Contents

Introduction

Critical Rayleigh

Velocity study

Unsteady case

Conclusion

Unsteady case





At the beginning, this study aims at refind the critical Rayleigh because an unsteady simulation better reflect the experiment condition. But near Rac, our computation didnt't converge. So We have chosen to present the results we have obtain: they permit us to understand what happen during the establishment of the rolls. The computation have been lead on a 20x40 grid at Ra=8000. The simulation method as consist on:

    • Initialise the velocity to 0 Initialise the temperature to: T=T2+(T1-T2)y/d

    • Start the iterations with a time step of 5s.

The results are plotted below: the pictures represent the temperature and the velocity each 5 seconds (the simulation time for the first picture is 5 s)


















We could notice that the rolls appear at about 40s of the simulation. At this time, 3 rolls shape. But the former study at Ra=8000 have shown only 2 rolls. So something happened: this difference is due to the stability domains. In the unsteady case, we have fall in a domain where 3 rolls are stable. But I donít want to develop this subject because this paragraph just aims at show what happen during the formation of the rolls.