Some limitations to measurements of turbulence micro-structure with hot and cold wires

Some of the problems associated with measurements of the small-scale characteristics of dynamic and thermal fully developed turbulence are considered in the context of a study of the tendency-towards-isotropy assumption, Taylor's hypothesis, and internal intermittency, taking advantage of the possibilities offered by the I.M.S.T. Air-Sea Interaction Simulation Tunnel. The paper reviews and discusses the choices of probes and operating parameters for hot-wire anemometers and cold-wire thermometers. A detailed study of the most important characteristic, the spectral signal-to-noise ratio, is presented. Problems specific to measurements with several wires are studied, especially the true resolution of one-point multi-wire probes and the errors due to calibration uncertainties for two-point measurements. Serious doubts are raised on the validity of gradient-probe data.