Joint UNDP/World Bank Energy Sector Management Assistance Program Activity Completion Report No. 062A/86 Country: ETHIOPIA Activity: AGRICULTURAL RESIDUE BRIQUETTING PILOT PROJECTS FOR SUBSTITUTE HOUSEHOLD AND. I,MUSTPAL FUELS VOLUME I - TECHNICAL REPORT DECEMBER 1986 Report of the joint INDPAVWdd Bank Energy Sector Management Asstance Program This document has a restricted distribution. Its contents may not be disclosed without authorization from the Govemment, the UNDP or the World Bank. ,NRgSY SECTOR GANAGEiMIVT ASSISTANCE PROORAM The Joint UNDP/World Bank Energy Sector Mapagement,Assistance ProgriiiF(ESMAP), started in April1983, assists countries in implerenting the main investment aid- policy recommendations of the Energy Sector Assessment Reports produced under another Joint UNDP/World Bank Program. 'ESMAP provides staff and consultant assistance in formulating' and justifying prority pre-investment and- investment projects and in providing management, institutional and policy support. The reports produced under this Program provide governments, donors and potential investors with the information needed to speed up project preparation and implementation. ESMAP activities can be classified broadly into three grout-j: Energy Assessment Status, Reportst these evaluate -achieve- ments in the year following issuance of the original assessment report and point out where urgent action is still needed; *- Project Form,1ation, and Justification: work designed to accelerate the preparation and implementation of investment" projects; and , . - Institutional and Policy Support:-this work also frequently leads to the- identification of technical, assistance packages. - ' v The Program aims to supplement, advance and strengthen the impact of bilateral and multilateral resources already available for technical assist~nce in the energy sector. Funding of the Program The Program is a major international effort and, while the core -finance has been provided by the UNDP and the World Bank, important financial contributions to the Program have also been made by a number of bilateral agencies. Countries which have now made or pledged initial contributions to the programs througlKthe UNDP Energy Account, or through other cost-sharing arrangements with UNDP, are the Netherlands, Sweden* Australia, Switzerland, Finland, United Kingdom,, Denmark, Norway, and New Zealand. Further Information For further information on thq Program or to obtain copieso of completed ESMAP reports, which ire listed at the end of this document, pleise contact: Division for Global and OR Energy Strategy and Interregional Projects Preinvestment Div. II United Nations Devel'opment, Energy Department Program Worl, Bank One United Nations Plaza 1818 R Street, N.W. New York, N.Y. 10017- Washington, D.C. 20433 ETHIOPIA AGRICULTURAL RESIDUE BRIQUEING PILOT PROECTS FOR SUBSTITUTE HOUSEHOLD AND INDUSTRIAL FUES IOLUNR I - TECHNICAL REPORT DECEBER 1986 CUamUCY AND FL EQUIVALES Currency 1 US$ 8 Birr 2.07 1 Birr US$0.48 Conversion Factors 1 toe = 42.74 thousand MJ a 42.74 GJ 1 MJ - 948 Btu - 239 kcal = 0.278 kWh Gross Heating Fuel Conversion Efficiency Major Fuels Value Household Industrial 0NJ/kAg) () CS) Woodfuels: Eucalyptus 17.8 25.1 70 Charcoal 29,0 300. 80 Commercial fuels: Kerosene 36,7/1 36.0 - Fuel oil 40.1/1 - 85 Electricity 36.6/kWh - 95 Briquettes: Coffee: parchment 19.3 28.8 80 husks 19.8 24.3 80 Cotton stalk 17.8 25.2 80 Wheat straw 18.0 27.0 80 Maize residues 16.7 28.5 80 ACOU ECAFOO Ethiopian ChipWood and Furniture Plant gCmC 8tbiopian Coffee Marketing Corporation EmEC Ethiopian National Energy Comission RBMAP UNUP World Bank Energy Sector Msnageent Assistance Program ET1ARSO Ethiopian Hardwood and Softwood Board Plant FANCDA Forestry and Wildlife Conservation Developi_t Authority mCTm Ministry of Coffee and Tea Development NoE Ministry of Mines and Energy MVW Ministry of State Farms 10 Applied Science Research Institute (The Netherlands) WDA Urban Dwellers Asociation VANU Wollega Agricultural Development Enterprise co centimeter db dry basis cmv Gross Besting Value ci Giga Joule ha hectare NWV Higher Heating Value kg kilogram kV kilowatt kWh kilowatt hour I liter LEEC Long Run Marginal Cost mcwb moisture content, vet basis NJ Mega Joule 0 & x Operations and Maintenance tone metric ton twe tonnes of wood equivalent toe tomnes of oil equivalent TABLE OF COUTEN Page I* INTRODUCTION AND SUMMARY .................................. 1 Background**o ......................................I Coffee Residues.....................*................... 3 Cotton Stalk Residues................................. 3 Wheat Straw 4......................................... ... 4 Maize 5eius.......................... Comparative Economics ....... .. . . . ... .. .................. 5 Marketing and Distribution....... . . ...... . ......o..o......o 9 Impleumentations..... ....... ........ .... ........... .............. 11 9 10 II. DENSIFIED AGRICULTURAL RESIDUES FOR UELWC OD SUBSTITUTIONoo. ....... o ............. . 12 Background ...oo...ooo.o oooooooo*oo #., 12 Fuelwood Supply and Demand.............................. 12 Need for a Fuelwood Substitute.......................... 13 Availability of Agricultural Residues................... 13 Suitability of Residues as a Fuelwood Substitute........ 15 Densification Technologyo........ ................ ........ooo 16 Pre-Processing..................................................... 16 collecti0n*****gon........ ..... ..... ................... 16 Storage..................................................... 18 claig..... ...... ... ... . ............ ......... ... c...... .... . ... 18 Feedingo..............................................oooo 19 Densification Equipment.................................. 19 Piston Press Briquetting . .. . . .. . ........ ..... ......... 19 Screw Press Briquetting ... ............. 21 Roll Briquetting.......... . .... . ...................... 21 PelLetizing. ~ ............................................. 23 Appropriate Densification System........................ 24 III. PILOT PROJECT FOR BRIQUETTING OF COFFEE RESIDUES .......... 27 Summary and Conclusions ....... 66... ......... ......0.. 27 Availability of Coffee Residues......................... 29 Sun-Dried (Unwashed) Coffee Residues.................... 29 Washed Coffee Residueso.................................. 30 Potential Pilot Projects for Coffee Residue Briquetting .so......o......oo...oo..o.o...oo...o....oo........ 34 Pilot Plant for Briquetting of Sun-Dried Coffee Husks and Washed Coffeeoo............................... 5 Site Selectiono.............................o.... . ..... 35 Plant Design and Equipmento..............................oooooo 37 Capital Costs........................................... 40 Annual Operating and Maintenance Cott s 41 Production Cossts. 41 Pilot Plant for Producing Dried Coffee Pulp............. 43 Capital Costs for Pulp Drying........................... 44 Annual O&M Costs for Pulp Drying y i ng.................... 44 Production Costs of Coffee Pulp Briquettes*............. 45 Pilot Plant for Briquetting of Coffee Parchment: Site S i . . e . . o . 46 Plant Design and Equipment.... .................... 47 Characteristics of Coffee Parchment Briquettes......., 50 Capital Costs 51 Annual Operating and Maintenance Costs***************** 52 Production Costs 53 IV. PILOT PROJECT FOR BRIQUETTING OF COTTON RESIDUESU......... 55 Summary and Conclusions....... 55 Availability of Cotton Residues iu....................... 56 Potential Pilot Projects for Cotton Residuestu.......es 58 Pilot Plant for Briquetting of Cotton Stalk Residues: Site Seletin.. 59 Collection and Storage of Residues********************&* 60 Plant Design and Equipmenu........... ............. .... 63 Characteristics of Cotton Stalk Briquettes ttes.....oo.. 67 Capital Costs 67 Annual Operating and Maintenance C o s t s 68 Production Costs 68 V. PILOT PROJECT FOR BRIQUETTING OF WHEAT RESIDU DS......o.... 70 Summary and Conclusions........ 70 Availability of Wheat Residues......o..............oo... 71 Potential Pilot Projects for Wheat Residues dues....... 73 Pilot Plant for Briquetting of Wheat Straw Residues: Site Selection..... ... . .. 74 Collection and Storage of Residues idu*..*o*.es.o.... 75 Plant Design and Equipment 79 Characteristics of Wheat Stalk Briquettes.......o...... 82 Capital Costs................................... 82 Annual Operating and Maintenance Costs.................. 82 Production Costs.oosts0......*o***oooo.* ooooo.o.o 84 VI. PILOT PROJECT FOR BRIQUETTING OF MAIZE RESIDJES........... 85 Summary and Conclusions..... 85 Availability of Maize Residues.................o........ 86 Potential Pilot Projects for Maize Residues dues......... 87 Pilot Plant for Briquetting Maize Residueseo*o........o. 88 Site aseto ......*........8 Collection and Storage of Residues.o.......o............ 89 Plant Design and Eq u i p m e nt..... 91 Characteristics of Maize Residues Briquettes............ 94 Capital Costs 94 Annual Operating and Maintenance Costs.................. 95 Production Coats................. .................... ... 96 VII. POTENTIAL DEMAND FOR BRIQUETTES...... ............... oqbo** 98 8usr ........................................................ 98 Household Sector.. ...... 98 Detailed Demand Information........................... 98 Demand-Side Issues.......................O 99 Comparative Economics ...... ......... .... ece. 102 Industrial Sector........ ............c........c.. 104 Potential Demand.................... 104 Comparative Economics .................... 108 Industrial Demand Issues... ...... .......... ..... 113 Externalities................ ........................... 115 Impact on Air Pollution in Urban Areas................ 115 Impact on Water and Lan4 Pollution.................... 115 Impact on Soil Nutritionflc oe.e...e..cc...c.eec:....e.. 125 Potential Fire Hazards................................ 116 VIII. MARKETING AND DISTRIBUTION STRATEGY ..... .... 117 Introduction ...............* ..... e. e.c..eoe.e.cec..o 117 Marketing Strategy...... eec..................... 117 Role of Intermediary OrganizationAse.c..........c...... 119 Ministry of Coffee and Tea Development .................. 119 Ministry of State Farms........ ......cce........c. 119 Ministry of Mines and Energy............................ 119 The Ministry of Industry ............0c.....0.. 0 .0.c.....cc 120 Other Organizationso.............o.....*.................. 120 Public Education Campaigno....................... ........ 121 IX. INSTITUTIONAL POLICY AND MANAGEMENT ISSUES. .............. 122 Ovwerview *o....eoo.o..ooo........oo...oo...o..e.oooo... 122 Plant Management e.e. oo................... eeeooc..o...oo... 122 Marketing Structuresooooo.......... e.eo.e.c.c.o o.o..c.. 123 Inter-Ministerial Coordinationo...... ..eo.....oo . .....eo 124 Overall Project Supervisiono...............o..o.....o..o 124 Project Monitoring and Evaluation....................... 124 X. IMPLEMENTATION ISSUES, BUDGET AND SCHEDULES ....e.cc..... 126 Introduction............................................. 126 Fundingo..... o.........o.......o............... .....oo.... 127 Plant ............................................ 127 Tender Documents/Equipment Specifications/ Procurement eooo ooo......oo.e...... ..................... 127 Site Selectionooo oooooo.................................. 128 Personnel Recruitment and Trainingo..................... 128 Organizational Structure................................ 129 Civil Workso....e..eo....e.e.......o..o.o ...eo.o....... . 129 Installation and Start-Upo.o.e.........o.....e.......... 129 Full-Scale Operationo....................o......... o.e.o. 130 Spare Parts .............. ccc...... cc..... cc. 130 Public Education ............. 130 Transportation ********oooo*oeo*... ....... .. .......... 130 Coffee..... ........................................... 131 esidu-pecific Consierations...................................... 131 Coffeet .............................................. 131 Cotton ................................................ 131 wheat ... * ~~~~~~~~~132 Naize ... , ~~~~~~~~~~132 Sensitivity Anlss..132 Project Budget and Disbursement Schedule................ 133 Monitoring and Supervision ...........................o. 133 Implementation Schedule...ee...... . ............ e.. ... 134 TOBLES 1.1: Briquettes as Household Fuelt Financial and Economic Cost Comparison..... .. .o. .e............. ........ 6 1.2: Briquettes as Industrial Fuels Financial and Sconomic Cost Comparison... . .. . ............. ......* 8 1.3: Foreign and Local Investment Costs for Pilot Briquetting Plants... 10 2.1: Availability of Residues: 1984/85 Harvest Season on...... 14 2.2: Estimated Range of Densification Costs for Extrusion Processeso c e s s es.................. .......... 25 3.1: Summary Data on Coffee Residues Briquetting Pilot Projects.... ....... 28 3.2: Summary of Coffee Residues Production................ 34 3.3: Coffee Residues-Chemical and Energy Analysis....sis.. 34 3.4: Coffee Processed at Dry-Pulping Stations for the Ethiopian Coffee Marketing Corporation................o 36 3.5: Dilla Coffee Husks Briquetting Pilot Plant: Construction and Equipment Cott s 40 3.6: Dilla Coffee Husks Briquetting Pilot Plant: Capital Costs (U8$) ................ . 41 3.7: Dilla Coffee Husks Briquetting Pilot Plant: 0&M Costs... 42 3.8: Financial and Economic Costs of Coffee Husk Briquettes..* 42 3.9: Dilla Coffee Pulp Drying Pilot Plant: Capit:l Costs (US$).*-.. .............. ...- . ...- . -- ...... 44 3.10: Dilla Coffee Pulp Drying Pilot Plant: OlM Costs ts.,..... 45 3.11: Financi.al and Economic Costs of Coffee Pulp Briquettes*** 46 3.12: Mercato Coffee Parchment Briquetting Pilot Plantt Construction and Equipment Costss ts.................... 50 3.13: Mercato Coffee Parchment Briquetting Pilot Plant: Capital Costs (US$) 51 3.14: New Site Coffee Parchment Briquetting Pilot Plant: Capital Costs (S)52 3.15: Mercato Coffee Parchment Briquetting Pilot Plants OlM Costs 52 3.16: New Site Coffee Parchment Briquetting Pilot Plant: OEM Css53 3.17s Financial and Economic Costs of Parchment Briquettes..... 54 4.1: Summary Data on Cotton Stalk Residues Briquetting Pilot Plant........ 55 4.2: Cotton Fiber and Stalk Residues Productiono.............. 57 4.3: Cotton Stalks Chemical and Energy Analysis............... 58 4.4: Cotton and Residues Production of the Middle Awash Enterprise (1343 / 8 4 ) 59 4.5: Collection Equipment and Costs for Cotton S8alk Residues ....... 61 4.6: Cotton Stalk Residues Briquetting Pilot Plant: Construction and Equipment Cott s 64 4.7: Cotton Stalk Residues Briquetting Pilot Plant: Capital Costs .............................o.. 67 4.8: Cotton Stalk Residues Briquetting Pilot Plant: O&M Costs and Briquette Transport Cost s 69 4.9: Financial and Economic Costs for Cotton Stalk Residues Briquettes................ .............oee.... 69 5.1: Summary Data on Wheat Straw Residues Briquetting Pilot Plant........................................... 70 5.2: Wheat Grain and Straw Residues Production................ 73 5.3 Wheat Straw - Chemical and Energy Analysis............... 74 5.4 Wheat and Wheat Stalk Residues Production of the Arussi Enterprises (1983/84).........................** .. 75 5.5 Collection Equipment and Costs for Wheat Straw Residues 77 5.6 Wheat Straw Residues Briquetting Pilot Plant: Construction and Equipment Cost s 81 5.7 Wheat Straw Residues Briquetting Pilot Plant: Capital Costs (US$) 83 Wheat Straw Residues Briquetting Pilot Plant: O&M and Briquette Transport Costs.- 83 5.9 Financial and Economic Costs for Wheat Straw J r i q u e t t e s ~~~~~~~~~~84 6.1 Summary Data on Maize Residues Briquetting Pilot Plant: Anger State Farm......................... 85 6.2 Maize and Residues Production.<........................... 86 6.3 Maize Residues - Chemical and Energy Analysis............ 88 6.4 Maize Data: Anger State Farm ........................... 89 6.5 Collection Equipment and Costs for Maize Residues........ 90 6.6 Maize Residues Briquetting Pilot Plant: Construction and Equipment Cott s 95 6.7 Maize Residues Briquetting Pilot Plant: Capital Cost (U8). 96 6.8 Maize Residues Briquetting Pilot Plant: Operating and Maintenance Cott s 97 6.9 Financial and Economic Costs for Maize Residues Briquettes.occ..... *******@* 00.0....... .........e... 97 7.1 Summary of Supply and Demand Prospects for Household Fuels in the Short Term (up to 1992)......... 99 7.2 Summary of Supply and Demand Prospects for Household Fuels in the Long Term (1993-2002)9...... 100 7.3 Comparative Financial an4d Economic Costs of Household Fuels and Briquettes in Addis Ababa*.......... 102 7.4 Actual and Forecast Industrial Energy Consumption#.....0. 104 7.5 (a) Level 1 Potential Industrial Briquette Demand: Plants Directly Surveyed......e............... . .. . . ..... 105 7.5 (b) Supplemental Level 1 Potential Industrial Briquette Demand: Plants Not Surveyed.................. 106 7.6 (a) Level 2 Potential Industrial Briquette Demand: Plants Directly Surveyed .............................. 107 7.6 (b) Supplemental Level 2 Industrial Briquette Demani: Plants Not Surveyed .................................... 108 7.7 Comparative Financial and Economic Costs of Conventional Industrial and Briquette Fuels............ 109 7.7 Comparative Financial and Economic Costs of Conventional Industrial and Briquette Fuels (continued)............ 111 7.8 Distances and Delivered Costs of Briquettes to End Use Siees 112 10.1 Project Budget... 139 10.2 Disbursement Schedule..... ...... .e... ...e... . ..... . 140 10.3 Implementation Schedule h.e d.u l......**o ....e..... .. 141 FIGURES 1.1 Financial Costs per Tonne of Briquettes from Pilot Plants Delivered to Addis A b a b a 4 1.2 Real Price Increases of Puelwood versus Cost of State Farm Briquettesi q u e tt.. .. .... .....ees........ 7 2.1 General Steps in a Densification Procecss........... s.... 17 2.2 Side View of the Piston Press Briquettor*e............r.. 20 2.3 Screw Press Briquettor...................e..... ......... 22 2.4 Roll Briquettt t or. 23 2.5 Pellet Mill..... .. 24 3.1 Mass Balance for Sun-Dried Coffee Processing************* 31 3.2 Mass Balance for Washed Coffee Processing...essing******. 32 3.3 Flow Diagram Coffee Residue Piston Briquetting Plant..... 38 3.4 Plot Plan for Coffee Residue Briquetting Plant*,*****.*** 39 3.5 Flow Diagram for Coffee Parchment Briquetting Plant...... 49 4.1 Schematic Diagram of Cotton Stalk Residue Storage ge...... 62 4.2 Flow Diagram - Cotton Stalks Piston Briquetting Plant l a nt.. . .... ..... . . ...... 65 4.3 Plot Plan for Cotton Stalks Briquetting Plant nt**.,, 66 5.1 Flow Diagram for Wheat Residue Briquetting Plant********* 78 5.2 Plot Plan for Wheat Straw and for Maize Stalks Briquetting Plant 80 6.1 Maize Residue Storage Flow Diagragm...................... 92 6.2 Maize Residue Briquetting Plant Flow Diagram*****ram***** 93 7.1 One-Hole Metal Cooking Stove............................. 101 10.1 Variations in Plant Production Capacity...ac. ity*........ 135 10.2 Variations in Briquetting Energy Requirements............ 136 10.3 Variations in the Discount Rate 137 10.4 Variations in the Briquetting Equipment Service Life..*.. 138 I. INTRODUCTION AND SUMIAtY Backgro-nd 1.1 In the energy sector, the most pervasive problem faced by Ethiopian households and industries is the increasing scarcity and cost of traditional fuels which account for over 90X of national energy consumption. By 1992, there will be a deficit of energy consumed beyond the sustainable supply equivaleet to more than 22 million tonnes of fuelwood. This deficit will severely erode the ecology, leading to massive deforestation and the resultant depletion of agricultural resources on which so much economic activity depends. Large-scale forestry programs are urgently needed to increase the supply of fuelwood and decrease the negative impact of deforestation on the agricultural environment. However, this is a long-term solution and, in the foreseeable future, the development and utilization of indigenous fuel substitutes must be vigorously pursued. 1.2 Agricultural residues from small farms are currently being used as a fuelwood substitute throughout Ethiopia. At least 3,300,X00 tonnes of surplus coffee, cotton, wheat and maize residues are produced annually, although not all are economically accessible. However, about 4 percent of cultivated land is occupied by State-owned Farms and processing facilities where surplus cotton, wheat, maize and coffee residues are available in large, accessible quantities and form a potential fuel source. This surplus amounts to a conservatively estimated potential of nearly 600,000 tonnes, which is equivalent to over 640,000 tonnes of fuelwood, or around 50Z of annual household fuel demand in Addis Ababa. If these residues are to be a viable fuelwood substitute, they must be collected and marketed to household and industrial consumers at a competitive price. 1.3 To investigate opportunities for residue utilization, a mission from the joint UNDP/World Bank Energy Sector Management Assistance Program (ESMAP) visited Ethiopia in November-December, 1984. 1I/ The mission's objectives were to: (a) verify the existence and availability of economicall- accessible quantities of agricultural residues on State Farms and at coffee processing facilities; and (b) identify potential means of collecting these residues. The mission concluded that handling, 1/ This report is based on the findings of a mission which visited Ethiopia from November 17 to December 7, 1984. The mission members were Messrs. Matthew Mendis (Mission Leader); Joe Leitmann (Researcher); Johannes van der Aar, Rienk Versluys, Hans Peter Buess, Klaas de Koning, Robert Stapel (Consultants). The report was authored by Messrs. Mendis and Leitmann. Secretarial support provided by Maria A. Cabezas. -2- storage and transport costs would have to be significantly reduced in order to economically utilize agricultural residues as a substitute fuel. Related work on the use of bagasse for energy and other purposes in Ethiopia has also been completed by ESMAP. 2/ 1.4 Most agricultural residues can be used as a fuelwood substitute in loose form. However, for reasons relating to transport, storage, marketing and stove or industrial boiler design, residue use can be enhanced and expanded through densification, e.g. in the form of briquettes that simulate firewood. For example, the cost of delivering baled wheat straw to the capital from a farm 300 km away would be a prohibitive US$144/tonne while transportation charget for a tonne of wheat straw briquettes would be only US$45. The ability to produce a marketable, densified fuel depends on the densification technology, chemical composition of the residues, their burning characteristics, the social/technical acceptability and, of course, their selling price. Regarding choice of technology, high-pressure briquetting results in a densified fuel with physical characteristics that are quite similar to existing domestic fuels. This and other available densification technologies are reviewed in Chapter 2. 1.5 Laboratory tests done specifically for this report have established that the physical and chemical composition of Ethiopian agricultural residue briquettes makes them a potentially attractive fuelwood and charcoal substitute. They appear to be an acceptable household cooking fuel as they are (1) easy to ignite, (2) suitable for local cooking practices, (3) have comparable or higher combustion efficiencies, and (4) exhibit favorable or comparable characteristics in relation -to existing cooking fuels. Industrially, briquettes can substitute for solid or liquid fuels in boilers with little or no equlipment modification in many cases. 1.6 In general, a project which seeks to produce agricultural residue briquettes entails residue collection, processing, densification, packaging, storage, transportation, marketing and distribution. A summary of the cost components of briquettes from various residues, on a per tonne basis, is presented in Figure 1.1. The detailed technical and cost parameters of pilot plants for supplying briqdetted coffee, cotton, wheat and maize residues are presented in Chapters 3 through 6, and are briefly reviewed below. Briquette demand considerations for potential household and industrial consumers, including their social and technical acceptability, are reviewed in Chapter 7. Issues and options relating to marketing and distribution of a price competitive fuelwood substitute are discussed in Chapter 8. Institutional issues and policies for managing briquetting facilities are considered in Chapter 9. Finally, 2/ Ethiopia: Bagasse Energy Survey, draft report of the Joint UNDP/IWorld Bank Energy Sector Management Assistance Program, December 1986. -3- implementation considerations, a sensitivity analysis, a budget and project schedulitng are outlined in Chapter 10. Coffee Residues 1.7 More than 220,000 tonnes of coffee residues are produced annually in Ethiopial on an energy basis, this is equivalent to 283,500 tonnes of fuelwood or 94,900 toe. To demonstrate the potential of coffee residues as a fuel substitute, two pilot projects have been identified: (a) a 5,000 tonne per year coffee husk briquetting plant near Dilla which can also be supplemented in the off-season with up to 1,000 tonnes of dry coffee pulp; and (b) a 2,500 tonne per year coffee parchment briquetting plant in Addis Ababa which can eventually be extended to a capacity of 5,ooo tonnes when more parchment becomes available. The delivered cost of these briquettes to a vendor or industrial user in Addis Ababa is estimated to range from $36/tonne for coffee parchment briquettes to $98/tonne for coffee pulp briquettes. Equivalently, on a useful energy 3/ basis for households, the cost ranges from $6.5O/GJ to $16.10/GJ. This compares to $18.60/GJ for woodfuels and $50.00/1J for charcoal. The total capital costs for a 5,000 tonne per year parchment briquetting plant is $370,380. One year's working capital for operating and maintenance costs and transportation costs amount to $132,160. The capital costs for a 5,000 tonne per year coffee husk plant are estimated at $516,850 while one year's working capital for X&M and transport are estimated at $290,650. Cotton Stalk Residues 1.8 More than 116,000 tonnes of cotton stalk residues are produced annually in Ethiopia; on an energy basis, this is equivalent to 109,600 tonnes of fuelwood or 45,600 toe. All of the residues are located on State Farms which currently burn them. A pilot project to collect, store and densify a small fraction of the cotton stalk residues is proposed at the Middle Awash Agricultural Development Enterprise, located some 300 km from Addis Ababa, to demonstrate their potential as a fuel substitute. A total of $1,149,360 is required to cover the capital costs while $399,810 is required for one year's operation, maintenance and transport costs in 3/ Useful energy is defined to be the gross energy of the fuel (GHV) times the energy conversion efficiency of the particular combustioz, device (i.e., cook stove, boiler, etc.) -4- order to finance a 5,000 tonne per year cotton stalk briquetting plant. These briquettes can be produced for $82 per tonne; with an additional $32/tonne to cover transportation, they have a delivered price in Addis Ababa of $114/tonne. On a useful energy basis for households, the cost of cotton stalk briquettes is $25.40/GJ. fgue 1.1 Rnancial cots Pew Tnne d ples From Pilot Plt Deleed to Adds Ababo Coo 140 - 130 120- 110 so. 80 40 20 - 1 10 10 Hudk Pup cotton Wea Molas Parrw Poer4re PAC * bV HO Cdbe k Oi SD. M- OomCbM C4ffe Pup a M.*. NO coton -Colton S9% 9t moo se VAt - VAWWgam. Okstl Moae -MoWReOsMMArot. Pe-Cdb. Oicten AdfO NAWSit. PM - Cdt.Paont. AdSAbtMeCdOOSt Wheat Straw Residues 1.9 More than 79,800 tonnes of wheat straw residues are available annually on Ethiopian State Farms, assuming that 501 are left on the ground for soil replenishment; on an energy basis, this is equivalent to 80,700 tonnes of fuelwood or 33,600 toe. A pilot project to produce wheat straw briquettes is proposed for Dixis, a State Farm located about 220 km from Addis Ababa. To finance a 5,000 tonne capacity per year briquetting plant, $1,111,400 is required in capital plus $415,960 for one year's working capital to finance the operating, maintenance, and transportation costs. Wheat straw briquettes can be produced for $81/tonne ex-factory; with an additional $34/Jonne for transport to Addis# the delivered cost to the capital is $115/tonne. On a useful energy basis for households, this amounts to $23.30/GJ. Maize Residues 1.10 On State Farms, 165,600 tonnes of maize residue (stalks, husks and cobs) are produced; on an energy basis, this is equivalent to 167,800 tonnes of fuelwood or 69,900 toe annually. A pilot briquetting plant for this residue is proposed at the Anger State Farm, located approximately 350 km from Addis Ababa. To construct and install a maize residue briquetting plant of 5,000 tonnes capacity will require $1,464,190 in capital plus $487,690 in working capital for one year's operating, maintenance and transportation costs. The financial cost of maize residue briquettes ex-factory would be $108/tonne; their delivered cost to Addis would be $140/tonne. On a useful energy basis for households, this results in a delivered cost of $26.40/GJ. Comparative Economics 1.11 In financial terms, all of the proposed briquetting plants produce a competitive household fuel which has less than or about half the useful energy cost when compared with charcoal. When fuelwood is taken as the comparative fuel, then only the coffee parchment and husk briquettes have a lower useful energy cost. The sensitivity analysis in Chapter 10 indicates that if utilization of equipment capacity is raised to 75X from the currently assumed level of 50X, 4/ then cotton stalk and wheat straw briquettes would also be competitive with fuelwood. In Table 1.1, a financial and economic cost comparison of briquettes (produced at 502 of plant capacity) and conventional household fuels i% presented. 1.12 Given 1985 market prices in Addis Ababa for fuelwood, the State Farm residues of cotton, wheat and maize presently have a higher household useful energy cost than fuelwood. However, fuelwood prices have been increasing at the real rate of 9.27 per year over the past 13 years, with the bulk of the growth occurring in the past decade. Given the projections of future supply and demand for fuelwood, this rate of 4/ This is a realistic assumption for pilot projects in Ethiopia, given residue collection constraints and unfamiliarity with the technology. - 6 - real price increases is expected to continue if not accelerate. Figure 1.2 illustrates the impact of relative real price increases of fuelwood on the competitive position of the State Farm residue briquettes. Asauming a relative real price increase of 71 per annum for fuelwood results in wheat straw briquettes becoming competitive in a little over two years, cotten stalk briquettes in four and a half years and maize briquettes in over five and a half years. If the relative inflation rate is 91 p.a., then wheat briquettes become competitive in under two years with cotton briquettes becoming competitive in a little over three years and maize briquettes be':oming competitive in about four and a half years. Table 1.1: BRIQETTIES AS HOUSEHOLD FUEL: FINANCIAL AND ECONOMIC COST COWARISON Fuel Delivered Cost Cost of Useful Energy (USS/tonne) (USSd6J) Financial Costs I. Comparative Fuel: Fuelwood a/ 83.25 18.60 Charcoal a/ 434.97 50.00 II. Briquette;: Coffee Parchment _/ 36.02 6.46 Coffee Husk 71.70 14.93 Cotton Stalk 113.94 25.40 Wheat Straw 114.67 23.30 Maize Residues 140.34 26.40 Economic Costs III. Comparative Fuel: Kerosene (USS/I) 30.19 IV. Briquettes: Coffee Parchment _/ 36.75 6.62 Coffee Husk 61.99 13.85 Cotton Stalk 113.18 25.23 Wheat Straw 105.86 21.77 Maize Residues 144.66 27.22 a/ 1985 market prices In Addis Ababa, b/ New site. Conclusions concerning the competitiveness of briquettes are based on the "Cost of Useful Energy" which takes the delivered cost (presented in $/GJ in Table 1.1) and modifies it by the burning efficiencies of the various fuels in cook stoves to come up with the useful cost of energy for a household. In economic terms, all types of briquettes have a much lower cost of useful energy than kerosene, which is the comparative fuel with guaranteed availability. Figure 1. 