Environmentally friendly turbine design concepts

DOE/ID/13741
July 1999

H
Y A Summary of
Environmentally Friendly
D Turbine Design Concepts
R Mufeed Odeh

O
P
O
W
E
R
U.S. Department of Energy
Idaho Operations Office

DOE/ID/13741

A Summary of Environmentally Friendly
Turbine Design Concepts
(Concepts Developed by Alden Research Laboratory, Inc.,
Voith Hydro, Inc. and their Teams)

July 1999

Prepared by

Mufeed Odeh

United States Geological Survey - BRD
S.O. Conte Anadromous Fish Research Center
Turners Falls, Massachusetts 01376

For the

United States Department of Energy
Idaho Operations Office
Advanced Hydropower Turbine System Program
Under Contract DE-AI07-99ID13741

ABSTRACT

The Advanced Hydropower Turbine System Program (AHTS) was created in 1994 by the
U.S. Department of Energy, Electric Power Research Institute, and the Hydropower Research
Foundation. The Program’s main goal is to develop “environmentally friendly” hydropower
turbines. The Program’s first accomplishment was the development of conceptual designs of
new environmentally friendly turbines. In order to do so, two contractors were competitively
selected. The ARL/NREC team of engineers and biologists provided a conceptual design for a
new turbine runner*. The new runner has the potential to generate hydroelectricity at close to
90% efficiency. The Voith team produced new fish-friendly design criteria for Kaplan and
Francis turbines that can be incorporated in units during rehabilitation projects or in new
hydroelectric facilities**. These include the use of advanced plant operation, minimum gap
runners, placement of wicket gates behind stay vanes, among others. The Voith team will also
provide design criteria on aerating Francis turbines to increase dissolved oxygen content.
Detailed reviews of the available literature on fish mortality studies, causation of injuries to fish,
and available biological design criteria that would assist in the design of fish-friendly turbines
were performed. This review identified a need for more biological studies in order to develop
performance criteria to assist turbine manufacturers in designing a more fish-friendly turbine.

This paper is a summary of final reports submitted to the U.S. Department of Energy’s
Advanced Hydropower Turbine System Program by ARL/NREC and Voith teams:

* ARL/NREC team report: “Development of a more fish tolerant turbine runner –
Advanced hydropower turbine project”, prepared by T.C. Cook, G.E. Hecker, H.B. Faulkner, and
W. Jansen. DOE Contract No. DE-AC07-95ID13383. Hereafter referred to as Cook et al.
(1997).

** Voith team report: “Development of environmentally advanced hydropower turbine
system concepts”, prepared by G.F. Franke, D.R. Webb, R.K. Fisher, D.Mathur, P.N Hopping,
P.A. March, M.R. Headrick, I.T. Laczo, Y. Ventikos, and F. Sotiropoulios. DOE Contract No.
DE-AC07-96ID13382. Hereafter referred to as Franke et al. (1997).

iii

ACKNOWLEDGEMENTS

This summary report was based on conceptual design work completed by Alden Research
Laboratory, Inc., Voith Hydro, Inc. and their respective teams* for the U.S. Department of
Energy’s Advanced Hydropower Turbine System Program. Special thanks to the Alden and
Voith teams for their review and valuable comments. Also, special thanks to many members of
the DOE’s Advanced Hydropower Turbine System Program’s Technical Committee for their
active participation, support, and review of this report.

* ARL/NREC team: Alden Research Laboratory, Inc. and Northern Research and
Engineering Corporation.

* Voith team: Voith Hydro, Inc., Normandeau Associates, Tennessee Valley Authority,
Harza Engineering Company, and Georgia Institute of Technology.

v

CONTENTS

ABSTRACT ...................................................................................................................................iii

ACKNOWLEDGEMENTS ............................................................................................................ v

INTRODUCTION........................................................................................................................... 1

BIOLOGICAL CONSIDERATIONS AND DESIGN CRITERIA................................................. 4
Injury Mechanisms .............................................................................................................. 7
Mechanical: Abrasion, Grinding, and Strike.............................................................. 8
Pressure ....................................................................................................................... 9
Cavitation .................................................................................................................. 12
Turbulent Shear Stress .............................................................................................. 13
A NEW TURBINE DESIGN CONCEPT..................................................................................... 15
Design and Evaluation Criteria ......................................................................................... 15
Development of a New Turbine Runner ........................................................................... 17
Preliminary Design.................................................................................................... 17
Geometric Design...................................................................................................... 18
Flow Assessment Using CFD ................................................................................... 19
Satisfying the Design Criteria ................................................................................... 20
NEW DESIGN CONCEPTS FOR EXISTING AND NEW TURBINES .................................... 24
Environmental Issues ........................................................................................................ 24
Design Concepts................................................................................................................ 25
Kaplan Turbines ........................................................................................................ 25
Francis Turbines ........................................................................................................ 29
Francis Turbine with Increased Dissolved Oxygen................................................... 32
CONCLUSIONS ........................................................................................................................... 33

GLOSSARY OF TERMS ............................................................................................................. 34

REFERENCES.............................................................................................................................. 37

TABLES

1. Criteria for design and evaluation of the new ARL/NREC fish-friendly turbine runner. ........ 16
2. Preliminary design of the new ARL/NREC runner. (Source: Cook et al. 1997) .................... 18

vii

FIGURES
1. Schematic diagrams of Francis and Kaplan turbines ................................................................. 3
2. Schematic diagram showing locations within a turbine system where
fish injury mechanisms are believed to occur ......................................................................... 7
3. Percent of fish mortality as a result of exposure to rapid pressure reductions
in laboratory tests .................................................................................................................. 11
4. Schematic of the new ARL/NREC fish-friendly turbine ......................................................... 19
5. New ARL/NREC runner leading edge velocity distribution, shown here at mid-span
between the shroud and hub. 3-D CFD results..................................................................... 21
6. New ARL/NREC runner pressure contours, shown here at mid-span between
the shroud and hub. 3-D CFD results................................................................................... 22
7. New ARL/NREC runner pressure change rate contours, shown here at mid-span
between the shroud and hub. 3-D CFD results..................................................................... 23
8. Schematic diagrams of a typical Kaplan turbine runner. (a) runner with gaps and
(b) gapless runner (fish-friendly)........................................................................................... 26
9. Elimination of wicket gate overhang in Kaplan Turbines........................................................ 27
10. Locating wicket gates properly behind stay vanes maximizes efficiency and minimizes
probability of strike ............................................................................................................... 27
11. Schematic of a rough weld joint smoothed over for fish safety............................................. 28
12. Turbine efficiency when operated with: (a) constant runner rotational speed, and
(b) adjustable runner rotational speed ................................................................................... 28
13. Resulting shape of reduced number of Francis turbine runner blades ................................... 29
14. Elimination of wicket gate overhang by using spherical discharge ring ................................ 30
15. Increasing the distance between wicket gates and the runner ................................................ 30
16. Providing mild pressure changes in short penstocks.............................................................. 31
17. Providing mild pressure changes in long penstocks............................................................... 32

viii

A Summary of New Environmentally Friendly
Turbine Design Concepts
INTRODUCTION

The development of an environmentally friendly hydropower turbine stems from the need
to continue using a reliable source of renewable energy along with maintaining a healthy
environment and a sustainable ecosystem. The U.S. Department of Energy (DOE), Electric
Power Research Institute (EPRI), and the Hydropower Research Foundation envisioned the
Advanced Hydropower Turbine System Program (AHTS) in 1993. The program was created in
1994 with the objective of developing new hydropower turbine designs that would minimize fish
injury and mortality, are environmentally friendly (i.e., maintain adequate water quality), and
produce hydroelectricity efficiently. The Hydropower Research Foundation, a non-profit
organization formed by the National Hydropower Association, provided matching funds from
industry to DOE for the conceptual design phase.

DOE issued a Request for Proposals for environmentally friendly turbine design concepts
in October 1994. Submitters were encouraged to be innovative and to start from ground zero.
Responses were received in February 1995 from companies, universities, state agencies, research
labs, and individuals. Proposals were reviewed and rated according to their suitability to the
AHTS Program’s objectives, engineering soundness, and environmental application.

Two proposals were chosen for funding in October 1995. One came from a team of
engineers and biologists at Alden Research Laboratory, Inc. and Northern Research and
Engineering Corporation (ARL/NREC team, DOE contract No. DE-AC07-95ID13383). Another
proposal came from a team led by Voith Hydro, Inc., and included Tennessee Valley Authority,
Harza Engineering, Normandeau Associates, and Georgia Institute of Technology (Voith team,
DOE contract No. DE-AC07-96ID13382). The two teams took two different approaches to
achieving the AHTS Program objectives. ARL/NREC proposed to design a new turbine runner,
whereas Voith chose to improve existing runner designs.

The proposal from ARL/NREC team outlined a method to design a totally new turbine
runner. Their idea was to start with a single-bladed impeller that is a combined screw/centrifugal
pump, which is widely used in the food processing industry to transport fish and vegetables with
minimal damage. This impeller is also used to pump fish safely around diversion structures and
bypass systems at some locations in the U.S. This innovative approach would later yield a multi
bladed turbine runner design that may be used in new installations or to replace existing turbine
runners, where feasible.

