(79e) CFD Study of the Effect of an Internal Device within a Cross Junction on Mixing Phenomenon

This study investigates the effect of an internal device within a cross junction with two inlets and two outlets on solute mixing phenomena employing ANSYS-Fluent 17.1. A three-dimensional turbulent flow and the Shear Stress Transport (SST) turbulence model together with the species transport model was simulated to predict the solute concentrations in the outlets. The results showed that intersecting the two inlet flows at 90^o in the cross junction tend to bifurcate them rather than mix them completely. However, with the implementation of an internal cylindrical device inside the cross junction, which is the main contribution of this work, inflows are divided into equivalent flows and nearly complete mixing can be obtained in the two outlet flows.

Introduction

Water distribution networks often employ two pipe junction devices: tee junctions for which no mixing problems have been reported, and cross junctions for which depending the number of inlets and outlets some mixing problems have been reported [1,2]. For the case of cross junctions, further classification can be outlined as follows: 1) one inlet and three outlets, 2) three inlets and one outlet and 3) two inlets and two outlets (with the flow entries possible at 90° or 180°). Network simulation software (e.g., Epanet) assume that mixing in pipe junctions is complete and instantaneous in all cases. However, the assumption of complete mix might pose a high risk in water quality prediction, in particular when the network is supplied by more than one source [1].

Some studies [1,2] have highlighted the importance of mixing on the concentration of diverse mixed solutions employing pipe junctions. These authors reported that depending on the incoming flow directions (90° or 180°), only two results can be attained: i) incomplete mixing, or ii) complete mixing. In the case of i), it was observed that regardless of the magnitude in the incoming flow in either of the inlets, mixing was incomplete for the fact that the cross junction bifurcates and reflects the flows, whereas for the case of ii) solute mixing has shown to be perfect owing to the fact that the both inflows shock in the cross.

Water quality in distribution systems is strongly influenced by mixing after the cross-pipe junctions. Adding chlorine to the drinking water reduces the risk of microorganism growth, eliminating potential risks to the health of the consumers. It has reported that concentration of microorganism (total coliform and Escherichia coli) can vary greatly in the water network system as a result of the poor mixing as well as the pipe material, and larger vulnerability zones can be identified because of incomplete mixing [3].

This work reveals the results of placing of an internal cylindrical device within a cross junction in order to improve mixing when is considered two inlets at 90°.

Methodology

Two cross junction systems with two inlets (North and West) and two outlets (South and East) at 90° with a nominal diameter of 1.5in (43.94mm) and a 500mm length were assembled digitally. The inlets were labeled as Q_N and Q_W, and outlets as Q_S and Q_E (Figure 1a and 1b). The first system consisted of a standard cross junction and the second one was equipped with an internal cylindrical device. The ratio of the inner tube and the outer tube used was 0.55. Different combinations of flow rates and chlorine concentrations were investigated to assess mixing at the cross junctions. Flow ratio was then defined for each scenario (Q_N/Q_W). Different combinations are shown in Table 1. It is important to point out that chlorine concentrations for the west inflow are higher than those for the north inflow in the entirety of the evaluated scenarios.

Figure 1c exhibits a comparison between the numerical results and data reported in the literature. It can be seen that good correlation exists between experiments and computations for which error differences not greater than 8% in all cases were observed.

Figure 1d represents dimensionless outflow concentrations (Q_S and Q_E) for the two analyzed pipe-line systems. All points in this graph were calculated as the ratio between the outlet concentrations computed from simulations, and the expected mixing outflow concentration (0.5) employing a cross junction. The results showed that concentration values lying in the interval 1.0±0.3 (blue-filled circles and squares) suggest that chlorine concentrations for this configuration has achieved nearly complete mixing in the outflows when the pipe-line system is equipped with the internal device, but on the other hand, the values represented by the black-filled circles and squares suggest that incomplete or poor mixing is present in the effluents.

Conclusions

To assume that complete mixing is achieved at standard cross junctions is a biased assumption. It was observed that varying incoming flows at 90^o and chlorine concentrations, solute mixing was shown to be incomplete. It was found that flows tend to bifurcate and reflect off rather than mix them completely. However, with the presence of an internal cylindrical device inside the cross junction, the cylinder section splits evenly the flow entering from the west, thus resulting in an improvement of the mixing at the outlets

References

[1] Mompremier, R. (2015). Difusión del Cloro en redes de tuberías a presión. UNAM, México.

[2] Ho & O’Rear (2009). Evaluation of solute mixing in water distribution pipe junctions. Journal / American Water Works Association, 101(9).

[3] Mompremier, R., Mariles, Ó. A. F., Bravo, J. E. B., Ghebremichael, K., & Martínez, A. E. S. (2018). Study of the effect of pipe materials and mixing phenomenon on trihalomethanes formation and diffusion in a laboratory-scale water distribution network. Water Science and Technology: Water Supply, 18(1), 183–192.

[4] Abraham, J. P., Sparrow, E. M., & Tong, J. C. K. (2008). Breakdown of laminar pipe flow into transitional intermittency and subsequent attainment of fully developed intermittent or turbulent flow. Numerical Heat Transfer, Part B: Fundamentals, 54(2), 103–115.

[5]Menter f. R., Langtry r. B., Likki s. R, Suzen y. B., h. P. G. (2004). A correlation-based transition model using local variables part i – model formulation. ASME-GT2004-53452, 1–11.

[6] Mompremier, R., Pelletier, G., Mariles, Ó. A. F., & Ghebremichael, K. (2015). Impact of incomplete mixing in the prediction of chlorine residuals in municipal water distribution systems. Journal of Water Supply: Research and Technology - AQUA, 64(8), 904–914.