2 REAL PRICE INCREASES OF RJELWOOD VERSUS COST OF STATE FARM BRIQUETrES 40 - 9% 38 7% 36- 34- ~32- 5% 30 i-t X . . - 26…-OrO ~24 22 V-A 20- 18 1985 1986 1987 1988 1989 1990 1991 1992 1993 1994 1995 YEAR Wodd Bank-30384:3 - 8 - 1.13 For irdustrial use, agricultural residues briquettes can be produced and delivered to users in Addis Ababa at a lower cost than most industrial fuels. Compared to fuel oil, maize residues briquettes are not competitive. For all other industrial fuels, all briquettes are less expensive substitutes. Economically, cotton and maize residues briquettes are more expensive than fuel oil but less than the other fuels. A financial and economic cost comparison of briquettes and industrial fuels is presented in Table 1.2. Table 1,2: BRIQUETTES AS INDUSTRIAL FUEL: FINANCIAL AND ECONOMIC COST COMPARISON Fuel Delivered Cost Cost of Useful Energy (USS/tonne) (USSi)J Financial Costs :. Industrial: Fuel Oil (USS/1) 0.29 8.51 Fuelwood 83.25 9.86 Charcoal 362.00 14.69 Electricity (US kWh) 0.06 17.54 II. Briquettes Coffee Parchment a/ 36.02 2.33 Coffee Husk 71.70 4.54 Cotton Stalk 113.94 8.00 Wheat Straw 114.67 7.94 Maize Residues 140.34 9.41 Economics Costs III. Industrial: Fuel Oil (USS/I) 0.27 7.92 Fuelwood 83.25 9.86 Charcoal 434.97 18.75 Electricity USS/kWh) 0.06 17.54 IV. Briquettes: Coffee Parchment a/ 36.75 2.38 Coffee Husk 61.99 4.92 Cotton Stalk 113.18 7.95 Wheat Straw 105.86 7.34 Maize Residues 144.66 9.70 a/ New site. In this table, the cost of useful energy reflects the conversion efficiency of the various fuels when used in boilers to give the true cost of energy to industrial consumers. If briquettes are delivered to - 9 - industrial facilities which are closer to the pilot plants than Addis then they become even more attractive in financial and economic terms as transportation costs are lower. The proposed pilot plants are a worthwhile investment because of their ability to provide an affordable fuel substitute to household or industrial users. Marketing and Distribution 1.14 In the area of marketing and distribution, price-competitive briquettes must be made available to household and industrial consumers by methods that are compatible with existing purchasing patterns in the open market. With the selection of Addis Ababa as the target household market, briquettes can be sold through private merchants who are ideally suited to marketing fuel in small, accessible and saleable quantities to families and individuals. For industries, the Ministry of Industry can arrange supply contracts so that plants can purchase a guaranteed output which is not consumed by households. Beyond this, the most important role for governmental involvement is in wholesale and test marketing, and in organizing a promotional public education campaign. By following this marketing strategy, briquettes will be economically priced based on the supply and demand of briquettes and other competing fuels. Implementation 1.15 In addition to project financing, a variety of institutional and implementation issues must be taken into consideration. These are addressed in Chapters 9 and 10. Once a well-managed set of pilot projects has been implemented, a desperately needed substitute household and industrial fuel can be provided to the Ethiopian econmy and the case for further expansion to cover the entire range and availability of agricultural residues can be evaluated. Budget 1.16 To implement all commercial-scale pilot briquetting plants, a total investment of US$6,521,550 is required to cover capital and one year's worth of operating, maintenance, and transportation costs. Of this amount, $5,014,710 are foreign exchange expenditures and $1,506,840 are local costs. The total capital investment is $4,719,810 of which foreign exchange requirements alone amount to US$3,793,300. A breakdown of foreign and local costs for each plant is presented in Table 1.3. A breakdown in greater detail of the capital, O&M and transportation costs is presented in Chapter 10 along with a disbursement schedule. - 10 - Table 1.3: FOCIŁN AND LOCAL INVESTMENT COSTS a/ FOR PILOT BRIQUETTING PLANTS Pilot Plant Foreign Local Total (USS)---i__ Coffeet Parchmont (Mercato) 321,710 55,360 377,070 Parchment (New Site) b/ 34,650 90,820 125,470 Husk 614,570 192,930 807,500 Pulp 106,110 76,990 183,100 Cotton 1,240,180 308,990 1,549,170 Wheat 1,199,930 327,430 1,527,360 Maize 149760 454,320 1.951.880 .Total 5,014,710 1,506,840 6,521,550 a/ Includes one year's charges for operation, malntenance and transportatlon, as well as capital Investments. b/ Includes only the marginal costs of expanding the plant to 5,000 tonnes por year capacity. A more complete project budget with foreign and local figures for capital, operating and transportation costs is presented in Chapter 10. Detailed budgets for each site are in Annex 1. The project will be financed by an IDA credit for the energy sector in Ethiopia. Conclusions 1.17 Agricultural residues briquettes are a viable economic alternative to increasingly scarce and costly fuelwood for both domestic and industrial fuel. If necessary, the entire production from the pilot projects can be eisily absorbed by industries that presently use fuelwood or charcoal, or that have the capability to utilize fuelwood or charcoal. While the acceptance of agricultural residues briquettes in households is still undetermined, tests in typical stoves, as well as laboratory analysis, indicate that they can be used as a substitute with no major problems. The pilot projects represent only a small fraction of the potential for agricultural residues utilization. Establishing a market for these residues will have the dual benefits of reducing pressure on scarce fuelwood supplies and increasing self-reliant economic activity in Ethiopia. - 11 - Recommendations 1.18 As a result of these conclusions, the following actions are recomnded: (a) proceed with the pilot briquetting plants in the following order of priority, according to their economic feasibility, managerial viability and future potential for expansion: (i) coffee parchment in Addis Ababa (Mercato site), (ii) coffee husks in Dilla, (iii) cotton stalks in Middle Awash, (iv) wheat straw in Dixis, (v) coffee parchment in Addis Ababa (new site), (vi) maize residues in Anger, and (vii) coffee pulp around Dilla; (b) obtain industrial purchase agreements so as to ensure a guaranteed demand for the output of the pilot briquetting plants; (c) establish a systematic program to introduce the use of .briquetted fuels to the domestic sector, including a feedback mechanism to gauge potential problems and improvements; and (d) establish an assessment system so that each of the pilot plants can be monitored and evaluated to help identify the most promising areas for future expansion. - 12 - II. DESIPIED ACICULTURAL RESIDUES FOR PUELVOOD SUBSTITUTION Background 2.1 Ethiopia's economy faces increasingly severe problems in the 1980s and beyond, primarily as a result of persistent drought and worsening terms of trade. In the energy sector, the cost of petroleum imports is an enormous burden, amounti-; to about half of foreign exchange earnings in 1982/83. The other, more pervasive problem is the increasing scarcity and cost of traditional household fuels, its massive impact on deforestation and the resultant depletion of agricultural resources on which so much economic activity depends. Major programs of reforestation are urgently needed to redress the imbalance between supply and demand for woodfuels, and to enhance the agricultural ecology. However, this is a medium- to long-term solution and, in the immediate future, the development and utilization of indigenous fuel substitutes must be vigorously pursued. Fuelwood Supply and Demand 2.2 The Ethiopian landscape has deteriorated rapidly during this century. The estimated cover of closed canopy forest has fallen from roughly 40X around the turn of the century to less than 31 now. Some 200,000 hectares and one billion tonnes of topsoil are lost each year as land is cleared and forests thinned to supply new areas for crops, animals and people, and to meet the growing demand for the country's principal fuel, firewood. 5/ With an average daily per capita consumption of 2 kg of fuelwood, biomass represents over 901 of final energy demand. With a growing population and decreasing fuelwood supply, the cost of firewood in Addis Ababa has soared from $9 to $83 per tonne between 1973 and 1985; similarly, charcoal prices have risen from $75 to a high of $435 per tonne between 1975 and 1985. 6/ The average household expenditure on fuel is over $1 per day at these rates. With an average per capita income of US$120 and an 8-person household, this amounts to over 351 of household income for an average Ethiopian family in Addis Ababa. 5/ Ethiopia: Issues and Options in the Energy Sector. July 1984, Joint UNDP/World Bank Energy Sector Assessment Program, Report No. 4741-ET and Bank mission estimate. 6/ Woodfuels market price assessment, April 1985. - 13 - Need for a Fuelwood Substitute 2.3 To compensate for firewood scarcity and high prices, growing amounts of natural fertilizers in the form of dung and crop residues normally returned to the soil are now being diverted to household cooking, reducing crop yields by more than one million tonnes of grain a year. If the present level of reforestation continues and no substitute fuels are introduced, in 20 to 30 years all but the least accessible pockets of forest will have been consumed and much of Ethiopia's center will be like Eritrea and Tigray today: subject to persistent drought, crop failure, famine and outmigration. To prevent this, 6 million ha would have to be planted in the next 30 years. However, it would take ten to fifteen years before any forestry efforts could even begin to improve the imbalance between fuelwood supply and demand. Therefore, it is of immediate importance to find substitute energy sources to ease the pressure on existing fuelwood plantations and natural forests and to provide time for reforestation programs to be implemented. Availability of Agricultural Residues 2.4 Nationally, 3,308,000 tonnes of residues from coffee, cotton, wheat and maize are theoretically available. This is equivalent to a potential of 3,319,000 tonnes of fuelwood or 1,359,000 toe, as indicated in Table 2.1. Roughly 4 percent of the cultivated land in Ethiopia is occupied by State-owned Farms and processing facilities where surplus cotton, wheat, maize and coffee residues form a potential residue source which is concentrated and can be easily collected. These crops yield approximately 581,400 tonnes of residues annually, 7/ equivalent to 639,700 tonnes of fuelvood or about 50X of total household fuel demand in Addis Ababa. The major alternative uses for these residues are: - grazing by roaming herds after harvest; - ploughing part of the residues to enhance soil structure and fertility; - using residues directly for energy; and - baling of wheat straw for cattle consumption due to shortages of cattle feed. 2.5 Even after taking into account these valid alternatives, it is estimated that virtually all of the coffee, cotton and maize residues and 7/ Ministry of State Farms field tests (December 1984 - February 1985) and mission estimates. - 14 - 501 of the wheat straw could be recovered and used as fuelwood substitutes. As Table 2.1 indicates, this amounted to a potential of 581,400 tonnes of surplus residue i the 1984/85 harvest season, or enough biomass to supply almost half of Atdis Ababa's annual household energy needs. Table 2.1: AVAILABILITY OF RESIDUES 1984/85 HARVEST SEASON Available Fusiwood Theoretical Range Crop Residue Area Residue a/ Equivalent TOE of Avallable Residue b/ (ha) -- -- -----(tonnes)----- Coffee wastes - 220,000 283,500 94.900 220,000 Cotton stalks 30,600 116,000c/ 109,600 45,600 244,720-350,000 Wheat straw: National d/ 648,500 701,400 ./ 692,730 288,550 420,840-1,227,450 State Farms 76,000 79,800 O/ 78,810 32,820 45,600-133,000 Malze residues: National d/ 597,500 2,270,500 2,233,510 930,190 1,195,000-10,898,400 State Farms 46,000 165,600 167,800 69,890 92,000-220,800 Total: National - 3,307,900 3,319,340 1,359,190 2,080,560-12,695,850 State Farms - 581,400 639,710 243,210 602,320-923,800 a/ Based on mission estimates and Ministry of State Farms and Mlnistry of Coffee and Tea Development data. b/ Based on developing country averages (Source: Bernard and Kristoferson, Agricultural Residues as Fuel In the Third World, Earthscan, 1985). c/ P1ter grazing. d/ 1981-1982 figures. _/ Assuming that 50% of the residues are left on the ground for soil replenishment. The figures on potential residue production levels used in this reiort were collected at selected State Farms and are thus primarily representative of a particular place and time. A nationwide residue utilization project would need to rely on representative data collected from all State Farms over several harvest seasons. - 15 - Suitability of Residues as a Fuelvood Substitute 2.6 In theory, most agricultural residues can be used as a fuelwood substitute in loose form. However, for reasons of transport, storage, marketing and stove design, residue use can be enhanced and expanded through densification, e.g. in the form of briquettes that simulate fuelwood. The ability to process and market suitable briquettes from residues depends on the densification technology, chemical composition of the basic material (especially lignin content), residue characteristics (moisture content, size, presence of foreign matter, etc.) and the social/technical acceptability of the final product. 2.7 Various technologies for densifying a variety of materials exist and are described later in this chapter. The choice of technology not only affects the physical integrity and burning characteristics of the briquette but also its price due to varying capital, operating and maintenance costs of equipment. Chemical composition will also determine a briquette's suitability for particular uses. For example, briquettes with a high ash content might not be a problem in domestic stoves but could cause serious slagging problems in industrial boilers. Excessive smoke may not be acceptable in an urban household but could be surmounted in a boiler where cleaning devices or secondary combustion can be utilized. A residue's moisture and lignin content will affect its quality as a briquetting material, though the former can be altered by drying or adding water. 2.8 Concerning acceptability of the final product, briquette quality requirements are less stringent for industrial as opposed to household applications. This is particularly true in Ethiopia where most cooking is done on simple stoves or over a three-stone open fire. In this case, a fuel must exhibit several characteristics, including: - yield a simmering fire with small flames; - be composed of a slow-burning material which does not disintegrate easily; - be easily extinguishable with leftovers recoverable for preparing the next meal; - marketable in small, usually daily, quantities. Extensive trials and tests are required to determine the social acceptability of briquettes for Ethiopian cooking practices and households. - 16 - Densification Technology 2.9 Densification is the general process of compressing suitable raw materials into a more compact or densif ied (i.e., higher density) form such as pellete, briquettes, logs, bales, etc. The resulting material is easier to handle, store and use. Densification techniques have been available for many years and were primarily used to produce animal feed, fertilizer, etc. More recently, densification has been used for compressing loose sawdust, shavings, and various agricultural residues to produce fuel pellets or briquettes that are suitable for use in existing solid fuel combustion systems. Particular examples of agricultural residues that have been successfully densified include coffee husks, rice husks, groundnut shells, sugarcane bagasse, cotton stalks, straw and stovers from wheat, barley and maize, residues from cottonseed, sunflower seed and coconut husks. In general, most lipnocellulosic (woody) biomass type agricultural residues are potential candidates for densification. 2.10 Densification processes fall into two broad categories: extrusion and batch processes. Extrusion processes, which include screw and piston extruders, pellet mills and cubers, force the feedstock through a hole or die by one of several processes. The die wall provides the back pressure for compression and creates friction between the sliding material which results in the transfer of heat to the material which *n turn helps promote binding of the material (through the release of its natural lignins). In batch processes, which include roll briquetting, baling and manual presses, the walls of the compression chamber provide the pressure and there is little or no friction heating. Materials densified in batch processes generally need the addition of artificiaL binders and are not as compact (dense) as extrusion processed materials. Pre-Processing 2.11 Prior to densification, agricultural residues have to undergo a number of stages including collection, storage, cleaning, drying, comminution and feeding. These stages are outlined in Figure 2.1. Depending on the residue, each of the above stages will require a certain degree of equipment, materials and labor. Collection 2.12 Depending on the agricultural residue, collection can be a major component of the densification process. For example, materials like cereal straw and stover and cotton stalks tend to be widely dispersed in the fields and must be collected and transported to a central location. Alternately, materials like coffee husks, sugarcane bagasse, rice husks, etc., are produced at central locations and do not present a major collection effort. - 17 - Figure 2.1: GENERAL STEPS IN A OENSIFICATION PROCESS Steps~ Raw Residues Cotton Stalks, Wheat Straw, Maize, Barley and Sorghum Stalk and Stoves, Coffee Husks, Pulp and Parchment, Bagasse, etc. Collection Balers, Choppers/Forage Wagons, Trucks, etc. Drying Natural (Solar) or Artificial (Process iteat-Drum Dryers, Pneumatic Dryers, etc.). Storage Open Piles, Sheds, Silos, etc. Cleaning Screens, Air Classifiers, Electromagnets, etc. Comalnution/'Nize Hamm_rmills, Choppers, Manual Shredders, etc. Reduction Feeding Belt Conveyors, Screw Conveyors, Pneumatic Conveyors or Manual Labor, etc. Dens!fication Piston Press, Screw Press, Pellet Mill, Roll Briquettor, Manual Press. Briquettes * Sequential order of process steps are not necessarily fixed. - 18 - Storage 2.13 The type of storage required will depend on the residue and the environmental conditions it is subjected to. Usually the residue will be stored in an open-air heap, a shed, a bin or within retaining walls or fences. If the collected residue is dry and open-air storage would result in the accumulation of moisture, then closed or sheltered storage is necessary. Conversely, wet residue can be reduced in moisture content through carefully managed open storage. Cleaning 2.14 Cleaning is necessary if the residue contains foreign materials (such as stones, soil, metal, etc.) that could damage the processing and densifying equipment. Cleaning can usually be achieved with pneumatic, mect?nical and/or magnetic screens. Drying 2.15 In general, most extrusion-type densification equipment requires that the feedstock be in the range of 10 to 20 percent moisture content on a wet basis (Z mcwb). If the moisture content of the feedstock is too high (above 20X mewb), the excess water becomes a superheated liquid because. of the high pressure required for densification and the resultant frictional heat buildup. The water will flash to steam as it exits the densifier and the pressure is lowered, usually exploding the briquette or pellet. For low pressure densification systems such as baling or manual presses, the principal concern regarding moisture is decomposition during storage. Stored at moisture contents above 20 percent for extended periodts, any biomass will begin to decompose, reducing its calorific value and posing a risk of spontaneous combustion. Because of this, drying of the residue prior to densification is required if the material as received is above 201 mcwb. The method of drying will depend on several factors including environmental conditions, initial moisture content of material, level of throughput, size of material, type of densifying equipment, etc. Comminution 2.16 Most densification equipment requires that the maximum particle size of the incoming feedstock be no more than 25 percent of the diameter of the resulting briquette or pellet. For example, a piston extruder producing briquettes 50 mm in diameter has a maximum particle size constraint of 12 mm. Feedstock size reduction is usually done with a hammermill. Various configurations of hammermills are available depending on the characteristics of the input feedstock and the required output. The cost of particle size reduction, including capital, energy, labor and maintenance costs can be relatively high, often as much as the densification process itself. - 19 - Feedins 2.17 Once the residue has been cleaned, dried and sized, it must be transported to the densifying equipment. This is generally done with mechanical or pneumatic conveyors. If the densification system is of low capacity, then manual labor might be substituted. However, this is generally not cost effective. Densification Equipment 2.18 As indicated earlier, there are two major categories of densification processes: extrusion and batch processes. The latter requires binders and results in a low density product which is not suitable for multiple handling and long distance transport. In most cases, the only available binder is starch based which has an alternative food value and which adds significantly to the cost of production. Thus, batch processing systems are not considered as a viable alternative in this study. 2.19 There are basically four main types of extrusion densification processes: (a) Piston press briquettors (b) Screw press briquettors (c) Roll briquettors; and (d) Pellet mills. 2.20 A brief description of each of these processes is described below. Piston Press Briguetting 2.21 In this extrusion process, illustrated in Figure 2.2, a reciprocating piston forces the feed material into a die, where pressure and friction heat the feedstock to 150-3000C before it is extruded through a die 25-100 m in diameter. In most cases, the die is water- cooled to reduce wear. The briquettes then enter a cooling line which, by friction, provides a back pressure on the material exiting the dies so that the cooling takes place with gradually decreasing pressure. A sudden pressure drop can cause the high temperature water to flash to steam, exploding the briquette. The back pressure can often be adjusted to allow optimum production for fuels with varying moisture contents. As they exit the cooling line, briquettes may be cut or broker off at any desired length. - 20 - hure2tSIDE VIEW OF TIM S - R.WCW- * \ , * , _~~~~~~~ CWa"d W-034 - 21 - 2.22 Briquettes produced by the piston press method tend to be highly durable. However, this durability is a function of the maximum barrel pressure encountered in tha production of the briquette. Barrel pressures in excess of 1,000 kg/cm' are recommended for the production of durable briquettes. Important aspects of the piston extrusion process are the wide range of capacities of available equipment, from 150 kg to 1.5 tonnes per hour. Piston press systems are characterized by generally high capital costs and moderate operating costs. Screw Press Briguetting 2.23 In this slightly different process, a conical screw forces the feed material into a die similar to that of the piston extruder. One type of screw extruder is shown in Figure 2.3. An important point regarding the screw extrusion process is the low production capacity, less than one tonne per hour to as low as 75kg/hr. This results in high labor costs and high capital costs per tonne of output. Another point is the high amount of friction heating by the screw, resulting in higher die temperatures and increased wear on the screw and die head. In some models, the screw and die head are water-cooled, but for certain processes, such as densifying rice hulls, the high temperature is desirable to facilitate densification, and the die may even be heated further with electric elements. In these cases, the die temperature can reach 400OC, causing carbonization of the outer layer of the briquette. The wear on the die caused by such high temperatures results in high maintenOd Cobe FPcessl Sun Dded41k Ctodo +wowE0m kg,Sun DX.d Cof Re dRS Sun Dded c4feBe=s *GW d coItes NA N*dSs Is iSA 8tvU Waft Br*-300501 - 32 - Figure 32: Mass Balance for Washed Codfee P sng Guen Bow Pu PaIchn t 65 65% apcm 2 I k PChmw 10% wb 10% Readv for bodt 10% Wuld Baruk-30050:2 - 33 - 3.12 As a result of the washed processing method, two distinct types of residues are generated. The first is the wet coffee pulp which consists of the epicarp that is removed at the washing stations in the coffee growing regions. For 100 kg of ripe cherries delivered to a washing station, 602 by weight end up as washed coffee pulp with the remaining 40X consisting of the green bean and endocarp or parchment. Of this, only 20 kg remain after sun-drying of the bean and parchment. This is then shipped to the washed coffee processing facility in Addis Ababa where the parchment is removed. The result is 16 kg of washed coffee beans ready for export and 4 kg of parchment as residue. A mass balance for the washed coffee process is shown in Figure 3.2. 3.13 With a current production of 13,000 tonnes of washed coffee beans, potentially 2,600 tonnes of dry coffee parchment (102 mcwb) are available at the washed coffee processing plant in Addis Ababa. Presently, most of the parchment is sold for 2 Birr per 40 kg sack to the local population who utilize the loose parchment as a cooking fuel. However, demand is irregular and storage of this bulky material is a problem. With washed coffee production projected to increase to 30,000 tonnes, the availability of coffee parchment residues will increase to 6,000 tonnes which is the energy equivalent of 6,200 tonnes of air-dried fuelwood or 2,600 toe. 10/ 3.14 An additional 45,000 tonnes of wet pulp (651 mewb) are produced at the approximately 180 washing stations in the coffee growing regions of Sidamo and Kaffa. The average production at most washing stations is between 150 to 500 tonnes of wet pulp depending on the capacity of the washing station. 3.15 Presently, the wet pulp is discharged into local streams where it generally clogs, forming a putrescent mass and producing a highly acidic effluent which pollutes the water, destroying any form of aquatic life and generating an offensive odor. Recovery of this pulp for fuel use would require collecting the residues as they are discharged from the pulping machine and processing them to greatly reduce the moisture content. Assuming that the moisture content is reduced to 121 mcwb for fuel use, approximately 18,000 tonnes of fuel quality residue would be available. This would have an energy equivalent of 18,200 tonnes of air- dried fuelwood or 7,600 toe. 11/ A discussion of potential ways for drying the residue is presented later. 10/ GHV of coffee parchment residues is 18.5 NJ/kg. 11/ GHV of coffee pulp is 18.0 MJ/kg. -34 - Potential Pilot Projects for Coffee Residue Bripuetting 3.16 Table 3.2 presents a summary of the coffee residue production in Ethiopia for the three main types of residuess husks, pulp and parchment. A chemical and energy analysis of coffee residues is presented in Table 3.3. The data indicate that coffee residues have relatively comparable characteristics to those of Eucalyptus slobulus, a typical Ethiopian fuelwood. The gross heating values of the coffee residues range between 18.0 and 19.3 NJ/kg, indicating that they could be an ezcellent fuelvood substitute. TabIe 3i2: SUPARY OF COFFEE RESIDUES PRODUCJION Quantity Produced Fuslwood (tonnes/year) Equivalent a/ Location and Residues at 12S mcwb (tosnas) TOE OIstrIbution Sun Dried coffef husks 200,000 206,700 86,10 At over 110 sun dried coffee decorticating plants locat"d In the coffee growing region sWshed coffee pulp 18,000 bl 18,200 7,600 At 0er 180 washing stations primarily In Sidamo and Koffs Wshed coff parchment 2,0C0 a/ 2,700 1,100 At the NWeato washed coffee processing plant In Addis Abab Total From Coffee ResIdues 220,600 283,500 94,900 a/ Fuelwood OVW 17.79 MJ/kgA t 15 mewb. b/ Assumes that 45,000 tonnes of 65% mcwb as produced Is reduced to 12% mcwb. c/ Present production site of parchment Is Addis Ababa. Table 3j3: COfFEE RESlOUES-O4ENICAL AND ENERr ANALYSIS Husks Parchment Pulp Eucalyptus moisture Content (5) a/ 12 10 12 1S Ash Content (S) N/A 0.33 NWA 0.29 Volatile Oontent (S) NWA 83.4 N/A 82.2 C (5) 47,8 50.1 N/A 47,46 H (5) 5,1 6.19 NWA 6.08 0 (5) 36.0 42.7 NWA 45.82 N (5) 3.0 0.24 N/A W4A Sulphuric Lignin (5) 20.6 25.1 NYA NA HHV NJ/kg 20.90 20.55 20.45 21.00 6HY NJ/kg 18.40 18.0 1t.a00 17.80 a/ Except for moisture content, measuroments are given on a dry basis. - 35 - 3.17 Husks represent over 90X of the coffee residue production. However, the extremely low bulk density (approximately 50-80 kg/m3) of the husks produced precludes their economic transport to large urban areas 300-500 km distant. Densifying or briquetting this material to a density of 500 to 600 kg/m3 would greatly reduce transport, handling and storage costs and facilitate their use as a domestic or industrial fuel. Washed coffee pulp, while amounting to only 18,000 tonnes of 121 mcwb residue, represents a growing source of residues which, under present practices, also constitutes a major environmental and ecological problem. Utilizing this residue as a source of energy would have multiple benefits. Finally, coffee parchment, while only a fraction of the available coffee residues, is strategically located in Addis Ababa and can be i_mediately used as a domestic or industrial fuel. In addition, production of this residue in Addis Ababa is projected to double over the next five years. Densifying this material would extend its market potential and enhance its efficient use. 3.18 Given the above, two distinct pilot projects are proposed for the densifying of coffee residues: (a) A pilot plant to produce fuel briquettes primarily from sun- dried coffee husks, which might also be supplemented with washed coffee pulp residues that are available in the selected region; and (b) A pilot plant to produce fuel briquettes from the washed coffee parchment that is available in Addis Ababa. Pilot Plant for Briguetting of Sun-Dried Coffee Husks and Washed Coffee Site Selection 3.19 There are 55 towns in the eight coffee producing regions containing a total of 110 sun-dried coffee decorticating plants. Coffee is processed in these centers for the Coffee Marketing Authority and for private merchants. Most of the sun-dried coffee arriving in Addis Ababa is likely to have been processed at these centers. Table 3.4 presents a record of the coffee processed at these centers for the Coffee Marketing Board in 1981/82 season, specifying the number of decorticating stations, the tonnes of coffee processed and the distance of these centers to Addis. 