A 70-day data set from bottom-mounted ADCPs on the two sides of the Faroe-Bank Channel was analysed using the recorded flow variance and echo intensity in the deeper reaches of the passage as proxies for turbulence. A consistent picture emerged, not least since the data losses (which were ascribed to turbulence-induced activation of the fish-elimination option in the ADCP software) could be shown to co-vary with the internal M-2 tide affecting the vertical shear, which in turn proved to be correlated with the flow variance.