(70e) Introducing a Dynamic Calculation Method for Pressure Relief Load and Orifice Sizing

This paper presents a dynamic calculation method for pressure relief load and orifice sizing.   This method is suitable for complex systems such as multi-phase, multi-component mixtures in pool fire scenarios such as initial oil-water-gas separation in the upstream area, but can also be extended to other types of problems, such as different types of equipment (i.e. tanks), and different types of scenarios, such as maximum heat input from a heater.

When complex systems relieve pressure due to excessive heat input such as external pool fires, for years most engineers have made several simplifying assumptions to allow for rapid determination of relief load, orifice sizing and subsequent activities in the over-pressure protection field.  How good are these assumptions?  Oftentimes, these are not good and result in systems which are undersized and calculations which provide little insight into what is physically going on during the transient event.

This is important because an undersized over-pressure protection system can result in catastrophic issues such as loss of containment resulting in fire, explosion or toxic effects. 

Also, lack of insight of the transient event does not allow one to know how long it will take before loss of containment is expected to occur.  Gaining knowledge of the dynamics allow for response to better mitigate impact of the event.

This problem is well suited to dynamic simulation.  Unfortunately, many engineers do not have access to dynamic process simulation software, which is generally more expensive than the more commonly available steady state process simulation software.

This method is one in which a base spreadsheet is used to determine the constant volume parameters, initial conditions and boundary conditions.  An appropriate time step is then chosen, and through transformation of variables the problem is converted to a constant flow rate process which then allows for solution in a steady state process simulator.  The process simulation results are then placed back in the spreadsheet, and the calculation is complete for that time step, and the starting conditions for the next time step are calculated.  After the pressure relief event occurs, the orifice size is automatically calculated at the end of each time step.  This process is repeated until enough results are available to have confidence that the maximum orifice size has been calculated and is the one which should be specified.

By using this method, one can know many items including the time it takes until the pressure relief event starts, the fluid being relieved during the event, the liquid level in the vessel, and the time versus temperature profile so important to the evaluation of the vessel integrity.  Sometimes the liquid in the vessel may initially swell and compress the vapor.  Therefore, the initial relief may be vapor, but if the liquid then completely fills the vessel, the relieving fluid will be liquid.  As boiling becomes more vigorous, it may next relieve a 2 phase mixture, with a final period of vapor release.  There may also be a supercritical relief phase. This sequence will not be understood and properly designed for unless a dynamic analysis is conducted.

Many times when pressure relief work by others is reviewed, the documentation is lacking or the underlying thoughts are not stated.  This can be problematic when the facility is expanded, since the basis needs to be recreated.  This method is thoroughly documented, including an extensive section on scenario selection which has about 100 scenarios identified and requires a written response to each one.

Several worked examples will be discussed, showing some of the significantly different results obtained with the more detailed method and the profound insights gained.  Use of this method will bring many of the benefits of dynamic simulation to the majority of engineers, without the associated cost, resulting in safer relief system design.