Jozef KRIŠ of Secondary clarIfIerS Research field: water

Ali G. GHAWi, J. KRiŠ

Improvement performance of Secondary clarIfIerS By a computatIonal fluId dynamIcS model

Key WordS

• Clarifier• Sedimentation• Sludge• Wastewater• Solids• CFD

aBStract

Secondary clarifier is one of the most commonly used unit operations in wastewater treatment plants. It is customarily designed to achieve the separation of solids from biologically treated effluents through the clarification of biological solids and the thickening of sludge. As treatment plants receive increasingly high wastewater flows, conventional sedimentation tanks suffer from overloading problems, which result in poor performance. Modification of inlet baffles through the use of an energy dissipating inlet (EDI) was proposed to enhance the performance in the circular clarifiers at the Al-Dewanyia wastewater treatment plant. A 3-dimensional fully mass conservative clarifier model, based on modern computational fluid dynamics theory, was applied to evaluate the proposed tank modification and to estimate the maximum capacity of the existing and modified clarifiers. A Computational Fluid Dynamics (CFD) model was formulated to describe the tank is performance, and design parameters were obtained based on the experimental results. The study revealed that velocity and (suspended solids) SS is a better parameter than TS (total solids), (Biochemical Oxygen Demand) BOD, (Chemical Oxygen Demand) COD to evaluate the performance of sedimentation tanks and that the removal efficiencies of the suspended solids, biochemical oxygen demand, and chemical oxygen demand were higher in the baffle.

Ali G. GHAWIemail:[email protected]:waterrecources,wastewatertreatment

Address:DepartmentofcivilEngineering,FacultyofcivilEngineering,AL-QadisyiaUniversity,Iraq

Jozef KRIŠemail:[email protected]:waterrecources,watersupplies,watertreatmentplan

Address: Department of Sanitary and EnvironmentalEngineering,FacultyofCivilEngineering,SlovakUniversityofTechnology,Radlinského13,81368,Bratislava,Slovakia

Vol. XIX, 2011, No. 4, 1 – 11

1. IntroductIon

Inwastewater treatmentplantsaswellas inavarietyof industrialprocesses, sedimentation tanks are used to separate suspendedsolids fromwater. Sedimentation by gravity is themost commonandextensivelyappliedtreatmentprocessfortheremovalofsolidsfromwaterandwastewater.Theincreasingconcernwhichisbeingvoicedas to thedestructionandpollutionofour environmenthasproduced agrowing awareness worldwide of the need for moreeffective wastewater treatment. In addition, the contribution ofeffluents and sludge to the spread of many types of human and

animalinfectionsisnowbeingquantified.Thishasemphasizedthevitalneedforimprovedwatersuppliesandsanitation,especiallyindevelopingcountries.Theoperationofwastewatertreatmentworksare therefore no longer the exclusive domain of the engineer andchemist; multidisciplinary teams of engineers and scientists arerequiredinordertomaximizethebenefitstothecommunitywhichshouldoccurfromtheinstallationofsewagetreatment.Inmodernsocieties propermanagement of wastewater is anecessity, not anoption.Wastewatercollectedfrommunicipalitiesandcommunitiesmustultimatelybereturnedtoreceivingwatersortotheland.Thecomplex question of which contaminants in wastewater must be

2011 SlovaK unIverSIty of tecHnoloGy 1

2 Improvement performance of Secondary clarIfIerS By a computatIonal ...

2011/4 PAGES 1 — 11

removedtoprotecttheenvironment-andtowhatextent-mustbeanswered specifically for each case. The answer to this questionrequires analyses of local conditions and needs, together withthe application of scientific knowledge, economic analysis, andengineering judgment based on past experience and considerationofnationalrequirementsandregulations.Upgradingofexistingwastewater treatmentplants (WWTPs)maybecome necessary for avariety of reasons. Growth within theservice area, or the desire to serve additional areas,may result inthe need to increase the capacity of an existing treatment facility.New,morestringent requirementsmaybe imposedonatreatmentfacility, resulting in aneed to upgrade treatment processes. Olderfacilitiesmayneedupgradingtoreplaceexistingequipmentthatnolongerfunctionsasintendedortoallowinstallationofnewer,moreefficient and cost-effective technology. In this case, the objectiveof the upgrading may be to improve plant reliability or / andreduce operating cost.Of course,more than one of these reasons

