MICROHYDRO SYSTEMS

Water power has been used in

energy production for centuries.
In more recent times water
power for grain milling was
practised in various parts of
Africa including Ethiopia where
grain mills began to be installed
from the mid 19th Century. It was
the primary source of electricity
in many parts of Europe during
the initial stages of the "electricity
revolution" last century. Along
with the European colonisation of
Africa came hydro power and
units were installed in many parts The choice of turbine will also be dictated by cost,
of the continent primarily for availability, and which devices can be manufactured or
power on farms and estates and maintained locally. Many designs require casting of the
also for mission hospitals and turbine runner and an advantage of the cross flow design
schools. is it can be fabricated from sheet metal in a relatively
basic workshop.
Technologies are available to
match the types of water Uses of power
resources available; these can
Hydro power has the advantage of being able to supply
be high, medium and low motive and electrical power needs. Milling, grinding,
head, and technologies have sawing, oil expelling and many other agro industrial
been designed to suit different power needs can be supplied by direct drive from the
site specifications. turbine shaft negating the need for more complicated
electrical generation and control equipment. When
electricity generation is included power can be supplied to
High head turbines: isolated villages, industrial needs further from the turbine
site and for remote schools, hospitals and commercial
 pelton premises. Battery charging is also an option.
 turgo

Medium head turbines:

 crossflow
 turgo
 Francis
 reverse flow

Low head turbines:

 Mitchell Banki cross
flow turbines
 propeller

 Kaplan
 High 20m - 100+m; Medium 10m - 20m; Low 3m - 10m The
Head definitions (approx):
table below shows the power estimation in kW for different heads and flow rates of a
micro-hydro plant.

Flow rate (m/s) 1 m/s 2 m/s 3 m/s 4 m/s

Head (m)

2 10 20 30 40

4 20 40 60 80

6 30 60 90 120

8 40 80 120 160

10 50 100 150 200

Types of installations
Run of river
Storage (dams, lakes)
Integrated (irrigation, water supply systems)

System size
Battery charging, 100W to 1kW
Mechanical power, 1kW to 30kW
Stand-alone hydro electric, 1kW to 200kW
Grid connected, 20kW upwards

Outline costs (excluding alternator and drive)*

2kW Cross flow $1000-2000 2kW Pelton $1000 – 4000

10kW Crossflow $2000 - 10000 10kW Pelton $2000- 10000

Environmentally friendly
Micro hydro, or small-scale hydro, is one of the most environmentally benign energy conversion
options available, because unlike large-scale hydro power, it does not attempt to interfere
significantly with river flows.
Fraenkel (1991)

When is hydropower micro?

The definition of micro hydropower varies in different countries and can even include systems
with a capacity of a few megawatts. One of the many definitions for micro hydropower is: hydro
systems up till a rated capacity of approximately 300 kW capacity. The limit is set to 300 kW
because this is about the maximum size for most stand alone hydro systems not connected to
the grid, and suitable for "run-of-the-river" installations.

WATER TURBINES

Introduction

The word turbine was coined in 1828 by Claude Burdin (1790-1873) to describe the
subject of an engineering competition for a water power source. It comes from Latin
turbo, turbinis, meaning a "whirling" or a "vortex," and by extension a child's top or a
spindle. Defining a turbine as a rotating machine for deriving power from water is not
quite exact. The precise definition is a machine in which the water moves relatively to
the surfaces of the machine, as distinguished from machines in which such motion is
secondary, as with a cylinder and piston. The common water wheel is a rotating
machine, but not a turbine. We shall discuss many types of water-driven prime movers
in this article, but mainly turbines, for which will explain the fundamental theory.

Water in nature is a useful source of energy. It comes directly in mechanical form,

without the losses involved in heat engines and fuel cells, and no fuels are necessary.
Solar heat evaporates water, mostly from the oceans, where it is mixed into the lower
atmosphere by turbulence, and moved by the winds. Through meteorological processes,
it falls on the earth as precipitation, on the oceans, but also on high ground, where it
makes its way downhill to the sea, with evaporative and other losses. A cubic metre of
water can give 9800 J of mechanical energy for every metre it descends, and a flow of a
cubic metre per second in a fall of 1 m can provide 9800 W, or 13 hp. The efficiency of
hydraulic machines can be made close to 1, so that all this energy is available, and it can
be converted to electrical energy with an efficiency of over 95%.

The disadvantage of energy from water is that it is strictly limited, and widely
distributed in small amounts that are difficult to exploit. Only where a lot of water is
gathered in a large river, or where descent is rapid, is it possible to take economic
advantage. Most of these possibilities are quite small, as are the hydropower sites along
the Fall Line on the Atlantic coast of the United States, or on the slopes of the Pennines
in England. These were developed in the early days of the Industrial Revolution, but are
now abandoned because their scale is not the scale of modern industry. Each site
provided a strictly limited horsepower, and in the autumns the water often failed. For
expansion and reliability, all were rapidly replaced by steam engines fueled by coal,
which were expandable and reliable. Today, hydropower usually means a large project
on a major river, with extensive environmental damage. The fall in head is provided by
a dam, which creates a lake that will be of limited life, since geological processes hate
lakes and destroy them as rapidly as possible.

Niagara Falls is an excellent example of a hydropower site. It is unique; there is only

one, and hardly anything else similar. The Niagara River carries the entire discharge of
the Great Lakes, about 5520 m3/s, and the concentrated elevation difference is about 50
m. The visible falls carry nothing like this much water today; most is used for power.
Hydropower could destroy the falls as a sublime view; we are lucky it has not. The
power available from this discharge and drop is 3.6 x 10 6 hp. The figures given in the
encyclopedia for the power available from the Canadian and U.S. power projects on
each side add up to considerably more than this. Perhaps they use more drop, or perhaps
they are just optimistic. The first large-scale hydropower development here was in
1896. This was also the site of Nikola Tesla's two-phase plant that pioneered polyphase
power in the U.S.

For comparison, the more than 190 million registered motor vehicles in the U.S.
probably have an aggregate power capability of neary 2 x 10 10 hp, equivalent to 5000
Niagaras. Hydropower and increasing population cannot coexist; the limits of
hydropower are fixed and obvious. It is really too bad that small-scale hydropower
projects are no longer economically viable. In 1920, about 40% of electric power in the
U.S. came from hydropower; in 1989 that percentage had dropped to 9.5%. It was not
that hydropower had decreased in absolute terms, but had remained roughly constant
while the total market had expanded greatly.

Power From Water

To get power from water, it must be extracted as the water is lowered in elevation. That
there is a current in a stream is obvious; it comes from the open-channel flow of the
water under the influence of gravity. If you hold a paddle-wheel in the stream, it will be
rotated and you can extract power from it. This is an elementary impulse turbine, a
rotating machine acted upon by the force of moving water. This is not a copious source
of power, but at least it does not involve the disturbance of the stream, which was
always of intererest to the neighbors. Wheels like this are called undershot, and were
very common at all dates, because they can be used without disturbing the stream.
These were the typical Roman water mills, described in Vitruvius. There are, however,
very few relics remaining from these mills, and their history is often seriously
misrepresented. The wheels drove horizontal millstones through gearing, usually by a
face and a lantern gear. All parts of a mill were made of wood, except for small pieces
of metal where desirable.

Roman water mills were suggested by the irrigation machinery that had been long used
in the arid Near East to raise water. The Egyptian shaduf, a counterweighted pot, was
millennia old, raising water by about 2 m, and sometimes arranged in cascade. This was
not, of course, a source of energy but a consumer of it, though it made the most of
human effort. A much later development was the chain of pots, saqiya or Persian wheel,
that rotated to raise pots of water to a height, where they were automatically emptied
into an aqueduct as they rotated. This could be operated by man or beast, but some
ingenious person fitted the rim of the wheel with vanes or paddles. When dipped into
the current, they rotated the wheel with sufficient force to raise the pots with no
assistance. This was the remarkable noria, probably the first application of water power,
developed in the fourth century BCE at an unknown location. The Syrian city of Hama,
75 km inland from the Mediterranean, received its water from aqueducts filled by norias
on the River Orontes. Famous norias also were built in Portugal.

