One site we are considering has a tidal race to one side of the property estimated to be about six knots, or 3 metres per second. This is a page of calculations based on this site.
This site was sold to someone else, so the notes made on tidal generation have been put here, as an alternative to deleting them. They probably could do with editing, but they're not likely to get edited any time soon.
The stiffest requirement is the electric car: we want to run our 32A 3-phase socket for 50% of the time. We also want to run a house requiring 30kWh per day for electricity and 40kWh per day for winter heating. But since the 3-phase socket consumes 21kW at 230v, running it for 12 hours will provide 252kWh, which is way more than we need.
The plan is two three-bladed Savonius turbines beneath a jetty in the tidal race. Although three-blade mills are less efficient, they don't mind being half out of the water. The Sandia report suggests an efficiency of about 0.15 at speeds between 0.4 and 1.0 of the radial velocity speed, for a three-blade turbine with a 0.15 overlap in the centre.
The theoretical power at 2m/s and 1000kg/m3 is 16kW per square metre of turbine in the flow. So if we're generating 0.15 of that energy we need 5 square metres. If we have a height of 2.4m this gives a radius of 1m. In a 2m/s current that gives a rotational speed of 2 radians per second, or 19RPM. At slowest speed of 0.4 of this, to get 20Hz generator output we need 316 magnets and 474 coils. A generator producing 18kW at 58.4v produces 308A, so each coil produces 650mA.
The calculations for a tidal system are very similar to those for a wind system, but with much more energy and rather less storage. So a six-knot tide will be at √1/2 of six knots for 50% of the time (look at a sine wave). Six knots is 3m/s, √1/2 of six knots is 2.1m/s. The same equation can be used for the slower speed and a Savonius generator:-
P = 0.37 ρ v 3 A
but now ρ is 1000, v is 2.1 and A is, say, 0.5m2. That gives a figure of 1.7kW for at the "cut-in" voltage and 3.5kW for the peak. An average of 2.5kW per rotor might be obtained, giving 5kW for two rotors for half of the time, or 60kWh over the 24 hour period. If the rotors are 1m high with a diameter of 250mm, then the outer edge will do 2m/s at a rotational speed of 4 radians per second, which is 76RPM. To get a 20Hz output, the rotor of the generator would need 32 magnets: to get 50Hz output, the rotor would need 80 magnets. To do this amount of work, the cut-in torque is 425Nm, and the peak is 875Nm. Obviously there'd be a longish shaft, so some pretty serious bearings and mounts would be needed to allow it to survive both storms and spring tides.
The really big advantage is that the storage requirements are massively reduced. Assuming the tides come perfectly regularly on a sinewave, then the time between the end of the last flow and the next ebb (or vice versa) is just three hours. Enough batteries to provide our average consumption for three hours is about 3kWh, or 250Ah of 12v batteries; to store half the output from a given tidal flow is 7.5kWh, or 625Ah of 12v batteries; and to provide our peak consumption (about 28kW) only needs 84kWh, or about 7000Ah of 12v batteries. Compared to the batteries for the solar/wind system, this is tiny.
Because there is just so much energy, heating and hot water can be done using the normal arrangements for storage heating: a tank capable of holding three hours of hot water, and electric heaters that store enough heat to keep warm for three hours. These technologies have been around since the 1970s, when Economy 7 electricity first appeared. Appliances that store heat for eighteen hours ought to manage for three.
This page is some notes on Domestic Power from Renewable Sources, and is written and maintained by Simon. At this stage these pages are constantly under revision. Thoughts and comments are welcome.