delays + final element travel + process response to the action.
Worked example
Separator: the pressure rises from the trip point (92 % of design) to design pressure in 90 s at the maximum blow-by rate.
| Element | Time |
|---|---|
| Transmitter response | 1 s |
| Logic solver (scan, output) | 0.5 s |
| Solenoid venting | 2 s |
| Valve stroke | 15 s |
| Pressure stops rising after closure | 5 s |
| Total | 23.5 s |
Against 90 s process safety time this leaves ample margin; a common company rule requires the response to be no more than half the process safety time. A 45 s valve would not meet that rule.
Choosing the setpoint
A setpoint too close to the limit leaves no time; too close to operation causes spurious trips. The setpoint, the response time and the process safety time must be decided together and written in the SRS.
In First Out AI
Fill "Process safety time", "Required response time" and "Trip setpoints" in the specification details and the stroke time per final element. The proof test draft uses the stroke time and response time as acceptance criteria.
Written for First Out AI as engineering guidance. It supports engineering work and does not replace it; check every value against your standards, certificates and company rules with a competent functional safety engineer.