Joint UNDP/World Bank Energy Sector Management …documents.worldbank.org/curated/en/105411468749750… ·  · 2016-07-17ghana saiwill residues utilization study volume i - technical

Joint UNDP/World BankEnergy Sector Management Assistance Program

Activity Completion Report

No. 074A/87

Country: GHANA

Activity: SAWMILL RESIDUES UTILIZATION STUDY

(VOLUME I - TECHNICAL REPORT)

OCTOBER 1988

Report of the Joint UNDP/Wdd Bank Energy Sector Management Assistance ProgramThis document has a restricted distnbution. Its contents may not be disclosed withoutauthorization from tne Government, the UNDP or the World Bank.

Pub

lic D

iscl

osur

e A

utho

rized

Pub

lic D

iscl

osur

e A

utho

rized

Pub

lic D

iscl

osur

e A

utho

rized

Pub

lic D

iscl

osur

e A

utho

rized

ENERGY SECTOR MANAGEMENT ASSISTANCE PROGRAM

PURPOSE

The Joint UNDP/World Bank Entrgy Sector Management AssistanceProgram (ESMAP) was started in 1983 as a companion to the EnergyAssessment Program, established in 1980. The Assessment Program wasdesigned to identify and analyee the most serious energy problems indeveloping countries. ESMAP was designed as a pre-invesetment facility,partly to assist in implementing the actions recommended in theAssessments. Today ESMAP carries out pre-investment activities in45 countries and provides institutional and policy advice to developingcountry decision-makers. The Program aims to supplement, advance, andstrengthen the impact of bilateral and multilateral resources alreadyavailable for technical assistance in the energy sector. The reportsproduced under the ESMAP Program provide governments, donors, andpotential investors with information needed to speed up project prepar-ation and implementation. ESMAP activities fall into two majorgroupings:

- Energy Efficiency and Strategy, addressing the institutional,financial, and policy issues of the energy sector, includingdesign of sector strategies, improving energy end-use, defininginvestment programs, and strengthening sector enterprises; and

- Household, Rural, and Renewable Energy, addressing the tech-nical, economic, financial, institutional and policy issuesaffecting energy supply and demand, including energy fromtraditional and modern sources for use by rural and urbanhouseholds and rural industries.

FUNDING

The Program is a major international effort supported by theUNDP, the World Bank, and bilateral agencies in a number of countriesincluding the Netherlands, Canada, Switzerland, Norway, Sweden, Italy,Australia, Denmark, France, Finland, the United Kingdom, Ireland, Japan,New Zealand, Iceland, and the USA.

INQUIRIES

For further. information on the Program or to obtain copies Afthe completed ESMAP reports listed at the end of this document, contact:

Division for Global and OR Energy Strategy, ManagementInterregional Projects and Assessment Division

United Nations Development Industry and Energy DepartmentProgramme World Bank

One United Nations Plaza 1818 H Street, N.W.New York, N.Y. 10017 Washington, D.C. 20433

GHANA

SAIWILL RESIDUES UTILIZATION STUDY

VOLUME I - TECHNICAL REPORT

OCTOBER 1988

AB8C - Architectural and Engineering ServicesATP - African Timber and Plywood (Ghana) Ltd.BRRI - Building and Road Research InstituteCIDA - Canadian International Development AgencyeCw - Electricity Corporation of GhanaEEC - European Economic CommunityEBP - Export Rehabilitation ProgramFAO - Food and Agriculture OrganizationFD - Forestry DepartmentFPIB - Forest Products Inspection BureauFPRI - Forest Products Research InstituteGIHOC - Ghana Industrial Holding Corporati"nCOG - Government of GhanaCRC - Ghana Railway CommissionCTMB - Ghana Timber Marketing BoardCWA - Cliksten West Afica Ltd.IIED - Interna' nal Institute for Environment and DevelopmentMIN - 1lim Tim_r Co., Ltd.MLNR - Ministry of Lands and Natural ResourceODA - Overseas Development AdministritionPNDC - Provisional National Defense CouncilSIPI. - Subri Industrial Plantations Ltd.STC - State Transport CorporationSTP - Specialized Timber Products Ltd.TDRI - Tropical Development Research InstituteTEDB - Timber Export Development BoardTVLC - Takoradi Veneer and Lumber Co., Ltd.VRA - Volta River Authority

ABBUEVITIE

a - annumAbs - abscluteADO - Automotive Diesel OilCIP - Cost, Insurance and Freightcm - centimeterC&E - Construction and EquipmentDCF - Discounted Cash FlowDM - Deutsche Mark1IRR - Economic Internal Rate of ReturnFAS - Free Aboard ShipPFl1 - Financial Internal Rate of ReturnFOB - Free on BoardfUa - from and atCJ - gigajouleGWh - gigawatt-hour

h - hourha - hectareHa - MercuryHHV - Higher Heating Valuehl - hectalitersHP - horsepowerIDO - Industrial Diesel OilIFO - Inland Fuel OilIgal - Imperial gallonin - inchkg - kilogramkJ - kilojoulekm - kilometerkPa - kilopascalkVA - kilovolt-amperekV - kilowattkWh - kilowatt-hour1 - literLHV - Lower Heating ValueLRMC - Long Run Marginal Costm - meterN - millionMCaI - megacaloriemcdb - moisture content, dry basisicwb - moisture content, wet basisMD1 - Medium Density Fiberboardmin - minuteNJ - megajoulemm - millimeterMo - monthMW - megawattM.T. - metric tonneNPV - Met Present ValueOD - oven dryO&M - Operations and MaintenanceRFO - Rosidual Fuel OilRWE - Round Wood EquivalentSCF -S tandard Conversion FactorSWE - Sol8i Wood Equivalentt - metric tonneTC - turtogeneratortoe - tonnes of oil equivalenttonne - metric tonneUSD - U.S. DollarWTP - Willingness-to-Payyr - year

micY AND FURL BQUIVALES

cumcY

1 VI$ - 150 Cedi

COUVERSON FACTORS

1 MJ 948 Etu239 Kcal

* 0.278 kWh

mcwb LHV HHVFuel () (NJ/kg) (NJ/kg)

Sawmill Residues 36 11.9 20.0Fuelwood. air-tried 30 13.1 20.0Sawdust Briquettes 5 18.9 20.0Charcoal 5 29.0 30.2Crude Oil -- 43.3Gas Oil (ADO) 43.3 45.5Industrial Diesel Oil (IDO) 42.1 44.6Inland Fuel Oil (IFO) 40.1 42.8Residual Fuel Oil (RPO) 39.8 42.5Electricity - 3.6 a/

a/ NJ/kVh

Costs of Utiization.............................................. 46Technical/Infrastructure Constraints to Residues

Utilization ......... 47Water Spray Lubricdtion of Saw Blades*................ 47Outside Storage of Sawdust........................... 47Boiler/Furnace Configuration......................... 47Boiler Efficiency......e........................... * *7

V. POTENTIAL ON-SITE ALTERNATIVES FOR IMPROVING AND/ORINCREASING USE OF WOOD INDUSTRY RESIDUES AS FUEL........ 48Summary ....................................................... 48On-Site Utilization................................................ 50

Backgroun.................................................. 50Sawmill Process ..................................... 50Cogeneration at Grid Connected Mills................. 57Cogeneration at Non Grid Connected Mills............. 64

Residue Handling and Combustion EfficiencyImprovements ................................................... 83~~~~~~~~ ~~~~~~~83Backgrounds ** **#*o ................e........................... 83Saw Guide and Sawdust Storage Improvements*........... 84Furnace Modifications for Sawdust Combustion .....0... 85Boiler/Furnace Efficiency Improvements................ 85

VI. POTENTIAL OFF-SITE ALTERNATIVES FOR IMPROVING AND/ORINCREASING USE OF WOOD INDUSTRY RESIDUES AS FUEL ........ 86Summary ....................................................... 86Off-Site Utilization............................................... 88

Background ......................e.... ,****oo 88Substitution of Sawdust for Oil Fuels Consumption.....*. 88

Financial and Economic Analysis...................... 88Substitution of Sawdust for Fuelwood Consumption*....0.. 90

Financial and Economic Analysis...................... 91Off-Site Conversion Alternatives e....................... 92

Background .................................................... 92Improved Charcoal Production............................ 93

Present .e00 ............. 93Improved Methods ......... 96Financial and Economic Analysis of ImprovedCharcoal Options ................ o *.. 106

Sawdust Briquetting ..... 109

Proposed Plants...................................... 117Charcoal Briquettes ........... 124

Background....................................... ....... 124Production Options/Economics......................... 125

VII. CONCLUSIONS AND REOOMENDATIONS FOR INCREASED AND,ORIMPROVED UTILIZATION OF WOOD INDUSTRY RESIDUES. ........ 128Suuunary... ...*.*............................................ ~128On-Site Direct Utilization.............................. 128

TABLE OF CONTS

EXECUTIVE SMAY.......... ....... ............. ........ ........... i

TM INTRODUCTION 1

Sackope o ....... ,............. ......... ......... ... ...... 3Scope of Suy3Organization of Report...B e p o rt.. ............ 000 e0e 4

II. TUE GHANAIAN WOOD PROCESSING INDDUSTRY ......oo........o.... 6Sector Performance...................................... 6Type, Capacity and Location of Wood Processing

Facilitiest ieooooooeee*o..... ......... o.o..o.... 10Industrial Timber Production.ootucti...oo.o. oono... o.... 13Wood Industry Trends Affecting Residues Disposition.... 15

Trend to Greater Value-Addedue-Atd..oeoo.oed.o. ooooo 15Trend Toward Exploitation of Secondary Species.ooooos 16

III. SUPPLY OF WOOD INDUSTRY RESIDUESo.oo... ......o.o......o.o. 17s.umry. ...... .... ............... ................... 1?Sources, Types and Characteristics of Residuestues.o.... 17Location of Residues...... i d u es.... .................... 21

Forest Residueso......*.**..... ................... oo 21Wood Processing Residuesi.dou esooo...oo.o.o.oo.o...o. 22

Quantities Producedod...... ...0.0.00.0... .0000 .0 0000..0 23Existing Sto c k p i le.... 00.000.........................0. 25Reliability of Supplies p p l ie.oo... o..... oo..o...o..o.... 26Present and Projected Surplus r p lu..o.......o.o.os...... 27

Present Surpluso............ o ..............eo.o ..oo 27Future Surplusr p l us.................................. 29

IV. DEMAND FOR WOOD PROCESSING INDUSTRY RESIDUES..*........ o 31Summry .... ............. ................................ 31On-site Energy Uses and Disposalo....oo.o.o.ooo...o..o 31

Steam and Process Heeatoooo...o....................... 31Cogenerationo.oo*oooo*o,ooooo00 0 34Charcoal Production..oood. u c t i on... ...... ......... 36D ispo sal..o..oooooooo. oo ... ooooooooo... o.... oo....oo 37

Off-site Energy Usesoosooooes...o.oos..o..ooooo..o..oooo 37Industrial and Co_mercial Heat Raising sing**...o.oo.. 37Domestic Cooking.ooki.n g .ooo.o.o...ooo . .oo..e... 38Charcoal Productionoo....................0...0...0..0 38Briuti qgooo u e t t i ngooooo*o***ooo* 38

Energy vs. Non-Energy Uses... ses..................0..0.. 38Secondary Manufacturing and Export p. o rt.........o... 39Cottage Industry Woodworking........o.o..oo.......... 39Particle and Fiber Board Productiono.o..o.ooo.o..o.oo 39

Summary of Present Residues Utili zation................. 39

Sawmill Process Heat...............*.***.,**,0.9......000 128Cogenerationo.........................0.0..o.o....... 131

Direct Utilization in the Industrial Sector............. 132Residue Substitution for Oil Fuels................... 132Residue Substitution for Fuelwood.................... 133

Conversion Alternativeso ........... o.o...eo...oo..o..o 133Improved Charcoal Making............................. 133Sawdust Briquetting.....0.o...0.o.00oq.e.eO.*0o.0.... .133Briquette Carbonization.............................. 134

National Investment Implications........................ 134Total Investment Potential........................... 134Competition for Residue Resources.................... 135Residue Utilization Priorities....................... 136

Uncertainties and Risk Factors.......................... 138Health of the Wood Processing Industryoo..00.0 ...00.. 138Location of Wood Processing Facilities............... 138Wood Supply/Demand................................... 138Oil Prices ...o...e.o..o...oo.. 4*o*o*ooooooo* 138Electricity Costs.o000* * oo....................... 139

Inveatment Recommendations.. .o.....o o.................. t39

Sawmill Process Beat........................ 139Sawdust Briquettingo................ 0000000000000000 139

Technical Assistance Recommendations.................... 140Pilot/Demonstration Projects .o....o........o...o.ooo. 140

Improved Solid Residues Carbonization ........0 0..... 140Briquette Carbonization. 0 0 0.000o0.00.0o0............o 141

Policy Recommendations....o............................. 141Sawmill Process Heato...................... .0.0...... 142Residue Conversiono................................... 142

Areas for Further Investigation........................s 142

TABLES2.1 Value Estimates for Industrial Forest Products in

Export and Domestic Trade.o.. o... ...... e........*o.*o 9

2.2 Industrial Timber Processing Facilities, 1986........... 102.3 Distribution of Sawmills by Size of Production, 1986.... 112.4 Distribution of Production by Sawmill Size, 1986........ 122.5 Annual Cut from High Forest............................. 142.6 Projected Distribution of the Industrial Hardwood

Timber Harvest...o.v.s.... .o......... 0.0..0..0 14

3.1 Densities and Moisture Contents of SelectedGhanaian Woods.....................s 18

3.2 Moisture Content of Typical Saw Timber Species Mix ....... 193.3 Fuel Characteristics of Selected Ghanaian Woods......... 203.4 Nominal Characteristics of Wood Processing

Industry Residues ..... .... 21

3.5 Sawmill Residue Production, 1986........................ 233.6 Combined Mill Residue Production, 1986.................. 253.7 Wood Processing Industry Residue Production, 1986 ....... 253.8 Surplus Sawdust Concentrations....0*. 0*..........0 .....o 29

3.9 Planned Additions to Wood Processing FacilitiesUtilizing Mill Residue for Process Heat Generation... 30

4.1 Wood Processing Facilities Utilizing Kill Residuefor Process Heat Generation, 1986.................... 32

4.2 Wood Processing Facilities Utilizing Mill Residuefor Co-Generation of Steam and Power, 1986........... 35

4.3 Wood Processing Industry Disposition of Residuesby Region, 1986......................... 40

4.4 End-Uses of Wood Processing Industry Residuesby !.ype, 1986........................................ 42

4.5 Disposi..on of Wood Processing Industry Residuesby End-Use, 96.........................4

5.1 Matrix of Technical Options for Improving and/orIncreasing On-Site Residue Utilization .. 60*-*4000.. 49

5.2 Wood Processing Facilities Visited by Miss9ion**on..... 505.3 Secondary Species Requiring Kiln Drying..o..oo...0..g 525.4 Summary of Capital and Annual Operating Costs for

Sawmill Process Heat Unit Production Model.oo..o... 545.5 Potential Gross Contribution to Forest Industry

Value Added through Lumber Kiln Dryingoo ing.......... 565.6 Financial Analysis Results, Swmill Process Heat

Unit Production Motl 575.7 Economic Analysis Results, Sawmill Process Heat

Unit Production Moee l 575.8 Summary of Incremental Capital and Annual Operating

Costs for STP Cogeneration Alternative 2......... 605.9 Summary of Incremental Capital and Annual Operating

Costs for STP Cogeneration Alternative 3......... 625.10 Pinancial Analysis Results, Grid Connected

Co-Generation Models (STP).... 63

5.11 Economic Analysis Results, Grid ConnectedCo-Generation Models (8TP ) 64

5.12 HIM Cogeneration Options Decision Matrix..0*0*.........0 675.13 Mim Area Electrical Demand and Consumption, 1985...5.... 695.14 Utilization of Residues at MIM 705.15 Charcoal Sales Prices, Mim Timber Co 705.16 Summary of Capital and Annual Operating Costs for

MIM Cogeneration Alternatives........o........ee... 195.17 Net Present Cost of MIM Alternatives with No Grid

Extension Assumed .............. . 815.18 KIM Alternative 3A vs. IA Financial and Economic

Analysis Beet ...... .....815.19 KIM Alternative 5B vs. Base Case B Financial and

Economic Analysis Results....... ..................... 836.1 Matrix of Technical Options for Improving and/or

Increasing Off-Site Residue Utilization.............. 876.2 Industries and Institutions Visited to Evaluate

Potential for Direct Utilization of Wood Residue..... 896.3 Financial Analysis Results, Oil to Sawdust-Fired Boiler

Conerson ....................................90

6.4 Economic Analysis Results, Oil to Sawdust-Fired BoilerConversion ......... ~ ........ .......................... 90

6.5 Financial Analysis Results, Fuelwood to Sawdust-FiredBoil'r Conversion............................ ........ 91

6.6 Economic Analysis Results, Fuelwood to Sawdust-FiredBoiler Conversion...... .........*......00.0.000..... 92

6.7 Earth Mound Charcoal Kilns, Ruia s i 966.8 Mill Residue Charcoal Marketing, Kumasi................. 966.9 Comparison of Charcoal Knl n s 996.10 Summary of Annualized Capital and Operating Costs

of Charcoal Production Alternatives.................. 1006.11 Financial/Economic Analysis Results, Beehive Brick

Kiln Charcoaling Improvement......................... 1086.12 Chaowus Ltd. Briquetting Machine Specificationsations*** 1106,13 Chaowus Ltd. Briquette Fuel Characteristicsoo0.00...0000 1116.14 Estimated Sawdust Briquette Demand m a nt.................. 1136.15 Financial and Economic Costs of Fuelwoodl.............o. 1136.16 Briquettes vs. Fuelwood Comparison in Bread Bakingt

Accraccra............................................ 1156.17 Briquettes vs. Fuelwood Comparison in Brick

Manufacturing, Cape Coast... 1156.18 Briquettes vs. Inland Fuel Oil Comparisont Accraa....... 1266.19 Characteristics of Proposed Sawdust Briquette

Manufacturing Plants..l a n ts.........................0 1186.20 Summary of Capital and Annual Operating Costs for

Proposed Briquetting Plantsa.n t.s.......o............ 1296.21 Summary of Briquette Production and Transport

Financial Costs........... ........................... 1216.22 Financial Analysis Results, Proposed Sawdust

Briquetting lns 1216.23 Estimated Wood Balance, 1986-20008..... 1226.24 Economic Analysis Results, Proposed Sawdust

Briquetting Plants ..... .... 1237.1 Net Economic Contribution to Export Lumber Unit

Value through Kiln Dr y i n g 1297.2 Sawmill Process Heat Investment Potentia1............... 1307.3 Total Investment Potential for Increased and/or

Improved Utilization of Wood Industry Residues....... 1347.4 Net Benefit/Residue Resource Ratio for Energy Uses

in Kumasi 1357.5 Residue Energy Utilization Priorities .................. 1367.6 Residue Utilization Profiles for Kumasim.... 137

FIEURES2.1 Timber Export Volumes, 1976-1986; 1990 (Projected)...... 72.2 Timber Export Values, 1976-1986; 1990 (Projected)....... 82.3 Distribution of Sawmill Capacity by Size, 1986.......... 123.1 Sawmill Residual Fractions vs. Recovery

Fraction, 1985 ..... 243.2 Surplus Residues by Type, 1986.......................... 28

3.3 Surplus Residues by Region, 1986 ....................... 284.1 Residue Utilization by Region, 1986 ..................... 414.2 Residue End-Us by Type, 1986........................... 434.3 Residue Disposition by End-Use, 1986 .................... 455.1 STP Ltd. Schematic - Alternative 1...................... 535.2 STP Ltd. Schematic - Alternative 2..................... 595.3 STP Ltd. Schematic - Alternative 3...................... 615.4 MIM Complex Daily Load Profile .......................... 665.5 Mim Timbers Ltd. Schematic - Base Case A/

Alternative 1A...........................725.6 Mim Timbers Ltd. Schematic - Alternative 2A ............. 745.7 Mim Timbers Ltd. Schematic - Alternative 3A............. 765.8 Him Timbers Ltd. Schematic - Alternative 4A............. 776.1 Typical Earth Mound Charcoal Production................. 946.2 Tropical Development Research Institute (TDRI) Steel

Charcoal Kiln ........... 986.3 Clay/Metal Charcoal Kiln (Subri Semi-Mobile Kiln)....... 1016.4 Missouri Charcoal Kiln (Subri River Project)............ 1036.5 Casamance Charcoal Kiln........... ...................... 1056.6 Beehive Brick Charcoal Kiln............................. 1076.7 Continuous Briquette Charcoaling Machine................ 126

BIILIOGRAPHY

MAPIBRD 20619: Chana Wood Processing and Other Industrial Centers

EXECUTIVK SUNKARY

1. Ghana harvested 1.07 million m3 (round wood equivalent) ofi3dustrial hardwood timber in 1986 and consumed or exported about 600,000m of timber products or logs. The difference, which can be consideredas wood industry residues, had an energy value of 94,000 toe. Aboutthree-quarters of these residues were utilized at varying efficiencies inthe domestic sector and in the wood processing sector itself. Theremaining 23 percent, with 22,000 toe energy value, found no usage.Efficient use of these residues as a source of energy, especially in thewood processing industry, could have a significant impact on thefinancial health and foreign exchange earnings of the sector. Thissector accounted for 6.2 percent of GDP in 1985 and approximately 6.4percent of total export earnings in 1986. Converted surplus residues,after accounting for mill consumption, could significantly contribute tothe energy demands of the commercial and household sectors. Thesepossibilities and the investment requirements necessary to realize themare discussed below.

Sector Performance (Chapter II)

2. Improving the performance of the Ghana forestry industry hasbeen the primary target of the Export Rehabilitation Program (ERP)financed by credits from IDA, the Overseas Development Adminstration andthe Canadian International Development Agency. As a result, productivityin the sector has demonstrated marked gains since 1984 after experiencinga sharp drop in output during the period 1976-1983. At present, onlyabout 55 percent of wood industry processing capacity is utilized. Inthe short term, further gains in capacity utilization are unlikely due toforest resource constraints. Total timber harvest is expected to remainclose to the 1986 level through the end of the century. However, twoindustry trends have been identified that have implications for residueproduction and consumption:

(a) increases in the level of domestic processing for higher value-added; and

(b) a shift to higher production of secondary (non-traditional)species.

The primary effects of these two trends will be to change the quantity,type and characteristics of the residues generated as well as most likelyincrease the on-site utilization of residues as a result of increasedneeds for process heat.

Supply of Wood Industry Residues (Chapter III)

3. Types. Wood industry residues can be bruadly classified intotwo major categories: solids (slabs, edgings, offcuts, veneer wastes andcores); and fines (sawdust, planer shavings and sander dust). Solids

- ii -

accounted for 79 percent of the residues produced while sawdust accountedfor the remaining 21 percent. Table 1 presents a summary of thedistribution of residues by type.

Table 1: PRODUCTION OF WOOD INDUSTRY RESIDUES, 1986(3 SUE)

Slabs & veneerEdgIngs Offcuts Sawdust Waste Cores Total

quantity 213,175 66,747 93,269 34,723 26,698 434,612

Percent 49% 1% 21% 9% 6% 100%

4. Location. The production of wood industry residues isprimarily concentrated in a few major wood processing centers. In fact,66 percent of the total residue production is concentrated within an 8 kmradius in the Kumasi area. An additional 23 percent is distributedbetween three ' other centers: Takoradi (9X); Mim (9Z); and Akim-rda(5X). The remaining 11 percent is scattered amongst eight otherlocations.

5. Future Supply. So long as the wood industry continues itsrecovery, total residue production is expected to remain fairly stablethough the types and characteristics of the residues could changeslightly. Increases in the level and quality of processing and shifts toharvesting of more secondary species will be the primary factorsinfluencing change in the characteristics of the residues. On the whole,it is expected that effects will be counterbalancing and that the energyvalue of residues will also remain fairly stable.

6. Surplus Residues. Only 23 percent of wood industry residuesare presently not utilized. The total surplus residues have an energypotential of 27,000 toe or equivalent to 13 percent of the 1985industrial woodfuel consumption. Green (wet) sawdust is the only residuein abundant surplus accounting for 84 percent of the total availablesurplus residues. Ninety percent of all sawdust produced is not utilizedand present disposal costs are estimated at approximately US$80,000 to125,000 per year. This does not account for the potential environmentaldamage as a result of burning (smoke) or dumping (leaching).

Demand for Wood Processing Industry Residues (Chapter IV)

7. End Uses. Table 2 provides a summary of the 1986 end-use ofwood proceising industry residues. At present only 23 percent of theresidues are consumed for on-site process heat generation orcogeneration. Approximately 32 percent is used off-site for firewood or

- iii -

charcoal production while 22 percent is used for non-energy purposes(e.g. local furniture, fencing, etc.).

8. On-Site Use. There are 21 wood processing facilities thathave wood-waste fired boilers for generation of steam and process heat.Almost all are combination mills producing sawn timber, plywood and/orveneer. These mills generally consume 50 to 60 percent of their woodresidues. There are four mills that have cogeneration equipmentinstalled on-site. However, only one currently operates in acogeneration mode; the others have either experienced equipment problemsor have not completed their installation. When cogenerating, these millscould consume most of their residues to meet their heat and electricityrequirements. At present, only two major mills are not connected to thegrid.

Table 2: END-USES OF WOOD PROCESSING INDUSTRY RESIDUES, 1986Cm SUE)

Fuel for MtIl Firewood/ Non-Energy SurplusPracess Heat/Cogeneration Charcoal Production Purposes Residues Total

Solids 92,235 136,468 96,341 16,177 341,343

Flues 6,674 3,195 - 83,400 93,269

Total 99,031 139,663 96,341 99,577 434,612(23S) (32%) a/ (22%) (23%) (100%)

a/ includes 3% on-site charcoal production at MIN.

9. Off-Site Use. More than half of the wood industry residues areused off-site; 29 percent is for energy and 22 percent for non-energypurposes. Of the residues utilized off-site for energy, about 28 percentis used directly as firewood for food preparation in the commercial andhousehold sectors. The majority, 70 percent, is converted to charcoal inprimitive earth mound kilns. The resulting output represents 2 percentof the total charcoal production in Ghana. Only 2 percent is sawdustwhich is converted to briquettes at a private plant in Oda for sale tobakers and brickmakers. Of more than 60 industrial sites surveyed in thevicinity of the wood processing industries, only one is a regularconsumer of unprocessed sawmill residues. Other industries usingwoodfuels obtain their firewood from the natural forests at competitiveprices. With the exception of a minor amount of sawdust for briquetting,no other sawdust is presently used off-site for energy.

- iv -

On-site Options for Residue Utilization (Chapter V)

10. Opportunities exist not only to utilize surplus sawdust butalso to improve the present use cf solid residues so as to obtain themaximum economic benefit from all residues produced. The mosteconomically attractive on-site options for improving and/or increasingresidue utilization are:

(a) Steam generation to meet process heat needs for kiln drying andwood treatment;

(b) Improved saw blade guide and sawdust storage/handling systemsto reduce water content in the sawdust residue and increase thenet energy available;

(c) Furnace modifications to enable substitution of sawdust forsolid residues; and

(d) Furnace improvements for greater combustion efficiency.

Investments in cogeneration systems proved to be marginal at best and arenot recommended. Table 3 summarizes the estimated financial and economicIRRs for the major investment. Detailed discussions and evaluations ofeach are presented in Chapter V and highlighted below.

Table 3: SUMMARY OF ESTIMATED FINANCIAL AND ECONOMIC IRR FOR ON-SITERESIDUE UTILIZATION OPTIONS

OPTION FIRR EIRR

Saw.ill Process Heat 27-56S 30-52S

Cogeneratlon at SrId Connected Mills Negative 4%8

Cogeneration at Non-grid Connected Mills (MIN) o/ Negative <12%

oZ Comparison Is against the option of grid extension costs.

11. Sawmill Process Heat. Results of the analysis on the possibleuses for residues at a typical large sawmill point to the potential forsignificant economic gain from oa-site generation of process heat forwood treatment and kiln drying. Major economic benefits are derivedfrom:

v

(a) Higher value-added through export of kiln dried products; and

(b) Increased range of economically exploitable species through logsterilization and lumber drying.

Financial and economic IRR exceeding 30 percent are possible. Nationalinvestment requirements in the range of US$ 7.4 to 13.8 million would berequired (depending on the level of value-added processing as3umed) withresulting net annual economic benefits in the range of US$ 2.7 to 4.5million respectively.

12. Sawdust Utilization: Benefits could be maximized if provisionsare made to utilize surplus sawdust as a fuel at mills. This wouldentail incorporation of measures to improve the fuel quality of sawdustsuch as more efficient sawblade guide systems and sheltered sawdusthandling and storage systems. Retrofit costs are minor compared to othersawmill capital investments. Additionally, provisions must be made toinstall boilers with appropriate combustion systems for direct sawdustutilization.

13. Coseneration. Almost all large sawmills in Ghana are connectedto the national grid and therefore receive the benefit of low cost hydro-power. The marginal financial cost of residue fueled cogeneratedelectricity at these large mills is estimated at 5.6 to 7.1 UScents/kWh. With present industrial tariffs set in the range of 3.5 UScents/kWh, there are no financial incentives for mill owners to undertakeinvestments in cogeneration. Lack of financial incentives is not theonlv barrier. Marginal economic costs of residue fueled cogeneration isestimated at 5.8 to 7.6 US cents/kWh which is still higher than thepresent estimated marginal cost of 5.2 US cents/kWh for grid system elec-tricity. In the short term at least, residue fueled cogeneration at saw-mills cannot be supported. In the longer term however, the marginal costof grid system electricity is expected to rise as all low cost hydropowersites in Ghana have been fully exploited. Based on data in a recentsystem expansion study for VRA by Acres International, an estimate ofLRNC for electricity in the range of 8 US cents/kWh is not unrealistic.Given this scenario, residue fueled cogeneration could be competitive.It must be pointed out, however, that the total potential power from thissource would likely not amount to more than 20 MW whereas the Acres studyindicates a need to bring on between 200 to 400 MW by the mid 1990s.

14. Two large mills are presently located off-grid. One, AfricanTimber and Plywood Ltd. (AT&P) located in Samreboi, is not expected to beconnected to the national grid in the foreseeable future. Its optionsfor electricity are either diesel generation or residue fueledcogeneration. AT&P is presently undergoing renovation with Bank ofScotland financing and expects to cogenerate its electricity when itrecommences operation in 1988. Clearly, in this case residue fueledcogeneration is more economic that diesel generation. The other mill,Mim Timber Co, Ltd. (MIM) located 47 kM from Sunyani, is scheduled to beconnected to the grid in 1989 when VRA undertakes a US$2.2 million grid

- vi -

extension. To date, construction has not yet begun on the extension. Ananslysis of the least-cost electricity supply option at MIM, assuminggrid extension to be uncommitted, confirms that grid extension is thebest choice. A 1.2 MW base/intermediate load residue-fueled cogenerationplant coupled with existing diesels for peaking power results inmarginally cheaper electricity when grid extension costs areconsidered. A US$2.1 million investment is required for the cogenerationplant yielding an economic IRR of only 12 percent. A recommendation fora cogeneration plant cannot be strongly supported given the additionalmanagement and risk factors associated with operating the cogenerationplant, the financial disincentives for MIN management, and the fact thatsocial benefits to the surrounding area are not fully provided.

Off-site Options for Residue Utilization (Chapter VI)

15. Several options are available for improving and/or increasingthe off-site utilization of wood industry residues. As indicatedearlier, over 70 percent of the residues used off-site for energy areconverted to charcoal in primitive, inefficient earth mound kilns. Over89 percent of the total Sawdust production is unutilized. And, thepresent known demand for s'wdust briquettes exceeds the supply. Giventhese facts, four key options were investigated in detail:

(a) Direct utilization of sawdust in industrial/commercial oil-fired or wood-fired combustion systems;

(b) Improved charcoal producti,n techniques;

(c) Increased sawdust briquetting capacity; and

(d) Introduction of briquette carbonization techniques.

A summary of the estimated financial and economic IRR for the first threeoptions is presented in Table 4. Detailed discussions and evaluations ofa range of these options are presented in Chapter VI and highlightedbelow.

Table 4: SUMMARY OF ESTIMATED FINANCIAL AND ECONOM'CiRR FOR OFF-SITE RESIOtE UTILIZATION OPTIONS

OPTION FIRR EIRR

Conversion of Industrial/Comercial Negative NegativeOil-fired Combustion Systems

Conversion of Industrial/Commercial 11-23% a/ 32-42% a/Wood-fired Systems

Improved Charcoal Production 300% 500%Techniques

Increased Sawdust Briquetting Capacity Negative-19% b/ 9-19% b/

I/ Represents the IRR from two of the most attractive candidates.b/ Variable depending on scale of project.

- vii

16. Substitution for Oil Fuels. The economics of substitutingunprocessed sawdust. or solid residues for oil fuels (RFO/IFO) inindustrial/comercial combustion systems in Ghana are unfavorable.Evaluation of a wide range of possible industrial/ccmmercial candidatesindicated that petroleum prices would have to rise above US$ 30/bbl in1986 dollars before savings from fuel costs could amortize the capitalcosts required for conversion. Alternately, capital costs would have todecline by 40 to 100 percent in order to yield favorable returns. One ofthe primary reasons for the negati -e economics is the fact that mostpotential industrial/commercial candidates have low utilization factorsassociated with their combustion equipment and therefore a relatively lowbase from which fuel savings can be derived.

17. Substitution for Fuelwood. The economic potential forsubstituting unprocessed sawdust for fuelwood consumption inindustrial/commercial combustion equipment is limited. Conversion ofmost fuelwood combustion equipment to utilize sawdust is generallyfeasible requiring modifications to the grate and feed systems. However,haulage costs for sawdust and the low financial costs of fuelwood limitthe possible candidates to those in the immediate vicinity of thesawmills.

18. Improved Charcoal Making. Almost all charcoal production fromsawmill residues is produced by the traditional "earth mound" method.The conversion efficiency of most of these charcoaling operations wasfound to be extr ,ely low. The earth mound technique is generally usedin forest and land clearing operations because it requires littleequipment and can move with the resource. Sawmill residues provide anexcellent opportunity to utilize improved charcoaling techniques, therebypotentia11y doubling the output of charcoal derived from these residueswithout increasing residue consumption.

19. After evaluating several improved charcoaling techniques, itwas determined that Beehive brick kilns and Casamance kilns were the mosttechnically and economically attractive options for charcoaling sawmillresidues in Ghana. If all residues presently carbonized in only theKumasi area were converted in Beehive brick kilns, charcoal output wouldincrease by 4,400 tonnes/yr or equivalent to an 80 percent increase overthe present charcoal production from residues. The estimated totalinvestment required to achieve this improvement is US$ 110,000 whichwould yield an economic rate of return near 500 percent. Clearly, aprogram to improve charcoal production from sawmill residues should behigh on the agenda.

20. Sawdust Briguetting. Loose sawdust is cumbersome to manage,transport and use and as stated earlier has limited off-site potential.When briquetted, the sawdust can be effectively transpor'-i, stored andutilized. The potential for sawdust briquettes has already beenestablished in Ghana with the operation of a 2,000 tonne/yr plant inOda. Present demand greatly exceeds supply. Total estimated demand fromjust the bakers and brick manufacturers in the main urban areas is in

- viii -

excess of 45,000 tonnes/yr which is about equal to the quantity ofbriquettes that could be produced if all surplus sawdust were to bebriqu4tted.

21. Evaluation of three different capacity screw press sawdustbriquetting plants indicates that definite economies of scale exist withrespect to this technology. A 14,000 t/yr plant in Kumasi iseconomically attractive with financial and economic IR estimated at 19percent. A plant of this scale would consume just over half the surplussawdust in the Kumasi area yet meet only one-third of the potentialdemand for briquettes from the bakers and brick manufacturers. Totalinvestment for the plant is US$780,000, of which 24 percent is localcosts.

22. Briguette Carbonization. While sawdust briquettes are anacceptable fuel-wood substitute in the commercial sector, they would facedifficulties penentrating the household market where charcoal andcharcoal stoves predominate. Sawdust briquettes produced by the heatextrusion screw press systems used in Chana could be carbonized either inseparate kilns or in a carbonization tunnel appended to the last stage ofthe briquetting machine. Carbonization by the kiln method is proven andis practiced in Japan and Taiwan. The carbonization tunnel is stillexperimental but it has potentially significant energy efficiencyadvantages. Assuming a 32 percent yield of briquettes to charcoal andaccounting for capital and operating costs results in charcoal briquettecosts in the range of US$ 80 to 125/tonne. With charcoal FOB exportvalue estimated at US$115 and charcoal prices in Takoradi ofUS$132/tonne, the possibility for a viable charcoal briquette marketexists. However, the relatively experimental nature of this technologyrequires follow-up investigation before actual investments in this areaare undertaken.