3.20 Given the above information, the Dilla processing plant in the Sidamo area was identified by the Ethiopian Ministry of Coffee and Tea Development as the best site for a pilot coffee husk briquetting project. The Dilla plant represents one of the larger coffee processing centers in the country. It is connected to the national power grid and is on one of the major radial truck routes to Addis Ababa. In additiont several washed coffee stations are located in the vicinity and these could provide supplementary quantities of washed coffee pulp for - 36 - briquetting. Washed coffee pulp is produced during the months of September to January while sun-dried coffee is processed from December to July. Table 3.4: COFFEE PROCESSED AT DRY-PULPING STATIONS FOR THE ETHIOPIAN COFFEE MARKETING COORORATION 1981/82 Location Distance from Erected To be erected Total Tonnes of coffee Region Town Addis Ababa very soon processed for ECMC (kmn) Kaffa Nizan 585 1 1 2 1,567 nf Jima 335 3 - 3 2,844 Kaffa Yebu 355 2 1 3 1,044 LImu Genet 410 1 - I 1,337 Kaffa Limu Seke 428 1 - 1 290 if 0Imbe 370 2 - 2 1,285 nf Agaro 380 10 - 10 1,840 Babu 375 - 1 1 Illubabor Tepl 575 2 1 3 1,454 Toba 405 5 - 5 934 Denbl 420 1 1 2 706 if echi 460 1 - 1 226 nf 3edele 480 2 - 2 257 lilubabor Chora 515 2 - 2 377 Yayu 560 3 - 3 17 Hurumu 580 3 - 3 328 tMotu 600 3 - 3 491 -- Gore 620 1 - 1 163 nf Nola 614 1 - 1 159 Supe 640 2 - 2 n.a. Iollega Gimbi 441 2 - 2 1,044 Guye 464 1 1 2 411 Enango 464 2 - 2 1,295 Goliso 501 2 - 2 602 kAKara 561 2 - 2 888 nf Begi 688 1 - 1 98 nf Kaka 602 4 - 4 367 Cidame 687 3 - 3 998 Wollega D/Dolo 650 4 - 4 734 n ZNugl 693 8 - 8 2,345 3.21 The site in Dilla selected by the Ministry of Coffee and Tea presently has four decorticating mills installed in an "industrial zone" that is owned by the Coffee and Tea Marketing Authority (CTMA). Additionally, two private mills are to be located in a separate building - 37 - adjacent to the CTMA mills. The capacity of the CTMA mills, when operated 24 hours per day, is 70 tonnes of coffee beans with a resulting 95 tonnes of residue. The maximum hourly residue production is therefore a little under 4 tonnes per hour. However, average daily production is about 23 tonnes of beans per day with a residue production of 30 tonnes per day. The annual residue production of the CTMA mills is approximately 5,000 tonnes. Thus, the CTKA mills in conjunction with the two private mills will produce more than adequate quantities of coffee husks to support a 5,000 tonne per year briquetting plant at the Dilla site. Plant Design and Equipment 3.22 Figure 3.3 presents a flow diagram for the proposed coffee husk briquetting plant. Each of the existing blow pipes coming from the coffee mills is connected to a cyclone separator (1). The cyclones discharge the husks onto a horizontal belt conveyor (2) which can discharge in both directions. When more residue comes from the mills than the briquetting plant can process, the husks are then discharged onto an intermediate storage heap by means of an inclined mobile belt conveyor (3). The mobile belt conveyor can be used to collect this material and feed it again onto the horisontal belt conveyor (2) when no residue comes from the coffee mills. This makes the briqttdtting and milling plants as independent as possible from each other. During normal operation the residue is charged directly into the briquettor's surge hoppers (4) by means of an inclined belt conveyor (5). The husks are then briquetted in two piston briquettors (6). After leaving the die the briquettes are pushed through a cooling line (7) of about 6 to 10 meters. At the end of the cooling line, the warm briquettes are allowed to fall into bags with the aid of a bagging station (8). The filled bags are then removed to a cooling and storage area capable of holding about 10 days of production, prior to being transported. 3.23 The plot plan for the briquetting plant is shown in Fig. 3.4. The briquetting plant requires a 40 by 25 meter covered building. The front wall of the building along the cooling area is covered only by open wire mesh to allow sufficient air circulation. Other side walls should allow for ample ventilation. 3.24 A list of equipment and costs for the pilot coffee husks briquetting plant is presented in Table 3.5. Two piston briquettors are required to meet the desired plant capacity of 5,000 tonnes per year (6,000 tonnes with the use of 1,000 tonnes of pulp). Each briquettor has a maximum capacity of 0.75 tonnes per hour for a total maximum capacity of 1.50 tonnes per hour. Thus, the maximum daily capacity, assuming the briquettors operate full time, is 36 tonnes per day which is in the range of the average daily production of 30 tonnes of residue per day from the mills and equal to half the maximum daily capacity of the mills. The plant is assumed to operate 200 days per season for 20 hours per day at an average capacity of 1.25 tonnes per hour. OUflCUUOO ~ ~ ~ ~ ~ ~ ~ ~ ~ ~ ~ ~ ~~~XMAOO4 ~~~~~~~~~~~~~~~~~~~~~~~~0 Wd u b BoUQep enpe e eee :rc eanOS 39 - Rgur 3A: Plot Plan for Cobe Residu Stuetlng Plant CdbeR~~~~~~Coe . * C4ffe. Re". Mft COW*o~XITIGCOFE ~IL 5 SCa~~~~~~~~~~~~~ ii1. '=~~~~~~~~~~~ L I *_I\ ,1-* - N6EMICOFfEMU I2 4Um %Wol Bonk-300044 - 40 - Table 3.5: DILLA COFFEE HUSKS BRIQUETTING PILOT PLANT: CONSTRWCT ION AND EQUIPNENT COSTS Item Local Foreign Total (USS) Oonstruction Site Preparation 11,00 - 11,600 Buildings 65,800 19,100 84,900 Housi n n - - - Constructlon Total 77,400 19,100 96,500 E4uIpM*nt Piston briquettors (2) - 172,000 172,000 Collectlon/storage equlpnent - 6,000 6,000 ConveFor$ - 20,000 20,000 S1Ila - 25,000 25,000 Electrical - 13,000 13,000 TrOnsftOrer - 7,C00 7,000 BaggIno statlon - 2,000 2,000 Workshop equIupment - 12,400 12,400 OtherMiscellaneous - 2,500 2,500 EquI pment Total - 259,900 259,900 3.25 Power requirements for the briquettors, conveyors, fans, lighting, etc., are estimated to be 180 kW. Each piston briquettor has 'a motor rated at 75 kW. An average energy input of 60 kWh per tonne of briquettes is assumed. A 500 kVA transformer is proposed to obtain power from the national grid lines located next to the proposed plant. Capital Costs 3.26 The total capital costs of the coffee husk pilot plant are $516,845. This includes costs for construction (site preparation and buildings), equipment, spare parts, equipment delivery, engineering, installation and contingencies. A summary of these capital costs by year and by foreign and local components is presented in Table 3.6. Detailed costs are presented in Annex 1. The total foreign costs are estimated at $421,350 while local costs are estimated at $95,500. The capital costs were annualized over the plant or equipment life at a lOZ discount rate. The total annualized capital costs are reported in -he last column of Table 3.6. Total annualized capital costs are estimated at $67,830. Assuming an annual production of 5,000 tonnes of briquettes, the capital charge per tonne of briquettes is $13.56. - 41 - Annual OperatinA and Maintenance Costs 3.27 The annual ObM costs for a 5,000 tonne per year husk briquetting plant are estimated at $117,352 per year. A summary of this estimate is presented in Table 3.7. Of the total, local costs are $88,763 and foreign costs $28,589. The major annual local costs are $18,900 for power and $20,600 for bags. Given a production of 5,000 tonnes per year the OM charge per tonne of briquettes is $23.47. Details of the annual 06M costs are presented in Annex 1. Production Costs 3.28 Average production costs for coffee husk briquettes are calculated based on the the capital and O0M cost estimates presented in Tables 3.6 and 3.7. Average financial production costs for coffee husk briquettes are $37.04/tonne while economic costs 12/ are $42.00/tonne. Table 3.6: DiLLA COFFEE HUSKS BRIQUETTING PILOT PLANT: CAPITAL COSTS (USS) (5,000 tonnrs/yr capacity) (USS) Total Costs Annualized Charges Item Local Foreign Total Local ForelIn Total Constructlon 71,400 19,100 96,500 9,090 2,240 11,330 Equrpment - 259,900 259,900 - 36,050 36,050 Spares (at delivery) - 26,000 26,000 - 3,550 3,550 Transport and DelIvery 1,8I0 21,600 23,400 250 2,950 3,200 Engineering and Installation 7,620 56,460 64,080 900 6,630 7,530 Contingencles _8.60 38,300 46,980 1.020 5.140 6.160 Total 95,500 421,350 516,850 11,260 56,560 67,830 12/ Economic costs are derived by utilizing a shadow factor of 1.33 in all foreign exchange expenditures and 0.50 on all local labor expenditures. In addition, economic prices of $0.397/1 for diesel and $0.063/kWh are utilized. -42 - Tdble 3.7: DILLA CtFEE HUSKS BRI19ITTINS PILOT PLANT: CU COSTS Local Forelgn Total (USS) A, Labort 28,000 28,000 Skilled 7,700 7,700 Unskilled B. Power: Electrlclty 18,900 18,900 Diesel C. Operation & Maintenance Fuel Lube Oil Spare Parts 25,990 25,990 Servioes 2,599 2,59 * BuIlIding Mwintenance materials 2,895 2,895 E. C1usuuables: ags 20,600 20,600 Other Operating Costs 0,694 25,990 106,684 Coniglency 8S069 2.S99 10,666 Total Operating Costs 08,763 28,589 117,352 3.29 For briquettes delivered to Addis Ababa, the transportation costs will add $34.66/tonne to the financial costs and $20.18/tonne to the economic costs. Thus, the delivered financial costs in Addis Ababa are $71.701tonne and the economic costs are $62.17/tonne. The financial and economic costs per unit of energy are $3.63/GJ and $3.15/GJ respectively. A summary of the financial and economic costs of coffee husk briquettes is presented in Table 3.8. Table 3.8: FINANCIAL AND ECONOMIC COSTS OF COFFEE HUSK SRIQUETTES Financial Economic Delivered Ex-Factory Delivered Ex-Factory Cost/raone (USS) 71.70 37.04 62.17 42.00 Cost/40 kg Bag (USS) 2,87 1.48 2.49 1.68 Cost/iJ (US$) 3.63 1.86 3.14 2.12 - 43 - Pilot Plant for Producing Dried Coffee Pulp 3.30 The average throughput of ripe coffee cherries at a new washing station is about 600 tonnes per year. This will result in a potential production of 360 tonnes of wet (652 mcwb) pulp or about 143 tonnes of dry (12X mcwb) pulp a year. With a washing season of about 120 days, there is an average production of 3 tonnes of wet pulp or about 1.2 tonnes of dry pulp per day. Such a limited pulp production does not justify the installation of a densification unit at these washing stations. Howevert the dried pulp could be collected to be densified at a central plant. Eventually, about 6 to 8 new washing stations in the Dilla area could be selected to produce 10 tonnes/day equivalent to 1,000 tonnes per year of dry washed coffee pulp for briquetting at the proposed Dilla sun-dried coffee husk briquetting plant. Since the washed coffee pulp is produced during the months when the sun-dried coffee mills are not operating, this residue could be used to increase the capacity of the proposed Dilla briquetting plant while simultaneously being tested in the pilot project for future large-scale production. 3.31 Collection of the washed coffee pulp will require that the material be first trapped in a sieve or perforated tray. The soaking wet material is then pressed in a simple manual press to extricate as such water as possible. The "squeeze dry" material is then sun-dried in wire mesh trays that are similar to those used to sun-dry the washed coffee beans. With regular manual turning of the pulp material, the pulp will dry lo about 12X mcwb in 4 to 5 days. Assuming a dry bulk density of 50 kg/im and a material depth after drying in the trays of 0.3 m results in approximately 333 m' of trays required to produce 1 tonne per day of dry pulp on a regular basis. 3.32 After sun-drying, the pulp is emptied into sacks for storage. Bach sack is assumed to hold 20 kg of pulp. The sacks are to be stored under a small storage shed capable of holding up to 500 sacks equal to an average of 10 days of production at each washed coffee station. The dried pulp is then to be collected once a week by a roving 50 kW - 2Wd tractor/trailer unit to be delivered to the pilot coffee husk briquetting plant at Dilla. 3.33 The cost of equipment required to produce and deliver 1,000 tonnes of dried coffee pulp to the pilot coffee husk briquetting plant is approximately $65,800 of which $26,800 is local costs. Collection equipment consists primarily of manual expellers, screens, drying trays, rakes, sacks, a tractor and two trailers. It is assumed that this equipment is used to support the production of the 1,000 tonnes of pulp from up to 10 washing stations each producing an average of 100 tonnes of pulp per season or 1 tonne per day. - 44 - Capital Costs for Pulp Drying 3.34 The total capital costs for coffee pulp drying are estimated at $107,630. A summary of this estimate is presented in Table 3.9. Detailed costs are presented in Annex 1. The capital costs were annuaiised over the plant and equipment life at a 10 discount rate. Total annualised capital costs are estimated at $17,385 resulting in a capital charge per tonne of dry pulp of $17.39. Table 3.9: DILLA COFFEE PULP DRYING PILOT PLANT: CAPITAL COSTS (USS) (1,000 tonne/yr capacity) Total Costs Annualized Charges Item Local Foreign Total Local Foreign Total Construction 11,600 - 11,600 1,360 - 1,360 Equlpment 26,800 39,000 65,800 5,020 6,430 11,470 Spares (at delivery) 2,680 3,900 6,580 460 660 1,200 Transport and delivery 1,570 2,550 4,120 270 430 700 Ergineerlng & Installation 630 9,110 9,740 070 1,070 1,140 Contingencies 4,330 5.460 9.790 720 860 1,580 Total 47,620 60,010 107,630 7,900 9,480 17,380 Annual O&M Costs for Pulp Drying 3.35 The annual 06M costs of drying and delivering 1,000 tonnes of washed coffee pulp to the coffee husk briquetting plant are estimated at $17,342. Therefore* production and delivery of dry coffee pulp has an average O0M cost of $17.34 per tonne. A sumuary of the annual 06& cost estimates for pulp drying is presented in Table 3.10. Details are presented in Annex 1. - 45 - Table 3.10: DILLA COFFEE PULP DRYING PILOT PLANT: ON COSTS Local Foreign Total (US$) -… A. Labor: Skilled 500 500 Unskilled 3,800 3,800 Be Operation & Maintenance: Fuel 2,618 2,618 Lube Oil 0,262 0,262 Spare Parts 680 3,900 6,560 Services 658 0,658 C, Building Maintenance: Naterlals 0,348 0,348 D. Consumabless Bags 1,000 1,000 1,000 Other 000 Operating Costs 8,986 6,780 15,766 Contingency 899 678 1.577 Total Operating Costs 9,8B5 7,458 17,342 Production Costs of Coffee Pulp Briguettes 3.36 The average financial and economic costs of drying and delivering the coffee pulp from the washed coffee stations to the pilot coffee husk briquetting plant at Dilla are estimated at $37.04/tonne and $41.81/tonne respectively. However, the dry pulp material must still be tensified in order to produce briquettes that can be transported to markets in Addis Ababa or elsewhere. Estimates of the average financial and economic costs of producing coffee pulp briquettes are calculated based on the same average production costs for producing coffee husk briquettes plus the average costs of drying the pulp. This assumes that both drying and briquetting capital and O&M charges are assessed against the coffee pulp briquettes. The resulting financial and economic costs are $71.76/tonne and $79.17/tonne respectively. The delivered costs in Addis Ababa are $106.42/tonne (financial) and $99.34/tonne (economic). A summary of the average production costs for coffee pulp briquettes is presented in Table 3.11. 3.37 For the purposes of the pilot project, the coffee pulp briquettes are to supplement the output of the Dilla coffee husk briquetting plant. As such, only the additional marginal costs of increasing the output of that plant from 5,000 to 6,000 tonnes per year - 46 - should be assessed against the production of the coffee pulp briquettes. Assuming no changes in plant life, the additional costs are the 0WM costs per additional tonne of briquettes produced. Thus, the marginal costs of producing .1,000 tonnes of coffee pulp briquettes are $58.20/tonne (financial) and $62.26/tonne (economic). The financial and economic delivered costs to Addis Ababa are $92.86 and $82.23 per tonne respectively. These marginal production costs are only 16X and 211 lower than their respective average costs. Table 3.11: FINANCIAL AND ECON041C COSTS OF COFfEE PULP BRICUETlES Finarn˘lal Economic Delivered Ex-Factory Delivered Ex-Factory Average Production Costs: Cost/tonne (SS) 106.42 71.76 99.34 79.17 Cost/40 kg Bag (USS) 4.25 2.87 3.97 3.18 Costai6 (USS) 5.51 3.71 5.14 4.10 Morainal Production Costs: Cost/tonne (USS) 92.86 58.20 82.23 62.06 Cost/40 kg (USS) 3.71 2.33 3.29 2.48 cost/NJ (USS) 4.81 3.01 4.26 3.21 3.38 Given the significantly higher costs of producing coffee pulp briquettes when compared to coffee husks, it is recommended that the coffee pulp briquetting project be implemented only after one year of successful operation of the coffee husk briquetting plant. This will allow for plant operators to become familiar with their equipment and eliminate additional pressure on plant equipment. While production costs for coffee pulp briquettes are significantly higher than coffee husk briquettes, the negative impacts that presently result from disposal of the washed coffee pulp warrant an examination of the potential for recovering and possibly utilizing this waste material. It is proposed that a more detailed cost/benefit analysis associated with recovering and briquetting this material be conducted during the demonstration phase of the pilot project. Pilot Plant for Briguetting of Coffee Parchment: Site Selection 3.39 The washed coffee parchment presents a unique opportunity for the production of fuel briquettes. Presently, all the washed coffee parchment is produced at the washed coffee processing plant in the Mercato area of Addis Ababa. Approximately 2,600 tonnes of parchment are - 47 - available annually from the processing of 13,000 tonnes of export quality washed coffee. 3.40 The maximum capacity of the Mercato plant is 8 tonnes of washed coffee (including parchment)/hour. However, the plant is operated at an average rate of 6 tonnes/hour. At this rate the plant will generate about 1.2 tonnes of parchment/hour. The plant operates between mid- November and mid-May. Operation is mainly geared to export deliveries which makes plant operations rather irregular. The coffee is processed as close to shipping schedules as possible in order to minimize storage time and thereby loss of quality. Therefore, when the plant is working it operates on a 24-hour basis till it meets its shipping target and then stands idle till it needs to meet its next shipping target. Given this intermittent operating scheme, a certain de8ree of storage of parchment will be necessary if the briquetting plant is to have a more even operating load. However, storage space is very limited at the Mercato site. 3.41 It is expected that washed coffee production will increase to 30,000 tonnes annually over the next five years with a resulting parchment production of 6,000 tonnes/year. The increase in washed coffee production cannot be accommodated at the Mercato site as there are space constraints, and the present layout of the plant is not suitable for increasing the processing capacity. Thus, a decision has been made to construct a new plant. The plant is to be located along the Debre Zeit road next to the Kaffa (husk) Cleaning Plant in an industrial area with good road and rail connections both into and out of Addis Ababa. The new plant will have sufficient space so that storage of parchment and briquettes will not be a primary constraint. However, the plant is not e-pected to be constructed until after 1987. All equipment presently in use at the Nercato site will be transferred to the new plant. 3.42 Given the above, it is proposed that briquetting equipment with a capacity of 1.2 tonnes/hour be initially installed at the Mercato plant. This will allow the parchment to be briquetted as it is produced and thereby minimize storage constraints. Given the 1.2 tonnes/hour briquetting capacity and assuming no constraints for parchment storage, no additional briquetting capacity will be necessary when the washed coffee processing plant is moved to its new site and its processing capacity increased to 30,000 tonnes/year. Requirements for storage space could be minimized if the processing of washed coffee were evenly distributed over a longer period during the year than is currently the practice. Plant Design and Equipment 3.43 The pilot coffee parchment briquetting plant presents an excellent opportunity Lo compare the field performance of two types of competing but different briquetting presses: the piston press and the - 48 - conical screw press. The general accessibility of this plant and its location in Addis Ababa will provide the opportunity for close monitoring of the performance of each system. It will also minimize the drawbacks associated with spare part availability and technical expertise necessary to maintain a multiple system. The screw press briquette, as explained in Chapter 2, has a 25 to 30 mm diameter as compared to the 70 mm diameter briquette of the piston press. Therefore, the screw press briquettes resemble thin fuelwood sticks and may be more acceptable as a domestic fuel. The screw press briquettes are also reported by TNO to have better combustion and handling characteristics as a domestic fuel. However, the exact performance and maintenance requirements of the screw press operating in a developing country context has not been definitively established. Also, having both the screw and piston press briquettes available in Addis Ababa would present an excellent opportunity to field study the acceptability of each type of briquette as a domestic fuel. Like the piston press briquettes, the screw press brdquettes can also be used as an industrial fuel with little or no modification to wood combustion equipment. 3.44 Figure 3.5 presents a flow diagram for the proposed pilot coffee parchment briquetting plant. The plant is to be initially located at the Mercato washed coffee processing plant. Coffee parchment exiting from the mills is conveyed pneumatically (1) to a buffer silo (2) which is capable of holding up to 8 hours of pgrchment production or about 9.6 tonnes. Given I bulk density of 50 kg/ma for the parchment, a buffer storage of 192 m is required. This is equivalent to a silo approximately 5 m in diameter and 10 m high. 3.45 From the buffer storage, the parchment is evenly distributed to the surge hoppers of the two briquettors (4 and 5) by means of a screw conveyor (3). The material from the surge hoppers is fed to the briquettors through a vibrating chute in order to avoid bridging of the material. Both the piston press (6) and screw press (7) briquettors are rated at approximately 0.75 tonnes per hour each. Thus, the combined capacity of the two briquettors is sufficient to meet the average output of 1.2 tonnes per hour of parchment from the mill. 3.46 The hot briquettes exiting from the briquettors are pushed through a cooling line (8) of about 4 m. The briquettes are then placed in a cooling heap for about 1 to 2 hours before they are loaded into 40 kg bags with the aid of the bagging station (9). 3.47 Space restrictions at the Mercato plant require that the briquetting plant be incorporated within the existing boundaries. Given that all salvageable equipment at the Mercato plant will be moved in 2 to 3 years, a substantial investment in plant building modifications or additions as a result of installing the briquetting equipment would be wasteful. Therefore, it is proposed that a simple open shed with fenced walls be constructed to temporarily accommodate the briquetting plant. This will also allow for good ventilation and cooling and minimize fire hazards. A more substantial building with parchment and briquetting storage capacity is proposed for the plant once it is moved to the new site. - 49 - Figure 3.5: Row Diagram for Coffee Parchnmt Bdquettlng Plant Ccft Paichrrnt from M-D o~~~~~~~~uf Stm _eSHoo 2 SMWovew>a Piston Press crquewtor 7 Piston Prm &4"Or~d Bnk-000. - so - 3.48 A list of the construction and major plant equipment requirements for the briquetting plant both at the Mercato site and proposed new site is presented in Tables 3.12. In addition to the piston and screw press briquettors, the plant will require pneumatic and mechanical conveyors, a bagging station and a full complement of spares and workshop equipment. 3.49 Power requirements for the briquettors, conveyors, fans, lighting, etc., are estimated to be 170 kW. An average energy input of 60 kWh per tonne of briquettes is assumed. Both the screw press and piston press briquettors have a rating of 75 kW. No electric transformer is required as low voltage power lines are accessible from the plant. Table 3.12: MERCATO COFFEE PARHENT BRIQUETTINO PILOT PLANT: CONSTRUCTION AND EQUIPMENT COSTS Iter Mercato New Site Construction Site Preparation 600 6,900 BuIlding/Housing 3,4C0 50,400 Construction Total 4,000 57,300 EquIpment Piston Briquettors 86,000 Screw Briquettors 63,000 Collection/Storage Equipment - Conveyors 8,300 Silos 25,000 Electrical 6,200 Transformer - Bagging Stations 1,200 Workshop Equlpment 2,400 Other/Miscellaneous 2,500 Elulpment Total 194,600 Characteristics of Coffee Parchment Briquettes 3.50 Coffee parchment briquettes produced from residue samples taken from Ethiopia had a bulk density of approximately 600 kg/mi and a moisture content of 5.36% mcwb. When stored in 90% humidity, the briquettes stabilized at 8.6% mcwb. The average tsh content was 0.33% db with &he ash reaching a softening point at 1,310 C and a fusion point at 19560uC. 3.51 In combustion tests conducted in a woodburning stove, the coffee parchment briquettes were easy to ignite, maintained their integrity during combustion and did not smoke excessively or emit noxious odors. They burned with a cooking efficiency of 28.8% as compared to - 51 - 25.1Z for a typical domestic fuelvood under the same test. Thus, coffee parchment briquettes were determined to have favorable transport, storage and combustion characteristics such that they are technically feasible for fuelwood substitution. More detailed test data on the briquettes are presented in Annex 2. Capital Costa 3.52 The total capital costs for constructing the coffee parchment pilot briquetting plant at the Mercato washed coffee procassing plant are estimated to be $311,750. However, given the constraints discussed earlier, the maximum capacity of this plant is expected to be 2,500 tonnes of parchment briquettes per year. When the plant is moved to the Debre Zeit road location and adequate building and storage capacity added to increase plant capacity to 5,000 tonnes per year, the additional capital costs are estimated at $58,630 for a total capital cost of $370,380. Tables 3.13 and 3.14 present a summary of the capital cost expenditures by year and by foreign and local components for construction of the plant at the other site. Details are presented in Annex 1. The total foreign costs are estimated at $299,610 while local costs are estimated at $12,140 for the plant at the Mercato site. The foreign and local costs of the plant constructed at the new site are $312,150 and $58,230 respectively. Annualized capital costs are calculated based on a discount rate of 101 and are reported in the last column of Tables 3.13 and 3.14. Total annualized capital costs for the coffee parchment briquetting plant are $41,060 at the Mercato site and $47,950 at the new site. This results in a capital charge per tonne of briquettes of $16.42 at the Merc*to site and only $9.59 at the new site. Table 3.13: HERCATO COFFEE PARCHENT BRIQUETTING PILOT PLANT: CAPITAL COSTS (USS) (2,500 tonnes/yr capacity) Total Costs Annualized Charges Local Foreign Total Local Foreign Total Construction 4,000 - 4,000 470 - 470 Equipmnt - 194,600 194,600 - 26,280 26,280 Spares (at delivery) - 19,460 19,460 - 2,620 2,620 Transport and Deivery 1,200 15,500 16,700 160 2,060 2,250 Engineering & Installation 5,840 42,810 48,650 690 5,030 5,710 Contingencies 1,100 27,240 28, 130 3.60 3,730 Total 12,140 299,610 311,750 1,450 39,610 41,060 - 52 - Table 3.14: NEW SITE COFFEE PARCHENT 9RIQTETINB PILOT PLAN: CAPITAL COSTS (USS) (5,000 tonn.s/yr eapacIty) Total Costs Annuallzed Costs Local Foreign Total Local Forelgn Total Construction 45,900 11,400 57,300 5,390 1,340 6,730 Equlpmwit - 194,600 194,600 - 26,280 26,280 spares at delivery - 19,460 19,460 - 2,620 2,620 Transport and delivery 1,200 15,500 16,700 160 2,080 2,250 Engineering and installation 5,840 42,810 48,650 690 5,030 5,710 Contingencies A.290 2880 3670 620 3.740 4.360 TotaI 58,230 312,150 370,380 6,860 41,080 47,950 Annual Operating and Maintenance Costs 3.53 Annual O&M costs for the briquetting plant at the Mercato location are estimated at $53,265 of which only $10,703 are foreign costs. Given a briquette production of 2,500 tonnes per year.. the O&M charge per tonne is $21.31. A summary of the O0tM cost estimate is presented in Table 3-15 with details presented in Annex 1. Table 3.15: NERCATO COfFEE PAROSENT BRIQUETTINS PILOT PLANT: ON COSTS Local Foreign Total --((USS t-- A. Labors 14,000 14,000 Skilled 3,850 3,900 Unskilled B. Power: Electricity 9,450 9,450 Diesel C. Operation & Maintenance: Fuel Lube 01I Spare Parts 9,730 9,730 Services 973 973 D. Building Maintenance Materials 120 120 E. Consumables: Bags Other Operating Costs 38,693 9,730 48,423 ContIngency 3,869 973 4 842 Total Operdting Costs 42,562 10,703 53,265 - 53 - 3.54 The annual 011 costs for the 5,000 tonne per year parchment briquetting plant at the new site are estimated at $108,158. Of this, only $21,406 are foreign costs with the remainder being local costs for labor,.power and materials. Given a briquette production of 5,000 tonnes per year, the 01M charge per tonne is $21.63. A summary of the OM cost estimate is presented in Table 3.16. Table 3.16: NEW SITE COFFEE PARCHMENT BRI9UETFINS PILOT PLANT: OW COSTS Local Foreign Total )(USS) A. Labor: 28,000 28,000 Skilled 7,700 7,700 Unskilled 3. Pottr: Electricity 18,900 18,900 Oiesel C. Operatlon 1 Maintenance: Fuel Lube Oil Spare Parts 19,460 19,460 Services 1,946 1,946 D. Building Maintenance: Materlals 1,719 1,719 E. Consumables: Bags 20,600 20,600 Other Operating Costs 78,865 19,460 98,325 Contingency 7.887 I-Aim 9,833 Total Operating Costs 86,752 21,406 108,158 Production Costs 3.55 Average financial and economic production costs for coffee parchment briquettes were calculated based on the data in Tables 3.13 through 3.16. Production costs were calculated for both a 2,500 tonne per year plant at the Mercato site and a 5,000 tonne per year plant at the proposed new site. Average financial production costs for parchment briquettes produced at the Mercato site are estimated at $37.73 per tonne and at the proposed new site at $31.22. Economic production costs are $43.67 and $34.50 respectively. The financial and economic costs per - 54 - unit of energy for coffee parchment briquettes produced at Mercato are $1.95/1J and $2.25/0J respectively. Similarly, the financial and economic costs per unit of energy for briquettes produced at the new 9ite are $1.62/03 and $1.79/GJ respectively. Since these briquettes are produced in Addis Ababa, only a 10 Birr ($4.80) financial charge and a 4.65 Birr ($2.25) economic charge per tonne were added to the delivered costs of the briquettes. A summary of the ex-factory and delivery costs of the briquettes is presented in Table 3.17. Table 3.17: FINANCIAL AND ECONOMIC COSTS OF PARCHMENT sRIQMuTES Financial Economic Delivered Ex-Factory Delivered Ex-Factory Mercato Plant Cost/tonne (USS) 42.53 37.73 45.77 43.52 C*st/40 kg (USS) 1.70 1.51 1.83 1.74 CostdGJ (USS) 1/ 2.20 1,95 2.37 2.25 Now Site Cost/tonne (USS) 36.02 3122 36.74 34.50 Cost/40 kg (USS) 1.44 1.25 1.47 1.38 Cost9J (USS) 1/ 1.86 1.62 1.90 1.79 a/ 6HN of coffee parchment brIquettes Is 19.33 GJ/tonne. - 55 - IV. PIUOT PRWJECT FOM IRIQUTT$ING OF COTTON RESIDUES Summary and Conclusions 4.1 More than 116,300 tonnes of cotton stalk residues (122 mcwb) are produced annually in Ethiopia. This is equivelent to 109,600 tonnes of fuelwood (152 mcwb) or 45,600 toe per year. All of the cotton stalk residues are located on large State Farms primarily in the Awash Region. At present, the residues are not utilized but are burnt and then ploughed under. Chemical analysis of the residues indicates that they would make an excellent fuelwood substitute. However, their low bulk density (140 kg/m3) requires that they be densified first in order to facilitate transport and handling. A pilot project to collect, store and densify a small fraction of the cotton stalk residues is proposed to demonstrate their potential as a fuel substitute. A farm in the Middle Awash Enterprise located about 300 km from Addis Ababa has been selected as a potential site for a pilot project. 