The Impact of Two-Phase Flow – Emergency Relief System Design

By Elizabeth Raines & Benjamin Doup, PhD

Ensure your emergency relief system is adequately sized by accounting for two phase flow

Introduction to Two Phase Flow

In order to ensure your vessel is appropriately protected from possible overpressurization scenarios, it is crucial to consider multiphase or two-phase flow. The presence of two-phase flow can increase the required size of your relief device, relief piping, and/or the effluent handling systems.

In the context of emergency relief system design, two‑phase vapor-liquid flow is very common due to the volume expansion of the initially all-liquid pool in the vessel caused primarily by vapor/gas generation and to a lesser extent by the reduced liquid density at increased temperatures. The extent of the volume expansion depends on the fill fraction (liquid volume fraction) of the vessel at the onset of venting, the two-phase flow regime of the fluid, and the rate of vapor/gas generation. When the fill fraction or the rate of vapor/gas generation increase, the likelihood of two‑phase flow in the relief system increases. The flow regime characterization depends upon the fluid, vessel geometry, and upset scenario.

Multiple two-phase flow regimes were modeled in the original DIERS research project [1] and are based upon a drift-flux modeling approach. Assuming uniform vapor generation, a fluid can be classified as churn-turbulent, bubbly, or homogeneous (or foamy). A homogeneous flow regime indicates there is no vapor-liquid disengagement. A bubbly flow regime indicates minimal vapor-liquid disengagement, and a churn-turbulent flow regime results in significant vapor-liquid disengagement. The models for these flow regimes have been benchmarked against large scale experimental data [1,2]. If the flow regime of a fluid is unknown, blowdown testing can be used to classify the flow regime (https://www.fauske.com/blog/flow-regime-determination-in-emergency-relief-system-design-blowdown-testing and https://www.fauske.com/blog/flow-regime-characterization-in-emergency-relief-system-design).

If the upset scenario is fire exposure, vapor/gas generation may preferentially occur at the vessel wall. This wall-boiling flow regime was developed for liquid filled storage vessels and relies upon natural circulation currents that aid in vapor-liquid disengagement [3‑5]. Care should be taken in applying this flow regime, as it is not always applicable. For instance, the wall-boiling regime is not applicable for smaller diameter vessels (due to merging of the two-phase boundary layers) or vessels with agitation (due to the inability to develop natural circulation currents). Interested readers should refer to Fisher and Forest [6] and Fauske [7] for additional discussions on the wall-boiling flow regime.

To investigate the impact of two-phase flow and the fluid flow regime on the required ideal vent area, the discharge flow rate and the vapor quality of the discharge flow, a case study is performed here that compares a hypothetical non-reactive fire exposure scenario to an agitated reactor vessel scenario.

Comparison of ERS Design Results Based on Flow Regime

This relief system design problem was evaluated within FERST Powered by CHEMCAD [2]. The Leung Omega methodology for vapor systems [3] was utilized with API 520/521 methodology [4, 5] for the fire heat input calculation. The following flow regimes were considered: vapor only, churn turbulent, bubbly, and homogeneous. The material properties were assumed to be those for dichloromethane and were obtained within FERST Powered by CHEMCAD. The vessel parameters utilized are shown in Table 1. Evaluations were conducted considering two different void fractions at the time of relief, and two different set pressures. Some example parameters that can be compared include ideal vent area/diameter, discharge mass flow rate, exit quality, and mass remaining at the turnaround time. Table 2 provides definitions for these parameters, and the results of the evaluations are documented in Table 3 through Table 5. 

Table 1: Vessel Parameters

Table 2: ERS Parameter Definitions

Table 3: Test Results – Vessel 100% Full with a 10 psig Set Pressure

Table 4: Test Results – Vessel 70% Full with a 10 psig Set Pressure

Table 5: Test Results – Vessel 100% Full with a 50 psig Set Pressure

Conclusion

Comparing the results in Table 2 through Table 4 provides an example of how two-phase flow (and specific flow regime) can impact the results of an ERS evaluation. The presence of two-phase flow can increase the required size of a relief device, and the capacity required for effluent handling systems. Tools like FERST powered by CHEMCAD make it easy to evaluate the impact of these parameters and allow the user to quickly perform sensitivity analyses to determine how things like void fraction or set pressure can impact vapor/liquid disengagement, and therefore the results in terms of parameters such as ideal vent diameter or discharge mass flow rate.

Contact us today to help you evaluate the potential presence of two‑phase flow and ensure that your entire relief system is properly sized to handle the worst-case situations. We would also be happy to provide you with a demonstration of FERST powered by CHEMCAD!

Contact: Elizabeth Raines at Eraines@fauske.com  

Resources

1. Fisher, H.G., Forrest, H.S., Grossel, S.S., Huff , J.E., Muller, A.R., Noronha, J.A., Shaw, D.A., and Tilley, B.J., Emergency Relief System Design Using DIERS Technology, The Design Institute for Emergency Relief Systems (DIERS ) – Project Manual, 1992.
2. Grolmes, M.A. and Fisher, H.G., “Vapor-Liquid Onset/Disengagement Modeling for Emergency Relief Discharge Evaluation,” AIChE 1994 Summer Meeting, 1994.
3. Grolmes, M.A. and Epstein, M., “Vapor-Liquid Disengagement in Atmospheric Liquid Storage Vessels Subjected to External Heat Source,” Plant/Operations Progress, Vol. 4, No. 4, 1985.
4. Fauske, H.K., Epstein, M., Grolmes, M.A, and Leung, J.C., “Emergency Relief Vent Sizing for Fire Emergencies Involving Liquid-Filled Atmospheric Storage Vessels,” Plant/Operations Progress, Vol. 5, No. 4, 1986.
5. Epstein, M., Fauske, H.K., and Hauser, G.M., “The Onset of Two-Phase Venting Via Entrainment in Liquid-Filled Storage Vessels Exposed to Fire,” J. Loss Prevention in the Process Industry, Vol. 2, 1989.
6. Fisher, H.G., and Forrest, H.S., “Protection of Storage Tanks from Two-Phase Flow Due to Fire Exposure,” Process Safety Progress, Vol. 14, No. 3, 1995.
7. Fauske, H.K, “Properly Size Vents for Nonreactive and Reactive Chemicals,” Chemical Engineering Progress, February, 2000.
8. FERST powered by CHEMCAD Version 1.0.0.14534, Fauske & Associates, LLC, 2020.
9. Leung, J. C., Simplified Vent Sizing Equations for Emergency Relief Requirements in Reactors and Storage Vessels,” AIChE Journal, Vol. 32, No. 10, p. 1622-1634, 1986.
10. API Standard 521, “Pressure-relieving and Depressuring Systems,” American Petroleum Institute, Washington, D.C., Seventh Edition, June 2020.
11. API Standard 520 Part 1, “Sizing, Selection and Installation of Pressure-relieving Devices,” American Petroleum Institute, Washington, D.C., Tenth Edition, July, 2020.