Conclusions (Chapter VII)

23. Investments. Investments in sawmill process heat generation,sawdust briquetting and improved carbonization systems provide the mostattractive options for improving and/or increasing the use of woodindustry residues. Total potential levels of investment exceed US$8million with economic rates of return ranging from 19 to 46 percent persub-project. If a high level of log treatment and kiln drying isassumed, total investments in just process heat equipment could approachUS$14 million. A summary of the total investment potential andassociated residue consumption is presented in Table 5.

- ix -

Table 5: TOTAL INVESTMENT POTENTIAL FOR INCREASED AND/OR IMPROVED UTILIZATION

OF WOOD INDUSTRY RESIDUES

Investment Annual ResidueInvestent Location Amount EliR NPV Consumption

Sawmill US$7.4 N

Process National at present wood 46S USS20.4 N up to 80,000 m3Heat Industry output

SawdustBriquetting Kumasi USS0.78 N 19% US$0.50 M 27,000 m3Plant

ImprovedCharcoalingof Residues National a/ up to USS0.11 M 490% USS1.11 N 64,000 m3

a/ Amounts given are for Investment for improved utilization of all residues presentlycarbonIzed in Kumasl.

24. Competition for Residues. Options for residue utilization aremutually exclusive to the extent that they might compete for the sameresidue resource in terms of type and location. The primary source ofcompetition would likely occur for sawdust in Kumasi if both maximumprocess heat generation and sawdust briquetting were promoted. The totalsawdust production in Kumasi would not necessarily be sufficient underthis scenario. Valuation of residue use for various options provides abasis for prioritization. The resulting residue utilization prioritiesare presented in Table 6. They indicate that sawdust should be usedfirst at the mill boilers to the extent determined by process heat demandand technical feasibility of sawdust combustion. Solid residues shouldbe used only to meet additional demand not met by sawdust. RemainirZsawdust should be briquetted if available in sufficient quantities torealize necessary economies of scale. Remaining solid residues should beconverted only in efficient charcoal kilns such as the Beehive brickkiln. An analysis of the situation in Kumasi based on thisprioritisation would allow for construction of the 14,000 t/yrbriquetting plant.

x

oable 6: RESIDUE ENERGY UTILIZATION PRIORITIES

Priority Ranking Utilization

1 Combust sawdust on-site for process heatgeneratlon/cogeneration,

2 Combust solid residues on-site for process

heat generatlon/cogeneration.

3 Convert solid residues to charcoal In

efficient kiIns.

4 Convert sawdust to briquettes.

5 Utilize solid residues as firewood.

Recommendations (Chapter VII)

25. Investments. As a result of this investigation the following

key investment recommendations can be stated:

(a) Primary attention should be focused on investments to promote

the on-site utilization of sawmill residues for generation ofprocess heat used in value-added processing;

(b) Attention should also be focused on the possibility of

investments in the 14,000 t/yr briquetting plant in Kumasi;

(c) Any investment in increasing residue utilization should

incorporate technical assistance and resources to minimize

water contamination of sawdust through improved saw blade

guides and storage and handling systems;

(d) Financing should be provided to institute a program to convert

all sawmill residue charcoaling operations to improved methods;

(e) If donor funding is made available, a small briquette

carbonization project should be instituted to demonstrate the

technical and economic feasibility of this option.

26. Policies. Government policies on residue utilization can have

a major effect on the ultimate disposition of waste wood resources. With

appropriate policies of fuels pricing, tax/fee levels, regulation, and

market information dissemination the government can guide the development

and use of this indigenous resource. Specific policy recommendations are

- xi -

discussed in Chapter 7. Key policies recommended to improve residueutilization include:

(a) Market development and information dissemination relating toexport opportunities for kiln dried and treated lumber;

(b) Differential taxation on value-added forest products, andphased extension of log export ban to wider group of marketablespecies (with attendant increase in requirements for lumbertreatment and drying);

(c) Institution of permit/fee systems for wood residue dumping anda ban on open residue incineration; and

(d) Provision of domestic loan financing and training schemes,possibly from National Energy Board resources, for improvedcharcoaling of wood industry residues.

27. Further Investigation. Logging residues constitute more than1.0 million mJ/yr of waste wood with an energy potential of more thandouble that available from the wood processing industry residues. Todate only a small fraction of these residues are recovered. Clearly, aninvestigation into the logistics and economics of recovering thispotentially more significant source of indigenous energy is needed tocomplement any efforts to improve and/or increase the utilization of woodresidues. Subjects to be addressed include methods of carbonizing wastewood from timber operations in the forest, incentives that would berequired to induce a shift of traditional charcoal-making into thelogging areas and present technical, institutional and infrastructuralconstraints to wide scale charcoaling of forest residues.

I. INTRODUCTION

Back!round

1.1 The Joint UNDP/World Bank Energy Sector Assessment Programconducted a comprehensive review of the Ghana energy sector in October,1985. The results of this review were contained in a final report issuedin November, 1986, entitled, Ghanas Issues and Options in the EnergySector. The assessment determined that, in 1985, energy end-uses in theeconomy were supplied primarily by biomass fuels (i.e., woodfuels andagricultural residues), 72 percent, and imported petroleum, about 24percent. Only 4 percent of end-use energy demand was met by electricityobtained primarily from two large hydropower plants on the Volta River.

1.2 While forest resources are ample in Ghana, they are subject tosignificant over-exploitation especially for meeting the large woodfuelneeds of the rural and urban populations. The Bank/UNDP EnergyAssessment concluded that this resource could be seriously depleted by.the early 1990s if deforestation pressures are not eased. Similarlyunsettling is the fact that in 1985 imported petroleum absorbed 26percent of the foreign exchange earnings of Ghana. The Assessment statesthat even if petroleum prices were to remain at two-thirds of their 1985levels and Ghana succeeded in increasing exports, petroleum imports wouldstill account for nearly 20 percent of foreign exchange earnings unlessexpansion of indigenous energy sources were to become viable. In helpingidentify possible options to improve the energy situation in Ghana, theAssessment identified the nearly 1.2 million tonnes/year of wood wastesgenerated in logging and sawmilling as a potentially important source ofindigenous energy. The Assessment estimated that recovery and efficientuse of these residues could reduce the offtake from the natural forestsby at least 10 percent, or 0.8 million tonnes/yr. 1/

1.3 Given the conclusions of the Ghana Energy Assessment, theBank's Energy Sector Management Assistance Program (ESMAP) is executing aseries of studies to assist the Government of Ghana (GOG) to exploit thepotential of its wood wastes. The first undertaken, as represented bythis report, was a study of the feasibility for increasing, and/4r.improving the use of sawmill and wood processing industry residues. 'hestudy was co-financed by the Canadian International Development Agency(CIDA) and was initiated in June 1986. A reconnaissance mission to Ghanafielded in that month identified the main issues to be addressed in thestudy. In addition, two Ghanaian consultants were engaged to conduct acomplete survey of wood processing industry residue production and energyconsumption patterns and also to identify and survey potential commercial

1/ Ghana: Issues and Options in the Energy Sector, Report No. 6234-GH,World Bank, November, 1986.

-2-

and industrial wood energy consumers. 2/ An ESMAP mission followed inOctober 1986, consisting of a mission leader, energy economist, woodindustry/residue specialist, biomass conversion specialist and biomasscombustion/energy applicat-,ns engineer. 3/ The data collected by boththe local consultants and the ESMAP mission have been extensivelyevaluated and the results are documented in this report.

Objectives

1.4 The main objectives of this study are to identify and evaluatetechnically and economically feasible opportunities for improving and/orincreasing the use of wood processing industry residues as an energysource. In this context, the specific issues evaluated include:

(a) The rate of production, availability and location of differenttypes of wood industry residues, and the characteristicsdetermining their energy potential;

(b) The technical and economic potential of different opportunitiesfor utilization of the residues both at the site of residueproduction as well as externally;

(c) The infrastructure, social and policy constraints presentlyinhibiting the increased use of wood industry residues as anenergy source; and

(d) The technical assistance and investment requirements necessaryto realise the economically feasible potential of Ghana's woodindustry residues.

1.5 On the basis of the evaluation of the above issues, the finalobjective of this ESMAP study is to formulate a comprehensive strategyand follow-up program to promote, where economic, the increased and

2/ "Survey of Wood Residue Generation and Utilization in Ghana," EsselBen Hagan (Consultant) and Martin Ben-Dzam (Consultant - WoodIndustries Specialist), Ru-Tek Consultants and Industries Ltd.,Kumasi, Ghana.

3/ This report is based on the findings of a mission which visitedGhana from October 17 to November 7, 1986. The mission members wereMessrs. Matthew Mendis (Mission Leader), Charles Feinstein (EnergyEconomist), Brian Hickman (Consultant - Wood Industry-Residues),Peter Neild (Consultant - Biomass Conversion), and Philip Trees(Consultant - Biomass Combustion/Energy Applications). The reportwas authored by Messrs. Mendis and Feinstein. Secretarial supportwas provided by Vernet Mason.

- 3 -

efficient use of Ghana's wood industry residues. In this regard,particular emphasis is placed on:

(a) Identifying and evaluating potential pilot/demonstrationprojects which would help establish the effectiveness andviability of increased and/or efficient residue utilization;

(b) Defining investment requirements for a program to increaseutilization of surplus residues both at wood processingfacilities as well as in other commercial and industrialenterprises;

(c) Defining appropriate policy instruments andorganizational/.nstitutional measures required to support therecommended program; and

(d) Identifying areas for further investigation and/or developmentwhich will help promote efficient use of wood residues.

Scope of Study

1.6 This study focuses only on the wood processing industryresi,ues. It does not consider forest residues produced in the processof logging or land clearing. Of the total estimated 1.2 million tonnesof wood residues produced in Ghana annually, wood processing industryresidues account for approximately 342,000 tonnes or about 28 percent.However, this represents the most readily accessible and inexpensivelyobtained portion. The forest residues, while constituting a significantquantity, require additional resources to collect and transport topotential markets. At present some unknown portion of these residues iscollected and converted to charcoal to supply the more luv ive urbanmarkets. Aspects for increasing the recovery of forest -,a, whileworthy of further investigation, are not covered in this -

1.7 In conducting this study, over 90 percent - the woodprocessing industries were directly surveyed to establish present andprojected patterns of residue production, consumption and disposal. Inaddition, surveys of energy consumption patterns at major industrial andcomercial enterprises in the vicinity of the wood processing areas werealso conducted to help identify opportunities for wood residueconsumption. The study also investigated present means for conversion ofresidues to either charcoal or sawdust briquettes for transport todistant markets. In the case of charcoal, on-site observations werecarried out to characterize charcoaling operations including conversionefficiencies, productivity and economics, with the intent of identifyingmethods for improving charcoal conversion. A detailed analysis of thesawdust briquetting operation in Oda, one of only a few commerciallyoperating plants in Africa, was also undertaken to assess the scope forsimilar briquetting plants elsewhere in Ghana.

-4-

1.8 The study considered the inputs of over sixteen GOG agencieseither directly or indirectly related to the forestry, wood processingand energy sectors. These inputs were essential in understanding theorganizational and institutional influences on residue utilization.Information on sector performance and trends was ottained to assist inthe evaluation. Finally, fuelwood and charcoal prices at thewholesale/retail levels were surveyed at the major urban markets toensure incorporation of current energy price information in thesubsequent evaluations.

Organization of Report

1.9 The report is structured to provide a clear picture of theoverall potential and investment requirements necessary to improve and/orincrease the use of wood industry residues as an economically attractiveenergy source in Ghana. Chapter II provides a brief profile of the Ghanawood processing sector including past, present and projectedperformance. Emphasis is placed on industry trends affecting residueproduction and disposition, especially as it relates to processing forgreater value added and exploitation of secondary species. Chapter IIIpresents information on the supply of wood industry residues includingsources, types, characteristics, locations, quantities, stockpiles andsurpluses. Chapter IV contains the demand side of the picture outliningpresent on-site and off-site uses including energy versus non-energy usesfor the residues.

1.10 Potential options for improving and/or increasing the use ofwood industry residues as a fuel are covered in Chapter V for options atthe mill sites and Chapter VI for off-site options. Both generic andspecific case studies are evaluated to determine technical, financial andeconomic feasibility. In the case of on-site options, emphasis is givento use of residues for process heat generation and to cogeneration ofelectricity. Economics of cogeneration at both grid and non gridconnected mills are evaluated. Residue handling and combustionefficiency improvements to increase the quality and usefulness ofresidues are also investigated. Off-site options considered included:substitution of sawdust for petroleum and fuelwood consumption; improvedcharcoal production; sawdust briquetting; and briquette charcoaling. Inall cases, detailed cost estimates for implementing the proposed optionare developed as a precursor to the financial and economic assessment.Key organizational and institutional factors that could alter the resultsare identified.

1.11 Conclusions regarding increased and/or improved utilization ofwood residues are summarized in Chapter VII. Economically attractiveoptions identified in Chapters V and VI are extrapolated to determinetheir national potential. Possibilities of cross competition forresidues are accounted for by prioritizing the alternative utilizationsusing a benefits/scarce resources ratio. Total national requirements to

implement all economically attractive options are then estimated.Uncertainties and risks associated with investments in this sector arebriefly discussed. Lastly, the chapter presents recommendations for keyinvestments and technical assistance. Pilot projects to demonstratetechnical and economic feasibility of options unfamiliar to Ghana areidentified along with pro,ected costs. Policy recommendations to helppromote the economically attractive options for improving and/orincreasing residue utilization are presented at the conclusion of thereport.

- 6 -

II. TME GCHEtAI VOOD PROCESSING INDVSTRY

Sector Performance

2.1 In parallel with the general decline in the country's economicsituation, the Ghana forest product sector experienced a sharp drop inoutput during the period 1976-1983. However, due largely to changes inGovernment of Ghana (COG) policies and increased donor assistance, thewood industry has demonstrated a substantial recovery since reaching alow point four years ago. A twelve year summary of the value ofindustrial forest product values is given in Table 2.1. Forestry andlogging contributed 5.9 percent of the 1984 GDP; 1985 figures areestimated at 6.2 percent of the total, or about 7 percent if themanufacturing wood industry is included. 4/

2.2 Export performance is shown graphically in Figures 2.1 and 2.2.The strength of the present recovery can be gauged by noting thatestimated 1986 forest product exports of US$ 48.0 million represent a 65percent increase over comparable figures for 1985, or a performance whichhas not been achieved in absolute terms since 1978. The drop in exportgrowth rates represented by the 1990 projections reflect limitations onsupply of the most readily marketable species, as will be discussedlater.

2.3 The COG has received credits for forestry rehabilitation anddevelopment from IDA totalling US$25.9 million for: (a) the ExportRehabilitation Project (Credit 1435/SP9-GH), and (b) the ExportRehabilitation Technical Assistance Project (Credit 1436-GH), and hasalso received credits granted by the Overseas Development Administration(ODA) and the Canadian International Development Agency (CIDA). TheExport Rehabilitation Program (ERP) concentrates on the establishment ofquick-disbursing IDA credits to the forest industry to (a) financepurchase of equipment, materials and spare parts to enable an increasein production and exports, and (b) restructure the timber marketingorganizations which would enable the forest industry to export with aminimum amount of Goverment control. A total of sixty-two mills havereceived credit assistance under IDA. In addition, the ERP has providedtwo Government-owned industries with technical assistance. ERPassistance has been targetted at removing bottlenecks in order to restorebasic log extraction and primary productive capacity.

4/ Draft Forest Sector Review, West Africa Projects Department,Agriculture Division, World Bank.

-7-

Figure 2.1: TIMBER EXPORT VOLUMES, 1976-1986; 1990 (PROJECTED)

1486

368

.6 77 78 79 88 81 82 83 84 8S586 87 88 89 98Year

23 Logs z Lumbera/ World Bank projection.b/ Lumber includes secondary wood products such as profiles, flooring

and furniture components.Sources TEDS; World Bank.

Figuie2.2: TIMBER EXPORT VALUES, 1976-1986; 1990 (PROJECTED)

538~28

* - -_ ___ __ U

76 7778 79 8881 828384 8586 8788 89913Year

a Logs E3 Lumber

al World Bank projection.b/ Lumber includes secondary wood products such as profiles, flooring,

and furniture components.Sources TUB; World Bank.

Table 2.1: VALUE ESTIMATES FOR INDUSTRIAL FOREST PRODUCTS IN EXPORT AND DOMESTIC TRADE(Millions of Current US Dollars)

Est.

Year 1974 1975 1976 1977 1978 1979 1980 1981 1982 1983 1964 1985 1986

Rcorded Exports 78.5 78.8 91.4 72.6 63.8 44.5 41.3 20.4 15.3 14.8 18.7 29.1 48.0

Domestic warket 29.3 34.6 61.8 111.1 64.9 63.6 64.4 49.0 46.9 27.6 30.3 49.5 -

TOTAL 107.8 113.4 153.2 183.7 128.7 108.1 105.7 69.4 62.2 42.4 49.0 79.6 -

Source: World Bank, TEDS.

- 10 -

Type, Capacity and Location of Wood Processing Facilities

2.4 The wood processing industry in Ghana consists, for the mostpart, of primary producers such as sawmills, plymills, veneer plants, andcombinations thereof. A small amount of secondary manufacturing existsproducing small items such as mouldings, broom handles, parquet flooringand knock-down furniture components. Some 72 of the active mills inGhana are straight sawmills, i.e. performing log break-down and producingsawn timbers for domestic consumption and export. An additional 21 millsproduce rotary veneer/plywood (principally for the domestic market)and/or sliced veneers (primarily for export). The plywood and veneermills are nearly always sited next to a sawmill; if not physicallylocated in the same complex, then common ownership and management assurethe input log supply. For this reason, sawmill facilities which includeplywood or veneer lines will be referred to as combination mills. Thenumber and regional distribution of the industrial timber processingfacilities is shown in Table 2.2. Not included in the compilation areapproximately 12 forest mills (so-called "bush mills") which arescattered mostly within the high forest zone surrounding Kumasi and whichemploy gasoline-engined mini-saws. Their small output and remotepossibility of residues recovery preclude study consideration and werethus not surveyed.

Table 2.2: INDUSTRIAL TIMBER PROCESSING FACILITIES, 1986

Rotary Veneer/ SlicedRegion Sawmills Plywood MIIls Veneer Mills

Ashanti 43 4 7Brong Ahafo 8 - 1 a/Central 3Eastern 6 1 2Western 12 4 2

Total 72 9 12

a/ On-site but not operational.

Source: Ru-Tek.

2.5 Based on the GOG's Ministry of Land and Natural Resources(NLIR) guidelines, sawmills can be classified as large, medium or smallaccording to their annual saw log input capacity:

- 11 -

Classification Annual Saw Log Input (m3)

Large >10,000

Medium 5,000 - 10,000

Small 5,OOO

Ghana sawmills have been classed in Table 2.3 according to their size ofestimated 1986 actual production. The Ashanti region, comprising 43sawmills almost all of which are located within an 8 km radius of eachother in the Kumasi area, is the dominant wood processing zone. TheWestern region is of second significance in terms of log volumesprocessed and is clustered around the centers of Sekondi-Takoradi,Samreboi, and Sefwi-Wiawso. The wood industry of the Brong Ahafo regionis dominated by the state-owned Mim Timber Co. Ltd., the largest singlesawmill in Ghana, while the mills in the Eastern region are located inAkim-Oda and Nkawkaw. The three wood processing facilities of theCentral region are situated in Dunkwa and Cape Coast.

Table 2.3: DISTRIBUTION OF SAWMILLS BY SIZE OF PROOUCTION, 1986

Mills with Annual Production of a/Regilon <5000 m3 5000-10000 m3 >10,000 u3 Total

Ashanti 14 11 18 43

Brong Ahafo 6 - 2 8

Central 2 - 1 3

Eastern 3 1 2 6

Western 3 2 7 12

Total 28 14 30 72

a/ Size classes according to saw log Input.Source: Ru-Tek.

2.6 Large sawmills having lO,OOO m3 or greater annual log inputcapacity account for 72.6 percent of the total installed sawmillcapacity, as shown in Figure 2.3. The five largest sawmills in Ghanaaccount for 25 percent of the total installed sawmill capacity.

2.7 Estimated actual 1986 industrial timber processing mirrorsinstalled capacities and points to the dominance of the large sawmills in

- 12 -

primary production and, hence, residue production. As shown in Table2.4, production at the large class sawmills in the Ashanti, Western andCentral regions account for nearly three-fourths of the total in thoseareas, while in Brong Ahafo the fraction exceeds 80 percent. Only in theEastern region does the share from large mills slip, and then only to61.7 percent.

Industrial Timber Production

2.8 Industrial timber harvesting occurs almost exclusively in thehigh forest zone which covers one-third (8.2 million ha) of the totalGhana land area of 23.9 million ha. Practically all existing forest isunder concession. The concessions have not been rationalized whichresults in certain mills experiencing a shortage of raw material inputand having to operate in others' concessions, often at considerabledistance from the mill. Present knowledge of standing wood volumes isincomplete and dates back to PAO surveys of 1980-1982. The OverseasDevelopment Administration is presently carrying out a new resourceinventory in the high forest zone and preliminary findings are expectedto be available in 1987.

Figure 2.3: DISTRIBUTION OP SAWMILL CAPACITY BY SIZE, 1986

- < 5.000 m3/yrOman (12.2%)

135,0000 U,000 -

mm 5,6~18000 ./y>x m3/ys 168,00 >i/yr

(72.6%)5

FIGURES Mg LOG INPUT IN M3/YM

Source: NLNR.

- 13 -

Table 2.4: DISTRIBUTION OF PROOUCTION BY SAWMILL SIZE, 1986(Percentage of Tota I Product ion)

Mills with Annual Production of a/

Region < 5000 SOOO-10,000 m3 > 10,000 o3

Ashanti 12.8 12.5 74.7

Western 7.4 17.8 74.8

Eastern 19.4 18.9 61.7

Central 27.1 - 72.9

Brong Ahafo 18.3 -- 81.7

a/ Size classes according to saw log input.

Source: Ru-Tek.

2.9 Approximately 180 species occur in the Chanaian forest, of

which a minority have commercial economic value. Silviconsult Ltd. hasclassified the species into three groups according to their degree of

commercialization. 5/ Group A comprises 40 species presently considered

commercial and for which doii,estic and export markets exist. Group B

contains 20 species presently considered marginally commercial. However,these species are considered potentially commercial and are likely to be

increasingly exploited in the near future as markets are developed.

Group C contains the remaining species which grow to a utilizable size.

A listing of Group A and B species by botanical and trade names is given

in Annex 3.

2.10 Fourteen of the 40 species in Group A species are consie red

highly desirable and are banned from export in log form. These so-called

"primary" species are indicated in Annex 3. The remaining species are

considered "secondary" and when harvested are often unrecorded as to

species by the Forest Products Inspection Bureau (FPIB). Thus an

accurate breakdown of annual cut by species is not possible at present.

A geieral comparison of the estimated 1985 commercial roundwoodproduction to the Silviconsult recommended annual allowable cut has been

made in Table 2.5. The comparison reveals that he recommended annual

allowable cut for the primary species of 185,000 ma was already exceeded

5/ The Forest Department Review and the Requirements of the Forest

Products Inspection Bureau and the Timber Export Development Board

(Draft Report), Silviconsult, Bjarred, Sweden, September 1985.

- 14 -

Table 2.5: ANNUAL CUT FROM HIGH FOREST IN RESERVES

1985 Sliviconsult AnnualEstimated RWE Allowable CutProduction bl Recommendation

Species Group (million m ) (mIllilon m3)

Group A 0.523 0.719of which 14 primary (0.225) (0.185)of which 26 secondary (0.298) (0.534)

Group 8 0.004 0.386

Unallocated by species a/ 0.400 -

TOTAL 0,927 1.105

a/ Unrecorded as to species.b/ 1966 RWE product estimated at 1.065 millIon m3.

Source: World Bank, Sliviconsult; TE0O; Mission estimates.

2.11 Table 2.6 depicts the present and projected future distributionof the industrial hardwood timber harvest. Figures for 1986 estimatethat exports itz log form accounted for 18.8 percent of the total harvestof ',065,000 e , and 37.1 percent of the roundwood equivalent (RaE) oftotal exports. As is reflected in the projections, by 1986 the totalhardwood timber harvest had essentially reached the Silviconsultrecommendation for annual allowable cut. Harvest projections thus assumeadoption of a sustained yield resource management strategy. Increases inthe total harvest would be dependent on substantial production occurringin the high forest areas outside reserves and in the estimated 52,000 haof timber plantations. However, neither of these resources issystematically managed at present.

Table 2.6: PROJECTED DISTRIBUTION OF THEINDUSTRIAL HAROWOOD TINDER HARVEST

(thousand m3)

Product Exports Log RWE of RWE TotelLumber a/ Veneers Plywood Exports Exports Domestic Harvest

1986 110 29 1 200 539 526 1,065

1990 124 33 2 251 589 511 1,100

1995 176 28 10 180 589 511 1,100

2000 210 20 30 80 555 545 1,100

a/ Includes secondary wood products such as profiles, flooring, and furniturecomponents.

Source: World Bank.

- 15 -

Wood Industry Trends Affecting Residues Disposition

2.12 Two trends in local wood processing have been identifiedthrough data analysis and industry discussions as having potentiallysignificant impacts on residues production and utilization. These are(a) a trend to further domestic processing for higher value-added, and(b) a shift to higher production of secondary species. Each is expectedto be a gradual trend occurring over the next five to fifteen years. Theprimary effects to be examined in succeeding chapters include changes inthe amounts, types and characteristics of the various residues generated,as well as changes in the demand for these residues within the mills as aresult of increases in processing.

Trend to Greater Value-Added

2.13 Unless new inventory estimates and resultant management plansindicate that a higher level of harvesting is sustainable, production andsector revenues will face a constraint in log availability. A shift togreater value-added per unit log harvested is therefore anticipated inorder to maintain growth in export value in 1990 and beyond. Specificmanifestations of the trend include:

(a) Greater processing of logs into export lumber, so that exportin log form is eventually phased out past the turn of thecentury;

(b) higher product recovery through improved sawmilling techniquesand machinery and installation of more re-saw equipment;

(c) increased secondary manufacture of wood products such asmouldings (profile boards), fiooring, broom handles andfurniture components.

2.14 The rapidity of this shift will be governed by:

(a) Pace of modernization and expansion of wood processingfacilities;

(b) Success in development of expanded markets for sawn andsecondary manufactured wood products;

(c) Pace of improvement of transport, handling and shippinginfrastructure for timber products;

(d) Rate of depletion of commercially valuable timbers, andresultant rise in economic pressures within the sawmillindustry;

(e) Political factors, such as a near ban on log exports asproposed in the Lagos Plan of Action for the Year 2000; and

- 16 -

(f) Government policies such as industrial timber stump&ge feelevels and value-added taxation.

Trend Toward Exploitation of Secondary Species

2.15 As noted, the wood processing industry is already facingconstraints on the availability of the readily exportable fourteenprimary species. These limitations are most evident in the Ashanti andBrong Ahafo regions, and sawmills relying on logs from these areas willbe forced to move to processing of secondary species or face declininglog throughputs. Mill managers in Kumasi report having to go 240 km one-way to log the high demand redwood species, approaching the limits ofeconomic logging given the high transport and logging road constructioncosts. The production manager at Mim Timber Ltd. forsees a 30-50 percentreduction in primary species throughput with a corresponding increase insecondary species production over the next five to ten years, while themanager of Specialized Timber Products in Kumasi envisions a virtuallycomplete shift at his mill in the long run.

2.16 A number of barriers on increased sales of secondary species,especially in export markets, presently exist:

(a) Uncertainty as to the resource base and total availability ofsecondary species. Many overseas customers are unwilling toaccept lesser known species unless a regular supply is assuredfor ten years.

(b) Perishability. Many of the secondary species are subject toattack by staining and rotting fungi unless specializedtreatment and drying measures are incorporated.

(c) Lack of knowledge among processors and customers of speciesworking properties and uses. The Porest Products ResearchInstitute (PPRI) has primary responsibility for investigatingand reporting on these characteristics.

(d) Conservatism of industry and variable market tastes. Marketdemands are dictated by the uncertain whims of "changingfashion," and the current "in" phase of whitewoods is ofuncertain duration.

The Timber Export Development Board (TEDB) is charged under PNDC Law 123with the responsibility "to develop markets for and promote the sale andexport of lesser known timber species." While by far the greater part ofsecondary species exports have been in log form, significant exports inproduct form are expected to begin by 1990.

- 17 -

III. SUPPLY OF WOOD INDUSTRY RESIDUES

Summary

3.1 Estimated total wood processing industry residue productionamounted to 434,612 m3 SWE in 1986, and was composed of slabs/edgings(49%), sawdust (21X), offcuts (15%), veneer wastes (9%) and cores (6%).In their green condition, the residues have an average density of 786kg/mr at a nominal moisture content of 36 percent (wet basis), implying anet energy value of 94,000 toe. Major concentrations of residueproduction occurring in the Kumasi and Sekondi-Takoradi areas coincidewith locations of major non-wood processing industries. Other locationswhere residues are found host little or no non-wood processing industrialactivity.

3.2 So long as the wood industry continues its upward trend ofrehabilitation as in the last three years, it can be safely assumed thatthe supply of residuals is reliable. Log throughput levels will continueto hold the greatest influence on the production levels of residues. Neteffects of trends to increasing product recovery, secondary manufactureand secondary species exploitation on the suitability and reliability ofwood processing residues as a fuel is not expected to be significant.

3.3 Sawdust is the only wood processing residue presently inabundant surplus amounting to some 83,400 mi SWE with a net energy valueof 18,000 toe. Ninety percent of all sawdust produced is unutilised atpresent. Surpluses of solid residues totallinj 15,000 m SWE areprimarily found at the isolated sawmill of Mim Timber Co.

Sources, Types and Chares-eristics of Residues

3.4 The residues produced by the saw, ply, veneer and secondarymanufacturing operations consist of:

(a) Slabs and edgings, including slicer boards and bark strips;

(b) offcuts;

(c) sawdust, including planer shavings;

(d) veneer waste (both green and dry), including dry plywood trim;and

(e) cores (bouls).

The residues are largely in the green condition, although some driedveneer, dry trim and sawdust is present in the veneer and plymills, and

- 18 -

some dried offcuts, sawdust and planer shavings is available in thesawmills and secondary manufacturing plants. The density and moisturecontent of the residues are significant when estimating their tonnage andassociated energy value.

3.5 Table 3.1 contains the densities and green moisture contents ofmany of the more common wood species found in Chana. The densities aregiven in the green and oven dry conditions. The densities and moisturecontents are shown for the average green condition and are not indicativeof changes in these properties due to handling or manufacture. Changesin properties can be expected due to: (a) The addition of water due toexcessive sawguide lubrication; (b) addition of water due toprecipitation when residues are stored in uncovered stockpiles;(c) natural drying in storage; or (d) forced drying in manufacturingprocesses or fuel drying facilities.

Table 3.1: DENSITIES AND MOISTURE CONTENTS OF SELECTED GHANAIAN NOOS

Green GreenMoisture Moisture

Densities Content Content

Oven Dry (kg/) _ Green (kg/n) (Dry Basis) (Wet Basis)

Species Low Avg High Low Avg High (%) (N)

African

Walnut 440 560 640 750 825 900 47.3 32.1

Afromosia - 650 - 950 1,050 1,200 61.5 38.0

Ayan 570 710 860 900 950 1,000 33.8 25.3

Bublnga 600 750 880 1,000 1,050 1,100 40.0 28.6

Candellei - 650 - 900 930 950 43.0 30.0

Ceiba 200 260 400 600 700 850 169.0 62.8

Cordia 160 230 290 700 750 80 226.0 69.3

DOnta 660 720 770 700 750 800 31.9 24.2

Edinam - 520 - - 9O0 - 73.0 42.2Eusri - 500 - 800 825 850 65.0 39.4

Guarea 550 600 700 8a0 900 1,000 50.0 33.3

Hyedus 660 750 880 1,000 1,050 1,100 40.0 28.6

Kyenkyen - 430 - 700 750 80o 74.7 42.8

Kyere 460 500 570 900 925 950 85.0 45.9

Mahogany 420 490 570 650 725 800 48.0 32.4

Mekore 510 590 690 850 9D0 950 52.5 34.4

MNnsonia 540 600 650 850 950 1,000 58.3 36.8

Odum 480 600 670 990 1,045 1,100 74.1 42.6

Sopele 490 620 720 690 890 1,065 43.5 30.31

Utile 450 590 700 750 825 900 39.8 28.5

Wawa 250 350 520 530 590 650 68.6 40.7

Source: Vaa.nfurh/Scheiber Holzatias, Veb Fachbuchverlag, Leipzig, 1974;

Mission estimates.

- 19 -

3.6 The utilization of mixed wood species within the mills mayresult in the accumulation of species mixtures or surges of one or twospecies. There is an advantage to mixing species when the mixture is tobe used as fuel in the green condition as the low moisture content (m.c.)fuels will assist the higher m.c. fuels to burn, and an average densityfuel will occupy less space and ease the fuel conveying and furnaceproblems which may attend very low density fuels, or very wet fuels. Theanticipated shift to secondary species indicates a trend toward lessdense and wetter species. In the short term, this trend will not be seenin certain concessions such as African Timber and Plywood (Ghana) Ltd.(AT6P) and Clikstan West Africa Ltd. (CWA). An evaluation of the averagemoisture content of green residues and the anticipated shift in aboutfive years has been calculated for certain mills surveyed. The analysisis based on their present and future species mix and a-pears in Table3.2. The increase in wet basis moisture content of approximately 1.5-2.5percent at some processing sites would have the effect of reducing lowerheating values (LHV) 6/ of undried residues by 2.8-4.7 percent.

Table 3.2: MOISTURE CONTENT OF TYPICAL SAW TIMBER SPECIES MIXCS)

1986 Est. 1991wet Dry Dry Wet

Mill Basis Basis BasEs Basis

Specialized Timber Prod. 61.1 38.2 - -

A.G. Timbers 49.5 32.9 - -

A,E. Saoud 57.0 36.3 - -

HIM Tiober Co. 52.3 34.3 58.5 36.9

Gliksten (W.A.) Ltd. 46.9 31.9 46.9 31.9

TVLC: Plywood 66.7 40.0 70.6 41.4

TVLC: Timber 52.8 34.5 56.5 36.1

TVLC: Combined a/ 45.4 31.2 48.5 32.5

a/ Includes dried residuals.

Source: Mission estimates.

6/ The LHV represents the energy content of a fuel after the heat ofwater vaporization is deducted.

- 20 -

3.7 The effect on the aggregate moisture content of using a portionof dried fuel is evident from the data for "TVLC Combined" when thefigures are compared to the plywood and timber green mixed speciesinputs. The average moisture content has been significantly reduced byincorporating the dried components.

3.8 The higher heating values (HHV) 7/ of Ghanaian hardwoods aredifficult to find in the literature, but the figures in Table 3.3 areavailable for the eleven species shown.

Table 3.3: FUEL CH^ACTERISTICS OF SELECTED GHANAIAN WOODS

Higher Percent AshSpecies Heating Value, 00 Volatile Content

(NJ/kg) (S) (S)

Wawa 20.17 82.9 1.3

Esa (Celtis) 19.22 81.1 2.1

Ekki 20.61 80.4 0.3

Kyeokyen 19.12 80.9 2.9

Sapele 19.66 81.3 1.0

Doha"s 20.35 81.0 O.S

Danta 20.22 81.2 1.3

Odum 20.24 75.2 2.7

Kane 20.77 82.2 1.0

Celba 18.59 81.7 3.2

Otie 19.19 76.4 3.6

Sources BMI; P-E International.

3.9 Estimated useful average characteristics for Ghana woodprocessing industry residues are summarized in Table 3.4.

7/ The MMV represents the oven-dry or calorific heat of a fuel.

- 21 -

Table 3.4: NOMINAL CHARACTERISTICS OFWOOD PROCESSING INDUSTRY RESIDUES

Oven Dry Density 503 kg/03

Green Molsture Content 36%(vet basis)

Green Density 786 kg/m3

Volatiles 81.3%

Ash Content 1.6%

Higher Heating Value 20.0 MJ/kg(oven dry)

Lower Heating Value 11.9 MJ/kg(at 36% mcwb)

Source: Mission estimates.

3.10 Other important physical characteristics of sawdust and veneerwastes found in Ghana should be noted. Due to the abnormally deep cutsmade in the large diameter logs and cants in Ghana, feed speeds are low,and therefore the sawdust is very finely divided. If not overly wetteddown, this fine sawdust is easily air conveyed and is a very combustiblefuel. It can present a disadvantage in a furnace in that it can becomeentrained in the combustion gases and carried up the stack, causing highemission levels and fly ash problems. Therefore, a furnace approporiateto the combustion of fine sawdust should be used.

3.11 Veneer wastes found at the combination mills come in long andthin sheets, tend to curl, are flexible, and tend to bulk up to largevolumes. These "as clipped" residuals are therefore difficult to handleand convey mechanically and labor-intensive manual methods are commonlyemployed.