may combine for aparticular plant. The subject of upgradingexisting wastewater plants is particularly important at this time.It is important both because of the large number of existingfacilities and because of the increasingly stringent requirementsimposed onwastewater treatment facilities. Anumber of studieshave investigated sediment distribution and flow patterns insedimentation tanks and clarifiers. Several of the studies (Krebs[1],Dahl,etal.[2],Krebs,etal.[3],BrouckaertandBuckley[4].Lakehal,etal.[5],JayantiandNarayanan[6],GhawiandKris[7,8,and9],havebeencarriedoutusingtheCFDmodel.The CFD study of asecondary clarifier at the Al-DewanyiaWastewaterTreatmentPlantinIraqwasundertakenwithaviewtoimproving its capacity to retain sludgeunderhighhydraulic loadconditions,whichhascomeunderpressureduetothegrowthintheprovisionofservices.Theobjectiveofthisstudyistoexaminethepossibilityofupgradingconventional secondary clarifiers in an operating wastewater

Fig. 1 Clarifier scheme at Al-Dewanyia Wastewater Treatment Plant.

2011/4 PAGES 1 — 11

3Improvement performance of Secondary clarIfIerS By a computatIonal ...

treatment plant by applying an energy dissipating inlet (EDI)(baffle) for theclarifier inlet. Inorder to achieve suchobjectives,field experiments and amathematical model (CFD model) wereconductedat themainwastewatertreatmentplantinAl-Dewanyiausing sedimentation tanks with and without EDI for secondaryclarificationofactivated-sludgemixedliquor.

2. materIalS and metHodS

TherearetwosecondaryclarifiersattheAl-DewanyiaWastewaterTreatmentPlant.Thesearecircularwitha30mdiameterby3.0mdeepwallandarecentrallyfed.Eachunitisnominallydesignedtohandle250m3/hrflow,withanequalflowrateofsludgerecycledto the activated sludge plant. The clarifiers at these plants arecentre-feedandperipheral-overflowclarifiers (Figure 1)designedfor optimum activated sludge secondary clarifier performance.Thetankgeometryandoperatingconditionsforbothclarifiersaresummarizedbelow:Clarifierdiameter=30mSidewalldepth=3.0mPeak Day Conditions: Influent Flow = 18,000 m³/d = 750 m3/h,MLSS=3,000mg/L,RASFlow=7,500m³/d=312,5m3/hSurfaceOverflowRate(SOR)=1.3 m/hSolidsLoadingRate(SLR)=132.6kg/m²/day“Typical” Settling Characteristics (from an example site with anSVIofapproximately150mL/g)

The performance and capacity of acenter-feed clarifier is verysensitivetotheintensityoftheinfluentjetsenteringtheclarifiers.Acenter-feedclarifiernaturallygeneratesastrong influent jetdueto its small center-feed area in acircular clarifier tank as showninFigure 1.The intensive center influentoftenbrings significantturbulenceintothesettlingcompartment,especiallyunderhighflowconditions.Toenhancethehydraulicefficiencyandcapacityofthecenter-feed clarifiers, one of the most important key issues is todevelopacenter-feedapparatus,whichcouldbeusedtoeffectivelyreducetheintensityofthecentralinfluentjetandturbulenceunderhigh-flowconditions.To enhance the hydraulic efficiency and capacity of the centerfeed clarifiers, the key is to develop anew center feed structure,which could be used to effectively reduce the strength of thecenter influent jet under high-flow conditions. To estimate theperformance enhancements resulting from the use of an energydissipatinginlet(EDI)inacircularclarifier,asetofcomputationalfluid dynamics (CFD) calculations were performed (Figure 2).In the first simulation, the flow through aclarifier equippedwith

