PERFORMANCECHARACTERISTICSANDDESIGN

RECOMMENDATIONSFORBIOMASSBURNINGSTOVESUSING
EARTHENCONSTRUCTIONMATERIALS

By
NicholasH.Schreiner

AREPORT

Submittedinpartialfulfillmentoftherequirementsforthedegreeof

MASTEROFSCIENCE

EnvironmentalEngineering

MICHIGANTECHNOLOGICALUNIVERSITY
2011

2011NicholasH.Schreiner


Thisreport,PerformanceCharacteristicsandDesignRecommendationsforBiomassBurning
StovesUsingEarthenConstructionMaterials,isherebyapprovedinpartialfulfillmentofthe
requirementsforthedegreeofMasterofScienceinEnvironmentalEngineering.

DEPARTMENT:

CivilandEnvironmentalEngineering

Signatures:

ReportAdvisor: _______________________________________

Dr.KurtisPaterson

DepartmentChair: _______________________________________

Dr.WilliamBulleit

Date: _______________________________________

ii

 TABLEOFCONTENTS

1 INTRODUCTION.....................................................................................................................................1
1.1 PROJECTDEVELOPMENT...............................................................................................................2
1.2 OBJECTIVES...................................................................................................................................4
2 PROJECTBACKGROUND........................................................................................................................5
2.1 STOVETHEORY..............................................................................................................................5
2.2 STOVEMATERIALS........................................................................................................................5
2.3 ROCKETSTOVES............................................................................................................................6
2.4 GTZUGANDASTOVES...................................................................................................................7
3 METHODS............................................................................................................................................10
3.1 USERSURVEYS.............................................................................................................................10
3.2 STOVECONSTRUCTION...............................................................................................................12
3.3 STOVEPERFORMANCETESTING.................................................................................................13
3.4 STOVETHERMALMODELING......................................................................................................14
3.5 MATERIALSTESTING...................................................................................................................15
3.5.1 Specificheat........................................................................................................................15
3.5.2 Thermalconductivity...........................................................................................................15
3.5.3 Compressionstrength.........................................................................................................17
3.6 STOVEEMISSIONSTESTING........................................................................................................17
4 RESULTSANDDISCUSSION..................................................................................................................18
4.1 USERSURVEYRESPONSES...........................................................................................................18
4.2 STOVEPERFORMANCECHARACTERISTICS..................................................................................18
4.3 STOVEBODYHEATLOSSANDTHERMALDISTRIBUTION............................................................20
4.4 MATERIALPROPERTIES...............................................................................................................23
4.5 STOVEEMISSIONS.......................................................................................................................23
5 DESIGNSUGGESTIONS........................................................................................................................28
6 CONCLUSIONS.....................................................................................................................................32
7 FUTUREWORK....................................................................................................................................33
8 REFERENCES........................................................................................................................................34
9 APPENDICES........................................................................................................................................36
A. FINITEDIFFERENCEMODELMATLABCODE....................................................................................36

iii

 B. DIGITALFILESONATTACHEDCD....................................................................................................38
C. HUMANSUBJECTSRESEARCHAPPROVAL:MICHIGANTECHNOLOGICALUNIVERSITY..................39

iv

 LISTOFFIGURES

Figure11Interventionopportunities...................................................................................................1
Figure12ProjectSite:MauritaniaandMali,SahelRegionofWestAfrica(Source:adaptedfrom
MillenniumEcosystemAssessment)..............................................................................................2
Figure13StoveconstructioninMali(photobyauthor)........................................................................3
Figure21RocketStove(source:AprovechoInstitute)...........................................................................7
Figure22ShieldedFireStove(source:adaptedfromMinistryofEnergyandMineralDevelopment,
RepublicofUgandaandGTZ2004)................................................................................................8
Figure23RocketLorenaStove(source:adaptedfromMinistryofEnergyandMineralDevelopment,
RepublicofUgandaandGTZ2004)................................................................................................8
Figure24Twopotshieldedfirerocketstove........................................................................................9
Figure31Schematicofshieldedfirerocketstoveconstructedforlaboratoryperformanceandemissions
testing(Alldimensionsincentimeters)........................................................................................12
Figure32ShieldedfirestoveduringaWaterBoilingTestfor1Lofwaterina30cmdiameterpot......13
Figure33Thermocoupleplacementandcomputationalelementnodesinthestovebody..................14
Figure34Conductiveheattransferthroughaseriesofplanarobjectsofvaryingthermalconductivity 16
Figure35Emissionssamplingsetupduringstovetesting;instrumentsamplinginletswerepositioned1
mabovepotatopstove..............................................................................................................17
Figure41Stovebodytemperaturedistribution(WaterBoilingTest:Run1)........................................21
Figure42Stovebodytemperaturedistribution(WaterBoilingTest:Run2)........................................22
Figure43AveragesofCO2concentrationofshieldedfirestoveemissionsoverthreecompleterunsof
theWaterBoilingTest(WBT).Errorsbarsrepresentonestandarddeviationaboveandbelowmean
values.........................................................................................................................................24
Figure44AveragesofCOconcentrationofshieldedfirestoveemissionsoverthreecompleterunsof
theWaterBoilingTest(WBT).Errorsbarsrepresentonestandarddeviationaboveandbelowmean
values.........................................................................................................................................24
Figure45AveragesofPM2.5concentrationofshieldedfirestoveemissionsoverthreecompleteruns
oftheWaterBoilingTest(WBT).Errorsbarsrepresentonestandarddeviationaboveandbelow
meanvalues................................................................................................................................25
Figure46MovingaverageofPM2.5concentrationofthreestonefireemissionsduringonecoldstart
test.Errorsbarsrepresentonestandarddeviationaboveandbelowmeanvalues........................25
Figure47MovingaverageofPM2.5concentrationofthreestonefireemissionsduringonecoldstart
test.Errorsbarsrepresentonestandarddeviationaboveandbelowmeanvalues........................26
Figure51Stoveinletandcombustionchamberdimensions(source:adaptedfromAprovechoInstitute)
...................................................................................................................................................29

Figure52Potskirtthicknessandstovebodydiameter.......................................................................30
Figure53RechoRoketStove(source:http://www.rechoroket.com/).................................................31

vi

 LISTOFTABLES
Table11WHOHealthStatisticsforMauritaniaandMali(2008)(WHOGlobalHealthObservatory
Database).....................................................................................................................................3
Table31Cookingpracticesandimprovedstovesurvey.....................................................................11
Table41Shieldedfirewaterboilingtestresults.TheWaterBoilingTestwasconductedinthreephases,
highpowercoldstart,highpowerhotstart,andsimmertoevaluatethestoveundervarying
operatingconditions.Thedurationofthefirsttwophaseswasdependentonthetimeneededto
bring2.5litersofwatertoboil.Thethird,lowpowerphasewas45minutesinduration.During
thistimewaterwaskeptjustbelowboilingtosimulateslowcookingoperations.Valuesshownas
temperaturecorrectedareadjustedtoastandard75Ctemperaturechange(25to100)...........19
Table42Comparisonofshieldedfirestoveandthreestovefireperformanceresults..........................20
Table43Heatlosstostovebodyrelativetototalenergyinput...........................................................22
Table44Mudmixturematerialproperties.........................................................................................23
Table45Thermalpropertiesofcommoncombustionchambermaterials(Andreatta2003)................23
Table51Inletandcombustionchamberdimensionsbasedoncookpotdiameter.Distheinlet
diameter,Histheheightofthecombustionchamber,Kisthedistancefromthetopofthefuelinlet
totheoutletofthecombustionchamber,andListhepotskirtgap(modifiedfromMinistryof
EnergyandMineralDevelopment,RepublicofUgandaandGTZ2004).........................................29
Table52Summarycomparisonofcookstovedesignimprovements...................................................31

vii

 ABSTRACT

Thisreportprovidesananalysisofthethermalperformanceandemissionscharacteristicsofimproved
biomassstovesconstructedusingearthenmaterials.Commonlyreferredtoasmudstoves,thistypeof
improvedstoveincorporateshighclaycontentsoilwithanorganicbinderintheconstructionofits
combustionchamberandbody.Whenlargequantitiesofthemudmaterialareusedtoconstructthe
stovebody,thestovedoesnotoffersignificantimprovementsinfueleconomyorairqualityrelativeto
traditionalopenfirecooking.Thisispartlybecauseasignificantamountofheatisabsorbedbythemass
ofthestovereducingcombustionefficiencyandheattransfertothecookpot.

Ananalysisofthethermalandmechanicalpropertiesofstovematerialswasalsoperformed.Amaterial
mixturecontainingaonetooneratiobyvolumeofhighcontentclaysoilandstrawwasfoundtohave
thermalpropertiescomparabletofiredceramicsusedinmoreadvancedimprovedstovedesigns.