Location of Residues

Forest Residues

3.12 Residues from forest operations are available when the treesare felled and are approximately equal to the volume of roundwoodextracted. Butt and top logs, branchwood and non-sawlog material left i the forest from commercial logging thus amounts to some 1.1 million m

- 22 -

annually, or 860,000 t/yr green weight. An unknown amount of theseresidues makes up part of the estimated 8.6 million t/yr nationalwoodfuels gross supply, 8/ but the greater portion remains in the forestto rot over a period of time. Mim Timbers Ltd. is presently the onlymill making use of some forest residues; Afrormosia branches of 40 cmdiameter and larger are fed to Scanstyle Ltd. for processing intofurnit3re parts. IiJgh financial haulage costs of ipproximately 1,500Cedi/m (US$ 10/mJ) for a typical 150 km haul limit greaterutilization. Assisted through a US$3.5 million FAO project, the ForestryDepartment has experimented with mobile metal kilns in the Subri Riverbasin as a means of efficient on-site carbonization of forest residues.The kilns' high cost and difficulty in moving them from logging site tosite along with the potential forest fire hazards have thus far severelylimited their application. However, the COG has created a corporation,Subri Industrial Plantations Ltd. (SIPL), for large scale clearing andreforestation within the Subri forest reserves. Aided by US$ 16 millionfinancing from the African Development Bank, SIPL plans to cut 4,000 haduring the first five years following start-up. Estimated yields perhectare are:

Commercial timber -- 86 m3

Carbonizable wood 160 .3

Residaal firewood 164 m3

Charcoal supply to the Sekondi-Takoradi area is thus expected to beboosted by 11,500 t/yr, and for firewood, 131,000 m3/yr. Although worthyof further examination, potential forest residue utilization is beyondthe scope of the present study.

Wood Processing Residues

3.13 Concentrations of residues occur in parallel with saw andcombination mill activity. In the case of mills located in the Sekondi-Takoradi area and, to a lesser extent, the Kumasi area, theseconcentrations coincide with locations of major non-wood processingindustries. The Akim-Oda, Nkawkaw and Dunkwa areas host only minor non-wood processing industrial activity, and in the isolated company towns ofHim, Sefwi-Wiawso and Samreboi the mills represent virtually the solesource of non-agricultural employment.

8/ Ghana Energy Assessment.

- 23 -

Quantities Produced

3.14 Information on residue production is based on a direct surveyof the 66 mills listed in Annex 2. Residue volumes were elicited byquestioning mill management on production figures and estimated numbersof carts or loads per day of each of the residue types. The datagathered at any one mill, however, may be very subjective and affected bythe vagaries of question interpretation, experience, mood, and timeavailability of the interviewee. The survey data has therefore beenutilized as follows. Data derived from the sawmill questionnaires hasbeen plotted on Figure 3.1 which shows fractions of slabs/edgings,offcuts and sawdust as ordinates plotted against the lumber recoveryfactor as abscissa. Fitting the curves to this plot results in a morestatistically powerful method to determine the sawmill residuals.

3.15 The average recovery factor (weighted by production level) wasdetermined for each region in Ghana and the annual residual volumedetermined by type as shown in Table 3.5. All volumes represent solidwood equivalent (SUE) volumes and can be converted to oven dry or greenweights using the nominal characteristic values in Table 3.4.

Table 3.5: SAWMILL RESIWUE PRODUCTION, 1986

Weighted Log Slabs/Edgings Offcuts Sawdust TotalAverage Inpt Frection Voaliu Fraction VolVe Voiure Resiues

Resgion Recovery (mJ) (a') uJ) (m') (a)

Ashanti 0.423 348,599 0.331 115,386 0.104 36,254 48,804 200,444Brong Ahafo 0.434 19,400 0.322 6,247 0.100 1,940 2,716 10,903central 0.485 22,644 0.284 6,341 0.086 1,947 3,170 11,548Eastern 0.451 32,549 0.310 10,090 0.096 3,125 4,557 17,772Western 0.585 72,668 0.216 15,696 0.064 4.651 10,174 30,521

495,860 153,850 47,917 69,421 271,188

Overall Weighted Average Lumber Recovery: 0.453

Note: All volumes expressed In solid wood equivalents (SWE).Source: Mission estimates.

3.16 The data from the plymill, veneer mill and combined millquestionnaires cannot be evaluated by similar methodology due todissimilarities between mills and the small number of mills (twelve)falling into these categories. Therefore, the residuals from this groupof mills was taken at face value from the questionnaires. The residuesfrom these mills are summarized in Table 3.6. Again, all volumesrepresent solid wood equivalents.

- 24 -

Figure 3,18 SAWMILL RESIDUAL FRACTIONS VS. RECOVERY FRACTION, 1985

.. . ...- /1 * * 158

: -1S -'1-

SS 1-- 1 1 Itl":t I1--~~

t ~~~~~FtAt t0JS-

3 ~ ~~~~ H-Fcs !,W

- 25 -

Table 3.6: COMBINED MILL. RESIDUE PRODUCTION, 1986 a/Im3 SWE)

Slabs a Sawdust VeneerEdgings Offcuts Shavings Waste Cores Total

59,325 18,830 23,848 34,723 26,698 163,424

af Includes plywood and veneer mills.

Source: Ru-Tek.

3.17 The total wood processing industry residues are the sum of thevolumes given in Tables 3.5 and 3.6 and are shown in Table 3.7. Slabsand edgings are the most abundantly produced residues at 213,175 m3 SWE,representing 49 percent of the estimated 1986 total residuals productionof 434,612 m3 SWE. Sawdust at 21 percent and offcuts at 15 percentfollow in terms of abundance, while veneer waste and cores are relativelyless common at 9 and 6 percent respectively. In their green condition,the total residue production has an energy value of 94,000 toe or 4percent of fuelwood primary energy production in Ghana.

Table 3.7: WOOD PROCESSING INDUSTRY RESIOUE PRODXCTION, 1986iM SWE)

Slabs & VeneerMills Edgings Offcuts Sawdust Waste Cores Total

Sawmills 153,850 47,917 69,421 - - 271,188

Combined 59,325 18.830 23.848 34723 26698 163.424

Total 213,175 66,747 93,269 34,723 26,698 434,612

IndustryPercentages 49% 15% 21% 9% 6% 100%

Source: Misslon estimates.

Existing Stockpile

3.18 The wood processing facilities in and adjacent to Kumasi andTakoradi commonly have depressed areas land-filled with sawdust. Suchstockpiles become contaiminated with dirt and water and inevitably bio-degrade with time and are generally not a satisfactory fuel source. The

- 26 -

use of sawdust residues stockpiled out-of-doors in depressions andravines as an energy source is not recom_ended. The operating costs torecover such residues, to process them to remove contaminants and to drythem in rotary dryers, and the high capital costs of the necessaryancillary fuel-handling equipment, are usually prohibitive. Capitalfacilities should only be planned based upon a confirmed ongoing supplyof sawdust residue which can then be handled following practicesappropriate to the process.

Reliability of Supplies

3.19 So long as the wood industry continues its upward trend ofrehabilitation as in the last three years, it can be assumed that thesupply of residuals is reliable. Supply of residuals for on-site usewould, of course, be assured as the mills control the utilization oftheir own residues. OnRoing supply of residues for off-site uses couldbe assured by contractual arrangements with producers.

3.20 Effects of the earlier noted wood products industry trends onthe reliability of residue volumes and characteristics can be summarizedas follows.

(a) Greater processing of logs into export lumber. Increases inthe volume of logs converted into sawn products will have theobvious effect of increasing the volumes of slab/edging,offcut, and sawdust residuals which are a direct by-product ofthe log break-down and cutting operations.

(b) Higher product recovery. Mill improvements increasing recoverytend to reduce the amount of residues. Examples are: reducedkerf diminishes sawdust volumes% improved edging practicesreduces edgings volumes; greater sawing accuracy increases thenumber of boards produced; accurate charging reduces round-uplosses; improved clipping reduces veneer clips; more accuratelay-up reduces dry trim, etc. As demonstrated in Figure 3.1,the effect on sawmill residues is strongest as regards slabsand edgings, and weakest with sawdust.

(c) Increased secondary manufacture. Secondary manufacture mayincrease some residues and diminish others. If small cuttingsare recovered from edgings and offcuts, these residues willreduce while sawdust and shavings increase. If additionalmouldings are produced, edgings may reduce and sawdust andshavings increase. As wood pieces for secondary manufacturemust be kiln-dried, the shift is toward drier and more finelydivided fuels and away from larger section residues.

(d) Increased secondary species exploitation. As previouslycalculated, larger proportions of the less dense and wetter

- 27 -

species will reduce green residue calorific values modestly.However, the kiln-drying requirements of many of the lesser-known species, to be discussed in Chapter 6, will increase theproportion of dry residues. In addition, the suitability of anumber of these species for rotary veneer production will alsoproduce more dry residuals.

The effects of these trends will be most evident in the larger operationswhere capital is available to be invested in upgrading and convertingplants. The smaller mills will continue to produce residues as they dotoday. Log throughput levels will continue to hold the greatestinfluence on the production levels of wood industry residues. Othereffects are, to some extent, counterbalancing, and their overall neteffect on the suitability and reliability of wood processing residues asa fuel is not expected to be significant on an industry-wide basis.

Present and Projected Surplus

Present Surplus

3.21 Sawdust is the only wood processing residue psesently inabundant surplus in Ghana, amounting in 1986 to some 83,400 m solid woodequivalent or 65,500 tonnes. Ninety percent of all sawdust produced isunutilized at present, which accounts for 8S percent of all surplusresidues. At its nominal moisture content of 36 percent (wet basis), thesurplus sawdust has a net energy value of 18,000 toe. Suqjpluses ofslabs/edgings, offcuts and veneer wastes amounting to 13,900 m are foundaS the isolated sawmill of Nim Timber Co. Ltd., with the balance of 2,270m of slabs/edgings found at other remote mills. Residual volumes bytype are graphed in Figure 3.2.

3.22 The regional distribution of the residue surplus is showngraphically in Figure 3.3.

3.23 Significant concentrations of surplus sawdust exist in thefollowing five centers in the estimated amounts given in Table 3.8.Kumasi contains 63 percent of all surplus sawdust and 69 percent of theexploitable concentrations.

- 28 -

Figure 3.2: SURPLUS RESIDUES BY TYPE, 1986

90W~7 8'

78

0~6

P ~40-

209 -o ~100~~

Slabs/ Offcuts Sawdust Veneer CoresEdgings Wastes

Figure 3.3: SURPLUS RESIDUES BY REGION, 1986

610-

3 ' .

050

* 40t

> ~10

Ashanti Brong Central Eastern WesternAhafo

3 Sawdust 3 Other

- 29 -

Table 3.8: SURPLUS SAWDUST CONCENTRATIONS

Center m3 SWE % of Total Surplus Sawdust

Kumasl 52,600 63

Mim 9,500 11

Sokondl-Takoradi 9,200 11

Nkawkaw 2,500 3

Ounkwa 2,500 3

91

Source: Mission estimates.

Future Surplus

3.24 IJnder a "business as usual" scenario (i.e. assuming no newinvestment in improved residues utilization), the composition anddistribution of surplus residuals would not be expected to changedrastically, with sawdust continuing to be the main unutilized residue.Quantities of sawdust will be most sensitive to log throu*hput levels. Aspecific exception to this pattern is found at Cliksten Wea Africa Ltd.in Sefwi-Viawso, where the anticipated shut-down of cognerationoperations will free an *nknown proportion of the sawdust presently usedas boiler fuel (2,575 mJ in 1986). Similarly, solid residues not usedfor woodfuel by the surrounding community will be surplus. 9/

3.25 Plans outlined by mill managers indicate a trend to increaseduse of wood residues for mill process heat generation, as depicted inTable 3.9. The additional heat demands will divert some off-site use ofsolid residues back to the mills. A decline in surplus sawdust volumeswill occur to the extent that sawdust is utilized in mill boilers.

9/ The anticipated re-start of African Timber and Plywood Ltd. is notexpected to affect surplus as the mill plans to use all its residuesfor cogeneration.

- 30 -

Table 3.9: PLANNED ADDWITIOS TO WOOD PROCESSINGFACILITIES UTILIZING MILL iRSIDUE

FOR PROCESS HEAT GENERATION

MiIIMill Type Descrlptlon

Specialized Timber Products, Ltd. S Installation of cogenerating 70 t/hr at 20Kumasi bar boiler and 480 kVA turbine-generator

due for completion In 1987, plus associatedsteam pits and 6 x I5Om3 klIns.

Nim Tlmbers Ltd. S/P/V Plans to expand present klIn capacity ofimim 240 m3 to I X 0 .3 wIth S years.

Hardwood Timber Products, Ltd. S 3 x 100 m3 kilns currently Idle due to highTakoradi operational cost of oil-fired boilers. Will

convert to wood-fired boilers In mid-1987.

WesXern Timbers Ltd. S 4 x 70 3 kilns to be installed In 1987.Takoradi Boiler on-site awaiting Installation.

John Sitar Co. Ltd. S/P/V Plans to Install 12 x 25 .3 kilns, steamTakoradi pits, 2 x veneer dryers, and 2 x boilers.

Scanstyle Furniture Ltd. F Current kiln capacity of 350 m3 to beMiii Increased by 28% In 1987.

Poku Transport Industrial P/V Plans to Install 2 x locomotive typeComPlex Ltd. boilers.Kumasi

Du-Paul Wood Treatment Co. Ltd. S New sawmill to be completed In 1987. WiIITakoradi utilize process heot for kiln drying

of lumber for moulding line.

A. E. Saoud Ltd. S New soamill under construction willKumasi include 1 x 15 t/hr boiler.

Note:

1. Mill types: S a SawmillP a PlywoodV a VeneerF a Furniture parts

Source: Ru-Tek; Mission estimates.

- 31 -

IV. DUIAnD 101 KOOD PROCESSING INDUSTRY ESIDUES

Summary

4.1 About 27 percent of the 1986 total residue production wasconsumed within the mills themselves, primarily for process heatgeneration. The largest fraction of residues, 50 percent, was utilizedoutside the mill complexes for both energy and non-energy uses. Thebalance, or 23 percent, was surplus and was disposed of by burning ordumping. With the exception of Mim Timber Co. in Brong Ahafo, virtuallyall solid residues are utilized. The Ashanti region is the majorconsumer of wood processing by-products, both for on- and off-site uses.

4.2 The use of slabs and edgings for firewood and charcoalproduction is the largest end-use of wood residues at 31 percent of totalresidue production. Fuel for mill process heat/cogneration, other (non-energy) uses, and unused surplus evenly split the remainder. The chiefnon-energy use of residues is as a raw material for cottage industrywoodworking.

On-site Energy Uses and Disposal

4.3 Approximately 118,000 m3 (27 percent) of the wood processingresidues produced in 1986 were consumed on-site, i.e. at the millsthemselves. Nearly 84 percent of this consumption was for internal steamand process heat raising pu poses, including cogeneration of steam andelectricity. About 10,750 m of the on-site total was carbonized at theMim Timber Co. Ltd. plant, the only mill in Ghana engaging in thisactivity. The balance o)f on-site consumption was for non-energypurposes.

Steam and Process Heat

4.4 As depicted in Table 4.1, 21 mill facilities have wood-wastefired furnace/boilers for the generation of steam and process heat. Theheat thus raised is utilized in:

(a) Steam/conditioning pits for preparation of veneer blocks andfor sterilization of log species subject to spore and fungusinfestation;

(b) rotary and sliced veneer dryers;

(c) dry-kilnas for the drying of wood stocks for secondarymanufacture, manufactured components and export lumber.

- 32 -

The majority of mills in this category presently combust 50-60 percent oftheir total residues. Almost all the mills utilizing process heat arecombination mills, indicating process heat demand in the straightsawmills is quite low.

Table 4.1: WOOD PROCESSING FACILITIES UTILIZINGMILL RESIDUE FOR PROCESS HEAT GENERATION, 1986

ResidueUt I lized

for No. ofProcess Wood-fired Boller

Residue Heat Soilers/ DateMIII Generated Generation Nameplate of

Name of MlII Type (m3 SWE) (*3 SWE) Ratings Manufacture

ASHANTI REGIONPoku Transport IndustrialComplex Ltd. Veneer/PlymlIl P/V 7,281 4,611 (1) 465 MW 1978Kumasl at 20 bar

Poku Transport & Sawmills Ltd.Kumesi S 4,866 a/ (1) 720 MCaI/hr 1982

Logs and Lumber Ltd,Kumasi S/P/V 25,365 14,744 (3) 1S t/hr (1) 1972

3 t/hr (2) 1982EJIsu Forest Products Ltd.Kumasi S/V 6,250 3.425 (1) Hot Water 1977

BoilerFyne LimitedKumasi S/V 6,658 4,019 (1) 6 t/hr 1982

at 3 barsWood Complex Kassl Ltd.Kumasl 5 7,349 a/ (1) 12 t/hr 1978

Atwima Timbers Ltd.Kumasl S 15,761 9,797 (1) 10 t/hr 1978

Lumber Processing Ltd.Kumasi P/V 11,080 6,023 (3) 4 t/hr 1977

eachNaJa David Veneer S PlywoodKumasl P/V 13,035 7,085 (1) 18 t/hr 1977

Wood Industries Ltd.Kumasl S 5,760 a/ (1) 10 t/hr 1975

A. 0. Timbers Ltd.Kumasi S/V 14,240 7,204 (1) 20 t/hr 1975

- 33 -

ERONG AHAFO lEGIONMlm Timber Company Ltd.Mi, S/P/V 45,377 3,173 (1) 6.5 t/hr 1979

at 23 barsScanstyle Ltd.Mim F 4,473 1,200 (3) Hot Water 1969

Boiler

CENTRAL REGIONInternational Hardwood Ltd.Ounkwa S/F 8,750 6,073 (1) 20 t/hr

at 5 bars 1957

EASTERN REGIONNovotex Ltd.Nkawkaw S/P/V 2,101 368 (1) 15 bars 1976

Oda Wood Complex Ltd.Akim Oda S/P/V 12,661 7,881 (1) 5653.3 1982

x 103 kJ/hr

WESTERN REGIONA.T. 6 P. b/Samrebol S/P 20,700 20,700 c/ (4) 7.5 t/hr 1948

eachGilksten W. A. Ltd.Safwl-Wlawso S/P/V 12,79M 12,793 cl (4) 15 bar 1950

T.V.L.C.,Takoradi S/P 4,740 2,220 (1) 10 bar 1972

Ghana Prime Wood Products Ltd.Takoradi S/P 10,806 8,096 (1) 10 bar 1972

Biblanl Industrial Complex Ltd.Biblani S 1,474 319 (2) 5 bar 1974

TOTALS d/ 221,360 99,031

/ Boiler not In use, but operatlonal.b/ Mill not In operaticn since February, 1986. 1985 estimate.cf Co-generation of steam and power.d/ Totals do not Include A.T. 6 P.

Note:1. Hili types: S a Sawmill

P a PlywoodV u VeneerF a Furniture parts

PB a Perticle board

Source: Ru-Tek.

- 34 -

Cogeneration

4.5 Table 4.2 shovs that four wood processing facilities haveturbo-generators or steam engines to enable use of high pressure steamfor electricity generation. Three are state-owned: Mim Timber Co.,African Timber and Plywood, and Gliksten West Africa. The fourth,Specialised Timber Products, is a private entity. Only Gliksten WestAfrica Ltd. is currently operative in a cogeneration mode, as explainedbelow.

4.6 Cliksten West Africa Ltd. CUA presently burns all of itsresiduals, including hogged offcuts and cores, in four 1950 vintagelocomotive-type boilers raising an average of 3.5 t/hr of steam at 15bars. The steam is passed through 625 kVA and 437.5 kVA turbo-generators. No records of power production are kept, however plantdemand is met through supplementation from a 690 kVA diesel set. Thenational grid passes within 1 km of the mill. An 800 kVA transformer ison-site and mill management hopes to connect to the grid in 1987 as soonas financing for the required second 800 kVA transformer can bearranged. At that time, local electricity production will cease and GWAwill combust wastes only to meet process heat demands.

4.7 Specialized Timber Products Ltd. Installation of a 6.4 t/hr at28 bar boiler and 480 kVA back-pressure turbine at this new sawmill is inprogress and will be completed in the first half of 1987. The five yearold equipment was acquired from a German mill for US$ 540,000 installed,approximately a 60 percent savings over comparable new units. The millplans to combust essentially all of its residues and thereby meet 80percent of its electrical energy demand plus the beat load from six 150a kilnas and two steam pits. According to the mill manager, the chiefmotivation for the investment is to obtain a degree of independence fromthe ECC grid supply which was judged too prone to interruption. The STPmanagement estimates the value of lost production due to electric supplyfaults in 1986 at Cedi 5 million (US$ 33,000). Synchronization for apower purchase/sell-back arrangement has been investigated, but it wasconcluded by STP that lack of VRA/ECG standards negate this option.

Table 4.2: WOOD PROCESSING FACILITIES UTILIZING WILL iESIDUE FOR00-GENERATION OF STEAM AND POWER, 1986

Nb eplate NameplateRssidue Capacity of CapacIty of Totalutilized Steam Engine/ DIesel Installed

Rasidue for Turbine Gen- GeneratIng GenerationGenerated Co-Generation erating Sets Sets Caeacity

Name of Mill (3 SWE/yr) (03 SWE/vr) (&VA) (kVA) (kVA)

Gliksten West Africa Ltd., a/ Sefwl-Wiawso 12,793 12,793 1,062.5 690 1,752.5

Specialized Timber Products Co. Ltd., b/Kumasi 20,736 20,736 480 National Grid 480

Mim Timber Co. Limited, Hli 45,377 0 420 c/ 3,126 3,S46

A.T. S P. Ghana Llmited, d/ Samreboi 20,700 20,700 3,750 850 4,600

s/ To be connected to national grid In 1987. Present condition of boilers Is poor.b/ Projection only. Equipment installation to be completed In 1987.

c/ Steam engine presently not operational.d/ Mill not In operation since February, 1986. Estimates are for 1985.e/ 3 x 1250 kVA turbine-generators; 1 - fair condition; 2 - poor condition.

Source: Ru-Tek; Mission estimates.

- 36 -

4.8 Mim Timber Co. Ltd. Steam raising equipment at Mim comprisestwo boilers, one wood-fired unit rated at 6.5 t/hr and 23 bars, and theother a 1 t/hr fuel oil-fired unit rated and 10 bars. A second wood-fired boiler, a 2 t/hr at 7 bar locomotive type, is slated forinstallation in 1987. High pressure steam from the large wood wasteboiler is designed to feed a 420 kVA steam engine-generator iiatalled in1980. However, during operation in the first year, lube oilcontamination of the feedwater resulted in damage to the boiler. Thesteam engine has not been operated since 1981 because the condensatecannot be recycled due to lube oil contamination and there is a watersupply shortage at Him during the 5 to 6 month dry season which will notpermit open cycle operation. At present, only 7 percent of the residuesare combusted for process heat; the balance is sold as firewood (18Z),charcoaled (24Z) or burnt off in a fire pit (52Z). Thus total electricaldemands for the mill, the Scanstyle furniture factory, Mim Agro Ltd. andthe surrounding community of 2.2 MW peak and 6.4 GWt. are met entirely bydiesel generation. A 47 km Sunyani to Mim grid extension is proposed for1989 at a capital cost of US$ 2.2 million (before firancing), financed bythe European Investment Bank.

4.9 African Timber and Plywood (Ghana) Ltd. The large capacity ATPsawmill has been shut down since February, 1986 and the plywood linesince mid-1984 because of financial and operational difficulties. Arecently signed US$ 38 million financing agreement with the Bank ofScotland provides for mill rehabilitation and operation under a five yearmanAgement contract with Marktrace Projects Ltd. The ATP power housecontains four 1948 manufacture 7.5 t/hr boilers, of which possibly twoare in operable condition. These feed three 1,250 kVA turbo-generators;only one is considered serviceable. Plans are to refurbish the boilerhouse under the supervision of the manufacturer, Babcock and Wilcox (UK)Ltd. At such time when the mill complex returns into production,essentially all its residues will be utilized for steam and powergeneration. The national grid is 28 km away, but there are presently noplans for its extension to Samreboi.

Charcoal Production

4.10 Thirty-one percent of the solid residues produced at Mim TimberCo. Ltd. were converted into charcoal at the mill in large earth moundkilns. Total charcoal production in 1986 is estimated at 865 tonnes,which translates to an apparent conversion efficiency on a dry weightbasis of 13.9 percent. Sixty percent of the charcoal thus produced issold to mill workers at 30 Cedis per 40 kg sack; the balance is sold tooutsiders at various prices and leads to a average sale price of 64Cedis/sack. As production costs are calculated by Mim Ltd. at 45Cedis/sack (assuming a zero opportunity value of the residues), net millincome is 19 Cedis/sack. If charcoal consumption patterns followestimated urban averages of 140 kg/person/year, then 6,180 persons meettheir domestic cooking needs from the residue-derived charcoal.

- 37 -

Disposal

4.11 Almost all solid residues (i.e. slabs/edgings, offcuts, andcores) which are not consumed at the mills are sold for off-siteconsumption. The main exception is Mim, where nearly 15,000 m* ofslabs/edgings and offcuts were burned in an open fire pit as a means ofdisposal.

4.12 Sawdust is dumped away in land depressions or ravines near themills using 5 m3 tractor-trailers and bulldozers. Depending on landavailability, the sawdust may be burnt in large piles. Financial costsof equipment, operations and labor for sawdust disposal range from US$1,000 to 8,000 per annum per mill, with the latter figure pertaining,tothe largest mills. Environmental costs and fire hazards are likelysignificantly greater. Hardwood Timber Products Ltd. of Takoradi dumpssawdust in city areas under city council permit, to the voiceddispleasure of nearby residents. A sawdust pile spontaneously combustedon the premises the week before the mission's visit. In Kumasi, landfilling around the Ahinsan-Kaase industrial area is no longer beingallowed by the landowners. A large capacity mill complex, A.E.Saoud/Lumber Processing Ltd., had a serious fire accident two years agoas a result of sawdust disposal by burning.

Off-site Energy Uses

4.13 The largest fraction (50 p-rcent), or some 217,000 m3, of the1986 processing residues found their way to gses away from the millcomplexes. Energy use consisted of 124,000 mn of slabs/edfings beingsold for firewood and charcoal production, plus about 2,800 m of sawdustprocessed into briquettes. The principal application of this energy wasfor food preparation in bakeries, local restaurants and households.However, the sawdust briquettes are being adopted for use in brickmakingkilns.

Industrial and Commercial Heat Raising

4.14 Of 61 industrial sites surveyed preliminarily for energy use,only one, Asokwa Brick and Tile Co. Ltd. in the Central Region, is aregular consumer of unprocessed sawmill residues. Ankaful Brick and TileCo. Ltd. is a regular user of sawdust briquettes. Other industriesutilizing woodfuels, which include soap and palm oil producers, brickfactories and gold mines, obtain firewood from the natural forest.

4.15 Bakeries, small restaurants ("chop bars") and fish smokerslocated within a 10 km radius of the major wood processing centers aremajor consumers of slabs/edgings for use as firewood. Sale prices at thesawmill gate are about 400-500 Cedi/tonne (US$ 3/tonne) in Kumasi and 600Cedi/tonne (U8$ 4/tonne) in Takoradi. In addition, about two dozenpublic boarding schools in these areas use from 60 to 450 tonnes/yr eachfor institutional cooking.

- 38 -

Domestic Cooking

4.16 The balance of residues consumed directly as firewocS ib usedin households for lomestic cooking on simple, inefficient stoves, oftenof the "three stone" variety. However, charcoal is the preferreddomestic fuel in urban areas, exceeding firewood consumption by nearly 10to 1 on a gross energy basis. 10/

Charcoal Production

4.17 An estimated 60 percent of the slabs/edgings sold as fuelwoodin Kumasi, and 80 percent in Akim-Oda, are carbonized by charcoal makersworking within a few kilometers of mill sites. Very little residuecharcoaling activity takes place near the mills in Sekondi-Takoradi, andthe town's main source of charcoal supply is from the forest areassurrounding Tarkwa. The solid residues are carbonized in so-called"earth mound" kilns, although in many cases the top covering iF actuallysawdust. In the Kumasi area in 1986, approximately 64,000 m of slabsand edgings were convertee to 5,400 tonnes of charcoal with a calorificvalue of 3,750 toe. An estimated 300 persons were employed in thisactivity at 50 sites, earning amounts ranging from 100-300 cedis perperson-day.

Briguetting

4.18 There is one briquette plant presently operating in Ghana,Chaowus Ltd. in Akim-Oda. The plant is owned and operated by a Taiwaneseentrepreneur and has been in production for approximately one and one-half years. Present actual production rate of 1,100 t/yr is only halfstated capacity due to operational inefficiencies. Chaowus obtainssawdust at no charge from Akim-Oda area sawmills, using their own 7 tonnetruck to haul the sawdust to the plant. Total sawdust demand, includinIamounts combusted to heat the sawdust drier, is approximately 2,800 mSUE at the current briquette production rate. The plant thus consumesabout 60 percent of the sawdust produced in Akim-Oda.

Energy vs. Non-Energy Uses

4.19 It is axiomatic that wood processing residues should be put totheir hi hest and best use consistent with economics. Approximately95,000 m of residues were put to non-energy uses which will be reviewedbriefly for comparison with value in energy uses. In the case of offcutsand veneer cores, non-energy applications are the dominant utilization,accounting for the disposition of 77 percent of the former and 96 percentof the latter.

10/ "Report of Pilot Survey on Puelwood and Charcoal Consumption inAccra", Government of Ghana National Energy Board, October 1985.

Table 4.3: VOWD ROCESSINB IISSSRTY DISPO6ITION OF RESIOUES 6Y tMEBlon, 1906t0m SW)

UtilIzed Inldn thm Wilts UtIlied Outside1 tb Wills _w1 usSlabs Swdust Vener Slabs Saedust e Slabs vener

ReolcnAlI I tdgings Offeuts ShavIngs Waste Cores Totals Edgings Offeuts Slavings Wast Coe Totals Edgings Otfeuts Snudunt Wste Totals

Swnelils 14,400 3,760 - - - 18,160 100,966 52,494 390 - - 133,870 - - 46,414 - 48,414Combined UD#DOO 3,8 3_418 28.398 - 45.201 5jj0 2.6? _ - 20,467 28.393 - - 4,197 - 4,197Totals 24,400 7,145 3,418 26,396 _ 63,361 106,045 35,361 390 - 20,46? 162,263 - - 52,611 - 52,611

kron AlastoSnWlls _ - - - - - 6,247 1,940 - - - 8,187 - _ 2,716 - 2,716Ccblned 10,739 ILI73 - - 13,912 8,054 - - - 8,054 8,055 4.603 94 1,249 23,411Totals 10.739 3,173 - - 13,912 14,301 1,940 - - - 16.241 8,055 4.603 12,220 1,249 26,127 >

Se I I s 2,663 1,423 209 - _ 4,295 1,498 524 - _ - 2,022 2,270 - 2,961 - 5,231Combne - - - - - - - - -Totals 2,663 1,423 209 - _ 4,295 1,496 524 - - - 2,022 2,270 - 2,961 - 5,231

Eastrnsemi I I - - - - - 10,090 3,125 693 - - 13,908 - - 3,864 - 3,864Combined 5.37 1.503 - 3.091 - 9-971 - - 2,112 - 2.678 4.790 - - 378 376Total s 5,377 1,503 - 3,091 - 9,971 10,090 3,125 2,805 - 2,676 18,698 - - 4,242 4,242

"esterm5 ellils 1,334 1,205 - - 2,539 14,362 3,446 - _ _ 17,606 - - tO,174 - 10,174CobIned 12,041 5,299 3,047 1.985 3.553 23.925 - - - - - - - - 1,192 - 1,192Totals 13,375 4,504 3,047 1,985 3,553 26,464 14,362 3,446 - _ _ 17,608 - - 11,366 - 11,366

Industry otals 56,544 17,748 6,674 33,474 3,553 118,003 146,296 44,396 3,195 - 23,145 217,032 10,325 4,603 83,400 1,249 99,577

Soure: mI ssIan estieates.

- 39 -

Secondary Manufacturins and Export

4.20 Approximately 8,200 m3 of solid residues were further processedwithin the mills to yield marketable products such as flooring, toolhandles, broomsticks, fencing materials and wooden crates. Offcuts ofhigh value species such as Afrormosia and Hyedua are ripped into stripsfor export, while cores (bouls) are sliced lengthwise and reassembled forstrip shipping. These remanufacturing operations are usually highlyprofitable.

Cottage Industry Woodworking

4.21 In addition to the industrial scale production of knock-dawnfurniture parts, there is a large and active cottage industry centeredaround the sawmills based on the reworking of offcuts. These 3 to 10 mancarpentry/furniture operations typically employ small band and circularsaws to produce low cost structural timber, chairs, tables, beds, chests,toys, etc. for domestic consumption. It is prima facie evident that agood deal of employment, value and income is being generated by theseartisan activies.

Particle and Fiber Board Production

4.22 Novotex Ltd. in Nkawkaw feeds most of its slabs/edgings andoffcuts to a domestic grade particle board plant which is part of thecomplex. About 1,350 m of solid residues were utilized in this ashionin 1986; the balance of the raw material input for the 8,000 m boardproduction was extracted from the forest and hogged. Western TimbersLtd. of Takoradi, the MLUR and SCM Engineering of West Germany arejointly investigating the feasibility of producing medium densityfiberboard (MDF). The highest grade can coztain up to 20 percent sawdustand has an export value of about US$300/m FOB. Other grades roughlycompete with domestic plywood and are composed of up to 50 percentsawdust. Although the mill manager asserts that MDF is in high demand asan export product, the high cost of shipping such a dense product and thecurrent world surplus of MDF manufacturing capacity will likely dim theeconomic prospects of such a venture. Nevertheless, should the schemeprove successful it should be considered a higher value use for on-sitesawdust residues than as fuel.

Summary of Present Residues Utilization

4.23 The 1986 utilization of wood processing residues is broken downby region in Table 4.3 and displayed graphically in Figure 4.1. As to beexpected from residue production statistics, the Ashanti region is themajor consumer of wood processing by-products both for on- and off-siteuses. However, the Western region utilizes a greater share of itsresidues within the mills, due in large part to the higher proportion offacilities equipped with boilers for process heat raising andcogeneration.

- 41 -

Figure 4.1: RESIDUE UTILIZATION BY REGION, 1986

250

150lee

p4 500

Ashanti Brong Central Eastern WesternAhafo

gg3 In-Mil1 Use 3Outside Mill JUse H Surplus

- 42 -

4.24 The end-use of residues is arranged by residue type in Table4.4 and Figure 4.2. The use of slabs and edgings for fuelwood is thelargest single employment of wood residues at 31 percent of the totalresidue production. Fuel for mill process heat/cogeneration, other (non-energy) uses, and unused surplus evenly split the remainder.

Table 4.4: END-USES OF NOMD PROCESSING INDUSTRYRESIDUES BY TYPE, 1986

(a' SWE)

Fuel for Furniture,MIII Process Flrewood/ Fencing, Surplus/

iHeat/Cogeneration Charcoal Prod. Export & Other Unused

Slabs/ 48,746 134,918 19,186 10,325 b/Edgings (49%) (97%) (20%) (10%)

Offcuts 9,046 1,550 51,548 4,603 cl(9%) (1%) (54%) (S )

Sawdust 6,674 3,195 ", - 83,400(7%) (2%) (84%)

Veneer Waste 33,474 - - 1,249 c/(34%) (1S)

cores 1,091 - 25,607 -(1I) (26%)

Totals 99,031 139,633 96,341 99,577(100%) (100%) (100%) (100%)

IndustryPercentages 23% 32% 22% 23%

a/ 2,805 m3 SWE for sawdust briquette production at Chaowus Ltd., Akio-Oda.b' 8,05S .3 SUE burned In firepit at Mim Timbers Ltd., Rim.c/ Burned In fire pit at Him TIlbers Ltd.

Source: Mission estimates.

- 43 -

figure 4.2t USIDUE MM-USE BY TYPE, 1986

140w ~1310

1ao11090

0 ~80

500D 405 30

0 ~20le0

Mdl Fuel FLrewood/ Other Surplus/Charcoal Unused

3g Slabs/ Offcuts E Sawdust dWste CoresEdgkngs Wae

- 44 -

4.25 Table 4.5 and Figure 4.3 re-organize this information by end-uses of the residues.

Table 4.5: DISPOSITION OF NOD PROMCSSING INDUSTRYRESIDUES BY END-USE, 196

(U SWE)

VeneerSlabs/Edgings Of fcuts Sawdust Waste Cores

Fuel for W111 Process 48,746 9,046 6,674 33,474 1,091Heat/Cogeneratlon (23%) (14S) (7%) (96%) (%)

Firewood/Charcool 134,918 1,550 3,195 - -

Production (63%) (2%) (3%)

Furniture, Fencing, 19,186 51,548 - - 25,607Export l Other (9%) (77%) (96%)

Surplus/ 10,325 4,603 83,400 1,249 -Unused (S0 (7%) (90%) (4%)

Totals 213,175 66,747 93,269 34,723 26,698(100%) (100%) (100%) (100%) (100%)

Source: MIssion estimates,

- 45 -

Figure 4.31 RESIDUE DISPOSITION SY ND-USE, 1986

2202003 2180

gV) g16014812010080

a' 60

*3 1134e 9-40

o 20 -Slabs/ Offcuts Sawdust Veneer Cores

Edgings Wastes

S Mill Fuel F i rewood / Oth.r X Surplus

- 46 -

The figures indicate that residues are utilized or disposed of asfollowgs

Residue Utilization or Disposal

Slabs/edgings Primarily sold for charcoal production orfirewood. Burned to raise steam for processheating of dryers and conditioning vats atcombination mills. Some minor secondarymanufacture. Very minor use for particleboard manufacture.