acenterinletpipeandopencenterwellwascalculated.Inthesecondsimulation,thecenterinletpipewasreplacedwithanEDI.DetailsregardingthedevelopmentoftheCFDmodelaregiveninthenextsection.The CFD calculations provide estimates of the solid effluentconcentrations,returnactivatedsludge(RAS)concentrations,sludgeblanketdepthandflowdistributions in theclarifiers.Performancecomparisonsweremadeonthebasisofthesecalculatedparameters.Usingthetraditionalinfluentstructure(asshowninFigure 2),thejetoftheclarifierinfluentthroughtheinfluentslotsisverystrongduetotheverysmallcrosssectionalareaoftheslots.However,ifthe cross sectional areaof the inlet slots is simply enlarged, flowshort-circuiting (or unevenly distributed flow) may occur amongtheslots.TodesigntheEDI,theflocwelldiameterwas7.9mandthedepth1.5m.Figure 2showstheactualEDIgeometrythatwastested.Thewastewatertreatmentplantwasoperatedatdifferentflowratesto determine the effect of Hydraulic Retention Time (HRT) andSurface Loading Rate (SLR) on the performance of the clarifier.Influentandeffluentsampleswerecollectedatdifferentoperatingperiods.The liquid temperature ranged from23-29 °Cduring the

Fig. 2 Modified center inlet structure (EDI).


2011/4 PAGES 1 — 11

experiment. The samples were analyzed according to proceduresoutlinedin“StandardMethodsForTheExaminationofWaterandWastewater” , 17th edition,APHA, (1989) [10] to determine thefollowing parameters: Suspended Solids (SS), Total Solids (TS),BiochemicalOxygenDemand (BOD),ChemicalOxygenDemand(COD), Volatile Suspended Solids (VSS), Total Volatile Solids(TVS)andSettleableSolids.

3. cfd modellInG

Several computer software programs have been developed forComputational Fluid Dynamic (CFD) modelling. In this studyFLUENT6.3andthe3Dk-εturbulencemodelintheEnvironmentalEngineeringModulewasused.DuringthisstudythehydraulicCFDmodellingbeganwith thedefinitionof the settling tankgeometry.Secondly, fluid characteristics and boundary conditions weredefined. Themomentum balance, including the turbulencemodelandcontinuityequations,werethensolvednumericallyforthetankusingthefinitevolumemethod.Finally, theobtainedsolutionwaspost-processed to be properly visualised. Common mathematicalhydraulic model equations used for CFD modelling include themomentumbalancesforanon-compressibleviscousmediaandthecontinuityequation[11].

(1) (2)

In the settling model an additional scalar equation was added toinclude the concentration of the solids. This convection-diffusionequationisasfollows:

(3)

The settling velocity was modelled using the Takács exponentialsettlingfunction,thisexpressionbeingintroducedintheresolutionoftheconcentrationequation.

(4)

Thestandardk-εeddy-viscositymodelwasusedtoaccountfortheturbulenteffects.Theturbulentviscositywasdefinedasafunctionof the turbulent kinetic energy kand its dissipation rate ε by theequation:

The distributions of kand εwere determined from the followingtransportequations:

(5)

(6)

Themodel constants (Cμ,Cε1,Cε2,σk,σε) in theaboveequationshave been determined from experimental data and are set to thestandardparameters[11]:Cμ=0.09,Cε1=0.1256,Cε2=1.92,σk =0.9,σε=1.3Gbdescribestheinfluenceofthebuoyancyeffectsandisdefinedasafunctionofthesuspendedsolidsconcentrationgradient:

Theconcentrationgradient,whichreachesmaximumvaluesattheinterface between the clear fluid and the sludge blanket, hindersturbulence. The source term Gb introduced in the turbulenceequationaddressesthismatter.ThevalueofC3ε,usuallyreportedasconstant,varieswiththeratioofthegravitydirectionparallelflowvelocitywithrespecttotheperpendicularflowvelocity:

Thelaterexpressionyieldsvaluesclosetoaunityfortheunstableareas, and tends towards zero for the stratified sedimentation.ABoussinesq-typeapproachalso implies that theeffectof sludgegravityisintroducedimplicitlyasafunctionoftheconcentrationofsuspendedsolids.Itsimplementationinthemomentumequationsiscarriedoutbymeansofthesourceterms:

(7)

Thedependenceoftheviscosityontheconcentrationisempiricallyinput at different concentration ranges. The effect of the scraperbladeshasbeenusuallyeitherneglectedor introducedasuniform

2011/4 PAGES 1 — 11


constant sources, especially in the modelling of acircularsedimentationtank.However,duetothesignificanceofthescrapersystem for acircular sedimentation tank, an additional sub-modelwas incorporated to better model the effects of the transport ofsolids.Theconveyingforceexertedonthefluidwasapproximatedas afunction of fluid the velocity, including aflow regime-dependentdragcoefficient:

(8)

Different flow rates were used in each continuous experimentduring which several samples were collected from the influentandeffluentof the tank.Thesampleswereanalyzed todeterminesuspended solids, total solids, biochemical oxygen demand andchemical oxygen demand. In addition, some samples were takenfromthesettledsludgetodeterminetheconcentrationofsolids.

