Depressurization System Safety Review: A HAZID-Based Approach

The revised design did not foresee a new incinerator. Most PSV relief streams were routed to the flare system through a knock-out drum, while selected pressure-control discharges were retained through the existing incineration route.

For the reviewed process area, PSV reliefs were routed directly to the flare, and absorber credit was not assumed for the emergency relief composition. The flare composition basis was therefore required to reflect the revised routing without relying on absorber treatment during relief conditions.

For ammonia-related relief systems, the review supported a rupture disc plus PSV arrangement as the provisional design solution, due to corrosion and leakage concerns affecting relief valve reliability. Vendor alternatives could still be considered if a more robust corrosion-resistant solution was proposed.

The review identified existing rupture disc plus PSV arrangements without a clearly documented basis. The decision was to move toward a consistent design philosophy by retaining rupture discs where justified and removing arrangements that did not meet the agreed philosophy.

For ammonia storage and anhydrous ammonia relief, the final disposal route was not treated as a simple routing decision. Flare routing, cold venting and direct venting options were considered, with the final decision linked to atmospheric dispersion modelling and consequence evaluation.

For 100% ammonia relief, a new dedicated cold vent was foreseen in the design. The associated dispersion modelling was identified as a required basis for confirming the suitability of this release path.

The relief load basis treated the relevant fire zones conservatively, including cases where fireproofing was not credited. Where elevated equipment or local fire exposure assumptions could affect relief loads, the design required further confirmation against actual site conditions.

Gas breakthrough scenarios were reviewed to confirm whether high-pressure gas could unintentionally pass into downstream low-pressure systems and create additional overpressure or relief load case. The engineering team was expected to identify relevant gas breakthrough cases and include them in the corresponding PSV datasheets during basic design package preparation.

Tube rupture scenarios were reviewed as part of the relief design basis. The engineering team confirmed that these scenarios had been evaluated. However, the owner was expected to identify any additional field configuration or reactor protection arrangement that may not have been reflected in the design.

The flare header network had to be finalized before the relief design could be considered mature. The review confirmed that the flare header network document was completed, but also identified that detailed engineering calculations should be checked against verified field configuration because routing documentation may not fully represent the existing installation.

The review established that PSV type selection could not be finalized until flare header backpressure was confirmed. The engineering team was expected to define the required PSV types after backpressure confirmation, and the owner would compare these requirements against the existing installed equipment.

Potential PSV chattering was not accepted as a generic assumption. A list of potentially affected valves was to be developed and reviewed with instrument discipline and relevant vendors before the selected valve arrangements could be accepted.

For ammonia storage pressure management, the use of pressure-control-based balancing rather than relying only on relief valves was considered. The decision was that this concept would require renewed HAZOP review before being accepted as part of the operating or protection philosophy.