Offcuts Primarily sold to furniture manufacturersand carpenters. Some use as fuel atcombination mills and cogeneration sites andfor charcoal production. Very minor use forparticle board manufacture.

Sawdust Usually disposed of by dumping or openburning. Minor use to raise steam atcombination mills and for briquetteproduction in Akim-Oda.

Veneer waste Burned to raise steam at combination mills.

Cores Re-sawn for export. Sold for domesticconsumption. Minor use as hogged boilerfuel at GWA and for charcoal production.

The types end locations of surplus residues are discussed in Chapter V.

Costs of Utilization

4.26 In Chanaian mills most residues are collected by hand, althoughsawdust may be handled by chain conveyors or pneumatic systems in some ofthe larger plants. Slabs, edgings and trim blocks are normally placed onhand carts or collected in trailers for transport out of the mill area.Sawdust is most frequently transported to an adjacent pile for landfillor burning. These practices have costs associated with the labor ofcollection and for tractor/truck operations and maintenance. These costsare essentially the same whether the residues are to be sold, dumped,burned or utilized to fuel an on-site thermal plant. The delivered costassociated with off-site use is therefore transport to the point ofconsumption.

4.27 Additional costs would be associated with mechanized fuelhandling at the thermal plant. For example, fuel silos, their unloaders,and fuel distribution conveyors may be required at the thermal plantwhich would need to be justified by labor savings or the value of

- 47 -

continuity of operation. If air conveying of sawdust and shavings isutilized, there may be an increase in horsepower to redirect the the fuelup into a fuel silo. Determination of the incremental horsepower is sitespecific and would result in an incremental increase in power costs whichshould be considered in the economic evaluation of the specific project.

Technical/Infrastructure Constraints to Residues Utilization

4.28 A number of technical and organizational constraints causesawdust residues to go unutilized, thus diverting solid residues fromalternative uses into use as boiler fuel. Rxisting technicalshortcomings, as discussed below, result in inefficient combustion of thesolid residue fuels.

Water Spray Lubrication of Saw Blades

4.29 Numerous mills employ water spray for the lubrication of bandand circular saw blades used in log break-down. The water is sprayed inexcessive amounts, at some mills visited being applied on the sawblade bya garden hose resulting in most of the water simply splashing off theblade and excessively wetting the sawdust. The over 70 percent moisturecontent (wet basis) of the sawdust makes it very difficult or impossibleto combust without expensive pre-drying, and is one cause of the lowutilization of sawdust as a boiler fuel.

Outside Storage of Sawdust

4.30 Want of demand for the fuel and lack of covered storagefacilities causes the waste sawdust to be stored out-of-doors, exposed tothe elements. Rainwater adds to the moisture content and difficulty ofburning.

Boiler/Furnace Configuration

4.31 Predominantly firetube wood-burning boiler/furnaces utilizedthroughout Ghana are designed to burn solid wood fuels. The large grateholes and insufficient airflow for suspension result in sawdust fuelsfalling through the grate. Proper support of sawdust combustion would inmost cases require consid-rable modification of grates and air feedsystems.

Boiler Efficiency

4.32 At some of the wood-fired boiler sites visited, reported woodconsumption and calculated heat loads imply boiler efficiencies as low as15 percent. The evident excess combust5 n air results from missingfurnace doors and inadequate control of tramp air volume anddistribution.

- 48 -

V. POTENTIAL ON-SITE ALTEREATIVRS FOR INPROVINC AND/ORINCREASING USE OF MOOD INDUSTRY RESIDUES AS FUEL

Summary

5.1 Increased process heat requirements and cogeneration optionscould technically utilize all of the present residue surplus, whiletechnical improvements in current residue handling, storage andutilization systems could make virtually 100 percent of the currentlywasted sawdust suitable as a boiler fuel. However, actual improvementsand increases in the use of the available residues will be dependent onthe financial costs and economics of doing so when compared to otheravailable alternatives.

5.2 This chapter outlines the components of the most promisingtechnical options, discusses the possible technical constraints andestimates the capital and operating costs required for implementation. Afinancial and economic evaluation of each option is presented followingthe technical discussions. The alternatives are developed based onactual sites and conditions observed during the field work rather than ongeneric plants based on average conditions in Ghana. The selection ofactual sites makes the analysis potentially useful. However, the resultsmust be carefully evaluated if the conclusions are to be generalizedacross the sector.

5.3 Several options for improving andlor increasing the on-site useof wood industry residues in Ghana were identified during the fieldmission. The most promising options at the mill sites included:

(a) Steam generation to meet increasing process heat needs for kilndrying and wood treatment;

(b) Cogeneration to meet both electricity and process heatrequirements;

(c) Improved sawblade lubrication systems and sawdust storage toreduce the water content in the sawdust residue and increasethe net energy available;

(d) Furnace modifications to enable direct combustion of sawdust;and

(e) Furnace improvements for greater combustion efficiency inexisting boiler equipment.

The first four options have the potential of increasing overallutilization of residues while the last, if implemented, may have theopposite effect. A summary matrix of the technically feasible optionsfor improving and/or increasing the on-site use of wood residues andtheir potential impacts is presented in Table 5.1.

Table 5.1 MATRIX OF TECHNICAL OPTIONS FOR IMPROVINB AND/OR

INCIEASINS ON-SITE R£SIDUE UTILIZATION

Impact on Residues

Options Primary Taroets Utilization Renuireuents Benefits

Steam generatlon Nedium to large scale Increase In on-site Capital for additional Increased value added for

for Increased mills facing declining demand for solid boiler and kiln drying export lumber and potential

processing and Inputs of primary species residues and possibly equipment. commercialization of secondary

kiln drying or desiring additional sawdust, species. Also potential for

revenues through Increased reduced disposal costs If sawdust

lumber processing. Is used.

Cogeneratlon Large scale mills with Increase In on-site Capital for high Lower expenditures for electricity

high process heat demands, demand for solid pressure boilers and especially if generated by diesel.

residues and possibly turbo-generatlon Also, potential for reduced residue

sawdust. equIpment, dlsposal costs.

Sawdust handling Mills that utilize Reduced on-site con- Technical assistance, Increased efficiency of on-site

and storage residues for o,% *te sumption of solid Improved saw blade combustion, higher revenues from

Improvements. energy, residues and potential cooling systems to from sales of surplus solid

off-site sales. reduce excess water In solid residues, lower costs for

residue, Improved storage disposal of sawdust.

and handling facilities.

Furnace MillIs with wood-burning Increase In on-site Technical assistance, Revenues from the sale of

Modification furnaces unable to burn demand for sawdust, boiler rehabilatlon and surplus solid residues.

sawdust and where there conversion and possibly

Is a ready market for additional residue

solid residues. handing equIpment.

Furnace Mills with In-efficient Reduced on-site con- Technical assistance Increased efficiency of on-site

efficiency wood-burning furnaces, sumptlon of residues for Improved combustion, higher revenues

Improvements end potential for combustion process from sales of surplus residues.

Increased off-site control.

sal.-. of solid

residues.

- 50 -

On-Site Utilization

Background

5.4 Opportunities for improving or increasing the on-siteutilization of wood residues were extensively investigated through pre-mission on-site surveys of more than 90 percent of the wood processingindustries in Ghana. This was then followed up by a second set of visitsto 13 facilities which account for about 50 percent of the total woodprocessing capacity in Ghana. A list of the facilities visited ispresented in Table 5.2.

Table 5.2: WOOD PROCESSING FACILITIES VISITED BY MIlSION

Ashanti Rei.on

Specializee Timber Products Ltd. (STP) KumasiA, G. TImbers Ltd. KumasiA. E. Saoud Ltd. Kumasi

Eastern Reglon

Oda Plywood and Veneer Co. Ltd. Oda

Brona Ahafo Region

Him Tirber Company Ltd., (K14) Him

Scanstyle Furniture Ltd. Mim

Western Region

African Timber and Plywood Ltd. (ATUP) Sasrebol

Gliksten West Africa Ltd. (GWA) Sefwi-Wlawso

Hardwood Tlmer Ltd. TUkoradiTakoradi Veneer and Lumber Co. Ltd. (TVLC) Takoradi

Western Tlmbers Ltd. TakoradiJohn Bitar Co. Ltd. SekondiProstea Goldfields Ltd. Prestea

Sawmill Process Heat

5.5 The most extensive on-site use of wood residues is for processheat needs, primarily for steaming vats, veneer and plyboard drying andlumber kiln drying. Total 1986 on-sit residue utilization for processhgat amounted to approximately 99,000 m% (including approximately 13,000m' used in cogeneration). There is a definite trend toward increased on-

- 51 -

site process heat demand. This trend is driven by: (a) the need formore on-site kiln drying and treatment of lumber primarily due to theincreased reliance on "secondary" species in the Brong Ahafo and Ashantiregions (due to a depletion of primary species) and (b) the desire of theproducers to obtain more added value for exported wood products. Theoption to establish or increase kiln drying and steaming vat capacity isprimarily limited to t~e large scale mills with log throughput capacitiesin excess of 10,000 m per year. At present only a minor percentage ofthe export lumber output is kiln dried. Assuming an increase in kilndrying to 60 percent of the export lumber would result in an increase inthe on-site demand for residues by as much as 48,000 m per yearrepresenting 48 percent of the total presently surplus residues. If 100percent of tise export lumber were kiln dried, the tot& residue demandfor kiln drying would increase by an estimated 80,000 m per year or 80percent of the available surplus.

5.6 Sawmill Process Heat Unit Production Model. The SpecialisedTimber Products Ltd. (STP) mill in Kumasi is selected as the model usedto evaluate the technical and economic potential of increased on-siteresidue utilization for sawmill process heat. The primary objective atSTP is to install a boiler to provide steam for sterilizing vats anddrying kilns which are required if the mill is to be capable ofprocessing secondary species and thereby increase its annual output.More generally, kiln-drying serves to add value to the export lumber.About 90 percent of the mill's production is for export. At present, STPhas no on-site steam generation capacity. The mill currently operatesone shift a day with the rysultant log input level of 43,000 ma/yr. STPhas a capacity of 70,000 m /yr with a two shift operation but it is notyet able to operate two shifts a year due to the shortage of primaryspecies logs for input.

5.7 STP has no concessions and has to purchase all of its logs.Present log throughput consists of 70 percent Wawa and 15 percent Utile,with the remaining 15 percent composed of Afromosia, Mansonia, Odum andienri. STP plans to produce a greater variety of secondary species andis currently installing kiln capacity to do so. A combination of anti-stain and anti-rot treatment is required in order to render a number ofthe secondary species whitewoods commercially viable. The dryingrequirements of the various secondary species is the focus of on-goinginvestigation; a partial listing of these requirements is given in Table5.3. STP is also installing steaming vats to sterilize mold spores insusceptible whitewood logs.

- 52 -

Table 5.3: SECONDARY SPECIES REQUIRING KILN DRYING

Species Group Present Uses

Aprokuma IV Light carpentry work; moulding

Emire lb Exterior Jolnery, flooring,plywood and moulding

Kyonkyon lIb Vcneer and plywood

Kyere IV Veneer and plywood

Ofram III Interior Joinery, moulding,plywood, furniture

Otle 11I Plywood face/core, lightconstruction, moulding andInterior fittings

Wawa a/ lb Light Joinery, plywood, veneers,boxes and crates, mouldings

a/ Drying required for mouldings.

Source: Mim Timbers; P4E International; TEDS.

STP currently purchases their electricity from the grid but is proceedingwith a cogeneration installation. The cogeneration option at STP is notconsidered in this section but will be addressed later. Only the directprocess heat requirements are investigated here.

3.8 3 Infrastructure Requirements. STP plans to install 6 dry kilnsof 150 m capacity each plu% two log steaming vats. It is estimated that3.7 tonnes/hr of 5.3 kg/cm' gauge saturated steam will be required tosupply the steaming vats and kiln driers. To satisfy this steamrequirement, a sawdust/waste wood-fired boiler and new boiler room willbe required. Condensate steam would be collected and returned to afeedwater tank. Wood residues would be manually fed to the furnace withoversize pieces being manually cut-up to suit the furnace requirements.The boiler is assumed to operate continuously for 50 weeks/yr. Fuelrequirements are estimated at 11,000 tonnes/yr of wood residues averaging38 percent mcwb. Total wood residue production at the mill is estimatedto be 48 3ercent of the log throughput which is equal to 22,800 tonnes/yr(70,000 m x 0.48 x 0.420 bone dry tonnes/m? / 0.62). Of the 22,800tonnes/yr residue, approximately 30 percent is sawdust equivalent to6,840 tonnes/yr. The remaining 15,960 tonnes/yr residues are made up ofedgings, off-cuts, slabs and rejects. A schematic diagram of theproposed alternative is presented in Figure 5.1 as STP Alternative 1.

- 53 -

Figure 5.1: STP LTD. - SCHEMATIC - ALTERNATIVE 1

FLASH STEAM

Alternative 1 09th

1.2 t/h VT

FUEL BOILER | _ CONDENSATEfUEL N (S.3kgIcml gauge 3..l RECEIVER

11000 ta saturated)@ 38%mcwb

l '~~~~~~~IN

FEEDWATER 2 8Vh

TANK

MAKE-UP WATER0.9t/h

5.9 Kiln Throughput. Kiln throughput is derived in Annex 5 as19,500 ,3 per annum based on a redwood/whitewood mix slightly differentfrom that planned at STP but which mirrors the national average exportcut. It is assumed that kiln dried lumber output will consist of: 29percent of one inch whitewoods requiring 10 days drying time; 29 percentof two inch whitewoods requiring 14 days drying time; and 42 percent ofredwoods requiring and average of 21 days drying time. The kilnthroughput represents 55 percent of the mill total lumber production,which is consistent with the expressed plans of the STP manager to dryhalf of his production.

5.10 Costs. A summary of the estimated c- ;nd annual operatingcosts for the on-site process heat model is pt_S___ id in Table 5.4. Theconstruction and equipment (C&E) costs are separated out to reflect thosecosts associated with the boiler and the kilns and vats. Total C&K costsfor the boiler including pipeworks, valves, ducts, electrical,instruments, spares etc., are US$333,000 while those for the kiln. andvats are US$350,000. The total installed capital costs for the entireproject is estimated at US$1.28 million. Details of the cost estimateare presented in Annex 5.

- 54 -

Table 5.4: SiJWMARY OF CAPITAL AND ANNUAL OPERATING COSTSFOR SAWMILL PROCSS HEAT UNIT PRODUCTION MOCEL

Local Costs Foreign Costs TotalItem VO00 USS) (000 USOS) ('000 USS)

Capital CostsConstruction

ofIlers S0.0 12.0 62.0KIls/Vats 15.0 5.0 20.0

65.0 17.0 82.0

EquipmentBoilers 271.0 271.0Ki In/Vats 330.0 330.0

601.0 601.0

Transport Chares a/ 59.3 58.3 117.6Engineering, Installation 181.1 297.5 478.6and Contingencies

Total Capital Costs 305.4 973.8 1279.3

Annual OpWrating CostsLabor 9.7 9.7Power 21.9 21.9e & M 15.0 15.0Consumables b/ 6.7 lS.0 21.7ContingencIes 6.8 - 6.8Total Annual Operating Costs 60.1 15.0 75.1

a/ Includes International freight and Insurance, local port charges andbank fees.

b/ Includes lost revenues from reduced slab sales.

5.11 Total annual operating costs are estimated at US$75,100 withthe largest single component being electric power requirements for thepumps, blowers and kiln air circulation fans. Only US$15,000 of thetotal is expected to be foreign costs. Details of the annual operatingcosts are presented in Annex 5.

5.12 Benefits. Air-drying makes essentially no contribution toproduct value added because the lumber will never reach the 10-12 percentmoisture content (dry basis) required in the European market. Local kilndrying can replace 8uropean kiln drying and potentially capture the valuegain. The gross contribution to forest industry value added throughlumber kiln drying is calculated in Table 5.5 and the resulting annualbenefit streams displayed in Annex 5. The percentage value added forkiln-dried lumber is based on estimates made by mill managers andexporters, which ranged from 15 to 30 percent for redwoods and 10 to 15percent for whitewoods. The redwood estimates include a 3 percentbenefit due to reduced mill-to-port transport costs; however, accordingto mill managers, drying has no effect on the transport economics of the

- 55 -

less dense whitewoods. The median figures thus obtained, 22.5 and 12.5percent respectively, are consistent with current export pricedifferentials quoted for kiln-dried vs. undried lumber. For example, theOctober 18, 1986 "Ghana Timber Export Market Report" issued by TEDB givesaverage U.K. prices per POB m3 for FAS Swietenia Mahogany of US$605 forkiln-dried and US$515 for shipping dry, a 17.5 percent gain. This doesnot include the estimated 3 percent of gross value savings in truckingcosts due to the lighter weight of the kiln-dry wood. Additionalbenefits not captured in Table 5.5 include the lower defect rate whenshipping kiln-dried, and the lower logging costs associated with therelatively more plentiful secondary species which may be renderedexportable through drying.

5.13 Financial Analysis. Results of the financial analysis aregiven in Annex 5 and summarized in Table 5.6. It shows the modelinvestment in kiln-drying capacity to be highly favorable from theentrepreneurial viewpoint. Financial rates of return are acceptable overthe range of estimated value-added percentages, the variable exhibitingthe strongest influence on rates of return in the sensitivity analysis.Discounted payback times of 2 to 5 years are appropriately short for thewood products industry, which faces rapidly changing market conditionsfor its products. European agency commissions, the TEDB export levy andmiscellaneous GOG export charges bave been netted out from the benefitstreams. The financial cost per m of product drying is derived in Annex5 as $16.60 for redwoods and $9.48 for whitewoods. The redwood figure isconsistent with estimates made in the World Bank Draft Forestry SectorReview (Annex VII) of $15.00-18.00/m3, kiln drying costs in WesternEurope of $25.00/p and up, and estmates made by Ghanaian mill managersof $15.00-25.001Ie.

.6sew,4mv wissu IMuw ISOi "

5f6 ae".- P0.1,91 *50O 400600A 6"40V smON z i, t 0 6je.430#60600 b4e4 Pe"5 Otdj Aq p*,4..dwA s..4l- 4.4 Si "ooem, - -.-.mwt , 4P_ "

____ _i..siat U9_5 455g,I) *Z . 4"P '_ 1-^ O

c-se sac 5~~~~~Stb c is $esa i-re sib.s 6its xessa xswar zs i.-aa %5 c ir sssu mmcc c-c ee e teSt e ae s e-e xe.t esat xess 10 *-c Ole e iS S onea'

ct s-u at e e~~~~~~~~~~~~~SS cc Ole ce6 xelss xeat es-el IS a's olau . an s se 0a

a** i*7C-W i-c 5-¢ :* 1S 1@ Z5r 5 @5

*ts srn b'srt s-t- * *S 4, Ole ii. ess es-at Xe1ss 1 Ole aasi c 0 sat nwseVsa *Sa *'- *0 55 *esi *5l' 9s-e- * *' Ole e CS **

se c-e ee ases a e~~~~~~~~~~~~~~~Uf Ole O-c *0 esct es-at 12slt N 0l-5 ersi S& 0 565 5"P'-*-ut s-~gt aj. e tsi .me c-c s-c i-c * es-st *s-*t * -ct VS el sn-e e "s i i -So

a's O-l e S, ts sea as sr * c- * e-st es,at e1-,s a a's Ole" "S slc em".. saa~~~~~~~~~. * - sew z *s - - w o *, 9. M-S1 Ima n O. oh * 10 sw J o Z_% 0 an tee*s

c-SC bt5 if'# 5 i i-ses*- I-s S es-si sat esut * baass *68 e e scm _ _c-ca~ ~~~~ a-c tas a ci * u am s-c es-r esx e*sswn ao **bsc * -i 0 C it sea e

ar c-u siS Sa e ao2 ell S., esst esat *w es 0n i c a * OS ceisa s-en te~~~vst c c Ju996 cS," *5W et' xel ware Xvt N 5%is sle5i a c age auss eVws

cit I S's e c sm S-c s-c se t es e.a zeSt N ie st b-elit c a e se*t W10""S* ' s-a t S' d* s * an s'Z i-ca S, es-e es-rt es-si s *'ua *'c is * an cc-is asue c-a ca c sc cuB aess .- ea ac-s Xa' ewS-s a Sss.i cct c * *W r* c-a s--a c-i ca i rot eo- s- eSt *VAS eaI es-es / 5c i w e.g Wa -O * 9l a e

*tb ,i *5 Ve CW on *,Z &,, a1 xesi wr x'est N 6Ol *' W *a a e6.1 10,- S., * * 69a **4 *b Z-% xee 391. cc' 3.St 4 slgz Ole * a 69a * _Om

e's l* ~S. e di ele Olc clc xeses es-ar es-st a c-cl., 0 C 0 W5na'* s Ol cc s u-s O-a le. *-g Xs," es-ar eslsi a c-s -- eS e a is *em *tC_916 %IC 9+.2, es 9% 91, *-t I., Xs," xsa nes 4 `S a 's b 't *' so * gt

i'S S-- ,el * * Its i-c S9 - i xs es" s-ar Xe-Ss a c-c ae e 5 "I e -_el.cts itesi j C C 14 989 sid *.5 s-ft eS-at c i c*sis *,%ft cc oac eapst nwt**

slit _ai i-I e *io *-$a Oei aur leds-egs es-an es-st a C-s ii 50 *I * C *-01'cii *c .s t u-s S'cZ n-il * SC c-s xes-r e." 1S-sU i a-ca cc us a n-i a *swqa

s-ass *.ce sis sica e ii,a s-as *'- s-s es-4n esaa -st N sla. c-s C * iS **l | * g * g X X~~~~~Z&imt s IS :z% 41.C 40lb xes ". 211.1 a S ali S" t St _

i-ac.s c-in-I c-ins a~~~~SCS u-s9 jes.iS i-am; c-gist aac. es-rn xstn Xs, a J,mess e.g1si 614 55ew5%s-tel u-aa s-ear sat, u-ic isis s-cm tt.*Za a-era es-rn es-a es-si a 4-%Q cl tait "I tisi c X-m*c-sc *%i-f t'-is uS st uSSS s-scs C--r fis ses esa ess t t-Ssg 4 Xc fra. ci t *'ss c * _q.us-usez _ zeca _e s _1 it i.a cc1tel n-1 S% ss s-.s slat i-sit ese. war _ess S i1si O ee lea sn See C

------* ------------------------- -- -* --- -- - - - ---- -#* v.43 01 0wa-- ------ ------------ ------- ozS *

frb ses a-ca _tm _r .s _- _-r a-_ esr es _____ -a. es_-t N * .a c- * sc X a>wt sin c *

SeU 4A 6- 11 -1-A 403 $~-MA -1 Pw6-wwqw:`% - - -e -1 so <s004AWA02 1 le _*. w ww.6000 *.4S*d 1* a p SP t@4 PopW 't* IAS_d Polt ej4 _ t'A -vta^ *-t_ cell _t p-su

jeS1 ;4 dw PUP 54.-) 559.0-03 hsi * 140u3 ' .'i P."'G 90't 9 -wd P-iS40uIO P:<e.(@ X

At poNp 514 a ud5 v i 1SSd Sviv-4Io

8NIAMO NuN1 V3V1 N0fl0W4J (13M1 3nlVA AVitiWI 1SMOS Q± wouIuAIWii SSOue lVuN3L0cd :~S* etqs±

- 57 -

Table 5.6: FINANCIAL ANALYSIS RESULTS, SAWMILL PROCESS HEATUNIT PRODUCTION MOCEL

Value-Added NPV FIRR Discounted Payback(S) (S) (years)

Low Case 1,487,890 27 <5

Medium Case 2,833,580 42 <3

High Case 4,180,270 56 2

5.14 Economic Analysis. Economic analysis results are given inAnnex 5 and Table 5.7 and show returns which exceed those found in thefinancial analysis. The 4 percent commission paid to overseas agents hasbeen considered a real resource cost and has been netted out from thebenefit stream; other export fees are domestic transfer payments and havebeen disregarded for the evaluation from a social viewpoint. Electricitycosts, charged at the marginal ECG industrial tariff rate in thefinancial analysis, are charged at the full marginal grid cost inthe economic analysis.

Table 5.7: ECONOMIC ANALYSIS RESULTS, SAWMILL PROCESS HEATUNIT PROOUCTION NOCEL

Value-Added NPV FIRR Discounted Payback

(S) (S) (years)

Low Case 1,664,890 30 <5

Medium Case 3,167,390 46 <3

High Case 4,670,670 62 2

Cogeneration at Grid Connected Mills

5.15 The potential for on-site cogeneration of electricity may beenhanced given the increasing demand for on-site process heat. Atpresent, none of the sawmills in Ghana that have access to the nationalgrid generate or cogenerate their electricity. There are several factorsthat have influenced this development. The primary reason was therelatively low cost of grid electricity in the past. Electricity tariffshave increased by nearly 400 percent since mid-1983. Even with theseincreases, grid electricity is still relatively inexpensive averae,ing

- 58 -

about 4.96 cedis/kWh (3.31 US cents/kWh) for a large sawmill under theindustrial tariff. Under the present circumstances, it is unlikely thatsmall scale cogeneration systems will be competitive. However, thepossibility for cogeneration of electricity at mills with log inputcapacities above 40,000 m' and with high process heat and electricitydemarids is worth investigating. One such mill in the Kumasi area, STP,is presently undertaking such a project utilizing a 5 year old 480 kVA(384 kW) cogeneration system purchased from a sawmill in West Germany atan installed cost of US$540,000.

5.16 There are only 6 mills with capacities above 40,000 m3 and withaccess to the grid. All of these mills are located in the Kumasi area.With the exception of STP, none are contemplating installing cogenerationequipment although three of these mills currently have process heatgenerating equipment installed. The total on-site residue consumptionpossible, assuming all of the six mills were to eventually installcogeneration systems to utilize all their residues, would account for110,000 m /yr of residue consumption equal to 40 percent of the totalsupply in the Kumasi area.

5.17 Project Description. The STP mill in Kumasi is again selectedto evaluate the technical and economic potential of cogeneration at gridconected mills. The objective in this case is to meet all process heatneeds and to generate as much of the mill's electricity requirements aspossible consuming all residues produced. Surplus electricity generatedduring the off-shifts could potentially be sold to the grid. Anevaluation of this prospect will be discussed later based on theresulting marginal costs of the cogenerated electricity.

5.18 Process heat demand at STP is estimated to be 3.7 tonnes/hr forthe kiln driers and steaming vats. The investment and operating costs forthe associated boiler, kiln driers and steaming vats were outlined in theprevious section. They are restated, net of costs of the dry kilns andsteaming vats, as STP Alternative 1 in Annex 6. In this analysis onlythe incremental investments and operating costs for generatingelectricity are relevant to the decision to invest in cogeneration. Twoalternatives are considered for the marginal analysis. The firstincorporates a back pressure and a condensing turbine with the exhauststeam from the back pressure turbine satisfying the process heat load.The second alternative incorporates only a condensing turbine with theprocess heat loads supplied directly by expanding the high pressure steamthrough a pressure reducing valve. Brief descriptions of both thesealternatives follow.

5.19 STP Alternative 2. In this option, c 8.5 tonne/hrsawdust/wood-waste fueled boiler generating 28 kg/cm gauge, 315 °Csteam is required. The steam is su.plied to a 288 kW backpressureturbogenerator (TG) exhausting at 5.3 kg/cm2 and a 575 kW condensingturbogenerator. An average load factor of 80 percent plus 13 daysscheduled downtime is assumed for all systems. Steam exhausted from thebackpressure TG is designed to satisfy the heat loads of the kiln driers

- 59 -

and steaming vats estimated at 3.7 tonnes/hr steam. An average of 3.1tonnes/hr of wet steam exhausting from the condensing TC will passthrough a condenser with the condensate being returned, along with thecondensate from the vats and kilns, to the boiler feedwater tank. Thecondenser heat is to be rejected through a mechanical draft coolingtower. A schematic diagram of the proposed alternative is presented inFigure 5.2.

Pigure 5.2: STP LTD. - SCH8MATIC - ALTERNATIVE 2

f LASH STfEAM

Alternative 2 09tEhA

BOILER PRESSURE TGFUEL (28kg/cm2 gauge 6.8Vh _ 1 . t

22,600 Va 31 Sa CONDENSING@ 38%mcwb TG WWCONDENSATE

t~~~~~~~~~~_t 3 RECEIVER

6.8 Vth COOLING OOWERU

FEEDWATER 3 WhTANK Q 8

MAKE-UP WATER0.9tSh

5.20 Wood residues would be received at the the boiler plant fromthe sawmill. Oversized residues would be screened and hogged and placedin a storage bin. Sawdust residues will be mechanically conveyed to thefurnace. The boiler and TC units are assumed to operate 3 shifts a dayfor 350 days/yr. Average power generation is 690 Kw vhich would meet allof the mill's requirements when it is in operation plus allow surplusesto be sold back to the grid. Total residue consumption by the boiler isestimated to be 22,800 tonnes/yr with average moisture content of 38Xmcwb. This is equivalent to all the residues produced on site whey themill is operating at its projected maximum capacity of 70,000 m logthroughput.

5.21 Costs. A summary of the estimated incremental capital andoperating costs for instituting this cogeneration option at STP ispresented in Table S.8. A detailed breakdown of the total andincremental costs is given in Annex 6. The incremental costs reflect theneed for a larger capacity and higher pressure boiler, the addition ofbackpressure and condensing TOe, condenser and cooling tower systems and

- 60 -

larger buildingst pipes, valves, tanks. etc. that are associated withthis option. Total incremental C&8 costs amount to US$814,000 with totalincremental investment costs equaling US$1,740,000.

Table 5.8s SU#4ARY OF INCREMENTAL CAPITAL AND ANNUAL OPERATINGCOSTS FOR STP CGEiNERATIOt ALTERNATIVE 2

Local Costs Foreign Costs TotalItemn (o000 USS) '000 US$) ('000 USS)

Incremental Capital CostsConstruction 29.0 8.0 37.0EquIpment 777.0 777.0Transport Charges a/ 76.7 75.4 152.1Engineernlg, Installationand Continogncles 284.5 489.7 774.2

Total Incremental Capital Costs 390.2 1350.1 1740.3

Incremental Annual Operating CostsLabor 0.9 0.9o & m 28.0 28.0Consumables b/ 29.6 28.0 57.6Contingenclies 8.6 8.6

Total Incremental Annual 67.1 28.0 95.1Operating Costs

a/ Includes International freight and Insurance, local port charges and bankfees.

b/ Includes lost revenues from reduced slab and offcut sales.

5.22 The total incremental operating costs are estimated atUS$95,100. This reflects the need for additional supervisory labor tooperate the larger high pressure boiler and higher operating, maintenanceand consumable costs associated with the boiler, turbogenerator and watertreatment systems. In addition, slabs and offcuts available for sale inthe Alternative 1 scenario are combusted in this alternative, and areaccounted for as a cost. Details of the estimated operating costs arepresented in Annex 6.

5.23 Benefits. The primary incremental benefit of this option isthe cost savings from on-site power generation and revenues from the saleof surplus electricity. It is estimaL.:d that a total of 4.55 GWh/yr (netof internal parasitic consumption) will be generated under this option.Assuming a buy/sell value equivalent to the average cost for gridelectricity at STP of 3.31 US cents/kWh results in a potential annualfinancial benefit of US$150,290. Other benefits include increasedreliability in electricity supply (assumes grid as back up) andelimination of environmental costs (non-quantified) of sawdustdisposal. The value of lost production due to supply interruptions wasput at US$30,000 per year by the STP management, but was riot included in

- 61 -

the analyses on the assumption that ECG reliability will continue toimprove.

5.24 STP Alternative 3. This option also calls for a 8.5 tonne/hr,28 kg/cm2 gauge, 315 'C steam boiler fueled by sawdust and waste wood.The boiler operates at an average 80 percent load factor producing 6.7tonnes/hr of steam. An average 3.5 tonnes/hr of high pressure steam issupplied to a 650 kW condensing turbine Yhile 3.3 tonnes/hr is expandedat a pressure reducing station to 6 kg/cm gauge to meet the process heatl.oads of the plant. The wet steam exhaust from the TG passes through acondenser with the condensate being returned to the boiler feedwatertank. Condenser heat is dissipated by means of a mechanical draftcooliag tower. Condensate from the kilns and vats is also recycled tothe boiler feedwater tank. The primary advantage of this design whencompared to STP Alternative 2 is that power generation is independent ofthe plant's need for process steam. Fluctuations in the process heatloads of the plant will, in Alternative 2, require either a reduction inpower output or a dumping of the exhaust steam from the backpressureTO. In Alternative 3, excess steam due to a drop in the process heatload could be channeled to the TG to generate additional electricity. Aschematic diagram of the proposed alternative is presented in Figure 5.3.

Figure 5.3: STP LTD. - SCHEMATIC - ALTERNATIVE 3

DESUPERHEATINGSPRAY WATERFLSSTA

Alternative 3 04 FlASH STEAMl

PRV

6.8 t/h 3.3 t/h12thBOILER

FUEL (28kg/cm? gauge22,800Va 31/aC)@ 38%mcwb -S20kW

COOLING OWER CONDENSATE

6.8 t 3h COOLING RECEIVER

FEL 1'W4TER 3.51/hFELt NO~~~~~~~~~~~~TER ~~~~~2.8 t/hTANK E

I MAKE-UPWATER0.5 Vh

5.25 Wood residue requirements, processing, handling, storage anddelivery are similar to STP Alternative 2. The boiler and TG unit areassumed to operate 3 shifts/day for 350 days/yr. The maximum possiblepower generation would be 650 kW which would occur if the condensing TCwere operated at full capacity. However, the average power output at 80

- 62 -

percent load would be only 520 kW. All wood residues generated on-sitewould be consumed by the boiler.

5.26 Costs. A summary of the incremental capital and annualoperati,ng costs when compared to STP Alternative 1 is presented in Table5.9. The incremental C&E costs amount to US$740,000 while the totalincremental capital costs are equal to US$1,587,900. Incremental annualoperating costs are equal to US$90,700 which is primarily due to theadditional costs for operation, maintenance, consumables and lost residuesales. Details of both the capital and annual operating costs arepresented in Annex 6.

Table 5.9: SUNMARY OF INCFREKNTAL CAPITAL AND ANNUAL OPERATINGCOSTS FOR STP 00-GENERATION ALTERNATIVE 3

Local Costs Foreign Costs TotalItem ('000 USS) ('000 USS) ('000 USS)

Increwntal Capitol CostsConstruction 21.0 6.0 27.0EquIpment 713.0 713.0Transport Charges I/ 70.4 69.2 139.6Engineering, Installation

and Contldgencles 259.9 448.4 708.3

Total Incremental Capital 351.3 1236.6 1587.9Costs

Incremental Annual Operating CostsLabor 0.90 & M 26.0 26.0Consumables b/ 29.6 26.0 55.6Contingencies 8.2 - 8.2

Total Incremental Annual 64.7 26.0 90.7Operating Costs

a/ Includes Intersatlonal freight arJ Insurance, local port charges and bankfees.

b/ Includes lost revenues from reduced slab and offcut sales.

5.27 Benefits. The primary incremental benefit of this option isthe cost savings from on-site power generation and revenues from the saleof surplus electricity. It is estimwted that a total of 3.21 Gth/yr(net) will be generated under this option. Assuming a buy/sell valueequivalent to the average cost for grid electricity at STP of 3.31 UScents/kWh results in a potential annual financial benefit of

- 63 -

US$1O6,11O. Other benefits are similar to those discussed under STPAlternative 2.

5.28 Financial Analysis of STP Alternatives. Financial analysisresults are derived in Annex 6 and sumarized in Table 5.10. Theanalysis shows that while Alternative 2 io the more ieasible of the twooptions, there are inadequate financial incentives for project adoptionby mill owners. Either the average industrial electricity tariff wouldhave to rise to approximately 7 US cents/kWh, or the incremental capitalcost of the installation would have to drop by 76 percent. In thiscontext, it should be noted that the second hand-equipment actually beinginstalled by STP was obtained at substantial discount relative to the newequipment considered in these models.

Table 5.10: FINANCIAL ANALYSIS I£SULTS,GRID CONNECTED COGENERATION MMOELS (STP)

Alternative 2 Alternative 3

NPV S(1,320,460) S(1,470,630)

FIRR Undef. Undef.