1

The Voith team submitted a detailed proposal aimed at reviewing existing engineering
and biological design criteria and available turbine technology, and proposed to make design
concepts that would lead to enhanced fish survivability. The unique capabilities of the Voith
team enabled them to study important environmental issues related to hydropower turbines,
evaluate mortality studies of turbine passed fish, and provide three new design concepts that can
be used for rehabilitation of existing turbines and in new hydroelectric facilities.
Environmentally friendly design concepts for Kaplan and Francis turbines were submitted to
DOE by the Voith team. Schematic diagrams of Francis and Kaplan turbines are shown on Figure
1 to familiarize the reader with the various components of these two typical designs. A
supplemental report on the third concept dealing with an environmentally friendly aerating
Francis turbine will be submitted at a later date.

In order to formalize new concepts and improve on existing designs both ARL/NREC and
Voith teams started their projects with a literature search to identify probable causes of fish
injury and mortality, as well as other environmental issues impacted by turbines. Both teams
reached the conclusion that the available literature lacks adequate biological information or
design criteria on which to base new turbine concepts. These biological design criteria include
quantitative values (or thresholds) for damage-causing mechanisms that a turbine designer would
take into consideration to make their turbine fish-friendly. For example, the cavitation
coefficient must meet a certain value, velocity shear stress in certain passage areas should be
limited to defined numbers of force per unit area, etc. As a result, DOE recommended that an
independent study of the biological design criteria needed for the design of advanced turbines be
performed. fada et al. (1997) provided a review of available literature, suggested provisional
biological design criteria, and gave recommendations as to what biological design criteria
relative to fish injury mechanisms need to be developed further. This can be accomplished by
simulating turbine hydraulics in the laboratory coupled with using fish species of interest for
testing.

Although gaps in biological information were presented, both teams identified biological
and engineering performance goals, based on existing information, for their new concepts and
proceeded with their assignment. Both teams made use of the Computational Fluid Dynamics
(CFD) method of solution to refine their conceptual designs.

2

14

13
12
1 2 1 2

3
12 12

3 10
7 11
9 13 10
4
7
11 5

8 6 14 9 8 15
12
3

4 5

6
15

16
Francis turbine Kaplan turbine

1. Generator Rotor 6. Stay Ring Discharge Ring 11. Lower Guide Bearing 1. Generator Rotor 7. Turbine Cover 13. Thrust Bearing
2. Generator Stator 7. Supporting Cone 12. Thrust Bearing 2. Generator Stator 8. Stay Ring 14. Oil Supply Head
3. Turbine Shaft 8. Guide Vane 13. Upper Guide Bearing 3. Spider 9. Guide Vane 15. Concrete Semi-spiral Case
4. Runner 9. Operating Ring 14. Spiral Case 4. Turbine Shaft 10. Operating Ring 16. Draft Tube Cone
5. Turbine Head Cover 10. Guide Vane Servomotor 15. Draft Tube Cone 5. Runner 11. Guide Vane Servomotor
6. Discharge Ring 12. Guide Bearing

Figure 1. Schematic diagrams of Francis and Kaplan turbines. (Source: Franke et al. 1997)

BIOLOGICAL CONSIDERATIONS AND DESIGN CRITERIA

The issue of safe fish passage dominated the decision of whether a new turbine design
concept was environmentally friendly. Fish passage is an important issue to many hydroelectric
plants’ operators. However, improving water quality of turbine discharge, such as increasing low
dissolved oxygen content, and plant operating conditions were also considered priorities.
Available information on fish injury and mortality was reviewed by both teams to assess the
types and causes of injury and to develop the criteria to be used for evaluating new designs.
Biological design criteria were needed to assist in establishing allowable limits of hydraulic
parameters that may contribute to new design concepts, fish mortality, and plant operation.

Power plant owners, Department of Energy, Electric Power Research Institute (EPRI),
Federal Energy Regulatory Commission, Fish and Wildlife Service, National Marine Fisheries
Service, and others have conducted studies designed to identify the levels of fish injury and
mortality, and precise causes of mortality as a result of passing through hydropower turbines.
Although findings from these studies are useful in establishing qualitative guidelines, their use
for predicting the performance of new designs is somewhat limited. This is due to the methods
used and the different objectives the studies set out to accomplish. Both ARL/NREC and Voith
teams reached this same conclusion regarding available information from past mortality studies.

Field studies have been used to identify injury of fish passing through turbines. However,
these can be complex, costly, and may yield results that can be biased by the mark/recapture
techniques used. Furthermore, the complex flow field inside the turbine system makes it nearly
impossible, without as yet undeveloped instrumentation, to accurately attribute observed fish
behavior and damage to a specific injury mechanism.

In 1987 EPRI conducted a review to identify turbine designs and their operating
characteristics that may contribute to the mortality of turbine-passed fish (EPRI 1987, reported in
Cook et al. 1997). The review indicated that, generally, rapid pressure drops (including
cavitation), higher head differential across the turbine, and low turbine efficiency may increase
fish mortality. However, the impracticality of locating damaging zones and observing injury
mechanisms within a turbine made it difficult to explain exact causes of fish mortality (EPRI
1987).

The EPRI (1987) review indicated that in Francis turbines the runner entrance (where
wicket gates, blades, and the runner's peripheral speed interact), higher peripheral runner speeds,
and greater wicket gate openings were correlated to higher fish mortality. Fish mortality did not
4

change with operating head in Francis turbines (similar mortality at 40 ft and 410 ft). In Kaplan
turbines comparing the peripheral runner speed and plant operating head with fish mortality
yielded little correlation (mortality at 20 ft and 110 ft was the same). This was in contrast with
the general belief that hydraulic head is a major contributing factor to turbine mortality.
However, the clearance between the blade tips and the discharge ring in a Kaplan turbine, where
fish could be caught, was of concern (EPRI 1987).

Another recent review of fish entrainment and mortality studies by EPRI (EPRI 1992,
reported in Cook et al. 1997 and Franke et al. 1997) included data from many projects with
riverine as well as anadromous fish species throughout the U.S. The EPRI study findings
showed that estimated mortality averaged 20% for Francis turbines, 12% for Kaplan turbines,
and 9% for bulb turbines, and that a wide variety of species suffer similar mortality rates in a
given turbine type. Several studies indicated lower rates of mortality for naturally entrained fish
compared to fish that were artificially introduced into the turbine system (averaging 6% mortality
for both Francis and Kaplan turbines). The EPRI review included studies of juvenile clupeid
species (American shad and blueback herring) conducted after 1987, which confirmed the higher
mortality rates in the case of Francis turbines compared with Kaplans (mortality was 16% for
Francis turbines and about 4% for Kaplans). The difference in mortality percentages in the more
recent studies was attributed to two factors (according to the authors of the EPRI 1992 review);
artificially entrained fish (i.e., test fish) were larger in size than naturally entrained ones, and in
later studies (beyond 1987) researchers had better handling and evaluation techniques.

The Voith team conducted their own review of available mortality studies in order to
arrive at design criteria on which they would base their new design concepts for improvements to
features of existing and new turbine designs (see Chapter 4, Franke et al. 1997 for details). The
multidisciplinary team looked into fish-damage-causing mechanisms and evaluated existing
injury and mortality data. That led to new understandings, the need for further testing of new
perceptions, and some conclusions for the new design concepts. Following is a summary of
some of the Voith team’s findings and opinions. Also, references are made to the Voith team’s
review where appropriate throughout this report.

x Injury and mortality mechanisms are dependent on the zone which the fish takes to
pass through the turbine system. At Wanapum Dam in Washington, fish that
passed through a zone near the turbine hub experienced 5% higher mortality than
fish that passed through the zone in the middle of the runner.
x Fish encountering the zone surrounding the blade sustain injury due to blade strike,
blade end gaps, and local fluid flow effects. However, quantifying exact sources of
5

turbine passed fish injury and mortality is difficult due to the lack of controlled
experiments. Also, observed injuries may be the result of multiple damage
mechanisms. And, most studies to date used juvenile salmonids of limited size
range and did not provide data regarding turbine operating conditions or the
location of test fish injection zones. Planned injury and mortality tests should take
into consideration the zone in which the test fish are to be released, turbine
operating hydraulic conditions, and use various fish species.
x Injuries caused by pressure appear to be related to the difference between the
acclimation pressure upstream of the turbine and the exit pressure within the draft
tube zone.
x Turbines can be designed to operate cavitation free while increasing power
production. Proper turbine operation at cavitation-free conditions will reduce
maintenance costs and fish mortality that is believed to be related to cavitation.
x A threshold value of the shear stress indicator (the indicator here refers to the rate
of deformation OR rate of strain of the fluid, dv/dy) was identified as 450 ft/s/ft
(using Computational Fluid Dynamics and existing literature). Values above this
rate are believed to cause mortality.
x Turbine operating point has significant effect on fish survival. Tests at Wanapum
Dam showed that peak fish survival did not coincide with peak efficiency, but
occurred at a discharge where the predicted blade strike probabilities were low and
before cavitation became significant. Analyzed mortality data showed no
conclusive evidence supporting the belief that maximum fish survival occurs at
discharges within 1% of peak efficiency. Data did not preclude the possibility that
maximum survival can occur at greater than peak efficiency discharge.
x Fish survivability in fish-friendly turbines ought to be evaluated at before and after
conditions (benchmarked) using the same hydraulic and biological evaluation
techniques.

