Scrupulous measurements and detailed data analysis of the torque in aswirling turbulent flow driven by counter-rotating bladed disks reveals anapparent breaking of the law of similarity. Potentially, such breakdown couldarise from several possible factors, including dependence on dimensionlessnumbers other that $Re$ or velocity coupling to other fields such astemperature. However, careful redesign and calibration of the experiment showedthat this unexpected result was due to background errorscaused by minutemisalignments which lead to a noisy and irreproducible torque signal at lowrotation speeds and prevented correct background subtraction normally ascribedto frictional losses. An important lesson to be learnt is that multiple minutemisalignments can nonlinearly couple to the torque signal and provide a dcoffset that cannot be removed by averaging. That offset can cause the observeddivergence of the friction coefficient C_f from its constant value observed inthe turbulent regime. To minimize the friction and misalignments, wesignificantly modified the experimental setup and carried out the experimentwith one bladed disk where the disk, torque meter and motor shaft axes can bealigned with significantly smaller error, close to the torque meter resolution.As a result we made precise measurements with high resolution and sensitivityof the small torques produced for low rotation speeds for severalwater-glycerin solutions of different viscosities and confirmed the similaritylaw in a wide range of Re in particular in low viscosity fluids.
展开▼