An outstanding water-raising project that should be more widely known was the
Artificio of Juanelo Turriano (? - 1585) at Toledo. Toledo is 600 m above the Rio Tajo,
and for centuries water had been laboriously carried up in leather bags on the backs of
mules. Turriano built an undershot wheel on the river that was equipped as a noria,
filling a basin on the bank. By means of reciprocating rods, a bucket filled from this
basin was rocked about an axis, tipping its water down a pipe into a following bucket.
This was continued up the slope to the city, each bucket rocking and alternately filling
and tipping into the next, in a mechanical bucket brigade to the top, doing the work of
600 mules.

These undershot mills were very sensitive to water level. To remedy this difficulty
where water level varied, wheels were mounted on pontoons that were moored in the
stream. Usually, the wheel was set between two boats, where the bearings would be
satisfactory and the water velocity the greatest. Mill boats were used on the
Guadalquivir at Códob, since the river level was quite variable with the season. Mill
boats were also used on the Thames in London, and beneath the Grand Pont in Paris, as
well as at Kön on the Rhine. If they were moored in the arches of a bridge they could
take advantage of the higher current there. The openings in medieval bridges were
usually small and inadequate, favoring this application. The invention of the floating
mill is traditionally ascribed to an emergency measure of the Roman general Belisaurius
in 537, while defending Rome from an Ostrogothic siege.

If you were permitted to construct a weir, and usually this involved an extended legal
wrangle over navigation rights, you could make available a few feet of drop all in one
place. If you made a spillway, you would have some rapid water at its base to put a
paddle wheel in. An alternative was to make the wheel move in a closely-fitting shroud,
or to provide it with buckets, so that the water would by its weight press down the
paddles. This is a completely different principle than the impulse machine, though many
machines derived energy from both sources. With care, most of the energy of the
descending water could be gathered, a distinctly more efficient operation than any
available type of impulse machine. Later, well-designed wheels with the water applied
at some height less than the wheel diameter were called breast-shot wheels because the
water entered part way up the diameter. A wheel of large diameter was usually
necessary to provide sufficient power at the desired slow speed. Breast wheels were the
most common type of mill wheel, even into the 20th century, since they were both
efficient and well suited to moderate heads.

The great weir on the River Dee at Chester supplied a head for the city's mills for many
years. Despite many orders for its destruction, it has lasted to the present. At Exeter in
Devon, there was great opposition to Countess Wear (an alternative spelling) built
across the Exe by the Countess of Devon that facilitated the collection of traditional
tolls on river traffic, as well as head for a mill. There were other weirs on this small
river, each supplying a mill, and a very early navigation canal was built to bypass the
weirs. Boats were often dragged upstream past a weir on water temporarily released by
a flash gate. Locks were a more recent affair, and now permit easy passage without a
waste of water.

Tidal mills were also built in favorable locations. Sometimes the rising and falling tidal
streams in an estuary could work an undershot wheel. More commonly, a tidal basin
was filled at high tide and then emptied past the wheel at low tide, giving two periods of
power a day for several hours in most places. The times of day were variable, but were
at least predictable. Tidal mills were on a much smaller scale than today's tidal power
schemes.

If the water was available at a sufficiently high level, perhaps through a canal or flume
from an upstream point, it could be introduced near the top of the wheel. These were
overshot wheels and, like breast wheels, used the weight of the water only, and were
quite efficient. Overshot wheels were used by the Romans wherever sufficient head was
available. At Barbegal, near Arles in southern France, a double row of 16 overshot
wheels, each 9 ft in diameter and 3 ft wide, was applied to flour milling. This
installation could provide flour for 80,000 people. At Laxey, on the Isle of Man, an
overshot wheel of 70 ft diameter, rotating at 2.5 rpm and developing 175 hp, was fed by
water under pressure that rose in a masonry column and was then led to the top of the
wheel in a wooden flume. The output of the wheel was used for pumping a lead mine.
This "Lady Isabella" wheel was built in 1854 or 1856, and served until the mine was
abandoned in 1929.

The water for an overshot wheel was often applied beyond the centre line, so that the
wheel rotated in the same sense as the water approached. If the water below the wheel
became too high, it would retard the wheel, an effect known as back watering. It was
easy to cure this by admitting the water on the other side of the centre, so the wheel
revolved in the opposite direction and the water below would aid the rotation, not
hinder it. The Laxey wheel was of this type, called a pitch-back wheel. Another help
was ventilating buckets that had openings to aid the discharge of water.

John Smeaton (1724-1792), one of the first modern Civil Engineers, and a very
successful one, founder of the Smeatonian Society that later became the Institution of
Civl Engineers, took great interest in improving water wheels, and in introducing cast
iron in their construction, just at the time when the rotary steam engine was coming into
being. Improvements in water wheels continued in the 19th century, by such engineers
as Robert Fairbairn, and were brought to a high state of excellence with metal replacing
most of the wood in both wheel and gearing. Since the wood around waterwheels is
alternately wet and dry, it is very subject to rotting, and this was a persistent problem.

The vanes, blades or buckets on a waterwheel were called floats, sometimes rungs.
Water was carried to and from the wheel in a mill race. The water arrived in a head
race, and left in a tail race. A head race was locally called a lade or a launder. A mill
race could be called a flume in the United States, or a leat in Britain. The water was
controlled by means of a sluice, sometimes redundantly called a "sluice gate." A sluice
is a vertically sliding gate, often operated by rack and pinion. A penstock may be a
sluice, but is also a closed channel through which water is delivered under pressure.

Water wheels can be run in reverse--driven, for example, by windmills, and function as
pumps. The most common water-lifting machine in Holland was the scoop wheel, very
much like a breast-shot wheel run in reverse. These were sometimes cascaded to give a
greater lift. For lesser discharges, an Archimedean screw rotated by a small four-sailed
windmill could be used, called a tjasker.

The Norse mill or Greek mill, was completely different, with a horizontal wheel and
vertical shaft called a tirl, that turned the millstone directly, without gearing. A stream
of water from a penstock was directed on the wheel, turning it by impulse. This was a
very primitive mill, but a quite practical one, which appeared in about 0 CE. It was not
displaced by the more elaborate Roman mill, with vertical wheel and horizontal shaft
with gearing, in many places. It is really not Greek nor Norse particularly, but widely
distributed in early communities. An later example was the Provençal mill from
southern France. Poncelet derived a design for an impulse turbine from this model, with
a horizontal metal wheel with curiously shaped floats fed by a penstock. These
horizontal mills were all impulse turbines, ancestors of the Pelton wheel. Although they
may bear a resemblance to Francis turbines, the resemblance is only superficial.

The uncertainty of water supply could be ameliorated by a mill pond, which stored
water to even out variable flows. Many mills had weir or dam, a mill pond, and a flume
to supply the wheel, making a very pleasant establishment. However, protracted drought
or winter frosts could bring the mill to a halt. In some cases, when a steam engine was
installed, it was used to pump water up to the mill pond, instead of being applied
directly to the machinery. This, of course, involved the least disturbance to the existing
machinery and allowed the steam engine to be immediately applied. When renovation
occurred, the opportunity was used to remove the wheel and connect the steam engine
directly to the machinery.

An alternative to turbines is the positive-displacement engine, in which a piston

reciprocates in a cylinder, valves admitting and releasing the water. This is especially
adaptable to high heads. One double stroke of the piston in effect lowers an amount of
water equal to the cylinder displacement from the high reservoir to the low, extracting
its potential energy. Care must be taken to cushion the shock of sudden opening and
closing of valves by air chambers and other means. These column-of-water engines
were extremely rare. The best known was built by Georg von Reichenbach of Bavaria
to power reciprocating brine pumps. A positive displacement rotary engine like the
Roots blower could easily be run in reverse on water pressure, and would, moreover,
function smoothly without shock. I do not know if this has been done. Reciprocating
engines have the advantage of being able to work at different speeds, which turbines
can do only with difficulty. Reciprocating pumps are much more common than
reciprocating engines.

