Radiant Floor With Phase Change

8/21/2019 Radiant Floor With Phase Change

1/29

ThermalDynamicsofRadiantFloorHeatingwith

WoodenFlooringSystems

ByWarren

Cent

Abstract

Radiantfloorheatingdesignsvaryinmaterialandconstructionmethods. Finiteelementmodelswere

developedusingexistingandproposedfloordesignswiththeintentofcomparingheatingperformance.

Performancemeasuresincludeduniformityoftemperatureacrossthesurfaceofthefloor,lengthof

timenecessarytoreachtheoptimumheatingrange,andrateoftemperaturedegradationafterthe

heatingsystemhasbeendeactivated. Flooringsystemsmodeledincludedslabongradeandwooden

joistandsheathing. Surfacematerialsincludeconcrete/mortar,tile,finishwood,andaderivativeof

paraffinwaxandpolyethylenerepresentingashapestabilizedphasechangematerial(SSPCM)below

thesurface.

The

repeated

heating

cycle

used

was

pumped

water

warmed

to

50C

followed

by

5

minutes

ofdeactivation. Thecyclerestartedfortwoadditionalcyclesandthesystemwasthendeactivated. This

studydiscoveredthelowertheconductioncoefficientofthesurfacematerialis,thelowerthe

temperaturevarianceisonthesurfaceandthesystemfoundalowerdecreaseintemperatureovertime

afterthesystemhadbeendeactivated. ItwasfoundthatintroducingaSSPCMtoafloorsystem

increasedheatcapacityandcreatedauniformtemperatureoverthesurfaceofthefloor.

Introduction

Methodsofradiantfloorheatinghavebeenusedinmanyformsthroughouthistory. Alongwith

providing

an

efficient

method

of

delivering

heat,

radiant

floor

heating

systems

provide

a

better

distributionofheat,idealforhumanbodies. Manysystemshavebeendevelopedutilizingbothpassive

andactivemechanicsthattrapanddirectheatthroughthesystem. Moderntechnologiesemploylaying

wireorpipeandencasingitwithinconcreteormortar,makinguseofelectricorheatedwatersupplies.

Researchhasbeenreportedonthedynamicsandeffectsofradiantfloorheatingwithdensematerials

suchasstone,concrete,tile,andmortarbutlesshasbeendonewithflooringsystemsinvolvingwooden

products. Inseveralrespects,woodensystemswouldbepreferredoverheaviersystems. Woodhas

lowerthermalconductivity,similartothatofinsulation,thanmanyotherconstructionmaterials,

allowingforaslowertransferofheatthroughthematerial. Also,manyprojectsthatwouldmakeuseof

radiantfloorheating,suchashomesandlowriseconstruction,usewoodastheirmainconstruction

material. Findingmethodsofutilizingradiantfloorheatingwiththeuseofwoodenmaterialswouldnot

requirelarger,heavierthermalmassingtobeusedinastructureandwouldmakeuseofthematerial

alreadyinservice. Findingmethodsofincorporatingradiantheatingintoexistingandfuturewooden

constructionprojectswouldcreateanadvantageforbothbuildingconstructionandqualityofhabitation

fortheusersofthespace.


2/29

Withadvancementsinwoodcompositetechnologies,arangeofshapesandmaterialcombinationsare

viable. Todate,mostofthesecompositesystemshavebeendevelopedtooptimizeengineering

mechanicalproperties. Anotherpossibilityinradiantfloormethodsistoincorporatephasechange

materialsintowoodcompositesasameanstomanipulatetheenergystorageanddynamicsofthe

system. Researchisneededtoidentifynovelsystemconfigurationsandmaterialsthatwouldoptimize

thermaldynamics

for

radiant

floor

heating.

