In order to measure the diffusion coefficient of a heated plane airstream, a single laser beam is passed through the jet, perpendicularly to the flow direction. The thermic turbulence in the airstream causes random fluctuations of the refractive index. Consequently, the beam direction undergoes, in the flow, random perturbations. After having traversed the jet, the beam produces a luminous trace on a photoelectric cell placed outside the jet. An experimental setup for measuring the probabilities of the beam impact positions on the cell is described. From the Markovian process model, applied along the whole random path of the beam, it has been possible to compute these probabilities by solving the Einstein–Fokker–Kolmogorov equation. The diffusion coefficient can be determined by adjusting the numerical solution to agree with the experimental results. In addition, the calculation procedure gives the order of magnitude of an integral scale, characterizing the dimension of the turbulent structures in which the propagation of light can be considered rectilinear. A good agreement between the results and the published data obtained by means of the cold-wire anemometer technique proves the validity of the method.

The difficulties of investigating thermal turbulence using a classical hot wire anemometer in flows with very high mean velocity and temperature are well known. We are thus concerned in this paper with the possibility of studying such a turbulence by means of a laser beam propagating through a heated free plane jet of air. The beam undergoes in the flow random perturbations which tend to deflect it around the path that it would have taken if it had propagated in a vacuum. A formula is derived which relates the random deflection to the random temperature fluctuations. The experimental set -up is described. The values of the mean temperature and velocity have been chosen so that comparisons between the results obtained by this optical method and thoses obtained by the classical hot wire are made possible. A close agreement between both sets of results proves the validity of the model.

No abstract

No abstract