FeedbackfrommudstoveusersinMauritaniaandMali,WestAfricawasalsocollectedduring
implementation.Suggestionsforstovedesignimprovementsweredevelopedbasedonthisinformation
andthedatacollectedintheperformance,emissions,andmaterialpropertiesanalysis.Design
suggestionsincludereducingstoveheighttoaccommodateusercookingpreferencesandlimiting
overallstovemasstoreduceheatlosstothestovebody.

viii

 1 INTRODUCTION

Threebillionpeople,approximatelyhalftheworldspopulation,dependonbiomassfuels(wood,crop
residue,animaldung,etc)astheirprimarydomesticenergysource(WHO2002).Often,thecombustion
ofthesefuelsisdoneindoorsoveropenfireswithlittleornomeansofventilation.Extendedexposure
tobiomassemissionssignificantlyincreasestheriskofacuterespiratoryinfections,chronicbronchitis,
andobstructivepulmonarydisease(Bruce,PerezPadillaetal.2002).TheWorldHealthOrganization
(2002)estimatesthatpoorindoorairqualityassociatedwithbiomassfuelcombustionisresponsiblefor
2.7%oftheglobalburdenofdisease.Womenandchildrenareespeciallyvulnerabletothehealthrisks
associatedwithbiomassfuelcombustionandpoorindoorairquality.Inmanypartsoftheworld,
traditionalgenderrolesassigntheresponsibilityofmealpreparationandchildcaretowomenthus
disproportionallyexposingbothtoharmfulemissionsduringcookingactivities.Womenandchildren
comprisethemajorityofthe1.6millionannualdeathsattributedtoindoorairpollution.Itisestimated
that56%ofthistotalarechildrenundertheageoffive(WarwickandDoig2004).

Inresponse,agrowingnumberofgovernmentalandnongovernmentalorganizationshaveparticipated
ininitiativestoimproveindoorairqualityandreducebiomassfuelconsumption.Interventionsinclude
improvedventilation,cleanerandmoreefficientfuelsources,andimprovedstovesthatincreasefuel
economyandreduceemissions.ExamplesofpotentialinterventionopportunitiesareshowninFigure
11.

Figure11Interventionopportunities

Stovedisseminationprogramshavebeenmetwithvaryinglevelsofsuccess.Asistruewithany
developmentinitiative,thesustainabilityofanimprovedstoveprojectisnotdrivenbythetechnology
alone.Social,cultural,andeconomicfactorshaveasignificanteffectonstoveuseandadoptionrates.
Themostsuccessfulimprovedstovesareeasytoconstructinlocalsettingsusingexistingtechniquesand
materialsandhaveclearadvantageswithrespecttofueleconomy,easeofuse,durability,and
cleanliness(Barnes1994).

Mudstovesareoneexampleofimprovedstovedesign.Theycombinehighclaycontentsoilwithsand
oranorganicbindertoformastovebody.Stovematerialsareoflittleornocostandrequirefew
specializedtoolsorskillstoconstruct.Thismakesthemattractiveinareaswherethereislittleeconomic

incentivetoadoptanimprovedstove.However,thesetypesofstovesareoftencriticizedforthepoor
thermalandmechanicalpropertiesofthematerialsusedandthenegativeeffectthiscanhaveonstove
performanceanddurability.

Thisstudyoffersananalysisofearthenmaterialsandtheiruseinimprovedstroveconstructionthrough
performancetestingmudstovedesignandadetailedanalysisofmaterialproperties.

1.1 PROJECTDEVELOPMENT

ThestudyofimprovedmudstovesdevelopedduringthetwoyearstheauthorspentasaU.S.Peace
CorpsvolunteerinMauritaniaandMali,WestAfrica.TheauthorservedinMauritaniafrom2008to
2009beforetheprogramwassuspendedduetosecurityconcerns.Hespenttheremainderofhis
serviceinneighboringMalifrom2009to2010.BothcountriesareinAfricasaridSahelregiontothe
southoftheSaharadesert.RelevantnationalstatisticsforbothMauritaniaandMaliaregiveninTable
11.

TheprojectbeganinMauritaniaasanattempttoreducestressonlocalfirewoodsourcesthrough
increasedhouseholdfueleconomy.InMauritania,mostcookingactivitiesareperformedoutsideand
indoorairqualitywasnotasignificantconcern.InMali,however,alargerproportionofcookingisdone
indoorsespeciallyduringthecoldandrainyreasons.Inthissetting,increasedemphasiswasputon
improvingindoorairqualityinadditiontofuelefficiency.

Figure12ProjectSite:MauritaniaandMali,SahelRegionofWestAfrica(Source:adaptedfromMillennium
EcosystemAssessment)

 Table11WHOHealthStatisticsforMauritaniaandMali(2008)(WHOGlobalHealthObservatoryDatabase)

Mauritania Mali
TotalPoulation 3,044,000 11,968,000
Grossnationalincomepercapita($) 1,970 1,000
LifeExpectancy(male/female)(years) 55/60 45/48
Under-five mortality rate (probability of dying by age 5 per
118 194
1000 live births)
Children aged less than 5 years with symptoms of Acute
Respiratory Infection taken to a health facility (%)
44.5 38.1(yr.2006)

Communitymembersinbothcountriesshowedlittleinterestinpurchasingmoreadvancedmetalor
ceramicstoves.Theywere,however,willingtogathermaterialsandassistintheconstructionofmud
stoves.Stoveswereconstructedonahousehold,basisusingdesignsdevelopedinUgandabythethe
GermanTechnicalCooperation(GTZ)andtheUgandanMinistryofEnergyandMineralDevelopment
(UgandaGTZ2004).

Figure13StoveconstructioninMali(photobyauthor)

1.2 OBJECTIVES

Aqualitativeassessmentofstovedesignandperformance,basedonsimpleobservationsinthefield,as
wellasfeedbackfromusers,promptedthismoredetailedanalysis.Withabetterunderstandingofthe
capabilitiesofmudstovesandthesuitabilityofearthenmaterialsinstoveconstruction,development
workersinthisregionofWestAfricaandelsewherewhoareinterestedinaddressingindoorairquality
andhouseholdenergyusecanmakeaninformeddecisiononwhetheramudstoveisanappropriate
interventiontool.Theobjectivesofthisanalysisarearethefollowing:

Objective1:Evaluatemudstoveperformancewithrespecttothermalefficiencyandemissions.

Objective2:Evaluatethemechanicalandthermalpropertiesofmudstovematerialmixtures.

Objective3:Providesuggestionsforimprovementstomudstovedesignusingexperimentaldataaswell
asfeedbackfromstoveusers.

 2 PROJECTBACKGROUND

Thischapterprovidesdetailsonthescientificbasisofbiomassburningstoves,severalstovedesigns
usedinthedevelopingworld,andmaterialsused.

2.1 STOVETHEORY

Improvedstovesaredesignedtoincreasefuelefficiencyandreducesmokeandharmfulemissions
associatedwiththecombustionofbiomassfuels.Thisisaccomplishedthroughimprovedthermal
efficiency,acombinationofenhancedcombustionefficiencyandheattransfer(Baldwin1987).
Combustionefficiencyistheratioofenergyreleasedduringcombustiontothetotalcombustibleenergy
potentialofthefuel.Thoughpoorcombustionwillnegativelyaffectastovesthermalperformance,itis
notassignificantafactorasheattransferefficiency.Energylossesduetoinefficientorincomplete
combustiontypicallyaccountforlessthan8%oftotalenergyinput(Baldwin1987).However,the
incompletecombustionofbiomassfuelswillproduceharmfulemissions.Theseincludecarbon
monoxide(CO),unburnedhydrocarbons(UHC),nitrogenoxides(NOx),smokeandsoot(Ndiema,
Mpendazoeetal.1998).

Improvedstovedesignsseektoincreasecombustionefficiencythoughanumberofdesignfeatures.
Adequatedraftandaproperairtofuelratioensureamorecompleteandefficientcombustionprocess.
Carefulconsiderationofthesize,geometry,andplacementofthefuelinletandcombustionchamber
providesincreasedcontrolofairflow.Ahotterfireisalsomoreeffectiveatconsumingcombustible
gases.Properinsulationaroundthecombustionchamberreducesheatloss,allowingthefiretoburnat
highertemperatures(Bryden,Stilletal.2005).

Stoveheattransferefficiencydescribestheamountofenergyabsorbedbythecookpotrelativetothe
amountofenergyreleasedbythecombustionprocess.Whencarefullycontrolled,evenanopenfireis
capableofhighcombustionefficiency(Bussmann1988).However,inanopenfire,radiationisthemain
mechanismofheattransfer,andapproximately2040%ofitsinputenergyislosttotheatmosphereas
hotgasemissions.Utilizingthesehotgasesthrougheffectiveconvectiveheattransfercansignificantly
improvethermalefficiency(Baldwin1987).

Inanimprovedstoveconvectionistheprimarymethodofenergytransferbetweenthemoving
combustiongasesandthestationarysurfaceofthecookpot.Convectionconsistsofenergytransfer
thoughbulkfluidmotion,advection,andthearbitrarymotionoffluidmolecules,conduction.Advection
isthemoreeffectivemechanismofenergytransferandisproportionaltothefluidvelocity(Incropera
2007).Therefore,designfeaturesofimprovedstovesseektomaximizethevelocityofthecombustion
gasesaswellasthepotsurfaceareaincontactwiththesegases.