4.2 A summary of the data relating to the proposed pilot project is presented in Table 4.1. The total capital costs for a 5,000 tonne per year production capacity (including collection, storage and briquetting) are estimated at $1,149,360. O&M costs are estimated at $240,420 per year and transport costs at 4159,400 per year, assuming all briquettes are delivered to Addis Ababa. Thus, a total of $1,549,180 is required in capital plus one year's OU& and transport costs in order to finance the cotton stalk pilot briquetting project. Of this total, $309,020 are local costs and $1,240,180 are foreign costs. Table 4.lt SUMMARY DATA ON COTTON STALK RESIOUES BRIQUETING PILOT PLANT Iter Value Total Capital Costs USS1,149,360 Annual Capital Charges S169,890 Annual 08M Costs S240,415 Annual Production Costs 5410,305 Annual Briquette Production 5000 tonnes Briquette Costs, Ex-factory 582.06 Annual Transport Costs $159,400 Briquette Costs, Addis Ababa 5113.94/tonne ˘HV of Briquettes 17.80 GJ/tonne Briquette Price Delivered, Energy Basis. $6.40/GJ - 56 - 4.3 Cotton stalk briquettes can be produced for $82.06/tonne (ex- factory costs). An additional $31.88/tonne is needed to transport the briquettes to Addis Ababa resulting in a delivered cost of $113.94/tonne. On an energy basis, the cost of cotton stalk briquettes delivered to Addis ababa is $6.40/GJ. This is $1.72/CJ more than the cost of fuelwood in Addis Ababa. Taking relative combustion efficiencies into account, the cost of the useful energy from the briquettes used in cookstoves is $25.40/GJ versus $18.60/G for fuelwood. The combustion behavior of the briquettes is reported to be more like that of charcoal which costs $37.50/GJ when used in stoves in Addis Ababa. Also the briquettes can be used as a potential substitute for industrial fuel where the cost of useful energy is $6.75/GJ to 8.51/GJ from fuel oil, $6.43 to $14.69/CJ from charcoal and $2.60 to $9.86/GJ from fuelwood as compared to US$5.76 to US$8.00/GJ for cotton stalk briquettes. 13/ 4.4 It is recommended that the pilot project to briquette cotton stalk residues be undertaken. The resulting cotton stalk briquettes can be used as a charcoal substitute in the domestic sector and as a substitute industrial fuel for industries requiring process heat that presently are using wood, charcoal or fuel oil. Availability of Cotton Residues 4.5 Ethiopia produces approximately 70,000 tonnes of raw cotton annually. 14/ All this cotton is grown by the Ethiopian State Farms with the Awash Corporation producing about 851 of the total and the South Corporation the remainder. State Farms cotton plantations account for 30,600 ha. Thus, the average raw cotton yield per hectare is approximately 2.29 tonnes. 4.6 At present, most of the cotton is harvested manually, but the Ministry of State Farms is evaluating the use of combine harvesting techniques. Harvesting generally begins in early November and is completed by the end of January. Once a field has been completely harvested, usually after two pass-throughs, it is opened to cattle grazing. The cattle consume the leaves and thin green branches leaving the main stalk standing. Grazing is generally complete in two weeks. Finally, around mid-February, the standing stalks are either manually cut or slashed about 10 cm above ground, collected into large heaps in the fields and burnt in order to control pathogen and insect infestation of 13/ Variations in useful energy costs are due to variations on delivered prices of the fuels. 14/ Based on 1980/81 and 1981/82 production reported in Ethiopia: Recent Economic Developments and Future Prospects, Report No. 4683a- ET, May 31, 1984, The World Bank, Washington, D.C. - 57 - the succeeding crop. Before the next crop is planted in May, it is common practice for the State Farm to have a two month closed seasaa, from March to April, during which all possible sources of infestation are eradicated. 4.7 The total cotton residues that are theoretically available are the stalk, leaves, branches and unblossomed bolls. However, given the current practice of grazing, only the main stalk remains as residue after grazing. Despite this, it is not recommended that grazing oe restricted in order to conserve residues. First, the quantity of material removed during grazing is less than 20X of the total dry weight of stalk residue left in the field. Second, the local cattle-owning populace have become accustomed to grazing privileges and depend on it to feed their cattle through the dry season. Restricting their access to the fields would result in bitter relations between the local farmers and the State Farms. Finally, grazing does not damage the main stalk, the most desirable part from an energy standpoint, and in fact helps in eaLng the cutting and nandling of the stalk for collection. 4.8 The cotton stalk residues that are currently burnt or ploughed under on the State Farms are estimated at approximately 116,280 tonmes per year at 121 mcwb 15/ cotton. The cotton stalks have an average moisture content of aboiut 351 mcwb when they are first cut 16/. However, left lying in the field they quickly dry down to approximately 121 mcwb with a resulting gross heating value of 16.8 MJ/kg. Thus, total cotton stalk residues are equivalent on an energy basis to 109,600 tonnes of fuelwood (151 mewb) or 45,600 toe per year. A summary of cotton production and the quantity and energy value of the associated residues is presented in Table 4.2. Table 4?2: COTTON FIBER AND STALK RESIDUES PRODUCTION Cotton Residue Fuelwood Production Production Equivalent (tonnes) (tonnes) (tonnes) TOE Awash Corporation 59,500 98,700 93,200 38,800 South Corporation 17,400 16,400 6,800 Total 70,000 116,100 109,600 45,600 151 Based on an estimate of 3.8 tonnes/ha of residues. 16/ Based on data reported by the chief agronomist of the Awash Enterprise. - 58 - 4.9 Ethiopian soils are relatively high in potash. Therefore, the plowing under of the burnt cotton stalks is not done to replenish the soil. Burning is strictly a pest control procedure. State Farm agronomists indicate that other measures of pest control can be undertaken to permit recovery of the cotton stalks for use as a fuel without adversely affecting soil nutrients. Potential Pilot Projects for Cotton Residues 4.10 A comparison of the chemical and energy analysis of wood and cotton stalk residues is presented in Table 4.3. The data indicate the similarity of cotton stalk residues to wood and therefore it has al excellent potential as a fuelwood substitute. The cotton stalks do have higher ash content, which may be of significance in some high temperature industrial combustion systems where slagging of ash may occur. However, this should not be a problem in most industrial boilers or burners or in domestic wood burning stoves. Table 4.3: COTTON STALKS-CH4EICAL AND ENERGY ANALYSIS Cotton Stalks Eucalyptus Globulus moisture Content (M) a/ 12.0 15.0 Ash Content (%) 3.16 0.29 Volatile Content (S) 75.8 82.2 Fixed Carbun N/A N/A C (S) 47.10 47.46 H (S) 5.99 6.08 0 (M) 43.90 45.82 N (%) 0.35 N/A Sulphuric Lignin. 21.5 N/A HHV MJ/kg 19.05 21.00 GHV iJ/kg 16.76 17,79 a/ Except for moisture content, all measurements are given on a dry basis. 4.11 The bulk density of cotton stalk residues collected in the field is approximately 140 kg/m3 or only 28% that of stacked wood. The cotton producing State Farms in Awash are about 300 km from Addis Ababa and those under the South Corporation are even farther away. The nearest potential major industrial consumer is 100 km away. Thus, economic transport and use of the cotton stalk residues require that the material be densified. - 59 - 4.12 Given the overall potential of cotton stalk residues as a fuelvood substitute, a commercial scale pilot project to collect, process and densify the cotton stalk residues is proposed. A pilot project producing approximately 5,000 tonnes of cotton stalk briquettes a year would provide an excellent opportunity to demonstrate the economic and resource potential of this residue while providing valuable information for future exploitation of the resource. Pilot Plant for Briguetting of Cotton Stalk Residues: Site Selection 4.13 Over 88 percent of the total State Farm cotton production is from the Awash Corporation. The cotton farms of the Awash Corporation are located approximately 300 km from Addis Ababa along the main western radial truck route. The road is paved and in excellent condition. Given these facts, The Ministry of State Farms selected the farms of the Middle Awash Enterprise as potential candidates for the pilot briquetting proJect. 4.14 The Middle Awash Enterprise has three central farms (Melke Were, lelke Sade and Amibara) and two more remote ones (Gewane and Dofan- Bolham). Table 4.4 lists the area planted with cotton, the average annual cotton yield and the the estimated annual cotton stalk residues available at these farms. It is proposed to locate the cotton stalk briquetting plant at one of these main farms. The exact location of the plant should be determined only after an analysis of each farm's management capability, quality of support services, land availability, logistics of transport of field residues to briquetting plant and subsequent transport of the briquettes to the main road. However, any one of the farms has sufficient residues to support the pilot briquetting plant. Table 4,4: COTTON AND RESIOUES PROOUCTION OF T1E MIDDLE AWASH ENTERPRISE (1983/84) Fanm Area Planted Cotton Yield Cotton Residues a (ha) (tonnes/yr) (tonnes/yr) MeI ke Were 3,018 7,237 11,470 telke Sade 2,232 6,643 8,480 Amlbars 2,437 4,747 9,260 Gewane 2,081 5,634 7,910 Dofan-Bolham 1,400 2,873 5.320 Total 11,168 27,134 42,440 a/ Eased on estimate of 3.8 tonnes/ha and residues of 12% mcwb. - 60 - Collection and Storage of Residues 4.15 It is assumed that cotton stalk residues will be obtained only after grasing of the cotton fields. Therefore, collection of the cotton stalk residues could commence in early January and continue until mid- April when field preparation must begin for the new crop. However, entymologists of the Agricultural Research Institute strongly advise against the collection of residues into mid-April as it might result in aggravating boll worm infestation. Given this constraint and considering the farm's requirements for pathogen and insect control, collection of all the cotton stalk residues must be done within a two-month period beginning in early January and ending in early March. This will allow the farm to continue its practice of having a "closed season" for eradication of pests. The two-month residue collection period may later be extended if subsequent investigations provide other alternatives for past eradication. 4.16 For collection of the residues it is proposed to cut the cotton stalks soon after grazing in the same manner as is currently done for open field burning. Since the moisture content of the stalks soon after cutting is far too high for storage or briquetting (approximately 351 mcwb), subsequent field drying is necessary. Field drying will be achieved by laying the stalks in swaths covering 4 to 6 rows and allowing them to dry for approximately 7 to 10 days. This should be adequate to reduce the stalk to 121 mcwb given the hot dry weather during this period. 4.17 Once dried, the stalks will be collected from the swaths and fed to a mobile chopper. The chopper, mounted on a tractor, will blow the chopped stalks into a trailer drawn by the same tractor. When the trailer is loaded it will be disconnected, left on the access road and replaced by an empty trailer. The filled trailers will be transported to the chip storage area of the briquetting plant where the residues will be unloaded and the empty trailers returned to the field. 4.18 It is assumed that a total of 8 tractors, 8 trailers and 5 choppers are required to collect 5,000 tonnes of cotton stalk residues within the allowable two-month window. A list of the types and costs of equipment required to collect the residues is presented in Table 4.5. The basic assumptions used to derive this list are: (a) Residue stalk production of 3,800 kg/ha at 12% mcwb; (b) available time for cutting and collection of residues is 2 months or 55 working days; (c) net working hours per day are 10 hours; (d) chopper capacity of 15 m3/hr or 2,100 kg/hr; - 61 - (e) bulk density of chopped stalks is 140 kg/m3; (f) trailer capacity of 20 m3 or 2,800 kg; and (s) a 151 forced outage rate of equipment. TASLE 4.5: COLLECTION EŁJIPWENT AND COSTS FOR COTTON STALK RESIDUES Item Oost (USS) Tractors (8) 102,000 Trailers (8) 70,600 Choppers (5) 32,200 204,800 4.19 Normal wood chopping equipment is assumed to be adequate. The field dried stalks are to be fed manually to minimize the inclusion of soil particles which will damage the chopping and briquetting equipment and increase the ash content of the briquetted fuel. 4.20 With stalk production of 3.8 tonnes/ha, an area of 1,316 ha net is required to produce 5,000 tonnes of residue. A gross area of 1,550 ha is required, assuming a tare of 15 percent. An average haulage distance of 0.3 km in the field and 1.5 km on the access road is assumed to bring the residues to the briquetting plant storage area. Given this distance, an average of 45 minutes per round trip per haul/tractor including unloading of trailers is assumed. Therefore, each tractor is capable of hauling 1,745 tonnes per season (550 hours/season x 0.15/0.75 hrs/trips x 2.8 tonnes/trip). Each chopper unit is assumed to return at the end of the Jay with a loaded trailer accounting for a ~otal of 770 tonnes per season (5 choppers x 2.8 tonnes/chopper x 55 days/season). Therefore, approximately 3 tractor and trailer units. are required to haul the remaining 4,230 tonnes of material per season. 4.21 A maximum storage requirement of approximately 30,000 m3 equivalent to 4,200 tonnes of chopped residues is necessary. This is based on the assumption of a 5,000 tonne per year briquetting plant operating over a 10 months or a 250 working day period. The plant is assumed to operate duting the residue collection season and therefore only an additional 195 days of operating feedstock is necessary after the collection period is over. - 62 - Igure 4.1: Schemotc Dbgrom of Cotton Stik Resdue Storage Rec*.*ngPf S : >~~~~~~~~~~~~~~~~~hc ,{ft I 1~~~~~~~~~~~~~~~~oct {~~~~~~~~SIJ1h RM¶ Dr RIQ lcKdn ;.cen R Pif SrORAGE 9e C4nv&ia/ Vu1d 9anIc-3MO&6 I I.r - 63 - 4.22 The sheer volume of the storage requirement necessitates open storage. It is proposed that the residue trailers arriving from the field be mechanically off-loaded into one of three receiving pits. From there, the residues are to be transferred by screw conveyors to mobile belt conveyors/stackers. The belt conveyors/stackers will create three semi-circular storage piles on three sides of the proposed briquetting plant. The storage piles should be covered with lightweight tarpaulins upon completion of the residue collection period. The stored residues can then be fed, during operation, to the briquetting plant day storage silos by means of two belt conveyors/stackers. A schematic diagram of the residue storage operation is presented in Figure 4.1. Plant Design and Equipment 4.23 Figure 4.2 presents a flow diagram for the proposed cotton stalk briquetting plant. The chopped cotton stalks are discharged from the day storage heap onto a belt conveyor (1). The material is then sent through a hammermill (2) to reduce the particle size to match the requirements of the briquettor. From the hammermill, the material is passed over a vibrating screen to separate out oversized pieces of material which are sent back to the hammermill. All other material is pneumatically conveyed (3) to the top of the briquetting surge hoppers (4). The cotton stalk residues are then fed to the piston press briquettors (5) by means of a screw conveyor located below the surge hoppers. The hot briquettes leaving the die are pushed through a cooling line (6) and then the discharged briquettes are manually loaded into 40 kg sacks at a bagging station (7). The bags are removed to a temporary storage area capable of holding about 10 days of production. 4.24 The plot plan for the cotton stalk briquetting plant is shown in Figure 4.3. The briquetting plant requires a 40 by 25 meter covered building. The front and side walls of the building should allow for ample ventilation. 4.25 A list of construction and plant equipment costs for the cotton stalk pilot briquetting plant is presented in Table 4.6. Besides the buildings required for the briquetting plant, a total of $140,600 has been allocated for the construction of housing for one plant manager, 18 skilled laborers, 32 unskilled laborers and some seasonal tractor drivers. This labor is exclusively for the collection of residues and production of briquettes. Total construction costs amount to $212,600. 4.26 Two hammermills and two piston press briquettors are required to meet the desired plant capacity of 5,000 tonnes per year. It is proposed that each hammermill and briquettor have a maximum capacity -of 0.75 tonnes per hour. Assuming the briquettors operate full time, the maximum daily capacity is 36 tonnes per day. However, given that the plant is scheduled to operate for only 250 days per year at two shifts per day (16 hours), the average operating capacity of the briquettors - 64 - will be 0.625 tonnes per hour. The addt ional capacity of the briquettors will be beneficial in the event of production stoppages due to equipment failure. Total equipment cost for the collection, storage and briquetting plant is $610,600. 4.27 Maximum power requirements for the plant (briquettors, hammermills, conveyors, fans, lighting, etc.) are estimated at 250 kW. Grid electricity is not available at the cotton farms of the Middle Awash Enterprise. Therefore, power will be supplied by two 125 kW diesel generator sets, each capable of taking the average load of the plant. This is necessary to prevent generator outages from stopping the entire operation of the plant. The average plant energy consumption is estimated at 70 kWh/tonne of briquettes produced. Table 4.6: COTTON STALK ESIDWES BRIQUETTING PILOT PLANT: CONSTRUCTION AND EQUIPMENT COSTS Item Local Foreign Total -~~ (USS )--__-_- Construction Site Preparation 10,000 - 10,000 Buildings 49,600 12,400 62,000 Housing 112,500 28,100 140,600 OonstructIon Total 172,100 40,500 212,600 Equipment Piston Briquettors (2) - 172,000 172,000 Collection Equipment - 204,800 204,800 Storage EquIpment - 17,000 17,000 Conveyors - 23,200 23,200 Screens - 3,000 3,000 Silos - 20,000 20,000 Hammermills - 21,000 21,000 Electrical - 12,200 12,200 Generators - 106,300 106,300 Bagging Stations - 2,400 2,400 Workshop Equ ipaent - 24,200 24,200 Other/lMiscellaneous - _ 4,500 4,500 Equipment Total - 610,600 610,600 Fgure 4± Row Dla am Coion Stalms Piston Briquetting Plant &g3 0 wppe 4 I * BeltChopperdo B.eIstnCPessSriqueto 1oo eHcom ot r 2 FinPWnfAoto Wokdd 8acnk-30067 1. Belt Oonveryor S. Piston Press 8riquettor 2. HNwermill 6. Cooling Line 3. Pneumatic Conveyor 7. sagging Station 4. Surge Hopper - 66 - Figure 4.3: Plot Plan for Cotton Stalks Btquetllng Plant Hommirrll Room 3.6 m High \ _8m , _ 40m 6 m . Pneumatic Cono 8eHCon n yo BR11ETNG PLANT / 26 m Chopped Cotton S WoUl Bank-3M8 - 67 - Characteristics of Cotton Stalk Briquettes 4.28 Cotton stalk briquettes produced from residue yamples taken from Ethiopia had a bulk density of approximately 532 kg/m , a moisture content of 5.88X mcwb, an HHV of 19.05 NJ/kg and a GlV of 17.80 NJ/kg. When stored in 901 humidity, the briquettes stabilized at 10.8Z mcwb. The average ash content was 3.16X db with the ash reaching a softening point at 7200C and a fusion point at 1,5600C. 4.29 In combustion tests conducted in a woodburning stove, the cotton stalk briquettes were easy to ignite, maintained their integrity during combustion and did not smoke excessively or emit noxious odors. They burned with a cooking efficiency of 25.2Z as compared to 25.1X for a tpical domestic fuelwood under the same test. Thus, cotton stalk briquettes were determined to have favorable transport, storage and combustion characteristics such that they are technically feasible for fuelwood substitution. More detailed test data on the briquettes are presented in Annex 2. Table 4.7: COTTON STALK RESIDUES BRIQUETTINS PILOT PLANT: CAPITAL COSTS CUSS) Total Costs Annualized Costs Local Foreign Total Local Foreign Total Construction 172,100 40,500 4,000 20,220 4.760 24,980 Equipment - 610,600 610,600 - 99,640 99,640 Spares (at delivery) - 61,060 61,060 - 9,500 9,500 Transport and Delivery 1,000 37,680 38,680 160 5,860 6,020 Engineering & InstallatIon 12,170 109,760 121,930 1,430 12,890 14,320 Contingencies 18,530 -85960 104,490 2,180 13,270 15,450 Total 203,800 945,560 1,149,360 23,980 145,910 169,890 Note: A more dotalled breakdown of this costing Is presented In Annex 2. Capital Costs 4.30 The total capital costs of the cotton stalk residues collection, storage and briquetting project are $1,149,360. A summary of this estimate by foreign and local components is presented in Table 4.7. The foreign costs are estimated at $945,560 while local costs are estimated at $203,800. The capital costs were annualized over the plant or equipment life at a 101 discount rate. The annualized capital costs are reported in the last column of Table 4.7. Total annualized capital - 68 - costs are estimated at $169,890. Assuming an annual production of S,OOO tonnes of briquettes, the capital charge per tonne of briquettes is $33.98/tonne. Details on the capital cost estimate are presented in Annex 1. Annual Operating and Maintenance Costs 4.31 The annual O&M costs for a 5,000 tonne per year cotton stalk briquetting plant are estimated at $240,415 per year. A summary of this estimate is presented in Table 4.8. Of the total, local costs are $97,222 and foreign costs $143,193. The major annual local costs are $54,300 for labor and $20,600 for bags. The main annual foreign costs are $61,060 for spare parts and US$54,604 for diesel generated elec- tricity. Details of the annual OM cost estimate are presented in Annex 1. Production Costs 4.32 Average production costs for cotton stalk briquettes were calculated based on the capital and O0M cost estimates presented in Tables 4.7 and 4.8. Average financial costs for production of cotton stalk briquettes are $82.06/tonne while economic costo are $95.45/tonne. For cotton stalk briquette. delivered to Addis Ababa, the transportation costs will add $31.88/tonne to the financial costs and $17.17/tonne to the economic costs. Thus, the delivered financial costs of cotton stalk briquettes in Addis Ababa are $113.94/tonne while the economic costs are $113.16/tonne. The financial and economic costs per unit of energy for delivered cotton stalk briquettes are $6.40/GJ ahd $6.36/GJ respectively. A summary of the ex-factory and delivered costs of cotton stalk briquettes is presented in Table 4.9. - 69 - Table 4.8: COTTON STALK RESI0UES BRIQUETTING PILOT PLANT: &NM COSTS AND BRIQUETTE TRANSPORT COSTS Item Local Foreign Total - tUSS) -- -- - A. Labor: Skilled 40,400 40,400 Unskilled 13,900 13,900 B. Power: Electricity Diesel 54,604 54,604 C. Operation & Maintenance: Fuel 8,228 8,228 Lube Oil 6,283 6,283 Spare parts 61,060 61,060 ServIces 6,106 6,106 D. Bullding Maintenance: Materials 6,378 6,378 E. Consumables: ebgs 20,600 20,600 Other 1,000 1,000 OperatIng Costs 88,384 130,175 218,559 ContIngency 8.838 13,018 21,856 Total Operating Costs 97,222 143,193 240,415 Transport 7,970 151,430 159,400 a/ a/ 95% of transport costs are assumed to. be foreign costs for truck depreciation spare parts and fuel. Table 4.9: FINANCIAL AND ECONOMIC COSTS FOR COTTON STALK RESIDUES BRIQUETTES Financial Economic DelIvered Ex-Factory Delivered Ex-Factory Cost/tonne (USS) 113.94 82.06 113.18 95.45 Cost/40 kg Sag (USS) 4.56 3.28 4.53 3.82 Cost/AJ (USS) 6.40 4.61 6.36 5.36 - 70 - V. PLW? PROJECT FMR BUQUETIEG Or HESAT RESIDURS Summary and Conclusions 5.1 More than 700,000 tonnes of wheat straw residues (101 mcwb) are produced annually in Ethiopia. This is equivalent to 693,200 tonnes of fuelwood (151 mcwb) or 288,500 toe per year. Of the total wheat straw residues, 123,000 tonnes or 17.61 are located on large State Farms primarily in the south. At present, these residues are not utilized but are burned and/or ploughed under. Chemical analysis of the wheat straw residues indicates that they would make an excellezt fuelwood substitute. However, the low bulk density of baled wheat straw (90 kg/m ) requires that they be densified first in order to facilitate transport and handling. A pilot project to collect, store and densify a small fraction of the wheat straw residues is proposed to demonstrate its potential as a fuel substitute. Dixis, a farm under the Arrusi Enterprise located about 220 km from Addis Ababa, has been selected as a potential site for a pilot project. 5.2 A summary of the data relating to the proposed pilot project is presented in Table 5.1. The total capital costs for a 5,000 tonne per year production capacity (including collection, storage and briquetting) are estimated at $1,111,400. O&M costs are estimated at $245,910 per year and transport costs at $170,050 per year, assuming all briquettes are delivered to Addis Ababa. Thus, a total of $1,527,360 is required in capital plus one year's O&M and transport costs in order to finance the cotton stalk pilot briquetting project. Of this total, $327,430 is local costs and $1,199,930 is foreign costs. Table 5.1: SUIMARY DATA ON W*EAT STRAW RESIOUES 8RI UETTINO PILOT PLANT item Value Total Capital Costs S1,111,400 Annual Capital Charges S157,400 Annual OIM Costs 5245,910 Annual Production Costs S403,321 Annual Briquette Production 5,000 tonnes Briquette Custs, Ex-Factory S80.66/tonne Annual Transport Costs S170,050 Briquette Costs, Addis Ababa $114.67/tonne 6iHV of Briquettes Te.05/GJ/tonne Briquette Price, Energy Basis $6.35/GJ - 71 - 5.3 Wheat straw briquettes can be produced for $80.66/tonne (ex- factory costs). An additional $34.01/tonne is needed to transport the briquettes to Addis Ababa resulting in a delivered cost of $114.67/tonne. On an energy basis, the cost of wheat straw briquettes delivered to Addis Ababa is $6.35/GJ. This is $1.67/GJ more than the cost of fuelwood in Addis Ababa. Taking relative combustion efficiencies into account, the cost of the useful energy from the briquettes used in cook stoves is $23.51/CJ versus $18.60/GJ for fuelwood. The combustion behavior of the briquettes is reported to be more like that of charcoal which costs $37.50/GJ when burned ir stoves ;n Addis Ababa. Also the briquettes can be used as a potential substitute for industrial fuel where the cost of useful energy from t! briquettes ranges from $5.59 to $7.94/GJ as compared with $6.75 tL $8.51/GJ for fuel oil, $6.43 to $14.69/GJ for charcoal and $2.60 to $9.86/GJ for fuelwood. 17/ 5.4 It is recommended that the pilot project to briquette wheat straw residues be undertaken. The resulting wheat straw briquettes can be used as a charcoal substitute in the domestic sector and as a substitute industrial fuel for industries requiring process heat that presently are using wood, charcoal or fuel oil. Availability of Wheat Residues 5.5 Recorded wheat production in Ethiopia for 1981/82 season was 701,460 tonnes. The recorded area under growth was 648,000 hectares for an average yield of 1.08 tonnes of wheat per hectare. 18/ The State Farms accounted for 123,610 tonnes or about 18X of the total production. Average State Farm yields were 1.3 tonnes/ha which were slightly higher than the national average. 5.6 Harvesting of wheat is accomplished by combines on most State Farms. Harvesting generally starts in early November and lasts for two months. After harvesting, where accessible, local villagers enter the fields to pick up lost grain. The remaining wheat straw and chaff is then either baled for use as animal feed (a practice that has been increasing recently due to the drought), ploughed under or burnt (in the case of infested crops) along with the stubble in order to prepare the fields for replanting. In some cases, local cattle are allowed to enter the fields to forage prior to burning. However, if the straw is to be baled, the cattle are kept off the field until after baling in order to prevent the straw from being trampled. Once the fields have been cleared 17/ Variations in useful energy costs are due to variations in delivered prices of the fuels. 18/ Ethiopia: Recent Economic Development and Future Prospects, Report No. 4683a-ET, May 31, 1984, The World Bank, Washington, D.C. - 72 - of the residues they are allowed to sit fallow until early May when preparation for replanting begins. Planting of the wheat generally commences in early June and continues into July depending on weather and other related conditions. 5.7 The total aboveground residues available after wheat harvests include the straw, stubble, chaff and lost grain. The actual quantity of* this material is dependent on several factors including plant variety, soil conditions, fertilizers, meteorological conditions, and harvesting techniques. As a result, total or gross wheat residue production varies considerably from about 1.5 to above 2.5 times the weight of the grain yield. The recoverable (i.e., through mechanical means) or net residues are generally only about 50 percent of the total available residues. Barnard and Kristoferson estimate wheat residue production to be in the range of 1.0 to 1.8 tonnes per tonne of grain. 