Flow Regime Determination in Emergency Relief System Design - Blowdown Testing

By Benjamin Doup Ph.D., Senior Nuclear and Chemical Engineer, Fauske & Associates, LLC

In the current issue, we provide practical guidance on how to determine the expected flow regime under emergency relief venting conditions.

The flow regime during venting can be determined by running a blowdown test in the Vent Sizing Package 2 (VSP2^TM). The VSP2 blowdown test procedures, test interpretation, potential missteps, and benchmark test results that are applicable for vapor systems will be discussed.

VSP2 Blowdown Test Procedures

The suggested approach is to simulate the upset scenario in the VSP2 and then depressurize the test cell using a vent located on the lid of the test cell. Figure 1 shows a schematic of the VSP2 setup in blowdown configuration. This configuration is applicable for many materials and tests. However, depending on the specific design of the blowdown test the following modifications may be investigated:

• If your material is not hazardous you may be able to route the vent line to an open vessel filled with room temperature water. The water will act as a quenching fluid for the hot effluent from the test cell.

• Multiple or larger vents on the containment vessel may be necessary in order to depressurize the containment vessel at a similar rate to the depressurization of the test cell.

• If the material being tested has a high vapor pressure at room temperature, valves may need to be installed on the fill line and vent line inside the containment vessel to prevent mass loss after the blowdown is complete.

Figure 1 Schematic of the VSP2 setup in blowdown configuration

The general blowdown test procedures are:

1. Use set temperature to determine properties that are required for all-vapor critical flow and bubble rise velocities
2. Determine bubble rise velocity (u_∞) for the material of interest and for both bubbly and churn flows using Equation 1
3. Determine critical all-vapor mass flux at the set conditions using Equation 2
4. Determine the superficial vapor velocity through the test cell (j_g) at disengagement using Equation 3
5. Target a j_g/u_∞ between 0.8-4 by adjusting the vent diameter and/or all-vapor discharge coefficient (i.e., length of the vent line)
6. Start the VSP2 blowdown test using a specific upset scenario procedures
7. Begin blowdown of the test cell and containment vessel when the set temperature is reached by opening the valves on the test cell and containment vessel.
   1. a. It may be necessary to begin the containment blowdown before the test cell blowdown to avoid crushing the test cell.
8. Close the valves when test cell pressure reaches ambient pressure, which often occurs within 8 seconds.

The bubble rise velocities are 

(1)

 where          C_∞ = bubble rise velocity coefficient, 1.18 for bubbly flow and 1.53 for churn flows,

 g = acceleration due to gravity, m/s²

 ρ_f = liquid density, kg/m³

ρ_g = vapor density, kg/m³

  σ = surface tension, N/m

The critical all-vapor mass flux at the set conditions can be estimated using

G_g =(2)

where           G_g = critical all-vapor mass flux kg/m²/s

  k = isentropic coefficient,

P_s = set pressure, Pa

The vapor superficial velocity through the VSP2 test cell can be estimated using

j_g  =        (3)

where     A_vent = effective flow area of the vent on the VSP2 test cell, in²

     A_x = cross sectional area of the VSP2 test cell, in²

The VSP2 Blowdown Test Interpretation

The interpretation of the VSP2 blowdown test is as important (if not more important) than the design and execution of the blowdown test and is necessary for determining the flow regime of your material. The main indicator is mass remaining in the test cell after the blowdown. The final mass is used to obtain an average void fraction. The final void fraction is then compared with the disengagement void fraction for the bubbly and churn flow regimes to obtain similarities. The disengagement void fractions for the bubbly and churn flow regimes are determined using Equation 4 [1].

j_g = 

where  C₀ = distribution coefficient, 1.01 or 1.2 for bubbly flow and 1.5 for churn flow

         = vessel average void fraction, 

Additional items that will aid in the interpretation of the blowdown test are estimating the mass loss from the test cell assuming all vapor flow and simulating the depressurization of the test cell. The simulation of the depressurization of the test cell allows the time dependent temperature and pressure measurements to be used in the comparison of churn and bubbly flow regime comparison along with the mass remaining in the test cell.

Potential Missteps

The results of the blowdown testing may not always be extremely clear. When disengagement occurs while the flow is sub-critical, the vapor superficial velocity can be much lower than the vapor superficial velocity used in the test design.
This will affect the expected vessel average void fraction for complete disengagement. In this case simulating the VSP2 blowdown test may be necessary in order to provide the temperature and pressure comparison and aid in helping predict the superficial velocity when disengagement occurs. Crushing the test cell indicates that the size or number of valves used to depressurize the containment may be too small. In this case, the test may need to be repeated using a larger number or size of valves to depressurize the containment vessel. Alternatively a heavier walled test cell could also be used.

Benchmark Test

Benchmark tests have been performed using tap water and soapy water. Table 1 shows the initial conditions and the results of the benchmark tests. Figure 2 plots Equation 4 and the final void fractions of the benchmark test. Dynamic simulations were performed to aid in the interpretation of the blowdown tests. These tests show that tap water is predicted to behave as churn flow with a distribution coefficient of 1.5 and soapy water is expected to behave as a foamy or bubbly flow with a distribution coefficient equal to 1.01, which is consistent with the large scale test results [1].

One of the reasons that the average void fractions for the tap water tests, derived from the final mass of the VSP2 blowdown tests, are above the churn-turbulent predicted average void fraction is that disengagement occurs before the depressurization is over. This leads to all-vapor flow during a portion of the depressurization and additional mass loss from the test cell. Figure 3 shows the comparison between the depressurization transient and dynamic simulations for the tap water and soapy water blowdown tests with 1/8” diameter vent lines. The good agreement between the depressurization data and the dynamic simulations provide further evidence that the flow regime classification based on the blowdown tests is consistent with large scale data.

Emergency Relief System Design Using DIERS Technology

For more information regarding relief system design, process hazard analysis, and other process safety testing news, subscribe below!

References
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.