Capital Cost Switching Value -76S --

Marginal Cost of ElectricityGenerated 7.1 UStAkWh 9.3 UStAkWh

5.29 Economic Analysis of STP Alternatives. Details of the economicanalysis of Alternatives 2 and 3 are givenin Anex 6 and summary resultspresented in Table 5.11. Alternative 2 is again the favored option, butat a marginal generation cost of 7.6 US cents/kWh the electricityproduced is not competitive with the marginal cost of grid hydropower of5.2 US cents/kWh. Equivalence would not be established unless theeconomic cost of the required incremental investment were to drop by 49percent. In the economic evaluation, the value of slabs and offcuts hasbeen raised to reflect the replacement value of their closest substitute,forest wood.

Cogeneration at Non Grid Connected Hills

5.30 At present all except three of the major sawmills are connectedto the national grid. )f the three non grid connected mills, only one,Cliksten West Africa Ltd. (GWA) in Sefwi-Wiawso, is presentlycogenerating electricity and steam when the plant is in operation. Inaddition, electricity is generated with diesels to meet some plantrequirements and the domestic requirements of mill personnel. However,

- 64 -

the national grid now passes within one kilometer of the mill and presentplans are to connect to the grid sometime this year. GWA has initiatedprocurement of the transformers to make the connection. Once connectedto the grid, GOA plans to stop cogeaerating steam and electricity and touse its residues for the generation of its steam requirements only. Thesecond mill with the capacity to cogenerate is African Timber and Plywood(Ghana) Ltd. (AT&P) in Samreboi. However, the plymill at AT&P has beenshut down since 1984 and the sawmill was shut down in early 1986. Themill is presently undergoing a major rehabilitation effort that is beingfinanced externally. Once complete: it is expected that the mill willuse its residues to cogenerate all its electricity requirements. Thereare presently no plans to bring the national grid, which is 28 km away,to AT&P.

Table 5.11: ECONOMIC ANALYSIS RESULTS,GRID CONNECTED COGENERATION MODELS (STP)

Alternative 2 Alternative 3

NPV S(836,000) S(1,185,120)

FIRR Undef. Undef.

CapI.sl Cost Switching Value -49%

Marginal Cost of Electricity 7.6 USt/kWh 10.1 USt/kWhGenerated

5.31 The Mim Timber Co. (NIM) is the only major operating mill whichpresently generates all its electricity with diesel fuel. MIM had a 420kVA steam engine/generator installed in 1980 but ceased to operate thesteam engine shortly thereafter because engine lube oil was contaminatingthe condensate which was subsequent'y damaging the boiler. MIM is 47 kMfrom Sunyani where the national grid will pass and VRA's present plansare to extend the grid to MIN in 1989 at a estimated capital cost ofUS52.2 million. The total 1985 electrical demand at MIM, the adjacentScanstyle furniture factory, Nim Agro Ltd. and the surrounding commnunitywas 2.2 MW peak and 6.4 Gwh per yea , all met by diesel generation. KIMpresently generates over 45,400 m of wood residues of which only 7percent is used to meet its process heat needs. Of the balance, 18percent is sold as firewood, 24 percent is converted to charcoal and 52'1' percent is burnt in open pits or disposed in landfills. The residuespresently burnt or disposed could, if converted at a 10 percentefficiency, supply approximately 7.7 Gwh/yr of electricity.

- 65 -

5.32 HIM Alternatives. Tne cituation at HIN is unique because ofthe proposed grid extension. Financial justification for extension ofthe grid to MIM has been based on the projected consumption ofelectricity by the MIM complex (mill, furniture factory and domesticconsumption). If the grid is extended to MIM, the alternative againstwhich cogenerated electricity must compete is grid electricity. However,if the grid is not extended to KIM, the alternative is dieselgeneration. To date, there seem to be conflicting views as to theinvestment commitment for the grid extension. The General Manager of VIAhas indicated that the investment decision is final and that the gKidwill be extended regardless of MIM's ultimate decision. If this is thecase, the investment for grid extension must be considered exogenous inthe economic evaluation of the cogeneration options. On the other hand,it is possible that extension of the grid could be postponed and theassociated investment saved if cogeneration at MIM is economically moreattractive than grid electricity. In this case, the investment for gridextension must be incorporated into the economic evaluation of thecogeneration options.

5.33 Evaluation of the cogeneration options is further complicatedby the fact that the electricity load profile at NIM has a wide variation(see Figure 5.4). After the planned mill modernization and installationof a new moulding line and additional kilns within two years, the, MIMcomplex (excluding Mim Agro) will have a 2.2 NW peak demand, an averagedemend of 1.4 NW (based on a weighted average operating period of 4,800hr/yr) and a continucus base load of 450 kW. Conversely, the steamprocess heat demand is relatively constant averaging 7.1 tonnes/hr duringplant operation and 4.4 tonnes/hr at all other times. Given thesecircumstances, several alternatives exist for design of a cogenerationoption. One option is to build a cogeneration plant which provides allprocess beat and baselintermediate load electricity while eithergenerating peak demand with diesel or, depending on the grid extensionassumption, obtaining it from the grid. A second option is to design acogeneration plant that can meet all electricity demand including peakdemand. A third option is to design a cogeneration plant thAt willconsume all available wood residues, provide all procesF heat needs andgenerate electricity in a manner most efficiently matched to the procesaheat demand profile while allowing the plant to either purchaseelectricity from the grid during periods of deficit and sell it to thegrid during periods of surplus. Another option might be to expandthrough a backpressure TG only that steam which is needed for processheat while' purchasing remaining electricity requirements from the grid.Five distinct options have thus been identified for detailedevaluation. However, the economics of two of the options are also testedunder the assumption that the inves.ment for the grid extension is a sunkcost. A summary of all the options considered in this evaluation ispresented in Table 5.12.

MIM COMPLEXDAILY LOAD PROFILE

Power (xOO kW)

28 -~ ~--1 it- ----- - - - ----- - - - -i--------l f

l F ~~PeakPeriod

10 . .. .... ....

'0 - - - --- - -- o ;S7V\ t \\&4 % > L : -C

0 6 10 is 20 24Time of Day

- 67 -

Table 5.12: 1IlM COGENERATION OPTIONS OECISION MATRIX

Assume No Grid Extension Assume Grld Extension(Real Resource Cost) (Sunk-Cost)

Base Case A100% Diesel GenerationProcess Heat Boller

Option #1 A Base Case iExtend GridNo on-Site Generation No On-Site GenerationProcess Heat Boller Process Heat Boiler

Option 12 A2.C6 M Wood-firtd Co-GenerationDiesel Back-up

OptIon 13 A1.2 #W Wood-fired Base Load Co-GenerationDiesel Peak Load Generation

Optloa 14 AExtend Grid260 KM wood-fired Co-Generatlon

Option ff A Option ff 8Extend Grid 1.2 1W Wood-fired Co-Generation112 1W Wood-fired Co-GeneratIon

5.34 Project Description. Nim Timber Co. Ltd., the largest a1.iglewood processing plant in Ghana, is a well managed state owned milllocated in the Brong Ahafo region. The mill production facilitiesinclude a sawmill which produces sawn timber and veneer flitches, aveneer slicing plant, a moulding plant and six lumber dry kilns.Adjacent to the mill is Scanstyle Him Ltd., the largest knock-downfurniture factory in Ghana. Also in the vicinity of the mill is Kim AgroCo. Ltd., a collection of privately owned farms.

5.35 At present, all electricity in the Mim area is generated bydiesel engines. The primary producers of electricity are Mim Timbers,Kim Agro, and Scanstyle. Kim Timbers has a total installed dieselcapacity of 3,126 kVA, Scanstyle has 600 kVA and Kim Agro, 640 kVA. In1985, Nim Timbers generated approximately 5.1 Glh of electricity andconsumed 424,055 Imperial gallons (Igal) of diesel fuel and 11,250 Igalof lubricating oil. At the same time Him Agro generated 6.5 KWh andconsumed 43,800 Igal of diesel oil and 2,080 Igal of lubricating oil,while Scanstyle generated 6.0 KWh burning approximately the same oil

- 68 -

quantities as at Mim Agro. Given diesel costs of 130 Cedis/Igal andlubricating oil at 809 Cedis/Igal, the fuel and lubrication costs alonefor generating electricity at Mim was about US$526,00O or equivalent tcU88.3C/kwh. Mim Timbers supplies some electricity to Scanstyle, to theresidences of the management of both Mim Timbers and Nim Scanstyle andfor some social services in Mim Town. A summary of the Him areaelectrical demand and consumption for 1985 is shown in Table 5.13.

5.36 Steam is raised in two facilities at Mim Timbers. The first isan oil-fired Cleaver Brooks firetube boiler which supports the existingsix lumber dry kilns; the second is a Lambion wood-fired furnace andSillar & Jamart high pressure boiler originally designed to provide steamto the now inoperative Spillingwerke steam engine and for the steamingvats and veneer driers. The Sillar & Jamart has a maximum capacity of6.5 tonnes/hr of 23 kg/cm2 gauge steam while the Cleaver Brooks is ratedat 1 tonne/hr at 10 kg/cm2. A third Kewanee-designed, Korean made wood-fired locomotive style boiler is on site but has Pot been installed.This boiler is rated at 2 tonnes/hr steam at 10 kg/cm .

5.37 Him Timbers has limited concessions and has nearly depletedmost of the economically available primary species. The managementrecognizes that operations must start shifting to process more secondaryspecies if the mill is to continue to operate at near productivecapacity. Mim management predicts a dramatic downward shift in the millsprocessing of primary species within the next 5 to 10 years. Toaccomodate this shift, Mim Timbers is planning to add more kiln dryingfacilities and steaming vats. Management estimates that an additional 32kilns will be required. Management is also emphasizing more secondaryprocessing and plans to enlarge the moulding plant. MIM's main sawmillis aging and is expected to be replaced in two to three years.

- 69 -

Table 5.13: MIN AREA ELECTRICAL DEMAND AND CONSUMPTION, 1905

Power DemandInstalled Peak

Use Copacity Demand(lcW) (kW)

SavoIII a/ 1,400

KiIns b/ 25

Veneer m I1 300

Moulding plant c/ 152

Shops 150 Combined:1600 kW

Offices 50

Water supply 92

Comunity 300

Him Scanstyle - 300

Min Agro 300

Total 2,200

Consumption (kWh/year)

Him Complex generation d/ 5,147,000

Scanstyle generation 600,000

Him Agro 650.000

Approximate Total 6,400,000

a/ Future replacement sawmill anticipated to have approximately thesame demand and consumption as existing.

b/ Kilns anticipated to Increase to 32 from current 6. Increase of 140Kd anticipated.

c/ Moulding plant anticipated to Increase from 152 to 450 kw withexpansion of capacity.

I/ Includes 360,000 kWh/year sold to Scanstyle.

Source: Mim Timber Co.

- 70 -

5.38 Mim Timbers has a processing capacity of 90,000 m3/yr of logthroughput. In yq86, Mim processed 81,000 mi of logs and, as a result,produced 45,400 m of wood resi ues. The average bone dry density of thewood was estimated at 590 kg/m witb an average moisture content of 34percent mcwb. With the shift to secondary species it is estimated thatthe average bone dry density will decline to 510 kg/m and the averagemoisture content will increase to 37 percent mcwb. A summary of the useof residues in 1985 and 1986 is given in Table 5.14.

Table 5.14: UTILIZATION Of RESIDUES AT HIM

1985 1986

(a3/yr) CS) (C3/yr) (t)

Boller Fuel 3,021 7.2 3,173 7.0

Sold Cmmerclally 7,355 17.5 8,054 17.7

Charcoal Production 9,807 23.3 10,739 23.7

Disposal by Burning 21.868 S200 23411 51.6

Totes 42,051 100.0 45,377 100.0

Source: Ru-Tek..

The Mim management indicated that firewood and charcoal were sold only tomeet the needs of the Mi community. Firewood was sold at 2000 cedis pertipper truck (about 5 mi) and 80 percent of the charcoal was sold atsubsidized prices to employees while 20 percent was sold to dealers at180 cedis per sack (about 40 kg). The sale prices for charcoal are givenin Table 5.15.

Table 5.15: COAL SALES PRICES, HIM TIMBER 00.

Produ.jtoo Share Consuer (C*qdtaS*ck)

10 Jr. & Sr. Staff 0

60 Other Staff 30

10 Scanstyle 100

20 Charcoal Dealers 180

Source: Nim Tlwer Co.

- 7) -

5.39 The Mim area presents an ideal situation to investigate thepotential for cogeneration because of: electricity and process heatrequirements; wood residue availability; management support at the mill;present high costs of diesel generation; and present isolation from thenational grid. A brief description and evaluation of the variousalternatives that are considered follows.

5.40 MIM Base Case A. This base case is defined to be 100 percentdiesel power generation with all process heat requirements generated withwaste wood. The national grid is assumed not to be extended. Futureelectricity consumption, excluding Him Agro, is estimated to grow to 6.7GWh/yr mainly due to the addition of kiln capacity, expanded mouldingplant and replacement sawmill. This represents a 0.9 GWh/yr increase or16 percent over the 1985 consumption level. An annual average powerconsumption of 1,406 kW and a peak of 2,200 kW is expected. Thisincludes a 300 kW load from Scanstyle. Electricity production willcontinue with existing diesel capacity with one-eighth of the dieselcapacity being replaced annually.

5.41 Following installation of 26 dry kilns additional to thecurrent 6, total process heat demand is expected to average 7.1 tonnes/hrwith peak demand reaching 10.1 tonnes/hr. In order to meet theadditional process heat requirements, a 1.6 tonne/hr, 10.6 kg/cm gaugesaturated steam wood fueled boiler will be installed to complement theexisting wood fueled Sillar & Jamart and Kewanee boilers. These threeboilers will have sufficient capacity to meet the peak process heatdemand from the kilns, steaming vats, veneer driers and veneer plant.Condensate would be collected and retur-4 to a boiler feedwater tank. Anew reinforced concrete building to house the new boiler is required. Itis proposed that the new boiler be manually fired to minimise capitalexpenditure and utilize the low cost labor. The estimated wood residuerequirement between the three boilers is 18,700 tonnes/yr averaging 34percent mcwb. This residue is available on-site. A schematic diagram ofthe process heat plant is presented in Figure 5.5.

- 72 -

Figure 5.5: HIM TIMBERS LTD. - SCUEMATIC - BASE CASE A/ALTERNATIVE IA

SILLAR & JAMART NEW LOW PRESSURE KEWANEE LOCOMOTIVEBOILER BOILER BOILER

(23 kg/cm2 gauge, 3250C) (10.6 kg/cm' gauge) (10.6 kg/cm2 gauge)

s.ltVh (<6.St/h) -1.6) V/h 2Vth (2Vth)j ~~~~~PRV

2.4 th (3.3th).4t/h (4.9 th)

16 kg/cm2 gauge 0.5 kg/cm2 gauge

Veneer Dryer - 2.7 tVh (3.2 t/h) 32 Kilns - 3.2 th (5 4 t/h)Veneer PlantSteaming Vats - 1.2 th (1. Sth)

Total - 2.7 t/h (3.2 th) Total - 4.4 t/h (6.9 tVh)

5.42 HIM Alternative lA/KIM Base Case B. This alternative requiresthat only the process heat plant be built and that all electricity willbe purchased from the national grid. The diesel generators will bemothballed with a base load capacity maintained for standby emergencypower. Given the assumption that the grid extension is not committed,this alternative is termed Alternative 1A and the cost of the grideztensien must be incorporated in the economic analysis. For the purposeof eco iomic evaluation, this alternative is considered "KIM Base Case B"when the investment for the national grid extension is assumed committed(i.e. a sunk cost). All aspects of the process heat plant are similar tothat outlined in "Base Case A". A schematic diagram of this alternativeis presented in Figure 5.5.

- 73 -

5.43 MIN Alternative 2A. This option requires that all electricityand process heat be generated with wood residues, and that the dieselgenerators be maintained only for emergency back-up power. It isdifficult to design a system that can operate efficiently throughout theload profile because of the wide variation between the base, intermediateand peak power loads at Mim (see Figure 5.4). The least cost solutionfrom an equipment standpoint would call for one single condensing TG setrated at 2.4 to 2.6 MU to handle the net 2.2 MW peak demand. However,operating a large turbine at below rated capacity a majority of the timewould incur serious operating efficiency penalties. Prom an energyefficiency standpoint, the design would dictate selecting two or threesmaller TG sets with the possibility of one or two being back-pressureturbines exhausting steam for use in the veneer driers, kilns and vats.The final design of the system would need to be based on a carefulevaluation of turbine costs and efficiency curves and the availabilityand opportunity costs of the input fuel (i.e., wood residues). Forpurposes of this analysis, the 2 turbine arrangement is chosen to derivea conservative costing and to allow maximum kWh production per tonne ofresidues. Full condensing capability is provided for efficiencypurposes, but turbine bleeds are provided to satisfy kiln needs for lowpressure steam At the final design stages the possibility of onecondensing and one back-pressure turbine would be evaluated before makingfinal hardware choices.

5.44 Another issue that has to be addressed at the final designstages is whether or not a sufficient water supply is available duringthe dry season to allow the use of water cooled condensers even with acooling tower. It is estimated that a 2.4 MS plant with condenser andmechnical draft cooling tower will require up to 1000 1/min of water tosatisfy all plant needs at full load. It is possible that an air cooledcondenser is the best choice for this site with an operating penalty ofincreasing the turbine exhaust pressure by 3.4 kPa (1 in. Hg Abs) to avalue of about 13.5 kPa (4 in. Hg Abs). The financial impact is notclear at this time as the river supply is at least 1 km away and thepotential need for additional piping and pumping capacity will impact anycost analysis. It is assumed that the five month dry season has a severeimpact on water availability such that the air-cooled condenser is chosenfor this analysis as the safe choice.

5.45 For this option, a new 24.5 tonne per hour, 23 kg/cm2 gauge,3250C wood/wood-waste burning boiler is required to supplement theexisting high pressure Sillar & Jamart boiler. A 24.5 tonne per hourboiler is required due to the relatively low steam temperature (325'C)being used in the power cycle. Depending on the efficiency of theturbines commercially available, the steam rate per kWh could be higherthan that used in this analysis (i.e. greater than 9.6 kg steam per kWh)and an even larger boiler could be required. In this scheme the Kewaneelow pressure boiler is not needed and should be sold to recover as muchcapital as possible. The higher pressure boilers would supply steam tothe turbine at a minimum rate of 9.6 kg steam per kWh. The total boilercapacity required is based on bleed steam being unavailable at all times

- 74 -

and the boilers being capable of supplying all kiln, vat, and dryer steamneeds. The kilns and vats would normally be supplied by turbine bleedsand supplemented by boiler steam only on an as-needed basis. Allcondensate would be collected for return to the boiler feed water system.

5.46 A new reinforced concrete or other suitable type building wouldbe provided as an extension to the existing boiler house. It is proposedto provide a simple mechanized fuel feed system consisting of a smallfront-end loader, a screen, a hogger and storage bin, and robustconveyors to feed the furnace. Oversize material wouid be rejected andwould require manual size reduction before being re-fed. Fuel storagewould consist of an enclosed part of the new building where the front-endloader could operate. A schematic diagram of this alternative ispresented in Figure 5.6.

Figure 5.6: MIM TIMBERS LTD. - SCHEMATIC ALTERNATIVE 2A

NeW NoH p4*l3 F a a t

(23k*%Ws' ug3)5 (23 3,a~.s2

l6.0t/h (24.5t/h) 4.4t/hj(6.St/h)

13.3t/h 1400kW 0 t/h design base "(20.9t/h) (2200kW)

VheWuovp *2.7tih(S.Zt" )2 KiIfl *).2Vbh(.4 tAl)

__________________ teamiungVs"U *2vh(lStA**Z. IUh 3.Z tPh Tomg 4 4 th (6.9 UN'

~~1~~IL ~COt4VSA?UII 4TM "rEegattum

*AVW8g (~4a"VW) .

5.47 This option would not require the continued operation of theexisting diesel generators except in an emergency. Therefore,synchronizing gear has not been included in the cost estimates. Thewood-fired plant would produce 6.01 million kWh per year net to users,and 34,,000 t/yr of steam supply to the kilns, vats, and veneer dryer.This will require up to 38,800 tonnes per year of wood residues at anaverage mcwb of 34 percent. Parasitic power requirements are estimatedat 10 percent of gross output and losses at 8 percent. The annual

- 75 -

capacity utilization factor for the plant would be about 31 percent withan assumed plant avai.ability of 88 percent (*6 weeks per year with 3weeks of scheduled down time and 3 weeks of unscheduled down time). The31 percent utilization factor is low primarily due to the large capacityrequired (1.4 MW above intermediate load) to cover a peak demand thataccounts for only 13 percent of the total power produced. 11/

5.48 MIM Alternative 3A. This option requires that wood be used toproduce all of the power needed to satisfy the base and intarmediateloads, and that peaking be supplied by the existing diesels. As the baseload plus intermediate load results in a maximum demand of 1 NW, a 1.2 MWcondensing TG set is proposed with a turbine bleed provided to supplykiln and vat steam. As an option for final design, the efficiency curvesfor two 600 MY turbines and one 1200 kW turbine should be compared usingload demand profiles to find the optimum solution. Also for final designpurposes, the possibility of a back-pressure turbine wiLh only partialcapacity for condensing supplemented by the condensing ability of thekilns and vats should be evaluated.

5.49 In this option, a new 13 tonne per hour boiler is required atthe steam conditions of the existing Sillar & Jamart boiler. The newboiler capacity is determined assuming a steam rate per kWh at 9.6 kgplus 10.1 tonnes per hour capacity needed fur process steam. The boilersize might have to be increased depending on the actual steam ratescommercially available for boilers in this size.

5.50 Following the same design direction as in Alternative 2A, anair cooled condenser is selected to help reduce the water consumption. A1.2 NM plant using a condenser and cooling tower would require 5001/min. The existing low pressure Kewanee boiler is not needed in thisoption and its sale is recommended. Similar boiler house additions, woodfuel storage, processing and feed systems are required as specified forAlternative 2A, but scaled down to match plant capacities. A schematicdiagram of this alternative is presented in Figure 5.7.

11/ If this plant were connected to the grid and the grid could acceptthe excess capacity during non-peak hours, the plant could supply9.50 million kWh per year to the grid based on an 881 availabilityfactor and a 10 hour per day non-peak period plus two days per weekwhen the mill is shutdown. As a power plant at this size canproduce up to 400 kWh per tonne of wood fuel at 34t mcwb, anadditional 23,750 tonnes per year, at a minimum, of wood fuel wouldbe required for a total annual waste wood require,ant of up to62,550 tonnes. This is approximately 23,000 tonnes more than thetotal residues produced at KIM.

- 76 -

Figure 5.7: MIM TIMBERS LTD. - SCHEMATIC-ALTE*MATIVE 3A

fEdW WIS PIS5MrASLA JOMAA

likelwmqnn (nkW PLM

10. Ot/h (13.lt/h) 4.4t/ (6.5t/h)

7.3t/h 770kW 0 th designb(9.5/h) (1000kV)

is _ 'Wreyw 0.5 IWmVT w Ouw *?tAV * I2%ZE" iln *) *th (S.4

VAW P1N

t~F .tml TOW .44 h (6.95 t

Se, 9*w eret e*Avvgp (fUimw.j -

5.51 In this option, the power plant would supply up to 5.25 millionkWh per year net to users satisfying the base and intermediate loads.The diesel requirement would be for 1.2 MW of capacity but would produceonly 833,000 kWh per yes, peaking load. Additional diesel service of617,000 kWh for scheduled and non-scheduled steam plant outages can bepredicted. Synchronizing gear is included in the plant costing to allowparallel operations, although the option of operating isolated equipmentshould be reviewed. The estimated combined availability/utilizationfactor for the steam plant is about 50 percent assuming 10 percentparasitic requirement and 8 percent losses. The estimated wood residuerequirement for this option is estimated to be 35,000 tonnes per year at34 percent mcwb.

- 77 -

5.52 MIM A'ternative 4A. In this option, a new 3.8 tonne perhour 12/, 23 kg/cp, 325C wood-fired boiler along with a back-pressureturbine would be provided. The existing Sillar & Jamart boiler wouldalso be retained. The system would allow for only back-pressureturbogeneration of up to 440 kW of electricity and enable the turbineexhaust to satisfy the low pressure process heat loads of the kilns andvats. The veneer dryer load would be satisfied by boiler steam. Thesteam rate for the turbine would be about 16.4 kg per kWh. Given steamrequirements in the plant, it is estimated that the cogeneration systemwould generate power at an average of 270 kW with an availability factorof 88 percent. Thus, total electricity generated would equal 2.09million kWh per year.

5.53 A new reinforced concrete, or suitable substitute, building tohouse the new equipment is required as an extension to the existingboiler house. It is proposed to hand-fire the new boiler to minimize theoverall cost and to utilize the low cost labor available. A schematicdiagram of this proposed alternative is shown in Figure 5.8.

Figure 5.8: MIM TIMBERS LTD. - SCHEMATIC - ALTERNATIVE 4A

3.2t/h (3.6t/h) 3.7t/h (6.5tih)

4.4t/h 270kW(6.9t/h)l

0.2t/h (2.7t/h) (423kW)

Vmw Ow *LltA0blO 33 Kilns D*t

cown~~~~~~~o maco'oes.salI ~, _______________ Stesosmfi VON f .1 2t1(.SAtO OqJg ToWl **.7t270Zt Toa *4.4 tAh(4t9

w S)-to m"um TO OAtS

12/ A 3.8 tonne per hour boiler with superheat and high steam pressuremight not be co_mercially available. A 4 to 6 tonne per hour unitmight be the only commercial option, hence the expense will be highfor the capacity needed.

- 78 -

5.54 This option would require the continued operation of theexisting diesel generators to provide the balance of power requirementsuntil connection of the mill to the grid. The boiler would consume up to19,000 tonnes of wood it 34 percent mcwb. Diesel generation initiallyand grid electricity eventually would be required for 4.61 million kWhper year. Synchronizing gear is included to allow parallel operation,although isolating systems must be evaluated as an alternative.

5.55 HIM Alternative 5A!5B. This 1.2 MW option is technically thesame as Alternative #3A, except that the grid extension is assumed to beundertaken and power can be fed in both -. ctions (i.e. MIM can feed thegrid surplus power during the non-peak 10 hours per day). The capitalcost for the steam plant will remain the same as in case 03A at US43.06million for the 1.2 MW condensing steam turbine-generator set, newboiler, building and appropriate auxiliaries. In this case, the singleturbine option is definitely the least cost solution and most efficientroute because the power plant is assumed to operate at a constant outputof 1.0 MW.

5.56 It can be assumed that the ne% wood-fired steam power systemcould supply up to 5.25 million kWh per year to satisfy most of thecomplex's base and intermediate load. The additional 617,000 kWh peryear needed due to scheduled and unscheduled power plant outages would beprovided by the grid along with an additional 833,000 kWh per yearrequired for peaking. The total of 1.45 million kWh covers peaking plusacLeduled and non-scheduled outages of the wood steam plant. Assumingthe power plant operates at full capacity (net output of 1.0 MW) 24 hoursper day, 7 days a week, 46 weeks per year, then the total output of theplant would be 7.73 GWh/yr. Thus, 2.48 GWh/yr will be available for saleback to the grid. Accounting for the 1.45 GWh/yr purchased from the gridfor peaking requirements and power plant outages results in a net supplyto the grid of 1.03 GWh/yr.

5.57 At a 34 percent mcwb, wood fuels in a steam plant of 1.2 MW canproduce between 350 and 385 kWh per tonne of fuel. Thus, to provide theadditional 2.48 million kWh to the grid will require an additional 7,100tonnes of wood (based on 350 kWh/t). As the plant will require up to33,000 tonnes for its ewn annual needs, the additional gent:ation forgrid feedback will increase the total annual fuel wood needs to about40,000 tonnes per year, which is slightly greater than the total woodresidues generated at the mill. Approximately 1,000 tonnes/yr of readilyavailable logging residues would have to be hauled to the plant fromforest areas about 100 miles away.

5.58 Costs. Estimated financial capital and annual operating costsare presented in Table 5.16 for the base case and each of the MIMcogeneration alternatives that were analyzed. Component-level costingsare contained in Annex 7. Capital requirements for the variouscogeneration alternatives range from US$1.6 to 5.2 million. The capitalcosts do not include costs of adding kiln or steaming vat capacity or anyother costs not directly related to the electricity and process heatgenerating plants.

Table 5.16: SUWSAY OF CAPITAL AM AMiAL OPERTIsB COSTSFOtR WIN COIENZMTION ALTERIUUTIVES (0000 US$)

OMSE CASE A ALTERNATIVE IAASf CAi- B ALTSATIVE 0A ALTERIATIVE 3A ALTERATIVE 4A ALTERNATIVE 3SK

Local Foreign Local Foreign Local foreign Local Foreign Local Foreign Local Foreign

Item costs Costs Totel Costs Costs Total Costs Costs Total Costs Costs Total Costs Costs Total Costs Costs Total

Capital Costso Construction 27.0 7.0 34.0 27.0 7.0 34.0 101.0 25.0 126.0 82.0 20.0 102.0 66.0 17.0 83.0 82.0 20.0 102.0

o Equipment - 385.0 365.0 -- 385.0 38.0 - 2420.0 2420.0 - 1425.0 1425.0 - 790.0 710.0 -- 1425.0 1425.0

o TranortChrge 8 38.0 37.4 75.5 38.0 37.4 75.5 238.9 234.7 473.6 1 .7 138.2 278.9 78.0 76.6 154.6 140.7 138.2 278.9

o Engilneering.festal lotion& Contingencies 155.0 264.0 419.0 155.0 264.0 419.0 830.4 1337.2 2167.6 484.3 76?.7 1252.0 212.4 419.5 631.9 484.3 767.7 t5.O

Total CapitalCosts 913.5 913.5 5187.2 3057.9 1659.5 3057.9

Differential frease Case 0.00 +4273.7 +2144.4 .746.0 +2144.4

Annual OperatingCosts

o Labor 12.2 1- 2.2 8.6 - 8.6 10.5 - 10.5 12.2 - 12.2 9.5 - 9.5 9.5 - 9.5

oPo b/ - - - 231.8 231.8 - - _- - 159.5 - 159.5 -35.6 -35.6

o 0 A W 75.3 304.2 c/ 379.5 15.0 - 15.0 87.2 18.0 105.2 66.1 107.7 173.8 25.0 - 25.0 53.0 53.0

o Constsbles - 461.3 d/ 461.3 -- 15.0 15.0 29.4 127.0 156.4 14.4 149.6 164.0 - 25.0 25.0 29.4 63.0 92.4

O Contingancles - - 85.3 - 27.1 - - 27.2 -- - 35.0 - - 21.9 - - 11.9

Total Annual

OperatIng Costs 938.3 297.5 299.3 385.0 240.9 131.2

Oltfewnttsl froDfe crt f - - - -- -- 439.0- -553.3 - - -697.4-- -807.1

D Includes International freight and lnurance local port cbrge and bank fee.

/ Includes costs for grid pr purchas.S/ Includes anal costs for reptceat of dil gerator capaity.

d/ includes annui cests for dIesol fuel.

- 80 -

5.59 Financial/Economic Evaluation of Mim Alternatives. Selectionof the most economically desirable NIM alternative may be dependent onwhether grid extension is considered a real resource cost or a sunkcost. In order to provide investment guidance relevant under bothassumptions, the following decision-making procedure is adopted:

(a) Evaluate all alternatives under the assumption that the gridextension has not been made and therefore is not a sunk cost.

(b) Choose the best alternative.

(c) If the alternative selected in (b) does not involve gridextension, then recommend this alternative.

(d) Otherwise, re-evaluate the options under the assumption thatthe grid extension is a sunk cost and recommend investment inthe best alternative.

5.60 Cost Minimization Under Assumption of No Grid Extension. Eachof the MIM alternatives is designed to provide the same level of serviceto the Mim areat (a) Meet the process heat demand for the steam vats,veneer dryer and 32 kilns to be in place within two years; and (b Meetthe electric power and energy demands of the upgraded sawmill facility,Scanstyle Ltd. and the surrounding community of 2.2 MW peak and 6.7GWh/yr. Since benefits have been equalized, a net present costminimization procedure was employed to rank alternatives for study ingreater detail. The net present value of each alternative, consideringonly the cost side, was computed as shown in Annex 7 using baselineassumptions over a 15 year project life at a 10 percent discount rate.Crid extension costs incurred in Alternatives 1A, 4A and 5A wereannualized over a 40 year lifetime. All financial NPV calculationsassume that the grid extension costs are not borne by MIM. Results ofthis comparision are shown in Table 5.17.

5.61 As expected, grid extension (Alternative 1A) is clearlypreferable to total diesel generation (Base Case A) from both financialand economic viewpoints. Given average industrial tariff levels of about3.5 US cents/kWh grid extension is also the least-cost among allalternatives consilered financially. However, the economic perspectiveis the more relevaat one for national planning purposes, especially sinceMIM is state-ownei. Alternative 3A, a residue-fired base/intermediateload unit with diesel peaking, is the least-cost option when economicpricing is applied. Alternative 2A is costly because of the poor loadfactor achieved by the peaking wood-fired turbogenerator, while thecomparison with Alternative 5A implies that diesels are a cheaper sourceof peaking power than the grid extension.

- 81 -

Table 5,17: I'T PRESENT OOST OF MIN ALTERNATIVESWITH NO GRID EXTENSION ASSUIIED

Not Present Cost('000 USS)

AIternati"e Financial Economic

Base Case A 8,051 7,318

IA 3,176 6,486

2A 7,463 7,387

3A 5,986 5,928

4A 3,492 6,471

SA 4,056 6,365

5.62 Alternative 3A Vs. 1A Marginal Cost Analysis. A marginal costanalysis was performed by subtracting the costs of Alternative IA from3A. The avoided grid extension and grid electricity costs then becomeeconomic benefits of the incremental investment in Alternative 3A overand above process heat requirements. Details of th- marginal analysisare found in Annex 7 and suanarized in Table 5.18.

Table 5.18: INI ALTERNATIVE 3A VS. IAFINANCIAL AND ECONOMIC ANALYSIS RESULTS

FInancial Economic

Not Present Value $(2,965,760) S293,180

IRR Undef. 12S

Discounted Payback Time 12 years

CapItal Oost Switching Value > -100% *14%

Draskeven Grid Electricity Price US 9.3 cents/Wh US 4.6 cents/kWh

5.63 Interpretation of the results is straightforward except for thecase of the breakeven values. They represent the marginal cost ofelectricity generated in Alternative 3A when avoided grid extension costs

- 82 -

are taken into account. In the financial case, the tariff charged at MIMfor grid electricity would have to rise from the eCC tariff figure of US3.5 cents/kWh to US 9.3 cents/hih for KIM to realize a savings fromAlternative 3A. MIM does not "see" the grid extension costs of US 3.6cents/kWh. In the economic evaluation, the effective cost of electricityat KIM is US 4.6 cents/kWh, cheaper than the estimated grid hydropowergeneration cost of US 5.2 cents/kWh. However, the investment of US$2.14million to obtain this savings yields a economic rate of return of only12 percent. This return is not sufficient to compensate for the risk andadditional management burden of the wood/diesel hybrid. Grid extensionis therefore recommended.

5.64 The effect of demand growth in the Mim area has not beenquantified in this simplified analysis. It is expected that growth inenergy dem&ad in the short run will be for nighttime community uses. Theincreased energy consumption could be accomodated at very low marginalcosts simply by burning more residues in the cogeneration system duringoff-peak hours. In the longer term, however, it can be expected that (a)on-site residue availability will become a constraint as totalelectricity consumption increases to 20 percent above the 6.7 GWh/yrfigure used in the baseline analysis, and (b) growth in demand will occurduring the on-peak hours when the sawmill is operating which nust be metby high-cost diesel generation. At the same time, demand growth lowersthe unit costs of the grid extension. The long term picture thussupports the grid extension recommendation.

5.65 Cost Minimization Under Assumption of Grid Extension. If thegrid is considered to be extended (i.e. a sunk cost) Alternatives 2 and 3can be immediately dismissed as peaking power requirements can be mosteconomically handled by the grid. On the basis of the cost minimizationresults in paragraph 5.60, Alternatives 4 and 5 are comparable ineconomic attractiveness. Alternative 5 was selected for closerexamination as it offers the potential for greater rtsidue utilizationand on-site electricity production.

5.66 Alternative 5B vs. Base Case B harginal Cost Analysis. Amarginal cost analysis was performed by subtracting the costs of BaseCase B from Alternative 5B. Costs include the opportunity values of thecharcoal and firewood being consumed in the cogeneration system. Inaddition, for maximum power production approximately 1,000 tonnes/yr oflogging residues would have to be hauled from the forest at a cost ofabout US$10/tonne. Benefits are the avoided costs of grid electricityplus amounts sold back to the grid due to on-site generation. 13/Details of the marginal analysis are found in Annex 7 and summarized inTable 5.19.

13/ Electricity sold to the grid is assumed to have the same financialand economic value as that purchased from the grid.

- 83 -

5.67 The resuits show that the economic cost of the electricityproduced is approximately the same or slightly higher than hydrogenerated power. An optimized systam in which residue consumptionexactly matched on-site residue availability could shave 10-20 mills offthese costs, an insignificant quantity at this level of investmentconsideration. Financial incentives for investment by Mim would notexist unless the cost of the marginal investment were to drop by 49percent or the electricity tariff to MIM were raised by 60 percent.