19/ Newcombe estimates the "technically accessible" or net recoverable residues in the case of Ethiopian State Farms to be 0.82 of the weight of the harvested grain. 20/ This translates to a net residue yield of 1.07 tonnes/ha given the average State Farms wheat yield of 1.3 tonnes/ha. Information provided during visits to a few State rarms during the 1984/85 harvesting season indicated a net residue production of 1.25 tonnes/ha, equivalent to a net residue to grain yield of 0.96. Recorded data obtained from the Dixis State Farm indicates an average net residue production of 2.1 tonnes/ha. However, grain yields at Dixis were also above average, reported at approximately 2.0 tonnes/ha, resulting in a net residue to grain yield factor of 1.05 which is consistent with the previously reported data. 5.8 Assuming a net residue to grain yield factor of 1.0 for Ethiopia, the available wheat residues can be estimated at approximately 700,000 tonnes based on 1981/82 production data. Wheat stalks have an average moisture content of between 15 and 201 mcwb when the grain is initially harvested. However, the stalk residues quickly dry down to below 10% mcwb within 5 to 7 days with a resulting gross energy heating value (CHV) of 17.58 NJ/kg. Thus, the annual available or net wheat straw residues are potentially equivalent on an energy basis to 693,200 tonnes of fuelwood (15X MCWB) or 288,500 toe. Just the net residues from the State Farms alone are equivalent to 122,150 tonnes of fuelwood or 50,840 toe. A summary of Ethiopian wheat production and the potentially available residues is presented in Table 5.2. 19/ Barnard and Kristoferson, Agricultural Residues as Fuel in the Third World, Earthscan, 1985. 20/ The Commercial Potential of Agricultural Crops: Case Studies on Cereals, Coffee, Cotton and Coconut Crops. Ken Newcombe, October 1984, Energy Department, The World Bank, Washington, D. C. - 73 - Table 5.2: WHEAT GRAIN AND STRAW RESIDUES PRODUCTION Wheat Grain Straw Residue a/ Fuelwood Equivalent (tonnes) (tonnes) (tonnes) TOE State Farms 123,610 123,610 122,150 50,843 Other 577.850 577.850 571,029 237,683 Total 701,460 701,460 693,179 288,526 a/ Assuming 50% retentlon for soil fertlilty. Potential Pilot Projects for Wheat Residues 5.9 A listing of the chemical and energy analysis of Ethiopian wheat straw residues and wood (Eucalyptus Globulus) is presented in Table 5-3. The data indicate that the wheat straw is chemically similar in characteristics to the wood with the exception that it has a significanitly higher ash content. The wheat straw is combustible and can be used as a fueltood substitute. However, the higher ash content may be of significance in some high temperature industrial combustion systems where the slagging of ash may occur. The ash should not be a problem when the wheat straw is used in domestic wood burning stoves or most industrial boilers or burners. 5.10 The bulk density of the wheat straw baled in the fields is about 90 kg/m3 or only 18 percent that of stacked fuelwood. The wheat producing State Farms are in the southern region of Ethiopia, 200 to 300 km away from Addis Ababa. Potential industrial consumers are 90 to 360 km away. Transport costs of baled wheat straw would add about $0.48/tonne/km as opposed to $0.15/tonne/km for briquettes or fuelwood. 21/ For a 300 km trip the transport cost of baled wheat would be $144/tonne as opposed to $45/tonne for briquettes or fuelwood. Thus densifying the wheat straw would significantly reduce transport costs and increase the economic radius in which the fuel could be competitive. It would simultaneousl; reduce handling and storage costs for users. 21/ Transport cost estimate based on truck tariffs of $2.18/vh/km on all-weather roads and $6.16/vh/km on unpaved roads and assuming a 70/30 split between paved and unpaved roads respectively. Trucks transporting baled wheat straw are capable of hauling 500 bales at 14kg/bale or 7.0 tonnes of wheat straw. Trucks transporting briquettes or fuelwood are capable of hauling 22 tonnes. - 74 - 5.11 Given the potential of wheat straw residues as a fuel substitute, a commercial scale pilot project to collect, process and densify the residues available at one of the State Farms is proposed. A pilot plant producing approximately 5,000 tonnes of wheat straw briquettes a year would provide an excellent opportunity to demonstrate the economic and resource potential of these residues while providing valuable informaticn for future exploitation of the resource at other State Farms and large private farms. Table 5.3: WHEAT STRAW - CHEMICAL AND ENERGY ANALYSIS Cotton Stalks Eucalyptus Globulus Moisture Content (5) a/ 10.0 lS.0 Ash Content (5) 4.5 0.29 Volatile Content (S) 78.1 82.2 Fixed Carbon (5) N/A N/A C (%) 47.5 47.46 H (S) 6.0 6.08 0 (5) 42.6 45.82 N (5) 0.42 N/A Sulfuric Lignin (5) 17.6 N/A i4V MJ/kg 19.50 21.00 GH NJ/kg 17.58 11.79 a/ Except for moisture content, all easurenents are given on a dry basis. Pilot Plant for Briguetting of Wheat Straw Residues: Site Selection 5.12 All of the State Farms' wheat production originates from the Southern Corporation which has three regional enterprises: Arussi, Bale, and Sidamo. The farms of the Arussi enterprise are closest to Addis Ababa, in the 200 to 350 km range. Those of Bale and Sidemo are considerably farther to the south. Therefore the Arussi enterprise was selected for closer evaluation. This enterprise has five separate farms with a total of 33,000 hectares of land under wheat and barley cultivation. The hectarage of each of the five farms is presented in Table 54. Out of these five the Dixis farm was selected as a prime candidate for a commercial scale pilot wheat straw briquetting plant. - 75 - Table 5.4: MHEAT AND WHEAT STALK ESIMUES PRODUCTION OF THE ARUSSI ENTERPRISE (1963/84) Wheat Stalk Fuelvood Area planted Grain Yield Residues Equivalent TOE In Wheat (tonnes/yr) (tonnes/yr) (tonnes) (tonnei) Dlxis 6,781 8,366 7,120 7,032 2,892 Gerodells 10,350 17,479 10,868 10,734 4,470 5.13 The Dixis farm is one of the largest (10,000 ha) of the five farms in the Arussi enterprise. It is also relatively close, 220 km,from Addis Ababa with only 70 km of unpaved road. The other large farm of the Arussi enterprise, Geredella, is almost 350 km from Addis Ababa with over 120 km of rough dirt road. Finally, baling of wheat straw is currently practiced at the Dixis farm in order to provide the wheat straw for animal feed to the Animal and Meat Corporation. Therefore, the farm management and personnel are familiar with baling equipment, techniques and requiremvnts. At present the Dixis farm receives about 10 Ethiopian cents per 1aale (equivalent to $7.14/tonne of wheat straw) taken by the Animal aud Meat Corporation. 5.14 The Dixie farm has a gross area of 10,755 hectares. Maximum usage was attained in 1980 when 10,000 ha were cropped. About 3,000 ha still have a severe water logging problem which results in low grain yields. Thus, plantings in the 1984/85 season amounted to only 6,000 ha with a grain yield range of 0.5 to 2.5 tonnes/ha and an average yield of 1.8 tonnes/ha. Plantings for the 1985/86 season are expected to be 7,300 ha at best, which should result in a small increase in the average yield. 5.15 The Dixis farm has a theoretical net residue availability of approximaately 14,600 tonnes of wheat straw, almost three times the requirements of the proposed pilot briquetting plant. Given this fact, along with its relative accessibility and its management and personnel experience, the Dixis farm was selected as the best site for the pilot wheat straw briquetting plant. Collection and Storage of Residues 5.16 Collection of the wheat straw residues for delivery to the pilot briquetting plant can be done in almost the same manner as is currently done for use of the wheat straw as an animal feed. Presently, all harvesting is carried out by combine harvesters. Harvesting starts at the end of October or early November and continues for about two months. In general, harvesting is completed by early January. The cutting height of the combines varies between 0.15 and 0.65 meters with a - 76 - mean height of 0.45 meters. The variation in cutting height is a result of the varying conditions of fields with regard to the presence of rocks and boulders that are potentially damaging to the combines and other field equipment. A concerted effort to rid fields of large rocks and boulders would have the double benefit of reducing damage to ploughing, planting and harvesting equipment while increasing net or recoverable straw yields. 5,17 At present, the recoverable straw is picked up and baled by means of mechanized square balers (John Deere 342A) producing bales of about 14 to 15 kgs with dimensions of 1.00 X 0.46 x 0.36 meters. The bales, which are ejected from the baler as it moves down a swath, are collected manually and placed in large piles at the perimeter of the field. The bales are then manually loaded onto trailers for shipment to the Animal and Meat Corporation. 5.18 To collect the wheat straw for briquetting, it is recommended that field stacks of about 240 bales be constructed at strategic locations in the field. Such stacks will match the load capacity of the trailers and help reduce carrying distances. It is proposed that balers as well as the trailers be drawn by 40 kW - 2WD tractors. Loading and off-loading of the trailers will continue to be done manually. Once filled, the trailers will be hauled to the storage site of the briquetting plant. 5.19 It is proposed that all baling operations be completed by the end of January so as not to disrupt normal field preparation activities. A total of 80 days with an average of 10 net working hours per day are assumed available for residue collection. Therefore, an average of 6.25 tonnes of wheat straw must be baled, collected and delivered per working hour. A list of the equipment required to collect 5,000 tonnes of wheat straw residues within the given period is presented in Table 5.5. The basic assumptions used to derive this list are: (a) Average wheat straw production of 2,100 kg/ha at 10X mcwb; (b) Available time for collection of residues is 3 months or 80 net working days at 10 net hours per day; (c) Average weight of bales is 14 kg/bale; (d) Average baler capacity is 200 bales/hr; (e) Trailer capacity is 240 bales/load; and (f) A 151 forced outage rate of equipment. - 77 - Table 5O5: COLLECTION EQUIPMENT AND COSTS FOR W)EAT STRAW RESIDUES Iten Costs (USS) Belers (3) 19,500 Trailers (5) 17,600 Wheel Tractors (6) 70.400 107,500 5.20 A baler operating 80 days per season with a given 15 downtime can produce 136,000 bales/season. For 5,000 tonnes of wheat straw, 360,000 bales are required, or 2.65 balers. Thr;s, at least 3 balers are required to meet the needs of the briquetting plant. Field transport of the bales will require at least 3 trailers. This is based on the assumption that each trailer has a capacity of 240 bales, requires 1 hour per cycle of loading and unloading and has a downtime of 151. 5.21 With a wheat straw yield of 2.10 tonnes/ha, an area of about 2,380 ha net is required to produce 5,00i tonnes of residue. A gross area of 2,800 is required assuming a tare of 152. An average haulage distance of 0.3 km in the field and 2.7 km on access roads is assumed to bring the residues to the briquetting plant storage area. Given this distance, an average of one hour per round trip per haul/tractor is assumed including hooking and unhooking of the trailers both in the field and at the storage site. Trailer capacity is 240 bales. With a 151 downtime, each tractor can transport 2,040 bales per day or 163,200 bales per season. With a required production of 360,000 bales per season, three tractors are required. Assuming that an unhooked trailer is always being loaded in the field and one is being unloaded at the storage site, a total of 5 trailers are needed. 5.22 The briquetting plant is assumed to operate 250 days over a 10 month period. With a collection period of three months or 80 working days, 301 of the wheat straw can be processed directly while 701 must be stoted for processing at a later stage. This is equivalent to 3,500 tonnes or 250,000 bales of wheat straw that must be stored. Given the average volume of each bale is 0.166 m3, the total storage space required is 41,400 m . Rg 51: Raow Dkam for Wheat Redo b*n rg Pkont Pneurmotc Corwvo 5 Fan CY I *_ Bet CovyrX ug Op * ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~ dCg 1ONo 1F)1 J~~~~ 3 Leaend 1. soft Conveyor 6. Surge Hopp 2. Bale B3reaker 7. Piston Briquettor 3. HammerIll 8 Screw Conveyor 4. Vibrating Screen 9. Coollng Line 5. Pneumatic Conveyor 10. Bagging Station - 79 - 5.23 The sheer volume of the storage requirement necessitates open storage. It is proposed that about 20 storage piles each approximately 7.2 m high (20 stacked bales), 14.6 m wid. (10 bales in alterating sequence) and 20 m long be created in the area around the briquetting plant. Each pile should be selarated by 5 m to create a fire break. Thus a total area of 10,000 m is required for storage spar^. The trailers arriving from the field are to be manually off-loaded and stacked. The storage piles should be covered with lightweight tarpaulins upon completion. During the collection period some bales can be directly delivered to the plant to meet its processing needs. After the collection period the bales can be transported to the plant from the storage piles by use of the field trailers. Plant Design and Equipment 5.24 Figure 5.1 presents a flow diagram of the proposed wheat straw briquetting plant. The baled residues are fed onto an in-plant belt conveyor (1) as they arrive from the fields, the outdoor storage piles or from the indoor day storage area. The indoor day storage area should be sufficient to store about one day's worth of material or about 20 tonnes (about 1,430 bales). This would require a storage space of about 60 m given a storage height of 4 m. The indoor belt conveyor will feed the bales to one of two bale breakers (2). Strings or wire must be removed from the bales manually prior to feeding the bale breaker in order to prevent choking of the bale breaker by the twine wrapping around the shaft. The bale breatrer is provided with a dust hood which is connected to a pneumatic suction pipe. Next, the residuas are discharged onto the floor and are manually fed to the hammermills (3). The hammermills reduce the particle size of the wheat straw to match the the requirements of the briquettor. The milled material is then passed over a vibrating screen (4) to separate out pieces of oversized material which are sent back to the hammermills. All other material is then pneumatically conveyed (5) to the top of the briquetting surge hoppers (6). The wheat straw residues are then fed to one of two piston press briquettors (7) by means of a screw conveyor (8) located below the surge hopper. The hot briquettes leaving the die are pushed through a cooling line (9) of about 10 m. The discharged briquettes are manually loaded into 40 kg bags at a bagging station (10). The filled bags are then removed to a temporary storage capable of holding about 10 days of production, prior to being transported. The plot plan for the wheat straw briquetting plant is shown in Figure 5.2. 5.25 A list of construction and plant equipment costs for the wheat straw pilot briquetting plant is presented in Table 5.6. Besides the building required for the briquetting plant, a total of $143,400 has been allocated for the construction of housing for one plant manager, 16 skilled laborers and 32 unskilled laborers and some seasonal tractor drivers. This labor is required exclusively for the collection of the wheat straw and the production of the briquettes. Total construction costs including the briquetting plAnt are $231,000. E C _____________F- I I.--  I - a 2 E I 0 I I--- EC I a A I I - -I- E Ł I E 0 bb I' - 81 - Table 5.6: WHEAT STRAW RŁSIDUES BRIQUETrING PILOT PLANT: CONSTRUCTION AND EQUIPMENT COSTS Item Local Foreign Total --_---CUSS)-- Construction Site Preporotlon 10,000 - 10,QOO Buildings 62,200 15,400 77,600 Housing 115,000 28.400 143,400 Construction Total 187,200 43,600 231,000 Equipment Piston Briquettors (2) - 215,000 215,000 Collection Equipment - 98,500 98,500 Storage Equipment - 9,000 9,000 Conveyors - 23,500 23,500 Scraens - 3,000 3,000 Silos - 20,000 20,000 Bale Breaker and Ha mermilis - 40,000 40,000 Electrical - 12,200 12,200 Generators - 106,300 106,300 sagging Stations - 2,400 2,400 Workshop Equipment - 24,200 24,200 OtheriMI scel I aneous - 5.000 s,oco Equipment Total - 559,100 559,100 5.26 Two piston press briquettors are required to meet the desired plant capacity of 5,000 tonnes per year. It is proposed that each briquettor have a maximum capacity of 0.75 tonnes per hour. Assuming the briquettors operate full time, the maximum daily capacity is 36 tonnes per day. However, given that the plant is scheduled to operate for only 250 days per year at two shifts per day (16 hours), the average operating capacity of the briquettors will be 0.625 tonnes per hour. The addi'ionsl capacity of the briquettors will be beneficial in the event of prod'tat on stoppages due to equipment failure. 5.27 Maximum power requirements for the plant (briquettors, hammermills, conveyors, fans, lighting, etc.) are estimated at 300 kW. Grid electricity is not available at the Dixis wheat farms. Therefore, power will be supplied by two 150 kW diesel generator sets, each capable of taking the full average load of the plant. This is necessary to prevent generator outages from stopping the entire operation of the plant. The average plant energy consumption is estimated at 85 kwh/tonne of briquettes produced. - 82 - Characteristics of Wheat Stalk Briquettes 5.28 Wheat straw briquettes produced from residue samp31es taken from Ethiopia had a bulk density of approximately 550 kg/m , a moisture content of 7.60X mcwb, an HEV of 19.50 MJ/kg and a GHV of 18.05 MJ/kg. When stored in 90% humidity, the briquettes stabilized at 10.0X mcwb. The average ash content was 4.50X db with the ash reaching a softening point at 1,110°C and a fusion point at 1,4100C. 5.29 In combustion tests conducted in a woodburning stove, the wheat straw briquettes were easy to ignite, maintained their integrity during combustion and did not smoke excessively or emit noxious odors. They burned with a cooking efficiency of 27X as compared to 251 for a typical domestic fuelvood under the same test. Thus, wheat straw briquettes were determined to have favorable transport, storage and combustion characteristics such that they are technically feasible for fuelwood substitution. More detailed test data on the briquettes are presented in Annex 3. Capital Costs 5.30 The total capital costs of the wheat straw residue collection, storage and briquetting project are $1,111,40C; An ;temization of these costs by foreign and local components is prehented in Table 5.7. The foreign costs are estimated at $222,990 while local costs are estimated at $888,410. The capital costs were annualized over the plant or equipment life at a 102 discount rate. The annualized capital costs are reported in the last column of Table 5.7. Total annualized capital costs are estimated at $157,300. Assuming an annual production of 5,000 tonnes of briquettes, the capital charge per tonne of briquettes is $31.48/ tonne. Details on the capital cost estimate are presented in Annex 1. Annual Operating and Maintenance Costs 5.31 The annual O&M costs for a 5000 tonne per year wheat straw briquetting plant are estimated at $245,916 per year. Details of this estimate are presented in Table 5.8. Of the total, local costs are $95,943 and foreign costs $149,973. The major annual local costs are $40,600 for labor and $20,600 for bags. The main annual foreiga costs are $66,200 for diesel generated electricity, $55,910 for spare parts and $7,000 for diesel fuel for collection of the residues. Details of the annual O0M cost estimate are presented in Annex 1. - 83 - Table 5.7: VIEAT STRAW RESIDUES DRIOUETTINB PILOT PLANT: CAPITAL COSTS (USS) Total Costs Annualized Costs Local Foreign Total Local Foreign Total Construction 187,200 43,800 231,000 21,990 5,150 27,140 Equipment - 559,600 559,100 - 87,100 87,100 Spares (at delivery) - 55,910 55,910 - 8,230 8,230 Transport and Delivery 1,700 37,650 39,350 250 5,610 5,860 Engineering & Installatlon 13,820 111,180 125,000 1,620 13,060 14,660 Contingencies 20,2tO 80,760 101.040 2.390 11,920 14,310 Total 222,990 888,410 1,111,400 26,250 131,170 157,432 Table 5.8: W1EAT STRAW RESIDUES BRIQUETriNg PILOT PLANT: O0N AND BRIQUEITE TRANSPORT COSTS Item Local Foreign Total (USb -- - A. Labor: 33,400 33,400 SkilIled 7,200 7,200 Unskilled B. Power: Electricity Diesel 66,198 66,198 C. Operation & Maintenance: Fuel 6,919 6,919 Lube Oil 7,312 7,312 Spare Parts 55,910 55,910 Services 5,591 5,591 D. Building Maintenance: Materials (3%) 6,930 6,930 E. Consumables: Bags 20,600 20,600 Other 13,500 13,500 Operating Costs 87,221 136,339 223,560 Contingency 8.722 13,634 22,356 Total Operating Costs 95,943 149,973 245,916 Transport Costs 8,500 161,550 170,050 a/ a/ 95S of transport costs are assumed to be foreign costs for truck depreciation, spare parts and fuel. - 84 - Production Costs 5.32 Average production costs for wheat straw briquettes were calculated based on the capital and O&M cost estimates presented in Tables 5.7 and 5.8. Average financial costs of production for wheat straw briquettes are $80.66/tonne while economic costs are $93.S8/tonne. For wheat straw briquettes delivered to Addis Ababa, the transportation costs will add $34.01/tonne to the financial costs and $12.28/tonne to the economic costs. Thus, the delivered financial costs of wheat straw briquettes in Addis Ababa are $114.67/tonne while the economic costs are $105.86/tonne. The financial and economic costs per unit of energy for delivered wheat straw briquettes are $6.35/CJ and $5.87/GJ respectively. A summary of the es-factory and delivered costs of cotton stalk briquettes is presented in Table 5.9. Table 5.9: FINANCIAL AND ECONOMIC COSTS FOR WHEAT STRAW BRIQUEllES Financial Economic Delivered Ex-factory Delivered Ex-factory Cost/Tonne (USS) 114.67 80.66 105.86 93.58 Oost/40 kg Bag (USS) 4.59 3.23 4.23 3.74 Cost/16J (USS) 6.35 4.47 5.87 S.18 - 85 - VI. PILOT PROJECT FOR BiUQUETTrIG OF MAIZE RESIDUES Summary and Conclusions 6.1 Nationally, 2,270,500 tonnes of maize residues are produced annually. This amounts to an equivalent of 2,233,500 tonnes of fuelwood or 930,200 toe. On State Farms alone, 174,800 tonnes of maize residues (stalks, husks and cobs) are produced, which is equivalent to 172,000 tonnes of fuelwood (15X mcwb) or 721,000 toe per year. In the Wollega Agricultural Development Enterprise, there is no alternative use for this biomass and on the Anger State Farm alone, four times the quantity of residues ia available that would be needed for a pilot briquetting plant. A summary of the data on maize residues and briquetting costs is presented in Table 6.1. Table 6.1: SUMMARY DATA ON MAIZE RESIDUES BRIQUETTING P' OT PLANT: ANGER STATE FARM item Value Residues available 21,900 tonnes Total capital costs USS1,464,191 Annual capital charges $214,010 Annual operating costs $328,290 Annual production costs $542,300 Annual production 5,000 tonnes Briquette cost, ex-factory $108.46/tonne Annual transport costs $159,400 Briquette cost, Addis Ababa $140.34/tonne Gross heating value (briquettes) 18.65 NJ/Kg Briquette price, energy basis S7.52/6J To construct and install a maize residue pilot briquetting plant will require approximately US$1,464,000 for capital costs and US$488,000 in first year operating, maintenance and transportation costs, for a total investment of US$1,952,000. Of this, US$1,498,000 are foreign costs and US$454,000 are local costs. Because of high field collection, processing and maintenance costs, briquettes produced by this project have the highest ex-factory cost (over US$108 per tonne) of all residues reviewed by this study. When delivered to Addis Ababa, the total energy price is US$7.52/CJ. This is also higher than the other residues but still lower than charcoal which costs the urban consumer U3$15.00/GJ. For this reason, the corn stalk briquetting project should be given the lowest pricrity when establishing the order of implementation for the pilot briquetting projects. - 86 - Availability of Maize Residues 6.2 Maize is one of the key staple crops grown on large-scale plantations managed by the Ministry of State Farms. An estimated 138,000 tonnes of maize are produced annually on 46,000 hectares of State Farms. Over half of this hectarage is contained within the Wollega Agricultural Development Enterprise which is located 300-400 km from Addis Ababa. Approximately 28,300 ha or about 80Z of the enterprise's 36,220 productive hectares are devoted to maize, with the remainder usetd for sorghum, soybeans and red peppers. Based on an average maise yield of three tonnes per hectare, approximately 101,880 tonnes of residue are available throughout the Follega Snterp:ise in the form of stalks (3.0 tonnes/ha) and cobs (0.6 tonnes/ha). A summary of maize production and the quantity and energy value of related residues is present in Table 6.2. Table 6.2: MAIZE AND RESIDUES PROOUCTION Residue: Residue: Residue; Fuelwood Production Total a/ stalks cobs Equlvalent bl Hectares (tonnes) (tonnes) (tones.) (tonnes) (m) ) TOE c/ All State Farms 46.00o 138,000 165,600 138,000 27,600 167,804 69,886 V.A.D.E. 28,3@O 84,900 101,880 84,900 16,980 100,220 41,739 a/ Ghoss heoting value (residue) a 17.51 NJ/kg t 12S mcvb bF Fueluood: 500 kg/.3; 17.8 NJ/kg at 15% wewb ,i TOE - 42.74 GJ 6.3 There are virtually no alternative uses for these residues so they are burned on the field. The fields are not grazed because the presence of tsetse fly in the bush prevents development of cattle berds. The region is scarcely populated and has sufficient supplies of fuelwood so residues are not used widely as a household fuel. The 10,000 permanent and up to 15,000 seasonal laborers at the enterprise use only sa11 amounts of maize residues. Small amounts of crop residues are bulldozed after harvest to form firebreaks in border fields. Thus, after harvesting and prior to ploughing, virtually all stalks and cobs are burnt to clear the fields. 6.4 Harvesting of maize begins in mid-November and can continue until the beginning of March, for a period of 3 to 3 and a half months. About 8,000 hectares of maize are harvested mechanically throughout the enterprise, with the remainder being picked manually. The primary reason for this it that much of the crop lodges around tasseling time in September, preventing mechanized harvesting. Secondarily, only 50 of - 87 - Wollega's 96 combines are operational for maize harvesting due to a lack of spare parts, mechanics and equipment suited to the terrain. 6.5 Mechanical harvesting is practiced on three farms (Loko, Ukie and Bello) which have little lodging; 70Z of their crop is picked by combines. While there is little difference in yield between mechanical and non-mechanical methods of harvesting, concentration and quality of residues are quite different. After combines have harvested a field, no stalks are left standing and cobs are widely dispersed throughout the plot, making residue collection difficult. Furthermore, mechanical harvesting uproots the stalks along with clumps of soil; collection of these stalks would result in the feedstock having a high soil content which, in turn, would damage the briquetting presses. Thus, farms which primarily harvest maize with combines would not be good sources of residues for densification. 6.6 The majority of maize hectarage is harvested manually by seasonal laborers. After the cobs have been picked, the stalks aze pushed to the ground to indicate which rows have been harvested and to facilitate burning later on. The cobs are collected in jute bags and carried to a central location where kernels are later recovered by miAnually feeding the cobs into semi-stationary combine harvesters. Cobs are piled behind the combine which then moves approximately 100 m down the field to the next processing location. Manual harvesting can be delayed due to the prevalence of weeds caused by irregular herbicide applications or by a shortage of labor since farm harvesting coincides with the household harvest. Still, manually harvested fields would be more suited to residue collection than mechanically harvested ones because residues are more centralized and less contaminated with soil. Potential Pilot ProJects for Maize Residues 6.7 A chemical and energy analysis of maize residues (stalk, husk and cob) is presented in Table 6.3. The data indicate that this crop residue has a low ash content, is high in carbon and has a relatively high gross heating value (17.5 MJ/kg), indicating that it could be an excellent fuelwood substitute. However, the residues are dispersed and cannot be easily used directly from the field. The main market, Addis Ababa, is over 300 km away from the nea'est farm in the Wollega Agricultural Development Enterprise. The nearby town of Nekempte is not considered to be a major market for maize briquettes because of the availability of other biomass fuels at low or nio cost. Thus, for purposes of storage and transportation, these residues require densification. - 88 - Table 6.0t MAIZE RESIDUES - CHEMICAL AND ENERGY ANALYSIS Corn Stover Eucalyptus Globulus Moisture content (%) a/ 12.0 15.0 Ash content (5) 3.45 0.29 Volatile content (5) 77.5 82.2 C (5) 47.1 47.5 H (5) 6.05 6.08 0 (4) 43.5 45.8 N (5) 0.40 N.A. Suphuric lignin (1) t7.2 N,A. HHV NJAKg 19.90 21.00 HV NJ/Kag 17.51 17.79 I Except for moisture content, all measurGments are given on a dry basis. 6.8 Civen the overall availability of maize residues f r use as a substitute for fuelvood, especially in light of the fact that they are currently being burned because th: e is no alternative use, a commercial scale pilot project to collect, process, densify and store maize residues is proposed. A pilot project producing approximately 5,000 tonnes of maize residue briquettes from stalk, husks and cobs would provi,le a much- needed fuelvood substitute to urban consumers, prevent the waste of this potential resource that currently occurs and provide valuable information for future utilization of this renewable energy source on a larger scale. Pilot Plant for Briguetting Maize Residues Site Selection 6.9 Over 60Z of maize hectarage on State Farms is located within the Wollega Agricultural Development Enterprises (A.D.E.) in an area where there are few significant uses for the residues resulting from maits production. Wollega consists of eight farms but, as previously noted, those which rely primarily on mechanized harvesting would not be suitable candidates for residue collection and briquetting. Given these facts, the Ministry of State Farms selected the Anger Farm, located 350 km from Addis Ababa, as the prime candidate for a commercial scale pilot briquetting plant. 