Objectives

Theoverallgoalofthisstudyistoexplorenewwaystoincorporateradiantfloorheatingwithwood

constructionmethodsfortwosituations:1)retrofittingtraditionallyframedwoodfloorsystemsand2)

incorporatingwoodenmaterialsintonewradiantfloorheatingsystems(aswellasphasechange

materials)withinnovativeassemblyconfigurations. Specificobjectivestoaccomplishthisgoalinclude:

1. identifykeyperformancetargetsforproposedradiantfloorsystemsincludingheattransferandstorageproperties

2. developconceptualdesignsofsystemsusingcurrentconstructionmethods3. conductparametricfiniteelementmodelstudiesofalternatesystemstodevelopdeeper

understandingsofrelativeadvantagesanddisadvantages

4. characterizethermaldynamicsofselectedsystemsandcontrastagainstothersystems5. Developrecommendationsofradiantfloorsystemsthatwarrantfurtherstudy

Approach

Publishedmathematicalandphysicalmodelswerereviewedtounderstandpreviouslyexploredmethods

ofradiantfloorheatingandtogiveinsighttothefiniteelementmodelingforthisstudy. Understanding

the

advantages

that

different

materials

bring

to

a

heating

system

is

important

in

the

pursuit

of

finding

betterradiantfloorheatingdesigns. Byalsoidentifyingtheshortcomingsofthesystemsandmodeling

methods,greateraccuracycanbeobtainedandencouragebetterresults.

Toapproachfindingthemostbeneficialsystemlayouts,severalmeansofmeasuringperformancewere

identified. Thecharacteristicsthatthefiniteelementmodeloutputsweremeasuredagainstincluded:

1. Lengthoftimethatasystemtakestoreachoptimumheatingtemperature2. Uniformityoftemperatureacrossthesurfaceofthefloor3. Slopeoftemperaturegradientovertimeastheheatingsystemisdeactivated

The

length

of

time

necessary

to

heat

a

system

to

an

active

heating

temperature

was

studied

because

a

rapidlyrespondingsystemcouldbebeneficialforsystemswithseverallocalizedzonesandsetback

thermostatcontrols.Ontheotherhand,asystemwithmorethermalstoragewouldtakelongertoreach

targettemperature,butcouldreducecyclingandtherebysavepeakenergyrequiredforstarting

circulationpumps.

Measuringtheuniformityoftemperatureacrossthesurfaceofthefloorisanimportantattributewith

regardtooccupantcomfort. Hotspotscanoccurwhenthesurfaceareaoffloordirectlyabovean


3/29

embeddedpipereceivesheatdisproportionallycomparedtotheareathatisnotdirectlyabovethepipe.

Findingsystemsthatencouragethoroughdistributionofheatoveritssurfaceduringitsheatingphase

wereconsideredbeneficial,anditwasquantifiedasthedifferenceintemperaturefromthewarmestto

thecoldestpointacrossthesurfaceofeachmodel.

Asa

flooring

systems

heat

source

is

deactivated,

the

heat

gained

within

the

material

will

be

extracted

bytheoutsideeffectsonthesystem. Althoughitisdifficulttomodelsaideffects,somesystemsstore

heatenergyandreleaseitslowerthanothersunderthesameloadingconditions.Thisisanissuefor

energyefficiencybecausetheheatsourcewouldhavetoreheattheentireflooringsystemoncethe

floorhasreleasedtheheat. Thesystemsthatretainheatbetterwouldrequirelessheatinginputfrom

theheatingsourceandtighterparameterscanbesettoensuregreaterthermalcomfortwithless

energyexpense. Thesystemsthathavelowerratesoftemperaturedegradationaftertheheatsource

hasbeenturnedoffwouldrankhigherthanthosethatdonot.

TraditionalRetrofit

Retrofitof

traditional

wooden

floor

systems

represents

a

huge

potential

market

for

radiant

floor

heating

systems. Betterunderstandingthedynamicsofthissystemisimportanttohavesuccessful

implementations. Manyhomesandlowriseconstructionprojectshaverealizedbenefitsfromthese

systems. Therearevaryingdegreesofimpacttoaprojectdependingontheradiantfloorheating

design,anditisimportanttoidentifyifspecificsystemshavethecreatelowerlevelsofimpact.

Structuralsystemsdovaryfromprojecttoproject. Knowingifaparticularsystemcanaccommodatea

radiantfloorheatingdesignduetoitsthermalandgravityloadingisimportanttoexploreandanalyze

beforeinstallationoccurs. Engineeringanalysisandjudgmentwillhavetobemadetodetermineifthe

structuralsysteminquestioncanaccommodatethepreferredheatingsystem. Someprojectsmaynot

allowfor

the

installation

of

specific

systems

and

it

may

be

more

advantageous

to

upgrade

the

structural

systemalltogether.