2.2 STOVEMATERIALS

Asignificantdisadvantageofanimprovedstoverelativetoanopencookfireisthebodyofthestove
itself.Regardlessofitsmaterialcomposition,thebodyofanimprovedstoveprovidesathermalmass
thatwillconductandstoreenergyfromthefirethatmightotherwisebetransferredtothecookpot.
Energyconductionintothestovebodyaccountsfor1442%ofinputenergy,thelargestofstoveenergy

losses(Baldwin1987).Selectingpropermaterialswhileminimizingtheoverallmassofthestovecan
reducetheamountofenergylosttothestovebody.

Thematerialthermalpropertiesrelevanttostovedesignaredensity(),thermalconductivity(k),and
specificheat(cp).Thermalconductivitydescribestheabilityofamaterialtotransferenergythrough
thermaldiffusion.Specificheatorheatcapacityistheabilityofamaterialtostorethermalenergyand
istypicallyproportionaltomaterialdensity(Incropera2007).Understeadystateconditions,thick,
densestovebodiesoflowthermalconductivitybuthighspecificheatwillconductlessenergyfromafire
thanathinner,lessconductivebody.However,itisunlikelythiswillcompensateforthelargeamount
ofenergyabsorbedwhilethebodywarms.Itisthereforeadvisabletousethin,lowdensitymaterialsin
stoveconstruction(Baldwin1987).

Metals,althoughrelativelylightweightanddurable,aresoconductivetheycanmakelightingand
maintainingfiresdifficult.Advancedstoveswillincorporateinsulativeceramics.Theuseoforganic
bindersandahightemperaturefiringprocess,producealightweight,porousmaterial.Theselow
densitymaterialshavealowheatcapacityandthermalconductivitybecauseofthesmallvoidsleftin
thematerialstructurebythefiringprocess(Still,Pinnelletal.2003)

Earthenmaterials,clay,sand,androck,arenotidealforimprovedstoveconstruction.Thesematerials
aredenseandhaveahighheatcapacity.Usingthemaspartofastovebodyislikelytodivertenergy
awayfromthefireandthecookpot(StillandRoth2010).However,earthenmaterialsarelowcostand
readilyavailableinmostareas.Usingthesematerialsinimprovedstovesseldomrequiresspecialized
toolsorcomplexconstructionprocesses.Simplemodificationstoearthenmaterialmixturescan
improvetheirmechanicalandthermalproperties,andmakethemamoresuitableoptionforstove
construction.

TheUgandaGTZstovesareanexampleofearthenormudstoves.Thesestovescombinehighclay
contentsoilandanorganicbindertoformthestovebody.Binderscanincludestraw,sawdust,crop
residueaswellaswoodashandpumice.Theadditionofbinderscanimproveamixturesmechanical
andthermalproperties.Theirfibrousnatureimprovesmaterialstrengthanddurability.Theyalso
displacetheclaymatrix,reducingdensity,heatcapacity,andthermalconductivity(Binici,Aksoganetal.
2007).

2.3 ROCKETSTOVES

TherocketstoveisanimprovedstovedesigndevelopedbyLarryWiniarskiandtheAprovechoInstitute.
ThestoveincorporatesanLshapedcombustionchamberandpotskirttoimproveheattransferand
combustionefficiencyduringcookingactivities.Thecombustionchamberconsistsofahorizontalfuel
magazineandverticalinternalchimney.Woodisfedhorizontallyintothefuelmagazineensuringeven
combustionfromoneendandamoreeasilyregulatedfeedrate.Theinternalchimneycreatesdraft,
acceleratingcombustionsgasesfromthefire.Thesegasesarethenforcedthroughtheskirtthat
surroundsthecookpot.Greaterconvectiveheattransferistheresultofimprovedadvectionand
increasedsurfaceareacontact(Bryden,Stilletal.2005).Aschematicofarocketstovecanbeseenin
Figure21.


Figure21RocketStove(source:AprovechoInstitute)

2.4 GTZUGANDASTOVES

IncooperationwiththeUgandanMinistryofEnergyandMineralDevelopment,theGermanTechnical
Cooperation(GTZ)developedadesignanddisseminationprogramfortwoimprovedmudstovedesigns.
Aspartofthisproject,thegroupdistributedadesignandconstructionguide,HowtoBuildtheImproved
HouseholdStoves(2004).Variationsonthetwodesignswereconstructedduringfieldworkin
MauritaniaandMali.Athirddesign,influencedbytheUgandaGTZmodels,wasalsodevelopedand
implemented.

Thefirstofthesedesigns,theshieldedfirestove,isasinglepotrocketstove.Itincorporatesthe
characteristicLshapedcombustionchamberwithapotskirt.Thestoveiscustombuilttothecookpot
mostfrequentlyusedinthehousehold.Clayandstraworanotherappropriatebinderaremixedand
packedaroundanLshapedformtoconstructthestovebodyandcombustionchamber.Thecookpot
itselfactsastheformforthepotskirttoinsureanexactfit.Aschematicoftheshieldedfirestovecan
beseeninFigure22.


Figure22ShieldedFireStove(source:adaptedfromMinistryofEnergyandMineralDevelopment,Republic
ofUgandaandGTZ2004)

Theseconddesign,arocketlorenamudstove,isamoreadvancedvariationoftheshieldedfirestove.It
incorporatesthesameLshapedcombustionchamberbutincludesspacefortwopotsconnectedby
channelsandachimneytoremovesmokefromthecookingspace.Itisconstructedfromthesamemud
mixtureastheshieldedfirestoveandisalsocustombuilttothedimensionsofthemostfrequentlyused
cookpots.AschematicoftherocketlorenastoveisprovidedasFigure23.

Figure23RocketLorenaStove(source:adaptedfromMinistryofEnergyandMineralDevelopment,Republicof
UgandaandGTZ2004)

Thethirddesign,atwopotshieldedfirestove,isavariationontheshieldedfireandrocketlorenastove
andwasdevelopedbytheauthorforimplementationinMali.ItincludesthesameLshapecombustion
chamberandchannelconnectedpotseatsastherocketlorenastovebuthasnochimney.Therounded
bodyissimilartothatoftheshieldedfirestoveandisformedaroundthepotsinsamemanner.

Figure24Twopotshieldedfirerocketstove

 3 METHODS

Thischaptercoversthemethodologiesusedinthisproject,specificallyprojectparticipantsurveys,stove
constructionandtesting,materialtesting,andthermalmodeling.

3.1 USERSURVEYS

Togainagreaterunderstandingofcookingpracticesintheprojectsitesandtoevaluatethe
performanceoftheimprovedmudstoves,usersurveyswereconductedwithcommunitymemberswho
showedinterestinconstructinganimprovedstove.Preandpostimplementationsurveyswere
conductedwiththeusersofthetwopotrocketstove,whileusersoftherocketlorenaandsinglepot
rocketstovesweresurveyedonlyafterstoveimplementation.

Surveyparticipantswerethewomenprimarilyresponsibleforcookingineachhome.Theywerefirst
approachedandaskediftheywereinterestedinconstructinganimprovedstove.Ifinterestwas
exhibited,thepreimplementationsurveywasgiven.Onemonthafterstoveconstructionwas
complete,theparticipantsweregiventhepostimplementationsurvey.

SurveyswereadministeredorallybynonnativebutcompetentWestAfricanBambaraspeakers.The
usersunderstandingoftheinterviewerstranslationofsurveyquestionsfromEnglishtoBambarawas
necessarytorespond.Incaseswherecomprehensionwasdifficult,athirdpartywasusedtoclarify
questionsandresponses.AllinterviewtechniqueswerecompliantwithMichiganTechnological
Universityhumansubjectsresearchguidelines.

ThesurveywasamodifiedversionofthequalitativeusersurveyoutlinedinthePartnershipforClean
IndoorAirs(PCIA)KitchenPerformanceTest(2007).TheKitchenPerformanceTestsurveyincludes
questionsnotrelevanttothetypeofstovesimplementedintheprojectaswellasseveralother
questionsrelatedtosocioeconomicandgenderissues.Althoughimportant,thesewereoutsidethe
scopeofthisstudyandwerenotused.Modificationsweremadetonarrowthescopeofthesurveyto
cookingpractices,fueluse,usersatisfactionanduserinput.Table31showsthequestionsusedinthis
abbreviatedsurvey.

10

 Table31Cookingpracticesandimprovedstovesurvey

PreImplementationSurvey PostImplementationSurvey

Questions
1. Whoisprimarilyresponsibleforcooking(age
andgender)?
2. Iscookingdoneinside,outside,orboth?
3. Isthekitchenseparatefromthemainhouse?
4. Whatkindofstove(s)isused?How
frequently?
5. Whosejobisittoobtaincookingfuel(age
andgender)?
6. Whereiscookingfuelobtained?Howfarisit
fromthehousehold?
7. Whatkindsofpotsareused?
8. Arepotlidsusedforcooking
9. Whatdoestheprimarycookdislikeabout
currentcookingarrangements?
10. Isthefamilyinterestedinusinganimproved
stove?
Questions
1. Howlonghasthefamilybeenusingthestove?
2. Howoftendoesthefamilyusethestove?
3. Whatkindsofpotsarebeingusedinthenew
stove?
4. Doesthefamilyuseanyotherkindsofstoves
inadditiontotheimprovedstove?
5. Isiteasierormoredifficulttocookwiththe
newstove?Why?
6. Domealstakelongertoprepareusingthe
newstove?
7. Arethereanycookingtaskseasierto
accomplishwiththenewstove?
8. Whatdoesthecooklikemostaboutthe
stove?
9. Isthereanythingthatthecookwouldchange
aboutthenewstove?