6

Injury Mechanisms
The survival of a turbine-passed fish is highly dependent on the path that the fish takes
through the turbine system (Franke et al. 1997; fada et al. 1997). Once a fish departs the forebay
and enters a turbine system it must contend with changes in physical geometry and flow
characteristics that are very rapid and believed to be injurious in certain zones along the path. An
illustration of the damaging zones within a turbine system is shown in Figure 2.

1 Increasing Pressure
2 Rapidly Decreasing
Pressure
3 Cavitation
4 Strike
5 Grinding
6 Shear
7 Turbulence

1
4 5
6
4
2 3

7

Figure 2. Schematic diagram showing locations within a turbine system where fish injury
mechanisms are believed to occur. (Modified from: fada et al. 1997)

The U.S. Army Corps of Engineers organized a turbine passage survival workshop in
1995 to identify causes of stress and injury to fish when passing through a hydropower turbine
system. Potential damage mechanisms were identified and loosely grouped into four categories;
mechanical, pressure, shear, and cavitation (USACE 1995). Mechanical causes include strike,
abrasion, and grinding. Pressure fluctuations, shear stress, turbulence, and cavitation are related
to flow characteristics.
After identifying the damage mechanisms, the next logical step would be to determine
biological design criteria that, when incorporated in new and rehabilitated turbines, would make

7

them more fish friendly. That necessitated a comprehensive literature review to identify existing
information that would lead to these criteria. Only laboratory experiments conducted to study
individual damage mechanisms under controlled conditions were reviewed (fada et al. 1997 and
fada 1998). The reviewers also briefly examined field techniques used to observe fish
movements in and out of turbine systems and to examine the resulting overall injury and
mortality. Among the most important findings of the review by fada et al. (1997) are:

x The least damaging turbine system design is one that directs the majority of the
migratory fish away from turbine intakes and towards their natural surface oriented
migration route;
x Shear stress and related turbulence are among the least understood of damage
mechanisms (see description below). Varying levels of shear stress and fish
response to them need to be studied in a laboratory setting;
x Further quantitative evaluation of indirect mortality, such as predation and disease,
of turbine passed fish is needed;
x Further understanding and data collection and analysis of fish trajectories inside
turbines are needed. Computational Fluid Dynamics is a valuable tool to
understand flow behavior inside turbines. CFD may be used to simulate fish as
passive objects in the flow field, given that data on fish behavior from field studies
are incorporated to calibrate the CFD model; and
x Further studies using hydroacoustic techniques and low-light underwater video are
needed to understand fish behavior and distribution as they approach turbine
intakes.

Following is a brief description of each of the damage mechanisms and some of the
related information presently available.

Mechanical: Abrasion, Grinding, and Strike
The rubbing action of a fish against a turbine system component or objects in the flow
field is referred to as abrasion, and can cause damage to the fish (USACE 1995). Abrasion
damage is dependent on flow discharge and velocity, number of turbine blades and spacing
between them, and the geometry of flow passages (USACE 1995). Data are not available to
identify the amount of or to distinguish injury due to abrasion.

Grinding injury can occur when a fish is drawn into small clearances (gaps of sizes close
to that of the fish) within the turbine system (USACE 1995). Gaps with high velocity zones that

8

may cause grinding injury are present between the turbine blade leading edge and the hub, the
blades and the throat ring, the wicket gates and stay vanes, and between the wicket gates and the
distributor ring (USACE 1995). Grinding injury can be documented by examining the fish’s
body for localized bruises, deep cuts, and even decapitation. However, precise prediction of
injury due to abrasion and grinding is not possible, and some of the fundamental symptoms of
grinding may also be caused by other fish injury mechanisms (fada et al. 1997).

A fish may be damaged when it collides with (strikes) a turbine system component. The
probability of a fish striking parts of the turbine system depends on several factors which include
the size of the fish, number of blades and their spacing, turbine speed, flow velocity and
discharge, among others. Several equations have been developed to calculate the probability of
strike in Francis and Kaplan type turbines (von Raben 1957 and Mont-n 1985, cited in fada
1997; USACE 1991, cited in Cook et al. 1997). Also, a new equation, based on the von Raben’s
model, was derived by the Voith team (Franke et al. 1997). These probability equations make the
assumption that a strike means serious injury or death, which may not always be true (Bell and
Kidder 1991, cited in fada et al. 1997). The probability of a fish dying from striking an object
within the turbine system is variable (Bell and Kidder 1991). A blade and a fish striking each
other (colliding) may cause scale and mucous loss, eye injury, and internal bleeding depending
on the velocities involved and the shape of the blade’s leading edge (Turnpenny et al. 1992).
Direct visual observations are not available to correlate mortality to strike (USACE 1995) and to
verify the strike probability models. Data on specific causes of mechanical injury to fish passing
through turbines are very limited and when compared to the field results, probability models
yield varying results.

Data relating fish mortality to entry into a water body showed that mortality varied
between 0% at 65 ft/sec and 100% at 145 ft/sec. Also, upon impact onto solid objects fish
mortality varied between 0% at 15 ft/sec and 100% at 95 ft/sec (USACE 1991, cited in Cook et
al. 1997). Data from EPRI (1987) indicated that mortality increases with runner peripheral
velocities; minimal mortality could be expected at runner peripheral velocities of 40 ft/sec or
lower in Francis turbines. The data in EPRI (1987) also showed that more strikes would occur at
higher tip speeds and that a peripheral runner velocity of 20 ft/sec or less may eliminate strike
mortality.
Pressure
Fish are subjected to rapid pressure changes throughout the turbine system. Damage due
to pressure is dependent on the amount and rate of change of pressure experienced by the fish as
well as the type of the fish. Physostomous fish, such as salmon and trout, have a pneumatic duct

9

that connects the swim bladder to the esophagus, which is used, along with the mouth, to rapidly
take in or vent gas (Lagler et al. 1962, cited in fada et al. 1997). Physoclistous fish, such as
perch and bass, do not have a pneumatic duct and must adjust their body’s gas content by
diffusion into the blood. Because this diffusion process may take hours, these fish are more
susceptible to damage due to rapid pressure decrease. Pressure changes felt by a fish are relative
to its acclimation pressure prior to entering the turbine system. These typically range from 15 ft
of water (21.2 psi or 146 kPa Absolute) at low-head plants to 170 ft of water (87.7 psi or 605 kPa
Absolute) at high-head plants (USACE 1995).

It is believed that fish are more sensitive to pressure decreases than pressure increases,
and that pressure-related mortality is due to injury to the swim bladder from decompression
(Tsvetkov 1972, cited in fada 1990 and in fada et al. 1997). Swim bladders in 10-cm perch
burst when pressure was reduced to 40% of acclimation values (Jones 1951, cited in fada 1990).
However, rainbow trout exposed to pressure increases of 35 to 185 psia (241 kPa to 1,275 kPa)
in less than one minute followed by near instantaneous depressurization exhibited normal
activity, and no mortality was attributed to the test conditions (Rowley 1955, cited in fada 1990).

Gradual pressure increases, up to 2064 kPa (300 psia), did not seem to cause significant
damage to sockeye salmon or six species of freshwater fishes (Harvey 1963 and Foye and Scott
1965, cited in fada et al. 1997). However, in both studies the rate at which pressure increased
was low (about 1 psi per second), which is unlike the rapid rate of pressure increase through a
turbine system. Whitefish fry and common carp exposed to a rapid increase in pressure from
atmospheric to 725 psia (4,997 kPa), followed by a 10-minute depressurization back to
atmospheric pressure experienced no mortality (Lampert 1976, cited in fada 1990). Alewives
pressurized to 50.7 psia (350 kPa), held for about 15 minutes, and depressurized back to 14.7
psia over a 2 minute period had difficulty maintaining horizontal disposition due to swim bladder
compression at first, but fully adjusted over the holding period (SWEC 1975). Overall, the test
fish mortality did not differ from the control fish mortality.

