(diff) ← Older revision | Latest revision (diff) | Newer revision → (diff)
Jump to navigation Jump to search

Front Page

Description

Test Data

CFD Simulations

Evaluation

Best Practice Advice

References

Particle-laden swirling flow

Application Area 3: Chemical, Process, Thermal and Nuclear Safety

Application Challenge AC3-12

Abstract

Swirling particle laden flows are found in various areas of process technology, e.g. swirl burners, pneumatic conveying in swirling pipe flow and gas cyclones. Already the single-phase flow is very complex due to the involved different scales of vortices, the strong anisotropy of turbulence and the mostly observed unsteadiness. In particle-laden swirling flows additionally the particle response to flow structures and turbulence becomes of importance which strongly depends on particle size. In order to provide a detailed set of data for validating numerical simulations, thorough measurements by Phase-Doppler-Anemometry were conducted in a laboratory-scale swirling flow system. The test section is a pipe-expansion flow with a co-annular inlet (diameter 64 mm) and a measurement section of 1500 mm length and a pipe diameter of 194 mm. Linear profiles of all velocity components of gas and particle phase were measured downstream of the inlet at 8 cross-sections between 3 mm and 315 mm. For the particle-phase also local size distributions were measured as well as the size-velocity correlations. Therefore, profiles of the number mean particle diameter (mean diameter based on particle counts) are available whereby the size segregation could be analysed. From the size-velocity correlations also the mean velocities for three size classes were determined, namely 30, 45 and 60 μm. Additionally, profiles of the stream-wise particle mass flux were determined. Two swirl cases were considered, which have about the same swirl number (i.e. about 0.5), but the ratio of secondary mass flow rate to primary mass flow rate was about twice in the second case. The particle mass flow rate was relatively low in both cases, so that the influence of the particles on the gas flow (i.e. two-way coupling) was relatively small.

Numerical computations performed for the two cases with the in-house finite-volume code FASTEST in connection with the k-ε turbulence model showed reasonably good agreement with the measurements. The particle phase was simulated by Lagrangian tracking, also yielding a quite good agreement with measured velocity profiles, the particle mass flux and the number mean particle diameter (Sommerfeld and Qiu 1993).

The swirling Case 1 was also used as a test case for validating "in- house" codes at the 5th Workshop on Two Phase Flow Predictions (Sommerfeld and Wennerberg 1991). Several groups have participated in these calculations and the results may be found in the Workshop Proceedings, including a description of the numerical methods applied. A brief summary of the results is also presented at the end of this document. As was to be expected, the scatter of the results from the different groups is not negligible.



Contributed by: Martin Sommerfeld — Martin-Luther-Universität Halle-Wittenberg

Front Page

Description

Test Data

CFD Simulations

Evaluation

Best Practice Advice

References


© copyright ERCOFTAC 2013