Observations
1. Whattypesofstove(s)ispresentinthe
kitchen?
2. Isthereanyevidenceforsuccessful
participationinanimprovedstoveproject?
Observations
1. Whattypesofstovesarepresentinthe
kitchen?
2. Doesthestoveappearasifithasbeenused
recently?
3. Doesitappearasifotherstove(s)arealso
beingused?Whatkindofstove(s)?
4. Whatkindoffuelisbeingused(species,
size)?

11

3.2 STOVECONSTRUCTION

Asinglepot,shieldedfirerocketstovewasconstructedforlaboratorybasedperformancetestingat
MichiganTechnologicalUniversity.ProceduresanddesigncriteriapresentedintheUgandanMinistryof
EnergyandMineralDevelopmentandGTZpublicationHowtoBuildtheImprovedHouseholdStoves
(2004)werefollowedduringthestoveconstructionprocess.Followingthedesignguide,criticalstove
dimensionsweredictatedbythesizeandgeometryoftheprimarycookpottobeusedduringstove
operation.

Becauseweighinginstrumentsarenotreadilyavailablewhenconstructingstovesinthefield,stove
materials,clayandstraw,weremixedbyhandonaonetoonevolumebasis.Drymaterialsweremixed
inbatchesandwaterwasaddeduntilthemixturereachedaworkableconsistency.

Sectionsof16cmdiameterPVCpipewereusedasformsforthefuelinletandcombustionchamber.
TheclayandstrawmixturewasplacedbyhandaroundthePVCpipetofromthestovebody.Thecook
potitselfwasusedastheformforthepotskirt.Theclayandstrawmixturewasplacedinthesame
mannerasthecombustionchambertoathicknessofsevencentimeters.Whenthepotwasremoved,a
smallgapwascutinthematerialtoformtheoutletforcombustiongases.Toensurestability,the
processwasconductedintwostagesovertwodays,allowinglowersectionsofthestovetodryfor24
hoursbeforeadditionalmaterialwasadded.Thestovewasallowedtodryforfourweekspriorto
testing.

Figure31Schematicofshieldedfirerocketstoveconstructedforlaboratoryperformanceandemissions
testing(Alldimensionsincentimeters)

12

3.3 STOVEPERFORMANCETESTING

TheproceduresoutlinedintheWaterBoilingTest(WBT)fromthePartnershipforCleanIndoorAir
(2007)wereusedtoevaluatethethermalperformanceofthestove.Thistestevaluatesastoveduring
threeseparatephasesofoperation:coldstart,warmstart,andsimmer.Duringthetest,theamountof
timeandfuelrequiredtobringameasuredvolumeofwatertoaboil,aswellasholditatasimmer,
wererecordedtodetermineoverallstoveefficiency.

ThreecompleterunsoftheWBTwereconductedinanenclosedspaceunderafumehoodinthe
foundryoftheMaterialsandMetallurgyBuildingontheMichiganTechnologicalUniversitycampus.The
equipmentusedincludedatypeKthermocoupleanddataloggertomonitorandrecordwater
temperatureandamassbalancetomeasuretheamountoffuelused.

Duringstoveconstruction,thermocoupleswerealsoplacedinthestovebodytomeasurethe
temperaturedistributionbetweenthecombustionchamberandtheoutsidewallofthestove.Three
typeKthermocouples(OmegaEngineeringInc.Stamford,CT)wereplacedjustinsidethewallofthe
combustionchamber,atthemidpointofthestovebody,10cmfromtheinsidewall,andattheoutside
stovewall,20cmfromtheinsidewall.Temperaturemeasurementsweretakencontinuouslyduring
stoveoperation,andrecordedeverytenseconds.

Figure32ShieldedfirestoveduringaWaterBoilingTestfor1Lofwaterina30cmdiameterpot

13

3.4 STOVETHERMALMODELING

Tomodeltemperaturedistributionandheatlossinthestovebodyduringoperation,aonedimensional,
tennodefinitedifferencemodelwasused.ThemodelwasoriginallydevelopedbyAndreattainhis
reportHeatLossfromStoves:Thermalpropertiesofinsulativebricks(2003)andwasadaptedtorunin
MATLAB.AcopyoftheMATLABscriptisavailableinAppendixA.

Themodelinputsinclude:therelevantthermalpropertiesofthestovematerials,stovedimensions,and
theambientenvironmentalconditionsduringstoveoperation.Assumptionsincludedaconstant
combustiongastemperatureandidealconvectionandradiationheattransferatthewallofthe
combustionchamberandtheoutsidesurfaceofthestove.

Themodelpredictedtemperatureasafunctionofoperationtimeattenevenlyspacednodesinsidethe
stovewall.Thepositionsofthesixthandtenthnodescorrespondedtothepositionsofthe
thermocouplespresentinthestovebodyduringperformancetesting.Temperaturedatacollected
duringmultiplestestrunsatthesepointswereusedwhencalibratingthemodelandcheckingfor
accuracy.Totalenergylosstothestovebodywasalsopredictedfromasummationofthetemperature
ateachnodemultipliedbythematerialsspecificheatandthetotalmassofthestovebody.

Figure33Thermocoupleplacementandcomputationalelementnodesinthestovebody

14

3.5 MATERIALSTESTING

3.5.1 Specificheat

Thespecificheatoftheclayandstrawmixturewasmeasuredusingacalorimeter.Smallsamples,3to5
cmindiameter,wereprepared.Aftermeasuringtheirmass,eachsamplewasplacedinanovenuntil
theyreachedatemperatureofapproximately120C.Whenremovedfromtheoven,sampleswere
placedinaninsulatedcontainercontainingameasuredvolumeofwater.Thesystemwasallowedto
reachequilibriumandthetemperatureofthewaterwasmeasured.

Becausethesystemisinsulated,itwasassumedthattheenergylostfromthesamplewasequaltothe
energygainedbythewater,andthesteadyflowthermalenergyequationcanbeapplied.Whereqis
energyinJoules,mismassinkg,CpisspecificheatinJ/kgK,andTistemperatureinKelvin.

q mc p (T f Ti )
(1)

mH 2O C p , H 2O (T f Ti , H 2O ) msample C p , sample (Ti , sample T f )
(2)
mH 2O C p , H 2O (T f Ti , H 2O )
C p , sample
msample (Ti , sample T f )
(3)

SolvingEquation2forCp,sampleprovidesanexpressionforcalculatingthespecificheatofthematerial
mixtureasshowninEquation4(Incropera2007).

mH 2O C p , H 2O (T f Ti , H 2O )
C p , sample
msample (Ti , sample T f )
(4)

3.5.2 Thermalconductivity

Theclayandstrawstovemixtureisneitheranisotropicnorahomogenousmaterial.Definitively
quantifyingthermalconductivitycanbedifficultforthistypeofmaterial.Insteadtwomethodswere
usedtoprovidearangeofappropriatevalues.

Thefirstmethodusedprinciplesofonedimensional,steadystateconductionthroughplanarobjects.In
thissystem,asampleofthematerialwasstackedinserieswithasectionoffoaminsulationofknown
thermalconductivity(Figure34).Thermocoupleswereplacedoneitherendandbetweenthetwo
objects.Thesampleandthesectionofinsulationwerethenplacedbetweenalargepotofwaterat
ambienttemperatureandablockofice.Theseservedasheatsinksforthesystemandremainedat
relativelyconstanttemperature,thewaternearroomtemperatureandtheicenearfreezing.The
systemwasallowedtoreachsteadystateconditionsandthetemperaturedifferentialacrossthe
materialswasrecorded.

Becausethetemperatureofthewaterandtheiceremainatconstanttemperatures,theheatfluxis
assumedtobeonedimensionalatthecenterandconstantthroughboththesampleandthesectionof
foaminsulation.Usingthemeasuredtemperaturedifferentialacrosstheinsulationandacrossthe
samplethethermalconductivitycanbedetermined.

15

AnadaptationofFourierslawofconductiveheattransfergiventheparametersofthissystemprovides
anexpressionfortheheatflux(Equation5).

k1 (T1 T2 ) k2 (T2 T3 )
q ''

L1 L2
(5)

WhereqisheatfluxperunitareaW/m2,Listhematerialthicknessinm,kisthermalconductivityin
W/mK,andTistemperatureinKelvin.Solvingtheequationfork2providesanexpressionforcalculating
thermalconductivity(Incropera2007).

k1 L2 (T1 T2 )
k2

L1 (T2 T3 )
(6)

Figure34Conductiveheattransferthroughaseriesofplanarobjectsofvaryingthermalconductivity

ThesecondmethodusedtheMATLABcomputermodeldescribedabove.Themodelrequiredvaluesfor
materialdensityandspecificheatinadditiontothermalconductivity.Havingpreviouslydetermined
densityandspecificheat,thermalconductivitywasallowedtovary.Themodelwasthencalibratedto
thetemperaturedatacollectedusingarootmeansquaredeviationapproach.