Swim bladder rupture and embolism are caused by suddenly and severely lowering the
pressure from the fish’s acclimated pressure (USACE 1991). Theoretical information on
mortality in salmonids, relative to pressure changes, indicated that when the minimum pressure is
30% of the acclimation pressure (i.e., Exposure Pressure/Acclimation Pressure ratio is 0.3), or
higher, no mortality is expected (USACE 1991). This general rule was supported by plotting
data from several fish mortality studies relating exposure of fish to minimum pressures below
their initial acclimation pressure; Figure 3. The few data showing mortality to the right of the 0.3

10

pressure ratio were from tests using physoclistous fish, such as bass and crappie, which are non-
anadromous.
100
90
80
70
Mortality (%)
60
50
40
30

20
10
0
0 0.2 0.4 0.6 0.8 1
Exposure Pressure/Acclimation Pressure

Figure 3. Percent of fish mortality as a result of exposure to rapid pressure reductions in
laboratory tests. (Source: fada et al. 1997)

In the 1995 Corps of Engineers Turbine Passage Survival Workshop, the following
conclusions were reached regarding the rate of pressure change in turbines:

x High head turbines are typically smaller units and have a high rate of pressure
change per unit time, and low head turbines are typically large units with a lower
rate of pressure change per unit time.
x Fish experience depressurization from as high as 88 psia (607 kPa Absolute) on the
upstream side of the runner to about 7 psia (48 kPa Absolute) on the discharge side
of the runner.
x Differential pressure per square inch of blade surface (Energy density) affects the
passage time through the runner, and the greatest rate of pressure change occurs in
0.1 to 0.2 seconds .
x In Kaplan turbines, generally the rate of pressure change across the blades is about
160 psi/sec, assuming a 75-ft head and a 0.2-second time period from high to low
pressures.

11

Cavitation
The presence of voids in the liquid has a damaging effect on marine and hydraulic turbine
propellers (Euler 1754, cited in Odeh 1988). Cavitation is the rapid vaporization and
condensation process of liquid. It normally occurs when the local pressure in the liquid drops to
or below vapor pressure, and with nuclei present in the liquid vapor cavities (bubbles) are
formed. These bubbles grow within the vapor pressure region and then become unstable and
collapse as they travel to areas with higher pressures. The collapse of bubbles can sometimes be
violent and cause noise, vibrations, pressure fluctuations, erosion damage to solid surfaces, and
loss of efficiency or flow capacity (Odeh 1988 and Tullis 1989). Cavitation damage can occur as
a result of high-pressure shock waves (that can reach up to 106 psi) or high-velocity microjets
shooting through the center of the bubble creating a local pit to the bubble’s adjacent solid
boundary (Tullis 1989).

Mortality in fingerling salmon was 50% when they were subjected to vapor pressure
followed by instantaneous return up to atmospheric pressure; the damage was attributed to the
high-pressure shock waves as vapor pockets in the test chamber collapsed (Muir 1959, cited in
fada 1990). Cavitation can also reduce turbine efficiency, which in some cases indicates an
increase in fish mortality (USACE 1995). Turnpenny et al. (1992) devised a spark–gap apparatus
to generate cavitation bubbles near the head and body of herring Clupea harengus and sole Sloea
solea. The apparatus in Turnpenny et al. (1992) was believed to have not generated the high
energy levels associated with cavitation bubble collapse that might be found in real turbines. The
freshly killed fish in their study showed no injury as a result of exposure to the bubble collapse
within their experimental apparatus.

A widely used non-dimensional cavitation parameter,V, can be defined as the ratio of
operating pressure conditions to the available gross hydraulic head, H, on the turbine runner.
This is expressed as V = (Hatm - Hs - Hv) / H, where Hatm the absolute atmospheric pressure (ft
absolute), Hs the turbine runner setting relative to tailwater level (ft), and Hv is the vapor pressure
(ft absolute). To avoid cavitation at a hydro plant, its operational Vmust be higher than its
critical value, Vcrwhere Vcr is when cavitation starts to be damaging to the turbine. The highest
fish survival at the Foster Project occurred when the turbine had an operational V almost one-half
to one-third the critical value (Bell 1981). Cavitation can also be minimized by properly
designing the runner geometry to minimize parameters governing cavitation, which include high
velocity/low pressure zones, surface irregularities, abrupt changes in flow direction, and location
or submergence (Cook et al. 1997).

12

Tests at Lower Granite Dam indicated no significant differences in fish injury and
mortality between the unit operating at best efficiency and under cavitation conditions
(Normandeau Associates et al. 1995). However, since cavitation and best efficiency conditions
occurred at the same flow, cavitation may have not been severe (Cook et al. 1997). Study results
at Lower Granite Dam showed 2 - 6% mortality, where 19% of the observed injuries were
attributed to pressure (Normandeau Associates et al. 1995).

Turbulent Shear Stress
Shear stresses in the flow field are a result of the change of velocity with respect to
distance, or the rate of deformation of the fluid. Shear stress is expressed as the force acting on
an area parallel to its direction (Gordon et al. 1992). The spatial change of velocity can be
attributed to both viscous forces and fluid flow properties, or fluid-induced forces due to its
acceleration and local turbulence (Franke et al. 1997). The highest values of shear stress are
found close to the interface between the flow and solid objects it speeds by, such as the blade
leading edges, vanes, and gates. The Voith team utilized CFD analysis to confirm the presence
of high shear stresses at these locations (Franke et al. 1997). Fish are believed to sustain injuries,
sometimes lethal, when they encounter zones of ‘damaging’ shear stress within the turbine
system; injuries are dependent on fish species, size, and the manner they enter the shear zone
(USACE 1995). Various researchers attempted to verify the limits of shear stress at which a fish
of certain size and species sustains injury using laboratory experiments; detailed reporting on
these can be found in fada et al. (1997). Some researchers introduced fish to a submerged water
jet at varying velocities, up to 120 ft/sec (Groves 1972; SWEC 1975; Turnpenny et al. 1992), and
others sent fish through a 14-inch pipe with varying size nozzles at the end of it, 4 and 6 inches
in two different tests (Johnson 1970a; 1970b; and 1972). Results from these experiments varied
according to the test fish size, species, and method of exposure. Tests with salmonids indicated
no mortality at submerged water jet velocities of 30 ft/sec (Groves 1972) or through the 14-inch
pipe at nozzle velocities of 67 ft/sec and less (Johnson 1970b). Tests with alewives and smelt
showed no signs of injury at jet velocities of 30 and 40 ft/sec (SWEC 1975).

Typical velocity changes across shear zones are on the order of 30 ft/sec, which is higher
than velocity gradients inside Kaplan turbines (USACE 1995). Shear stress zones are also
associated with vortices within the flow field. Most Kaplan turbines have gaps near wicket gates
and runner blades, and leakage from these and non-optimal turbine operation produce flow
separation which creates vortices with high shear stress zones (USACE 1995). Quantifying these
high shear stress zones can assist in designing and operating a turbine so that shear stress zones
are minimized and fish survivability is enhanced. For example, maximizing the blade tilt and

13

matching its leading edge angle to the incoming velocity vector minimizes vortices in a Kaplan
turbine, which reduces shear stress zones (USACE 1995). Vortices in the draft tube swirl also
have associated shear stresses and may be a primary source of shear stress damage to fish in
Francis turbines (USACE 1995).

14

A NEW TURBINE DESIGN CONCEPT
(Developed by ARL/NREC team)

A new “fish-friendly” turbine runner must have characteristics that are superior to
existing turbine designs that are known to adversely affect fish mortality. In order to achieve that
Alden Research Laboratory, Inc. and Northern Research and Engineering Corporation
(ARL/NREC team) re-evaluated existing fish mortality studies and gathered information on the
causes of injury to turbine passed fish; see the previous section “Biological Considerations and
Design Criteria”. Their evaluation of available information was used to identify criteria for
designing and evaluating the new runner and its potential to pass fish without injury.

The ARL/NREC team based their concept for the new turbine runner on a commercially
available pump that is used to pump fish and vegetables with minimum damage. The team used
a one-dimensional computer model for evaluating the power performance and a two-dimensional
computer model to develop the new runner geometry. Finally they performed three-dimensional
Computational Fluid Dynamics (CFD) analyses, a mathematical modeling technique, for three
design iterations of the new runner. The basic design assumptions were evaluated and operating
conditions were predicted for the new turbine. The detailed calculated flow conditions were
compared with the fish survival design criteria and geometric changes were made until the
criteria were satisfied.

Design and Evaluation Criteria
Available information on fish injury and mortality provides an aggregate view of what
happens to turbine-passed fish. Historically, mortality studies were conducted for reasons other
than the establishment of quantitative design criteria to be used for fish-friendly turbines.
However, several design criteria based on currently available biological information were chosen
by the team to provide guidance to design, improve, and evaluate their new runner. A list of the
criteria that were considered by the ARL/NREC team for design and evaluation are shown in
Table 1 below.

