Experimental Investigation on the Effects of Surface Roughness on Microscale Liquid Flow.

Microfluidics has become of interest recently with shrinking device sizes. Roughness structures left from machining processes on the inside of tubes and channels that were once not a concern may now create relative roughness that exceeds 5%. Confusion still exists in the literature as to the extent of the effects of roughness on laminar flow. This work aims to experimentally examine the effects of different roughness structures on internal flows in high aspect ratio rectangular microchannels. A total of four test sections were fabricated to test samples with different patterned rough surfaces, and to also vary the two opposite surfaces forming the long faces of the channel. These test sections allowed the same roughness samples to be tested at varying relative roughnesses and allowed a systematic study on their effects on pressure drop. The first test section looked at sawtooth effects on laminar flow. The second looked at uniform roughness on laminar flow. The third looked at sawtooth roughness in turbulent flow, and the fourth looked at varying pitch sawtooth roughness in laminar flow. Rough surfaces were formed in one of two ways. The first involved making structured repeating sawtooth ridges with a ball end mill on a CNC machine. The second was using sandpaper in a crosshatch pattern to make a more unpatterned roughened surface. In this study, the Reynolds number was varied from 30 to 15,000 with degassed, deionized water as the working fluid. The experimental uncertainty in the experimental data is at worst 7.58% for friction factor and 2.67% for Reynolds number. Roughness structures varied from a lapped smooth surface with 0.2 mum roughness height to sawtooth ridges of height 117 mum. Hydraulic diameters from 198 m to 2,349 mum were tested. The highest relative roughness tested was 24.8%. As a result of the first and second experiments, it was shown that using constricted parameters, sawtooth and uniform roughness performance could be predicted in the laminar regime. In the third experiment, it was shown that certain sawtooth roughness samples cause the results to converge to a single line for friction factor. In the fourth experiment, the pitch of sawtooth elements was shown to be a key parameter in showing when each parameter is applicable. It was found that roughness has an effect even at relative roughness values less than 5%. Lapped smooth samples showed no departure from macroscale theory at all channel diameters tested, which implies that no departure from continuum mechanics occurred at the length scales tested. This fit with what was expected. Early transitions to turbulence were seen however, showing decreasing transition Reynolds number with increasing relative roughness. The lowest turbulent transition occurred at a Reynolds number of 210, with a relative roughness of 24.8%. Most all of the roughness structures studied were found to have experimental results that were well predicted with the use of constricted parameters. However, samples with roughness elements placed at higher pitches were seen to have experimental results approaching theory calculated with root parameters of the channel.