16

3.5.3 Compressionstrength

Compressionstrengthoftheclaystrawmixturewasmeasuredusingsamplespreparedinconcrete
molds.Thesamplesmeasuredapproximately10cmindiameterand20.5cminlength.Cylinderswere
allowedtocureforaperiodofatleast30dayspriortotesting.Testcylinderswereloadeduntilfailure,
onaWarnerandSwaseyCompressionTestPressandtheaverageoftherecordedpeakloadswereused
todeterminetheoverallcompressivestrength.

3.6 STOVEEMISSIONSTESTING

Stoveemissionsweremeasuredinconjunctionwithstovethermalperformancetesting.Theobserved
pollutantswerecarbonmonoxide(CO),carbondioxide(CO2),andparticulatematteratorbelow2.5
micrometersinaerodynamicdiameter(PM2.5).ATSIQTrakairqualitymonitor(TSI,St.Paul,MN)was
usedtomeasureCOandCO2concentrationsatintervalsof30seconds.MeasurementsofPM2.5
concentrationsweretakenwithaTSIDustTrakmonitorsettorecordatintervalsofoneminute.Bothair
qualitymonitorswerefactorycalibratedbeforetheiruseinthisstudy.Measurementsweretaken
continuouslyduringallthreephasesofallthreefullexperimentalrunsofthePCIAWaterBoilingTest.

Figure35Emissionssamplingsetupduringstovetesting;instrument
samplinginletswerepositioned1mabovepotatopstove

17

 4 RESULTSANDDISCUSSION

4.1 USERSURVEYRESPONSES

Userresponsestotheadministeredsurveyprovidedcontexttostoveconstructionactivitiesaswellas
constructivefeedbackrelevanttoviablestovedesignimprovements.Inpreimplementationsurveys,
participantsgaveinsightintotypicalcookingpracticesandfueluse.Cookingactivitiesandfuelgathering
aredonealmostexclusivelybywomen.Travelofuptosevenkilometersisrequiredtoobtainadequate
firewood.Cookingisdoneprimarilyindoorsoverthreestonefires.Usersappreciatedtheportability
andtheeaseofuseofatraditionalthreestonefire,butweredissatisfiedwiththeamountofsmokeand
poorfueleconomy.

Inpostimplementationsurveys,participantscritiquedtheshieldedfirerocketstove,rocketlorena
stove,andtwopotshieldedfirerocketstove.Responsespertainingtodesignfeaturescommontoall
threestovetypeswereconsideredfordesignimprovements.Thesearediscussedingreaterdetailin
Chapter6:DesignSuggestions.

4.2 STOVEPERFORMANCECHARACTERISTICS

ResultsoftheWaterBoilingTestfortheshieldedfirestovecanbeseeninTable41.Theresultsare
separatedintothethreesectionsrepresentingeachphaseofthetest,highpower(coldstart),high
power(hotstart),andlowpower(simmer).

Asinglevalueofthermalefficiencyderivedfromacombinationofthetestphasesisnotanadequate
indicatorofstoveperformancebecauseitdoesnotconsidertheproductionofsteam.Thermal
efficiencyindicateshoweffectiveheatenergyfromthefireistransferredtothecookpot.Highthermal
efficiencymayalsocoincidewiththeproductionofexcesssteam.Energycarriedawaybysteamcannot
beutilizedinthecookingprocess(Balis,Ogleetal.2007).Forthisreasonspecificfuelconsumptionis
alsopresentedastheamountoffuelconsumedperunitoutputorinthiscase,theamountofwood
burnedtoboiloneliterofwater.

Othermetricsincludefirepower,simplyarepresentationofhowquicklythestovereleasesenergyfrom
fuel,andturndownratio,therelationshipbetweenthepoweroutputsduringthehighpowerphasesof
thetestandthelowerpower,simmerphase.Valuesshownastemperaturecorrectedareadjustedtoa
standard75Ctemperaturechange(25to100).Thisallowsforcomparisonsbetweenstovestested
underdifferentambientenvironmentalconditions.

18

 Table41Shieldedfirewaterboilingtestresults.TheWaterBoilingTestwasconductedinthreephases,high
powercoldstart,highpowerhotstart,andsimmertoevaluatethestoveundervaryingoperatingconditions.
Thedurationofthefirsttwophaseswasdependentonthetimeneededtobring2.5litersofwatertoboil.
Thethird,lowpowerphasewas45minutesinduration.Duringthistimewaterwaskeptjustbelowboilingto
simulateslowcookingoperations.Valuesshownastemperaturecorrectedareadjustedtoastandard75C
temperaturechange(25to100).

HIGH POWER HIGH POWER LOW POWER

SHIELDED FIRE STOVE
units (COLD START) (HOT START) (SIMMER)
Time to boil min 25.3 15.2 -
Temp-corrected time to boil min 24.2 14.5 -
Burning rate g/min 27.1 26.7 25.4
Thermal efficiency % 13% 18% 21%
Specific fuel consumption g/liter 289.3 171.5 20556.7
Temp-corrected consumption g/liter 276.6 163.1 -
Energy Input MJ 11.2 7.0 19.4
Firepower watts 7771 7656 7283
Turn down ratio - - - 1.08


Theincreasingtrendinthethermalefficiencyanddecreasingtrendinspecificfuelconsumptionofthe
shieldedfirestovebetweenphasesofthetestsuggestsasignificantamountofheatlosstothestove
body.Asthetestbegins,thestovebodyisatambienttemperatureandasignificantamountofenergy
islostwhileitwarms.Thoughthestovebodyisnotespeciallyconductive,itsmassandspecificheat
capacityaresignificantenoughtoabsorbandretainheatthatmightotherwisebeusedtoheatthecook
pot.Efficiencyincreasesasthestovebodyreacheshighertemperaturesandlessenergyisabsorbed.

Asameansofcomparison,theresultsofacoldstarttestforatraditionalthreestonefirearepresented
inTable42.Thetestwasperformedonlyonceandisnotmeanttobestatisticallyvalid,butitdoes
provideasimplebenchmarkforcomparison.

19

 Table42Comparisonofshieldedfirestoveandthreestovefireperformanceresults

THREE STONE SHIELDED FIRE

HIGH POWER (COLD START)
units FIRE STOVE
Time to boil min 49.6 25.3
Temp-corrected time to boil min 46.1 24.2
Burning rate g/min 14.2 27.1
Thermal efficiency % 18% 13%
Specific fuel consumption g/liter 367.3 289.3
Temp-corrected consumption g/liter 341.4 276.6
Energy Input MJ 12.1 11.2
Firepower watts 4078 7771

Acomparisonoftheperformancedatashowsthat,inacoldstartscenario,athreestonefirewill
outperformtheshieldedfirestovewithrespecttothermalefficiency,althoughitdoestakemuchlonger
toreachboilingtemperature.Atlaterphases,theshieldedfirestovedoesofferasmallimprovementin
thermalefficiencyandspecificfuelconsumption.Overall,theshieldedfirestoveisatasignificant
disadvantageduetothemassofitsstovebodyandofferslittleimprovementoverathreestonefire
withrespecttothermalefficiency.

4.3 STOVEBODYHEATLOSSANDTHERMALDISTRIBUTION

StovebodytemperaturedatacollectedduringthefirstandsecondrunoftheWaterBoilingTestsuggest
thematerialmixtureisoflowthermalconductivitybutrelativelyhighspecificheatcapacity.Theplotted
datainFigure41andFigure42showthetemperaturegainatthreepointsinsidethestovewall.In
boththeruns,thetemperatureinthemiddleofstovebodydoesnotbegintoincreasesignificantlyfor
thefirst30minutesofthetest.Itthenincreasesexponentiallyfortheremainderoftherun,indicating
slowthermalconductionfollowedbythermalabsorptionandstorage.

Thetemperaturesatthemidpointofthestovebodypredictedbythenumericalheatlossmodelare
comparedtothemeasuredvaluesduringthefirstandsecondrunsofthewaterboilingtestarealso
showninFigure41andFigure42.Themodelusedmeasuredvaluesofdensity,specificheat,anda
varyingvalueofthermalconductivitythatwascalibratedtofittheoutputtothemeasuredtemperature
data.ThemixturematerialpropertiesusedinthemodelarepresentedinTable44.

Thenumericalmodelwasalsousedtoestimatetotalheatlossintothestovebodyforallthreerunsof
theWaterBoilingTest.InTable43heatlossisexpressedasapercentageoftotalenergyinput.Design
changesthatreducewallthicknessandoverallstovemassshouldresultinreducedheatlosstothe
stovebodyandanincreaseinthermalefficiency.