6.10 Anger Farm comprises 6,700 ha of productive land, with 5,700 ha devoted to maize. The layout of the land and lodging of the plants prevent direct harvesting by combine for most of the maize plots. Farm management would like to substitute wind-affected maize with dwarf - a8 - varieties and rotate crops to reduce weeds and pests, but due to limited resources and strong demand for continued maize production, these plans are long-range. Further, Anger possesses hydropower potential in the form of a waterfall with a 7 m heid located near the farm's headquarters; a small, non-functional hydropower plant is already in place. Crop and residue production is shown in Table 6.4. Table 6.4: MAIZE DATA: ANGER STATE FARM Crop Yield Residue Yield Total Residues Fuelwood Equlvalent TOE No (tAha) (t/ha) (tonnes) (tonnss) (tonnes) 1,110 2.5 3.0 3,300 3,246 1,352 1,434 3.0 3.6 5,200 5,115 2,130 3,203 3.5 4.2 13.400 13,182 5.490 5,677 21,90G 21,543 8,972 Source: Chief Agronomist, Anger State Farm, 1984 and World Bank estimates. Because of quantity and availability of these residues and the lack of other end uses for the maize wastes, Anger State Farm has been selected as the site for the pilot maize briquetting plant. Collection and Storage of Residues 6.11 By modifying the present harvesting system, maize residues can be concentrated for easier collection. It is proposed that stalks no longer be pushed to the ground but ahould be left standing so that they can be manually cut 5-15 cm above ground level. The stalks would be piled in swaths of 2.0-2.5 m, with each swath representing 6 rows of maize. This system of cutting and piling stalks ij already practiced in a limited way on the farm for construction of firebreaks and thus is a familiar task. 6.12 Assuming no intervening rain, the stalks could be processed immediately after windrowing. St4lks from each swath would be fed manuslly into a tractor-drawn chopper, chopping the stalks into 5 - 10 cm pieces and blowing them into a trailer pulled by the same tractor. Manual feeding is advised to avoid inclusion of soil particles that could damage the briquettor. Should rain increase the moisture content of the stalks to above 12X mcwb, then field drying will be necessary prior to collection and chopping. 6.13 Cobs and husks should be recovered from the heaps left by the combines. They will be placed in jute bags which can then be emptied - 90 - into tractor-drawn trailers similar to those used for chopped stalks. The trailers, filled with chopped stalks, cobs or husks, will be hauled to the briquetting plant. Some residues will be supplied directly to the plant or stored overnight for processing the next day. However, the majority of material can be baled and stored for future processing once collection has been finalized. The estimated time for collection is three months (80 net working daya) and, with storage, the plant can continue to operate for another 7 months. 6.14 A list of the costs and types of equipment required to collect 5,000 tonnes of maize residues is presented in Table 6.5. Table 6.5: COLLECTlON EQUIP.4ENT AND COSTS FOR MAIZE IESIOlUES I ten Cost a/ (USS,) Tractors (9) '22,385 Bulk trailers (9) 81,817 Choppers (5) 32,918 Bags for cobs (1,t0) M9 Tots I 238,110 a/ All collection equipment Is foreign costs. The basic assumptions used to derive this list are: (a) Stalk production of 3,000 kg/ha and cob production of 600 kg/ha; (b) Available time for cutting and collecting residues is 80 net working days at 10 net hours/day; (c) Chopper capacity is 1,500 kg/hr at 65% efficiency; (d) A six-row swath averages 4.2 meters and yields 1,260 kg per 1 km swath; and (e) Trailers have a loading capacity of 3 tonnes of cobs and husks. 6.15 With a collection period of three months and the plant operating for ten months of the year, 30% of the material can be processed directly while 70% must be stored for processing at a later stage. Since storage of bulky material is expensive, it is recommended that the residues pass through high compression balers at the plant - 91 - site. Thus, most of the trailers coming in from the field will discharge their load onto an inclined belt conveyor which transports the residues to a storage bin over the baling presses. With press cepacity related to 7tS of raw material supply, balers capable of handlins 4.5 tonnes/hour will be needed to produce about 200 bales an hour. During the collection season, 160,000 bales will be produced and stored near the plant. Bales will be transported to piles manually, with the help of trolleys and a rubber belt conveyor. Transport of baled and unbaled residues to the briquettor is discussed in the following section. A schematic diagram of the residue storage operation is presented in Figure 6.1. Plant Design and Eguipent 6.16 Figure 6.2 presents a flow diagram for the proposed maize residues briquetting plant. For direct processing, the self-discharging trailers arrive from the field and unload directly onto a concrete slab running alongside an undergro' conveyor; after the collection period, all material is supplied in L._es from the storage area. Although this supply can be done manually (at a rate of 60-70 bales per hour), it is anticipated that one of the available tractors and two trailers will assist for 300 hours per season. After removal of baling wire or twine. the bales are placed on the underground belt conveyor. 6.17 After the residues, baled or loose, are fed onto the belt conveyor (1), they are transported to a bale breaker (2) in the case of bales or to the by-pass chute in the case of loose material. The bale breaker is provided with a dust hood which is connected to the pneumatic suction pipe. Next, the residues are discharged on the floor and sent through hammermills (3) by means of an inclined below-grade conveyor (4). The hammermills reduce the particle size of the stalk and cobs to match briquettor requirements. From the hammermills, the material is passed over a vibrating screen (5) to separate out oversized pieces of material which are sent back to the hammermills. All other material is then pneumatically fed (6) to the top of the briquetting surge hoppers (7). The maize residues are then fed to the piston press briquettors (8) by means of a screw conveyor (9) located below the surge hopper. The hot briquettes leaving the die are pushed through a cooling line (10) and fall into 40 kg bags at a bagging station (11). The filled bags are then removed to temporary storage capable of holding about 10 days production, prior to being transported. Figure 6.1: Maie Residue Stomge Row Diagram bkK*O Rat I B 1~~~~Tpo Loek T=rso FROM FELD SKERAG 709E Tr WVid dork-3M11 Figure 62 Maize Reddue Buiquefflng Plant Flow Dkgam | \/ ~~~~~~~~6 SLAg.HOppw Sole,t &cwvx" b Sn 7LseHPe Resldues Bt Co ~ Han 4 3 Rm/Or S Wbdd arida-3008012 - 94 - 6.18 A list of plant equipment and construction costs is presented in Table 6.6. Plant capacity is based on a total production of 5,000 tonnes of briquettes per year to be produced in 260 days of 16 hours/day. The installed capacity will be 1.5 tonnes/hour, allowing for 20X downtime, resulting in a net capacity of 1.2 tonnes per hour. Thus, two piston press briquettors will be required, each with a maximum capacity of 0.75 tonnes per hour and an average operating capacity of 0.6 tonnes per hour. The additional capacity will be beneficial in the event of production stoppages due to equipment failure. Housing is a significant construction cost; it is a necessary expense to attract and shelter permanent staff at the W.A.D.E. farms which are far from population centers. 6.19 Maximum power requirements for the plant (briquettors, balers, hammermills, conveyors, fans, lighting, etc.) are estimated at 300 kW. Grid electricity is not available at any of the Wollega farms. Therefore, power will be supplied by two 150 kW diesel generator sets, each capable of taking the average load of the plant. This is necessary to prevent generator outages from stopping the entire operation of the plant. The average power consumption is estimated at 90 kWh/tonne, including the energy to bale 702 of the residues that will go to storage. characteristics of Maize Residues Briquettes 6.20 Maize residues samples taken from Ethiopia had a bulk density of 99 kg/m ; briquet es from this material had a bulk density of approximately 511 kg/mr and a moisture content of 5.701 mcwb. When stored in a humid atmosphere, corn stover briquettes stabilized at a moisture content of 9.41 mcwb after 16 days. The average ash content was 3.45X db with the ash reaching a softening point at l,100lO and a fusion point at l,3100C. 6.21 In combustion tests conducted in a woodburning stove, maize residue briquettes were easy to ignite, maintained their integrity during the combustion process and did not smoke excessively or emit noxious odors. They burned with a cooking efficiency of 28.5Z as compared to 25.11 for a typical domestic fuelwoot under the same test. Technicallyt maize residue briquettes seem to be a good candidate for fuelwood substi- tutLon because of their favorable burning and storage characteristics. Detailed test data on their use are presented in Annex 2. Capital Costs 6.22 Total capital costs for maize residues collection, storage and briquetting are $1,464,200, including construction (housing for workerst plant buildings and site preparation), equipment and engineering costs. A summary of these costs, including foreign and local components, is - 95 - provided in Table 6.7. A more detailed breakdown of costs is presented in Annex 2. Foreign costs are estimated at $1,165,800 while local costs are estimated at $298,400. The capital costs were annualized over the plant or equipment life at a 10 discount rate. The total annualized capital costs are reported in the last column of Table 6.7 and are estinAted at $214,000. Assuming an annual production of 5,000 tonnes of briquette., the capital charge per tonne of briquettes is $42.80. Table 6.6: MAIZE RESIOUtS URIQTE1TING PILOT PLANT: COSRUCTION AND EQUIPAENT COSTS Iter Cost (USS) construction: Site preparation 10,000 Plant buildings 82,200 Employee housilg 222,100 Construction Total 314,300 a/ ulspment: Piston briquettors (2) 206,400 Storage equlpment 37,200 Conveyors 23,500 Screens 3,000 Silos 20,000 Bale breskersheamoermi ls 40,000 Electrical 12,200 Transformer/generators 127,500 Bgging stations 2,400 Vorkshop equipment 24,200 Other 5,000 Equipment Total 501,400 b/ a/ Of which US$253,500 Is local and US$60,800, foreign expenditure. b/ All equlpment costs are foreign exchange expenditures. Annual Operating and Maintenance Costs 6.23 The annual operating and maintenance costs for a 5,000 tonne per year maize residues briquetting plant are estimated at $328,300. A summary of this estimate is presented in Table 6.8. Of the total, local costs are $148,000 and foreign costs are $180,300. The major annual - 96 - local costs are $83,600 for labor and $20,600 for bags. The principal foreign costs are $74,000 for spare parts, $70,100 for diesel fuel for the electric generatoro and $11,700 for diesel fuel for residue collection equipment. Details of the O&M costs are presented in Annex 1. Table 6.7: MAIZE RESIDUES BRIQUETTING PILOT PLANT: CAPITAL COSTS (USS) Total Costs Annualized Charges Item Local Foreign Total Local Foreign Total A. Construction 253,500 60.800 314,300 29,aIO 7,140 36,920 S. Equipment - 739,500 739,500 - 120,250 120,250 C. Sparse at delivery - 73,950 73,950 - 11,470 11,470 D. Transport and delivery 2,700 51,470 54,170 420 7,980 8,400 E. Engineering & installation 15,040 134,120 149,160 1,770 15,750 17,520 F. Contingencies 7 105980 13311 3.200 16.26D 19,460 Total 298,360 1,165,820 1,464,190 35,170 178,850 214,020 Production Costs 6.24 Average production costs for maize residues briquettes were calculated on the capital and O&M costs estimates presented in Tables 6.7 and 6.8. Average financial costs of production for maize stalk/cob/husk briquettes are $108.46 per tonne while economic costs are $122.03/tonne. For maize residues briquettes delivered to Addis Ababa, the transportation costs will add $31.88 per tonne to the financial costs and $22.62 to the economic costs. Thus, the delivered financial costs of maize stalk briquettes in Addis Ababa are $140.34 per tonne while the economic costs are $144.65/tonne. The financial and economic costs per unit of energy for maize residues briquettes are $7.53/CJ and $7.57/CJ respectively. A summary of the ex-factory and delivered costs of maize residues briquettes is presented in Table 6.9. - 97 - Table 6,8: MAIZE RESIDUES BRIUETTING PILOT PLANT: OPERATING AND MAINTENANCE COSTS Local Foreign Total A, Labor: Skilled 40,800 40,800 Unskilled 42,800 42,800 B. Power: Electrlcity Diesel 70,125 70,125 C. Operation & Maintenance: Fuel 11,669 11,669 Lube Oil 8,179 8,179 Spare Parts 73,950 73,950 services 7,395 7,395 D. Building Maintenance: Materials 9,429 9,429 E. Consumables: Bags 20,600 20,600 Other 13,500 13,500 Operating Costs 134,524 163,923 284,447 Contingency 13.452 16,392 29,845 Total Operating Costs 147,976 160,315 328,292 Table 6.9: FINANCIAL AND ECONOMIC COSTS FOR MAIZE RESIDUES BRIQUETTES Financial Economic Delivered Ex-Factory Delivered Ex-Factory Cost/tonne CUSS) 140.34 108.46 144,66 122.03 cost/GJ CUSS) 7.52 5.82 7.76 6.54 Cost/Useful GJ a/ 26.39 20.42 27.23 22,95 a/ With a stove conversion efficiency of 28.5S. - 98 - . VII. POTCTMAL DEoWanD FOM DIQURTNB summary 7.1 A siseable market for agricultural residues briquettes exists, given the current household and industrial demand and pricing structure. In the household sector, firewood, animal dung, crop residues and charcoal accounted for 99.62 of fuel consumption or 21.99 million tonnes of wood equivalent (twe) in 1982. By 1992, this is expected to climb to 28.77 million twe, of which 22.35 million twe will represent a deficit beyond sustainable supply. In the test market for briquettes; i.e. the Addis Ababa peri-urban area, the biomass fuel deficit will be 940,000 twe or 747,000 tonnes of briquettes annually. In financial terms, briquettes have a significantly lower delivered cost when compared to charcoal and most forms are competitive with fuelwood. In the economic analysis, briquettes are priced, on a useful energy basis 22/, well below kerosene which is the most reliably supplied comparator. 7.2 In the industrial sector, total energy consumption for process heat is equivalent to a potential demand for 338,000 tonnes of briquettes annually. However, without major capital investments, 224,000 tonnes of briquettes can be technically substituted for other fuels. This potential demand is expected to rise to 566,000 tonnes by 1992. A detailed June 1985 survey of potential industrial users has indicated an im-ediate, economically feasible demand for 94,150 tonnes of briquettes. In financial terms, all briquettes except those made of maize residues have a lower useful energy cost than fuelwood, electricity and fuel oil for industrial purposes. Furthermore, all briquettes are financially less costly than charcoal. In economic terms, even maize residues briquettes are competitive with all industrial fuels. Thus, given the overwhelming market for substitute household and industrial fuels and the price competitiveness of briquettes in both markets, the pilot projects' production is assured of strong, long-term demand. Household Sector Detailed Demand Information 7.3 In the short term (to 1992), the estimated household demand for fuelvwot will be 28.3 million twe annually which will cLimb to 35.9 million twe in the long term (1993-2002). Table 7.1 provides a regional 22/ Useful energy .is defined to be the gross energy of the fuel times the efficiency of conversion of the particular combustion device (i.e. cook stove or furnace). - 99 - breakdown and a summary of the contribution of new supply in the short term. Table 7.2 presents simil- information for the long term scenario,. Despite the development of new energy sources, there will still be a 22.4 million twe deficit in the short ran and a 12.1 million twe annual deficit daring the decade 1993-2002. Therefore, there will be a persistent, massive household demand for a fuelwood substitute throughout Ethiopia wall into the future. Table 7.1: SUWMARY OF SUPPLY AND EMMAND PROSPECTS FOR HWSEHOLO FUELS IN THE SHORT TERM (up to 1992) Annual Contribution Possible by 1992 From New Sources Achievable New Biomess Plantatlons F fuels, Petroleum Estimated Other Tree Planting & Electricity Need In 1992 Ready for Harvest a/ Supply Deficit b/ (ha'OO0) (twe million) (haO00O) (twe mIlilon) (twe million) (twe million) Nthern Region Pr-i-urban 240 1.77 15 0.11 0.35 1.31 Rural 1,406 8.62 186 1.14 - 7.48 Southern Region Porl-urban 270 3.34 30 0.35 2.01 0.94 (Including Addis Ababa) Rural 938 14.54 124 1.92 - 12.62 Total 2,850 28.27 355 4.09 2.36 22.35 a/ Planted by 1981. b This Is a deficit In fuelwood or other high quallty cooking fuel and will, In practice, be met at whatever level by a combination of dung, crop residue, twigs and leaves and by simply making do wlth less cooking fuel. Source: Ethiopia Energy Assessment, 1984. Demand-Side Issues 7.4 Consumer demand for briquettes as a household fuel substitute will be influenced by their social acceptability, their compatibility with currently used cooking stoves and, of course, their price, which will be discussed in the following section. 7.5 For briquettes to be an acceptable household cooking fuel, they must: 1) be easy to ignite, 2) be suitable for cooking local foods, and - 100 - 3) exhibit favorable or comparable characteristics (low smoke output, no offensive smell, low ash, rugged storeability) in relation to existing cooking fuels. Regarding combustibility, laboratory tests conducted by TWO of The Netherlands with briquettes made from Ethiopian agricultural residues have shown that they are generally easy to ignite and burn well. In fact, they exhibited a much more flexible power output range than was possible with eucalyptus wood under similar tests. Given their combustion performance, the briquettes are more comparable to charcoal than fuelwood. Table 7.2: SUMMARY OF SUPPLY AND DEMAND PROSPECTS FOR HOUSEHOLD FUELS IN THE LONG TERM (1993-2002) Annual Contribution Possible by 2002 From New Sources Achievable New Biomass Plantations I Fuels, Petroleu Estimated Other Tree Planting & Electricity Need In 2002 Ready for Harvest a/ Supply Deficit b/ (ha'000) (twe million) (ha'o000) (twe million) (twe million) (two million) Northern Region Pert-urban 482 3.53 324 2.38 0.71 0.44 Rural 1,560 9.57 910 5.58 - 3.99 Southern RegIon Pri-urban 550 6.49 206 2.50 3.12 0.87 (Including Addis Ababa) Rural .5B0 16.33 920 9.51 - 6.82 Total 4,172 35.92 2,360 19.97 3.83 12.12 a/ Planted by 1997. b/ This Is a deficit in fuelwood or other high quality cooking fuel and will, In practice, be met at whatever level by a combination of dung, crop residue, twigs and leaves and by simply making do with less cooking fuel. Source: Ethiopia Energy Assessment, 1984. 7.6 For cooking local foods, a flame is preferred, especially for preparing injera, so fuelwood is used for this purpose while the preparation of wat (stew) and coffee can be done with wood or charcoal. Residue briquettes burn with a good flame and, thus, can substitute for all types of cooking fuel. The TNO tests indicate that water can be boiled using briquettes in a traditional stove in 8-11 minutes compared to 10 minutes with eucalyptus, and that the power output of a briquette flame ranges from 3.7 to 4.9 kW compared with 4.8 kW for eucalyptus. - 101 - 7.7 As for burning characteristicse combustion tests indicated that residue briquettes produced neither significantly more nor lese smoke that fuelwood, did not exhibit any noxious odors and did not fall apart during the cooking process. The briquettes tested did have a slightly higher (in the case of coffee parchment) to much higher (in the case of maize residue) ash content than eucalyptus. Thus, use of briquettes will be marginally less convenient in that cooks will have to clean out thrir stove grates more frequently. Also, briquettes which were stored in a moist environment (20°C at 90X relative humidity) absorbed water, thus increasing their moisture content from around 62 mewb to 9-11 mcwb over a 16-day period. However, this is not likely to be a problem for Ethiopian households as (1) most low income families purchase only enough fuel for a few days' use, and (2) such exceedingly wet climatic conditions are unusual in most of Ethiopia and certainly in Addis Ababa, except possibly during the rainy period. In addition, most fuelwood used in the domestic sector is in the range of 15 to 301 mcwb due to the relatively short time period between which the fuel is harvested and actually consumed. Thus, even the higher moisture contents of the briquettes should not pose a problem regarding stove efficiencies, flame temperatures, smoke, etc. 7.8 As combustion tests were performed in a one-hole metal stove with a grate and without a chimney (see Fig. 7.1) which is similar to urban household woodstoves used in Ethiopia, it appears that agricultural residue briquettes can substitute for fuelwood in existing cookstoves with no modifications. A principal difference will be the need to clean the stove more frequently due to the higher ash content present in briquettes. The testers, TMO (The Netherlands), believe that briquettes will also burn well in existing charcoal stoves; this conclusion is based on similar work done with briquettes in the "Thai bucket" charcoal stove. No tests were conducted with briquettes burning in a three-stone open fire; however, given their similarity in so many respects to eucalyptus, no problems are anticipated substituting them in this most basic type of "stove" or the typical injera cooker/oven. Figure 7.1: ONE-HOLE METAL COOKING STOVE -4 Front View Cross-section - 102 - Comparative Economics 7.9 When consumors purchase fuel, they face a market price (c/kg) but the most accurate measure of the price of a fuel is the cost of the useful energy that it provides (U8$/GJ). Based on financial costs, (i.e. market prices) households are currently paying 8.3 ˘/kg. for fuelwood. Table 7.3: COMPARATIVE FINANCIAL AND ECONOMIC COSTS OF HOUSEHOLD FUELS AND BRI9ETTES IN AWIS ABADA Cost of Conversion Cost of Cost/unit off Energy Efficiency c/ Useful Energy (USt/kg) MJ/kAg) (USSiGJ) (S) (USS,6J) Financial: Briquettes: Coffee parchment q/ 4.3 19.3 2.20 28.8 7.65 Coffee parchment bl 3.6 19.3 1.86 28.8 6.46 Coffee husk 7.2 19.8 3.63 24.3 14.93 Cotton stalk 11.4 17.8 6.40 25.2 25.40 Wheat straw 11.4 18.1 6.30 27.0 23.33 Maize residue 14,0 18.7 7.53 28.5 26.40 Wood fuels: Eucalyptus d/ 8.3 17.8 4.66 25.1 18.60 Charcoal d/ 43.5 29.0 l5.OO 30.0 =/ 50.00 Ecoom Ic Briquertes: Coffee parchment a/ 4.6 19.3 2.37 28.8 7.90 Cof fee parchment bl 3.7 19.3 1.90 28.8 6.62 Coffee husk 6.2 19.8 3.14 24.3 13.85 Cotton stalk 11.3 17.8 6.36 25.2 25.23 Wheat straw 10.5 18.0 5.88 27.0 21.77 Malze residue 14.5 18.7 7.76 28.5 27.22 Alternative fuel: Kerosene 39.88 f/ 36.7 10.87 36.0 30.19 a/ Mercato site. b/ mmw site. c/ These are laboratory efficiencies which are used to provide relative cost comparisons. d/ boodfuels Market Price Assessment, April 1985. e/ Based on efficlencles In Improved charcoal stoves. Average for charcoal stoves currently In use In Ethiopia was estimated at 24% by the World Bank Energy Assessment Mission. f/ Price In t/l and GHV In NJ/I. - 103 - Using their financial costs, coffee parchment and husk briquettes are 1.1 to 5.60 less per kg than fuelvood and the cotton, wheat 2nd maize residue briquettes are more expensive. lowever, using the more representative cost of useful energyt the financial price of fuelwood is US$18.60/GJ. At this value, all the coffee briquettes are less expensive while the cotton, wheat and maize briquettes are costlier than fuelwood. If the price of fuelwood were to increase by 251 then wheat residue briquettes would be less expensive; wood prices would have to jump by 37Z for cotton stalk briquettes to be competitive and 42U for maize residue briquettes. Given the rate of increase in the market price of fuelwood over the past five years of 6.5X per year, wheat briquettes will be competitive in 4 years, cotton in 5 and maize in 6 years. 7.10 If charcoal is used as the reference fuel, then all briquettes are financially more attractive. In terms of costs per unit sold, charcoal is more than three times as expensive as the highest priced briquette. Even when using the cost of useful energy figures, charcoalt, with its higher conversion, is currently priced 901 higher than the most expensive briquette. Thus, in financial terms, all types of briquettes represent significant savings over charcoal while the coffee, parchment and husk briquettes are also less costly than fuelvood. 7.11 A comparison based on the economic costs of the briquettes against those of kerosene is preiented in Table 7.3. Kerosene is used as the comparator for two reasons: (1) the present economic costs for fuelwood and charcoal are undetermined; and (2) kerosene is the only domestic cooking fuel currently available on a basis where its supply can meet demand. The prescut market prices of fuelwood and charcoal are not representative of the economic costs because they do not accurately account for replacement costs, and the supply of fuelwood to the market is regulated. The costs of fuelwood from proposed peri-urban plantations is not used in this analysis because this fuelwood is not presently available and its supply will not be sufficient to meet the total fuelwood demand. The economic cost of kerosene is derived by removing subsidies and shadow pricing the foreign exchange components of the retail price. 7.12 Using kerosene as the comparator, the most expensive briquette is about 302 less costly on a per NJ basis. On a cost per unit of useful energy, kerosene, with its higher conversion efficiency, is still US$3.00/GJ more expensive than the most expensive briquettes and over US$24/GJ more than the least expensive briquette. Therefore, all types of briquettes are economically viable ir the domestic sector when compared to kerosene. - 104 - Industrial Sector Potential Demand 7.13 The tctal 1982 energy demand in the industrial sector was reported at 167,000 toe. The distribution of this demand between biomasu, electricity and petroleum is presented in Table 7.4. Almost all the biomass energy is from fuelwood and charcoal which are in scarce supply. The biomass energy in all cases is used to provide process heat. Approximately 501 of the electricity consumption is estimated to be for process heat from electric boilers while over 901 of the petroleum consumption is also estimated to be for process heat, with the cement industry being the major consumer. Table 7.4: ACTUAL AND FORŁCAST INDUSTRIAL ENERGY CONSUMPTION ('000 TOE) 1982 1992 Slomams Foels 29 96 Electrici fy 35 Petroleum 103 213 Total 167 407 S Share of Total Energy ConsoptIon 2.1 3.7 Sourc: Ethlopia: Issues and Options In the Enemy Sector, July 1984, Joint UNOP/World Bank Energy Sector Assessment Program, Report No. 4741-ET. 7.14 Agricultural residue briquettes can be used to substitute for a significant portion of present industrial energy consumption. Using the total energy consumption for process heat, there is a potential demand of over 338,000 tonnes of briquettes. However, not all of the process heat demands can be readily substituted by briquettes. For example, in the cement industry briquettes can substitute for only about 201 of the process heat requirements without major capital investments. However, industries with fuel oil boilers can, with some equipment modification, convert completely to the use of briquettes. Based on information derived in two recent surveys of Ethiopian industry, it is assumed that 252 of the total industrial electricity demand and 501 of the total petroleum demand can technically be substituted with briquettes. This, along with 1002 of industrial biomass demand, results in a potential technical demand of 224,000 tonnes of briquettes. Total industrial energy demand is expected to increase to 400,000 toe by 1992. Using the same assumptions as above results in a potential technical demand in 1992 of 566,000 tonnes of briquettes. Table 7.5 (a): LEVEL 1 POTENTIAL INDUSTRIAL BRIQUETTE DENAND Plants Directly Surveyed Briquette Current Fuel Current Fuel Cost Demond Cost of Briquettes Plant (s) Location (type & osunt/year) (USS/year) (tonnes/year) (USS/tonne) ((USS/year) National Alcohol Addis Ababa Fuel"woo" 193,801 2,000 34.32 66,640 Distillery 5,400 r Ethiopian Hardwood Addis Ababa Fuelwoos 19,560 550 34.32 18,361 and Softwood Board 1,5C0 Plant (ETHARSO) Edget Oil Mill Addis Ababa Fuelwo1 49,527 660 34.32 22,651 1 ,380.f Private Brick Addis Ababa Fuel"oos 437,696 Kilns 8,000 i' Fuel all 74,435 288,000 liters TOTAL 5 512,131 3,625 34.32 124,410 Tobacco Barns 220 km east of Fuelwood / 150,620 1,225 97.73 119,719 (Awara Nelka) Addis Ababa 3,440 ,r Tobacco OBans IS0 km east of Fuel woos 43,059 430 102.76 42,131 (Nbra Era) Addis Ababa 1,150 us Li'me Kilnas 120 km west of Fuel oil 145,000 810 124.16 100,570 Addis Ababa 561,000 liters Charcoal 78,2'2 bl 43.98 35,624 525 tonnes (est.) Tobacco Barns 255 km south Fuelwoo 141,190 1,700 46.11 78,387 (Awase) of Addis Ababa 5,000 0 Total Surveyed Level 1 Potential Demand 11,000 .1 3,440 3 Is demand; actual fu"lwood supply is 860 P', so some tobacco leaves rot In field. b/ Price of bulk charoaol In Abbo reported by Lime Kiln Plant manager to be S149.07 per tonne. While this sees low, It Is ued as the least cost source for charcoal. - 106 - Table 1.5 (b): SUPPLFWENTAL LEVEL 1 POTENTIAL INIOUSTRIAL BRI9UETTE OE4AND Plants not Surveyed Current Fuel Current Fuel Briquette Demand Plant (type and annual amount) (cost per year) (tonnes per year) United lI and Fuclwood, 21,000 03 N,A, 7,600 Soap Mill + (Electric DLR) (Addis Ababa) Akakl Oil Will Fuelvood, 3300 0 N,A. 1,200 (Addis Ababa) 150 tonnes fuel oll + (Electric BLR) Nezrawl KIba Noug Fuelwood, 2400 3 N.A. 800 oIl MliI + (Electric BLR) (Nazareth) Arsine MItta Fwalwood, 1200 0 N.A. 400. oil will + (Electric BLR) (Nazareth) Total Supplemental Levl I Potentlal Demands 10,000 Overall Level 1 Potential Demond (Surveyed plus Supplemental): 21,000 7.15 The economically feasible demand for agricultural residue briquettes, given the cost of producing and transporting the briquettes, is less than 501 that of the potential technical demand. A detailed survey of potential industrial users was conducted, as part of this study, in conjunction with the Ethiopian Ministry of Energy and Mines and the Ministry of Industry. The resulting consultant's report is summarized in Annex 3. The survey indicated a current potential economic demand of approximately 94,150 tonnes. This demand was further categorized into two levels depending on the technical ease with which potential industries could convert to utilize the briquettes as a fuel. Level 1 demand consists of those industries which required no major modifications to convert to briquettes. These were generally industries currently using, or with a capability of using, either fuelwood or charcoal. Level 2 contains those industries which would require some modifications to combustion equipment in order to utilize the briquettes as a fuel. Potential Level I annual demand was estimated at 21,000 tonnes of briquettes (equivalent to the total output of all the briquetting projects) while Level 2 demand was estimated at 73,150 tonnes. The list of Level 1 and 2 industries is presented in Table 7.5 and 7.6. 