Impulse Turbines: The Pelton Wheel

The impulse turbine is very easy to understand. A nozzle transforms water under a high
head into a powerful jet. The momentum of this jet is destroyed by striking the runner,
which absorbs the resulting force. If the velocity of the water leaving the runner is
nearly zero, all of the kinetic energy of the jet has been transformed into mechanical
energy, so the efficiency is high.

A practical impulse turbine was invented by

Lester A. Pelton (1829-1908) in California
around 1870. There were high-pressure jets
there used in placer mining, and a primitive
turbine called the hurdy-gurdy, a mere
rotating platform with vanes, had been used
since the '60's, driven by such jets. Pelton
also invented the split bucket, now
universally used, in 1880. Pelton is a trade
name for the products of the company he
originated, but the term is now used generically for all similar impulse turbines.

A diagram of a Pelton wheel is shown at the right. The wheel of pitch diameter D has
buckets around its periphery, so spaced that the jet always strikes more than one at a
time. The buckets have the form shown at the
upper left, where the water enters at a splitter and
is diverted to each side, where the velocity is
smoothly reversed. Bucket sizes are from 2.5 to 4
times the jet diameter. The total head supplying
the nozzle is h, the sum of the pressure head and
the approach velocity head. The theoretical jet
velocity is V = √(2gh). Let's analyze an ideal
wheel, and assume that this is actually the jet
velocity. The peripheral velocity of the runner is
u.

The vector diagrams at the left show how the velocity is transformed by the runner. For
simplicity, we assume that all velocities are in the same straight line. The relative
velocity of approach to the runner is V - u. We assume that this velocity is reversed, so
that the final velocity is V - 2u. The force F on the runner is the rate of momentum
change, or F = ρ[V + (V - 2u)]Q = 2ρ(V - u)Q, where ρ is the density and Q the volume
rate of flow of water. The torque on the runner is T = FD/2 = ρD(V - u)Q. When the
runner is stopped, the torque has its greatest value, ρDVQ. When the peripheral velocity
of the runner is equal to V, the torque is zero. The torque curve is a straight line
between these points.

The power P = Fu or Tω. In terms of u, P = 2ρu(V - u)Q. By taking the derivative of P

with respect to u and setting it equal to zero, we find that maximum power occurs when
u = V/2, and this power is ρV 2Q/2, or ρghQ. This is the energy content of the water
from the jet, so the efficiency is unity, with all the energy of the jet turned into shaft
output. For any velocity u, the efficiency is η = 4u(V-u)/V 2. It is zero for u = 0 and for u
= V. This analysis should have been clear and easy to follow. It illustrates the princple
of the Pelton wheel very well, and actual wheels are not too far from ideal. When a
Pelton wheel is working close to maximum efficiency, the water drops easily from the
wheel, with a little turmoil, but not much velocity.

As an example of an actual Pelton wheel, one worked for a time generating electricity in
Southern California with the following specifications. Pitch diameter, 162" (2.06 m);
operating speed, 250 rpm (26.18 rad/s); head, 2200' (670.6 m). The theoretical V is
√(2gh) = 114.6 m/s, while the peripheral velocity u = 53.86 m/s. Then, 2u = 108 m/s,
very close to V and probably closer to the actual jet velocity. This wheel probably
developed about 60,000 hp on a flow of around 7 m3/s. The ratio of the runner velocity
u to the ideal jet velocity √(2gh) is usually denoted φ. For a Pelton wheel working at
maximum efficiency, φ is about 0.5.

The conduit bringing high-pressure water to the impulse wheel is called the penstock.
This was strictly just the name of the valve, but the term has been extended to the
conduit and its appurtenances as well, and is a general term for a water passage and
control that is under pressure, whether it serves an impulse turbine or not.

Reaction Turbines: The Lawn Sprinkler

By contrast with the impulse turbine, reaction turbines are difficult to understand and
analyze, especially the ones usually met with in practice. The modest lawn sprinkler
comes to our aid, since it is both a reaction turbine, and easy to understand. It will be
our introduction to reaction turbines. In the impulse turbine, the pressure change
occurred in the nozzle, where pressure head was converted into kinetic energy. There
was no pressure change in the runner, which had the sole duty of turning momentum
change into torque. In the reaction turbine, the pressure change occurs in the runner
itself at the same time that the force is exerted. The force still comes from rate of
change of momentum, but not as obviously as in the impulse turbine.

The duty of the lawn sprinkler is to spread water; its energy output as a turbine serves
only to move the sprinkler head. It is a descendant of Hero's aeolipile, the rotating globe
with two bent jets that was quite a sensation in ancient times, though this worked with
steam, not water. The lawn sprinkler seems directly descended from Rev. Robert
Barker's proposed mill of 1740. He used two jets at right angles to the radius. A later
improvement fed water from below to balance the weight of the runner and reduce
friction. Barker's mills only appeared as models, and were never commercially offered.
The flow of water in a lawn sprinkler is radially outward. Water under pressure is
introduced at the centre, and jets of water that can cover the area necessary issue from
the ends of the arms at zero gauge pressure. The pressure decrease occurs in the
sprinkler arms. Though the water is projected at an angle to the radius, the water from
an operating sprinkler moves almost along a radius. If you have such a sprinkler, by all
means observe it in action. The jets do not impinge on a runner; in fact, they are leaving
the runner, so their momentum is not converted into force as in the impulse turbine. The
force on the runner must act in reaction to the creation of the momentum instead, which
is, of course, the origin of the name of the reaction turbine.

A two-armed runner of a rotating lawn sprinkler is shown at the right. Conditions at the
ends of the two arms are the same. The jet at the end of an arm is projected at an angle β
with a perpendicular normal to the radius from the centre of rotation, in the direction of
the rotational velocity u2 = ωr. The space velocity V2 is the vector sum of v2 and u2,
which makes an angle α with u 2. When the runner is stopped, V2 = v2. As the runner
speeds up, V2 moves closer to a radial direction. When it reaches the radial direction,
there is no longer a component normal to the radius and, therefore, no accelerating
torque. It is easy to see that the torque will be a maximum when the runner is stalled.

To find v2 in terms of p1, we shall use Bernoulli's theorem. However, energy is not
conserved between the axial point 1 and point 2 at the end of the arm, since the water
does work in passing from one point to the other. There is a reaction force of magnitude
ρV2 in the opposite direction to V 2. The movement of point 2 is in the direction of u 2, so
the rate of doing work is ρV2u2cos α. Dividing by ρg to express this work as head, we
find that a head of V2u2cos α/g must be subtracted from the difference of the heads at
points 2 and 1. Since z2 = z1, and V1 = 0 at point 1, and p = 0 at point 2 (we are using
gauge pressures), we get V22/2g = p1/ρg - V2u2cos α/g.

From the vector triangles, we find that (dropping the subscript 2 for the moment) V 2 =
u2 + v2 - 2uv cos(180° - β) = u2 + v2 + 2uv cos β, and also that V cosα = v cos β + u.
Substituting for V2 and V2cosα in the above equation, we find that (v 2 - u2)/2g = p1/ρg =
h, which simplifies to v 2 = 2gh + u2. In this equation, h is the supply head, which may
include approach velocity if it is to be considered.

Now we can find everything we need as a function of u 2, or of the angular velocity ω. In

particular the component of V perpendicular to the radius is V' = V cos α = v cos β + u.
The corresponding reaction force is obtained by multiplying by ρ, and by the total
discharge Av, where A is the total area of the jets on all arms. The torque, then, is T = -
ρAvr(v cos β + u), where v = √(2gh + u2) and u = ωr.