Asfarasfiniteelementmodelingisconcerned,manyofthethermalpropertiesareresearchedandthe

propertiesarepublishedforconventionalbuildingmaterials.

WoodComposites

Compositematerialsofferawidevarietyofstructuralandthermaladvantagesinfloorcrosssections.

Ratherthanbeinglimitedbyshapeduetotheresource,compositescanprovidenovelgeometriestofor

thebenefitofthesystem. Thisstudywillhelpgiveguidancetofutureresearchtodevelopcrosssections

toaccommodate

radiant

floor

heating

layouts

and

to

utilize

the

advantages

available

to

them.

Finiteelementmodelingofcompositematerialscanbechallengingduetocomplexcrosssection

geometriesandavailablepropertiesforsomematerials.

PhaseChangeMaterials


4/29

ThecaseforphasechangematerialsisstrongbecauseSSPCMscandramaticallyincreasetheenergy

storagecapabilitiesofthefloorsystemsbymakinguseofitslatentheatcapacityduringphasechange.

Phasechangematerialswouldincreasethesystemsabilitydramaticallytoprovidewarmthtoaspace,

butitspropertiesmustbecontrolledandunderstood. Modelingcangiveinsightsonappropriateways

tointroduce

phase

change

materials

into

a

radiant

floor

heating

system.

LiteratureReview

HealthandComfortConsiderations

Currentinterestinradiantfloorheatingmethodscomesfromitspotentialhealthbenefitsaswellas

bettermanagementofenergyandresourcestocreateabetterlivingenvironment. Radiantfloor

heatingcanbemeasuredforthermalcomfortineverydaylivingenvironments(Song,2005). Beforethe

layoutofwiringandpiping,atmosphericcontrolsandsensors,andmaterialpropertiesareeven

considered;radiantfloorheatingpracticescanbeshowntoachieveindividualthermalcomfortbetter

thanconventional

heating

systems

(Woodson,

1999).

That

alone

could

be

the

best

reason

for

investigatingthemeansofimplementingsuchsystemsinfutureprojects.

Thehealthbenefitsthatradiantfloorheatingprovidesarefoundfromtheidealheatingcurve(Figure1),

whichisacharacterizationoftheidealtemperatureforthehumanbodybasedonspecifiedheightsfrom

thefloor(Woodson,1999;Song,2005). Generally,becausethefeetareindirectcontactwiththe

ground,theylosegreaterquantitiesofheatduetoconduction. Theheadhasthelargestamountof

bloodvesselspersurfaceareaonthebodymakingitoneofthemoreheatedareasofthebody. These

twoconditionscreateboundsforanidealdistributionofheattoprovidetothebodygivenaheightfrom

thefloor. Byapplyingtheboundaryconditionstothestructureofthecurve,agreatertemperatureis

needed

closer

to

the

ground

and

a

lower

temperature

is

needed

furthest

from

the

ground

with

a

smoothtransitionbetweenthem.

Figure1:RadiantFloorHeatingvs.IdealHeatingcurve Figure2:ForcedAirHeatingvs.IdealHeatingCurve


5/29

Theseboundsoftheidealheatingcurvearenotmetwithconventionalheatingsystems(Figure2). A

naturalactiontoexpectfromheatedliquidsandgasesisthatthewarmedparticleswillrisecausingthe

coolerparticlestofall. Bythisthermodynamicunderstandingofparticlebehavior,aheatedroomwould

locatecoolerairnearthefeetandwarmerairnearthehead. Thisbehaviorproducestheinversetothe

idealheatingcurvesrequirements. Manyconventionalsystems,suchasforcedairandbaseboard

heating,are

subject

to

produce

the

reverse

ideal

heating

curve

while

radiant

floor

heating

produces

the

idealcurvecloserthananysystem(Woodson,1999). Byheatingthefloorfirst,thefeetreceivetheheat

firstandastheheatrises,itiscooledcausingtheupperregionsoftheroomtobecooleraspreferred. It

isbythistransferofheatthatwegainunderstandingforradiantfloorheatingsusetodayandinthe

past.

