Considering the vast, economical, untapped mini- and micro-hydro power potential of Ethiopia, this paper envisages the design of a micro hydel scheme encompassing two options, namely one with dam and another without dam. The first option (with dam) comprises two Kaplan turbines, each with a power output of 750kW and the second involves a Cross flow turbine with a power output of 35kW, and both have been designed. The second option (without dam) has been considered to utilize the perennial discharge in the Seka river stream and this design is based on the flow duration curves obtained after taking in to account the rain fall data for this region from three rain gauge stations situated at Shebe, Dedo and Seka. The entire ranges of accessories have been designed, and the cost estimation for cross flow turbine along with the local manufacturing strategy has been discussed.

Key Words: Micro Hydel, Design, cross flow turbine, Kaplan Turbine, Seka Water fall, Ethiopia

Introduction

Ethiopia’s per capita consumption of modern energy is 20kgoe (kgs of oil equivalent), which is about 2% of the world average. Per capita consumption of electricity was estimated to be 28kWh [1] equivalent to about 1% of the world average. The extent of electrification across the country is depicted in Table 1.Per capita GDP also stands at about US$100, thereby presenting a major financing challenge for the provision of modern energy to the rural majority. Only 13% of the population has access to electricity. But Ethiopia is blessed with enormous hydro energy resources, the gross theoretical potential (650Twh/year) being the second largest in Africa. Technically feasible potential is stated to be 260Twh/year, of which 10% represents the potential for small-scale hydro installation, refer Table 2.Hydro output in 1999 was about 1.6TWh, a minute fraction of the assessed potential. Ethiopia needs affordable energy to increase agricultural productivity and food distribution, deliver basic educational and medical services, and establish adequate water supply and sanitation facilities as well as to build and power new job creating industries.

In Ethiopia, one of the energy policy objectives is to ensure a reliable supply of energy at the right time and at affordable price, particularly to support the Country’s agricultural and industrial development strategies. Enhancing and expanding the development and utilization of hydropower is one of the priorities of the energy policy. Hydro Power, a renewable energy, features high in Ethiopia’s power sector. It’s continued development is perceived as essential given the extremely low level of current electricity generation, demand forecasts at hand, and the abundance of hydro power resources [1]. It can be part of an overall rural development plan [2-3]. The significant positive social and economic benefits include i) promotion of local industry ii) use of indigenous labor and materials iii) reduction of firewood consumption iv) raising income level of rural population v) saving in transmission line costs to remote rural areas vi) low operating cost and less skilled labor requirement vii) mitigation of population drift towards urban areas, and viii) development of potential for tourism. Local manufacturing and innovations have the potential to reduce the capital costs involved in small hydropower generation.

In this background, considering the vast, economical, untapped mini and micro hydel power potential, this paper envisages the design of a micro hydel scheme for Seka river stream encompassing two options one with dam and another without dam, namely

1. Two Kaplan turbines with the power out put each of 750Kw
2. The second option involving the cross flow turbine with power out put of 35.Kw to utilize perennial discharge in the Seka river stream, based on flow duration curves obtained taking into account the rainfall data available.

Description of project area (Seka River)

The Seka waterfall is found in Jimma zone, Oromia region 366 km south west of Addis Ababa. The Seka River is part of Gilgel Gibe drainage basin. The catchments area of the river up to Seka fall is 220km2, the elevation is between 1800 & 2600m. The selected dam site is about 1.5km west of Seka town. In the Seka

basin, the rainy season normally begins in mid May and continues through the early October. The heavy and mostly constant rainfall is from July to September. The annual lean flow period is from December to June. Mean annual rainfall of the area is 1820mm. Local centers of population are towns of Seka, Dedo, Shebe and other small towns. From the total population of Seka Chekorsa Woreda, 96.63% is living in rural areas with scattered individual homes through out.

Option I – With dam using Kaplan turbines

The rainfall data had been collected previously from the three rain gauge stations namely Seka, Dedo and Shebe which are found in Seka Chekorsa Woreda. The rain fall data for Seka is presented in Table 3. Based on the rainfall data, a suitable concrete gravity dam and its associated parts have been designed. The dam ensures availability of water at a rate of 6.8m3/s and at a head of 33.6m The installed capacity of the project is fixed at about 1.5MWe which is going to be produced by two Kaplan turbine units, each of 750kw capacity. The essential features of the dam designed are listed in the following Table 3.