20

 Figure41Stovebodytemperaturedistribution(WaterBoilingTest:Run1)

21

 Figure42Stovebodytemperaturedistribution(WaterBoilingTest:Run2)

Table43Heatlosstostovebodyrelativetototalenergyinput

Units WBTRun1 WBTRun2 WBTRun3

EnergyInput MJ 33.19 41.72 37.75
EnergyLosstoStoveBody MJ 3.74 3.99 3.66
PercentofEnergyInput % 11.3 9.6 9.7

22

4.4 MATERIALPROPERTIES

TherelevantthermalandmechanicalpropertiesoftheclayandstrawmixturearegiveninTable44.
Thermalpropertiesforfiredceramicsandothermaterialsthathavebeenusedintheconstructionof
combustionchambersandimprovedstovebodiesarealsoincludedforcomparisoninTable45.Glass
woolisincludedtorepresentanidealinsulatorbutitislikelyimpracticalforstoveconstruction.

Withrespecttothermalproperties,theclaystrawmixtureiscomparabletoclaywithsawdustandclay
withvermiculitebricks.Iftheclaystrawmixturewasexposedtoasimilarfiringprocess,itislikelythe
thermalpropertieswouldimprovefurther.Combustionasaresultofstoveoperation,aswellas
breakdownduetonaturaldecompositionovertime,mightalsoconsumesomeofthestrawinamanner
similartoafiringprocess.Thiswouldreducedensityandimprovethematerialsthermalproperties.

Table44 Mudmixturematerialproperties

3
Density() 881 Kg/m
SpecificHeat(Cp) 881 J/KgK
ThermalConductivity(k) 0.0620.142 W/mK
2
CompressiveStrength() 1.45 N/mm

Table45Thermalpropertiesofcommoncombustionchambermaterials(Andreatta2003)

Density SpecificHeat ThermalConductivity

Material
(kg/m3) (J/KgK) (W/mK)
OrdinaryBrick 1600 840 0.700
GuatamalanBaldosa 1691 812 0.219
85%pearlite15%clay
439 921 0.128
(760Cfiringtemperature)
50/50sawdust/clay 729 701 0.081
85%vermiculite15%clay 559 698 0.120
ElCocoBaldosa 1328 835 0.181
Pumicebrick 770 835 0.107
Glasswool 40 700 0.038

4.5 STOVEEMISSIONS

Fromthedatacollectedbyemissionsmonitoringequipmentduringperformancetestingoftheshielded
firestove,movingaveragesofpollutantconcentrationsoverallthreecompleterunsofthewaterboiling
testweredetermined.PlotsofCO2,CO,andPM2.5aregiveninFigure43,Figure44,andFigure45
respectively.PlotofCOandPM2.5concentrationsforacoldstarttestofthreestonefirearealsoshown
inFigure46andFigure47respectively.

23

 2000
HighPower HighPower
LowPower
(ColdStart) (Hot Start)
(Simmer)

1500
CO2 Concentration(ppm)

1000

500

0
0 10 20 30 40 50 60 70 80
Time(min)

Figure43AveragesofCO2concentrationofshieldedfirestoveemissionsoverthreecompleterunsofthe
WaterBoilingTest(WBT).Errorsbarsrepresentonestandarddeviationaboveandbelowmeanvalues.

250
HighPower HighPower
LowPower
(ColdStart) (Hot Start)
(Simmer)
200
COConcentration(ppm)

150

100

50

0
0 10 20 30 40 50 60 70 80
Time(min)

Figure44AveragesofCOconcentrationofshieldedfirestoveemissionsoverthreecompleterunsofthe
WaterBoilingTest(WBT).Errorsbarsrepresentonestandarddeviationaboveandbelowmeanvalues.

24

 40
HighPower HighPower LowPower
(Cold Start) (Hot Start) (Simmer)

30
PM2.5(mg/m3)

20

10

0
0 10 20 30 40 50 60 70 80
Time(min)

Figure45AveragesofPM2.5concentrationofshieldedfirestoveemissionsoverthreecompleterunsofthe
WaterBoilingTest(WBT).Errorsbarsrepresentonestandarddeviationaboveandbelowmeanvalues.

20
18
16
14
12
CO(ppm)

10
8
6
4
2
0
0 5 10 15 20 25 30 35
Time(min)

Figure46MovingaverageofPM2.5concentrationofthreestonefireemissionsduringonecoldstarttest.
Errorsbarsrepresentonestandarddeviationaboveandbelowmeanvalues.

25

 2.5

2
PM2.5(mg/m3)

1.5

0.5

0
0 5 10 15 20 25 30 35
Time(min)

Figure47MovingaverageofPM2.5concentrationofthreestonefireemissionsduringonecoldstarttest.
Errorsbarsrepresentonestandarddeviationaboveandbelowmeanvalues.

Withanoverallmedianvalueof757ppm,theCO2concentrationsintheshieldedfirestoveemissionsdo
notvaryappreciablythroughthephasesoftheWaterBoilingTest.Relativetootherindoorair
pollutants,theadversehealtheffectsassociatedwithextendedexposuretoCO2arelowbutcaninclude
eye,noseandthroatirritation,mentalfatigueandheadache.CO2concentrationsinnormalindoor
environmentsvaryfrom350to2500ppm(Seppnen,Fisketal.1999).Concentrationsmeasuredinthe
shieldedfirestoveemissionsdidnotexceedthisrange.

AnincreasingtrendinCOconcentrationcanbeobservedthroughtheprogressionoftheWBT.COisa
productofincompletecombustionanddespitetheincreasedefficiencyatlaterstagesofthetest,CO
concentrationscontinuedtorise.AsimilarriseinCOemissionsduringimprovedstoveoperationwas
observedinNdiema,Mpendazoeetal.1998.However,theauthorsobservedaneventualpeakand
subsequentdeclineinCOconcentrationsafterthestovehadbeenoperatingforperiodofapproximately
25minutes.Thismaycorrespondtothestovebodyreachingaconstanttemperatureandpeak
operatingefficiency.Thelargemassoftheshieldedfiremudstoveandsubsequentenergyabsorption
preventeditfromreachingaconstanttemperatureandasteadystatecondition.

Carbonmonoxideisabsorbedeasilyintotheblood.Itcanimpairapersonsabilitytofunctionbut
presentsthegreatestthreattothosewithcardiovascularproblems(Mihelcic,Zimmermanetal.2010).
ThemedianCOconcentrationduringstovetestingwas80ppm.TheU.S.EnvironmentalProtection
Agency(EPA)ambientairqualitystandardsare35ppmforonehourofexposureand9ppmforeight
hoursofexposure.COemissionsfromtheshieldedfirestovealsosignificantlyexceededthoseofthe
threestonefirewhichhadamedianvalueof3ppm.

26

PM2.5concentrationsdecreasedsteadilyduringstoveoperation.Spikesintheplotteddatacorrespond
tofireignitionatthestartoftestphasesortheadditionofnewfuel.Particulatematterisalsoaproduct
ofincompletecombustion.Thedatasuggeststhatstoveefficiencyhasagreatereffectonparticulate
matteremissionsthanonCOemissions.

Findparticulatemattercancausesignificantdamagetotherespiratorysystem(Mihelcic,Zimmermanet
al.2010).ThemedianPM2.5concentrationduringstovetestingwas1.96mg/m3comparedto0.21
mg/m3forthethreestonefire.TheEPAstandardsets35g/m3and15g/m3asrecommendedPM2.5
concentrationlimitsforexposureperiodsof24hoursandoneyearrespectively.Itisclearlyevidentwhy
cookingonsuchstovesresultsinconsiderableirritationintheshorttermandrespiratoryimpactsinthe
longterm.

27

 5 DESIGNSUGGESTIONS

Objectivesofanyimprovedstoveprojectshouldbelocalviability,economicfeasibility,anddesirability.
Mudstovesarenotanidealstovesolution.Moreadvancedstovesoffergreaterfueleconomyandmore
significantimprovementinbiomassemissions.However,whentechnical,material,andfinancial
resourcesarelimited,mudstovesmaybeanappropriatesolution,anddespitetheirdisadvantages
certainimprovementscanbemade.Withinformationgatheredfromtheperformanceanalysisofthe
GTZUgandashieldedfirestove,theevaluationofmudmixturematerialproperties,andresponsesfrom
usersinpostimplementationsurveys,designsuggestionsformudstovebuildersweredeveloped.

Inthisstudyandothersithasbeenobservedthatunventilatedimprovedstoves(stoveswithouta
chimneyorotherdirectmeansofsmokeremoval)mightnotappreciablydecreasecombustion
emissionsdespiteanincreaseinthermalefficiency.Infact,insomecasescertainemissionsmight
actuallyincreaseduringtheoperationofanimprovedstoveversusanopenfire.Ahuja,Joshietal
(1987)notedadecreaseintotalsuspendedparticulateemissionsbutanincreaseincarbonmonoxide
emissionsforthemoreefficientstovesevaluatedintheirstudy.HighlevelsofCOwerealsoobservedin
cookstoveemissionsbyNdiema,Mpendazoeetal.(1998)aswellasthisstudy.

Ideally,anunventilatedorchimneylessimprovedstovewillbeusedoutsideorinawellventilated
indoorspaceseparatefromlivingquarters.However,thisisnotalwaysthecircumstancesunderwhich
improvedstovesareimplementedandoperated.Ifsignificantimprovementstostoveemissionsarenot
possibleduetocertaindesignconstraints,asmaybethecasewhenusingsimpleearthenmaterialsin
stoveconstruction,itmaybeadvisabletoputincreasedfocusonimprovingthermalefficiency.In
additiontoreducingbiomassfueluse,increasedefficiencymightalsoreducecookingtimeandthe
userslengthofexposuretocombustionemissions.Thiswouldrequirefurtherobservationswithusers
inthefieldasnolaboratoryanalysiscouldadequatelysimulatetraditionalcookingpractices.

