Hydro · Run-of-river
Take a portion of a stream, run it down a pipe, spin a turbine with it, and return it to the channel a hundred feet lower. Nothing is flooded, nothing is stored, and the stream below the powerhouse looks much as it did before. This is what most working micro-hydro actually is.
The definition
Run-of-river means the system takes what the stream is currently carrying and has no way to save any of it for later. Natural Resources Canada describes the arrangement plainly: most micro-hydro systems operate run of river, no large dams or storage reservoirs are built, no land is flooded, and the majority use only a fraction of available stream flow, which keeps environmental impact smallMicro-Hydropower Systems">[1].
The trade this makes is worth stating clearly. Storage would let you hold water back overnight and release it during the evening peak, matching supply to demand. Without it, the turbine produces the same amount at four in the morning as it does at dinner time, whether anyone needs it or not.
In exchange you avoid building a dam, and everything that follows from that: the engineering, the liability, the permitting, the sediment, and the flooded acre. For most properties this is an easy trade, which is why run-of-river is the configuration to assume unless something specific rules it out.
Between the intake and the tailrace lies a stretch of channel carrying less water than it did naturally. That stretch is the system's real environmental footprint, and it is why the fraction you divert and the distance you carry it both matter. A short diversion taking half the flow disturbs far less than a long one taking nearly all of it.
The component chain
Natural Resources Canada lists the principal components of a micro-hydro system, and the sequence below follows theirs[1]. Not every site needs all eight. A short, steep installation may run the penstock directly from the intake with no canal at all.
Where water leaves the stream. A low weir raises the level enough to feed the diversion reliably without impounding anything meaningful behind it.
Carries water from the intake toward the forebay, usually along the contour at a gentle grade. This section is about conveyance, not pressure.
A small tank where sediment settles and a screen catches leaves, needles and gravel before they reach the turbine. It also keeps air out of the penstock when flow changes suddenly.
The pressure pipe running down the slope to the turbine. Everything the site has to offer arrives through here, minus whatever friction takes on the way.
A shelter holding the turbine and generator. On a household system this can be a weatherproof box the size of a garden shed, sited above any flood the stream can produce.
The device that keeps output stable by sending unused power somewhere harmless. It is not optional equipment, and it gets its own guide in this section.
The channel returning spent water to the stream. Its outlet needs protection from erosion, because water leaving a turbine arrives with energy still in it.
Wire from powerhouse to house. Often a long run, and long runs at low voltage lose power to resistance the same way a small penstock loses head to friction.
Notice that the same problem appears twice. Water losing head in a narrow pipe and electricity losing voltage in a thin wire are the same kind of loss, and both are solved by spending more on the conductor. Where the powerhouse sits on the slope is a decision that trades one against the other, because moving it down the hill buys head and adds wire.
Why start here
A diversion that returns water promptly and impounds nothing is a smaller regulatory proposition than a dam. It still requires water rights and approvals, and it is still illegal to take surface water without them, but the review is a different order of magnitude.
A blocked intake stops the turbine. A burst penstock empties a pipe onto a hillside. Neither releases a stored volume of water downhill at once, which is the failure a dam makes possible and the reason dams are engineered the way they are.
Money goes into penstock, turbine and wire rather than into concrete and earthworks. Pipe can be bought in stages, extended later, and upgraded in diameter without rebuilding the whole project.
Natural Resources Canada characterizes micro-hydro as proven technology with excellent reliability, low maintenance cost, and a service life of 20 to 30 years[1]. None of that describes an experiment.
Where the money goes
People arrive at micro-hydro thinking about turbines. The turbine is a machine you order. The penstock is a decision that has to be made about your specific hillside, it usually costs more than the turbine, and it determines how much of the head you measured actually reaches the runner.
Friction in a pipe rises steeply as diameter falls. Undersize it and a meaningful share of the site's head is spent pushing water through pipe rather than turning a machine. That loss is permanent, it recurs every second the system runs, and the only remedy is to dig the pipe up and replace it with a larger one.
Sizing it properly, choosing material and pressure rating, burying it or laying it on the surface, and protecting it from freezing are covered in the intake and penstock guide. What matters here is the budgeting instinct: when the estimate feels tight, buy the bigger pipe and the smaller turbine, never the reverse.
High-head, low-flow sites cost less to develop than low-head, high-flow ones, because penstock, turbine, intake and spillway are all smaller for the same output[1].