A lawn sprinkler I have at hand has r = 75 mm, β = 110° and A = 8 mm 2. Let us first
find the free-running speed for a supply head of 300 cm. The free-running speed
corresponds to zero torque, when V is radial, or when v cos β + u = 0. Here, this
condition is u = 0.342v, or v2 = 8.549u2. Since v2 = 2gh + u2, we have 7.549u2 = (2)
(980)(300), or u = 279 cm/s. Since r = 7.5 cm, ω = 37.2 rad/s, or n = 355 rpm (n is the
usual symbol for angular velocity in rpm). This value will not be reached, of course, due
to friction, but it is very reasonable. Under a tenth of the head, the free-running speed
would be 112 rpm.

Let us assume that the actual speed under 300 cm head is 200 rpm, and find the
corresponding torque. For this runner, T = 0.60v(0.342v - u) cm-dyne. 200 rpm is 20.94
rad/s, so u = 157 cm/s. Then, we find that v = 783 cm/s, from which T = 5.17 x 10 4 cm-
dyne, or 53 cm-gm, as it is usually stated. This will be the torque required to turn the
runner at that speed. The discharge Q = Av = (0.08)(783) = 62.6 cm 3/s, or 3.76
liters/min. At stall, v = 767 cm/s, and the stall torque will be 12.07 x 10 4 cm-dyne, or
123 cm-gm. Note that the rotation of the runner causes v to be larger than it would be
without the rotation; the runner is acting as a centrifugal pump.

The water is projected with a radial velocity of V sin α = v sin β = 0.940v. Let's assume
that it is projected horizontally, although in the actual sprinkler the jets are aimed
upwards a little. Then the radial velocity will be 736 cm/s. If the height of the runner is
1 m above the ground, it is easy to calculate that the water will be projected to a radius
of 3.3 m. If the water were spread uniformly over this disc, the watering rate would be
6.6 mm per hour.

If we were more interested in power than in watering, β could be made 180°, and the
area of the jets could be increased, partly by multiplying the number of jets. If the
angular velocity of the runner could be such that v = u, the water would drop directly
down, and the efficiency of the turbine would be a maximum. However, we must have
v2 = 2gh + u2, so this condition cannot exist. All that can be done is to make u as large
as possible, but this is not very satisfactory. This is the reason Barker's mills are not
often seen these days.

An interesting related problem is the "reverse sprinkler," also known as Feynman's

sprinkler, though he did not originate the problem or give a solution (see Reference).
Imagine a sprinkler like the one described above immersed in water and suppose the
water is sucked out of it. There is a slight impulse towards the side where the water
enters when flow begins, and an equal reverse impulse when the flow stops, but for an
ideal fluid there is no steady torque. It is interesting to analyze this problem. The
momentum given to the fluid on entry is the same that is given up when the fluid turns
and moves into the supply pipe, so there is no net torque. There was a considerable
amount of argument concerning this problem in the 1980's, but the confusion over it is
much older.

Energy Relations
To analyze power turbines, we'll use Bernoulli's theorem in the form derived for the
water sprinkler, written between the inlet and outlet of the turbine runner. This is (p 1/ρg
+ z1 + V12/2g) - (p2/ρg + z2 + V22/2g) = hL + (V1u1cosα1 - V2u2cosα2)/g. The terms on the
right are the head loss in the supply piping, h L, and the utilized head h" which depends
on conditions at inlet and outlet. There is a velocity triangle at inlet and outlet relating
V, u and v, as in the case of the lawn sprinkler. There, we could use it to eliminate V
from this equation in order to find v.

Practical turbines cannot be well described by a single streamline, but we can still think
of average values of V1 and V2, u1 and u2, and the other quantities, and find out how
they vary with the size of the turbine, the rotational speed, the developed horsepower,
and the applied head. In fact, the theory of practical turbines is difficult, so tests are
actually used to determine the variables under study. Experiments are made on models,
and scaled up to full size, so understanding the relation between model parameters and
full-size parameters is important.

The principle of a power turbine is to direct the incoming water tangentially by

stationary vanes, and then to have it pass to the moving runner where it exerts forces on
the runner vanes while its pressure decreases from the input head to zero. Since the
pressure varies, the turbine must flow full. The exit velocity is not zero, but most of the
kinetic energy can be recovered in a draft tube where the water is decelerated. In 1826,
Bénoit Fourneyron (1802-1867) developed an outward-flow turbine that was efficient,
but the mechanical arrangements were poor, since the runner was on the outside. Jean
V. Poncelet (1788-1867) designed an inward-flow turbine in about 1820. S. B. Howd
took the design to the U.S. and patented it in 1838 (I do not know if he made any
improvments, or just filched the idea). In 1849, James B. Francis (1815-1892) improved
Howd's turbine, where the water entered horizontally through guide vanes that gave
them a whirl, then passed into the runner and was diverted downwards. His hydraulic
experiments on turbines at Lowell, Massachusetts, are famous. James Thomson of
Belfast, brother of Lord Kelvin, made important improvements to the inward radial flow
turbine, in the shape of the vanes, control, and other matters. Practical reaction turbines
are now all inward-flow machines. The water enters through a snail-like scroll case, and
exits below near the axis.

The runner of a Francis turbine is illustrated at the left. Its basic dimension is the
diameter D. The shape of the vanes cannot be well represented, but they are designed
for smooth flow at the design speed and head of the turbine. The plan view at the right
shows how the guide vanes in the stator direct the water onto the moving runner, acting
like nozzles. The water follows the dotted path in space from the inlet at 1 to the outlet
at 2. Relative to the runner, it flows parallel to the vanes, exerting the force that creates
the output torque. In this diagram, it is easy to imagine the velocity triangles at input
and output, which will be similar to those for the lawn sprinkler.

The vertical section at the left shows that the flow is not completely radial, as it was in
the earliest Francis turbines. During its passage through the runner, the water is diverted
axially, and exits at the bottom of the runner. This, of course, complicates our analysis,
but nothing is really fundamentally changed. The mixed flow allows a more efficient
turbine by making the exit smoother. The interested reader should try to find
illustrations of actual Francis runners to appreciate their complex shape. It is clear that a
single streamline is not sufficient to describe their action!

The distinction between radial and axial flow has a great effect on the appearance of the
turbine, but it does not affect its fundamental behavior. Hydraulic turbines can be made
that are almost completely axial flow, the runner taking the form of vanes perpendicular
to the axis, well-described by the term propeller turbines. An example is the Kaplan
turbine, invented by Victor Kaplan (1876-1934) and first put into service in 1912-13,
with movable blades that rotate, or "feather," to handle different conditions, the key to
making an efficient propeller turbine. In fact, the guide vanes of a Francis turbine are
usually movable for the same purpose. A turbine without such adjustments will work
efficiently only at its design speed and head. The water is given a swirl at the top of a
Kaplan turbine that is taken out by the propeller.

Torque is the rate of change of angular momentum, just as force is the rate of change of
linear momentum. When a fluid exerts a torque on a turbine runner, the reaction is a
change in angular momentum of the fluid. Fluid is given angular momentum by the
guide vanes which, ideally, is destroyed by the torque exerted on the runner. With some
machines, however, the water at the exit may still have considerable angular
momentum, and the energy in this motion is energy that does not appear at the shaft.
Where velocity in the exit fluid is part of the desired output (as with a fan), vanes to
straighten out the flow help to recover some of the energy that would otherwise be lost.
When a draft tube at the outlet of a turbine is filled with water, the pressure is less than
atmospheric at the turbine outlet. This, of course, is a desired effect so that advantage
can be taken of the whole drop in head even when the turbine is above the level of the
tail water. In a Kaplan turbine, the effect of the runner was to reduce the pressure even
more, below the vapor pressure of water at that temperature. Small bubbles of vapor
were produced, and when they reached higher pressure collapsed explosively, damaging
the runner. This is called cavitation and is a hazard in all hydraulic machinery where the
pressure may drop sufficiently. This almost caused the failure of the Kaplan turbine, but
the problem was eventually solved by taking care with the turbine setting and other
details.