3.1 Boundary conditions

All theboundariescorresponding to theconcretesurfacesweremodelled using thewall functions provided by FLUENT,withasurface roughness parameter set to 0.5 mm. The free liquidsurface was represented as arigid frictionless surface. Theflow boundary conditions were set by specifying the masswithdrawal rates.Thus theoverflowrateswerespecifiedat thecomputationalcells,andtheunderflowratewasdistributedoverarowofcellscorrespondingtothesludgewithdrawalarea.Thefeed inlet to the clarifier was allowed to satisfy the materialbalancebyspecifyingafixedpressureatthecellscorrespondingto the location of the feed slots. Flow rates to be used in themodel were determined from measurements conducted on theclarifieron10July2009.

4. eXIStInG clarIfIer performance

As shown in Figure 3, the existing secondary clarifiers at theAl-DewanyiaWastewaterTreatmentPlant,oftenexperienceaveryhigheffluentTSSduetotheimpactofamassivesludgeinventory.In the overloaded clarifiers, the effluent TSS (and BOD) isextremelysensitivetoanyminorvariationsintheplantflow.Thisisbecausethetopofthesludgeblanketisclosetothesurfaceandcaneasilybecarriedovertheeffluentweirs.Theoverloadedconditionscan often cause alarge unexpected loss of bio-solids from thesecondarytreatmentprocess.The flow capacity for the two existing clarifiers studied rangesfrom500to750m3/hrduetovariationsintheprocessparameters(MLSS).Theclarifiersareunabletoachievetheirexpecteddesignflowof750m3/hrdueprimarilytothethickeninglimitationoftheclarifiers.Theperformanceandcapacityofacenterfeedclarifierisverysensitivetothestrengthoftheinfluentjetsintotheclarifiers.Atraditional center feed clarifier naturally generates astronginfluent jetdue to its smallcenter feedarea.Thus, itoftenbringssignificant turbulence into the settling compartment, especiallyunderhighflowconditions.Theexperimentsconsistedoffiverunswithdifferentinfluentflowrates to the simulate actual operating conditions of the secondaryclarifierintheplant.Eachcontinuousrunlastedforaminimumof5hours.TheinfluenttotheclarifierwasthemixedliquorfromthesecondcompartmentofahighrateaerationtankattheAl-Dewanyiasewagetreatmentplant.TheoperatingconditionsduringthetestingperiodarepresentedinTable 1.Fromtheabovetableitisclearthattherewerenotmanyfluctuationsintheinfluentcharacteristics,i.e.themixedliquorsuspendedSolids(MLSS), which could affect the performance of the tank duringtesting period. Similar to the efficiency of SS removal, theBODandCOD removal efficienciesweremore or less constant duringthe operating period at each flow rate. This emphasizes that the

Fig. 3 Overloaded clarifier operations.

Tab. 1 Operating conditions during conventional settling tank experiment.

Q HRT* MLSS SVI Temperature, (°C)(m3/hr) (hour) (mg/l) (ml/g) Liquid Air150 2.17 2085 149 29.0 33.8200 0.87 2170 148 24.8 28.6250 0.65 2770 130 30.0 33.6300 0.47 3120 131 30.6 34.3350 0.33 2390 125 27.0 30.6

*HRT:HydraulicRetentionTime


2011/4 PAGES 1 — 11

tank performancewas stable during the period studied.Also, therelationship between (Hydraulic Retention Time) HRT and theremoval efficiency of both SS andTS is shown inTable 2 andFigure 4.ItisclearfromFigure4thatwhilethe%SSremovedincreasedas the HRT increased, the% TS did not show asimilar trendsince the%TSwas almost constant, if not slightly decreasing,as HRT was increased. This may indicate that biologicalactivities took place in the sedimentation tank, especiallyduring longer HRTs, thus transforming the biological SSinto dissolved solids. Such transformations would ultimatelyincreasetheTSconcentrationduringlongerHRTs,i.e.,decreasethe%TSremovalefficiency.ThisemphasizestheimportanceofevaluatingsedimentationtankperformancebasedonSS(ratherthan TS) as is usually reported in the literature. The effect inthe case of the relationship between the SLR and the removalefficiencyofSSandTSiscontrarytothatobservedforHRTasshowninTable 2andFigure 5.