15

Table 1. Criteria for design and evaluation of the new ARL/NREC fish-friendly turbine runner.
Criteria Description Value Chosen Reasoning
Fish-friendly turbine runner A new runner design Project’s objective
Hydraulic design parameters Flow { 1,000 ft3/sec Representative of most hydroelectric
(28.3 m3/sec) turbines installed in the U.S., including
Kaplan and Francis Tube turbines.
Head { 75 ft to 100 ft
(23 – 30 m)
Turbine operating efficiency 85% minimum Efficiency for most turbines peaks at
90% to 93%. 85% was chosen so the
(3-D calculations included
new runner can be competitive with
scroll case and draft tube)
existing designs.
Peripheral runner speed Less than 40 ft/sec Reduces strike injury and minimizes
(preferably 20 ft/sec) shear stresses and vortices between
moving and stationary parts
Minimum pressure 10 psia (68.8 kPa) Downstream migrating fish are typically
found within the top 34 ft, i.e., at 30
psia (206 kPa), and mortality occurs
when pressure drop is more than 30% of
acclimation pressure.
Rate of change of pressure Less than 80 psi/sec (550.3 Assuming fish injury occurs at a
kPa/sec) pressure rate of 160 psia/sec in Kaplan
turbines.
Shear stress indicator (Rate Less than 15 ft/sec/in (180 Tests of alewives, a fragile fish, at ARL
of Strain, du/dy) ft/sec/ft OR 180 m/sec/m) with 15 ft/sec/in did not cause injury.
Number and total length of Minimize Fewer blades and shorter leading edges
leading blade edges reduce probability of strike
Clearance between runner 2 mm or less Small clearances reduce possibility of
and fixed turbine housing mechanical injury. 2 mm is less than the
components 3 mm gap chosen by the USACE for
testing in a Kaplan turbine.
Flow passage Sizes Maximize Large amounts of water between blades
should reduce abrasion injury by
keeping fish away from the blades
Flow control and plant Maximize distance between Kaplan turbines are more fish-friendly
configuration runner and wicket gates and than Francis turbines. A small distance
minimize travel time from between wicket gates and the runner in
(Not tested for during this
intake to runner Francis turbines may increase the
phase of the AHTS project)
chance for abrasion and grinding injury.

16

Development of a New Turbine Runner
A commercially available screw/centrifugal pump impeller was selected for initial
evaluation; the selected impeller was based on performance comparison of six different pump
models. The chosen single-bladed impeller had a long leading edge, a large flow passage, and
hardly any gaps, and has been proven safe for the transport of fish and vegetables with minimum
damage (Johnson et al. 1993; EPRI 1994, and other ARL studies cited in Cook et al. 1997). The
impeller is clog-free, gentle, and fairly electrically efficient (80% when used for solids handling
and 75% when used for fish). Also, the combined screw/centrifugal pump is currently used in
some fish diversion and bypass systems, such as the U. S. Bureau of Reclamation’s Red Bluff
Diversion Dam on the Sacramento River (Johnson et al. 1993 and Liston et al. 1997). Biological
data from this and other studies conducted at Alden Research Laboratory, Inc. showed this pump
to be effective and safe to transport live fish. An impeller model with the highest operating
efficiency was chosen for the initial evaluation; an important parameter in selecting the initial
geometry is for the new runner to be competitive with efficiencies of existing turbines.

The design process, using NREC’s computer design software, included three stages: (1) a
one-dimensional power performance model was used to obtain overall dimensions of the runner;
(2) geometric design, quasi-three-dimensional flow model was used to arrive at an optimal
runner shape; and (3) three-dimensional CFD analysis was used to provide an accurate
assessment of flow characteristics inside the turbine and finalize the runner design.

Preliminary Design
During the preliminary design stage, the pump impeller performance in the turbine mode
was analyzed. Peak electrical efficiency reached 79% at 1000 cfs, 96 ft of head, and the rotor
diameter was about 22.2 ft. A new design was needed because this efficiency was well below the
desired value of 85%, and the efficiency was reduced drastically when the unusually large rotor
diameter was made smaller. This meant a new runner design had to be developed, and two- and
three-bladed runners were compared with the large pump impeller operated as a turbine, see
Table 2. Here, although Case 3 was chosen for further analysis because of the lower number of
blades, Case 2 may be used if a smaller diameter runner is desired.

17

Table 2. Preliminary design of the new ARL/NREC runner. (Source: Cook et al. 1997)

Number Runner Runner Rotational Overall
Design of Diameter Length Speed Headb Efficiency
Case Description Blades (ft) (ft) (rpm) (ft) (%)

Scaled up
Impeller as a
1 Turbine 1 22.2 10.8 61.2 96 79

New Turbine
2 Design 3 16.2 12.3 73 84 90

New Turbine
3 Design 2 17.5 13.3 68 85 89d
a
Runner diameter at best efficiency.
b
Head between scroll case inlet and tailwater.
c
Efficiency includes estimates for draft tube and scroll case losses
d
For Case 3, the overall efficiency would be reduced by about 1% with a 30% reduction in runner
diameter (12.3 ft diameter and 9.4 ft length).

Geometric Design
Using quasi-three-dimensional flow modeling, a detailed geometric design analysis was
conducted to predict the velocity and pressure distributions along the blade, hub, and shroud
surfaces of a new runner. This assisted in avoiding designs that produce turbulent flows with
vortices (which cause high head losses) and low pressure zones (where cavitation damage may
occur). The final runner design for this conceptual phase is shown in Figure 4. Flow analyses
were performed for various iterations of this design; each had two blades and a shroud attached
to the blade edges. Geometric refinements were made until the overall design criteria (Table 1)
were satisfactorily met by the final design. The final geometric design refinements were made
with the assistance of the CFD modeling of the fluid flow inside the turbine.

18

Figure 4. Schematic of the new ARL/NREC fish-friendly turbine. (Courtesy Alden Research
Laboratory, Inc.)

Flow Assessment Using CFD
Once the overall dimensions were established and local refinements in the runner
geometry were made to avoid severe turbulence and head loss-producing flow characteristics,
VISIUNTM (a 3-D CFD program developed by NREC) was used to analyze the new runner, Case
3 design in Table 2 above. CFD programs solve the complicated Navier-Stokes equations
governing fluids in motion by numerically integrating flow properties, such as velocity and
pressure, over very small areas within a grid system throughout the flow field. Using an iterative
procedure and starting with known hydraulic boundary conditions (e.g., head, velocity, and
pressure at inlet and exit), the CFD program solves the equations over the entire grid system
19

using an iterative procedure. Finally, the results are displayed to allow evaluation and to make
alterations to enhance the design.

In their flow analyses, the ARL/NREC team optimized the runner design by continuously
making judgements regarding the turbine operating efficiency and fish survivability. To
simultaneously satisfy both criteria sometimes created a conflict, and a “balanced design” was
sought rather than favoring one aspect over the other. Design criteria outlined in Table 1 were
used as guidelines to make judgements and alterations to the design. Among the chief
contributing factors to optimizing the design were: (a) avoiding flow separation to minimize
losses and turbulence; (b) keeping pressures above the set minimum and the rate of change of
pressure was to be kept below the set value to prevent fish injury due to decompression; (c)
balancing factors that may affect the peripheral speed, such as head, blade shape, runner
diameter, and number and length of blades to minimize potential fish injury; and (d) minimizing
high shear stress zones. Because large flow passages means lower number of blades, this led to
having longer blades to extract the available energy from the flow. Also, to avoid excess loading
and rate of change of velocity and pressure on the blades, they were wrapped around the hub.

Improvements were made and a final design was obtained. Results of the analyses of
three different runner designs indicated that a two-bladed, 17.5 ft diameter and 13.3 ft long
runner will be 90% efficient (at 84 ft head and 70.1 rpm rotational speed) and is expected to
provide safe fish passage.

Satisfying the Design Criteria
The new runner had to meet the engineering and biological design criteria in order to be
considered a viable new concept for further development as a fish-friendly hydropower turbine.
Preliminary two-dimensional and advanced three-dimensional CFD analyses were performed to
determine overall performance and flow characteristics, respectively. Important findings that
resulted from this conceptual design phase of the new runner included:

x The final design was a vertical shaft runner with two blades, 17.5 ft diameter, and
13.3 ft long runner. The runner blades will be 4 inches thick with the trailing edge
rounded. (The one-bladed pump impeller became unusually large and inefficient
when evaluated as a turbine.)

x The turbine will have a mixed flow inlet with the inlet blade tip angle set tangent to
the relative flow, and the exit blade angle set differently at the hub surface compared
to at the shroud surface.

x Predicted performance efficiency was 90% at 84 ft head and 1000 cfs, exceeding the
set criterion of 85%. This means this runner should be competitive with traditional
turbines’ operational efficiencies.
x Peripheral runner speed was 64 ft/sec. This was higher than the maximum design
criterion set at 40 ft/sec. Peripheral speed is fixed by the head and runner diameter.
20

x The minimum flow passage was 36 in. (i.e., a sphere of 36-in. diameter can pass
through the smallest zone within the runner). Because of the large amounts of water in
a flow passage of this size, fish will be kept away from the blades and their injury
reduced.

x A shroud was fixed to the blades’ edges to rotate along with them, eliminating
clearances between the runner and fixed surfaces. This eliminates the possibility of
fish being caught in gaps that may cause grinding injury.

x The rate of change of velocity with respect to distance (shear indicator, du/dy) away
from a solid boundary remained below 2 ft/sec/in., well below the maximum of 15
ft/sec/in. allowed. This was throughout the flow field outside a boundary layer, 2–3
inches thick, at the blade surfaces. Inside this layer the shear stress is not believed to
increase damage to fish beyond any mechanical type injury resulting from contact
with the blades.