The power developed by a turbine is P = ρgQh", where Q is the flow through the
runner. The power in the inlet water is P' = ρgQh, so the efficiency of the turbine is e =
h"/h. Strictly speaking, this is the hydraulic efficiency eh. The mechanical efficiency em
is the fraction of the runner power that is delivered at the shaft, lessened because of
friction. Hydraulic machines have a mechanical efficiency of 0.95 to 0.98 for large
machines, usually closer to the higher figure. Some of the supplied water may leak by
the runner and do no work. The volumetric efficiency is ev = (Q - QL)/Q, where QL is the
leakage. This is also quite high in practical machines. The overall efficiency is e =
P/ρgQh, where P is the shaft power output. P is usually expressed in hp (550 ft-lb/s) or
watt. 746 W = 1 hp.

If f is the fraction of the runner inlet that is open, then the inlet area is A = fπDB =
fπmD2, where m = B/D. The radial velocity at inlet is written V r = V1cosα1 = C1√(2gh).
Therefore, Q = AVr = KqD2√h, which shows how Q depends on the size of the machine
and the head. The power output will be P = eρgQh = eρgK qD2h3/2 = KpD2h3/2. The
constant Kp will be the same for machines of the same design.

We have already mentioned the parameter φ with relation to the Pelton wheel. Using it,
we can express the rotational speed in rpm by n = 60u/πD = (60φ√(2g)/π) h 1/2/D. We
note that n√(P) = const. x h 5/4, so that the combination n√P/h 5/4 will be a constant for a
particular machine, or machines similar to it. This relation between speed, power and
head for a turbine is very useful. The value of the expression is called the specific speed
ns, but it is not really a speed, and it should be noted that the expression is not a
dimensionless number. Also, the speed and power used in it should be the speed and
power for maximum efficiency, so that the constants are really constant. Even with
these cautions, it is a valuable way to classify turbines.

In United States practice, P is in hp and h is in feet, and n s is quoted assuming these

units. In metric practice, P is in kW and h is in metre. This gives a larger n s, which can
be reduced to the U.S. practice by multiplying by 0.2626.

Impulse turbines have low ns, from 1 to 10. A typical value for a Pelton wheel might be
4. Francis turbines have a specific speed of from 10 to 100, while Kaplan turbines give
from 100 up. These are values for well-designed, efficient machines. Of course,
monsters could be made with very different specific speeds but they would not be
satisfactory. If you know the head, speed and power desired, it is easy to find the
general type of turbine that would prove satisfactory.

Suppose you have a head of 2200 ft available, and want a 250 rpm machine delivering
60,000 hp. The specific speed will then be (250)√(60,000)/22005/4 = 4.06. This points to
a Pelton wheel, and, in fact, one was used under these conditions. If you have a head of
89 ft available on the Susquehanna at Conowingo, PA, and want 54,000 hp at 81.8 rpm,
then the specific speed is 69.5, clearly pointing to a Francis turbine. In the machine
used, D = 18 ft. At Rock River, IL, a head of only 7 ft is available. 800 hp at 80 rpm is
required. The specific speed is 199, clearly indicating a Kaplan turbine, which was
installed. The maximum horsepower that can be developed can be estimated by the
discharge Q and the available head. In engineering units, P max = 3.65Qh hp, where Q is
in cfs and h is in feet. The speed often depends on the speed for a directly-coupled
alternator. If N is the number of poles and f the frequency, then n = 60f/(P/2) = 120(f/P).
For f = 60 Hz and N = 24, n = 300 rpm. Alternators have from 12 to 96 poles, usually,
so rotational speeds will range from 600 rpm to 75 rpm.

References
R. L. Daugherty and J. B. Franzini, Fluid Mechanics, 6th ed. (New York: McGraw-Hill,
1965). Chapters 15 and 16 deal with turbines, but the theoretical background is
scattered in several previous chapters, chiefly 6 and 14. Machines are the most difficult
part of engineering fluid mechanics, but the most fascinating part. It is remarkable that
the conservation laws go so far to explain turbines.

J. Reynolds, Windmills and Watermills (New York: Praeger, 1970). Very well
illustrated; covers all kinds of historic mills.

M. Vitruvius, De Architectura (Cambridge, MA: Harvard University Press, 1934. Loeb

Classical Library #280). Vol II, Book X, Chapter V, pp. 304-307.

S. Strandh, A History of the Machine (New York: A&W Publishers, 1979). pp. 92-108.
Well-illustrated.

P. N. Wilson, Water Turbines (London: HMSO, 1974). A Science Museum Booklet.

Well-illustrated.

A. Jenkins, An Elementary Treatment of the Reverse Sprinkler, Am. J. Phys. 72, 1276-
1282 (2004).

TURBINES

A turbine converts energy in the form of falling water into rotating shaft power.
The selection of the best turbine for any particular hydro site depends on the site
characteristics, the dominant ones being the head and flow available. Selection
also depends on the desired running speed of the generator or other device
loading the turbine. Other considerations such as whether the turbine is
expected to produce power under part-flow conditions, also play an important
role in the selection. All turbines have a power-speed characteristic. They will
tend to run most efficiently at a particular speed, head and flow combination.

A turbine design speed is largely determined by the head under which it

operates. Turbines can be classified as high head, medium head or low head
machines. Turbines are also divided by their principle way of operating and can
be either impulse or reaction turbines.

high head medium head low head

impulse Pelton cross-flow cross-flow

turbines
Turgo multi-jet Pelton

Turgo

reaction Francis propeller

turbines
Kaplan

The rotating element (called `runner') of a reaction turbine is fully immersed in

water and is enclosed in a pressure casing. The runner blades are profiled so that
pressure differences across them impose lift forces, like those on aircraft wings,
which cause the runner to rotate.

In contrast a impulse turbine runner operates in air, driven by a jet (or jets) of
water. Here the water remains at atmospheric pressure before and after making
contact with the runner blades. In this case a nozzle converts the pressurised low
velocity water into a high speed jet. The runner blades deflect the jet so as to
maximise the change of momentum of the water and thus maximising the force on
the blades.

Impulse turbines are usually cheaper then reaction turbines because there is no
need for a specialist pressure casing. nor for carefully engineered clearances, but
they are also only suitable for relatively high heads.
 Impulse turbines

Impulse turbines are generally more suitable for micro-hydro applications

compared with reaction turbines because they have the following advantages:

 greater tolerance of sand and other particles in the water,

 better access to working parts,
 no pressure seals around the shaft,
 easier to fabricate and maintain,
 better part-flow efficiency.

The major disadvantage of impulse turbines is that they are mostly unsuitable for low-
head sites because of their low specific speeds too great an increase in speed would be
required of the transmission to enable coupling to a standard alternator. The crossflow,
Turgo and multi-jet Pelton are suitable at medium heads.

Pelton turbine

A Pelton turbine consists of a set of specially shaped buckets mounted on a periphery of

a circular disc. It is turned by jets of water which are discharged from one or more
nozzles and strike the buckets. The buckets are split into two halves so that the central
area does not act as a dead spot incapable of deflecting water away from the oncoming
jet. The cutaway on the lower lip allows the following bucket to move further before
cutting off the jet propelling the bucket ahead of it and also permits a smoother entrance
of the bucket into the jet. The Pelton bucket is designed to deflect the jet through 165
degrees (not 180 degrees) which is the maximum angle possible without the return jet
interfering with the following bucket for the oncoming jet.

Runner of a Pelton turbine

 source: http://www.hydrowest.com/runners1.htm

bucket shape
source: http://europa.eu.int/en/comm/dg17/hydro/layman2.pdf

In large scale hydro installation Pelton turbines are normally only considered for
heads above 150 m, but for micro-hydro applications Pelton turbines can be
used effectively at heads down to about 20 m. Pelton turbines are not used at
lower heads because their rotational speeds becomes very slow and the runner
required is very large and unwieldy. If runner size and low speed do not pose a
problem for a particular installation, then a Pelton turbine can be used efficiently
with fairly low heads. If a higher running speed and smaller runner are required
then the are two further options:

- increasing the number of jets.