Thegoodperformanceofthesedimentationtankduringthisstudyispossiblydue to thegoodsettleabilityof thebiologicalsolidsasindicated by the SludgeVolume Index (SVI) values being in theoptimumrangeof(125-149ml/g)aspresentedinTable1.

5. reSult and dISSocIatIon

5.1 performance of clarifiers with an optimized Influent Structure

Figure 6presentscomparisonoftheComputationalFluidDynamic(CFD) modelling results for flow and solids fields between thecentre-feed clarifier described in Figure 1, in which there is noenergydissipatingapparatusaroundtheverticalcentre-feedpipe.Figures 6(a) and 6(b) present the velocity and solids fields inaselected vertical slice of the tested clarifiers. In the modelpredicted velocity fields, each velocity vector originates at agridpoint used in the CFD model. The length of each vector isproportional to the magnitude of the velocity determined by themodelfor thecorrespondinggridpoint,andis inaccordancewiththe3.0cm/sscaleindicatedinthefigures.Thefiguresalsopresenthe simulated solids fields in an identical vertical section of themodel. In this figure the contour lineswith interval of 100mg/LindicatestheSuspendedSolidsconcentration.Inacentre-feedclarifier,itisnotveryeasytoenforceflowevenlyenteringtheclarifieralongtherimofanenergydissipatingcolumnunlessenough resistancealong the radialdirectioncanbecreatedwithin the device. However, the high resistance along the radialdirectioncannotbegeneratedthroughsimplyreducingthesizeornumberoftheinletports,whichwouldincreasetheflowintensity

Tab. 2 Performance of conventional settling tank in SS and TS removal.

Q (m3/hr)

HRT (hour)

q=SLR =SOR (m3/m2.hr)

SS removal (%)

TS removal (%)

150 15.54 0.26 94.8 59.6200 11.66 0.35 94.7 67.4250 9.33 0.43 94.1 62.0300 7.77 0.52 93.6 66.0350 6.66 0.61 94.1 68.1

Fig. 4 Performance of conventional settler at different hydraulic residence times.

Fig. 5 Performance of conventional settler at different surface loading rates.

2011/4 PAGES 1 — 11


enteringintotheclarifiers.TheEDIisabletosimultaneouslysatisfybothof theenergydissipatingprinciples, i.e.alargeaccumulativespaceofinletportsandauniformflowdistributionamongalloftheinletportsduetothemultilayerflowimpingement.Figure 6consistsofthetwopartsof6(a)and6(b)withrespecttothetwotestedclarifierswithandwithnotheEDI,respectively.AsshowninFigure 6(a),theCFDmodellingresultsfortheclarifierequippedwithasimplecentreinfluentpipeindicate:1. Thestronginfluentjetthroughtheinletports(2)penetratesthe

entire radius of the flocculationwell (3) and impinges on theinnersideofthewell(3)duetothelackofeffectivemomentum/energydissipating facilitieswithin the flocculationwell.Afterimpinging on the flocculationwell, the influent flow deflectsand forms avery strong downward current toward the sludgeblanketandclarifierfloor(5).

2. Significant reverse flow is predicted underneath the strongsurfaceinfluentjetduetotheshearsbetweenthem.

3. Apinchedclarifierinfluentflowunderthebafflelip(3)canbeobserved due to themassive sludge inventory in the clarifier.

Thedensityoftheforwardcurrentismuchclosertothewatersurfacethanthatpredictedunderalowerflowconditionduetothebuoyancyimpactofthethicksludgeblanket.

As shown in Figure 6(b), the modelling results for the clarifierequippedwithaEDI(8)indicate:1. Due to the small influent ports the strong influent jet (2)

continuously impinges on the multilayer perforated columns(8)oneafter theother.Thevelocitiesof the influent jetshavebeen substantially reducedbefore and after going through theports(9)inthelastperforatedlayer(8).Theresistancecreatedby themultiple perforated columns (8) forces the influent jetto be sufficiently distributed along the vertical and tangentialdirectionsbeforeitentersintotheflocculationwell(3).

