Validation Case: Propane Jet Flow Simulation

The aim of this validation is to compare the simulation results from SimScale using the Multicomponent feature in its proprietary solver, Multi-purpose, with the experimental results in the study done by Schefer titled “Data Base for a Turbulent, Nonpremixed, Nonreacting Propane-Jet Flow“.

The objective is to understand the flow patterns and concentration distribution of the propane jet in the presence of co-flowing air. The geometry considered is a simple pipe to benchmark the evolution of the propane jet. This type of analysis is useful in the design process of combustion chambers to analyze the gas flame or even the gas injection.

View Project

Geometry

The model consists of a small extended pipe for fuel injection and flow development, and a pipe with a larger diameter where a mixture of gases has to be modeled.

Pipe model with jet flow entrance — Figure 1: Pipe model with the extended region for the jet inflow

The dimensions of the pipe can be seen in the table below:

Dimension	Value \([m]\)
Jet pipe length	0.29
Pipe length	2
Jet pipe diameter (D)	0.0052
Pipe diameter (D)	0.30

Table 1: Jet pipe and pipe dimensions

Analysis Type and Mesh

Analysis Type: Steady-state, Multi-purpose with the Multicomponent model and K-Epsilon turbulence model

Mesh and Element Types: The mesh is a Cartesian hexahedral mesh, created using SimScale’s Multi-purpose solver.

Mesh Sensitivity

The Multi-purpose meshing algorithm, utilizing hexahedral cells, was employed to generate the mesh for this simulation. A refinement level of 5 was selected, and two region refinements were implemented: one to enhance the mesh quality near the jet flow region (Target cell size: 0.0008 \(m\)) and another in the main pipe ((Target cell size: 0.01 \(m\)).

Mesh Type	Number of cells	Element Type
Automatic Level 5	5095357	3D Hexahedral

Table 2: Mesh data for the propane jet validation case

multi-purpose Mesh within the flow domain — Figure 2: Mesh within the flow domain with fineness level 5 and region refinements

mesh refinement around the jet region from fine to coarse — Figure 3: Multi-purpose meshing performed on the pipe with refinement around the jet region

Simulation Setup

The pipe consists of a mixture of Propane, Oxygen, and Nitrogen gases. Each of these gases is treated as a separate ‘component’, with the carrier fluid being air. The physical properties of the different gases as well as other simulation parameters are listed below.

Fluid:

Propane (C₃H₈)
- Molar mass: 44.1 \(kg/kmol\)
- Dynamic viscosity \((\mu)\): 0.0000172 \(kg/s.m\)

Oxygen (O₂):
- Molar mass: 32 \(kg/kmol\)
- Dynamic viscosity \((\mu)\): 0.0000172 \(kg/s.m\)

Nitrogen (N₂):
- Molar mass: 28 \(kg/kmol\)
- Dynamic viscosity \((\mu)\): 0.0000172 \(kg/s.m\)

Boundary Conditions:

Boundary condition overview simscale pipe and jet inflow and outflow — Figure 4: Boundary condition overview where the flow goes from left to right

The model’s boundary conditions are different for each region of the domain. The smaller tube represents the region where the propane jet enters, and the larger tube represents the region where air enters, including oxygen and nitrogen as components. As the simulation is compressible, the temperature values were kept constant at 295 \(K\) for the boundary conditions. An ambient pressure was set for the outlet region, and a backflow fraction corresponding to the composition of the air was set, which means that in the case of a backflow, only air is allowed to flow back into the domain.

Boundary Condition	Value	Mass Fraction (O₂)	Mass Fraction (N₂)	Mass Fraction (C₃H₈)	Temperature \([K]\)
Velocity inlet (Jet Flow)	54 \(m/s\)	0	0	1	295
Velocity inlet (Air)	9.2 \(m/s\)	0.23	0.77	0	295
Pressure outlet	101325 \(Pa\)	0.23	0.77	0	295
No-slip pipe walls	–	–	–	–	295

Table 3: Boundary conditions for the pipe domain

Result Comparison

A comparison of the propane jet flow obtained from SimScale against the reference results¹ is given below. On the lower axis, the plot shows the relationship between the coordinates of the pipe from the entry of the jet into the larger tube \((x)\), normalized by the jet tube diameter \((D)\). SimScale results for the mass fraction of propane as a function of the pipe coordinate show an excellent correlation with the experimental data.

Propane mass fraction against pipe center length simscale — Figure 5: Centerline profile of C₃H₈ mass fraction

Axial and radial velocities were also analyzed to assess the profiles and compare the results obtained from SimScale. In this regard, figures 6, 7, and 8 demonstrate that the software effectively represents the physical phenomenon, particularly when examining velocity patterns, although the maximum axial velocity for this case is slightly lower, with an error margin of approximately 4%.

Figure 6: Centerline profile of mean axial velocity

mean axial velocity simscale vs schefer propane jet diffusion comparison — Figure 7: Radial profile of mean axial velocity at x/D=4

mean radial velocity simscale vs schefer propane jet diffusion comparison — Figure 5. Radial profile of mean radial velocity at x/D=4.

Figure 9 illustrates the jet contour observed in SimScale’s online integrated post-processor, where the analyzed variable represents the mass fraction of propane entering the tube. It shows the mass fraction of propane and its dissipation with the mixture of nitrogen and oxygen gas.

C3H8 mass fraction contour observed while mixing in air — Figure 9: C3H8 mass fraction contour as observed in the integrated post-processor

The jet’s flow development aligns with the expected velocity profile, with Figure 10 showing a pronounced velocity gradient at the jet’s center.

C3H8 velocity magnitude contour observed while mixing in air — Figure 10: Velocity magnitude contour.

References

Schefer, R.W., 2001. “Data base for a turbulent, nonpremixed, nonreacting propane-jet flow”. Combustion Research Facility, Sandia National Laboratories, Livermore, CA.
Dibble, R. W., Hartmann, V., Schefer, R. W. and Kollmann, W., “Conditional Sampling of Velocity and Scalars in Turbulent Flames using Simultaneous LDV-Raman Scattering,” Exper. Fluids 5, pp. 103-113 (1987).
Dibble, R. W., Kollmann, W., and Schefer, R. W., “Conserved Scalar Fluxes Measured in a Turbulent Nonpremixed Flame by Combined Laser Doppler Velocimetry and Laser Raman Scattering,” Combust. Flame 55, pp. 307-321 (1984).

Note

If you still encounter problems validating you simulation, then please post the issue on our forum or contact us.

Last updated: December 4th, 2024

Did this article solve your issue?

How can we do better?

We appreciate and value your feedback.