ConstructionMethods

Radiantfloorheatinghasbeenconstructedusingmanydifferentmediumsandmethodsthroughout

humanhistory. Traditionally,methodshaveincludedguidinghotairunderneaththecrawlspaceofan

elevatedroombetweenjoistsorbetweenthelayersofmasonrywalls(Song,2005;Song2008). With

theintroductionofelectricalwiring,copperpiping,andcastinplacematerials,radiantfloorheating

methodshavediversifiedandevolvedtoprovideefficientsolutionstoheatinglivingenvironments.

OneofthemostpopularandmostexploredmethodsofradiantfloorheatingusedtodayistheOndol

Systemwhichincorporateshydronicpipingcastwithinaconcreteormortarbed. ThenameOndol

comesfromtheoriginalKoreanpronunciationGudeul(guundol)meaningheatedstone.The

pronunciationovertimehastransformedintoOndol. Thismethodofheatinghasbeenknownforover

2000years(Song,2005).

Periodically,waterisheatedandpumpedthroughthenetworkandconvectivelytransfersheattothe

surroundingmaterials

(Cho,

2003).

Electric

wiring

systems

are

also

used

as

heating

elements.

Rather

thanconvectivetransference,electricalresistanceisusedtogeneratethermalenergyatthesource.

Bothpopularsolutionshaveadvantagesandsuitdifferentscenarios,butthemainfocusinthisstudywill

behydronicsystems.

Modeling

Thepotentialbenefitsofradiantfloorheatinghavecultivatedtheinterestofresearchtodiscoverbetter

methodsofdeliveringheatandwithbetterenergyefficiency. Arangeofmodelingmethodsareuseful

toolsinthisregard. Therearetwovantagepointsthatallowresearcherstodevelopenergyefficient

designs;macroandmicroscalemodeling. Macroscalemodelinghelpsresearcherstounderstandhow

thebuiltenvironmentasawholewillperformwithallthesystemschosentomaintainthecomfortof

thespace. Bymodelingcharacteristicssuchasthebuildingenvelope,orientation,materials,daylighting,

naturalandHVACventilation,exteriorclimate,andoccupantloading;softwareprogramssuchas

EnergyPlusandDOE2,tonameafew,canaccuratelyevaluateoptionsofhowtobesttodesignaspace

(Crawley,2005).


6/29

Microscalemodelingcanbeasusefulintandemwithmacroscalemodeling. Microscalemodelingis

usedtoevaluateindividualsystemsandtheabilitiesofspecificcomponentsofthosesystemstofurther

theefficiencyofthewhole. Strictlyfromamathematicalstandpoint,systemsandcomponentsmaybe

modeledtoinvolvethemostefficientlayoutandmaterialtypegivenspecifiedboundaryconditionsto

helpguidethedevelopmentofafiniteelementmodel(ZaheerUddin,1997). Macroscalemodelingcan

playan

important

role

by

eliminating

characteristics

of

the

heating

system

that

cause

little

effect

despite

modificationandbyisolatingwhichattributesaffectthesystemsperformancegreatly. Ifahydronic

systemisselected,severalimportantattributestosimulateincludethicknesscovering,coveringmaterial

properties,andfrequencyofpipingloopsalongthelengthofthespace. Whilethereareothermaterial

andsystempropertiesthatcharacterizeasystemandgaugeitsefficiency,theseareamongthemost

importanttoconsiderwhendevelopingamacroscalemodel(Sattari,2006).

Methodsofimprovingtheaccuracyandperformanceofradiantfloorheatingsystemsinvolveincreasing

sensorsandcontrolscapabilitiesoftheheatingsystem(Cho,2003;Cho,1999;Song,2008). Inmost

traditionalheatingsystems,theconventionalonoffthermostatistheextentofthesystemssensory

ability.Methods

to

increase

the

sensory

capabilities

include

using

the

conventional

thermostat

in

tandemwiththermostatswithinthefloorsystem. Workingthroughanalgorithmestablishedwithinthe

heatingsystem,iftheairorslabtemperaturedropsbeloworexceedsthethermalboundaries,the

systemisrespondstopreventneedlesslossandrechargeofheatintheheatingelements(Cho,1999).