2. DuetoThedownwardcurrentthedeflectionoftheinfluentjeton the flocculation well (3) , has been significantly reduced,sincethemomentumoftheinfluentjetiseffectivelydissipatedbyapplyingtheEDI.Thecircularbottom(10)forcesalloftheinfluentflowtogothroughthestaggeredports(9)andprevents

Fig. 6 Performance before (a) and after (b) central inlet retrofit.


2011/4 PAGES 1 — 11

the flow from short circuiting between the inlet ports (2) andflocculationwell(3).

3. Thepinchedflowunderneaththelipofthebaffle(flocculationwell)(3)hasbeeneliminated,andthelevelofthedensityoftheforwardcurrent ismuchcloser to theclarifierfloor(5)duetotheloweredturbulenceandthewellcontrolleddispersedsludgeblanketintheclarifier.

4. Thesignificantreverseflowunderneaththesurfaceinfluentjetpredicted in theexistingclarifiershasbeenalmosteliminated,since the significantly slowed influent jet generates amuchweakershearinfluenceontheambientflow.

Theexistingclarifiershaveflowcapacitiesofapproximately750(m3/h) under the normal process condition, which ismost of theyear.Theoptimizedclarifierscanachieveaflowcapacityofaround1300(m3/h),whichis30%higherthanthatoftheexistingclarifiers.The performance of the EDI (Baffle) was examined by applyingnine different influent flow rates ranging from 150 m3/hr to

350m3/hr in separate mathematical model runs. The duration ofeachcontinuous runwas at least5hours,duringwhichdifferentsampleswerecollectedfromthe influentandeffluentof the tank.The main parameters (i.e. SS, TS, BOD, etc.) were determined,and the removal efficiencies were calculated at different influentflow rates. The performance was stable during each operatingperiod studied. The values of HRT in the tank were calculatedforeachmathematicalmodelrunasillustratedinTable3,andthecorrespondingSLRvaluesarepresentedinTable 4.The relationships between HRT and the removal efficiencies ofboththeSSandTSwereestablishedaspresentedinFigure 7,fromwhich it isclear that the removalefficiency increasesas theHRTincreases.Figure 8 shows the relationship between the SLR andremovalefficiencies forboth theSSandTS. It isevident that theefficiencyoftheremovaldecreasesasSLRincreases.Suchtrends

Tab. 3 Operating conditions during EDI (Baffle) experiment. Q

(m3/hr)HRT

(hour)MLSS (mg/l)

SVI (ml/g)

Temperature, (°C)Liquid Air

150 2.04 1735 116 24.7 25.2175 1.22 2470 122 28.8 29.3200 0.82 2172 461 22.2 27.2225 0.61 1784 476 21.9 27.6250 0.51 2256 147 22.2 28.0275 0.44 2308 208 23.5 27.9300 0.38 1561 547 21.7 28.1325 0.34 2494 128 24.9 27.9350 0.31 2093 107 23.8 26.8

Tab. 4 Performance of EDI (Baffle) in SS and TS removal.Q

(m3/hr)HRT (min)

SLR (m3/m2.hr)

SS removal (%)

TS removal (%)

150 2.04 0.24 97.7 56.1175 1.22 0.28 97.5 69.2200 0.82 0.36 97.9 54.6225 0.61 0.39 97.5 43.8250 0.51 0.425 97.1 64.2275 0.44 0.461 96.7 47.6300 0.38 0.52 94.7 46325 0.34 0.563 97.2 64.8350 0.31 0.607 96.2 66.1

Fig. 7 Performance of EDI (Baffle) at different hydraulic residence times.

Fig. 8 Performance of EDI (Baffle) at different surface loading rates.

2011/4 PAGES 1 — 11


aresimilartothoseobservedintheconventionalsedimentationtankregarding thepercentage removalofSS andTS in relation to theHRTandSLR.Inthesemathematicalmodelrunsontheupgradedsedimentation tank,similarobservations to thosemadeduring theexperiments on the conventional sedimentation tankwere evidentregardingthetrendsinTS,BOD,andCODremoval.