Table 3: General information for proposed dam and its associated parts

Design of Kaplan turbine

Kaplan turbine is an axial flow turbine, suitable for low heads and high discharges. In this turbine, water first enters in a radial direction through guide vanes, which act like nozzles, and then flows through the runner in an axial direction. Based on the available data i.e. discharge and head Kaplan turbine and its associated auxiliaries are designed. The overall generation efficiency has been taken as 67% [4].
Power delivery =p= = 0.67*33.6*(2*3.4)*9.81*1000=1.5*106 watt=1.5MWe

I. Determination of size and hydraulic design of Kaplan turbine.

From the design charts [ 4 ], for a head of 33.6m the specific speed is obtained as 132 and corresponding to the head and discharge available, the speed of the turbine is estimated as 1000rpm. For this specific speed of 132, the speed coefficient turns out to be 1.42 and this directly determines the diameter of the rotor as 0.7m. The energy coefficient and flow coefficient are estimated to be 0.491 and 0.241 respectively while the unit speed and unit discharge are calculated to be 120.87 and 1.19 respectively. The different dimensions of the Kaplan turbine can be estimated based on the previously obtained rotor diameter [5], and are depicted in Fig.1 .The twist profile for the blade arrived at by applying free vortex design is highlighted in Table 5.and the blade chord, pitch and number are listed in Table 6. The profile of volute casing is also shown in Table 5. and is sketched in Fig.2. The particulars of the draught tube designed with an efficiency of 91.8% are included in Table 6.

II. Mechanical Design

The tangential force on each blade is estimated to be 7.96kN. There are three types of stresses at the root of the blade namely stress due to bending moment arising from hydraulic pressure and

weight of blade, stress due to centrifugal force and torsion stress due to transmitted torque. After considering all these the resultant stress at blade root is found to be 31MPa. To transmit the power output from the turbine, the shaft diameter is designed to be 90mm after making all the design checks. The details of the coupling for direct connection with generator and other auxiliaries for an integrated safe design are given in Table 6.

A custom built generator is proposed considering that there is no danger of winding failure at over speed, which is specified as 180% on a continuous basis. The overall arrangement is depicted in Fig.3.

Option II - Cross flow turbine

This option has been proposed without a dam, but to utilize the perennial discharge in the Seka river stream. The rainfall data that has been gathered from three rain gauge stations has been processed on an average basis. The flow duration curve thus obtained along with the one showing the power output per unit head are depicted in Figs. 4(a)& (b). Considering an average 86% flow duration and taking the discharge available at 0.55m3/s and using the available head of the waterfall at the project site as 10m, a Cross flow turbine with an indirect coupling (to the generator using a belt drive) has been proposed, based on the nomogram available [6] and also considering the suitability of cross flow turbine for the specific speed (m.kW) range of 55 -200 [4] The actual specific speed, ns of the cross flow turbine designed in this study turns out to be 159

A cross flow turbine is a 2 stage impulse turbine with a drum shaped runner having two parallel discs connected together near their rims by series of curved blades. Its runner shaft is horizontal to the ground in all cases. In operation, a rectangular nozzle directs the jet to the full length of the runner. The water strikes

the upstream blade and imparts most of (72-75%) its kinetic energy. It then passes through the runner and strikes the down stream blade on exit, imparting a smaller amount of energy (25-28 % of the total energy) before leaving the turbine. A feature such as vacuum enhancement by using a draught tube is necessarily expensive as it requires the use of air seals around the runner shaft. In this case, an air valve is installed to control the pressure in the casing to keep the runner from being submerged. Sophisticated machine can attain efficiency as high as 85% with simpler ones in the range 60-75% [7, 8].

Design of cross flow turbine

To exploit maximum performance, turbine hydraulic efficiency is estimated as =92% and this has been used to design an optimum configuration for the Cross flow turbine.
Power out put= Q*H*g* *ρ = 0.92*0.55*10*9.81*1000=50kW

The above power calculation is used, just to fix parameters of turbine (flow angle, fluid angle, absolute and relative velocity etc). But to get the actual power out put, efficiency is conservatively taken as 65% because the efficiency of locally made cross flow turbines usually ranges between (60 to75percent). Hence the realistic power output of the turbine would be at least 35kW. With this data, the speed of the turbine has been fixed as 400 rpm, from the Nomogram [6] available for Cross flow turbine.