Table 5.19: MIN ALTERNATIVE 5B VS. BASE CASE B

FilNANCIAI AND ECONOMIC ANALYSIS RESULTS

Financial Econ=mIc

Net Present Value S(139,310) S(24,270)

lRR os 8S

Discounted Payback Time

Capital Cost Switching Value -49t -9S

Marginal Cost of US 5.6 cents/kwh US 5.8 cents/Kwh

Electricity Generated

Residue Handling and Combustion Efficiency Improvements

Background

5.68 The Mission noted a number of low-cost improvements to residuehandling, storage and utilisation systems which, if implemented, couldsignificantly increase the energy value of the residues. In particular,sawdust is greatly underutilised as an on-site energy fuel, accountingfor only 7 percent of mill process heat-raising due to introducedmoisture and inability to properly combust in commonly found mill boilerfurnaces. Substitution of surplus sawdust for solid residues in millboilers would free up slabs, edgings and offcuts for alternative energyand non-energy uses, creating additional value. Sufficient suXplussawdust exists to potentially replace virtually all of the 58,900 a' SUEof solid residues burned for mill process heat/cogeneration.

5.69 While costs for these improvements are comparatively small, themotivation for their implementation will be primarily felt at mills whereincreasing process heat or cogeneration needs result in a tightening inthe total residue supply/demand balance. Detailed costings andcost/benefit comparisons have therefore not been carried out for the

- 84 -

individual improvements. Rather, their costs have been factored into theoverall investment requirements of the various process heat generationand cogeneration schemes reviewed above, and in certain of the off-sitesubstitution options evaluated in Chapter 6.

Saw Guide and Sawdust Storage Improvements

5.70 Dramatic improvements to saw guides over the last fifteen yearshave permitted many benefits to accrue to the sawmilling industry. Theseinclude reductions in saw kerf, improved sawing accuracy and improvedsmoothness of sawn surfaces, all of which lead to increased lumberrecovery. Spin-off advantages of the above are reductions in sawdustvolumes generated, and the production of finer and drier sawdust. Thesespin-offs ease sawdust conveying problems and improve its quality as afuel or as a feedstock for briquetting plants.

5.71 Low Volume Water Application. Reductions in the amount ofwater utilized to lubricate saws with existing saw guides can be obtainedthrough simple measures to control the volumes and to apply it to thesaws efficiently. The simplest and least costly method of the above isto introduce the saw water directly and accurately onto the saw via spraynozzles and to turn off this water when the blade is not in the cut.Depending on water quality it may be necessary to introduce filters toclean the water supplied so that it does not plug the nozzles. Suchsprays could be added for less than US$50 per bandmill or circularheadrig installation (or about US$100 if filtration is necessary).

5.72 The best method to reduce water consumption is to introduce itinto a milled groove located within the saw guide. This method depositsa film of water directly on the saw blade which is to be preferred tospraying or splashing it onto the blade which allows most of the water tobounce off the surface. The costs to convert a 3,300 m bandmill toquick-change guides would be approximately US$2,700 for complete guideassemblies and would require two man-days local labor to perform thefield installation. A guide resurfacing machizc costs about US$700. Anexample of such guides is shown in Annex 8.

5.73 Also included in Annex 8 are sketches from U.S. Patent3,623,520 for circular and bandsaw guide apparatus which clearly show themethodology of introducing water to the guides. This information isuseful to anyone considering converting his own equipment. Typicalselling price for guide arms for circular edgers for sawing wood up to300 m thick is about US$370 each. Two guide arms are required persaw. Maintenance equipment consisting of a babbit melting pot(electric), moulds and a guide surfacing machine with jigs would costabout US$1,100. It is estimated that this equipment could be produced inGhana at 25 percent of the U.S. costs.

5.74 ADO Application. Alternatively, diesel oil (ADO) can beapplied to blade surfaces using a felt oil applicator. One suchinstallation observed at Western Timbers Ltd. utilized locally devised

- 85 -

bandsaw guides made of Ekki wood, a very dense secondary specie. ADOconsumption was reported as 0.5 Igal/day. Hlowever, ADO application maystain certain whitewoods.

5.75 Sawdust Storage Improvements, Of course, sawguide improvementsare useless as an energy enhancing measure if excessive water is re-introduced into sawdust through exposure to the elements. The obviousremedy is to construct i,mple shalters to prevent rainwater from soakingthe residue&.

Furnace Modifications for Sawdust Combustion

5.76 Un-wetted, fine-grained sawdust is a readily combustible fuelprovided the proper type of furnace grate and combustion air circulationsystems are employed. As most sawmill furnaces currently found in Ghanaare primarily designed to burn solid fuels, modifications are requiredfor sawdust substitution. Both the extent of the required retrofits andspecific and depend on the installed equipment base. Generalrequirements comprise:

(a) Installation of pin-hole (in small furnaces) orinclined/stepped grates for proper combustion support. Use ofsolid fuel grates, such as bar or large-hole types, result in alarge portion of the sawdust falling unburned through thegrate.

(b) Increase in forced/induced draft airflow to provide for sawdustsuspension. In some cases, an increase in fan size/horsepowerwill be sufficient. Other installations will require re-working of the ducting and air circulation.

Given relatively low labor cc-ts, manual systems for sawdust feed intothe furnace are appropriate for all but the largest installations.

Boiler/Furnace Efficiency Improvements

5.77 Measures to improve overall efficiency of wood combustion arealso very site-specific, however, several generic suggestions arewarranted. For any firewood burning installation, the greatest problemis overstoking of the furnace so that the boiler attendants do not haveto add wood so often. This results in great fuel/air imbalances suchthat incomplete combustion produces many gaseous unburned hydrocarbonproducts (with significant calorific value) that are blown ou. the stackand lost. The remedy lies in both operator training and proper boilerattendant supervision. Efficiencies can be doubled, and sometimestripled, by learning and implementing proper stoking techniques.

5.78 The second major correctable problem is excess air infiltrationthrough missing fire doors, doors that are installed but left open,broken or missing air control dampers, and missing gaskets, unpatchedcracks, etc. These are all easily correctable and produce at least 25-30percent savings with minimal investment.

- 86 -

VI. POTSWTIAL OFF-SITS FLURIYTIVRS FOR INPROVING AND/OtIUGRISIUnG TuB u OEF OVOD IUDUSTRY RESIDUES AS FMUL

Summary

6.1 Off-site options for utilization of residues were extensivelyevaluated by the Nissioj. Of the total mill residues produced in 1986,29 percent or 124,000 m /yr was used off-site for firewood or to producecharcoal and about 2,800 m of sawdust was processed into fuelbriquettes. At present, most of the solid residues are utilized for foodpreperation in bakeries, chop bars and households while the sawdustbriquettes are used primarily by co_mercial bakeries and more recently inbricknaking kilns. As was pointed out in Chapter 3, sawdust is the onlyresidue in significant surplus for increasing off-site residueutilization. The primary technical option for off-site directutilization is modification of existing oil-fired or wood-firedindustrial/co nercial combustion systems to utilize the surplus sawdust.

6.2 At present, the majority of off-site residue utilization isaccomplished only after conversion primarily of the solid residues tocharcoal and, more recently, small quantities of sawdust to briquettes.Within this context, three possible technical options exist to improveand/or increase the indirect off-site use of mill residues:

(a) Improved charcoal production techniques;

(b) Increased sawdust briquetting capacity; and

(c) Introduction of briquette carbonization techniques.

6.3 A sumary matrix of the technically feasible options forimproving and/or increasing the off-site use of wood residues and theirpotential impacts is presented in Table 6.1. The sawdust briquettingoption technically has the best potential for increasing the use ofpresently surplus residues. Improved charcoal making could increase thesupply of residue-derived charcoal ',)y over 80 percent. Very littlepotential exists for increasing the direct utilization of surplus sawdustresidues primarily because of technical and economic constraints. Briefdescriptions of the key technically feasible options are presented intnis chapter along with schematic diagrams, equipment and operating costestimates and financial/economic analyses. The alternatives aredeveloped based on actual sites and conditions observed during the fieldwork rather than on generic plants based on average conditions inGhana. The selection of actual sites makes the analysis potentiallyuseful. However, the results must be carefully evaluated if theconclusions are to be generalized across the sector.

Table 6.1: MATRIX OF TECHNICAL OPtIONS FOR IWtIN6 AjRINZASINB OFF-SITE RESIWE UTILIZATION

Impact on ResiduesOttions Prlmary Tara.ts Utilizatlon Requirement anef Its

Direct Utilization

Conversion of Commercial/industrial Increased off-site demand Technical assistance and Savings from costs of Importedexisting ol1-fired boilers In the vicinity for unprocessed residues, capital for conversion of fuels or high cost solid wood-Industrlal/comercIai of the major wood especially sawdust, oil-fired combustion fuels,boilers, processing centers, equipment.

Conversion

Improved charcoal Independent charcoal Increased charcoal supply Capital for Improved Increased efficiency andproduction, makers using earth/ from converted solid charcoal making system, productIvity ia charcoal making,

sawdust mound cher- residues, technical assistance Improved and more consistentcoaling technique, and training, charcoal quallty, increased

revenues to charcoal makers.

Sawdust briquetting Production from sur- Dmand for sawdust which Capital for construction Beneficial use for presentlyplus sawdust available Is presently only residue and operation of briquet- discarded restaues. Revenuesat the wood processing In abundant surplus. ting factorles and to briquette producers andcenters adjacent to assistance for expansion reduced disposal costsIndustrial and urban of briquette market. to mills. Potential forcenters, prlmartly economic use of residuesKumasi, Takoradi and in distant markets. IncreasedAkim Oda. combustion efficiencyf

when compared to wet sawdust.

Briquette Product) n from sur- Demand for sawdust which Same as above Including Same rs above Including highercarbonization plus sawdust to Is presently only residue additional capital for value for charcoal.

provide an additional In abundant supply. construction and operat-supply source of char- Ion of a briquettecoal for the domestic carbonization process,sector.

- 88 -

Off-Site Direct Utilization

BackRround

6.4 Off-site opportunities for direct residue utilization wereconsidered by the Mission in the industrial, commerical and householdsectors of the economy. Of the total residues utilized off-site, about30 percent is used directly as firewood primarily for food preparation inthe commercial and household sectors. Hardly any mill residues arepresently utilized directly in the industrial sector for process heat.The Mission visited twenty non-wood processing industries in the Kumasi,Accra/Tema, Takoradi, Prestea and Cape Coast areas covering food,beverage and agricultural product processing, brick and tilemanufacturing, paper recycling, textiles, golo refining, andinstitutional services. These industries can be divided in to two maincategories according to their present fuel use: oil consumers andfuelwood consumers. A list of the industries and institutions visited andthe type of fuel they presently use is shown in Table 6.2. Detailedinformation on their pattern of fuel useage is contained in Annex 9.

Substitution of Sawdust for Oil Fuels Consumption

6.5 Industries that presently utilize oil fuels for steamgeneration or direct heat are generally not readily adaptable to directfiring of sawdust or solid residues. In all cases examined, the costsfor modifications and additional equipment required would not justify thesavings realized from substituting for petroleum fuels, primarily due tothe small scale, low utilization factors and type of equipment being usedin Chanian industries. In addition, the current low financial price ofUS$101 per tonne of fuel oil and economic price of US$76 per tonne(Accra/Tema FOB export value) makes competition with oil firingdifficult. Engineering designs and costings for conversion at threesites are discussed in Annex 10.

Financial and Economic Analysis

6.6 Detailed financial and economic analyses may be found in Annex10. None of the fuel oil substitution projects studied were viable fromeither analysis viewpoint. Results of DCP analyses at three sites aresummarized in Tables 6.3 and 6.4. Calculated breakeven values indicatethat oil prices would have to rise above US$30/barrel (in 1986 dollars)for these peojects to show positive returns.

- 89 -

Table 6.2: INiSTRIES AND INSTITUTIONS VISITED TO EVALUATEPOTENTIAL FOR DIRECT UTILIZATION OF WOOD RESIDUE

Facility Location Fuel Used

Kumasi Breweries Ltd. Kumasi RFO

Gulness Ghana Ltd. Kumasi Elec,ORF0

Komfo Anokye Teaching Hospital AL

Appiah Menka Complex Ltd, Kumasi Wood

Ashanti Ol Mills Ltd. Kuasi Wood/Palm Waste

Cocoa Processing Company Ltd. (WAN) Takoradi IFO

Cocoa Processing Company Ltd. (CPC) Takoradi ADO

GI0OC Paper Conversion Co. Ltd. Takoradi Paper Waste/ADO

cermic Coordetro Ghana Ltd. Taiorad a/

Prestea Goldfields Ltd. Prestea Wood

Lever Brothers Ghana Ltd. Teme FO

Food Specialities Ghana Ltd. Toem IBO

Ghana Toxtiles Manufacturing Co. Tmeo IFO

Tome Textiles Ltd. Tema IFO

GIHOC Brick & Tile Ltd. Acca iDOhlood

Ankaful Brick Ltd. Cape Coast Wood/Briquettes

oa Plant under constructIon. Information on fuel use could not be obtained.

- 90 -

Table 6.3: FINANCIAL ANALYSIS RESULTS, OIL TO SAWDUST-FIRED BOILER CoNVERSION

kPV I RR coapital CostSwitching Value

Site (S) (5) (5)

Kumsl Breweries Ltd. (1,023,120) Undef. -62

Cocos Processing Co. Ltd. (WAN) (432,890) Undef. -41

Lever Bros. Ghana Ltd. (2,297,340) Undef. -98

Source: Annex 10.

Table 6.4 ECONOMIC ANALYSIS RESULTS, OIL TO SAWDUST-FIRED BOILER CONVERSION

Capital Costsite NPV IRR Switching Value

(S) (S) (5)

Kumasl BrewerIes Ltd. (1,143,000) Undof. -72

Cocoo Processing Co. Ltd. (VAN) (491,870) Undef. -44

Lever Bros. Ghana Ltd. (3,628,330) Undef. -100

Source: Annex 10.

Substitution of Sawdust for Puelwood Consumption

6.7 Industries and comercial operations that currently use forest-dervied fuelwood could utilize the solid residues from the woodprocessing industries if these were available. However, most are unableto use the sawdust directly which is the residue that is primarilyavailable in surplus. Many of these candidates, especially the bakeries,brick factories and fish smokers, could easily convert to usingsawdust briquettes. In fact, most would prefer the briquettes overfuelvood. A discussion of this aspect is presented later in thischapter.

6.8 Many of the wood consuming industries visited had small Lambiontype furnaces which could be converted, with some minor modifications, todirect firing of sawdust. However, the scale of plant operations in mostcases was such that potential wood savings were generally not sufficient

- 91 -

to recover the capital investment required to install the requiredfurnace modifications and sawdust handling bUd storage systems. Adiscussion of the technical and economic parame;ers considered at some ofthe sites visited is presented in Annex 11.

Financial and Economic Analysis

6.9 Financial and economic analysis details for some of the morefavorable sites are shown in Annex 11 and summarized in Tables 6.5 and6.6. The Appiah Menkah Complex project is marginal from the investor'sstandpoint, while the conversion of the Prestea Goldfields Ltd. oreroaster is a feasible, though small, undertaking. Neither entity facesthe long-run cost of fuelwood production, and therefore the rates ofreturn to these conversions are boosted by about 20 percentage pointswhen economic pricing is applied.

Table 6,5: FINANCIAL ANALYSIS RESULTS, FIJELWOOO TO SAWDUST-FIlEDBOILER COWNERSION

CapItal CostSite NPV IRR Switching Value

(5) (N) (5)

Applah-Nsnkah complex Ltd. 5,730 11 +5

Presteos oldflelds Ltd. 20,470 23 --

Source: Annex 11.

Table 6.6.: ECONOMIC ANALYSIS RESULTS, FUELNOOD TO SAWDUST-FIREDBOILER CONVERSION

Capital CostSIte NPV IRR Switching Value

(5) (N) (S)

Applah-Minkah Complex Ltd. 90,370 32 +92

Prestea Goldflelds LSt. 54,990 42 -

So_rce; Annex 11.

- 92 -

Off-Site Conversion Alternatives

BackRround

6.10 A large portion of the the solid wood residues that are usedoff-site for energy is first converted to charcoal while all of thesawdust used for energy off-site is first converted to solidbriquettes. Charcoal conversion is accomplished primarily in traditional"earth mound" kilns operated by independent charcoal makers workingadjacent to or within a couple of kilometers of the mills that supply theresidue. The conversion erf.ciency of most of these charcoalingoperations was found to be cxtroaely low, in the range of 11 to 20percent (charcoal yield on a dry weight basis). The average conversionefficiency around the Kumasi area is estimated at approximately 16percent, based on estimates of solid residues consumed and charcoalproduced. Charcoal making techniques are available that could inureasethe conversion efficiency to 25 to 30 percent resulting in a neardoubling of the charcoal supplied from wood residuets.

6 11 The potential supply of surplus sawdust is estimated at 83,400;I SWE or 65,500 tonnes. Over 60 percent of this surplus is located inthe Kumasi area. As was indicated earlier in this chapter, the use ofunprocessed sawdust in the industrial sector is limited due to economicconstraints. At present only about 2,800 m3 of sawdust is used off-sitefor energy, and the sawdust must first be briquetted prior to acceptanceby consumers. There is presently only one briquetting plant in Ghana,located in Oda, and it is owned and operated by a private entrepreneur.This factory has a maximum capacity of 2,200 t/yr but presently producesat the rate of 1,100 t/yr. It is unable to meet the total demand forbriquettes from the bakers and brickmakers alone which is estimated at45,000 tJyr or equivalent to over 100 percent of the surplus sawdust.

6.12 Given the above, two technical options immediately presentthemselves for further evaluation: improved charcoal making techniques,and increased sawdust briquetting capacity. In addition, the largedemand and relatively high price paid for charcoal makes the possibilityof converting sawdust to charcoal briquettes also potentiallyinteresting. In fact, the possibility is presently being considered bythe owner of the Oda briquetting plant. This section brieflyinvestigates these three options outlining the technical possibilities,associated investment, operating costs and financial and economicreturns.

- 93 -

Improved Charcoal Production

Present Method

6.13 Earth Mound. Almost all charcoal produced from -sawmillresidues in Ghana is produced by the traditional "earth mound" methodalthough in most cases the top covering is actually sawdust. The mainadvantage of the mound system is that it requires no capitalinvestment. The primary disadvantages are that earth mound kilns arehard to control, are susceptible to the elements and have generally verylow yields with unpredictable charcoal quality.

6.14 Slabs, edgings and, in some cases, bouls are obtained fromsawmills by independent charcoal makers who purchase the residues by the"trailer" or "tractor" load from middlemen. The size, shape andcharacteristics (moisture content, density, percent bark) of the millresidues vary considerably. In most cases, the charcoal makers simplyhorizontally stack the residues in an appropriately sized area whichmight be slightly depressed. Vo cutting, pre-drying or sorting of theresidues is practiced and species are mixed indiscriminately. The moundsare generally rectangular in shape with the initial foundations of thestack built so as to permit some air flow under the stack. Thereafter,the wood is stacked as tightly as possible with the top of the moundsloped slightly, about 10 degrees. Once complete, the mound of stackedwood is then covered with dirt and sawdust. In some cases, greenvegetation (normally grass), if availatle, is first placed over the stackprior to covering with the dirt and sawdust. The sawdust is alsoobtained from the nearby sawmills and is used because it is light, easyto shovel and is obtained for a small transport cost. However, thesawdust is not a good covering material because it is porous and allowsair to infiltrate the mound causing local flare-ups which not only reducecharcoal production but could destroy the entire mound if unchecked.These flare-ups are extinguished either by covering the area with freshsawdust or by dousing it with water and then covering it with sawdust. Asketch of a typical charcoal mound observed in Ghana is presented inFigure 6.1.

6.15 Total time per cycle of charcoal production varied from twoweeks to a month for some of the larger mounds. In the case of the largermounds, harvesting of charcoal begins as early as 10 days after ignitionof the mound and continues for three more weeks. In such cases, theyield and quality of the charcoal vary ccasiderably with the progress oftime. In all cases, the charcoal makers had to live at the site of theircharcoal making operations. Most operations were family run with womenand children involved in the various operations.

- 94 -

Figure 6.1s TYPICAL MOUND CHARCOAL PRODUCTION

When dirt is used forthe covering, then a matof reeds are often usedto keep the charcoal clean. Exhaust Vents

Pile of wood

Inlet Vents *.''m ~~~~~~Dirt or SawdustInlet Vents > X t ^ Covering

- Fire StartingChannel

- 95 -

6.16 Observations from the production of residue charcoal at threetypical sites in the Kumasi area is presented in Table 6.7. Site #1represen s a small operation run primarily by women who operate only onemound at a time. Site #2 is a more typical small family operation withtwo to three mounds in varying stages of operation. Site #3 represents arelatively large operator with stockpiles of wood residues being airdried prior to carbonization. This type of operator is an exceptionrather than the rule. Not surprisingly, site #3 has the highest charcoalyield per unit of wood input and the highest return to labor. Site #2,which is more typical, has a charcoal yield of 16 percent with a laborproductivity of 12.6 t/man-yr. Site #1 was strictly a "hand-to-mouth"operation with the revenues of the previous charcoal operation requiredto finance the delivery of the next batch of slabs and edgings, whichwhen delivered were stacked for charcoaling thus reducing the time forair drying. In this case the charcoal yield was observed to be 11 percenton a dry basis. The technical characteristics of the charcoal producedat each of these sites is presented in Annex 12.

Table 6.7: EARTH "FNM CHARCOAL KILNS, KUMASI

Observation Site #1 Site _ 2 S;te #3

Input wood (m3) 10 46 58

Cost of wood Input (Cedi) 4,500 22,000 21,300

Output charcoal (tonnes) 0.6 3.5 6.1

Yield, dry weight basis (S) a/ 11 16 20

Charcoal sales price (Cedilkg) 8.3 11.4 6.7

Gross sales revenue (Cedi) 7,500 40,000 40,300

Gross profit (Cedi) 3,000 18,000 19,000

Labor Input (man-days) 30 100 60

Return to labor (CediMan-day) 100 180 317

Labor productivity (t/man-year) 7.2 12.6 36.6

a/ Yields calculated based on 50S Wawa/50 redwood Input wood mIxhaving an average dry density of 510 kg/m3.

Source: Mission estimates.

- 96 -

6.17 Production Costs. Total financial production costs at atypical earth mound operation were estimated at 8.6 Cedis/kg (US$57.61)of charcoal produced. A yield of 16 percent (dry weight basis) and laborproductivity of 17 tonnes of charcoal per man-year were used indetermining typical production costs. The resulting estimates andunderlying assumptions used to derive the financial and economicproduction costs are presented in Annex 13.

6.18 Marketing. Prices received by the charcoal makers varieddepending upon the quality of the charcoal and the location relative tothe retail markets in town. A typical marketing pattern is detailed inTable 6.8. Charcoal is wholesaled in sacks weighing between 35 to 40kg/sack. Final consumers apparently measure purchase quantities involumetric terms, as the less dense residue-derived charcoal sells for a12 percent premium on a weight basis over the heavier forest-derivedcharcoal from the high forest/savannah transition zone.

Table 6,8: MILL RESIDUE CHARCOAL MARKETING, KUMASI

Ced IAkg

wholesale price 8.6

Transport to market 1.1

Retailer's margin 1.7

Retail price 11.4

Source: Mission estimates,

Improved Methods

6.19 Options. The quality of the charcoal required for domesticconsumption is generally not extremely high. Volatile contents in therange of 20-30 percent are desirable in order to allow easy ignition andsome flammability (Earthscan). Given this criteria, carbonizationtemperatures in the range of only 300 to 400 °C are required withresulting theoretical charcoal yields of 30-50 percent (dry weight basis)in laboratoty retorts. Actual yields in field operated kilns would be inthe range of 20-30 percent.

6.20 Several improved charcoaling methods with resulting higheryields are available and are in use throughout the world. All of theimproved methods have one factor in common in that they require somecapital investment in equipment and some training of operators to utilizethe kilns efficiently. In addition, some kilns require significantprepration of raw wood such as cutting due to size limitation and dryingto mazimize yields. A brief evaluation of some of the most promisingimproved charcoeling methods, within the context of residues from Ghanasawmills, is presented in this section. The improved methods consideredare:

- 97 -

(a) TDRIt"Ghana Mini" metal kilns;

(b) Subri clay-metal (Posse) kiln;

(c) Missouri kiln;

(d) Casamance kiln; and

(e) Beehive brick kiln.

The first three options have been operated experimentally in Ghana anddata on their costs and operating characteristics were obtained foranalysis. The last two options are charcoaling techniques that have beengaining widespread use in other developing countries but have not beenintroduced in Ghana to date. A brief discussion of each,of these optionsis presented below along with associated capital and operating costs.

6.21 TDRI/Chana Mini Metal Kilns. The typical TDRI metal kilnconsists of two interlocking cylindrical sections and a conical cover.The cover is provided with four equally spaced steam release ports whichmay be closed off with plugs as required. The kiln is supported on eightair inlet/outlet channels, arranged radially around the base. Duringcharring, four smoke stacks are fitted onto alternate air channels.Total volume of the kiln is approximately 7 m3. Wood used in the kilnshould be air dried for a minimum of 3 weeks and sized from 450 to 600 mmlong and 200 mm in diameter. Wood with a length or diameter greater thanspecified should be split before use (TDRI). A sketch of the TDRI metalkiln is shown in Figure 6.2.

6.22 The Ghana Mini very closely resembles the TDRI metal kiln withthe exception that it has only one cylindrical section and is slightlysmaller in total volume. Both the TDRI and Ghana Mini have beendemonstrated in Ghana. A TDRI kiln is being operated using sawmillresidues at a sawmill in Kumasi while several Ghana Minis are inoperation at the Subri River Project where forest residues are beingcarbonized.

6.23 The total cycle time for loading, carbonization and unloadingof the metal kilns is approximately three days. Labor requirements areestimated at one person per kiln. Charcoal yield is estimated at 27percent on a dry weight basis. A summary of the characteristics of themetal kiln in comparison to other improved kilns is presented in Table6.9. The annualized capital and operating costs for converting in metalkilns all of the mill residues presently converted to charcoal in theKumasi area is sumnarized in Table 6.10. Details of these estimates andunderlying assumptions are given in Annex 13.

- 98 -

Figure 6.2: TROPICAL DEVHWPMN RESEARCH INSTIUTE (TDRI) (U.K.)ST8 CHARCOAL KILM

_ ~~~~~~~~~~2250mm O._

~~ > _ _ > > ~~~~3 pieces N each "Ring"ican be rolledthrough 11! the Foreqsfiro

"IN . 4-Vents &Vent Lids

4 - Vents

4 - stacks (are slid over4 of the vents, then after1 1/2 days are lifted offand slid on the other 4 - vents.)

9 99 -

Table 6.9: COWARISON OF CHARCOAL KILNS

Weight Cycle Capital Cost EstimatedVolume Rocovery a/ Time Local Imported Total Life

No, Type Kilon 3 of space 5 (days) (Codi) (SU.S.) (5U,S,) (years)

1 Hound 15 to 200 16 7-30 - - Nil N/A

2 TORI 7,2 27 3 90,000 850 1,450 S

3 Ghana Mini 6.8 27 3 82,000 740 1,280 5

4 MIssouri 220 18 b/ 21 3,607,000 9,220 33,260 530/

5 Clay/Metal 11,2 20 5 43,000 550 840 3

6 Casamance 30-100 25 4-7 - 50 S0 2

7 Brick (Beehive) 5b 30 7-8 280,000 300 2,170 3

a Weight of charcoal divided by the oven dry weight of wood used.b/ Actual performance averae of 10 fIrlngs; kIln doors were warped.et Idea I recovey.

source: mIssion estimates.

- 100 -

Table 6.10: SUM6ARY OF ANNUALIZED CAPITAL AND OPERATING COSTSOF CIURCOAL PRt2OCTION ALTERNATIVES

SUmRIALTERNATIVE MOUND TDRI METAL CLAY-METAL CASAMANCE BRICK SEEHIVE

FIn Econ Fin Econ Fin Econ Fin Econ Fin Econ

Capital (Sx103 ) - - 53.9 51.3 45.4 43.6 4.0 3.7 43.7 39.7

Wood Input a/ 196.6 417.8 196.6 417.8 196.6 417.8 196.6 417.8 196.6 417.8(Sxl03)

Labor (SM103) 114.5 100.7 50.8 44.7 73.4 64.6 81.2 71,4 32.4 30.3

Malntenance - - 33.8 30.7 6.8 6.3 - - 9.8 8.6

(SX105)

Total (Sx103 ) 311,1 518.5 335.1 544.5 322.2 532.3 281.8 492.9 282.5 496.4

CharcoalProduction b/(Tonnes/yr) 5,400 8,850 6,400 8,200 9,800

Unit Costs(5/Tonne) 57.6 96.0 37.9 61.5 50.3 83.2 34.4 60.1 28.8 50.7

Cedis/kg 8.6 5.7 7.5 5.2 4.3

a/ All alternatives are assumed to convert 64,000 m3 SWE/yr of wood. Wood costs are estimated at 450Cedis/tonne at the s"wmiil gate. The cost of wood delivered to the charcoalers averages at 600CedIs/tonne.

b/ Charcoal production varies as a function of conversion efficiency.

6.24 Subri Clay-I4etal Kiln. The Subri clay/metAl kiln is basicallyan improved pit kiln with metal pipes for ducting and a steel roof forcovering of the pit. The walls of the pit are made of pounded clay tohelp reinforce them. The dimensions of the kiln vary but are generallyrectangular with pit depth of no more than 2 meters. The primaryadvantage of the clay/metal kiln is that it is less costly than the TDRItype kiln. It is considered a semi-mobile kiln requiring only thetransport of the metal parts and the digging of a new pit each time thecharcoaling operation is to be moved. A sketch of a typical Subri clay-metal kiln is presented in Figure 6.3.

- 101 -

Figure 6.3: CLAY/METAL CHARCOAL KILN (SUBRI SEMI-MOBILE KILN)

Clay & Dirt Packed inthe ends before firing.

/ Firing Window

/------_4 Steel sheets2thk x 1200

00 ~x 2400 Ig.

Wooden Pegs

3 Ducts/side Approx.130 x 200 size

Clay sealingbetween steelsheets

1800

Approx.

- 102 -

6.25 The Subri clay-metal kiln can be dimensioned to take longpieces of wood unlike the TDRI circular kiln. The pit type kiln iseasier to load, seal and manage than the mound type. Its performance isless vulnerable to the elements and it does not require the same vigilantattention as a mound operation. However, the pit kiln requiresconsiderable initial excavation and is only justified when the kiln is tobe repeatedly used. The walls of the pit are susceptible to collapsepresenting a hazard to those that have to work in the pit. The wallsalso absorb a lot of the condensates and leach them back especially ifthere is heavy rainfall which can also cause an accumulation of water inthe pit.

6.26 The pit type kiln tends to produce a relatively higher portionof partially carbonized pieces of wood. It takes approximately five daysfor a complete carbonization cycle primarily due to the longer coolingperiod required. The high insulation properties of this type of kilnallow for higher carbonization temperatures and therefore "higherquality" charcoal (i.e. higher carbon and lower volatile content). Theaverage efficiency of the Subri clay-metal kilns operated in Ghana werearound 20 percent chercoal yield on a dry basis. A comparison of the keycharacteristics of this kiln with other improved kilnas considered ispresented in Table 6.9. A summnary of the annualized capital andoperating costs for converting in clay-metal kilns all of the millresidues presently converted to charcoal in the Kumasi area is presentedin Table 6.10. Details of these estimates and underlying assumptions aregiven in Annex 13.

6.27 Missouri Kiln. Tse Missouri kiln is a high volume concretekiln with steel doors. The kiln was developed in the United States forlarge charcoal making operations. It is a permanent structure requiring alarge capital investment. The kiln is generally rectangular in shapewith a vaulted roof and can vary greatly in size with larger onesreaching 12 m long, 7 m wide and 4 m high. The kiln usually has fourchimneys on either side that are made of metal. The primary advantage ofthe kiln is its scale which allows the use of mechanical loading andunloading therby reducing labor costs. It is also designed to operate incold weather and to withstand the extreme differences in temperaturebetween the inner and outer walls of the kiln. If operated properly, thekiln can achieve yields of 30 percent or more. A sketch of a Missourikiln is shown in Figure 6.4.

6.28 A Missouri kiln was constructed and operated at the SubriRiver project in Ghana. However, the kiln is not in operation nowbecause of several problems experienced due to the warping of the steeldoors and cracking of the concrete walls and roof. While the Missourikiln has a potentially high yield it requires a 3 weeks cycle time forcharcoal production, a high level of mechanization and careful processcontrol. In addition, the Missouri kiln is highly capital intensive. Acomparison of the key characteristics of this kiln is presented in Table6.9. Production cost for charcoal were not determined for this type ofkiln because it was judged to be technically inappropriate forapplication in Ghana.

- 103 -

Figure 6.4: MISSO0UI CHARCOAL KILN (SUBBI RIVER PROJECT)

t, C; Steel Stacks

:h ~~~~~~~Reinforced0. ( g @ X Concrete Walls

0 ~~~~~~~~~~~~~~~& Roof

Steel Doors

aw d~~~~~~~~~~

- 104 -

6.29 Casamance Kiln. The Casamauce kiln is an improved type ofearth mound kiln developed in Senegal. The kiln is generally circular inshape and has a capacity of 30 - 100 m3 with a diameter of 6 - 10 m. Itresembles the Sweedish mound kiln but the wood is stacked horizontallyrather than vertically. The kiln is built up of radially placed logs.The initial layer has substantial gaps between logs to allow circulationof air. The sucessive layers are built around a central vertical post tocreate a dome shape. A circumferential ventilation chamber is createdaround the outside of of the kiln at ground level by leaning short sticksagainst the stack. The pile must then be covered with a layer ofvegetation and earth. A lighting hole in the center is created byremoving the center post. A particular feature of the Casamance kiln isthat it can incorporate a chimney attached to the side of the kiln tocondense the pyrolitic tars produced during the devolatilization phase ofthe process. The chimney is usally made of welded oil drums or steelpipe of similar dimension. A sketch of a typical Casamance kiln is shownin Figure 6.5.

6.30 Provisions for collection and storage of the condensible tarsmust be incorporated if a chimney is used with the kiln. The recovery ofthese tars is often cited as an advantage of the Casamance kiln.Mowever, their use is rather limited, primarily as a wood preservative,while their disposal could be a major environmental problem. If thetemperature in the chimney falls below 100 eC, the resulting condensateis primarily water, contaminated by acetic and other pyroligneousacids. The useable tars and oils are too diluted to be economicallyrecovered. Therefore, the use of a chimney for tar collection requiresgood design and skill when operating in order to realize the benefitsfrom tar collection. It ip estimated that 40 liters of liquids and tarscan be collected per 100 ml (stacked) of wood carbonized.

6.31 The main advantages of the Casamance kiln when compared to atraditional earth mound kiln is the shorter cycle time and thepotentially higher efficiency. Total cycle time is estimated to take 4 -5 days with charcoal yields of 20-25 percent on a dry weight basis. Theprimary disadvantage of the Casamance kiln is the skill required to buildand operate che kiln, the importance of proper sealing of the kiln andthe need for good air circulation in the kiln to obtain the expectedyields. Considerable cutting and sizing of wood may be required in thecase of sawmill residues, given the importance of kiln construction.Finally, if tar recovery is incorporated, provisions for collection andstorage of the tars must be provided, increasing the capital cost of theoperation. A comparison of the key characteristics of the Casamance kilnis presented in Table 6.9. A summary of the annualized capital andoperating costs for converting in Casamance kilns all of the millresidues presently converted to charcoal in the Kumasi area is presentedin Table 6.10. Details of these estimates and underlying assumptions aregiven in Annex 13.

- 105 -

Figure 6.5: CA8AI A CE CHACOAL IKILN

ff, , -- ,s--

I' . - __

I ,. * . . .. ..

tt _, -- _ '"-'^

'-- -- "

* - "

I~ b

- 106 -

6.32 Beehive Brick Kilns. The Beehive brick kiln is Most widelyused in Brazil. These kilns can vary in size from 10 to 100 m with themost typical averaging 50 m3. The 50 m3 kilns have a circular base ofabout 5 m in diameter, a cylindrical wall 2 m high with a dome roofabout 2.5 m above the base. In general, the kilns are designed with twodoors diametrically opposite each other, one for charging and a smallerone for unloading. Four to six brick chimneys, evenly spaced around thekiln, are connected to the base of the kiln. Several brick-sized airinlets are maintained at the base of the cylindrical wall and in the domefor control of air flow. These air inlets are opened or plugged up asrequired during the charcoaling process. Once constructed, the bricksurface must be plastered, usally with mud, to seal the kiln. A sketchof a typical Beehive kiln is shown in Figure 6.6.

