Water Mills in Central Crete is also a CD3WD document, and this one is based upon the his travel and study of agriculture and watermills in Greece.
This has a lot of good technical information, and I have provided an excerpt below:
Water Mills in Central Crete
The energy, or capacity for doing work possessed by a stream of water, can have three manifestations, These are its elevation, its pressure and its velocity. Energy loss, that is dissipation into the unavailable form of heat, is associated with high velocities. The Cretan water mill shows an appreciation of these factors.
Flow along the aqueduct is at a low velocity and constant high elevation. Flow down the tower is also at a low velocity. Loss of elevation in this process is compensated by an increase in pressure and there is little dissipation of energy. At the nozzle there is a transition from high pressure to high velocity manifested by the formation of the jet. The jet impinges on the wheel producing forces which cause its rotation. In the absence of test facilities it is interesting to attempt to make an estimate of the hydraulic performance of a Cretan water mill. This can be done if the combination of water tower and orifice is treated as a flow meter. Assumptions are needed, for the jets observed were less perfect than those one would found in a laboratory. Far from showing a vena-contracta there was a tendency for the jet to diverge and to lose solidity. Under such conditions hydraulic coefficients must be assumed and can only be based on the observer’s engineering experience. The figures which follow are consistent with velocity and discharge coefficients of 0.91 and 0.87 respectively and with a wheel efficiency of 0.43.
Since the author was on a journey of reconnaissance and not resident near to a working mill, he was not able to get a full set of corresponding dimensions and velocities on any single mill. If the mill was running, it was not practical to enter the pit to measure the orifice and the wheel. Alternatively, if the mill was at rest, the pit could be entered, but the speed would not be known. Nevertheless, since little variation in speed was observed in the mills found to be in motion and only minor differences of wheel and orifice were measured in mills at rest