I. Determination of the size of cross flow turbine

A Photograph of the water flow through the cross flow turbine runner is reproduced here in Fig. 5 from Osberger layout [9]. Treating this turbine as a 2-stage velocity compounded impulse turbine, the velocity triangles for flow

in the successive 2 blades passes in the runner are depicted in Fig.6 for optimum performance.

Larger diameter for the runner results in a smaller length of the runner. The advantage that can be obtained from a larger diameter is that,
the mechanical stress by water pressure would be reduced. But it will cost more to build mainlydue to heavier drive component required because of the slower speed and greater torque, when compared to a smaller machine of same power. However, using a larger diameter runner results in a machine having much higher durability and a much greater potential for increasing the efficiency beyond the apparent fixed limit 87%. The biggest concern with a runner, this larger in diameter, is the loss in head due to the bigger inlet. So an attempt has been made to reduce the diameter of the runner by increasing its length. The design methodology adopted and the different dimensions of the cross flow runner obtained are incorporated in Table 7 and presented in Table 8.

II. Design of nozzle

For maximum efficiency, the runner is designed for a single blade operation [9]. The nozzle curvature is selected as two times the runner diameter, to give a nice long gentle sweep on the blade Considering mechanical clearances and a long gentle sweep, arc of nozzle is taken as 73o as shown in Fig7. The angle () which the water hits the blade is calculated as 16o.as measured relative to the blade angle (300) which it self, is relative to the periphery of the runner. A simpler arrangement is to use a flow control valve as nozzle, to adjust the angle of attack. The nozzle parameters are listed in Table 8.

III. Over all dimension of Cross flow turbine
A schematic of the turbine arrangement in two different views is presented in Fig.8. The actual dimensions of the turbine and associated

components designed are highlighted in Fig.9. The runner can even be partitioned into two parts, so that 1/3, 2/3 or the full length may be used at a time depending on the load condition. The performance characteristics of the turbine are such that efficiencies of 83-84% have already been reported over a wide flow range of 25-100%, see Figs 10(a) and (b).

Manufacturing of cross flow turbine blade

Using the rolling die, it is possible to form the blade from flat stock and finally cut to the required length and later machining to the required accuracy (73.46o). There is another alternative, which is much simpler to arrive at or near the same result. The dimension and the angles in the design represent the near optimum dimension angles to satisfy mechanical advantages & unrestricted passage of the water. These dimensions are fairly fixed, therefore cannot be arbitrarily changed without some decrease in efficiency. As a result, there are definite dimensional relation ships between the various components of the runner. We can use this to a great advantage because a supply stock is all around us and readily available in the form of steel pipe. The blade can be fabricated by cutting a pipe length wise, in strips and machining to the required accuracy and radius. The runner diameters usually range between 305mm to 813mm in the length and usually most of them are made from 4-inch steel pipe, however there is no hard and fast rule to fabricate always from 4” steel pipe [9]. Some relevant dimensions are given in Table 10.

Cost estimation

Total cost including manufacturing for locally made cross flow turbine and all of its associated

parts with locally available material is estimated as 15,000 birr. The cost and the efficiency of cross flow turbine depends on its sophistication. For example if it is associated with draught tube it costs much. The same capacity Cross flow turbine, if it is purchased from foreign country would cost as much as 80,000birr.

CONCLUSION

One of the proposals outlined here can be feasibly implemented depending on the availability of financial resources with private sector parties / individuals/ power sector companies. The design of the systems and associated components has been carried out to minimize both hydraulic and mechanical losses. The salient feature of this project is that it can be implemented easily; especially the second option of the design i.e. the cross flow turbine. This turbine and its associated parts can be made with locally available material in local workshops. Since the design is affordable with minimum cost, both the private sector and the government have the chance to implement this project to improve the supply and lessen the shortage of power generation in rural areas of this region. The capacity of this project with second option is such that it can serve on an average 175 families. Other special feature of this cross flow turbine is that it is particularly relevant in the Ethiopian context and need to be widely publicized; manufacturing strategies firmed up and standardized to reduce the overall cost associated with its fabrication.