A close look at the velocity distribution near the blade shows the flow is proceeding in
the downstream direction and does not separate from the blade between the hub and the shroud,
Figure 5. Further geometry refinements can still be made to reduce velocity decrements at the
hub and shroud, that cause flow separation, which cause higher velocity gradients (i.e., higher
strain rate and as a result higher shear stresses).

Figure 5. New ARL/NREC runner leading edge velocity distribution, shown here at mid-span
between the shroud and hub. 3-D CFD results. (Source: Cook et al. 1997)

21

x Pressure distribution and change rate throughout the runner are important factors in
assessing the turbine for engineering and biological performance. Local pressure at
or below vapor pressure causes cavitation, which is undesirable in turbo-machinery
and is believed to cause damage to fish. Rapid pressure changes are also believed
to cause injury to fish. Pressure distribution and its decrease from runner inlet to
exit were found to be reasonably uniform, Figure 6. The minimum absolute
pressure, 8.6 psia, was found to be at the trailing edges of the blades. Although less
than the 10 psia design criterion the 8.6 psia minimum value found was associated
with only 0.0001% of the total volume of the water passing through the runner and
cavitation is not expected anywhere in the runner.

Figure 6. New ARL/NREC runner pressure contours, shown here at mid-span between the
shroud and hub. 3-D CFD results. (Source: Cook et al. 1997).

About 1% of the total volume of the runner flow passages was found to experience
pressure change rates greater than the maximum design criterion of 80 psi/sec. The pressure
change rate remained below the limit throughout the runner except in small areas on the suction
side of the leading edge of the blade, Figure 7.

22

Figure 7. New ARL/NREC runner pressure change rate contours, shown here at mid-span
between the shroud and hub. 3-D CFD results. (Source: Cook et al. 1997)

CFD analysis of the new runner design concept showed that it would perform well and is
not likely to injure fish passing through it. It can be used to replace existing turbine runners
where fish injury is of primary concern. However, a reduced power output may be the result
because the new runner may not have the same flow capacity or efficiency as the outgoing one.
An ideal situation for existing application is one where the plant had an open turbine bay that was
designed for the purpose of plant expansion. This new runner can be used in new applications,
or situations where minimum discharge requirements must be achieved by the power plant
operator. It can also be used downstream of return pipes of diversion systems (e.g., flow from
bypass fish screens).

There were no wicket gates or scroll case included in the analysis of the new runner. This
will be accomplished in the next step of the turbine design process. It is expected that this new
turbine design will have a draft tube expansion similar to that which was used during the new
runner development study.

Next, to be certain of this new runner’s abilities to generate power efficiently and pass
fish with minimum injury, a pilot test is planned. The ARL/NREC team will be designing a 42-
inch-diameter runner and test stand to be located at the Alden Research Laboratory, Inc. facilities
in Holden, Massachusetts. The planned pilot test of the new runner is a cost-shared project
between the ARL/NREC team and DOE’s AHTS Program.

23

NEW DESIGN CONCEPTS FOR EXISTING AND NEW TURBINES
(Developed by Voith Hydro, Inc. and their team)

The Voith team took a different approach to accomplishing their assignment. Their
objectives were broader and dealt with several environmental issues, including fish passage and
water quality (i.e., dissolved oxygen). They accomplished this by assembling a multidisciplinary
team to analyze existing data to guide them towards new design concepts.

Environmental Issues
The team studied various environmental issues associated with hydropower turbine
applications throughout the United States. These included fish passage, dissolved gasses in
turbine discharge, and minimum flows downstream of power plants. The team compiled a
database on 2,555 hydroelectric dams and examined another with over 6,000 turbine
manufacturers’ entries. Following is a brief summary of their findings (Franke et al. 1997).

x A large share of the hydroelectric power in the U.S. is generated by low and
medium head Francis turbines (about 23% of the total design discharge), mostly
found in the eastern and central states. Most of the flow passes through low head
axial turbines, such as Kaplan turbines, which are typical of installations throughout
the Western states (31% of total power and 57% of total discharge on the West
coast). In the southeast, 48% of the design discharge passes through low head axial
turbines and 22% in low head Francis turbines.

x Turbine sizes were found to be evenly distributed through the U.S. (about 27-29%
in each size category of 2, 2-4, and larger than 6 meter diameter). More of the
smaller turbines are found in the Upper Midwest.

x Upstream and downstream passage of salmon species is a concern at hydropower
sites on both the East and West coasts, anadromous American shad are important to
the East coast, and passage of freshwater fish is of significance to the Upper
Midwest and inland states.

x Dissolved oxygen is an important issue to the Southeast and Ohio valley states, and
is considered significant for the Great Plains and Northeast.

x Dissolved oxygen and minimum flow problems were found to occur at sites with
low plant factor, below 0.35. Francis turbines were found at projects with low plant
factor (about 80% of the capacity and 67% of the design flow). Plant factor is
defined as the yearly power produced (kWh) divided by the product of the plant
capacity multiplied by the operating hours in a year.

As a result of their investigation and after consultation with the AHTS Program, the
Voith team identified three specific objectives to be achieved. These consisted of providing new
24

design improvements that can be applied to a specific turbine site to make it more
environmentally friendly. The three were:

1. Provide new Kaplan turbine design features to increase fish survivability.
2. Provide new Francis turbine design features to increase fish survivability.
3. Provide new Francis turbine design features to increase dissolved oxygen in the
turbine water discharge. The Voith team will submit a supplemental report on this
third objective at a later date.

Design Concepts
Faced with the same dilemma of lacking biological design criteria as others (e.g.,
ARL/NREC team), the Voith team made use of the available fish mortality studies to aid them in
providing new design concepts. The Voith team also made extensive use of CFD analyses of all
aspects and components of Kaplan and Francis turbines. The team used their experience and
CFD tools to further their studies and to develop an understanding of turbine flow velocity and
pressure distributions and how these may lead to fish injury. Independent investigations of basic
turbine flow physics and issues dealing with low dissolved oxygen and ways to mitigate it were
also conducted.

The design concepts provided here can be used for both rehabilitating existing turbines as
well as new turbines in order to improve their compliance with the new age of environmental
awareness and safe fish passage. These new concepts would also benefit the hydropower plant in
more ways. The Voith team believes that incorporating the suggested design modifications
would result in a more efficient operation; more generated power, and reduced operation and
maintenance costs.

Kaplan Turbines
An environmentally friendly Kaplan turbine is one that generates power efficiently,
passes fish safely, and costs less to operate and maintain. Following is a list of design concepts
that was suggested by the Voith team in order to make existing and new turbine designs more
fish and environmentally friendly.

1. A turbine should be operated at high efficiency with no cavitation and reduced back-
roll; reducing the probability for fish injury and decreasing runner replacement costs,

2. Removing the gaps within a turbine system eliminates the added probability of fish
injury and enhances the turbine efficiency. Eliminating gaps at the wicket gates or
between the blades and the hub and discharge ring is believed to minimize fish injury
due to grinding. Side by side comparison of a typical Kaplan runner and a fish-
friendly Kaplan runner are shown on Figure 8. The gaps were removed by changing
the shape of the hub and discharge ring from the cylindrical-spherical-conical shape to
one that is all spherical, and recessing the blades into the discharge ring.
25

(a) (b)

Figure 8. Schematic diagrams of a typical Kaplan turbine runner. (a) runner with gaps and (b)
gapless runner (fish-friendly). (Source: Franke et al. 1997)

3. Eliminate wicket gate overhang. Eliminating the overhang of wicket gates by
changing the shape of the discharge ring from cylindrical to spherical results in
eliminating the gaps between the wicket gates and the discharge ring. Leakage
through gaps causes strong vortices with high shear stress that can potentially injure
fish. Reducing the wicket gate overhang will also increase the efficiency of the power
plant by reducing losses caused by the leakage at the wicket gate/discharge ring gap,
see Figure 9.

26

Conventional design
Fish-friendly Design

Figure 9. Elimination of wicket gate overhang in Kaplan Turbines. (Source: Franke et al. 1997)

4. Properly place wicket gates and stay vanes to minimize the potential for fish injury
due to strike and flow behavior induced stresses. Use a hydraulically smooth stay
vane and place it relative to the gates in such a way as to provide efficient operation
of the turbine and decrease fish injury. Flow visualization tools such as CFD can help
optimize the placement of these two important components of the turbine system to
minimize fluid disturbances and the potential mechanical strike for different gate
openings, Figure 10.

Stay Vane

Conventional design
Wicket Fish-friendly design
Gate

Figure 10. Locating wicket gates properly behind stay vanes maximizes efficiency and
minimizes probability of strike. (modified from: Franke et al. 1997)

5. Use environmentally friendly lubricating fluids and greases. Use a biodegradable
fluid in the hub and greaseless wicket gates bushings. This prevents pollutants from
being discharged into the water, enhancing water quality for the aquatic habitat
downstream of the power plant.