5.2 comparison between conventional and edI (Baffle) Sedimentation tanks

Inorder toperformsuchacomparison, theremovalefficiencyforSShasbeendeterminedforboth typesofsettlersat fivedifferentinfluentflowratesrangingfrom150m3/hto350m3/h.Comparingtheresultsobtainedfromoperatingthemathematicalmodelofthetankasaconventionalsedimentationbasinandasahighratesettler(EDI),i.e.withouttheEDI(baffle)andwiththeEDI(baffle),itisapparentthatduringanoperationwiththeEDI(baffle)theefficiency oftheSSremovalisbetterthanincaseofconventionaltankby2%

-3%,whichisamarginalincreaseinefficiency.However,thetankwith an EDI (baffle) was capable of maintaining high removalefficiencies even when the biological solids had ahigh SVI asshowninTables 1 and2,knowingthathighSVIvalues(>200ml/g)areindicativeofpoorsludgesettleability.ThemeritsoftheEDI(baffle)ismoreapparentwhensettlingratherthan thickening is controlling the tank design. Thismay indicatethattheapplicationofanEDI(baffle)inthesecondaryclarificationofbiologicalsludgemaynotbeasadvantageousasitsapplicationin the primary clarification ofwastewater solids.However,whensecondary clarifiers are overloaded or suffer from rising sludgeproblems, upgrading of such clarifiers using an EDI (baffle) isdefinitely advantageous.This is in addition to savings in costs oflandareacoveredbysettlerswhichismuchlessincaseofanEDI(baffle) than in case of conventional type gravitational settlingtanks.BasedontheresultsobtainedfortheEDI(baffle),astatisticalmodelcouldbeformulatedbyapplyingalinearregressionanalysisto the relationship between the SLR and%SS removal.Figures 9, 10, and 11 illustrate the relationship obtained which could beexpressedbythefollowingequation:

% SS removal = 98.26 -1.39 SLR ........................................ (9)

ThisisastatisticalmodeldescribingtheremovalefficiencyofSSintheupgradedEDI(Baffle).Similarly,thefollowingequationswhereobtainedforBODandCOD:

% BOD removal = 96.20 - 1.01 SLR .................................... (10)

% COD removal = 95.50 - 0.8 SLR ...................................... (11)

Fig. 9 Effect of surface loading rate on SS removal of EDI (Baffle).

Fig. 10 Effect of surface loading rate on BOD removal of EDI (Baffle).

Fig. 11 Effect of surface loading rate on COD removal of EDI (Baffle).


2011/4 PAGES 1 — 11

6. concluSIonS

Thefollowingconclusionscanbedrawn:1. The EDI (baffle) has proved effective in improving

theperformance of the secondary sedimentation of biologicalsolids at the studied surface loading rates in arangeof0.2 to1.6m3/m2hr.

2. In comparisonwith aconventional settler, the EDI (baffle) islessaffectedbyoverloading.Ifthedesignsurfacesloadingratecriteria for conventional settling tanks are used for designinghigh-rate settlers, the latter should perform better within therangeofsurfaceloadingratesnormallyusedinpracticaldesign.

3. The efficiency of the removal of solids increases with theincreaseofHRTanddecreaseofSLR.

4. Theefficiencyofsuspendedsolidsremovalisabetterparameterfordescribingtheperformanceofsedimentationtankscomparedto total solids. Meanwhile, the biological transformations ofsolids in the secondary sedimentation tank could contributetoBODandCODwhichresults inhigherBOD/SSandCOD/SS ratios in the effluent than in the influent.This emphasizesthe uniqueness of SS as abetter parameter in performanceevaluation.

5. The main advantage of an EDI (baffle) in the secondarysedimentationofbiologicalsolidslies initscapabilitytocopewithplantoverloadingconditions.Suchsettlerscouldbeeasilyinstalled in an existing rectangular sedimentation tank asasolutiontorisingsludgeproblemsatminimalcostcomparedto other solutions such as increasing tank depth, addition ofchemical coagulants,...etc. Installation or removal of an EDI(baffle) would not interfere with the normal operation ofexistingsedimentationtanks.

7. acKnoWledGementS

The article has been written with the support of the VEGA1/1143/11grantresearchtaskssolvedattheDepartmentofSanitaryEngineering of the Faculty of Civil Engineering of the SlovakUniversityofTechnology,Bratislava.

NomenclaturesSymbol Description

FAvolumeforceterm(N/m3)whichiszeroinboththexandydirections.