NOMENCLATURE

bo Runner length,m C Absolute velocity of the fluid, m/s
Cd Coefficient of discharge of nozzle Cx Component of C in flow direction, m/s
Cy Tangential component of C, m/s H Head of water available, m
D Diameter of rotor, m N Rotational speed, rpm
P Power output, W Q Discharge, m3/s
So Thickness of water jet, m U Blade velocity, m/s
g Acceleration due to gravity, m/s2 W Work output, J/kg
w Relative velocity of fluid, m/s H Hydraulic efficiency of turbine
 Density of fluid, kg/m3  fixed blade angle, 0
 Rotor blade angle, 0 A Cross sectional area of jet, m2
ns Specific speed (N*0.001* P^0.5)/ (H ^1.25)

REFERENCES
1. Mengistu Tefera, Power Sector Reforms and Promotion of Renewable Energy and Efficiency in
Ethiopia, African Energy Policy Research Network, No.309, March 2004.
2. J.A.Veltrop, “Hydro Power: Needs, Challenges and Opportunities” in Hydro Power ’92, E.Broch and
D.K.Lysnake(ed) Proc. Of the 2nd International Conference on Hydro Power, Lillehamer, Norway, 16-
18June, 1992, A.A.Balkema Publishers, USA.
3.Moniton, L., Lenir, M and Roux, J., Micro hydroelectric power stations, John Wiley & Sons,
Chichester, 1984.
4.Mechanical Design and Manufacturing of Hydraulic machinery, Mei Zu-Yan (Ed) , Hydraulic
Machinery Book Series, Avebury Technical 1991.
5. Fang Qing – jiang, Construction of Axial flow and Diagonal flowTurbines, in Mechanical Design and
Manufacturing of Hydraulic machinery, Mei Zu-Yan (Ed) , Hydraulic Machinery Book Series, Avebury
Technical 1991, pp 47-79.
6. Adam Harway, Andy Brown, Priyantha Hattiarachi and Allen Inversin, Micro Hydro Design Manual:
Aguide to small-scale waterpower schemes, ITDG Publishing, 1993.
7. Arter, A., “The split flow turbine –Cross flow development results in a new design” in Proc. Of 3rd
International Conference on small hydro power, Hangzhou Regional Center for small hydro power,
Hangzhou, China, 1984.
8. An Li-rang, Lu Li, Investigation of Hydraulic characteristics of Cross flow turbines, Hydro Power
Equipment, 1986, No.1 (in Chinese).
9. Joe Cole, cross flow Turbine Abstracts.htm, March 2004.

Table 1: Electrification at national and Regional levels (Ethiopia, 1999) [1]
Indicator National* Tigray Afar Amhara Oromia SNNP Somalia
1) Population in millions 61.7 3.6 1.2 15.1 21.1 12.1 3.7
2) Regional Population as % of
National total 100 5.8 1.9 258.7 35 19.6 6
3) Urban population as % of
Regional total**. 14.6 16.7 7.6 10.1 11.5 7.4 13.9
4) Population in Electrified
town (in million). 8 0.47 0.05 1.13 1.82 0.53 0.24
5) Percent population with
accesses to electricity 13 13 4.2 7.5 8.6 4.4 9.5

Table 2: Hydro potential in Ethiopia

Table 5: Variation of flow parameters from hub to tip and Volute casing profile

Table 6 - Kaplan Turbine auxiliaries

Discharge, Q=0.55m3/s Head, H=10m and Speed, N=400rpm
Drunner=
tjet=1/5 Drunner= mQ=Cd*4.43 = 0.55m3/s
U1=U4=

bo= =0.606m
C2=C3,

C1 =Cd
W1st stage = 6U2 & W 2nd stage = 2U2.
Power out put(max)= Q*H*g* *ρ = 0.55*10*9.81*0.92*1000 =50kW
Power out put(actual)=35.1kW Wtotal=
W1st stage = 68J/kg, W2nd stage=22.70J/kg
W2=U3Cy3-U4Cy4 =U3Cy3=U22= 22.7
U2=4.765m/s = , D2= 0.227mω1=7.77m/s and 1=31.43o 1= 16.800
=4.05m/s, ω2=Cx2=4.05-C2= = 6.25m/s
2=40.3o , 2=3=90
ω3=Cx3 =Cx4=4.05 m/s =C4 α4=90o
ω4= =7.9 m/s, 4= 30.88o

Table 7:Design parameters of the Cross flow turbine

Table 9: Cross flow turbine auxiliaries

Table 10 : Common pipe sizes which can be used to make cross flow turbine blades

Fig.5 flow through the runner Fig .6 velocity triangle

Fig7 disposition of nozzle with respect to Fig.9 Over all dimension of cross flow runner of cross flow turbine turbine