7.16 The industries identified in Tables 7.5 and 7.6 do not represent a complete list of potential Level 1 and 2 industries since it was based on a limited survey that was conducted over a two week period. However, the list does represent a significant portion of the potential candidates that are within 300 km of any of the proposed briquetting pilot projects. Table 7.6 (o): LEVEL 2 POTENTIAL INDUSTRIAL 19RlETTE CEMAND Plants Directly Surveyed Cost of Fuel Oil Fuel Oil Brlquette Cost of Modification Displaced Savings Demand Briquettes Plant Location (USS) (l iters/year) (USS/year) (tonnes,yeor) (USS/yeor) Cemnt Plant Addis Abab 135,000 20% - 1.68 mililon 434,203 4,000 137,280 EthlopIan Hardwood Addis Ababa 100,000 a. 726,000 + 187,638 2,100 74,818 and Softwood Board 750 tonnes fuelwood 19.S6 Plant (ETHARSO) 207,206 Ethiopian Chipwood Addis Ababa 70,000 324,000 83,739 750 25,740 and Furniture Factory Plant (ECAFC0) cment Plant DireDawa - 380 km east 100,000 205 - I moillon 258,460 2,150 221,918 of Addis Ababa Cewnt Plant "uger - 100 km Nw 250,000-400,000 200 - 6.5 millIon 1668 million 15.000 1,260,000 of Addis Ababa Total Surveyed Level 2 Potential Deand: 24,000 el Thls represent the current mix of fuels consumed. Increasing substitution of briquettes to 50S at the Addis cement pilnt will generate additional demand of 6,000 tonnes per year. - 108 - Table 7.6 _b.s SUPPLMENTAL LEVEL 2 INDUSTRIAL 9RIQUETlE DEMAND Plants not Surveyed Fuel Oil Substituted Annual Amount Cost Nadification Cost Annual Briquette Plant (tonnes) (USS) (lUSS) Demand (tonnes) St. Ge.orges Breery 1,450 392,850 180,000 3,500 Addis Tyre 3,000 813,000 180,000 7,330 Akaki Textiles 5,000 1.25 millon 270,000 12,000 Modjo Textiles 900 225,350 100,000 2,200 Dire Dwa TextIles 7,000 1.75 IIIIon 360,000 17,000 Total Supplemental Level 2 Potential Demends 42,030 Overall Level 2 Potential Demand (Surveyed and Supplemental): 68,000 Comparative Economics 7.17 It is first necessary to compare the final cost of useful energy from briquettes versus that from conventional fuels in order to identify potential industries for conversion. Factors that enter into the determination of the final cost of the useful energy include the: - difference in the delivered cost of fuels; - relative conversion efficiencies; and - additional capital and O&M costs. 7.18 A summary of the range of delivered prices for conventional fuels is presented in Table 7.7. The delivered price of fuelwood and charcoal to industries varies according to the location of the industry, the distance the fuelwood is transported and the degree of subsidy that is inherent in the government allocation and pricing process. Industrial fuelwood prices were reported as low as $26.09/tonne for the ETHARSO plant in Addis Ababa which buys from the government fuel depot, to over $86.70 for the private brick kilas outside Addis Ababa. Charcoal price distributions were equally wide from as low as $149/tonne for the Ambo Lime plant to over $362/tonne at the open market price. Industrial fuel oil prices also vary slightly but only due to differences in transport cost. Retail prices were reported at $0.23/1 at Assab $0.29/1 at - 109 - Nazareth. 23/ Electricity prices are assumed, for the purposes of this analysis, to vary between the present industrial tariff of $0.03/kWh and the expected new tariff $0.06/kWh. Table 7.7: COWARATIVE FINANCIAL AND ECONOMIC COSTS OF CONVENTIONAL INDUSTRIAL AND BRIQUETTE FUELS Production Costs Del ivered Ccnversion EEF Cost of useful Fuel USS/1 costs (W) energy (USS/GJ) Financiol Costs: Fuelwood: (S/T) (14,3 NJ/kg) 26.09-98.74 a/ 70 2.60 - 9.86 Chwacol (S/F) (29.0 NJ/kag) 149.07 - 362.00 a/ 80 6.43 - 14.69 Fuel oil (S/I) (40.1 NJ/I) 0.23 - 0.29 85 6.75 - 8.51 Electricity (USS/T)t (3.6 IJ/kWh) 0.03 - 0.06 95 8.77 - 17.54 Dliv ered Ex-factory to Addis Briquettes: Coffee Parchment (19.3 NJ/kg) 31.22 36.02 b/ s0 2.02 2.33 Coffee Husks (19.8 NJ/kg) 37.04 71.70 80 2.34 4.54 Cotton (17.8 NJ/k) 82.06 113.96 80 5.76 8.00 Wheat Straw (18.1 NJ/kg) 80.66 114.67 80 5.59 7.94 Maize Stalks (18.7 NJ/kg) 108.46 140.34 80 7,27 9.41 ' Open narket price. bl Assures a minimum of BIO/tonne for transport within Addis Ababa. 23/ Energy Assessment Report. - 110 - 7.19 The delivered cost of agricultural briquettes like fuelwood and charcoal is sensitive to transport distances. Table 7.7 gives the range of ex-factory costs for the different briquettes and the estimated delivered costs to potential industries in Addis Ababa. The delivered financial costs of the briquettes vary from a low of $36.02/tonne for the coffee parchment briquettes produced in Addis Ababa to a high of $140.34/tonne for the maize stalk and stover briquettes produced at Wollega. 7.20 A more accurate comparison of fuel costs is the cost of the useful energy that results after conversion. Conversion efficiencies vary by fuel type and by ccuversion process. Average conversion efficiencies by fuel type were estimated for typical Ethiopian systems and are presented in Table 7.7. The resulting cost of useful energy is also presented in Table 7.7. All agricultural briquettes except maize stalk briquettes have a lower useful energy cost than fuelvood if the delivered cost of fuelwood purchased by government permit is taken as the representative financial cost. All briquettes including maize stalks have a lower useful energy cost than the cost of charcoal to private industries. Taking the government subsidized prices for fuelwood and charcoal would result in only the coffee parchment and husk briquettes being competitive. In comparison to fuel oil at $0.29/1, all briquettes except maize stalks have a lower useful energy cost. The same is true for electricity at $0.03/kWh. However, if electricity is priced at $0.06/kWh, then all briquettes including the maize stalks are competitive. 7.21 The useful energy costs of the briquettes based on ex-factory costs are also shown in Table 7.7. The transport differential adds as much as $2.37/CJ to the cost of the useful energy from the briquettes. Thus, using the briquettes at locations closer to their production sites rather than in Addis Ababa would only improve their cost competitive- ness. The estimated delivered costs of briquettes for the potential level 1 and 2 industries identified earlier are shown in Table 7.8. 7.22 The economic costs of the briquettes, both ex-factory and delivered to Addis Ababa, are presented in Table 7.7 along with the economic costs for conventional industrial fuels. Economic costs for industrial fuelwood and charcoal are assumed to be the open market price even though this underestimates the actual economic costs for these fuels. Economic costs for fuel oil are derived by removing any taxes and shadow pricing, the foreign exchange components of the retail price. The economic cost of electricity is based on the estimated LRMC for grid- supplied electricity. 7.23 The comparison of the delivered economic costs indicates that all briquettes are less costly than all of the conventional industrial fuels except for cotton and maize residue briquettes which are slightly more expensive than fuel oil. When compared to charcoal or electricity, even maize residue briquettes, which have the highect cost, are about SOZ less expensive. - 111 - Table 7.7 COsMARATIVE FINANCIAL AND ECONOMIC COSTS OF CONWENTIONAL INDUSTRIAL AND 8RIQUETlE FUELS (contInued) Production Costs Delivered Conversion EFF Cost of useful Fuel USS/T Coats (W) energy (USS/UJ) energy (USSdGJ) De livered Economic Costs to Addis Fuelwood (S/T) (14,3 NJ/kg) 27.69 83.25 70 9.86 Charcoal (S/T) (29.0 NJ/kg) - 434.97 80 18.75 Fuel Ol (S/i) (40.1 NJ/I) - 0.27 85 7.92 Electricity ($/kWh) (3.6 NJ/kwh) - 0,06 95 17.54 Delivered Ex-factory to Addis 9riquettes: (USS/T) (USS/T) Coffee Parchiunt (19.3 NJ/kg) 34.50 36.75 80 2.23 2.38 Coffee Husks (18.4 NJ/kg) 41.81 61.99 80 2.65 3.92 Cotton (17.8 NJ/kg) 95.45 113.18 80 6.70 7.95 Wheat Straw (18.1 NJ/kg) 93.58 105.86 80 6.48 7.34 Maize Stalks (18.7 NJ/kg) 122.03 144.66 80 8.18 9.70 - 112 - Table 7?8s DISTANCES AND DELIVERED COSTS OF BRIQUETTES TO END USE SITES Production Distance Delivered Cost Briquettes Site Aware Mtlka (km) (USS/tonne) Coffee (Addle) Addle a/ 34,32 Muger 100 52,39 Ambo 120 43.98 Nazareth 110 47,61 Mura Era 150 52.44 Awara Molka 220 68,75 Dire Dave 380 80.21 Coffee (Dilia) Addle Ababa 330 71.70 Awes 75 46,11 Huger 430 88.97 Wheat (Dixie) Addls Ababa 20 113.71 Naza;eth 90 101.02 Nuger 300 121.31 Nure Era 130 99.63 Awara lllka 200 110,58 Dlre Dawa 360 129.71 Cotton Residues Dire Dews 130 97.76 (Awash) Aware Melka 100 94.14 Nura Era 175 103.20 Addis Ababa 270 113.94 Muger 370 126.75 Aubo 390 129.16 Maize (Nskeapte) Ambo 130 124.16 Adis Ababa 260 140.34 luger 270 141.07 a/ Assume f lat rate charge of 10,00 Blrr/tonne for delivery wIthIn AddIs Ababa metro area. 7.24 Accounting for only delivered costs and conversion efficiencies of briquettes versus the conventional fuels does not give an accurate indication of the true costs of the useful energy. The additional capital and O&M requirements for converting to the use of briquettes must also be accounted for. These costs are especially significant when considering conversion from fuel oil or electricity. Conversion of fuel oil systems to use of briquettes can vary from a simple retrofit of a gasifier/burner system to replace the fuel oil burner to a complete replacement of the entire combustion system. Conversion from electricity to use of briquettes will require the complete replacement of all process - 113 - heat generating equipment and could be quite costly. However, most industries that now have or will have electric boilers also have to maintain a fuelvood or 'zel oil-fired back-up boiler. Both fuel oil and electric systems also have lower operating labor and wsintenance covt requirements. Alternately, wood and charcoal fueled systems will require little or no equipment modifications and little or no additional operating labor and maintenance requirements. In the case of wood fueled systems, the OM requirements may even decline because of the lower volume, higher calorific value, lower moisture content and uniformity of briquettes. 7.25 Estimates of the capital and OUM costs for converting to briquette use for the various industrial combustion systems surveyed by the consultant were presented in Tables T.5 and 7.6. The data indicate that for industries with electric boilers it is economic to convert to briquettes when the briquettes can be delivered for less than $115.03/tonne. 24/ For industries with fuel oil boilers, briquettes must be delivered for less than $105.79-119.06/tonne in order for them to be competitive. 25/ For the purposes of this analysis, wood and charcoal systems are assumed to entail no additional capital or OM costs. Therefore, the competitive costs of briquettes are $108.39/tonne for wood priced at $75/tonne and $128.28/tonne for charcoal priced at $200/tonne. Industrial Demand Issues 7.26 The above discussion clearly demonstrates that agricultural briquettes are technically feasible and economically competitive with conventional fuels in a broad spectrum of industrial applications. However, there is no guarantee that any significant industrial demand will evolve without a concerted effort to convince potential industries to convert to briquettes. One major reason is the newness of the fuel to Ethiopia and the lack of knowledge and confidence regarding its use. Another is the continuing existence of price subsidies in some conventional fuels. A third reason is a lack of recogniton of the preseft and anticipated future shortages of wood and charcoal fuels. In addition to information and economic distortions, lack of a clear insticutional framework to iatroduce and foster the use of bricuettes will also inhibit demand. The following is a brief discussion of some of the demand side issues that need to be addressed. 7.27 Information/Education. Fostering the use of briquettes in the industrial sector will require a focused informational and educational program aimed at potential users. Emphasis must be placed on convincing 24/ Includes four man-years per year of labor at US$1,SOO/m-y plus US$14,000 for typical modification in Ethiopia. 25/ Includes two man-ye#rs per year of labor at US$1,500/m-y plus US$7,000 for typical modifications in Ethiopia. - 114- both managerial and technical personnel of the economic advantages of converting to the use of briquettes. Measures should be identified for overcoming or dealing with some of the disadvantages of briquettes such as the relatively high ash contents, higher generation of particulates, necessity to isolate briquettes from direct contact with water, etc. Most aspects of the information/education need. for introducing briquettes are to be addressed in the Bank's proposed Ethiopia Energy Project. 7.28 Technical Assistance. At present no organised means to provide technical assistance to industries willing to convert to briquettes exists. Technical assistance both for conversion and during the initial periods of briquette use will ensure more efficient and proper use of the product. Provisions for technical assistance in this area are incorporated in the Bank's Ethiopia Energy Project. 7.29 Capital for Conversion. Industries that require modification of equipment and/or additional facilitiea in order to use briquettes will require some capital. Availability of capital to the industrial sector if restricted would constrain potential conversion. The Bank's Ethiopia Energy Project proposes to make financing available for industrial conversion to use briquettes. 7.30 Reliability of Supply. Industries that make a capital investment or give up fuelwood or charcoal allocations in order to convert to briquettes will be concerned about the reliability of briquette supply. Also, the uniformity of briquette quality, given process needs, may be a concern for industries with no previous experience with this fuel. 7.31 Electric Boiler Program. The present Ministry of Industry program for converting industrial steam generation equipment to electric boilers, if continued, may inhibit the demand for briquettes in some industries where the technical and financial potential to convert to briquettes exists. At least one industrial facility that has both an electric boiler and a fuelwood-fired back-up boiler should be used for a direct comparison of the relative economics of electricity vs. briquette firing. Preliminary analysis suggests that the increased 06M costs associated with firing briquettes will typically add US$6.00/tonne to the price of the briquettes, or US$0.32/CJ to the useful energy cost of the briquettes when compared to electric boiler use. A similar comparison between briquettes and fuel oil results in an increase in the energy cost of the briquettes of between US$0.16 and US$0.20/CJ. At the delivered prices projected for the briquettes at most sites, briquettes should still prove to be the least cost fuel, if no capital-intensive modifications are required to enable their use. 7.32 Institutional Arrangements. An effective institutional framework will be necessary to ensure that briquettes will be distributed equitably during periods of shortage, that prices will be effectively controlled, and that contractual agreements will be honored. Industries - 115 - committing to the use of briquette., especially those that undertake investments, will welcome if not require such institutional/governmental assurances. Externalities Impact on Air Pollution in Urban Areas 7.33 Currently, burning biomass and industrial smoke contribute to air pollution in Addis Ababa. Substituting briquettes for some of the fuelwood and other biomass which is currently burned will probably have a neutral impact on outdoor air pollution. A substantial portion of families in urban areas cook outdoors or in well-ventilated areas. However, for those families who cook inside and are exposed to smoke and fumes, the impact of briquette substitution on indoor air quality is uncertain. However, for those currently using leaves, twigs, bark and other lower quality biomass, use of briquettes may in fact improve their air quality. For both industrial and household use of briquettes, monitoring and evaluation of smoke emissious should take place as part of pilot project implementation. Measures to accommodate this activity are incorporated in the Bank's Ethiopia Energy Project. Impact on Water and Land Pollution 7.34 If coffee pulp is incorporated in the Dilla briquetting site then there will be a substantial reduction in water pollution in that area as pulp currently clogs and pollutes streams in the Dilla region. Otherwise, the pilot projects are unlikely to have either a positive or negative impact in this area. Impact on Soil Nutrition 7.35 To assess this issue, each residue must be esamined separately. Coffee parchment and husks are not presently returned to the soil; thus, their removal from processing facilities will not affect soil nutrition at coffee plantations. Cotton stalks are currently burned on the field as a pest control measure. According to State Farm agronomists, the ash that remains after burning is not important to cotton plant nutrition as abundant nutrients are already available in the soil which is continually replenished through silt deposits from the Awash River. Wheat straw is either left on the field to be plowed under prior to planting or a large percentage may be baled for animal feed. In assessing overall availability of wheat residues, it was assumed that 501 of the straw would be left on the field for plowing under; this was acceptable, from a soil nutrient standpoint, to State Farm agronomists. Maize residues in the Wollega Agricultural Development Enterprise are currently burned and their ashes are not judged by State Farm agronomists to be important to soil nutrition. Elsewhere, the maize stalks and cobs are either burned in the field or collected for use as a household fuel, - 116 - as is the case around Assela. Overall, the selective removal of residues does not appear to pose a threat to soil nutrition. However, careful monitoring and evaluation of soil nutrition and structure and subsequent crop yields must be done during the pilot phase of the briquetting project in order to assess future prospects for expansion. Provisions to accomplish this by using State Farm agronomists and Ministry of Agriculture research units to monitor soil conditions and productivity are incorporated in the briquetting component of the Bank's proposed gthiopia Energy Project. Potential Fire Hazards 7.36 Fir# hazard precautions need to be taken for the storage of the raw material as well as for the final product, both at the briquetting site and with potential users. By storing the residues at low moisture cont uts, the risk of spontaneous combustion following fermentation will be eliminated. To prevent possible fire losses, storaga piles will be separated to reduce the risk of a fire spreading. Adequate fire prevention systems and regular inspections of stock should be a part of each site's routine. Fire extinguishing equipment must be supplied to each plant as part of the pilot project. In the households, there should be little fire hazard as few families will be storing large quantities of briquettes at any one time. In any event, the fire hazards should be, if anything, less than those associated with fuelwood or charcoal. Fire prevention for industrial users should be based on similar practices for the control of charcoal and fuelwood fire hazards. - 117 - VIII. MARKETING AND DISTIDUTION STRATEGY Introduction 8.1 The primary objective of this project is to provide a price competitive substitute for fuelvood to Ethiopian households and industry. The marketing strategy adopted should make briquettes available to the consumer by methods that are compatible with existing purchasing and distribution patterns in the open market and at prices that are competitive with existing alternatives. This chapter will outline a proposed marketing strategy, review the role of intermediary organizations and suggest the components of a public education campaign. Marketing Strategsy 8.2 The objective of the marketing strategies for agricultural residue briquettes should be to introduce the use of briquettes simultaneously to the domestic and industrial sectors. In the domestic sector, marketing should initially concentrate on household consumers of fuelwood and charcoal primarily in Addis Ababa. By initially concentrating on the household market in Addis Ababa, existing retail centers can be used and the largest single national market, concentrated in a relatively small area, will be accessible. Rural households will benefit indirectly through a reduction in the "urban shadow" effect. Further, many rural and semi-urban areas are not full participants in the cash economy and a marketed substitute fuel would not be appropriate for their economic situation. Finally, by limiting the market to the capital area, research on social and technical acceptability of briquette fuels can be much more easily conducted and analyzed, with results becoming available as well as subsequent program improvements being made more quickly than in a widely scattered set of markets. 8.3 The potential industrial market has already been identified in Chapter 7. The firms are divided into two levelsz those which could use briquettes without investing in factory modifications (Level 1); and those which would have to modify their equipment (Level 2). Level 1 industries are the most likely candidates for initial marketing as they will not need to make modifications to equipment in order to utilize briquettes. However, Level 2 firms with aggressive management may seek to undertake plant modifications in order to use briquettes and they should not be excluded from purchasing. 8.4 If briquettes are to reach low-income households, they must be retailed in small quantities as most poorer families buy no tore than a few days' fuel supply at a time. Further, briquettes should be sold in neighborhood markets and in the kabeles so that they are accessible to low-income purchasers. Currently, the private sector is the principal - 118 - intermediary for retailing household fuels in Ethiopia. Private retailers sell charcoal and fuelwood in small quantities through open markets which exist in even the smallest and poorest neighborhoods. Given that there are large numbers of these private retailers, their prices are subject to open competition and their mark-ups are reasonable, e.g. 15-252 for fuelwood. They provide an adequate brokerage between the supply and demand for household fuels. Thus, private merchants would be ideally suited for marketing briquettes in small, accessible and saleable quantities to households. 8.5 The most important role for governmental involvement is in wholesale and test marketing. The GOE, through the MME and ENEC, will be involved in consumer acceptance testing of briquetted fuels as part of the World Bank's Energy Project cooking efficiency program. The results from this test marketing will provide useful feedback to pilot briquetting plants that are associated with MSF and MCTD facilities. They, in turn, could promote wholesaling by encouraging private merchants to purchase briquettes in bulk at the pilot plantgate. Once production processes are smoothed to provide a secure supply of briquettes acceptable to households, wholesaling could be expedited by having the MSF and/or NCTD manage bidding for large lots of briquettes at bulk marketing facilities in Atdis Ababa. 8.6 For industrial users, large quantities of briquettes can be easily substituted for fuelvood or charcoal. As such, packaging is not the primary concern. The main consideration here is that a guaranteed demand for specified quantities of briquettes be established. This will require coordination between the Ministry of Industry, potential industrial consumers and briquette suppliers. Industries that are able to convert to briquettes and that are closest to the briquette pilot plants should be given priority. Initially, supply/demand contracts should be established between the designated briquette suppliers and the Ministry of Industry, which is responsible for coordinating the entire Ethiopien industrial sector. Gradually, as supply bottlenecks are removed, industries should arrange for their own purchases by bidding on the open market. When this occurs, briquettes for industrial use will be priced at their true economic value through competitive bidding. 8.7 The pricing of briquettes will be determined on the open market by supply and demand of briquettes and other competing fuels. Presumably, there will be variations in prices for briquettes from different residues because they will have different calorific values and burning characteristics. Once household and industrial users perceive these differences, they will express their preferences through the marketplace for specific briquetted residues. Therefore, while final prices cannot be estimated for each briquetted residue, the retailing system envisioned should allow for the briquettes to be sold for what the market can bear. - 119 - Role of Intermediary Organizations 8.8 Limiting the number of intermediaries between plantgate and final market will be important if prices are to be kept competitive with fuelwood and charcoal. As indicated earlier, this can be achieved by selling the briquettes directly to private retail and wholesale merchants operating in the open market. The potential marketing and distribution role of some of the key organizations involved in the briquetting project is reviewed below. Ministry of Coffee and Tea Development 8.9 A subsidiary of the Ministry of Coffee and Tea Development (MCTD), the Ethiopian Coffee Marketing Corporation (ECMC), is responsible for the marketing of coffee as well as residues. The latter consists primarily of coffee parchment which is currently sold "as is" form the Mercato Processing Plant in Addis Ababa. The ECKC had plans to sell the entire stock of this residue to the Ethio-Briquette Corporation, a private venture funded by the Ethiopian Agro-Industrial Development Bank. The status of this arrangement was unclear at the time this study was undertaken. In any event, the ECMC is now very keen on assuming a degree of responsibility for producing and marketing briquettes from coffee r 'idues. As such, it could be involved in managing the bidding process for wholesaling of coffee residue briquettes either to Ministry of Industry and/or private merchants. Ministry of State Farms 8.10 The Ministry of State Farms (MSF) has a Marketing Services Department which confines its activities to ex-farmgate sales. It is not involved in wholesaling or retailing, nor does it anticipate involvement in these areas. Much of this is done by the Agricultural Marketing Corporation, a part of the Ministry of Agriculture. However, although both entities operate their own fleet of trucks, they also have to hire vehicles, indicating a shortage of transport. Therefore, the NSP is not equipped to actually distribute and market briquettes. Their most appropriate role is in the collection, storage and processing of the State Farm residues. Ministry of Mines and Energy 8.11 The Ministry of Mines and Energy (MME) and the Ethiopian National Energy Commission (mNEC) which it coordinates, have been involved in the agricultural residue project since its inception. - 120 - Organizationally, they are the governmental bodies responsible for renewable energy research and development projects. Institutionally, they have the greatest awareness of the household energy situation and options facing the country. Finally, the MME has no direct financial ties or interest in the marketing process. Therefore, they are the most likely impartial candidate for undertaking and assessing the trial marketing and consumer acceptance program for briquette fuels. The Ministry of Industry 8.12 The Ministry of Industry, which is responsible for coordinating industrial investment and growth in Ethiopia, has an interest in reducing the energy costs of the industries that it is directly and indirectly responsible for. In this regard, the Ministry could be made responsible for introducing briquettes to potential industrial users. As a potential industrial demand sufficient to cover the entire output of the pilot plants has been identified among interested industries 26/, a guarantee should be obtained from the Ministry of Industry to ensure that the briquettes will be used in the industrial sector if they are not consumed by households. The Ministry should be given the responsibility for arranging individual supply and demand contracts with pilot briquetting plants and specific industries. Other Organizations 8.13 The Forestry and Wildlife Conservation Development Authority (FAWCDA) has a Wood and Charcoal Production, Processing and Marketing Enterprise (WCPPME) with 33 distribution centers operating in Addis Ababa which provided up to 202 of the 1984 supply of charcoal to the capital. Despite severe shortages of charcoal and subsidized retail prices, the WCPPME has such a backlog of charcoal at its depots (over 125,000 bags of 40 kg in late 1984) that it cannot accept further supplies. This is caused by WCPPME's restrictive retailing practices, including: limited opening hours and the requirement for an identification card and letter of approval to buy from local authorities. Further, customers cannot select bags even though weight and quality vary greatly, cannot buy twice in one day and must buy 40 kg bags only (and have cash for that size purchase, which is usually beyond most household's ability). Given this situation, WCPPME would not be a satisfactory mechanism for marketing briquettes. 8.14 The Urban Dwellers Association (UDA). a municipally-sponsored community development organization, has offices in all 284 administrative 26/ See Chapter 7 ant Annex 3. - 121 - neighborhoods of Addis Ababa. They are beginning to retail household necessities through their offices but the organization and management of this system is not fully in place. Despite an expressed interest in fuel marketing, they are not yet managerially or financially equipped to market firewood and charcoal, let alone briquettes. Public Education Campaign 8.15 Regardless of wholesale and retail arrangements, a public awareness effort should be prepared and undertaken prior to and during the introduction of briquetted fuel. The most apppropriate institution for this would be the MME's ENEC and Project Management Unit. The overall theme of the campaign should be the substitution of non-woodfuels for firewood and charcoal so as to conserve the ecosystem for future generations. The links between choice of fuel, energy availability and food production should be made clear. Specifically, the public awareness effort should concentrate on: - the proper use of briquettes in traditional and improved cookstoves; - storage, handling and re-use of briquettes; and - energy conservation through specific food preparation and cooking techniques, e.g. simmering, using lids and aluminum pots, hayboxes, etc., all of which will increase the cooking efficiency of briquettes as well as other fuels. After an intensive initial publicity campaign, the product should begin to sell itself. Periodically, publicity efforts can be mounted to promote specific modifications and to advise consumers of recommendations that may arise after more lengthy experience with briquettes as a household fuel. This is provided in the Bank's Ethiopia Energy Project as part of an overall cooking efficiency and consumer acceptance program. 