U Theaverageflowvelocityvector(m/s)P Theaveragepressure(Pa)η Dynamicviscosity(Pa.s)ρ Density(kg/m3)t Time(s)Cμ Amodelconstantk Theturbulentkineticenergy(m2/s2)ε Thedissipationofturbulentenergy(m2/s3)C Theconcentrationofsolids(mg/l)Us Thesettlingvelocity(m/s)σc TheSchmidtnumber(0.7)νt TheturbulentviscosityUso Thereferencesettlingvelocity(m/s)

rh,rpInducethedominationofthefirstandthesecondtermforthefallingandtherisingpart

Cns Thenonsettleableconcentration(mg/l)CD Thedragcoefficientρ Thefluiddensity(kg/m3)Vt Theblade-to-fluidrelativevelocity(m/s)A Thescraperdisplacementarea(m2)CFD ComputationalFluidDynamicsRTD ResidenceTimeDistributionSS SuspendedSolidsTSS TotalSuspendedSolidsEDI EnergyDissipatingInletHRT HydraulicRetentionTimeBOD5 BiochemicalOxygenDemandTVS TotalVolatileSolidsCOD ChemicalOxygenDemandVSS VolatileSuspendedSolidsSOR SurfaceOverflowRateQ Discharge(m3/s)

2011/4 PAGES 1 — 11


REFERENCES

[1] Krebs,P. (1991).Thehydraulicsof final settling tanks,Wat.Sci.Tech.,23Kyoto,1037-1046.

[2] Dahl, C. Larsen, T. & Petersen, O. (1994). Numericalmodellingandmeasurementinatestsecondarysettlingtank,Wat.Sci.Tech.30(2),219-228.

[3] Krebs,P.Vischer,D.Gujer,W.(1995).Inlet-structuredesignfor final clarifiers, Journal of Environmental Engineering,121(8),558-564.

[4] BrouckaertC.J.BuckleyC.A.(1999).Theuseofcomputationalfluid dynamics for improving the design and operation ofwaterandwastewatertreatmentplants.Wat.Sci.Tech.,40(4-5),pp.81-89.

[5] LakehalD.,KrebsP.,KrijgsmanJ.RodiW.(1999).Computingshearflowandsludgeblanketinsecondaryclarifiers.J.Hydr.Engrg.,125(3),pp.253-262.

[6] Jayanti S. Narayanan S., (2004). Computational studyof particle-eddy interaction in sedimentation tanks,J.EnvironmentalEng.,130(1),37-49.

[7] GhawiA.HadiJozefKris(2007a).”DesignandOptimizationof aSedimentation Tank in Slovakia with CFD Modeling”10th International Symposium on Water Management andHydraulic Engineering 2007 with special emphasis onthe impact of hydraulic engineering construction on theenvironment,4–9September2007.Šibenik,Croatia

[8] GhawiA. Hadi, Jozef Kris (2007b).ANumericalModel ofFlowinSedimentationTanksinSlovakia,ThirdInternationalPhD Symposium in Engineering, 25-26 October 2007,HungaryUniversityofPollackMihályFacultyofEngineering,Pécs,Hungary

[9] Ghawi A. Hadi, Jozef Kris (2007c). Improved, Modeling,Simulation and Operational Parameters of Settling Tank.6th InternationalConferenceofPhDStudents,UniversityofMiskolc,Hungary,12-18August2007,pp.69-75.

[10] APHA (1989), “Standard Methods for the Examination ofWaterandWastewater,”17thedition,AmericanPublicHealthAssociation,Washington,D.C.

[11] WilcoxD.C.,TurbulenceModellingforCFD,DCWIndustries,Inc.,LaCanada,California,(1998).

[12] Pollert, J. jun., Pavlíčková, I., Todt, V.: OptimalizacedosadzovacíchnádržíÚČOVPrahamatematickýmmodelem(Optimalization of secondary clarifier Praha by theemathematicalmodel).In:Městskévody,2010,Brno,ARDEC,2010,113-121,ISBN978-80-8602071-6.

[13]Pollert,J.jun.,Koníček,Z.,Thoeye,Ch.,Boonen,I.,Gunther,P.:Optimalizationofsecondaryclarifierusing3Dmodelingofsludge.In:WorldWatercongressandExhibition-Abstracts.Oxfod:IWA,2008.

Documents

Jozef KRIŠ of Secondary clarIfIerS Research field: water