6.33 The primary advantage of the brick Beehive kiln is that onceconstructed it requires a low labor input, produces a consistent qualitycharcoal and has a relatively high efficiency. A 50 m3 kiln has anaverage cycle time of 7-8 days and requires one trained laborer to load,operate and discharge. However, if the wood pieces are large (above 2 min length and 0.5 m in diameter) additional labor would be required forwood preparation. The average yield is reported at 30 percent on a dryweight basis, some 80 percent higher than that obtained from earth moundkilns. Given the refactory nature of brick kilns, carbonization athigher temperatures can be achieved resulting in "high quality"charcoal. In Brazil, most of the charcoal produced in brick kilns isused by the steel industry. The quality of the charcoal can becontrolled to meet less stringent domestic needs by reducing air inputduring the heating phase with the beneficial result of higher yields. Acomparison of the key characteristics of the Beehive kiln is given inTable 6.9. A summary of the annualized capital and operating costs forconverting in brick beehive kilns all of the mill residues presentlyconverted to charcoal in the Kumasi area is presented in Table 6.10.Details of these estimates and underlying assumptions are given in Annex13.

Financial and Economic Analysis of Improved Charcoal Options

6.34 It would be expected that stationary type kilns would befavored in circumstances where wood for carbonization is availableessentially on-site, in the form of sawmill residues, rather than inforest areas distant from population concentrations and points ofconsumption. The results of the financial/economic analyses are inaccord with this expectation, with the Bethive brick kiln having thelowest cost of charcoal production both on a financial and an economicbasis at US$28.83 and US$50.46 per tonne, respectively. It is followedby the Casamance kiln with production costs calculated as 19 percent(financial) and 19 percent (economic) higher than the Beehive kiln, andthen the TDRI metal kiln at 31 and 22 percent higher. The traditionalearth mound kiln occupies last place, having an economic cost of charcoalproduction some 90 percent higher than the Beehive type.

- 107 -

Fisure 6.6: BEEHIVE BRICK CHARCOAL KILN

IL -I I 8

- 108 -

6.35 A switch in charcoaling regimes for the sawmill residuespresently carbonized in the Kumasi area can be viewed in cost/benefitterms. However, whereas the "before project" scenario is clearly theexisting earth mound method, choice of a new and improved method musttake a number of social variables into account. The capital requirementsof Beehive kilns are beyond the means of the traditional charcoalers; bynecessity, brick kiln operation will be either contracted to privateentrepreneurs sited on or adjacent to the mills or a side business of themill owners themselves. Beehive kilns would require about 50 workers fortheir operation and displace about 300 traditional charcoal makers, or anet employment loss of 250 persons. Adoption of the best low capitalalternative, Casamance kilns, would result in a net displacement of about100 persons.

6.36 Civen the very high financial/economic returns which will bedemonstrated, either the brick or Casamance kiln type may be suitable forintroduction. The Beehive kiln type was chosen for more detailedcost/benefit analysis.

6.37 Financial Analysis. The financial analysis of Beehive vs.earth mound kilns is contained in Annex 13 and summarized in Table6.11. Using the going wholesale price for earth mound-produced residuecharcoal of 8.6 Cedi/kg gives a rate of return of 292 percent and simplepayback time of 4 months, indicating that the proposed improvement shouldbe readily taken up by the private sector. The gains result primarilyfrom the superior charcoal yield and efficiency of labor utilizationcompared to the traditional method.

Table 6.11: FINANCIAL/ECONOMIC ANALYSIS RESULTS,BEEHIVE BRICK KILN CHARDOALING IMPROVEMENT

Financial Economic

NPV 5694,769 S 1,110,261

IRR 292% 490%

Simple Payback Time 4 months 2 months

- 109 -

6.38 Economic Analysis. Details and results of the economicanalysis are also displayed in Annex 13 and Table 6.11. The economicevaluation differs from the financial perspective mainly in the costassociated with the input wood residues. Given that residue-derivedcharcoal is replacing forest-derived charcoal at the margin, the cost ofresidues for carbonization is adjusted upwards to reflect the additionaleconomic opportunity cost of forest wood consumption of US$5.00/tonne.As the conversion efficiency of mound kilns on a wet weight basis isestimated at about 10.6 percent, over 9 tonnes of wet residues go intothe making of each tonne of residue charcoal. A charcoal economic valueof US$96.02/tonne results after adjustment for domestic currencyovervaluation is included. Using the economic value of charcoal thusfurther increases the attractiveness of the proposed intervention from aneconomy-wide standpoint.

Sawdust Briguetting

Existing Plants

6.39 There is one briquette plant presently operating in Ghana,Chaowus Ltd. in Akim-Oda. The plant is owned and operated by a Taiwaneseentrepreneur and has been in production for approximately one and one-half years. Two screw press briquette machines are employed with a ratedproduction capacity of 150 kg/hr of dried sawdust briquettes each or 7.2t/day if operated continuously. Present actual production rate of 1,100t/yr is only half stated capacity due to operational inefficiencies. Inparticular, the presses are idle for one ouJt of three shifts per day dueto lack of sawdust drying capacity. The plant operates six days per weekand employs 24 men, 6 of them skilled or semi-skilled.

6.40 Sawdust Supply. The plant obtains sawdust at no charge fromAkim-Oda area sawmills, using their own 7 tonne truck to haul the sawdustto the plant where it is dumped in a large pile outside the entrance doorto the enclosed plant building. Two laborers then reload the sawdustinto the truck at intervals determined by the briquette productionrate. The truck then backs into the infeed end of the building and dumpsthe sawdust on the floor. Two men shovel this sawdust onto a smallvibrating screen which filters out lumps, stones, sticks, etc.

6.41 Sawdust Drying. A short screw conveyor elevates the sawdustfrom below the screen and feeds it into a piped air stream at theentrance to a vertical "flash tube" dryer. The dryer is heated by thecombustion gases of a rudimentary sloping-grate furnace fed by hand withfirewood and broken briquettes. Gas temperature at the inlet to thevertical section of the dryer is nominally 200 eC to a maximum of250 'C. The dryer is supposed to reduce the wood moisture content to 10percent moisture content (wet basis). However, in the wet seasons thedryer cannot remove enough water so the plant must run a "graveyardshift" drying sawdust and not producing briquettes. The pre-dried

- 110 -

sawdust is then stored and is re-dried nrior to briquetting during thefollowing two shifts. Following the flash tube dryer the sawdust isseparated from the drying gas flow in a cyclone. The sawdust falls outof the bottom of the cyclone into a metering bin. Many glowing emberscarried over from the firewood furnace find their way to the metering bincausing occasional flare-ups. The gases exhaust from the cyclone upwardin to the upper part of the building resulting in significant dustemissions within the workplace.

6.42 Briquette Extrusion. The sawdust feeds out of the metering binto each of the two screw extruders. Each screw is one piece with asurface of welded-on hard stellite which needs to be replaced every 2 to3 days; the welding operation takes about 30 minutes. The briquettes areextruded continuously through a die but they butt into a deflector plateand are broken off at prescribed intervals. The briquettes could be madeto any specified length. Specifications of the briquetting machinery, asobtained from the manufacturer's brochure copied in Annex 14, are givenin Table 6.12. Total plant power demand is calculated as 38.1 kU, withelectrical consumption averaging 110 kWh per tonne of briquettesproduced.

Table 6.12: CHAOWUS LTD. ORIQUETTINg MACHINE SPECIFICATIONS

Manufacturer: Ming-YI Iron Works Co. Ltd, Taiwan

Screw Presses: 2 x IS HP each at 1,150 rpm

Briquetting Capacity: 2 x 150 kg/hr each

Pre-heoaters: 2 x 2 kVA ecn

Forming Muff Heaters: 2 x 4 kVA each

Pre-Weot Temperature: 200-220 *C

6.43 Briquette Characteristics. The 450 mm long briquette producedis cylindrical in shape, approximately 55 mm in diameter with 4triangular keys approximately 4 mm high spaced equally around the outsidediameter and running longitudinally along the outer surface. Thecontinuous hole in the center is 22 -m in diameter. The briquette isextremely dense, hard and strong. Measured briquette fuelcharacteristics are shown in Table 6.13.

6.44 Briquette Marketing. The 1.6 kg briquettes are bundled inpackages of 10 and sold at prices ranging from 60 to 80 Cedis/bundle,with the average delivered price in the Accra/Tema area of 70Cedis/bundle corresponding to US$ 29.17/tonne. Delivery is made by the

- I1I -

Chaowus Ltd. tip truck. The majority of the product is sold to theAccra/Tema bakers' unions; Ankaful Brick Ltd. in Cape Coast has recentlyconverted from firewood to briquettes to fire its two brick kilns andconsumes 5 tonnes per week. The acceptance of the briquettes h&s createda demand that exceeds the supply capability of Chaowus, and --the bakersare not happy with the plant's inability to provide enough briquettes.

Table 6.13: CHAOWUS LTD. BRIQUETTE FUEL CHARACTERISTICS

Density, oven dry 1,281 kg/03

Moisture Content, dry basis 4.9$

Moisture Content, wet basis 5.1$

Density at 5.1% mcwb 1,350 kg/0

Volatiles Content 0.9%

High Heating Value 20.1 NJ/kg

Lower Heating Value (9 5.1% mcwb) 18.9 NJ/kg

Source: Annex 12; Mission estimates.

6.45 Sug8ested Operational Improvements. Given the evidentunsatisfied demand, plant economics could be substantially enhanced ifproduction bottlenecks could be eliminated. The following suggestionswere made to the Chaowus management at Akim-Oda as a means to improvingplant efficiency and output:

(a) Move the outfeed storage pile to the inside of the plant toavoid the problem of the sawdust being rained upon. The unusedspace in the building would allow for this storage.

(b) Move the dryer furnace to increase the burning time between thefurnace outlet and the sawdust introduction and thereby reducethe problem of carry-over of burning embers.

(c) Insulate the flash tube dryer, even if only with a sand-filledsheet metal cylinder around each of the two flash tube runs.

(d) Move the dryer cyclone to the center of the metering bin toeliminate the necessity of a worker climbing the ladder andplowing the sawdust in the bin towards the screw at the far endof the bin.

- 112 -

(e) Extend the dryer cyclone exhaust pipe so the exhaust gases(smoke) are emitted outside instead of inside the plant.

Market Potential

6.46 Supply. The supply of sawdust briquetts in Ghana istheoretically constrained by the availability of s3urplus sawdust. Totalsurplus sawdust in 1986 is estimated at 83,400 m3. If all of this isconverted to briquettes, this would be equivalent to approximately 47,000tonnes/yr of briquettes with an energy value of 22,140 toe/yr orequivalent to 11 percent of the total non-residential fuelwoodconsumption in 1985. If sawdust haulage costs are to be avoided,production of briquettes must occur in localities containing significantconcentrations of surplus sawdust. The areas having surplus sawdust inexcess of 2,500 m /yr were given in Table 3.8. Of the 83,400 m ofsurplus in 1986, 52,600 m is located in the Kumasi area. A little over9,000 @ is found both at Nim and at Sekondi-Takoradi. Approximately2,500 m is found at Nkawkaw and Dunkwa. These five areas account for 91percent of the total surplus sawdust, equivalent to a potential of 43,000tonnes of briquettes.

6.47 Demand. The primary use of sawdust briquettes would be tosubstiFute for fuelwood consumption mainly in the industrial andcommercial sectors where the premium qualities of the briquettes arevalued. To date, the briquettes produced by the Chaowus plant in Odahave been primarily consumed by commercial bakers and at a few brickkilns. Other potential users include fish smokers, chop bars,institutional kitchens, and other small commercial fuelwood consumers.While use of briquettes is possible in the household sector, it isunlikely to be widely used domestically due to unfamilarity with theproduct, the tendency for smoking (compared to charcoal) and flamelesscombustion of the briquettes, and the expected higher cost of thebriquettes. It is also unlikely that large industrial consumers offuelwood would be a major source of demand for briquettes unless thebriquettes are equal or less in price, on an energy adjusted basis, thanbulk costs for fuelwood. Conversion to use of briquettes in industrialboilers and furnaces currently using fuel oil would be uneconomic for thesame reasons that conversion to firing with unprocessed sawdust,discussed earlier, is uneconomic.

6.48 The potential demand for briquettes from the bakers and brickmanufacturers alone, located only in the main urban areas, is estimatedto be 45,000 tonnes per year. The basis for this estimate is presentedin Table 6.14. The potential demand from the bakers and brickmanufacturers alone exceeds the potential supply of briquettes by 2,000tonnes per year. If the fuelwood consumption of fish smokers and chopbars were added, the potential demand for briquettes would increase toover 500,000 tonnes/yr.

6.49 Willingness to Pay. Information obtained from the presentusers of briquettes, the bakers and brick manufacturers, indicated a

- 113 -

strong preference for the briquettes based on a number of technicalreasons. The main advantages of the briquettes were their uniformquality, higher heat (due to lower moisture content) and cleaner burningproperties (less smoke and ash). The primary disadvantage cited bycurrent users was the reliability of supply. Most users of thebriquettes indicated a willingness to pay a 10 to 20 percent premium (onan energy equivalent basis) over their cost of fuelwood.

Table 6.14: ESTIMATED SAWDUST BRIQUETTE DEMAND

Rat$ of EstimatedNumber Total Fuelwood Briquette Briquetteof a/ Consumption to Fuelvood Demand

Application Producers ('000 t/yr) Consumption ('000 t/yr)

Bread Baking 6,575 52.6 2:3 35

Brick/Tile Manufacture 38 19.8 1:2 10

a/ Fuelvood users In urban areas only.Sources "Report on Pilot Survey on Fuelwood and Charcoal Consumption In Accra,"

Government of Ghana National Energy Board, October 1985; Missionestimates.

6.50 The market price for fuelwood varies throughout Ghana,depending in part on the distance of the market from the source offuelwood supplies. A summary of wholesale fuelwood costs in major urbancenters is presented in Table 6.15. The highest prices occur in theAccra/Tema area where financial costs are US$25/t and economic costs areUS$27/t. The equivalent economic cost per unit of energy is 19.2 UScents/GJ. The Sekondi/Takoradi and Cape Coast areas represent the nexthighest areas for fuelwood with financial and econimic costs in the rangeof US$16/t and of US$19/t, respectively. The Kumasi area has financialand economic costs of fuelwood estimated at US$10/t and US$14/trespectively.

Table 6.15: FINANCIAL AND ECONOMIC COSTS OF FUELMOOD

Area Financial Costs Economic Costs(USS/t) a/ (USS/t) a/ (USS/GJ)

Accra/Tewo 25 27 1.93Kumsi 10 14 1.00Sekondi-Takoradi 16 19 1.36Cape coast 14 17 1.21

a/ Bas. on air-dried weight typically 30% mceb.Source: Mission estimates.

- 114 -

6.51 Estimates of the "willingness to pay" or breakeven price forbriquettes consumed by bakers in the Accra/Tema area and brickmanufacturers in the Cape Coast area are presented in Table 6.16 and 6.17respectively. In the case of the bakers, the breakeven price isUS$37.50/t of briquettes. However, interviews with several of the bakersindicated a willingness to pay of up to a 20 percent premium for thehigher quality characteristics of the briquettes including ease ofhandling and storage, faster ignition and heatin8 of ovens, consistentfuel characteristics resulting in more predictable quality of bakedproduct and less fouling of ovens due to better combustion of briquettesand lower residue ash. Assuming a conservative 10 percent premium overthe energy equivalent breakeven price with fuelwood results in a unitvalue of US$41.67 per tonne of briquettes. 14/

6.52 The breakeven price for briquettes consumed by the brickmanufacturers in Cape Coast is considerably higher than that for thebakers in Accra/Tema. At present, the brick makers are paying US$33.33/tfor briquettes from Oda delivered to their plant in Cape Coast. However,the use of briquettes have several operational advantages which result ina net financial benefit to the brick makers of approximately US$27.92/tof briquettes used or an equivalent willingness to pay for briquettes ofUS$61.25/t. The most significant benefits are reduced labor costs andincreased yields of finished bricks per kiln firing. Labor requirementswere reduced by over two-thirds primarily by eliminating the splittingand chopping of fuelwood and reducing the mass of material needing to bestored and conveyed. Also, because of the higher and more uniformtemperature of the heat generated by the briquettes, the yield ofacceptable bricks increased by some 25 percent. With fuelwood firing thethe yield was lower primarily due to tverfiring of bricks near thefurnace and underfiring of bricks at the top of the kiln. A comparisonof the costs and benefits associated with fuelwood and briquette use bythe brick makers is presented in Table 6.17. In addition to these twoquantifiable financial benefits, the brick manufacturers indicated thatthe laborers prefered to work with the briquettes because of their easierhandling characteristics. The smaller size of the briquettes allows themto be placed completely inside the furnace area and thereby reduce theamount of heat escaping at the mouth of the furnace. With firewood, thetemperature near the furnace was extremely uncomfortable causing thelaborers to pitch the firewood from a distance into the furnace. Thispitching of the firewood caused noticable damage to the furnace doors andkiln.

14/ Figure obtained from bids of 100 Cedi per 16 kg bundle of briquettesin bidding simulations with bakers.

- 115 -

Table 6.16: BRIQUETTES VS. FUELWOOD COIIPMISONIN BREAD BAKING, ACCRA

Fu.lwood Briquettes

Fuel Input/45 kg sack flour a/ 48 kg 32 kg

Cost of Fuel Input/45 kg sack flour 180 Cedis 120 Cedis

Fuel Financial Cost/tonne 3,750 Cedis 3,750 Cedis

Net Briquette Financial Benefit/45 kg Sack Flour 60 Cedis

Net Briquette Financial Benefit/tonne 1,875 Cedi/t

ireakeven Price: 3.750 Cedi/tWillingness to Pay + Benefit: 1,875 Cedi/tof Consumer Total: 5,625 Cedl/t

(USS37.50/t)

a/ Based on Inputs required per 45 kg sack of flour,

Source: Mission estimates.

Table 6.17: BRIQJETTES VS. FUELWOOD COMPARISONIN BRICK MANiFACTURING, CAPE COAST

Fuelwood Briquettes

Fuel Input a/ 9.5 t (I' GJ LiV) 4.8 t (91 GJ iJV)Cost of Fuel bl 1,789 Cedi/t 5,000 Cedi/tTotal Cost of Fuel Input 17,000 Cedi 24,000 CediFiring Time 3 days 1.5 daysCycle Time 7 days I daysFiring Labor 27 man-days 9 man-daysCost of Labor Input 6,000 Cedi 1,900 CediBrick Recovery 3,900 bricks 4,900 bricksValue of Recovered Bricks C/ 89,700 Cedi 112,700 CediNet Briquette Financial Benefit/Firing - 20,100 CediNet Briquette Financial Benefit/Tonne - 4,188 Cedi/t

(US$27.92/t)Oreakeven Price: 5,000 Cedi/tWillingness to Pay + Benefit: 4.188 Cedi/tof Consumer

Total: 9,188 Cedi/t(USS61 .25/t)

a/ Based on single firing of 5,200 bricksb/ Prices for fuel delivered to brIckmaker.cS/ 23.0 Cedi/brick sales price.Source: Mission estimates.

- 116 -

6.53 In estimating briquette consumers' willingness to pay, it isrelevant to consider whether present fuelwood consumers would be betteroff if they were to use petroleum fuels as a substitute for firewoodinstead of briquettes. This comparison has been calculated forbriquettes vs. inland fuel oil, the most economic petroleum fuel in smallscale heat raising installations, and is presented in Table 6.18.

Table 6.18: BRIQUETTES VS. INLAND FUEL OIL COMPARISON, ACCRA

IFO Briquettes

Heat Value (MJ/kg) 42.8 18.9

Financial Cost (USS/Tonne) 111 49.02 a/

Economic Cost (USS/Tonne) 86 37.98 a/

a/ Breakeven values on a unit energy basis.

Source: Mission estimates.

The comparison shows that briquettes could cost the consumer up toUS$4S.02/tonne and still be cheaper to use than IFO, thus supporting theUS$41.67/tonne willingness to pay figure for briquettes. On an economicbasis, the opportunity cost of IFO consumption of US$37.98 (FOBAccra/Tema) is below the opportunity cost of briquette consumptionestimated as:

Financial WTP : $41.67/tonneSCF adjustment : x 0.88

$36.67/tonne

Plus:Opportunity cost of

fuelwood consumption : $ 5.00/tonneAdjustment for 3:2 ratio: x 1.50

(fuelwood:briquettes) $ 7.50/tonne

Total Economic WTP : $44.17/tonne

However, given the small scale of the individual bakery operations andthe investment requirements for IFO combustion retrofit, oil-firing hasnot been considered as the next best opportunity when computing briquettesubstitution benefits. The economic analyses have therefore usedUS$44.17 as the unit benefit valuation.

- 117 -

6.54 The sawdust briquettes have combustion characteristics that aremore akin to low grade domestic charcoal than firewood. The possibilityof substituting briquettes for charcoal consumption in some cases shouldbe investigated. If briquettes were accepted as a charcoal substitute,the willingness to pay for briquettes would approach US$55/t in Accra andUS$85/t in Takoradi (based on financial costs for charcoal adjusted forheating values).

Proposed Plants

6.55 The operatiou of the Chaowus Ltd. sawdust briquetting plant inOda has provided several important pieces of information relating to thefeasibility, design and operation of such plants and the subsequentmarketing and use of sawdust briquettes in Ghana. The evalLtion of theoperations of the Oda plant indicated the possibility of severalimprovements to increase the overall productivity of the plant whilesimultaneously reducing plant operating cost. The Oda plant has alsodemonstrated the viability of operating the briquetting equipment inGhana. The success of the Oda plant has led Chaowus Ltd. to considerconstructing a larger briquetting plant in Kumasi. The exact status ofthis proposal is unknown at this stage.

6.56 Three briquetting plants of varing sizes and locations areproposed to evaluate the economics of briquette production. The smallestplant, with a capacity of 3,500 t/yr briquette output, is considered tobe located in Takoradi. The plant will require about 6,700 m /yr ofsawdust which is about two-thirds of the surplus sawdust available in theSekondi-Takoradi area. The other two plants are considered located inthe Kumasi area with the Kumasi-A plant having a capacity of 7,000 t/yrbriquette output and the Kumasi-S plant having a capacity of14,000 t/yr. The sawdust requirements of the larger plant are estimatedat 27,000 m3, slightly more than 50 percent of the surplus sawdustavailable in Kumasi. Characteristics of three proposed plants arepresented in Table 6.19.

- 1l8 -

Table 6.19: CHARACTIUSTICS OF PROPOSED SAWDUSTBRIQUETTE MANUFACTURING PLANTS

No. Rated Assumed Plantof Output/ Output/ Output/ Plant Sawdust

Plant Screw Press Press Year o/ Consumption/Year b/Location Presses (kg/hr) (kg/hr) (tonnes) (tonnes) (m' SWE)

Takoradi 4 150 120 3,500 4,878 6,207

Kumasi-A 10 150 120 7,000 9,758 12,415

Kumaesl-B 20 150 120 14,000 19,515 24,828

a/ Based on 3 shift, 6 day/week operation 50 weeks/year with 80Y plantavailability factor.

bl Based an Input sawdust at 36% mcwb and 786 kg/M3, and output br I quettes at 8%mcwb. Amounts do not Include Internal plant sawdust consumption for sawdustdrier.

So.ree7 Mission estimates.

6.57 Costs. A summary of the capital investment and annualoperating costs for the three proposed briquetting plants evaluated ispresented in Table 6.20. A detailed breakdown of these costs andsupporting equipment manufacturer's quotation are given in Annex 14. Thecapital investment required for the 3,500 t/yr briquette plant inTakoradi is estimated to be almost US$300,000. The investmentrequirements for the 7,000 t/yr Kumasi-A plant is expected to beUS$460,000 and for the larger 14,000 t/yr plant, US$780,000. Annualoperating costs range from US$68,000 for the Takoradi plant to US$143,000for the large Kumasi plant. Economies of scale are apparent in both theinvestment and annual operating costs. A four fold increase in thecapacity of the plant results in only a two and two-thirds increase inthe capital investment and only a doubling of the annual operating costs.

6.58 Estimates of transporting the briquettes to the market are alsopresented in Table 6.20. The estimates represent expected costs oftransporting all the briquettes to consumers in the Accra/Tema area wherefuelvood prices are highest. The briquettes from the Takoradi plant areassumed to have an average one-way transport distance of 240 km whilethose produced in Kumasi have a transport distance of 285 km.

6.59 Table 6.21 summarizes the factory gate and delivered financialcosts for briquettes produced at the three proposed plants. Again, theeconomies of scale are evident in the factory gate production costestimates. Briquettes produced at the 3,500 t/yr plant in Takoradi havea financial breakeven cost of US$32.53. Those produced at the largerplants in Kumasi decrease in cost to US$22.13/t at tbe 7,000 t/yr plant

Table 6.20: SAWR OF CAPITAL AND Nm L oPEERA?T- comTS OOSTS FOR FPOSED WliquETTtilS PLAiTS

TAKORADI KWASI-A KIdMS ILoceal Costs Foreign Cost Total Local Costs Foreign Costs Total Local Costs Foreign Costs Total

Item (lOwO uSS) ( 000 iiS S * (I'00Uts S) (0005 uS O) O S000 iiS) ( uS S) (OM00 tiS * tooo us S) (1011 to S S)

Capital CostsConstruction 51.6 13.0 64.6 64.3 15.0 79.3 116.9 28.0 144.9

Equiplnt - 121.0 121.0 - 224.0 224.0 - 307.0 387.0

Transport Charges I/ 12.3 16.0 28.3 22.8 29.6 52.4 39.4 51.1 90.5

Engaineeinog Installation 10.8 67.8 76.6 13.9 90.t 104.6 26.9 130.2 1572

and ontingencies

Total Capital Costs 292,5 440.3 Moe d

Annual Operating CostsLaor 12.0 - 12.0 t4,g - 14.0 23.2 - 23.2

POWr 27.8 - 27.8 23.7 _ 23.7 41.3 - -

0 a 14 7.7 12.9 20.6 14.2 II'5 31.7 23.1 34.9 56.0

Cons4mbles 0.4 1.1 1.5 0.9 2.0 3.7 I.i 5.3 ?.l

Continoencies - 6.2 - - 7.4 - - 13.0

Total Annual Operating Costs 68.1 61.3 142.6

iriquette Traeort Costs 533. 127.7 2r5.4

a/ Includes international freight ad insurane, local port chargs and bak fees.

- 120 -

and US$18.95 at the 14,000 t/yr plant. Transportation costs add anadditional US$15.36/t to the briquettes produced in Takoradi while ittakes US$18.24/t to transport briquettes from Kumasi to the Accra/Temamarket. However, the possibility exists of marketing the briquettescloser to the proposed factories as transport costs add significantly tothe delivered costs of the briquettes. The economics of this marketingaspect are discussed later.

6.60 Financial Analysis. The estimated willingness to pay of thebakers of US$41.67/tonne has been taken as the marginal demand price forpurposes of the financial cost/benefit evaluation. Results aresummarized in Table 6.22 and details of the analysis may be found inAnnex 14. The Takoradi plant is shown to be an infeasible investment,and the smaller of the two Kumasi-situated plants is only marginallyprofitable. The Kumasi-B plant is able to exploit economies of scale tothe extent sufficient to show a 19 percent financial rate of return. Perthe critical parameter analysis, the profitability of the Kumasi-Binvestment is maintained under varying assumptions of capital cost,briquetting machine lifetime, and briquette demand price. Sawdust isassumed to be available for the cost of haulage: Given the lack ofalternative markets for surplus sawdust, it is unlikely that the supplyprice will rise sufficiently to significantly impact briquettingprofitability.

6.61 The effect of marketing the briquettes at plant gate wasexplored by (a) netting out all briquette transport costs and (b)downward adjusting the assumed briquette demand price to reflect lowerfuelwood prices in the areas surrounding the factories. This analysiswas performed for the Takoradi and Kumasi-B plants and is presented inTable 6.22 and Annex 14. The gains in reduced transport costs from localmarketing do not offset the loss in revenues from the assumed lowerbriquette prices of US$26.40/t and US$16.50/t in Takoradi and Kumasi,respectively. The results suggest Accra-Tema, with approximately 75percent of the urban population of Ghana, as the primary market for thesawdust briquettes. This is in accordance with general experience withcharcoal: Woodfuels having a higher energy content per unit weight tendto have a comparative advantage in urban markets distant from woodsources. More limited markets may be found among Takoradi and Kumasiusers willing to pay a substantial premium for the superior combustioncharacteristics of the briquettes compared to firewood. Plans of theChaowus management to market briquettes in the Northern region do notappear feasible given the long transport distances and probable lowwillingness/ability to pay of the consumers.

- 121 -

Table 6.21s SUMARY OF BRIQUETTE PRODUCTION AND TRANSPORT FINANCIAL COSTS

Plant Annual Capacity Production Costs Transport Costs Total Delivered

(t/yr) (US$/t) (USS/t) (USS/t)

Takoradl 3,500 32.53 15.36 47.89

Kumasi - A 7000 22.13 18.24 40.37

Kumasi - B 14,000 18.95 18.24 37.19

Table 6.22: FINANCIAL ANALYSIS iESULTS, PROPOSED

SAWDUST 8RI3QtETTING PLANTS

Takoradi Kumasl-A Kumasl-i

(3,500 t/yr) (7,000 t/yr) (14,000 t/yr)

Marketed in Accra/Tema

NMPV (204,800) 553,860 S498,240

FIBR Undef. 12% 19%

Capital Cost Switching Value -50% +8% +445

Briquettor Life Switching Value 20 yr 8 yr 4 yr

Discounted Payback Time 20 yr 16 yr 7 yr

Briquette ireakeven Price 547.89 $40.37 537.19

Marketed at Plant

NPV ($202,020) -- ($327,350)

FIRR Undef. -- 2%

Capital Cost Switching Value -47% -28%

Briquettor Life Switching Value 20 yr -- 20 yr

Discounted Payback Time 20 yr - 20 yr

Briquette Breakoeven Price S 32.53 518.95

- 122 -

6.62 As depicted in Table 6.23 national wood demand is forecastunder a trend-based scenario to rise at 3.6 percent per annum. Financialanalysis results were therefore tested under the assumption that woodsupply is approximately fixed, leading to a real price increase of thebriquettes of 3 percent/yr or 81 percent higher in 20 years. Under thisscenario all three plants are feasible, having an FIU of 14, 22 and 28percent respectively.

Table 6823: ESTIMATED WOOD EALANCE, 1986-2000

1986 1990 1995 2000(miliIon m3 RUE)

Fuelwood consumption 12.0 13.4 15.4 17.6Industrial wood consumption a/ 1.S 1.5 2.0 2.0

Total 13.5 14.9 17.4 19.6

Wood Increment b/ 17.6 16.9 16.6 16.0

Balance 4.1 2.0 (0.8) (3.6)

a/ Industrial wood consumption Includes only the portion of woodutilized for Industrial manufacture and energy generation.

b/ increwent flgure Is taken from a working document prepared Inconnectlon with Bank report: Ghana: Issues and Cotions in theEnergy Sector, May 1986 (Report No. 6234-&il).

Source: World Bank.

6.63 Economic Analysis. The estimated economic willingness to payof the bakers of US$44.17/t has been taken as the marginal benefit forthe economic cost/benefit evaluation. Results are summarized in Table6.24 and details of the analysis are given in Annex 14. Similarconclusions as to the desirability of the proposed investments followfrom the economic analysis as from the financial perspective. The maindifferences in the financial and economic benefit streams are due tovarying electricity tariffs charged to the variously sized plants. TheTakoradi plant is assumed to fall under the ECO "Commercial .- Less than100 kVA maximum demand" tariff which gives a higher average cost per kWhthan the estimated economic LRMC of supply. In the case of the largerKumasi plants, both qualifying for special industrial tariffs, thesituation is reversed; the economic cost of electricity exceeds thefinancial cost.

- 123 -

Table 6.24: ECONONIC ANALYSIS RESULTS,PROPOSED SAWDUST BRIQUE9TTINS PLANTS

Takoradi KumasI-A Kumasi-B(3,500 t/yr) (70000 t/yr) (14,000 t/yr)

Marketed In Accra/Toma

NPV ($11,200) 599,100 S500,330

EIRR 9S 13S 191

Capital Cost Switching Value -3% +15% +48%

Briquettor Life Switching Value 13 yr 7 yr 4 yr

Discounted Payback Time 20 yr 13 yr <7 yr

Briquette Breakeven Value S44.64/t S42.60/t S40.07/t

Marketed at Plant

iNPV 73,710 - $163,520

EIRI 14% - 13%

Capital Cost Switching Value +19% - +16%

Briquettor Life Switching Value 6 yr - 7 yr

Discounted Payback Time 13 yr -- 13 yr

BrIquette ireakeven Value S28.32/t -- $20.69/t

6.64 Local marketing for briquettes is more attractive in the socialcost/benefit analysis than under pure private market assumptions,especially in the case of the Takoradi plant. However, it is still moresocially profitable to market the Kumasi-B plant output in the Accra-Temaarea than in Kumasi.

6.65 As in the financial analysis, assumptions about the future rateof woodfuels price inflation have a substantial effect on the economicanalysis results. Economic internal rates of return increase to 21, 24and 29 percent respectively when the value of briquettes inflates at 3percent real per annum.

- 124 -

Charcoal Briquettes

Background

6.66 There are a number of methods that have been proposed for theproduction of charcoal briquettes. Essentially there are two majordifferences between methods: one is to carbonize the sawdust beforebriquetting it; while the other is to briquette the sawdust beforecarbonizing it. In the first method, the loose sawdust is normallycarbonized in a continuous retort. The resulting char is collected,cooled in a water quench tank, mixed with an organic binder and extrudedinto briquettes. The briquettes are then stacked on steel carts andpushed into a drying kiln which is fueled by the hot pyrolysis gases fromthe retort. This process is used in North America to produce charcoalbriquettes from sawdust for the barbecue market. The resultingbriquettes are costly primarily due to the use of starch binders.

6.67 Alternately, it is technically possible to convert to charcoalthe sawdust briquettes that are produced by the heat extrusion screwpress briquetting machines presently used at the Chaowus plant in Oda andproposed for the briquetting plants evaluated earlier. Similarly producedbriquettes are routinely converted to charcoal in plants in Japan andTaiwan. The sawdust briquette, because of the high pressure andtemperature under which it is initially produced, is sufficiently bondedto maintain its integrity during the charcoaling process. Uniformcarbonization is aided by the fact that the briquette is hollow in themiddle. The charcoaled briquette looks much like the sawdust briquetteexcept it is black and has a rough and somewhat irregular surface.

6.68 For carbonization, the briquettes are placed in small, fourwheel enclosed steel carts. The carts have lids which are tightly sealedto eliminate the addition of air which would ignite the briquettes duringcarbonization. The carts are fed through a "hot tunnel" in a continuoustrain. The tunnel is heated by an external source: either electricity,firewood or fossil fuels. The carts pass through the tunnel at a speedsufficient to allow carbonization. After the carts exit the tunnel theyare allowed to cool before the charred briquettes are taken out andbagged. In some operations in Japan, the sawdust briquettes arecarbonized in batch mode. The briquettes are piled on a steel base and asteel dome is lowered over the pile to seal it. External heat isprovided for carbonization and the charred briquettes are allowed to coolprior to raising of the steel dome.

6.69 It is estimated that the theoretical maximum -recovery forcharcoal with 15 percent volatiles from sawdust briquettes isapproximately 42 percent on a dry weight basis. This recovery rate isbased on actual tests with briquettes obtained from the Chaowus plant inOda. A more practical recovery rate for an industrial kiln type operationis estimated at approximately 32 percent. Recovery rates would varydepending on the actual quality of charcoal required. Higher quality

- 125 -

charcoal (i.e., lower volatile contents) would have lower recovery ratesand vice versa.

Production Options/Economics

6.70 Chaowus Ltd. indicated that a plant producing 4 tonne/day ofcharcoal using the steel cart method would require a capital investmentof approximately US$100,000. No estimates on operating costs wereavailable. However, assuming two additional unskilled laborers pershift, annual O&M costs of 5 percent of capital and energy costs of aboutUS$10/tonne of charcoal results in an estimated production cost for thebriquettes of $29/tonne in addition to the cost of producing thebriquettes. Taking the financial production cost at the Takoradi plantof US$32.53 and accounting for conversion to charcoal at 32 percentresults in a overall charcoal briquette production cost of approximatelyUS$137/tonne of charcoal. This is slightly higher than the financialcost of forest derived charcoal in the Takoradi market of US$132/tonneand considerably higher than the price of US$80/tonne for charcoal in thedistant Accra/Tema market. If charcoal is produced from briquettes atthe large Kumasi-B briquetting plant, the estimated production cost forcharcoal briquettes drops, due to the lower production coat for sawdustbriquettes, down to US$92/tonne. Again, this is higher than thefinancial price of wood charcoal in the Kumasi market of US$66/tonne.

