Two flow rigs, one for water and one for air, were circulated in an intercomparison of laser doppler and hot wire anemometry measurements. Complex systematic differences in results were observed for not only the two types of measuring system but also between instruments of different design based on the same principle. Additionally, the inability to measure sufficiently close to the wall led to unacceptable differences between the integrated flow and that measured by an accompanying flow meter.
Using the measured inlet flow profiles and the geometries of the flow rigs, and intercomparison of flow computations was made using different computer codes for the same two down stream planes measured in the experimental programme. The results showed considerable dispersion particularly in the turbulence parameters. Even when two models used nominally identical computer codes there were significant differences in results beacuse of grid size effects, especially near the walls. Moreover, the integrated flow varied by up to 18%.
A comparison of the theoretical and experimental results suggested that on average the former differed from the latter by about 4%. It is therefore unlikely that current flow computations can be used to limit the need for experimentation.
Funding SchemeCSC - Cost-sharing contracts
G75 0QU Glasgow