Stoveheightisonedesignparameterthatcanbeadjustedtoimprovethermalefficiency.TheGTZ
Ugandashieldedfirestovedesignguidesuggestsastovebodyandinternalcombustionchamberheight
2.5timestheheightofthefuelinlet.Intheirevaluationofrocketstoveoperatingvariations,Hudelson
etal.(2002)observedanincreasingtrendinthermalefficiencywithareductioninstoveheight.
Combustionefficiencymaysufferascombustiblegasseshaveareducedamountofspaceforsecondary
andmorecompletecombustion,butreducingthestoveheightwillputthepotincloserproximitytothe
firethusincreasingradiativeheattransfer.

Stoveusersalsotookissuewithstoveheightinpostimplementationsurveys.InMalithestapledishisa
cornflourbasedmixturecalledtoh.Duringpreparation,tohrequiresrigorousstirringforittosolidify.
Atallerstoveandaraisedpotheightwouldmakepreparationeasierastheuserdoesnothavetobend
overasfar;yet,Maliansdislikedtheaddedheightandpreferredalowerpotheightwhilecooking.

Asmallerinletdiameterresultsinincreasedefficiencyandlowercombustiongasloses(Hudelson,
Brydenetal.2002).Thiswasapointofcontentionbetweenstovedesignersandstoveusersin
MauritaniaandMali.Traditionallylargediameterwoodwascollectedforcookfiresasittendstoburn
forlongerperiodsoftimebeforeadditionalwoodisneeded.Attemptstosuggestasmallerstoveinlet
diameterandtheuseofsmallerfuelweremetwithresistance.Theinabilitytoburnlargepiecesof
woodwasacomplaintseeninseveralpostimplementationsurveys.Ifpotentialstoveusersare
receptivetousingsmallerfuelsizethenreducingthestovesinletdiameterisadvisable.Itisalso
recommendedthataconstantcrosssectionalareabemaintainedbetweenthestoveinletandthe

28

internalcombustionchamber.Anymodificationsmadetotheinletdiametershouldbeappliedtothe
combustionchamberdiameteraswell.

Table51andthecorrespondingdiagraminFigure51showstoveinletandcombustionchamber
dimensionsbasedonthecookpotdiameter.ThedataismodifiedfromtheGTZUgandaguidetoshow
adjusteddesignparameters.AsPVCpipeorothermaterialswillbeusedastemporaryconstruction
forms,itmaybeadvisabletoselectinletandcombustionchamberdimensionsthatmostcloselymatch
locallyavailablematerials.

Table51Inletandcombustionchamberdimensionsbasedoncookpotdiameter.Distheinlet
diameter,Histheheightofthecombustionchamber,Kisthedistancefromthetopofthefuelinlet
totheoutletofthecombustionchamber,andListhepotskirtgap(modifiedfromMinistryof
EnergyandMineralDevelopment,RepublicofUgandaandGTZ2004).

Inlet/Combustion
PotDiameter Chamber PotSkirtGap
ChamberDiameter 2 K=D/2(cm) H=K+D(cm)
(cm) Area(cm ) (L)(cm)
(D)(cm)
?20 12.0 113 6.00 18.00 1.6
2127 14.0 154 7.00 21.00 1.5
2830 16.0 201 8.00 24.00 2.0
3135 16.0 201 8.00 24.00 1.5
3640 18.0 254 9.00 27.00 1.9
4145 18.0 254 9.00 27.00 1.6
4650 20.0 314 10.00 30.00 1.9

Figure51Stoveinletandcombustionchamberdimensions
(source:adaptedfromAprovechoInstitute)

29


Significantheatlossintothestovebodywasobservedduringstovetestingaswellasinthenumerical
temperaturedistributionandheatlossmodel.Reducingtheoverallmassofthestovewilldecreasethe
amountofenergyconductedandstoredinthestovebody.Inadditiontoreducingoverallstoveheight,
wallthicknessshouldalsobeminimized.Whencustomsizingastovebodytothedimensionsofthe
userscookpot,theGTZUgandashieldedfirestovedesignguidesuggestsadding20cmtothepots
diametertosettheoverallstovediameter.Thissignificantlyincreasesthestovesmassandthermal
capacity.

Theadditionalthicknessisdesignedtoaccommodatethepotskirt.Thepotskirtisanimportantdesign
featureasitimprovesconductiveheattransferbyacceleratingcombustiongasesthroughasmallgap
surroundingthepot.Adding10cmtotheeithersideofthestovebodyallowsforan8or9cmthick
skirt(aftercuttingawayasmallamountofmaterialforthecombustiongasoutlet).Giventhemeasured
mechanicalstrengthoftheclayandstrawmixture,thisisanexcessiveamountofmaterial.Adding10
cmtotheoveralldiameterofthestovebody(5cmoneachside)shouldbeadequate.Thiswillallowfor
3to4cmthickpotskirt.TheadjustmentstothestovebodythicknessandpotskirtcanbeseeninFigure
52

Figure52Potskirtthicknessandstovebodydiameter

Anotheroptionistoremovetheclaypotskirtalltogetherandusescrapmetaltoconstructtheskirt.
Thestovebodywillthenonlyneedtobewideenoughtosupportthepot.TheRechoRoketstoveisan
exampleofamudstovesimilarindesigntotheGTZUgandashieldedfirestovebutitinsteadutilizesa
potskirtconstructedfromscrapmetal(usedcans,sheetmetal,etc).

30


Figure53RechoRoketStove(source:http://www.rechoroket.com/)

Table52Summarycomparisonofcookstovedesignimprovements

Engineeringbasedstovedesignimprovements Userbasedstovedesignimprovements

Increasedstoveheight Decreasedstoveheight

Smallerfuelinletdiameter Increasedfuelinletdiameter

Smallersizedfuel Largersizedfuel

31

 6 CONCLUSIONS

Theuseoflarge,heavystovesconstructedwithearthenmaterialsisnotanidealinterventionmethod
whenaddressingenergyuseandindoorqualityassociatedwithbiomasscombustion.Insome,ifnot
manycases,thesestoveswillhavepoorerthermalperformanceandincreasedemissionsconcentrations
whencomparedtototraditionalopenfires.Ifappropriateandsustainableinthecontextoftheproject
site,smaller,moreadvancedstovesconstructedusingmetalandfiredceramicsaremorelikelyto
providethedesiredimprovementsinfueleconomyandairquality.

Despitetheirdisadvantages,thestrongstigmaagainstearthenandmudstovemaybeunjustified.
Earthenmaterialsareoflowcostandsimpletoworkwith.Inareaswherethereislittleincentive,
economicorotherwise,topurchaseanimprovedstove,motivatingindividualstoconstructtheirown
usingavailablesoilandorganicmaterialmaybeeasier.Thematerialanalysisshowedthatwhenan
adequateamountoforganicmaterialismixedwiththesoil,themixturewillhavethermalproperties
comparabletomoreadvancedceramics.Ifoverallstovevolumeiskepttoaminimum,avoiding
excessiveheatabsorptioninthestovebody,thesestoveswilllikelyprovideincreasedperformanceover
traditionalopenfirecooking.

Userinvolvementandinputisvitaltothesuccessandsustainabilityofanimprovedstove
implementationproject.Thesuggesteddesignimprovementwillnotprovideanidealizedstovewith
respecttoefficiencyandemissionscharacteristics.Optimizingstovedesignbasedsolelyonatechnical
analysisofthermalperformanceandmaterialpropertieswillnotguaranteeadoptionanduse.
Compromisesshouldbemadebetweenengineeringknowledgeanduserinputandfeedback.

Inadditiontotheirinvolvementinthedesignandconstructionprocess,stoveusersshouldalsoreceive
trainingonadditionalmethodsofimprovingfueleconomyandindoorairquality.Thisshouldinclude
suggestionsforincreasingventilationinthecookingspace,properfiretendingtechniques,andfuel
selection.Collaborativediscussionsontheconnectionbetweenpoorindoorairqualityanditsadverse
healtheffectswillalsolikelyleadtoincreaseduserinvolvementandinnovation.

32

 7 FUTUREWORK

Additionalstovesusingvariationsinthesuggesteddesignimprovementsshouldbebuiltandtested.
Themathematicalheatlossmodelpresentedinthisreportcanbeusedtopredictstovebodyheatloss
andtemperaturedistributionforvariousdesignvariations,butthisshouldonlybeapreliminarytest.
Becauseofthehighamountofvariabilityassociatedwiththebiomasscombustionprocess,thereisno
substitutionforfullscaleperformancetesting.

Additionaltestingshouldalsobeconductedinthefield.TheWaterBoilingTestandotherlaboratory
basedperformanceevaluationsarenotadequaterepresentationsofstoveoperationandcooking
practicesinrealworldsettings.ThequalitativesurveyportionoftheKitchenPerformanceTest(KPT)
wasusedinthisstudytosolicituserfeedback.Thefulltestalsoincludesaquantitativemeasureof
householdfueluseoveraprescribedperiodoftimebothbeforeandafterstoveimplementation.The
KitchenPerformanceTestprovidesarealworldassessmentofastovesefficiencywhilegainingvaluable
userfeedback.