6.71 The Mission's consultants, after evaluating the sawdustbriquetting process, have proposed an alternate means for continuousproduction and carbonization of the briquettes. The proposed system,shown in Figure 6.7, requires coupling a tunnel kiln at the exit of thebriquette machine. The hot, freshly produced sawdust briquettes wouldimmediately enter a sealed tunnel kiln which is heated by pyrolysis gasesrecovered downstream of the kiln. The length of the kiln would bedimensioned to allow sufficient carbonization time and the temperature ofthe kiln controlled to achieve the desirable chatZoal quality. Theconsultants have estimated that the capital cost for a 7 tonne/daycharcoal production capacity would be approximately US$125,000.Additional labor and energy costs would be minimal due to the coupling ofthe kiln directly to the output of the briquetting machines and the useof the pyrolysis gases for heating the kilns. However, equipment O0Mcosts would be higher due to the mechanized nature of the process. Forthe purposes of this analysis it is assumed to be 10 percent which islikely to be an upper limit. With these assuptions, the production costof charcoal from the sawdust briquettes is approximately US$16/tonne.Given the production cost of sawdust briquettes at the proposed Takoradiplant, the resulting cost of charcoal briquettes would be approximatelyUS$124/tonne. The cost of the charcoal briquettes produced in Takoradiis marginally below the {inancial price for charcoal in the Takoradimarket. The production cost of charcoal briquettes from the Kumasi-Bplant is estimated at US$79/tonne which is still higher than thefinancial price for charcoal in the Kumasi market.

op.i cr l J+o nri

'J~t t VEC49 420 clenf

: |6_^N1evf2O it z\ /2sb;~+cnsc| >w m zr |E eslc >'

t \L Ole*> Volsrne: tominbup"CW^ au

I cz v irl

ns~~~~~~~~~~~ 1t. nn1

.~~~~~~~~~~~~~~~~~~~Tp coiR*tiS,thrw bt;4iclldi.6qdb4tin ;un

Co9TJb 4SoSoO BICZA E TTE cld*2cooL:tJq ic4I

- 127 -

6.72 The continuous sawdust briquette carbonizing process proposedby the consultants has not been demonstrated on a commercial scale. Also,as indicated earlier, the demand for sawdust briquettes is greater thanthe potential supply. Civen both these facts and the marginal economicsof converting briquettes to charcoal it would be unadvisable to seriouslyconsider the commercial production of charcoal briquettes in Ghana.

- 128 -

VII. flOES AND COECLUJICS FOM INCREASED AND/OI INPROVEDUTILIZATION OF VOOD INDUSTRY URSIDUES

Su _ ry

7.1 Investments in sawmill process heat generation and sawdustbriquetting have been identified which could economically utilize thebulk of the 18,000 toe energy contained in the present sawdust surplus.Total potential levels of investment exceed US$8 million with economicrates of return ranging from 19 c:o 46 percent per sub-project. Actuallevels of investment for sawmill process heat are dependent on the marketfor treated and dried lumber and the level of presently unused boilercapacity. The investments will maximize economic benefits if provisionsare made to utilize sawdust as a residue fuel insofar as possible.Needed improvements include upgraded sawblade guides, better sawdusthandling and storage, and furnace grate and airflow systems enablingdirect sawdust combustion. A briquetting plant investment is contingentupon significant unused sawdust quantities remaining in Kumasi afteramounts for on-site combustion are subtracted out. Unutilized sawdustwill continue to be found at smaller mill sites having no process heatdemand, while both solid residues and sawdust will be significantlyunderutilized at Nim Timber Co. and Cliksten Vest Africa Ltd.

7.2 Solid residues not finding their way into high value non-energyuses should be carbonized by the most efficient means practicable.Beehive brick kilns are recommended for charcoal production from sawmillresidues. Total investment requirements for carbonizing in brick kilnsall residues presently charcoaled in Kumasi is estimated at US$110,000with economic rates of return from 300 to more than 500 percent.

On-Site Utilization

Sawmill Process Heat

7.3 Results from analysis of a typical large sawmill point to thepotential for significant economic gains from increasing on-site use ofresidues for process heat generation. Major economic benefits to bederived include:

(a) Higher value-added through export of kiln-dried products, thusincreasing foreign exchange earnings in the forestry sector;

(b) Increased range of economically exploitable species through logsterilization and lumber drying, thus reducing logging pressureon progressively scarce primary species and maintaining sawmillcapacity utilization.

- 129 -

Financial and economic rates of return exceeding 30 percent arepossible for the required investments in boiler equipment, steaming vatsand dry kilns. The potential net economic contribution per unit ofexports is estimated in Table 7.1.

Table 7.1: NET ECONOMIC CONTRIBUTION TO EXPORT LUMBERUNIT VALUE TROUG KILN DRYING (S/r 3)

Redwoods Whitewoods

Value Added 55.05 21.00

Drying Cost 19.18 10.96

Net Contribution 35.87 10.04

Source: Annex 5.

7.4 Investment potentials and resulting effects on sector earningsand residue utilization are summarized in Table 7.2 for three exportscenarios. For purposes of illustrating potentials, all export lumber isassumed to be kiln-dried. Scenario A is based on current export outputand shows a annual net economic contribution of US$2.7 million from aninvestment of US$7.4 million. In this case, the economic net presentvalue over the estimated 15 year life of the investment would amount toUS$20.4 million. Scenarios B and C illustrate the effects of long-termtrends toward increased log processing and improved product recoveryoccurring in conjunction with industry modernization. Estimates havebeen derived by applying scaling factors to the sawmill process heatmodel discussed in Chapter 5.

7.5 Estimates of scale of investment potential and benefits fromthis residue use are, by necessity, proximate. The scope for millprocess heat utilization will be primarily export market-driven, andaccurate forecasts of European acceptance of kiln-dried Ghanaian productswould require a greater level of understanding of the lumber trade thancovered by the Mission. In addition to marketing intelligence,complementary inputs are required to assure quality control, properpackaging, and shipping infrastructure.

7.6 On the basis of statistics compiled on present on-site residueconsumption, it would appear that there exists substantial unused boilercapacity at a number of the combination mill sites. Observations made atmills visited confirm this view; in many instances operating pressuresindicated steam outputs of 40 to 60 percent of boiler capacity. In thiscircumstance, required investments would be lower due to the pre-existingboiler capacity, and rates of return to the incremental kiln/vatinvestments would be higher.

Table 7.2: SAWNILL P0CES HEAT INVESTEIT POTENTIAL

Net UnitContribution Potential

Project Export to Forest Product Net Annual Estimated Estimated AnnualLumber Output Economic Value Economic Investmnt RequIrement Rquirmet Reldue( 000 *3) (S/0 3) Contribution for 100% Kiln-Drying for 100% Kiln-Drying

Sconario Redwood -itewood Redwood Whitewood (million S) (million S) ( 000 t/6) (000 m3 /Vr)

A. 1985 Projected 66 33 35.87 10.04 2.7 7.4 63 80(66%) (34S)

B. Scenario A plusall export logsconverted to 74 100 35.87 10.04 3.7 11.S go 133export lumber (42S) (58%) a-

C. Scenario B plusImproved product 89 121 35.87 10.04 4.5 13.8 120 162recovery (42%) (s%a)

e/ Based on estimated 587/i 3 for redwoods and SSO/m for whitewoods.

Source: Table 5.5; Mission estimates.

- 131 -

7.7 Future investments in on-site process heat generation willmaximize economic benefits if provisions are made to utilize sawdust as aresidue fuel insofar as practicable. As discussed in Chapter 5, thiswill entail improvements in sawblade guides, sawdust handling andstorage, and installation of boilers with appropriately designed gratesand airflow systems to enable direct sawdust combustion. These featuresare incorporated in modern sawmill boiler/furnace systems, but would haveto be retrofitted to the majority of existing installations. The presentfinancial opportunity cost to sawmill owners of combusting solid residuesinstead of sawdust in their boilers is low. The effective cost is thesales value of slabs and edgings for firewood at about US$3-4 per greentonne. As the economic value of solid residues as a substitute forforest wood or as a feedstock in efficient stationary charcoal kilnas isconsiderably higher, tax/regulation incentives may be warranted topromote the on-site utilization of sawdust instead of the solid residues.

Cogeneration

7.8 Grid Connected Mills. Almost all large sawmills in Ghana areconnected to the national grid, and therefore receive the benefit of lowcost hydropower, The mission estimates that electricity could beproduced at these mills in residue-fired turbogenerators operating in acogeneration mode for approximately 5.6 to 7.1 US cents per kWh. Withpresent industrial tariffs set in the range of 3.5 US cents/kWh, there isno financial incentive for mill owners to make the investments that wouldbe required. While one mill (Specialised Timber Products Ltd.) in Kumasiis in the process of installing a 480 kVA boiler/turbine, thecircumstances are atypical as the cogeneration equipment was acquired atconsiderable discount on the second-hand market.

7.9 Mill owners wishing to sell excess self-generated electricityto the grid could be offered a tariff in which sell rates exceed buyrates. However, the economics (i.e. from a national energy planningperspective) are unfavorable. Based on an analysis performed at twosites, the marginal economic cost of residue fueled electricitygeneration of 5.8 to 7.6 US cents/kWh is higher than the estimatedmarginal grid cost of the present 1,072 MW hydro-based system of 5.2 UScents/kWh. In the short-term at least, sawmill cogeneration investmentscannot be recommended.

7.10 Taking the longer term view, it is well established that thelow cost hydropower sites are fully exploited, and that additions tohydro capacity will entail higher marginal costs. The least cost hydroexpansion unit, the site at Bui as identified by Acres International forVRA, is costed at 7.9 US cents/kWh. Other expansion alternativesexamined by Acres include coal-fired and oil-fired thermal generation.The preliminary evaluation made herein indicates that sawmill-basedcogeneration would be competitive with the alternatives just mentioned.The availability of off-shore natural gas and the ultimate costs ofelectricity generated from this source remain to be defined.

- 132 -

7.11 Projections made by Acres forecast a need to bring 200-400 MWof generation capacity on-line in the mid-1990's. While wood residueelectricity may be the least-cost expansion alternative at that time, itshould be pointed out that the total installed boiler/turbine capacityrequired to combust half of the residues presently produced in Ghanawould not amount to 20 MW. A decision whether to pursue the wood residuealternative in light of its small contribution relative to the nationalelectricity picture is beyond the scope of the present effort, but shouldbe considered by national energy policy/planners.

7.12 Mon Grid Connected Mills. Two large mills are located off-grid. The African Timber and Plywood mill in Samreboi is well suited forcogeneration, and private financing has recently been secured to renovateits nearly 40 year old powerhouse. The situation at Mim Timber Co. Ltd.is unique in that a US$2.2 million grid extension is planned, butconstruction has not yet begun. A search for the least-cost electricitysupply option confirms grid extension as the best choice. A 1.2 MWbase/intermediate load residue-fired cogeneration plant coupled to theexisting diesels for peaking power appears to be a marginally cheaperalternative when grid power and extension costs are considered. However,the required US$2.14 million extra investment for electricity generationcapacity yields an economic internal rate of return of only 12 percent.Consideration of risk and management burden of the wood/diesel hybridsystem leads to the grid extension conclusion.

7.13 The same 1.2 MW cogeneration plant, when connected to the gridand operated in a base load mode, could produce electricity at a10 percent higher cost than the present hydro system. A decision toinvest in this project should be made using the same rationale asouttined in paragraphs 7.10 and 7.11. In addition, the considerationsshould include the potential contribution to network stability fromdistributed power production fed-in at a grid endpoint.

Direct Utilization in the Industrial Sector

Residue Substitution for Oil Fuels

7.14 The economics of substituting unprocessed sawdust for fuel oils(RFO/IFO) in industrial heat-raising applications are unfavorable. Ghanais a net exporter of these fuels from the Tema refinery and the low FOBprices received on the world petroleum products market are reflected intheir opportunity cost of consumption. Petroleum prices would, in mostcases, have to rise above US$30/barrel in 1986 dollars before the capitalcosts of conversion could be amortized by the fuel savings. Theeconomics of substitution for the higher cost gas oil were not evaluatedas the small scale of installations burning this fuel make economicconversion unlikely. In any event, switch-over to Inland Fuel Oil is themore probable economic course of action in this instance.

- 133 -

Residue Substitution for Fuelwood

7.15 The scope for substituting unprocessed sawdust for fuelwood incommercial/industrial boilers is limited. The greatest potential existsat sites close to sawmill operations where transport and handling costscan be minimized. Neverthless, economic benefits are small and arefurther limited by the small scale of coumercial operitions which combustfirewood. Boiler efficiencies observed at many of these sites suggestthat significant savings in fuelwood consumption could be achievedthrough low cost furnace improvements and operational procedures.

Conversion Alternatives

Improved Charcoal Making

7.16 Between 25 to 30 percent of all solid residues produced by themills goes into charcoal making where it is carbonized using the sameinefficient earth mound technology as found in the Transition zone andother traditional charcoal making areas. As the wood residues areavailable from a stationary source, higher efficiency carbonizationtechnologies can be applied leading to an increase in charcoal yield ofover 80 percent. If all solid residues presently carbonized in Kumasi,the area accounting for approximately 60 percent of all residueproduction in Ghana, were converted in Beehive brick kilns an increase incharcoal output of 4,400 tonnes/yr would result. A capital investment ofUS$110,000 would be required with an economic rate of return near 500percent. Three-quarters of the capital requirement is for local costs.

Sawdust Briquetting

7.17 Evaluation of three different capacity screw press sawdustbriquette plants demonstrates that definite economies of scale exist withrespect to this conversion technology. It is estimated that productionin Kumasi at the 14,000 tonnes of briquettes per annum level could beprofitably marketed to bakers and brick factories in the Accra/Temaarea. A plant of this scale would consume just over half the surplussawdust available in Kumasi yet meet only a third of the demand forbriquettes among urban bakers and brick makers alone. Total briquettedemand in this sub-sector is estimated at 45,000 t/yr. Jue to the higherheat value and superior burning characteristics of the briquettes, aminimum of 21,000 t/yr of fuelwood would be saved. Total investment forthe briquette plant is estimated as US$780,000. Under the assumptionthat consumers would be willing to pay a 10 percent premium over theenergy content adjusted price of fuelwood for the briquettes' ease ofhandling and combustion benefits, the investment yields an economic rateof return of !9 percent.

7.18 Costs for transport of sawdust briquettes to the Accra/Temapoints of coasumption exceed the briquette production cost in the largest

- 134 -

size plant investigated. Only road transport has been considered in theproject evaluation. However, when the Central railroad line isrehabilitated, transport costs could drop by perhaps 25 percent. Theresulting drop in the delivered cost in Accra/Tema of nearlyUS$5.00/tonne would substantially improve the economics of sawdustbriquetting.

Briquette Carbonization

7.19 A carbonization tunnel appended to the last stage of thesawdust briquette production line could carbonize the briquettes atrelatively low cost. Charcoal FOB export value of US$115/tonne andTakoradi market prices exceeding this figure suggest that a viable marketexists for briquette charcoal. However, the expected charcoal yield of32 percent or less in a commercial process means that the charcoalbriquette production cost would be a minimum of three times the sawdustbriquette cost of US$32/tonne in small scale plants. As the suggestedcarbonization process is experimental, investments in ciharcoal briquetteproduction are not proposed at this time. Follow-up investigations arerequired to define the technology and economics.

National Investment Implications

Total Investment Potential

7.20 If capital is not considered to be a constraint, the decisionrule is to accept all projects giving a rate of return greater than theopportunity cost of capital of 10 percent assumed in this study.Investments meeting this criteria are sumarized in Table 7.3. In viewof the low levels of capacity utilization presently found in Ghanaianindustry, rehabilitation investments will be found the most profitable.

Table 7.3: TOTAL INVESTMENT POTENTIAL FOR INCREASED AND/OR IMPROVEDUTILIZATION OF WOOD INDUSTRY RESIDUES

Investment Annual ResidueInvestment Location Amount EIRR WV ConsumptionSeamiII USS7.4 MPromess National at present wood 46% USS20.4 M Up to 80,000 m3Heat Industry output

SawdustBriquetting Kumasi US$0.78 M 19% US$0.50 M 27,000 m3Plant

ImprovedResidues National a/ up to USS0.11 N 490% USS1.11 M 64,0OO m3Carbonization

a/ Amounts given are for investment for improved utilization of all residuespresently carbonized in Kumasi.

- 135 -

Competitition for Residue Resources

7.21 Schemes for residue utilization are mutually exclusive to theextent that they compete for the same residue resource in terms of typeand location. The primary competition for residue resources would occurfor sawdjst in the Kumasi area. Total sawdust production in Kumasi of56,400 m , almost all of which is presently in surplus, wuuld not besufficient to provide for the consumption in both maximum sawmill processheat utilization and sawdust briquetting unless solid residues weresubstituted in the first application.

7.22 In order to provide a tool for prioritizing residue resourceutilization, a net benefits/scarce resources ratio was computed for anumber of alternative residue uses. As explained in Squire and Van derTak, the B/R ratio is a transformation of the NPV criterion and isanlogous to the (Benefits - Operating Costs)/(Capital Cost) ratio usedfor project ranking whea there is a capital budget constraint(Gittinger's Net Benefit-Investment Ratio). If the denominator isexpressed in the value of consumption numeraire, a unitless ratioresults. Conparisons can be made more immediately recognizable if theratios are computed per tonne of residue as is shown in Table 7.4. Inthis case, the values represent breakeven residue worth, or thewillingness to pay at which the residue utilization investment has a zeroup,.

Table 7.4: NET BENEFIT/RESIDUE RESOURCE RATIO FOR ENERGY USES IN KUMASI(S/tcmqe) a/

on-SiteBoller Charcoal c/ Briquette dVFuel bV Firewood Feedstock Feedstock

Financial

Solid Residues 23 (36) 3 10 0

Sawdust 23 (36) 0 Not Examined 3

Economic

Solid Residues 21 (41) 8 18 0

Sawdust 21 (41) 0 Not Examined 3

a/ Residue Inputs have been normalIzed to green tonnes,b/ Figures In parentheses are breakeven values derived from NPV analysis of sawmill

prOCess heat model. Figures to the left are based on breakeven values with RFO on anet energy equlvalent basis.

c/ Assumes use In the highest yielding alternative charcoal production method, I.e.eehIve brick kiln, operating at Kumasl sawmills.

d/ Based on Kumasl-B plant marketing production In Accra/Teo.m

- 136 -

Residue Utilization Priorities

7.23 Interpretation of Table 7.4 gives the following priorities forthe utilization of residues for energy purposes in cases where there arealternative uses for the same residue:

Table 7.5: RESIDUE ENERGY UTILIZATION PRIORITIES

PriorityRanking Utilizatlon

1 Combust sawdust on-site for process heatgeneratlon/cogeneration

2 Combust solid residues on-sIte forprocess heat generatlon/cogeneration

3 Convert solid residues to charcoal Inefficient kilns

4 Convert sawdust to briquettes

5 Utilize solid residues as firewood

7.24 The following conclusions are drawn from the above ranking:

(a) Sawdust should be used in sawmill boilers to the extentdetermined by demand for process heat and technical feasibilityof sawdust combustion. Solid residues should be used to meetthe balance of heat demand.

(b) Remaining sawdust should be briquetted if it exists in amountssufficient to allow briquetting on an economic scale; such asituation exists in Kumasi.

(c) Remaining solid residues will be most profitably used ifconverted in efficient charcoal kilns such as the Beehive brickkiln.

The effect of these prescriptions is illustrated in Table 7.6, whichgives a sample "before and after' residue utilization pattern. Thereasonableness of the underlying assumptions needs to be tested throughexamination of the export market for dried lumber and the existing dryingcapacity.

- 137 -

Table 7.6s RESIDUE UTILIZATION PROFILES FOR KUMASI('000 m3 SWE)

VeneerSolids Waste Sawdust

1986 Pattern

Mill Boiler Fuel 25.1 28.4 3.4

Charcoal Production 63.6 - -

Firewood 42.4

Non-Energy 62.2 - 0.4

Surplus -52.6

193.3 28.4 56.4

Potential Future Pattern

Mill Boller Fuel a/ 33.1 28.4 29.0

Charcoal Production b-

Firewood

Briquetting - - 27.0

Non-Energy b- 0.4

Surplus - _

I/ Assumptions:70S of export lumber from Kumasi.60% of export lumber dried.75% of new process heat demand met by sawdust.

b/ Charcoal/Non-Energy split not profiled.

7.25 Non-energy uses for residues were briefly reviewed in Chapter4. The basic conclusion is that non-anergy uses for solid residues suchas re-manufacture for export or cottage industry use should be consideredof higher value than energy uses. Prospects for utilization of largeamounts of sawdust for the production of fiber board are not favorabledue to the present world market situation for this product, and non-energy uses for sawdust have not entered into the estimations.

- 138 -

Uncertainties and Risk Factors

Health of the Wood Processing Industry

7.26 West Africa held 30 percent of the world tropical hardwoodtrade in the early 1960s, of which Ghana was a major player. Thedownward slide which characterized the Ghanaian wood products industry inthe period 1975-1983 has reduced Ghana's role to one of relativeinsignificance on a world scale; at the same time, upward potentialexists through recapture of lost export markets. While the presentindustry recovery is vigorous, a serious downturn could jeopardize thewood residue utilization investments planned on the basis of a healthylevel of wood processing activity. An investment in a sawdustbriquetting plant, as a case in point, might cease to be viable ifresidue production were to fall off drastically. A preference for exportof logs instead of products would similarly reduce residue production,but with negative implications for domestic income and employment.

Location of Wood Processing Facilities

7.27 The location of the wood resource base has shifted over time asthe most easily exploitable areas are logged. Coupled with the on-goinggrid extension program and transport network improvements, there is someincentive for wood processing industries to locate in areas closer to thetimber resource. Residue production would become less physicallyconcentrated in areas like Kumasi, although the changes would occur onlyslowly over a long time and would not be expected to affect the basicviability of investments.

Wood Supply/Demand

7.28 The sustainability of wood production for industrial and fueluses and its balance with demand have been taken as given in most of theresidue utilization analyses. Increasing wood scarcity would have theeffect of raising the importance of obtaining value-added for woodproducts, carbonizing wood residues efficiently and sawdust briquetting,and thereby raise the returns to these activities. Investment viabilitycould be threatened to the extent that wood resource shortages lead to adecline in residue availability.

Oil Prices

7.29 The assumed downside potential for petroleum prices is not sogreat as to make fuel oil more economic than on-site wood residues forsawmill process heat raising. On the opposite side, a rise in oil pricesincreases the value of wood residues as a fuel to the extent that woodand petroleum are substitutable fuels.

- 139 -

Electricity Costs

7.30 Electricity costs have been assumed to increase in line withthe general rate of inflation as per the Acres generation planningstudy. However, long run marginal costs can be expected to evolve inrelation to demand growth as the current hydro capacity becomesstrained. More or less rapid electricity demand growth will advance orretard, respectively, the timeframe in which residue-fired cogenerationmay become competitive with other electricity sources as discussed inparagi ;ph 7.11.

Investment Recommendations

Sawmill Process Heat

7.31 Priority investment attention should be focused on theutilization of sawmill residues for on-site generation of process heatfor lumber drying and treatment. Complementary efforts will be requiredto determine:

(a) target export markets for treated and kiln-dried lumber andneeded promotion efforts;

(b) species demand and drying requirements;

(c) quality control requirements;

(d) packaging and shipping infrastructure requirements; and

(e) marketing structures through which output from smallerexporters can be passed through larger mills for furtherprocessing.

The Timber Export Development Board should serve as the coordinating bodyfor these information gathering efforts.

Sawdust Briguetting

7.32 Second priority investment consideration should be given to a14,000 t/yr sawdust briquetting plant in Kumasi. It is anticipated thatfinancing could be raised from private sector sources. Investment shouldproceed only in conjunction with:

(a) assurances for long-term sawdust supply from cooperating mills;and

(b) briquette market promotion efforts.

- 140 -

The briquette marketing effort could be undertaken with the assistance ofa qualified local consulting organization engaged to perform asubstantial portion of the market survey and evaluation work.

Technical Assistance Recommendations

7.33 Future investments in sawmill boiler/kiln capacity expansionshould incorporate the following technical assistance components in orderto maximize effective residues utilization:

(a) Saw blade guide improvements;

(b) Residues handling and storage improvements;

(c) Furnace modifications for sawdust combustion; and

(d) Boiler efficiency improvements.

These measures will maximize the utility of sawdust as a boiler fuel,thus freeing solid residues for other energy or non-energy purposes.Boiler efficiency improvements, among others, at non-wood processingsites are being addressed under the on-going ESMAP Industrial EnergyEfficiency Improvement project. Coordination of these efforts with anysawmill energy efficiency upgrades is required.

Pilot/Demonstration Projects

Improved Solid Residues Carbonization

7.34 A component in improved charcoal making from solid sawmillresidues is recommended for inclusion in the proposed ESMAP GhanaCharcoal Production Improvement project. Kumasi is a highly suitablearea for the introduction of efficient Beehive brick kilns. Such kilnscan be constructed either at or adjacent to the sawmills using locallyavailable materials and can be operated by charcoalers retained by thesawmill. Alternatively, private developers could be awarded exclusivecontracts for the purchase and carbonization of the surplus solid residueoutput at each sawmill. The main ESMAP input will be to provideassistance to determine the appropriate organizational structure,identify interested sawmill operators or local entrepreneurs, and draftsuitable agreements. Specifically the project component will:

(a) Provide and present detailed information as to the expectedfinancial returns to the adoption of improved residuecarbonization regimes, including capital, material, andtraining requirements and charcoal yields;

- 141 -

(b) Survey and evaluate local sawmill operators and entrepreneursfor their interest and capabilities in improved residuecharcoal production;

(c) Demonstrate the construction and operation of the brick(Beehive) kiln on or adjacent to a suitable sawmill, andcompare with the present method;

(4) Determine the most viable organization of production, i.e.either directly managed by sawmill operators or byentrepreneurs;

(e) If the results of (b)-(t) indicate contractual arrangements asbeing preferable, assist in the drafting, negotiation andfinalization of fixed term renewable contracts for the purchaseand utilization of sawmill wastes;

(f) Provide technical details as to design, construction andoperation of the brick (Beehive) kilns;

(g) Develop a cadre of trained personnel, possibly at FPRI, capableof assisting and training others in the construction, operationand maintenance of the brick kilns; and

(h) Identify follow-on training requirements and financing needs.

Briquette Carbonization

7.35 Depending on donor interest, the technology and economics ofproducing charcoal from sawdust briquettes could be investigated througha pilot/demonstration project. The existing briquette factory in Oda issuggested as a demonstration site. The Chaowus plant owner has alreadybegun examining available processes and machinery for briquettecarbonization, which would have multi-country application if provedfeasible.

Policy Recommendations

7.36 Government policies on residue utilization can have a largeeffect on the ultimate disposition of the waste wood resource. Thepolicy instruments of fuels pricing and tax/subsidy levbls may be used todirect investment into economically desirable forms of residueutilization, while regulation can be applied to prohibit certain wastefulpractices. Lack of information among producers and consumers is anothertype of market failure which can be rectified through governmentintervention. Specific policy recommendations for each of the residueutilizations that have been identified as economically desirable aregiven below. When applicable, cross references to corroborating Bankreport recommendations are indicated in parentheses.

- 142 -

Sawmill Process Heat

7.37 Utilization of residues, especially sawdust, in sawmill processheat applications may be promoted by:

(a) Improved coordination between FPRI and TEDB concerning speciescharacteristics, uses and treatment requirements, andwidespread dissemination of research results to producers andoverseas agents (Ghana Forestry Sector Review).

(b) Increased flow of marketing intelligence from TEDB to producersconcerning treated and kiln-dried product export opportunities.

(c) Establishment of kiln-dried lumber quality control andpackaging guidelines, and provision of port infrastructure forcontainerized cargo handling (Ghana Forestry Sector Review).

(d) Differential taxation on value-added forest products (ChanaForestry Sector Review).

(e) Phased extension of log export ban to larger numbers of species(with attendant increase in log sterilization and productdrying requirements) as they become commercially accepted(Ghana Forestry Sector Review).

(f) Institution of permit/fee systems for sawdust dumping and a banon open sawdust incineration.

Residue Conversion

7.38 Conversion of residues to more economically useful forms may bepromoted by:

Provision of domestic loan financing, which may include theNational Energy Board Energy Fund or the National Board forSmall Scale Industries; in the case of residue charcoalingoperations, loans should only be made to suitably qualified andtrained entrepreneurs.

Areas for Further Investigation

7.39 An investigation into economic options for the utilization oflogging residues is recommended and has been incorporated as a majorcomponent of the proposed ESNAP Ghana Charcoal Production Improvementproject. In detail, the component wills

- 143 -

(a) Develop a method to carbonize waste wood from timber operationsin the forest; this includes a technical and economicassessment, logistic requirements, and timber companies' andGovernment's inputs and requirements;

(b) Identify the manpower requirements to carry out theseactivities, and determine how and to what extent these can bemet by charcoalers from the Transition zone;

(c) Determine the training needs and potential incentives fortraditional charcoalers to start operating in the timberextraction areas;

(d) Determine the regulatory, economic and social requirements tobe met to reduce traditional charcoaling operations in theTransition zone, and enable charcoal production in the timberextraction areas;

(e) Design an appropriate energy pricing and tax policy to support(d);

(f) Prepare a scheduled and costed plan of action.

- 144 -

BIBLIOGRAPHY

1. Ghana: Issues and Options in the Energy Sector, Report No. 6234-GH,World Bank, November 1986.

2. Draft Ghana Forestry Sector Review, West Africa Projects Department,Agriculture Division, World Bank, 1986.

3. "Survey of Wood Residue Generation and Utilization in Ghana," FinalReport, Ru-Tek Consultants and Industries Ltd., Kumasi, Ghana,November 1986.

4. "Ghana Sawmill Residues Utilization Study," Report X7280/1, SandwellSwan Wooster Inc., Vancouver, January 1987.

5. The Forest Department Review and the Requirements of the ForestProducts Inspection Bureau and the Timber Export Development Board(Draft Report), Silviconsult, Bjarred, Sweden, September 1985.

6. "Report on Pilot Survey on Puelwood and Charcoal Consumption inAccra," Government of Ghana National Energy Board, October, 1985.

7. EEC Trade Promotion Project "Ghanaian Timber Marketing Study," P-EInternational Operations Ltd. for Ministry of Finance and EconomicPlanning, London, 1981.

8. Ghana Generation Planning Study, Acres International Ltd.,Toronto, January 1985.

9. "Economic Appraisal Report on Rehabilitation of Central and EasternLines," RITES, New Delhi, October 1986.

10. "Timber Export Development Board Law," P.N.D.C.L. 123, ProvisionalNational Defence Council of Ghana, October 1985.

11. "Ghana Timber Export Market Report," Issue No. 11, Timber ExportDevelopment Board, Takoradi, Ghana, October 1986.

12. "Ghana Hardwoods," Ghana Timber Marketing Board, Takoradi, Ghana.

13. "Report on Export Permits," Forest Products Inspection Bureau,Takoradi, Ghana, October 1986.

14. "Charcoal Production in Steel and Brick Kilns - Technology Transferin Ghana," Building and Road Research Institute, Kumasi, Ghana.

15. Ghana: Towards Structural Adjustment, Report No. 5854-GH, WorldBank, October 1985.

- 145 -

16. Wagenfurh/Scheiber Holzatlas, Veb Fachbuchverlag, Leipzig, 1974.

17. "Production of Bricks by Hand Moulding Method at the Asakwa BrickFactory," Building and Road Research Institute, Kumasi, Ghana.

18. "3000 kW Power Plant Fired by Wood Waste: Preliminary Planning forMim Timber Co. Ltd.," Spilling Consult AG, Wohlen, Switzerland,June 1977.

19. "Charcoal Production in Developing Countries," SwedforestConsulting AB for SIDA, Stockholm, March 1983.

20. Charcoal Making in Developing Countries, Technical Report No. 5,Karthscan/IIED9 London, 1986.

t~~ ~ ~~~~~~~~~~~~ S10N .M$O ' sz

B U KINA F ASO~ --

L...oo R e g io A n Up pr W es t

( Brong e g A o Re

Bi Looi- ,-W

< \ ~~N o r t h e r Reg n t,f

C O T E 2 Xi$$

D' I V 0 1 R E

t ~~~~~~WOOD PROCESSING ANC~ OTHER- EKaN -EO~O OS INDUSTRIAL CENTERS

.win ~ T - 0 Gr0

-rs WOse D _0O04 D

.win~~~~~~~~~~~~~.

r~~~~~~G o> n gne

. ~~~~~~~~~~0 00 D0 00 , 00 100 TG

| 5. 2* / . r , i c

Activities Completed

Country Project Date Number

Energy Effieiency end Strategy

Africa Regional Participants' Reports - Regional Power Seminaron Reducing Electric System Losses in Africa 8B88 087188

8angladesh Power System Efficiency Study 2185 031/85Sotsvna Pump Electrification Prefeasibility Study 1/86 047/86

Review of Electricity Service Connection Policy 7t87 071/87Tuli Block Farms Electrification

Prefeaibility Study 7/87 072/87Burkina Technical Assistance Program 3/86 052/86Burundi Presentation of Energy Projeuts for the

tourtb Five-Year Plan (1983-1987) 5185 036/85Review of Petroleum Import and Distribution

Arrangements 1/84 012/84Costa Rica Recmeended Technical Assistance Projects 11184 027/84Ethiopia Power System Efficiency Study 10/85 045/85The Gambia Petrol -j Supply Management Assistace 4/85 035/85Chana Energy Rationalization in the Industrial

ectct of Ghana 6/88 084/88Guinea- Recommended Technical Assistance

8! sau Projects in the Electric Power Sector 4/8S 033/85Indonesia Energy Efficiency Improvement in the Brick,

Tile and Lime Industries on Java 4/87 067/87Power Generation Efficiency Study 2/86 050/86

Jamaica Petroleum Procurement, Refining, andDistribution 11/86 061/86

Kenya Power System Efficiency Report 3184 014/84Liberia Power System Efficiency Study 12/87 081/87

Recommended Technical Assistance Projects 6/85 038/85Madagasear Power System Efficiency Study 12/87 075/87Malaysia Sabah Power System Efficiency Study 3/87 068/87Mauritius Power System B ficiency Study 5/87 070/87Panama Power System Loss Reduction Study 6/83 004/83Papuea new nergy Sector Institutional Review: Proposals

Gunea for Strengthening the Department ofMinerals and Energy 10/84 023/84

Power Tariff Study 10/84 024/84Senegal Assistance Given for Preparation of Documents

for Energy Sector Donors' leeting 4/86 056/86Seychelles Electric Power System Efficiency Study 8/84 021/84Sri Lanka Power System Loss Reduction Study 7/83 007/83Syria Electric Power Efficiency Study 9/88 089/88Sudan Power System Efficiency Study 6/84 018/84

Managemen Assistance to the Miaistry ofEegy and wining 5/83 003/83

Togo Power System Efficiency Study 12/87 078/87Wood Recovery in the Nangbeto Lake 4/86 055/06

Uganda Energy Efficiency in Tobacco Curing Industry 2/86 049/86Institutional Strengthening in the Energy Sector 1/85 029/85

Zambia energy Sector Institutional Review 11/86 060/86Zimbabwe Power Sector Hanagement Assistance Projects

Background, Objectives, and ork Plan 4/85 034/85Power Systpu Loss Reduction Study 6/83 005/83

Household, Rural ,ad Renevable Energysurundi Peat Utiization Project 11/85 046/85

Improved Charcoal Cookstove Strategy 9/85 042/85Cote Improved Siomasa Utilization-Pilot Projects

dlIvoire Using Agro-Industrial Residues 4/87 069/87Ethiopia Agricultural Residue Briquettings Pilot Project 12/86 062/86

Bagasse Study 12/86 063/86The Cambia solar Water seating Retrofit Project 2/85 030/85

Solar Photovoltaic Applications 3/85 032/85Clobal Proceedings of the ESMP Eastern & Southern Africa

Household Energy Planning Seminar 6/U8 085/88Jamaica FIDC Sawmill Residues Utilization Study 9/88 088

Charcoal Production Project 9/88 090/88Kenya Solar Water Heating Study 2/87 It6/87

Urban woodfuel Developeent 10/87 076t87Malawi Technical Assistance to Improve the Efficiency

of Fuelwood Use In the Tobacco Industry 11/83 009/83Mauritlus Bsagasse Power Pc Atial 10/87 077/87Niger Household Energy Conservation and Substitution 12/87 0o2/87

Improved Stoves Project 12/87 080/87Peru Proposal for a Stove Dissemtnation Program

in the Sierra 2/87 064/87Rvanda Improved Charcoal Cookstove Strategy 8/86 059/86

Improved Charcoal Production Tecbaiques 2/87 065/87Senegal Industrial Energy Conservation Project 6/85 037/85Sri Laoka Industrial Energy Conservation: Fea6ibility

Studies for Selected Industries 3/86 054/86Sudan Wood Energy/Forestry Project 4/88 073/88Tanzania Woodfuel/Forustry Project 8a88 086/88Thailand Accelerated Dissemination of Im;roved Stoves

and Charcoal Kilns 9/87 079/87Rural Energy Issues and Options 9/85 044/85Northeast Region Village Forestry and Woodfuel

Pre-Investment Study 2/88 083/88Uganda Fuelvood/Forestry Feasibility Study 3t86 053/86