Having two or more jets enables a smaller runner to be used for a given
flow and increases the rotational speed. The required power can still be
attained and the part-flow efficiency is especially good because the wheel
can be run on a reduced number of jets with each jet in use still receiving
the optimum flow.
 - twin runners.

Two runners can be placed on the same shaft either side by side or on
opposite sides of the generator. This configuration is unusual and would
only be used if the number of jets per runner had already been
maximised, but it allow the use of smaller diameter and hence faster
rotating runners.

Turgo turbine

The Turgo turbine is an impulse machine similar to a Pelton turbine but which
was designed to have a higher specific speed. In this case the jets aimed to
strike the plane of the runner on one side and exists on the other. Therefore the
flow rate is not limited by the discharged fluid interfering with the incoming jet (as
is the case with Pelton turbines). As a consequence, a Turgo turbine can have a
smaller diameter runner than a Pelton for an equivalent power. With smaller
faster spinning runners, it is more likely to be possible to connect Turgo turbines
directly to the generator rather than having to go via a costly speed-increasing
transmission.

Turgo runner blades and water jet

source: http://europa.eu.int/en/comm/dg17/hydro/layman2.pdf

Like the Pelton, the Turgo is efficient over a wide range of speeds and shares
the general characteristics of impulse turbines listed for the Pelton, including the
fact that it can be mounted either horizontally or vertically. A Turgo runner is
more difficult to make than a Pelton and the vanes of the runner are more fragile
than Pelton buckets. At one time they were exclusively made by Gilbert, Gilkes
and Gordon a UK manufacturer who owned the patent rights, but they are now
manufactured in several other countries.

The Ghatta and the Multi-Purpose Power Unit

The Ghatta is a traditional Nepalese waterwheel with a vertical axis. The water enter the
waterwheel from above. The turbine is made out of wood to enable simple building and
repair techniques to be used. A consequent of this design are low efficiency and power
output (maximum 12 kW). Out of this traditional Ghatta the improved Ghatta was
developed. The wooden waterwheel was improved and replaced later with a steel one with
round buckets. This improved the momentum transfer of the water and doubled power
output.

Multi-Purpose Power Unit

The Multi-Purpose Power Unit (MPPU) is chronological situated in between the

Ghatta and the improved Ghatta. The name multi-purpose refers to the
construction of the MPPU which enables the connection of various machinery to it.
The concept of the MPPU is basically the same as that of the improved Ghatta: a
vertical axis with a fixed and a rotating grinding stone. Technical complexity, power
output and price are in between those of the improved Ghatta and crossflow
turbines.

All components are of steel instead of wood, water supply is improved and friction
losses are reduced compared to the improved Ghatta.

Design philosophy was to produce a device as cheap and simple as possible.

Special attention was given to transportability.

Crossflow turbine

Also called a Michell-Banki turbine a crossflow turbine has a drum-shaped runner

consisting of two parallel discs connected together near their rims by a series of
curved blades. A crossflow turbine always has its runner shaft horizontal (unlike
Pelton and Turgo turbines which can have either horizontal or vertical shaft
orientation).

Operation:

In operation a rectangular nozzle directs the jet onto the full length of the runner. The water strikes
blades and imparts most of its kinetic energy. It then passes through the runner and strikes the bla
again on exit, impacting a smaller amount of energy before leaving the turbine. Although strictly cl
as an impulse turbine, hydro dynamic pressure forces are also involved and a mixed flow definition
would be more accurate.

Part flow efficiency:

A high part-flow efficiency can be maintained at less than a quarter of full flow by the arrangement
flow portioning illustrated in the figure. At low flows, the water can be channelled through either tw
thirds or one third of the runner, thereby sustaining a relatively high turbine efficiency.
 Part-flow efficiency of a partitioned crossflow turbine
based on http://www.ossberger.de/1-3b.html

Reaction turbines

The reaction turbines considered here are the Francis turbine and the propeller turbine. A
special case of the propeller turbine is the Kaplan. In all these cases, specific speed is high,
i.e reaction turbines rotate faster than impulse turbines given the same head and flow
conditions. This has the very important consequences that a reaction turbine can often be
compiled directly to an alternator without requiring a speed-increasing drive system. Some
manufacturers make combined turbine-generator sets of this sort. Significant cost savings are
made in eliminating the drive and the maintenance of the hydro unit is very much simpler. The
Francis turbine is suitable for medium heads, while the propeller is more suitable for low
heads.

On the whole reaction turbines require more sophisticated fabrication than impulse turbines
because they involve the use of larger and more intricately profiled blades together with
carefully profiled casings. The extra expenses involved is offset by high efficiency and the
advantages of high running speeds at low heads from relatively compact machines.

Fabrication constraints make these turbines less attractive for use in micro-hydro in
developing countries. Nevertheless because of the importance of low head micro-hydro, work
is being undertaken to develop propeller machines which are simpler to construct. Most
reaction turbines tend to have poor part-flow efficiency characteristics.

Francis turbine

Francis turbines can either be volute-cased or open-flume machines. The spiral casing is
tapered to distribute water uniformly around the entire perimeter of the runner and the guide
vanes feed the water into the runner at the correct angle. The runner blades are profiled in a
complex manner and direct the water so that it exits axially from centre of the runner. In doing
so the water imparts most of its pressure energy to the runner before leaving the turbine via a
draft tube.

The Francis turbine is generally fitted with adjustable guide vanes. These regulate the water
flow as it enters the runner and are usually linked to a governing system which matches flow
to turbine loading in the same way as a spear valve or deflector plate in a Pelton turbine.
When the flow is reduced the efficiency of the turbine falls away.
 Francis turbine
(source: adapted from http://lingolex.com/bilc/engine.html & http://starfire.ne.uiuc.edu/~ne201/1996/schoenau/wheel~1.html)

Propeller turbine

The basic propeller turbine consists of a propeller, similar to a ship's propeller, fitted inside a
continuation of the penstock tube. The turbine shaft passes out of the tube at the point
where the tube changes direction. The propeller usually has three to six blades, three in the
case of very low head units and the water flow is regulated by static blades or swivel gates
("wicket gates") just upstream of the propeller. This kind of propeller turbine is known as a
fixed blade axial flow turbine because the pitch angle of the rotor blades cannot be changed.
The part-flow efficiency of fixed-blade propeller turbines tend to be very poor.

Kaplan
 Kaplan turbine
Source: http://www.uni-muenster.de/Energie/wasser/technik/technik.html

Large scale hydro sites make use of more sophisticated versions of the propeller
turbines. Varying the pitch of the propeller blades together with wicket gate
adjustment enables reasonable efficiency to be maintained under part flow
conditions. Such turbines are know as variable pitch or Kaplan turbines.

Reverse pump turbines

Centrifugal pumps can be used as turbines by passing water through them in reverse. Research is
currently being done to enable the performance of pumps as turbines to be predicated more accur

The potential advantages are the low cost due to mass production (and in many cases also local
production), the availability of spare parts and the wider dealer/support networks. The disadvantag
are the as yet poorly understood performance characteristics and very poor part-flow efficiency. V
companies have used pumps as turbines at various times, but the technology remains unproven a
relatively poor in performance.

Micro hydropower basics

Drive systems
Other documents in this series:
Introduction
Civil work components
Turbines
Drive systems (this document)
Electrical power
Measurement of head
Measurement of flow

Back to the micro hydro home page

The drive system transmits power from the turbine shaft to the
generator shaft or the shaft powering an other device. It also has the
function of changing the rotational speed from the one shaft to the other
when the turbine speed is different to the required speed of the
alternator or device.

The following options can be considered for micro hydropower drive

systems:

 direct drive,
 flat belt and pulleys,
 V or wedge belt and pulleys,
 chain and sprocket,
 gearbox.