27

6. Polish the surfaces. Keep surfaces smooth on the turbine’s stay vanes, wicket gates
and draft tube cone. Welds on the various parts of a turbine system can be made
smoother to reduce abrasion injury to fish, Figure 11. In certain areas where the
velocity is low smoothing the surfaces and weld may not be a necessity and could be
costly.

Conventional design
Fish-friendly design

Figure 11. Schematic of a rough weld joint smoothed over for fish safety (modified from:
Franke et al. 1997).

7. Use an advanced control system to operate the hydropower plant electrical
components efficiently, which is also believed to be more fish friendly.

x Runner rotational speed and generator speed can be adjusted to maintain turbine
operation at the “fish friendly” point at any required discharge. Electrical
conversion equipment is available to make it permissible for the turbine to operate
with adjustable speeds yet maintain its peak hydraulic efficiency, Figure 12. It is
recommended that the addition of this type of equipment be accompanied with a
new runner upgrade at the same time.
Discharge Coefficient

Energy (or Head) Coefficient

(a) (b)

Figure 12. Turbine efficiency when operated with: (a) constant runner rotational speed, and (b)
adjustable runner rotational speed. (Source: Franke et al. 1997)

28

x Ensure cam optimization to provide maximum efficiency operation and minimize
flow stresses by maintaining turbine blade and wicket gates positions for
maximum efficiency, and perhaps minimal fish injury.
x Install sounding devices to give warning when the trash racks need cleaning.
Clean trash racks minimize flow disturbance and allow surface oriented fish to
enter the intake from its upper portion, therefore minimizing blade tip strike that
may occur when fish are forced to enter at the bottom of the intake.

8. Draft tube piers. Total removal of draft tube piers may not be a possibility due to
structural reasons. However, design the draft tube piers to be hydraulically smooth
(round nose) to reduce flow separation and possibility of strike.

Francis Turbines
An environmentally friendly Francis turbine is one that also operates at optimized
hydraulic conditions. It would have a low number of blades, high efficiency with no cavitation,
reduced back-roll, and would have well designed wicket gates’ interaction with the discharge
ring and stay vanes. These conditions are believed to reduce the probability of fish injury.
Following are the primary issues to making a Francis turbine an environmentally and fish
friendly one.

1. Low number of blades. This reduces the probability of strike and maximizes the size
of flow passages, which also minimizes the probability of abrasion damage to fish. A
lower number of blades results in having longer blades to maintain the same capacity,
power production, and minimize cavitation, see Figure 13.

- - - - Conventional Design
Fish-friendly Design

Figure 13. Resulting shape of reduced number of Francis turbine runner blades. (Source:
Franke et al. 1997)
2. Use a thicker blade edge. Using a thicker blade entrance edge would produce a
runner with fairly flat efficiency performance characteristics related to the head. This
means entrance edge will not cavitate at high heads and flow separation may not
occur. As a result injury due to flow stresses is minimized. Also, a thicker edge may

29

enhance the chance that fish will be carried around the edge rather than collide against
it, lowering the probability of strike.
3. Reduce wicket gate overhang, increase wicket gate to runner distance, and align
wicket gates with stay vanes. Eliminating the wicket gate overhang will increase the
turbine efficiency and reduce gaps that cause vortices created by leakage. Eliminating
the gaps is also expected to prevent fish injury due to grinding. See Figure 14.
Increasing the distance between the edge of the wicket gate and the runner can be
achieved by enlarging the pin circle diameter. This would also reduce the probability
of the fish grinding between the trailing edge of the wicket gate and the runner, see
Figure 15. Alignment of the wicket gates with the stay vanes (Figure 10), at least at
one gate opening, can be achieved in existing Francis turbines but will require
changes to other components in the turbine system.

Conventional design

Figure 14. Elimination of wicket gate overhang by using spherical discharge ring. (Source:
Franke et al. 1997)

Conventional design

Figure 15. Increasing the distance between wicket gates and the runner. (Source: Franke et al.
1997)
4. Use greaseless and self-lubricating wicket gate bushings, where the grease is an
integral part of the bushing.
30

5. Provide smooth surfaces on stay vanes, wicket gates and upper draft tube cone to
reduce potential abrasion and descaling damage to fish. Examples include restoring
damaged surfaces, use of special coatings, and reduced weld roughnesses, see Figure
11.

6. Operate the turbine with adjustable speeds. As in the case of Kaplan turbines,
operating a Francis turbine with adjustable rotational runner speeds may result in
reducing the probability of strike, shear stress zones, cavitation, and pressure
fluctuations. Figure 12 above shows how the combination of adjustable speeds along
with adjustable gates enables the turbine to operate at its optimum point at various
hydraulic conditions.

7. Use an advanced turbine control system. Adjustable speeds, variable speed generator,
clean trash racks, and optimized multiunit operation are important conditions to
making a turbine unit more environmentally and fish friendly.

8. Minimize pressure changes experienced by turbine-passed fish. For new power plant
designs it is recommended to provide fish with a passage route that minimizes sudden
changes in pressures. Figure 16 shows the difference between a conventional plant
design (Figure 16a) and one that is more fish friendly (Figure 16b). In Figure 16a fish
would be acclimated at high pressures prior to entering the penstock and are exposed
to much lower pressures on the downstream end of the turbine in a very short period
of time. However, in Figure 16b a safer route would be provided; fish would travel
from a zone of low acclimation pressure, through higher pressures for a short time,
and back to a low pressure region within the tailrace. Figure 17 also shows another
pressure related fish-friendly design. Here (Figure 17a), the fish are in the pipe for a
long period of time and would acclimate to the high pressure inside the pipe, only to
experience sudden reduction in pressure after passing through the turbine. This may
cause injury due to sudden decompression. In the fish-friendly design shown in
Figure 17b, the fish remain at low pressure in the pipe, travel through high pressure
regions for a short time, and go back to the low pressure in the tailrace.

Figure 16. Providing mild pressure changes in short penstocks. (Source: Franke et al. 1997)

31

Figure 17. Providing mild pressure changes in long penstocks. (Source: Franke et al. 1997)

Francis Turbine with Increased Dissolved Oxygen
A supplemental report will be submitted at a later date by Voith Hydro, Inc. The report
will include design concepts relating to aerating Francis turbines to increase dissolved oxygen in
the turbine discharge. Voith will also include a discussion on using advanced CFD modeling of
test conditions at Wanapum Dam.

32

CONCLUSIONS

DOE’s Advanced Hydropower Turbine System Program achieved its initial objective.
Two contractors provided new turbine system design concepts that can be utilized in the
development of new hydropower turbines as well as rehabilitating existing facilities. The
ARL/NREC new runner design concept predicts efficient power generation and fish friendliness.
If successful, Voith’s new concepts would also make it feasible to obtain power efficiently while
making new and existing traditional turbines more environmentally and fish friendly.

The next step is moving forward with prototype testing of the new design concepts to
demonstrate their effectiveness. DOE and the ARL/NREC team have initiated steps to design
and test a prototype turbine similar to the one described in this report. The new ARL/NEC
turbine will be hydraulically and biologically evaluated at the Alden Research laboratory, Inc.
facilities in Holden, Massachusetts. Voith Hydro, Inc. is in the process of testing some of their
new design concepts already implemented at power plants in the Pacific Northwest, such as at the
Wanapum Dam on the Columbia River, in Washington.

Once a proof of concept test has been conducted and results are favorable, DOE hopes to
issue a Request for Proposals for final engineering design and full-scale prototype testing at an
existing hydropower facility. If funding is available, this activity is planned to start during the
fiscal year 2000.

33

GLOSSARY OF TERMS
** (Modified from Cook et al. 1997 and Franke et al. 1997) **

Abrasion damage - Damage to fish resulting from rubbing contact with moving or stationary
objects in a turbine flow passage (USACE 1995).

Absolute pressure - Atmospheric pressure plus gauge pressure.

Anadromous fish - Fish that ascend rivers from the sea to breed.

Atmospheric pressure - The force per unit area of air; varies with elevation; at sea level
atmospheric pressure (1 atmosphere) equals 14.7 psia (101.3 kPa).

Best efficiency point - The operating point at which a turbine produces the highest ratio of
power output relative to the flow through the unit and the net head across the unit.

Bulb turbine - Axial flow turbine that has the generator, enclosed in a bulb-shaped housing,
within the water passage to the runner (USACE 1991).

Cavitation - Formation and implosion of water vapor bubbles that occur in water. The formation
occurs when local pressure is reduced to vapor pressure.

Cavitation Coefficient - The ratio of operating pressure conditions to the available gross
hydraulic head, H, on the turbine runner; V = (Hatm - Hs - Hv) / H, where Hatm the absolute
atmospheric pressure (ft absolute), Hs the turbine runner setting relative to tailwater level (ft),
and Hv is the vapor pressure (ft absolute).

Clupeid species - Family of fish that includes several anadromous species such as alewife,
blueback herring, and American shad.

