**6. References**

78 Advanced Fluid Dynamics

1. Outer borders of the Jet-A spray trajectories created as a result of fuel jet disintegration in the cross flow of cold air at elevated pressure of 5 atm were measured by application the high speed imaging technique that allowed obtaining series of instantaneous images of the fluctuating spray. Locations of the liquid column breakup points (CBP) were determined using the light guiding technique that make mass of liquid illuminated

2. Crossing air flow had core turbulence ~4% and thickness of the boundary layer near the

3. Both injectors used in the study had the same diameter of the orifice *d=0.47mm* and a different shape of the internal path (i.e., sharp and round edge orifice) were manufactured using the same equipment and technology. They were installed with

4. Application of light guiding technique significantly improved accuracy of the jet in cross flow column breakup point (CBP) determination especially at elevated Weber number (We>200) when traditional shadowgraph methods are not effective because of

5. CBP was found to be strongly dependent upon velocity of the jet and internal turbulence of liquid inside the orifice. Jet injected from the sharp edge orifice disintegrates earlier compared to the round edge orifice. Dependence of the CBP location upon temperature of injector is much stronger in the sharp edge orifice

6. CBP locations were well correlated while converted to the non-dimensional form of characteristic time against the liquid Reynolds number. In fact, CBP location determined in this study were found to be 1-4 diameters of the jet downstream from the injection

8. Spray penetration into the cross flow was found to be proportional to square root of momentum flux ratio of the fuel jet to crossing air in the investigated range between *q=5* and *q=100* due to self explained dependence of droplet penetration upon the jet

9. Spray created by the sharp edge injector penetrated 12% further into the cross flow than from the round edge orifice. This observation was attributed to a larger droplet size

11. Simple correlations for the spray trajectories were obtained using only two empirical coefficients. One of them corresponded to the shape of the injector internal path and the other one only adjusted shape of the logarithmic function that determined average or maximum penetration of the spray and was independent of the injector

created by sharp injector and, possibly by the higher velocities of some droplets. 10. Good agreement between the spray trajectories obtained using high speed imaging technique used in the current study and borders of the spray measured by the processing of the PDPA data. It was found that that the maximum spray penetration determined as *Xmax=Xmean+ 2.8RMS* is equal to the border determined at the level of

orifice which is much closer than it was reported in the previous studies (z/d~8). 7. Spray trajectories were found to be independent upon Weber number in the investigated range between *We=400* and *We=1600* due to only shear breakup mode of

from inside fluoresce till the moment jet losses its continuity.

presence of droplets in high density around the liquid column.

rectangular channel walls ~3mm.

compared to the round edge orifice.

liquid jet disintegration.

design.

velocity at the point of injection.

10% threshold of the PDPA data rate maximum.

orifices openings flush with the channel wall.

**4. Conclusions** 


**0**

**5**

*Spain*

**Influence of Horizontal Temperature**

**Gradients on Convective Instabilities**

Since Bénard's experiments on convection and Rayleigh's theoretical work in the beginning of the twentieth century (1)-(2), many experimental, theoretical and numerical works related to Rayleigh-Bénard convection have been done (3)-(10) and different problems have been posed depending on what is to be modelled. Classically, heat is applied uniformly from below and the conductive solution becomes unstable for a critical vertical gradient beyond a certain

A setup for natural convection more general than that of uniform heating consists of including a non-zero horizontal temperature gradient which may be either constant or not (11)-(29). In those problems a clear difference is marked by the fact that the fluid is simply contained (11)-(19), where stationary and oscillatory instabilities appear depending on the multiple parameters present in the problem: properties of the fluid, surface tension effects, heat exchange with the atmosphere, aspect ratio, dependence of viscosity with temperature, etc., and the case where the fluid can flow throughout the boundaries (29), where vortical solutions can appear reinforcing the relevance of convective mechanisms for the generation of vertical vortices very similar to those found for some atmospheric phenomena as dust devils or

The case where the fluid is simply contained displays stationary and oscillatory instabilities. This problem has been treated from different points of view: experimental (11)-(18) and theoretical, both with semiexact (20)-(21) and numerical solutions (40)-(28). This case contains applications to mantle convection when the viscosity is large (45; 52) or it depends on

There are not experiments yet for the case where a flow throughout the boundaries is allowed, only observations of atmospheric phenomena (30; 33; 34; 36; 37), and theoretic numerical

In this work we will review this physical problem, focusing on the latest problems addressed by the authors on this topic, where a non-uniform heating is considered in different geometrical configurations, and we will show the relevant results obtained, some of them

in the context of interesting atmospheric and geophysical phenomena (30; 36; 37).

**1. Introduction**

threshold.

hurricanes (29)-(31).

temperature (19).

results (29; 31).

**with Geophysical Interest**

H. Herrero, M. C. Navarro and F. Pla

*Universidad de Castilla- La Mancha*

Wu, P.-K., Kirkendall, K. A., Fuller, R. P., & Nejad, A. S. (1997), Breakup Processes of Liquid Jets in Subsonic Cross-flows. *Journal of Propulsion and Power*, Vol. 13, No. 1, pp. 64 -73.