Direct drive

This system is only for the case where the shaft speeds are identical
because it uses a flexible coupling to join the two shafts together
directly. The advantages are low maintenance, high efficiency (>98%)
and low cost. The only disadvantage is that the alignment is far more
critical than with an indirect drive.
 Flat belt and pulleys

Modern flat belts run at high tension and are made of a strong inner
band coated with a high friction material such as rubber. They have
higher efficiencies than V-belts drives and run cleaner (ie with less
rubber dust). One pulley must have a slightly convex profile (crowned)
which together with good alignment, keeps the belt in position in either
vertical or horizontal use.

The main disadvantage is that a higher tension is needed than with

other drives (two tons is not unusual) which means that the bearings
suffer high loads, sometimes requiring additional layshafts to be used
or standard alternators to be fitted with heavier duty bearings. Also their
availability in some areas is less good than that of V-belt drives.

Flat belts generally require narrower pulleys than the equivalent multi
V-belt with advantages in cost and reduced overhang. Their maximum
speed ratio is around 5:1.

"V" or wedge belts and pulleys

This is the most common choice for micro hydropower schemes up to

100 kW. A major advantage is that these belts are very well known
because of their extensive use in all kinds of small industrial machinery,
hence they are also widely available.

V-belts differ from flat belts in that the frictional grip on the pulley is
caused by wedging action of the side walls of the belt within the pulley
grooves. Therefore less longitudinal tension is required to maintain the
grip and less radial load is imposed on the shaft and bearings.

Usually a number of V-belts are run side by side in multiple-governed

pulleys. Matched sets of belts are required to ensure even tension and
these sets can be difficult to obtain in some countries. At higher powers
and torques multiple V-belt installations can become cumbersome with
eight or more large belts and very wide pulleys.

The tolerance of misalignment of V-belts is very good but efficiency is

low than other types of belt at around 85-95 %. At very low powers the
low efficiency can be a problem and timing belts are often preferred.
Maximum speed ratio is around 5:1.

Timing belt and sprocketed pulley

These drives are commonly found on vehicle camshaft drives and

involve toothed belts and pulleys. They are very efficient (about 98%)
and clean running. The belt tension is lower than in any other belt drive,
giving reduced bearing loads, but the belts do not slip on overload so
cannot protect the shafts and bearings.

The main disadvantages are the cost of belts and pulleys and the low
availability. They are specially worth considering for smaller drives (less
than 3 kW) where efficiency is at a premium. Speed ratio can be up to
10:1.

Chain and sprocket

Chains can have a very high efficiency but only at some sacrifice of
lifetime. Long-life chain drives tend to be similar in efficiency to belt
drives. Chain drives are not recommended because of their high cost,
poor availability, the need to replace sprocket wheels periodically and
the difficult lubrication requirements. Very high speed ratios of greater
than 20:1 can however be achieved.

Gearbox

Gearboxes are used with larger machines when belt drives become too
cumbersome and inefficient. Problems of specification alignment,
maintenance and cost rule them out except in cases where they are
specified as part of a turbine-generator set.

Micro hydropower basics

 Electrical power
Other documents in this series:
Introduction
Civil work components
Turbines
Drive systems
Electrical power (this document)
Measurement of head
Measurement of flow

Back to the micro hydro home page

Machinery can be driven directly by a turbine as in traditional corn mills and

many modern timber sawing mills, but converting the power into electricity does
have several advantages. For instance, it enables the use of all types of
electrical appliances from lighting to electric motors and the flexible positioning
of the appliances to wherever a power point can be set up near or far from the
turbine. A device which converts a generator. The most common type of
generator produces alternative current and is known as an alternator.

As a brief reminder of simple electrical theory, the flow of electricity or current

(symbol I) is measured in amps (A). The potential difference (V) or pressure of
the electricity is measured in volts (V). The power (P) measured in watts (W) or
more often kilowatts (1 kW=1000 W), is equal to volts× amps. The resistance
(R) of a circuit is a measure of how well the electricity is being conducted (a
poor conductor has a high resistance). Resistance is measured in Ohms (W )
and is equal to the potential difference (or voltage drop) divided by the current.
Capacitance (C) expresses the degree to which energy is being stored in an
electric field rather than being available to do work and inductance (L) is similar
to capacitance but refers to magnetic fields.

AC and DC

Two types of current are produced by electrical generators, either alternating

current (AC) or direct current (DC). In the case of AC a voltage cycles
sinusoidally with time, from positive peak value to negative. Because the
voltage changes its sign the resulting current also continually reverses direction
in a cyclic pattern. DC current flows in a single direction as the result of a
steady voltage. DC is not usually used in modern power installations except for
very low-powered systems of a few hundred watts or less.

Alternating voltage can be produced in a stationery coil or armature by a

rotating magnetic field (figure18b), but more usually a coil is rotated in a
stationary magnetic field (figure 18a). The magnetic field can be produced
either by a permanent magnet or by another coil (i.e an electro-magnet) know
as a field coil (as in figure 18c and 18d) which is fed by direct current known as
the excitation current. A generator supplying alternative current is described as
an alternator to distinguish it from a machine designed to supply DC current
which is know as a DC generator or dynamo.

Figure 18 Alternator configuration

Current flow when a voltage difference is place across a conducting body. In

AC circuits the magnitude and timing of the current cycle relative to the voltage
cycle will depend on whether the conductivity body is resistance, inductive,
capacitive or some combination of these elements.

The first three cases in figure 19 show that for different types of load, the
current cycle either (a) stays in phase with the voltage, (b) lags behind the
voltage by a phase angle of 90° or (c) runs ahead ("leads") by a phase angle of
90° . Circuits in which the load causes the current and voltage to be out of the
phase are said to have reactive loads. Generally a load is a combination of
resistance, capacitance and inductance, described by a term impedance
(symbol Z) and causes a phase difference between current and voltage of
angle f.

Power factor

Figure 19d shows the current-voltage characteristic of a circuit where the load
causes the current I to lag the voltage V by an angle f (the load in this case is
predominantly inductive). Because the current and voltage are not perfectly in
phase, the useful power available is reduced and is proportional to the cosine
of the phase difference, so the power usefully consumed by the load is V× I×
cosf , although the power supplied is V× I. The power not consumed is simply
being shunted back and forth between supply and load. The ration of useful
power to total supplied power is called the power factor and is numerically
equal to cosf
 Figure 19 Principles of reactance

Micro hydropower basics

Measurement of head
Other documents in this series: Within this document:
Introduction Dumpy levels and theodolite
Civil work components Sighting meters
Turbines Water-filled tube and pressure
Drive systems gauge
Electrical power Water filled tube and rod
Measurement of head Spirit level and plank
 Measurement of flow Maps
further reading Altimeters
Back to the micro hydro web portal

Introduction

Several methods exist for measurement of the available head. Some

measurement methods are more suitable on low-head sites, but are too
tedious and inaccurate on high-heads. If possible, it is wise to take
several separate measurements of the head at each site. Always plan
for enough time to allow on-site comparison of survey results. It is best
not to leave the site before analyzing the results, as any possible
mistakes will be easier to check on site.

A further very important factor to be aware of is that the gross head is

not strictly a constant but varies with the river flow. As the river fills up,
the tailwater level often rises faster then the headwater level, thus
reducing the total head available. Although this head variation is much
less than the variation in flow, it can significantly affect the power
available, specially in low-head schemes where every half metre is
essential. To assess the available gross head accurately head water
and tailwater levels need to be measured for the full range of river
flows.

Dumpy levels and theodolite

The use of a dumpy level (or builder's level) is the conventional method
for measuring head and should be used wherever time and funds allow.
Such equipment should be used by experienced operators who are
capable of checking the calibration of the device.