Critical sigma - The runner sigma at which cavitation is initiated; the value of critical sigma is
lower than the value of plant sigma, with the difference representing the margin of safety against
cavitation.

Decompression - Lowering of pressure from the value at which fish are acclimated.

Discharge coefficient - QZd is dependent on the turbine discharge, Q, the rotational speed, Z, and
Q
the turbine diameter, D. Where: QZd .
ZD 3

Discharge ring - The stationary cylinder surrounding the blades.

Francis turbine - A reaction turbine, named for the inventor, in which water passes through the
34

runner first in a radial and then axial direction.

Grinding damage - Damage to fish drawn into narrow gaps between turbine components. Gap
damage to fish is considered to be the same as grinding damage.

Head coefficient - OR Energy Coefficient, EZd , is dependent on the net head, H, the acceleration
of gravity, g, the rotational speed, Z, and the turbine diameter, D.
gH
Where: EZd .
Z D

2

Kaplan turbine - A propeller turbine, named for the inventor, with runner blades which are
adjustable in angle when the unit is in operation (adapted from USACE 1991). Flow
downstream of the wicket gates is turned in an axial direction prior to reaching the runner.

Leading edge - The edge of the runner blade on the upstream side of the runner.

Leakage - Flow through gaps between rotating and stationary turbine components.

Mechanical injury - Damage to fish resulting from abrasion, grinding, and/or strike (as defined
elsewhere) on turbine rotating or stationary components.

Naturally-entrained - Fish that were present in a water body and have passed through a hydraulic
turbine on their own volition.

Navier-Stokes equations - The set of partial differential equations that govern the unsteady,
incompressible, viscous flow of fluids.

Peripheral runner speed - The speed at which the outside edge(s)of the runner travels (tip speed).

Plant sigma - The value of sigma for the site conditions and turbine setting.

Pressure damage - Damage to fish resulting from increases or decreases in the pressure from
values to which the fish have acclimated.

Pressure Surface - The side of the runner blades with higher pressure than the other side.

2S
Rotational Speed – Turbine runner rotational speed, Z. Where: Z RPM ˜ .
60

Runner sigma - The value of sigma at referenced to a particular place elevation of the runner;
usually calculated at the centerline or bottom of the runner in a vertical axis turbine, or at the
highest point of the runner in a horizontal axis turbine.

35

Shear damage - Damage to fish resulting from their passing through regions of rapid velocity
changes (shear zones).

Shear zone - Adjacent flow regions in a turbine having different velocities.

Shroud - Surface of revolution associated with the tips of the turbine blade.

Sigma - A non-dimensional parameter (V) representing pressure conditions at a turbine runner;
the ratio of the pressure above vapor pressure on the underside of the runner and the total
pressure across the runner.

Strike damage - Damage to fish resulting from their direct collision with rotating or stationary
turbine components.

Suction surface - The side of the runner blade with lower pressure than the other side.

Turbine efficiency - Percent of useful shaft power developed relative to hydraulic input power.

Turbine (runner) setting - The runner elevation relative to the tailwater level.

Vapor pressure - Pressure at which water vaporizes (boils); varies with water temperature.

Velocity Shear Stress - Stresses in the flow field resulting from the change of velocity with
respect to distance, or the rate of deformation of the fluid. Shear stress is expressed as the force
acting on an area parallel to its direction.

Vortex - Water flowing in a well defined circular motion with a lower pressure in the center;
vortices may be caused by flow disturbances at the runner's leading edge and by leakage through
gaps (USACE 1995); vortices in the turbine draft tube are caused by residual swirl leaving the
runner (USACE 1995).

Wicket gate - One of a series of gates in the flow passage leading to the runner which regulates
quantity and direction of water; the series of movable, flow-regulating gates impart rotation to
the flow (adapted from USACE 1991).

36

REFERENCES

Bell, M.C. 1981. Updated compendium on the success of passage of small fish through turbines.
Prepared for U.S. Army Corp. of Engineers, North Pacific Division, Portland, OR.

Bell, M.C., and J.C. Kidder. 1991. General discussion. Section I in Revised Compendium on
the Success of Passage of Small Fish Through Turbines. M.C. Bell (ed.). U.S. Army
Corps of Engineers, North Pacific Division, Portland, Oregon. 83 p.

fada, G.F. 1990. A review of studies relating to the effects of propeller-type turbine passage on
fish early life stages. North American Journal of Fisheries Management 10:418-426.

fada, G. F., C. C. Coutant, and R. R. Whitney. 1997. Development of biological criteria for the
design of advanced hydro power turbines. DOE/ID-10578. Prepared for Office of
Geothermal Technologies, U.S. DOE, Idaho Falls, ID.

fada, G.F. 1998. Better science supports fish-friendly turbine designs. Hydro Review, Vol.
XVII, No. 6, November 1998. pp 52-61.

Cook, T.C., G.E. Hecker, H.B. Faulkner, and W. Jansen. 1997. Development of a more fish
tolerant turbine runner – Advanced hydropower turbine project”, prepared for DOE
(Idaho Operations Office), contract No. DE-AC07-95ID13383.

Electric Power Research Institute (EPRI). 1987. Turbine-related fish mortality: review and
evaluation of studies. EPRI AP-5480, project 2694-4 final report, Palo Alto, CA.

Electric Power Research Institute (EPRI). 1992. Fish entrainment and turbine mortality review
and guidelines. EPRI TR-101231, project 2694-01 final report, Palo Alto, CA.

Electric Power Research Institute (EPRI). 1994. Update on fish protection technologies for
water intakes. EPRI TR-104122, project 2694-01 final report, Palo Alto, CA.

Euler, L. 1754. Histoire de l’Academie Royale des Sciences et Belles Lettres 10, Berline, pp.
227-295.

Franke, G.F., D.R. Webb, R.K. Fisher, D.Mathur, P.N Hopping, P.A. March, M.R. Headrick, I.T.
Laczo, Y. Ventikos, and F. Sotiropoulios. 1997. “Development of environmentally
advanced hydropower turbine system concepts”, Voith Hydro, Inc. Report No.: 2677-
0141. Prepared for the USDOE (Idaho) Contract No. DE-AC07-96ID13382.

Foye, R.E. and M. Scott. 1965. Effects of pressure on survival of six species of fish.
Transactions of the American Fisheries Society 94:88-91.

37

Gordon, N.D., T.A. McMahon, and B.L. Finlayson. 1992. Stream hydrology: An introduction
for ecologists. John Wiley Sons, New York, NY. 526 p.

Groves, A.G. 1972. Effects of Hydraulic Shearing Actions on Juvenile Salmon. Summary
Report. National Oceanic and Atmospheric Administration Northwest Fisheries Center,
National Marine Fisheries Service, Seattle, WA.

Harvey, H. H. 1963. Pressure in the early life history of sockeye salmon. Ph.D Thesis.
University of British Columbia, Vancouver, British Columbia.

Johnson, R.L. 1970a. Fingerling fish mortalities at 57.5 fps. Report No. 22. August 1970. U.S.
Army Corps of Engineers, North Pacific Division, Portland, OR 14 p.

Johnson, R.L. 1970b. Fingerling fish research effect of mortality of 67 fps velocity. Report No.
23. December 1970. U.S. Army Corps of Engineers, North Pacific Division, Portland,
OR 6 p.

Johnson, R.L. 1972. Fingerling fish research, high-velocity flow through four-inch nozzle.
Report No. 24. April 1972. U.S. Army Corps of Engineers, North Pacific Division,
Portland, OR 3 p.

Johnson, P. L., A. Watkins, C. N. Liston, and R. C. Kristof. 1993. Use of pumping as an
alternative to gravity diversion at Red Bluff Diversion Dam, California. Pages 141-146.
Proceedings of fish passage policy and technology symposium. Bioengineering Section,
Am. Fish. Soc., Portland, OR.

Jones, F. R. H. 1951. The swimbladder and the vertical movements of teleostean fish. I.
Physical factors. Jour. Expt. Biol. 28:553-566.

Lagler, K.F., J.E. Bardach, and R.R. Miller. 1962. Ichthyology. John Wiley Sons, Inc., New
York, NY. 545 p.

Lampert, W. 1976. Experiments on the resistance of fish to rapid increase in hydrostatic
pressure. Journal of Fish Biology 8:381-383.

Liston, C., C. McNabb, S. Borthwick, W. Frezell, and S. Atkinson. 1997. Status and recent
results of the Red Bluff research pumping plant program on the Sacramento River,
California. Fish Passage Workshop, Section 3. Milwaukee, Wisconsin.

Mont-n, E. 1985. Fish and turbines: fish injuries during passage through power station turbines.
Vattenfall, Stockholm, Sweden. 111 p.

Muir, J. F. 1959. Passage of young fish through turbines. Proc. Amer. Soc. Civil. Engineering
85(PO1):23-46.

38

Environmentally friendly turbine design concepts

More Related Content

Viewers also liked (15)

Similar to Environmentally friendly turbine design concepts (20)

Recently uploaded (20)

Environmentally friendly turbine design concepts