8.16 Additionally, a small campaign needs to be targeted to potential industrial users of briquettes. Emphasis should be placed on convincing both managerial and technical personnel of the economic advantages of converting to briquettes from conventional fuels. Technical assistance should also be provided to industries which plan on using briquettes so that they can properly store and utilize the product. Appropriate assistance should also be arranged for industries which require boiler and other modifications in order to burn briquettes, as well as other fuels, more efficiently. This is also provided for in the Bank's proposed Ethiopia Energy Project. Finally, the potential exists for significant fuel savings through the promotion of appr4priate industrial energy conservation measures. It. INSTITUTIOUAL POLICY AND M&UCUT ISSUES Overview 9.1 Through the Central Planning Supreme Council, the Ministries of Coffee and Tea Development and of State Farms have inherent control over their respective facilities and residues. Thus, they have a legitimate need to closely manage densification at coffee plants and individual farms. In the past, both ministries have held the institutional perspective that residues are wastes to be disposed of, not to be collected and utilized. Their organizational mission has been oriented towards the production of food, not energy. However, this project now requires that they be involved, to varying degrees, in two new activitiesS densification of fuel briquettes and wholesale marketing of a household/industrial fuel. To ensure that these new objectives are comprehensively and consistently undertaken, new organizational arrangements will be necessary for project implementation. This chapter outlines some possible arrangements affecting (1) the briquetting plants themselves, (2) wholesale marketing systems, (3) inter-ministerial coordination, and (4) overall project implementation, management and monitoring. Finally, some policy issues which any management system may have to address during and after project implementation will be briefly discussed. Plant Management 9.2 For optimal operating efficiency, pilot projects will need to coordinate with their source of residue supply but will have to maintain day-to-day operational independence. Plant, equipment, stores, workshops, officest personnel and housing should be under the exclusive management of the pilot project. Finance, administration, purchasing and training should also be kept separate. However, for planning purposes involving both the production and briquetting operations, cooperation is needed because: (a) pilot plant construction should be undertaken so as to minimize interference with farm activities or coffee processing; (b) the period available for residue collection by the pilot plant depends on the actual timing of harvesting or processing and land preparation operations by the farm; (c) for some repairs, material or equipment, the pilot project may have to rely occasionally on facilities available at the farm or coffee plant, and vice versa; and - 123 - (d) some pilot plant operations, e.g. the storage of cotton stalks which may pose a pest problem, will need to be supervised by farm personnel. 9.3 To minimize potential friction or rivalry between an individual pilot project and its residue supplier, several actions should be takens (a) pay scales of both organizations should be identical; (b) farm or coffee plant personnel should be involved in the detailed planning of pilot projects which affect their facilities; (c) farm and coffee plant managers and employees should be afforded special recognition for their role in producing both food and energy for the country and for producing an economic product from previously discarded waste byproducts; and (d) ongoing coordination committees should exist at each site to work out operational details requiring coordination. Marketing Structures 9.4 Various possibilities for briruette marketing and distribution have been discussed in Chapter 8. The concern here is with institutional issues that may affect marketing structures. To briefly examine this aspect of the project, marketing is separated into two systems: (1) from the farm- or plantgate to Addis Ababa or industrial users, and (2) from central distribution points in the capital to consumers in various neighborhoods. 9.5 Briquettes must be delivered in bulk by reliable, scheduled transport from the pilot projects to Addis Ababa or industries. The Ministry of State Farms has neither the capability nor the interest in being responsible for briquette distribution. The Ministry of Coffee and Tea Development would like to participate in the wholesale marketing of both parchment and husk briquettes. Both perspectives can be accommodated on the bulk delivery side of marketing by negotiating a regular schedule of pickups and deliveries with the Ministry of Transportation, either jointly or separately, using their fleet of trucks as well as private contractors that they rely on during peak periods. Though produced during the harvest season, briquettes may have to be stored until the end of harvesting as trucking will be concentrated on hauling the harvest. Once residue briquettes have reached Addis, they must be stored temporarily unless a direct truck-to-distributor link can be established. For industrial users, storage will be the responsibility of each industry. - 124 - Inter-Ministerial Coordination 9.6 Ultimately, policy coordination between ministries is achieved by the National Committee for Central Planning. However, for implementa- tion and operational matters relating to the pilot briquetting project, coordination between the participating ministries will be accomplished through a Project Management Unit (PMU) to be established under the MME/ENEC. The PMU will have the overall responsibility for coordinating, supervising and implementing the energy components of the proposed Ethiopia Energy Project through Project Implementation Units (PIU) of the Ministries and agencies concerned. Details on the functions, structure and responsibilities of the PMU and PIUs is presented in the Ethiopian Energy Project Staff Appraisal Report. Overall Project Supervision 9.7 Apart from overall policy coordination, the technical supervision of both project implementation and initial project operation are key institutional tasks which must be assigned. It is recommended that a consultant be hired as supervising engineer for all pilot plants to assist in and review the planning, procurement, training, construction, installation, start-up and production phases of the project. Draft terms of reference for the supervising engineer are provided in Annex 6 of Volume III. The supervising engineer would monitor and evaluate production over a period of three crop seasons. He would work directly with plant personnel but would also be responsible for keeping the concerned ministries and P W informed of problems and progress. The supervising engineer would also be the contact point for World Bank project supervision and monitoring, through the ENEC. Project Monitoring and Evaluation 9.8 The pilot briquetting projects will provide a unique opportunity to determine more accurately the actual costs and benefits (i.e., social, environmental, financial and economic) associated with the briquetting, distribution and use of a range of agricultural residues. However, a compreh-ensive monitoring and evaluation program must be established to capture this potential. The program must evaluate all aspects of the briquetting project from collection of residues through the actual briquetting process to final use of the briquettes. Results of the monitoring and evaluation program should help clarify issues relating to: impacts on soil structure and nutrition; crop productivity; farm management; briquetting technology; transportation constraints; marketing and distribution strategy; consumer preferences; end-use impacts, etc. Data from the ongoing monitoring and evaluation could also be used to improve initial technical, managerial, institutional and - 125 - marketing arrangements. The end result of the program will be to provide the data for future expansion of the production and. use of agricultural residue briquettes by helping identify the residues, technologies, strategies and policies that are most beneficial to Ethiopia. - 126 - X. INNTATIE ISS1=U BUDGET AND SCHDUlLING Introduction 10.1 This chapter outlines specific implementation issues, reviews a sensitivity analysis of key variables, presents the overall project budget, and proposes a timetable for implementing the various pilot briquetting plants. General implementation considerations are discussed, ranging from funding through personnel recruitment and training to marketing and distribution issues. Then, essential sits-specific issues are briefly considered for each of the four residues. A sensitivity analysis which gauges the importance of production capacity, discount rate and equipment service life is conducted. A budget and schedule for overall project implementation, including plant as well as collection equipment operations, are outlined. Finally, provisions for project monitoring and supervision are discussed and the scope of responsibi- lities for a supervising engineer is suggested. 10.2 The implementation phase of the project cycle begins at the point where funding is secured and continues through to monitoring and evaluation of the actual project. During the different stages of this phase, an awareness of the obstacles to and opportunities for implementation can determine the project's degree of success. Thus, important issues involving each of the following stages are identified below: - Funding; - Plant design; - Tender documents/equipment specifications/procurement; - Site selection; - Personnel recruitment and training; - Organizational structure; - Civil works; - Installation and start-up; - Full-scale operations; - Spare parts; - Public education; - Transportation; and - Marketing. To the greatest extent possible, the issues which have been identified in these stages were taken into consideration in the brief project outlines which are presented in this document. During detailed plant design and actual project implementation, they should be reviewed, reassessed and dealt with when appropriate. - 127 - Pundins 10.3 Through its Energy I credit to the Government of Ethiopia, the World Bank/IDA plans to finance a significant portion of the costs of coumercial scale pilot briquetting plants. DANIDA, the Danish aid organization, has expressed strong interest in meeting both the foreign and local costs of the proposed residue briquetting plants. If DANIDA funds are channeled through the Bank, then standard Bank project development, supervision and loan disbursement procedures can be followed. If another funding mechanism is chosen, then close coordination should be ensured between the Bank and DANIDA. Additionally, a loan agreement between the World Bank and the GOS should be reached which maintains the technical, economic and managerial integrity of the residue briquetting projects, and is satisfactory to both parties. Plant Design 10.4 Basic plant designs have been worked out in this report but a complete and detailed design for each site can only be presented when more details about funding, equipment and plant location are available. For example, the design of the required buildings is only detailed as far as overall dimensions are concerned. Actual Ethiopian building standards and practices must be accounted for in the detailed design and costing of the project. Compatibility of capacity and performance for the various equipment components must be ensured. Ideally, contracts should be awarded to suppliers who can provide a complete plant because this ensures that the various components are well intePrated. Tender Documents/Equipment Specifications/Procurement 10.5 Preliminary equipment specifications have been included in this report to expedite the implementation process (see Annex 4). These are provided as guidelines which can be used to develop the detailed tender documents necessary for project implementation. In addition, these documents should include the conditions of supply, delivery times, the scope of activities to be carried out by the supplier after delivery and any required guarantees. For procurement, it is understood that the GOE has proposed and IDA has agreed that the IL0 procurement service will be utilized.. To assist this process, a list of potential supplies for major plant equipment is presented in Annex 6. - 128 - Site Selection 10.6 This report identifies those farms and facilities which are considered by both the ESNAP mission and the GOE to be best suited for initiating pilot projects for the densification of biomass. It will be up to the selected farms and facilities, in conjunction with the relevant ministries and the supervising engineer, to determine the actual plant location, taking into account - seasonal and overall residue availability; - the shortest possible in-farm transportation distances; - future cropping patterns; - yield expectations; and - overall plant or farm lay-out. Personnel Recruitment and Training 10.7 Ultimately, each plant's viability will rest on hiring of the proper personnel and their provision with the necessary instruction on system operations, equipment use and machine maintenance and repair. According to the ECMC, recruiting personnel with the required managerial and technical skills will not be a problem for the proposed coffee residue plants in Dilla or Addis Ababa. However, the employment of apprcpriate staff for the plants on the State Farms will be more problematic due to the remote locations of these plants. The pro3ect has provisions to fund adequate housing facilities which need to be constructed before or along with the plant. Other infrastructural requirements and fringe benefits made available to State Farm personnel should be investigated. Talented and motivated plant managers should be hired first so that they can be involved in the final project design and implementation, including the recruitment of well-qualified employees. As this is a high-priority project for the nation's energy sector, it is assumed the GO will give commensurate assistance with the mobilization of necessary human resources. 10.8 It is unlikely that qualified personnel for all aspects of each project can be found. Therefore, suitable candidates should be made available from the MCTD, the NSF, the MME and other ministries for training in specific skills that will be required for project implementation. In-country training is preferred to overseas arrangements as relevant information may be residue- or site-specific, costs need to be kept down and trainers should be aware of local conditions. Much of the equipment-related training should be arranged for via conditions in the tender documents. - 129 - Organizational Structure 10.9 Though this has been discussed in Chapter 9, its importance should be stressed with regard to project implementation. A managerial system must be devised for operations within each plant. This could be prepared by each newly-hired plant manager, MME/PMU staff and the supervising engineer in accordance with relevant Ethiopian personnel practices. The status of plant/farm managerial relations and coordination must be delineated. Appropriate farm management, ministerial and project personnel could work out these details. Inter- plant relations and operations within the overall project should be determined with regard to budgeting, disbursement of funds, supply of spare parts, specialized technical expertise, marketing, pricing, etc. The structure of interaction between the overall project and relevant ministries should be coordinated through the MME and might include the MCTD, XSF, Ministry of Transportation, KELPA and the Ministry of Agriculture, among others. Finally, there must be an organizational link between the project and the World Bank during the implementation and monitoring stages. Civil Works 10.10 Once the location of the plant has been determined, site preparation and building activities can commence. It is assumed that all building and civil engineering activities can be carried out by local contractors and/or governmental construction services. For housing, standards should be similar to those presently applied on state farms and construction should be completed when permanent personnel are employed. For plant buildings, the supplier of the plant will have to provide basic dimensions with his bid. Within two months after ordering, the supplier must provide details on required foundations, electrical cable, gutters, etc. in order to facilitate building design. The equipment supplier should be directed to design the plant in such a way that all components are self-supporting. Installation and Start-Up 10.11 The installation of the plants' equipment should be carried out under supervision of the supplier, assisted by maintenance and operational personnel who have been hired for the future plant as well as the supervising engineer. When equipment installation is completed, the supplier should make sufficient trial runs to test the system's performance. Residues will have to be available for this purpose. Then, initial operations can start. During this stage, the supplier will have to prove that the plant in question achieves the req:.ired capacity and that the briquettes meet agreed-upon quality standards. - 130 - Full-Scale Overation 10.12 After the performance teats, full-scale operation should begin. Having received in-country training, plant personnel should be able to operate and maintain all equipment as well as the entire collection/preparation/densification/storage system. However, it may be advisable to receive select assistance from suppliers during the first year of operations for equipment that proved troublesome during trial runs. Spare Parts 10.13 Frequently used spares should be stocked at the plant sites while those which are not in great demand could be obtained from the dealers when necessary. For critical spare parts which are generally not available in Ethiopia from private dealers, sufficient stocks should be maintained to cover at least one year of operation. Replacements should be ordered quarterly. If similar equipment is installed in all plants then there will be a greater pool of parts within the country. For both collection and plant equipment, reconditioning of worn-out items should be carried out at the plant itself whenever possible, with assistance from State Farm workshops, when necessary. Public Education 10.14 Prior to marketinR and concomitant with trial production of briquettes, a public education campaign should be instituted in Addis Ababa to sensitize consumers to the availability and use of this new fuelwood substitute. The campaign does not have to be extravagant as (1) there are limited media and organizational channels available in Ethiopia, and (2) with fuelwood becoming so scarce, any competitively priced substitute is likely to do well in the marketplace if it exhibits the required characteristics for domestic and/or industrial use. Transportation 10.15 The lack of reliable transport could be a bottleneck when it comes to project implementation. First, seafreighted equipment has to be delivered from the port of Assab to the various project sites. Then, once the plants are operational, their residues must be trucked to the market in Addis. Currently, a substantial portion of Ethiopia's transportation resources has been redirected to the top priorities of drought relief and population resettlement. This will continue to have an impact on transport availability for the foreseeable future. The - 131 - alternative would be to bring materials by train from Djibouti to Awash station (in the case of the cotton pilot plant) or to Addis for tranoshipping to the other sites. Additional transport difficulties may arise in delivering finished briquette. from the pilot plants to the capital. The tightest squeese will occur during harvest season when trucking is devoted to hauling the crop. As this io a high priority project for the GO0, the hauling of residue briquettes should be evaluated on a priority basis by the Ministry of Transport. An agreement or contract could be formulated to guarantee regularly scheduled pick-ups from each plant's briquette store. Marketing 10.16 To ensure successful marketing of the product, well-organised distribution mechanisms must be in place prior, to initial production. Depending on the agreements reached with individuals, organisations and/or governmental units, storage facilities may need to be arranged within the target market. For Addis Ababa households, FAWCDA's new charcoal warehouses may provide an option if they are not being fully utilized. In case initial marketing delays are encountered which result in a backlog of briquettes, an agreement should be obtained from the Ministry of Industry to guarantee purchase of surpluses for use in industries capable of burning briquettes which were initially not selected to receive them. Residue-Specific Considerations Coffee 10.17 Several issues need to be considered prior to and during implementation of the various coffee residue components. Overall, the NCTD, HSI and MM3 will need to coordinate closely on issues relating to briquette marketing, distribution and wholesale pricing. This coordination should be undertaken within the proposed P1U of the MMB. At the Mercato site, provisions should be made the appropriate technical personnel to monitor and evaluate the comparative performance of the piston and screw press briquettors. In addition, adequate plans must be prepared for the transfer of equipment from the Mercato to the Kaffa Cleaning Plant site when the new washed coffee processing plant is moved to the new site. At Dilla, the production of 1,000 tonnes of washed coffee pulp briquettes should commence only after the plant is operating smoothly for at least one year on coffee husks. Cotton 10.18 Two key issues may affect implementation of the cotton residue briquetting pilot plant. First, there are disease transmission - 132 - considerations associated with cotton storage. The Swiss Tropical Institute has determined that the heat or pressure involved in briquetting cotton stalks kills off pests. Thus, the storage of finished briquettes should not pose a problem. However, there is less certainty as to whether chopping will completely prevent the breeding of Pectinophora and Diparopsis pupae. Thus, the chopped cotton stalk piles need to be closely monitored to assure that there is no outureak of the pink or Sudan boll worm. If this is done in conjunction with agronomists at the farm and enterprise level, then pest control can be arranged quickly, should it become necessary. Second, residue collection should be well-timed so that it occurs after, and does not preclude, grazing of the fields by cattle. Friction with migratory herders which has been previously encountered when grazing was restricted could lead to production delays. Wheat 10.19 Prior to the briquetting of wheat straw, two matters should be arranged. An agreement should be reached with the Livestock Corporation's Animal Feed Enterprise as to which portions of the Dixis Farm it will still have access to for the baling of straw to be used as animal feed. As the MSF encompasses both the Livestock Corporation and the Southern Corporation, this matter can be resolved internally. Next, there is a shortage of skilled and manual labor in and around the Dixis Farm. The project includes provisions for hiring and accommodating both types of personnel but they will have to be recruited and placed at the site in advance. Maize 10.20 Several implementation issues need to be considered specifically at the maize residue briquetting site. First, changes in the current harvest practice need to be instituted. Workers must no longer bend the harvested stalks over and will need to pile cut stalks in swaths. The former requires some training and supervision while the latter is already being done in a limited fashion for the construction of firebreaks and, as such, is a known mode of operation. Next, adequate labor should be recruited and retained during the harvest season as there have been shortages at this time in the past. Finally, the potential development of hydropower to replace part or all of the diesel generating sets should be investigated thoroughly once the plant has entered into smooth operation. Sensitivity Analysis 10.21 A series of analyses were conducted to determine the sensitivity of delivered costs for each briquetted residue to changes in production capacity; energy requirements per tonne of briquettes; discount rate for capital charges; and the service life of briquettors. - 133 - The most significant result of these tests was that an increase in production capacity from 5,000 to 8,000 tonnes per year at each site should reduce the delivered cost of briquettes by more than 25X. Increasing the process energy requirements per tonne of briquettes by 501 increases the delivered costs of the briquettes by a maximum of about 61. Raising or lowering the equipment discount rate by 21 did not have a substantial impact (less than a 3S change) in the delivered cost of briquettes. Varying the service life of briquettors over the range from 6 to 20 years also had only a marginal impact on delivered costs. A summary of the sensitivity analysis is presented in Figures 10.1 through 10.4. A tabular set of sensitivity analysis data is presented at the end of Annex I. 10.22 The theoretical maximum production from each briquetting site is 13,140 tonnes per annum, with each plant having an output of 1.5 tonnes per hour, operating 8,760 hours per year. For operational and managerial reasons, the technical maximum is assumed to be 801 of this figure, or 10,512 tonnes per year. Thus, in the proposed project where plants are producing only 5,000 tonnes of briquettes per year, the briquetting equipment is working at slightly less than 501 of technical capacity, which is assumed to be reasonable, given the remote location of plants and the use of an unfamiliar technology. If production were increased by 3,000 tonnes per plant annually, each site would be producing at about 751 of technical capacity and the delivered cost of cotton stalk briquettes would decrease by 27Z, wheat straw briquettes by 261 and maize residue briquettes by 251. In terms of comparative energy value, the cotton briquettes would come out at $18.50 per useful GJ, wheat at $17.20/GJ and maize at $19.50/GJ. Then, both cotton and wheat residue briquettes would be more cost competitive than fuelwood which is priced at $18.60/useful GJ in Addis Ababa. Project Budget and Disbursement Schedule 10.23 A breakdown of each project's foreign and local costs for capital items, operating costs and transportation charges is presented in Table 10.1. The overall project budget would be approximately $6.52 million, of which $5.01 million are foreign and $1.51 million are local costs. A more detailed plant-by-plant budget is available in Annex 1. The project disbursement schedule is summarized in Table 10.2. Monitoring and Supervision 10.24 To review and coordinate the activities of all pilot plants, a supervising engineer will be hired to facilitate implementation and to implement a technical monitoring and evaluation program over a period of three crop seasons. This person should be a qualified mechanical engineer with some basic knowledge of electrical engineering. The - 134 - supervising engineer will be involvd in the following phases of implementations - manufacture, works assembly and training of local labor; - erection, installation, test runs and comissioning; - the breaking-in period; and - three seasons worth of production. Throughout these phases, work will include regular reporting on project progress to the Vorld Bank. A more detailed set of terms of reference is provided in Annem 6. Igpleuentation Schedule 10.25 Below is the implementation timetable for the overall comercial scale pilot briquetting plant set of projects. - 135 - Fiallz 10.1. VARIATIONS IN PLANT PRODUCTION CAPACmY 400 360 300 150 X 100 _ ISO .. 50 c I I d 100 200 30 40 60 60 7 8 CVAPAf (TONSWY") A COfteHu* - DuO eL Cofhe FkXchmen - Now SROe e Co000e Parchment - Metot0 a Ctn9 Sota - AuIsh P ComnSto&Sltoer Wad 8e*-308B - 136 - Pipu?e 1042 VARQIONS IN BRIQUETflNG ENERGY REQUIREMENTS 1W0 140 F 130 120 110 100 so A e- 0 -k 40 *~~~~ C.e.u.k-OF &6 D % 2040 t. %c*LatG IN PowE REJRMneN A Cofeke-OIMO & C4oeS PaxCtwWa -tWSf C: CvfoPorImtent - mercdot a Cdtbn Slb -b Ed -30Uot P~~ Corn Stots& P~~~ *'1~VAeor*-Mwo,o - 137 - Fi ure 10.3 VARIAltONS IN THE DISCOUNT RATE 170 160 _ 140 - 130 120 _ m.__ '10 _ 100 90 80 . 70 60 50 40 .. .....********O*O 6% 8% 10% 12% 14% 16 18% 20% DscOUNT RAE (%) A. Ccofe HUsk - Vill & Cals. Parchment - Now Slte C: Couse Paprchmet - Mercato D Colton Stks -D s EU Wheat Mai3e P CoinSt9s & Stoqir Wodd Bank-30384:O - 138 - Figure 10.4 VARLATIONS IN THE BRIQUETTING EQUIPMENT SERVICE UFE 140 130 - 100 t 210 __ - _ _ - - 0 80 A 70 - 60 40 6 8 10 12 14 16 18 20 SERa fl fE (YEARS) 5-~~~CL A- Cdae Husk - DDb k Cdffee Parchent - Now Sit C: Coffee Parcment - Mecto Da Cotton Sktal- E Wjheat Moe E, Com Saks & StW nk 1 1 Table 10.1t PROJECr BUDGET Coffee Parchment Coffee Husks Cotton Wheat maIze ItemPlont Hurcato New Slte a/ Dila Pulp b/ Awash Dixis Anger Total I. Capital: Foreign 299,610 12,540 421,350 60,010 945,560 688,410 1,165,820 3,793,300 Local 12,140 46,090 95,500 47,620 203,00m 222,990 298,370 926,510 Subtotal 311,750 58,630 516,850 107,630 1,149,360 1,111,400 1,464,190 4,719,810 2. OperatIng Costs: Fore?gn 10,700 10,710 28,590 13,170 143,190 149,970 180,310 536,640 Local 42,620 44,130 88,760 27,640 97,220 95,940 147,960 544,290 Subtotal 53,320 54,840 117,350 40,810 240,410 245,910 328,290 1,000,930 I s-a 3. Transportation: Foreign 11,400 11,400 164,630 32,930 151,430 161,550 151,430 684,770 Local 600 600 8,670 1,730 7,970 8,500 7,970 36,740 Subtotal 12,000 12,000 173,300 34,660 159,400 170,050 159,400 720,810 4. Total Investment: Foreign 321,710 34,650 614,570 106,110 1,240,180 1,199,930 1,497,560 5,014,710 Local 55,360 90,820 192.930 76,990 306,990 327,430 454.320 1*506.8't Grand Total 377,070 125,470 807,500 183,100 1,549,170 1,527,360 1,951,860 6,521,55O a/ Incremental costs to Increase output from 2,500 to 5,000 tonnes per annum. bJ CollCtlon equlpment and incremental costs to produc 1,000 tonnes of pulp briquettes. Table 10.2: DISUIENT SOEUMLE Ye I Year 2 Yer 3 Ycr 4 Project Foreign Local Total Foreign Local Total Foreign Local Total Fore1gr Local Total Coffee Hussk 421,350 95,500 516,850 193,220 97,430 290,650 Coffee Parchmnt 299,610 12,140 311,750 22,100 43,220 65,320 12,540 46,090 58,630 22,110 44,73 66,840 (lsrcatc & Ne Site) marcato NOW Site Cotton Stalk 945;5t0 203,800 1,149,360 294,620 105,190 399,810 Wheat Straw 818,410 222,990 1,111,400 311,520 104,440 415,960 maize Residues 1,165,820 298,370 1,464.190 331,740 155.950 487,690 a Coffee Pulp __ _60.010 47.620 107.630 46.100 29.370 75.410 Total 720,960 10t7,640 828,600 2,049,290 567,440 2,616,730 1,844,510 601,710 2,446,220 399,940 231,050 630,000 Running total , costs 720,960 107,640 828,600 2,770,250 675,000 3,445,330 4,614,760 1,276,610 5,891,550 5,014,700 1,507,660 6,521,550 Running total, output (tonnas) - 7,500 17,500 26,00 Table 10.3: 111FUWIATION SOEUL row I Y7W 2 Yew 3 project lias 1 2 3 4 567591 011112 1 2354 56 7869 to 1 12 1 23 4 56 769 10 it12 DssigniAgieeroing ..* Tendering (specSf * iclatn5, biddIng. evalwatIon) Plant (a) mNaufactwre ----h-- ........... (b) Dal ivwy msa (e1 Installation es.. (d) Start-up '.) Ful I-*caIo O pe ation ---- --- --- ---- --- --- --- ---- --- --- --- *....Coollectloa ealuspent.... (a) Del iverye..aaafe (C) Opeation -:a:aaaa:a: -----------m::saua........ .....Trallog- maae.ee.e FAV - a ~* cofee bask *parchment a cotton stalk . abet straw a seize reslduas 4 coffee pulp