ThePartnershipofCleanIndoorAirhasalsoproducedaControlledCookingTest(CCT)whichmaybe
valuablewhenevaluatingtheperformanceofearthenstoves.TheControlledCookingTestinvolves
preparingastandardmealcommontothecommunitiestargetedinthestoveimplementationproject.
Conductedinacontrolledsetting,theCCTallowsforsimultaneousemissionstesting.

Furtheranalysisshouldalsobeconductedwithvariationsinmaterialmixtures.Theclaycontentof
theseearthenmixturesislikelytohavethegreatestaffectonthematerialsthermalproperties.
Increasingtheratiooforganicmattertoclaycontentmayimprovethermalperformance,butthismay
alsodecreasethemechanicalstrengthanddurabilityofthemixture.

Differenttypesofbindersbothorganicandinorganicshouldbeevaluated.Strawwastestedinthis
studybecauseitwasreadilyavailableinbothMauritaniaandMali.Cropresidueandothertypesof
plantmattermayalsomakesuitablebindersforamudmixture.Inorganicmateriallikepumicestoneis
lightweightandporous.Ifavailableintheprojectsite,itislikelytoimproveamixturesthermal
properties.Theuseofrefuselikepaperandplasticmayalsoresultinasuitablemudmixtureinaddition
toremovingmaterialfromthewastestream.

33

 8 REFERENCES

(2002).TheWorldHealthReport2002:ReducingRisks,PromotingHealthLife.Geneva,Switzerland,
WorldHealthOrganization.

(2004).HowtoBuildtheImprovedHouseholdStoves,RepublicofUgandaMinistryofEnergyand
MineralDevelopment,GermanTechnicalCooperation.

Ahuja,D.R.,V.Joshi,etal.(1987)."Thermalperformanceandemissioncharacteristicsofunvented
biomassburningcookstoves:Aproposedstandardmethodforevaluation."Biomass12(4):247270.

Andreatta,D.(2003)Heatlossfromstoves:Thermalpropertiesofinsulativebricks.

Baldwin,S.F.(1987).Biomassstoves:engineeringdesign,development,anddissemination.Arlington,
Va.,USA
Princeton,N.J.,USA,VolunteersinTechnicalAssistance;
CenterforEnergyandEnvironmentalStudies,PrincetonUniversity.

Balis,B.,D.Ogle,etal.,Eds.(2007).TheWaterBoilingTest(WBT),HouseholdEnergyandHealth
Programme,ShellFoundation.

Balis,B.,D.Ogle,etal.(2007).TheWaterBoilingTest(WBT)V3.0,HouseholdEnergyandHealth
Programme,ShellFoundation.

Balis,R.,K.Smith,etal.(2007).KitchenPerformanceTest(KPT)V3.0,HouseholdEnergyandHealth
Programme,ShellFoundation.

Barnes,D.F.(1994).Whatmakespeoplecookwithimprovedbiomassstoves?:acomparative
internationalreviewofstoveprograms.Washington,D.C.,WorldBank.

Binici,H.,O.Aksogan,etal.(2007)."Thermalisolationandmechanicalpropertiesoffibrereinforced
mudbricksaswallmaterials."ConstructionandBuildingMaterials21(4):901906.

Bruce,N.,R.PerezPadilla,etal.(2002).Thehealtheffectsofindoorairpollutionexposurein
developingcountries.Geneva,Switzerland,WorldHealthOrganization.

Bryden,M.,D.Still,etal.(2005).DesignPrinciplesforWoodBurningCookStoves,AprovechoResearch
Center,PartnershipforCleanIndoorAir,ShellFoundation.

Bussmann,P.(1988).Woodstoves:TheoryandApplicationsinDevelopingCountries.FacultyofApplied
Physics,EindhovenUniversityofTechnology.Ph.D.:182.

34


Hudelson,N.,K.M.Bryden,etal.(2002).GlobalModelingandTestingofRocketStove
OperatingVariations.Ames,IA,IowaStateUniversity,AprovechoResearchCenter.

Incropera,F.P.(2007).Introductiontoheattransfer.HobokenmNJ,Wiley.

Mihelcic,J.R.,J.B.Zimmerman,etal.(2010).Environmentalengineering:fundamentals,sustainability,
design.Hoboken,NJ,Wiley.

Ndiema,C.K.W.,F.M.Mpendazoe,etal.(1998)."Emissionofpollutantsfromabiomassstove."Energy
ConversionandManagement39(13):13571367.

Seppnen,O.A.,W.J.Fisk,etal.(1999)."AssociationofVentilationRatesandCO2Concentrationswith
HealthandOtherResponsesinCommercialandInstitutionalBuildings."IndoorAir9(4):226252.

Still,D.,M.Pinnell,etal.(2003)."Insulativeceramicsforimprovedcookingstoves."BoilingPoint:A
practioner'sjournalonhouseholdenergy,stovesandpovertyeducation(49).

Warwick,H.andA.Doig(2004).Smokethekillerinthekitchen.London,UK,ITDGPublishing.

35

9 APPENDICES

A. FINITEDIFFERENCEMODELMATLABCODE

rho=input('Input material density (kg/m^3): ');

c=input('Input material specific heat (J/Kg*C): ');
k=input('Input material thermal conductivity(W/m*C): ');
d=input('Input combustion chamber diameter (m): ');
h=input('Input combustion chamber height (m): ');
l=input('Input wall thickness (m): ');
Tamb=input('Input ambient air temperature (C): ');
Tcomb=input('Input combustion gas temperature (C): ');
totaltime=input('Input total stove operataion time (min): ');

deltat=10;
tend=time(length(time));
area=pi*d*h;
deltal=l/9;
alpha=k/(rho*c);
deltamax=0.5*deltal^2/alpha;
m=rho*deltal;

for j = 2:10;
t(j)=0;
end

qout=0;

for j=1:length(time);

t(1)=Tcomb-tamb;

for i=2:9

%Variable Conductivity
%t(i)=t(i)+deltat*k*(1+t(i)*1.8/2220)/deltal*((t(i-1)-2*t(i))+t(i+1))/(m*c);
%Constant Conductivity

t(i)=t(i)+deltat*k*(1)/deltal*((t(i-1)-2*t(i))+t(i+1))/(m*c);

end

t_mid_model(j)=t(5)+tamb;
t_out_model(j)=t(10)+tamb;
ho=1.7*t(10)^0.25;
qlostconv=ho*t(10);
qlostrad=5.67e-8*((t(10)+20+273)^4-293^4);

%Variable Conductivity

36

4/28/11 12:19 PM H:\MATLAB\Stove_Heat_Loss_Model.m 2 of 2

%t(10)=t(10)+deltat/(m*c/2)*(k*(1+t(9)*1.8/2220)/deltal*(t(9)-t(10))-
qlostconvqlostrad);

%Constant Conductivity
t(10)=t(10)+deltat/(m*c/2)*(k*(1)/deltal*(t(9)-t(10))-qlostconv-qlostrad);
qout=qout+area*deltat*(qlostconv+qlostrad);
hoeff=(qlostconv+qlostrad)/t(10);

end

qbody=area*m*c*(t(2)+t(3)+t(4)+t(5)+t(6)+t(7)+t(8)+t(9)+t(1)/2+t(10)/2);
totalqlost=(qout+qbody)/10^6

37

 B. DIGITALFILESONATTACHEDCD

FileName Description

User_Surveys.pdf Completeusersurveys

Performancecalculationsfromthewaterboiling
WBT_datacalculation_Stove_3.0.xlsx
testfortheshieldedfirerocketstove
Performancecalculationsfromthewaterboiling
WBT_datacalculation_3stone_3.0.xlsx
testforthethreestonefirestove
Temperaturemeasurmentsatthermocouples
StoveBodyTemperature.xlsx
buriedintheshieldedfirerocketstovebody
Matlabprgramfileofthefinitedifferencemodel
Stove_Heat_Loss_Model.m
forstoveheatloss

Stove_Heat_Loss_Model.pdf Finitedifferencemodelcodeinpdfformat

Carbonmonoxideandcarbondioxideemissions
CO_CO2_ShieldedFireRocketStove.xlsx
fortheshieldedfirerocketstove
PM2.5concentrationsfortheshieldedfirerocket
PM2.5_ShieldedFireRocketStove.xlsx
stove
Carbonmonoxideandcarbondioxideemissions
CO_CO2_3StoneFire.xlsx
forthethreestonefire

PM2.5_3StoneFire.xlsx PM2.5concentrationsforthethreestonefire

Specific_Heat_Calorimeter.xlsx Calorimeterdataforspecificheatcalculation

Temperaturedataforthermalconductivity
Thermalconductivity.xlsx
measurement

CompressionStrength.xlsx Dataofcompressionstrengthtests

38

 C. HUMANSUBJECTSRESEARCHAPPROVAL:MICHIGANTECHNOLOGICALUNIVERSITY

39