A longer penstock reaches lower ground and gains head, then gives some back to friction along the way. There is an optimum, and it is rarely the bottom of the property.
High initial development cost, very low annual operating and maintenance cost[1]. Build it properly once and it asks little for decades.
Living with it
Our example creek delivers 810 watts, hour after hour, indefinitely. A household does not consume 810 watts steadily. It consumes 200 watts for most of the night and 3,000 watts for twenty minutes when someone runs a well pump and a kettle at the same time.
A small battery bank resolves this. It absorbs the surplus through the quiet hours and releases it during the peaks, which is a different job from what a battery does in a solar system. There it covers days without sun. Here it covers minutes of heavy use, against a source that never stops refilling it.
Natural Resources Canada frames the alternative honestly: a stand-alone system without storage has to be sized to meet peak consumption if you want independence, and a battery-based system is what lets a smaller installation cover those peaks[1]. Sizing a hydro system for its worst instant rather than its average is how modest sites get talked out of being developed.
810 watts sounds small next to a 6,000 watt solar array. Across a day it is 19,440 watt-hours. The array in the solar section, fifteen panels on a northern site, produces 9,600 watt-hours on an average December day.
A turbine that fits on a bench beats fifteen panels in the month that matters, and it does so without a bank sized for four days of bad weather. This is why the hub calls water the best resource on the site.
Constant does not mean unchanging. The same creek carrying 320 gallons a minute in August may carry several times that during snowmelt, and a system designed around the August figure simply spills the surplus past the intake in April. That is the correct behavior, and it is the reason the previous guide insists on measuring the driest month rather than the prettiest one.
Checking the estimate
The screening estimate from the previous guide is worth testing against an independent formulation before anyone spends money on it. Natural Resources Canada gives the physical version in metric units[1].
P = Q × H × g × e
P in kilowatts, Q the usable flow in cubic meters per second, H the gross head in meters, g the gravitational constant at 9.8, and e an efficiency factor of 0.5 to 0.7.
Our creek in metric: 150 gallons per minute is 0.0095 cubic meters per second, and 60 feet of gross head is 18.3 meters. At the conservative end of the efficiency range, 0.0095 × 18.3 × 9.8 × 0.5 gives roughly 850 watts.
The Department of Energy shortcut gave 810 watts working from net head. The two agree within five percent, which is as close as two estimates with different assumptions have any right to come.
Working backward, the divide-by-ten shortcut implies a system efficiency near 53 percent when applied to net head. That sits at the bottom of the 50 to 70 percent band the Department of Energy considers representative of micro-hydro, which is a useful thing to know about a rule of thumb. It is not optimistic.
The gap between 50 and 70 percent is not a rounding allowance. It is turbine choice, how well that turbine matches the site's head and flow, generator losses, and penstock friction, and those are real decisions rather than a fudge factor. Matching the machine to the water is the subject of the turbine guide, and it is where a site moves from the bottom of that band toward the top.
Honest limits
Plenty of water and almost no drop means large pipe, large machinery and a poor return on both. Some low-head machines exist, but the economics are unkind and a storage scheme or a different source usually wins.
Seasonal watercourses that run hard in spring and go dry in August produce a system that is idle when demand is highest. Measuring the driest month, as the previous guide insists, catches this before the money is spent.
Good water half a mile from the house means a transmission run that costs real money and gives back real losses. Distance is survivable at higher voltage, but it changes the project from modest to substantial.
A stream carrying leaf fall, gravel or glacial silt will clog an intake faster than anyone wants to clear it, and abrasive sediment wears runners out. Intake design can manage this, and it raises both the cost and the chore list.
Catchments that rise several feet within hours of rain put intakes and powerhouses at risk. Everything has to be sited and anchored for the flood the stream produces once a decade, not the one it produces most weeks.
It is illegal to take surface water without a license or approval, and no amount of head and flow changes that. This is the first question to settle, not the last.
Next
Run-of-river describes how water moves through the property. It says nothing about which turbine converts it, how large the penstock has to be, or what keeps the electrical output stable. Those are the guides that follow.
If the site has more water than drop, or if power on demand matters more than continuous output, the small impoundments guide covers the alternative and is honest about what a dam costs in obligations. It comes next in reading order.
Sources
Head, flow and output figures in the examples are illustrative and carried forward from the measuring guide to keep one property in view across this section. Cost figures are deliberately omitted because published per-kilowatt estimates age quickly and site conditions dominate them. Get quotes for your own site and your own pipe run.