PTQ Q3 2022 Issue

against fire while the PSV is removed. A spare PSV with interlocking isolation valves, as described in Section 4, would also be a robust alternative. Imagining an operator instructed to operate a bypass at the top of a 100-foot tower during an impending overpres- sure scenario should help bring some perspective to this practice. The risk of the operator intervention failing must be considered, without any credit provided by limited occu- pancy. Stationing an operator at a PSV bypass for cases with high-severity consequences is unlikely to satisfy client risk tolerances without additional safeguards. Recommendations Without careful design considerations and additional safe- guards, it is unlikely the practice of stationing an operator at the PSV bypass will be as reliable as a PSV and can expose the operator to unnecessary risk. If the requirement for removing the PSV while the ves- sel is in service is identified in the early design phases, the installation of a 100% capacity spare PSV with interlocking valves or a three-way changeover valve can allow seam- less switch-over during operation See API-520 Part 2 Section 8.3.4 for additional details. 9 This would avoid any need to station an operator at the bypass or overpressure protection. Basing a project schedule and budget on the assumption that using an operator stationed at a bypass will be sufficient will likely result in schedule delays as the shortcomings are uncovered An alternative could be to provide an adequately sized, remotely activated hand valve on a separate PSV bypass to depressurise the vessel to flare. A full port remote hand valve might be sized in series with a downstream restric- tion orifice to help restrict flow. The overall system, includ- ing any automatic or manual activation, would need to be assessed for reliability. In some cases, a single PSV can be appropriate if the protected vessel can be isolated and de- inventoried when the PSV is removed. Fire, thermal relief, and valve leakage considerations are common drivers for the requirement to de-inventory the vessel after it is iso- lated and the PSV removed. If it is unavoidable to shut down the unit when a PSV requires servicing, and the owner is intent on using the practice of stationing an operator at the bypass, it is recom- mended to: • Ensure the practice is allowed under the local jurisdiction • Ensure adequate engineering assessments of operator response time, sizing of the bypass, and reliability/risk are completed. Provide additional safeguards as required • Ensure the bypass valve is identified as a safety-critical element as part of the MOC process and that adequate testing/maintenance is performed

• Rotate the operator at the bypass to avoid fatigue over a shift. Conclusions T his article has demonstrated that the mere existence of a PSV bypass is not sufficient to ensure that operator inter- vention via the bypass is adequate for overpressure protec- tion. If it is expected that the frequency of PSV maintenance is higher than the unit turnaround, it may be prudent to include a spare PSV and avoid the practice of stationing an operator at the bypass. Many vessels will have the potential for a pool fire, and it would be impossible to ensure a safe egress route for the operator at the top of the vessel in the event of a fire. Even if there was not a fire case (in most other upsets), it is not expected that the operator will have sufficient time to act based on industry-standard response time. It is also expected that operator intervention alone will not be suffi- ciently reliable unless additional SIS safeguards are present. Basing a project schedule and budget on the assump- tion that using an operator stationed at a bypass will be sufficient will likely result in schedule delays as the short- comings are uncovered. If it is unavoidable to rely on an operator stationed at the PSV bypass (for example at an existing facility), then careful design evaluations and risk review are required. In the author’s experience this practice would not be appropriate for most overpressure cases, and it should find limited application for cases with only minor unmitigated consequences that have sufficient response time and bypass capacity. * Note: Chatter is the opening and closing of a pressure-relief valve and a very high frequency (on the order of the natural frequency of the valve’s spring mass system). References 1 “API STANDARD 520, Sizing, Selection, and Installation of Pressure- relieving Devices, Part 1- Sizing and Selection,” 2020. 2 M. Hellemans, The Safety Relief Valve Handbook, Elsevier, 2009. 3 CCPS, Guidelines for Pressure Relief and Effluent Handling Systems, Wiley, 2017. 4 “API STANDARD 521, Pressure-relieving and Depressuring Systems,” 2020. 5 “ASME BPVC.XIII, ASME Boiler and Pressure Vessel Code. Rules for Overpressure Protection,” 2021. 6 “ASME BPVC.1 ASME Boiler and Pressure Vessel Code. Rules for Construction of Power Boilers,” 2021. 7 Pipeline Simulation and Integrity Ltf, “Testing and Analysis of Relief Device Opening Times,” 2002. 8 ABSA, “Overpressure Protection Requirements AB-525, Ed. 2,” 2021. 9 “API-STANDARD 520, Sizing, Selection, and Installation of Pressure-relieving Devices, Part II - Installation,” 2020. Jonathan Webber is a principal process engineer with Fluor Canada Ltd, where he is a member of the Fluor Global Overpressure Protection SME group. He holds a M.Eng (McGill) and Ph.D. (Dalhousie) in Chemical Engineering, and has over 16 years of experience in EPC. Email: Jonathan.Webber@Fluor.com

PTQ Q3 2022