Dumpy levels are used with staffs to measure head in a series of

stages. A dumpy level is a device which allows the operator to take
sight on a staff held by a colleague, knowing that the line of sight is
exactly horizontal. Stages are usually limited by the length of the staff
to a height change of no more than 3 m. A clear unobstructed view is
needed, so wooded sites can be frustrated with this method.

Dumpy levels only allow a horizontal sight but theodolite can also
measure vertical and horizontal angles, giving greater versatility and
allowing faster work.
 Sighting meters

Hand-held sighting meters measures angle of inclination of a slope

(they are often called inclinometers or Abney levels). They can be
accurate if used by an experienced person, but it is easy to make
mistakes and double checking is recommended. They are small and
compact, and sometimes include range finders which save the trouble
of measuring linear distance. The error will depend on the skill of the
user and will typically be between 2 and 10 %.

Water-filled tube and pressure gauge

It is probably the best of the simple methods available, but it does have
its pitfalls. The two sources or error which must be avoided are out of
calibration gauges and air bubbles in the hose. To avoid the first error,
you should recalibrates the gauge both before and after each major site
survey. To avoid the second, you should use a clear plastic tube
allowing you to see bubbles.

This method can be used on high-heads as well as low ones, but the
choice of pressure gauge depends on the head to be measured.

Water filled tube and rod

This method is recommended for low-head sites. It is cheap,

reasonably accurate and not prone to errors. In this case, if more
bubbles are trapped in one rising section of the tubes than in the other,
then the difference in vertical height of the sets of bubbles will cause an
equal difference in the head being measured, though this is usually
insignificant. Two or three separate attempts must be made to ensure
that your final results are consistent and reliable. In addition the results
can be cross-checked against measurements made by another
method, for instance by water filled hose and pressure gauge.

Spirit level and plank

This method is identical in principle to the water filled tube and rod
method. The difference is that a horizontal sighting is established not
by water levels but by a carpenter's spirit level placed on a reliably
straight plank of wood as dewscribed above. On gentle slopes the
method is very slow, but on steep slopes it is useful. Mark one end of
plank and turn it at each reading to cancel the errors. The error is
around 2%.

Maps

Large-scale maps are very useful for approximate head values, but are
not always available or totally reliable. For high-head sites (>100 m)
1:50,000 maps become useful and are almost always available.

Altimeters

These can be useful for high-head pre-feasibility studies. Surveying

altimeters in experienced hands will give errors of as little as 3% in 100
m. Atmospheric pressure variations need to be allowed for, however,
and this method cannot be generally recommended except for
approximate readings.

Micro hydropower basics

Measurement of flow
Other documents in this series: Within this document:
Introduction Measuring weirs
Civil work components Stage-discharge method
Turbines 'Salt gulp' method
Drive systems Bucket method
Electrical power Float method
Measurement of head Current meters
Measurement of flow
further reading
Back to the micro hydro web portal

The purpose of a hydrology study is to predict the variation in the flow

during the year. Since the flow varies from day to day, a one-off
measurement is of limited use. In absence of any hydrological analysis,
a long-term measuring system may be set up. Such a system is often
used to reinforce the hydrological approach and is also the most
reliable way of determining actual flow at a site. One-off measurements
are useful to give a spot check on hydrological predictions.

The flow measuring techniques described here are:

 the weir method,

 stage control method,
 the salt gulp method,
 the bucket method,
 the float method,
 current meters.

Measuring weirs

A flow measurement weir is a weir with a notch in it through which all

the water in the stream flows. The flow rate can be determined from a
single reading of the difference in height between the upstream water
level and the bottom of the notch (see figure 40). For reliable results,
the crest of the weir must be kept sharp and sediment must be
prevented from accumulating behind the weir. Sharp and durable crests
are normally formed from sheet metal, preferably brass or stainless
steel, as these do not corrode.

Weirs can be timber, concrete or metal and must always be oriented at

right angles to the stream flow. Siting of the weir should be at a point
where the stream is straight and free from eddies. Upstream, the
distance between the point of measurement and the crest of the weir
should be at least twice the maximum head to be measured. There
should be no obstructions to flow near the notch and the weir must be
perfectly sealed against leakage.
 Rectangular notch measuring weir
source: http://home.carolina.rr.com/unclejoe/weir.html

Temporary measuring weirs are used for short-term or dry-seasoned

measurements and are usually constructed from wood and staked into
the bank and stream bed. Sealing problems may be solved by
attaching a large sheet of plastic and laying it upstream of the weir held
down with gravel or rocks. It is necessary to estimate the range of flows
to be measured before designed the weir, to ensure that the chosen
size of notch will be correct.
The use of permanent weirs may be a useful approach for small
streams, but larger streams might better be measured by staging
(explained below).

Stage-discharge method

Once set up, this method provides an instant measurement of the flow
at any time. It depends on a fixed relationship between the water level
and the flow at a particular section of the stream. This section (the
contour section) is calibrated by taking readings of water levels and
flow (stage and discharge) for a few different waterlevels, covering the
range of flows of interest, so as to build up a stage-discharge curve.
During calibration the flow does not have to be measured at the contour
section itself. Readings can be taken either upstream or downstream
using, for instance, a temporary weir, as long as no water enters or
leaves the stream in between. The stage-discharge curve should be
updated each year. A calibrated staff is then fixed in the stream and the
water level indicated corresponds to a river flow rate which can be read
off the stage-discharge curve.

'Salt gulp' method

The `salt gulp' method of flow measurement is adapted from dilution

gauging methods with radioactive tracers used for rivers. It has proved
easy to accomplish, reasonably accurate (error <7 %), and reliable in a
wide range of stream types. It gives better results the more turbulent
the stream. Using this approach, a spot check of stream flow can be
taken in less than 10 minutes with very little equipment.

A bucket of heavily salted water is poured into the stream. The cloud of
salty water in the stream starts to spread out while travelling
downstream. At a certain point downstream it will have filled the width
of the stream. The cloud will have a leading part which is weak in salt, a
middle part which is strong in salt and a lagging part which is weak
again. The saltiness (salinity) of the water can be measured with an
electrical conductivity meter. If the stream is small, it will not dilute the
salt very much, so the electrical conductivity of the cloud (which is
greater the saltier the water) will be high. Therefore low flows are
indicated by high conductivity and vice versa. The flow rate is therefore
inversely proportional to the degree of conductivity of the cloud.

The above argument assumes that the cloud passes the probe in the
same time in each case. But the slower the flow, the longer the cloud
takes to pass the probe. Thus flow is also inversely proportional to the
cloud-passing time. Detailed mathematics will not be covered here
because the conductivity metre is usually supplied with detailed
instructions.

The equipment needed for `salt gulp' flow measurement is:

 a bucket,
 pure table salt,
 a thermometer (range 0 - 40° C),
 a conductivity meter (range 0-1000 mS),
 an electrical integrator (Optional).

Bucket method

The bucket method is a simple way of measuring flow in very small

streams. The entire flow is diverted into a bucket or barrel and the time
for the container to fill is recorded. The flow rate is obtained simply by
dividing the volume of the container by the filling time. Flows of up to 20
l/s can be measured using a 200-litre oil barrel.

Float method

The principle of all velocity-area methods is that flow Q equals the

mean velocity Vmeans times cross-sectional A:

Q=A × Vmean (m3/s)

One way of using this principle is for the cross-sectional profile of a

stream bed to be charted and an average cross-section established for
a known length of stream. A series of floats, perhaps convenient pieces
of wood, are then timed over a measured length of stream. Results are
averaged and a flow velocity is obtained. This velocity must then be
reduced by a correction factor which estimates the mean velocity as
opposed to the surface velocity. By multiplying averaged and corrected
flow velocity, the volume flow rate can be estimated.

Current meters

These consist of a shaft with a propeller or revolving cups connected to

the end. The propeller is free to rotate and the speed of rotation is
related to the stream velocity. A simple mechanical counter records the
number of revolutions of a propeller placed at a desired depth. By
averaging readings taken evenly throughout the cross section, an
average speed can be obtained which is more accurate than with the
float method.