UFR 4-02 Evaluation

From QNET
Jump to navigation Jump to search

Front Page

Description

Test Case Studies

Evaluation

Best Practice Advice

References




Confined coaxial swirling jets 

Underlying Flow Regime 4-02               © copyright ERCOFTAC 2004


Evaluation

Comparison of CFD calculations with Experiments

Comparison, done by Lei et al.[1], of the numerical simulations of the new ASM with experimental data performed by Vu & Gouldin [8] are shown on Figure 4 derived from [1]. General good agreement of axial and tangential velocities as well as pressure profiles is achieved, however in the downstream region x/Ri = 5.77 discrepancies between axial velocity profile and the measured data exist. The axial velocity profiles in the region r/Ri > 1.5 are quite uniform, this tendency is also observed on the figure 4d for turbulent fluctuating velocities. In the region of r/Ri < 1 axial velocity increases with the radial distance reaching maximum at r/Ri = 1. Local minimum is observed at the location around r/Ri = 1 ÷ 1.5. For the tangential velocity (Fig. 4b), local minimum value appears, which is caused by jets interaction and mixing and that minimum is flattened out with axial distance.

UFR4-02 TestCase Fig4.gif

U4-02d32 files image045.gif

U4-02d32 files image040.gif U4-02d32 files image038.gif

Calculated profiles of turbulent fluctuating velocities depicted on Figure 4d are in good agreement with experiment in region r/Ri > 1.5. In the central region around the axis (r/Ri < 1.5) U4-02d32 files image046.gif are seriously underestimated but in the downstream region differences are smaller. Figure 5 shows comparison of the calculated axial velocity by the new ASM and the k-ε model with the data measured by Vu & Gouldin. Both calculated data for axial velocities are close to each other and generally in good agreement with experiment, however, for the downstream region x/Ri > 3.67, especially around the jets centreline (r/Ri < 1) both models overpredict experimental data, but the new ASM gives better predictions than KEM.

Comparison of the calculated mean axial velocity profiles along the centreline with the same experiment was also presented by Sharif & Wong, and the results are depicted on Figure 6. It shows that ASM [15] turbulence model gives similar prediction results as RSTM, while NKEM calculation results for axial velocity are seriously overestimated.

U4-02d32 files image048.gif


The comparison of measured by Vu & Gouldin and calculated by Sharif & Wong axial turbulence intensity U4-02d32 files image049.gif is shown on Figure 7. It is seen that axial turbulent intensities are very low and almost uniform for the outer jet. The very high increase over the inner jet is observed with maximum value peak at jet interface location. All considered models estimate well outer jet region while failing to predict this high level of axial turbulence intensity, although the correct tendency can be seen. It has to be mentioned that reported experimental uncertainty for turbulence intensity is from ± 20 ÷ 50%.

The comparison of the predicted radial profiles for the axial and azimuthal velocities and static pressure at different axial locations with experiment are shown on Figure 8. From this figure it is seen that all models are close to each other in prediction of velocities as well as pressure and well predict the free vortex motion of the outer jet, however, in the region of jet interface all models underpredict the azimuthal velocities especially at downstream locations and the NKEM slightly overpredicts the pressure for most regions.

© copyright ERCOFTAC 2004



Contributors: Stefan Hohmann - MTU Aero Engines


Front Page

Description

Test Case Studies

Evaluation

Best Practice Advice

References