Othermethodsofincreasingtheefficiencyofradiantfloorheatingsystemsbymeansofsensory

capabilitiesincludepredictableactionsbymeansofmeteorologicaldata. Recommendedcyclesof

heatingareencouraged,characterizedbylengthandduringsettimeswithina24hourperiod,basedon

thetimeofyearandthehighandlowoutsidetemperaturesforthelocationthesystemisoperatingin

thatday. Bypredictingthehighandlowtemperaturesforthecoming24hourperiod,theheating

system

can

accurately

shift

its

heating

cycles

duration

and

activation

time

to

accommodate

the

weather

(Cho,2003).

EnergyEfficiency

Radiantfloorheatingsystemsgenerate,store,andsupplyheatbetterthanconventionalsystemsand

withtheintroductionofmaterialwithbetterthermalstoragecapabilitieswouldamplifythebeneficial

effects. Materialsabsorbandtransferthermalenergyatdifferentrates. Thosethatdischargethermal

energyatslowerratesenablethespacetostayheatedwhilerequiringlessoftheheatingsystemto

chargethespaceitself. Astateofmatterthattransfersheatwell,whilemaintainingaconstant

temperature,isduringthephasechangeofthatmaterial. Introductionofphasechangematerialsto

radiantfloor

heating

systems

could

enhance

thermal

storage

of

heat

extend

beyond

the

floor

and

can

beapartofthewallsandceilingscapabilitiesaswell(Neeper,2000;Khudhair,2004). Byincreasingthe

thermalstoragecapabilitiesofthespace,peakheatingdemandscanbereduced,withoverallenergy

savings. Manymethodshelpradiantfloorheatingsystemsperformefficientlyandreliably.

Incorporatingabalancedmixtureofmethodswillproduceperformanceresultsformanydifferent

projectsanddesigns.


7/29

ModelDevelopment

ThefiniteelementmodelingprogramusedinthisresearchwasADINA. Thefloorcrosssectionsthat

weremodeledincludedexistingconventionalradiantfloorheatinglayoutsandretrofitinstallation

designs(Sattari,2006;Ho,1995;Zhang,2006;Woodson,1999). Atraditional100mmconcreteslabon

gradewas

determined

as

the

base

line

flooring

system

to

compare

the

wooden

flooring

systems

to.

Othersystemsincludeslabongradewithinsulationunderlayment,conventionalwoodenjoistsand

sheathingtoppedwithavarietyoffinishmaterialsincludinghardwoodboards,tile,andShapeStabilized

PhaseChangeMaterial(SSPCM)overlaidwithsubfloorsheathing.

Forsimplicityofconstructionandefficientuseofcomputerdatastorage,finiteelementmodelswere

developedfora0.5meterlengthsectionforthefloorsystemscrosssection. Thislengthwaschosen

because:

1. Itallowsforaconventionalwoodenjoistfloorsystemtoberepresented2. Itallowedforavarietyoflateralspacingdistributionsofpipe(100mm,150mm,200mm)3. Symmetrycanbeutilizedtoextenttheheatingresultsalongtheexpanseofafloor

Severallateraldistributionsofpipingweremodeledinpreviousstudies(Song,2005;S.Sattari,2006)that

rangedfrom50to300millimeters. Spacinglengthsbetween100to200mmwereabletoshowthebest

uniformityofheat.

Verticaldistributionswerealsovariedintheslabongradesystemstomonitoritseffectonthesystem.

Heightsvariedfrom37.5to75mmfromthetopoftheslab. Allotherfloorsystemsdidnotvarythe

verticaldistributionbecausetheencapsulationlayerwasreducedinsizetopreventexcessweighton

theexistingstructuralsystem,leavinglittleroomtoexperimentwiththeheightofthepipingwithinthe

layer.

Oneretrofitsolutioninvolvedutilizingtheradiationemittedfromtheheatedpipingasthemainsource

ofheat(Woodson,1999). Thissolutionisoflowerimpacttoanexistingprojectbecauseitdoesnotrely

onintroducingheavierthermalmassing. Thesolutioninvolvesrunninganaluminumtroughhungfrom

thejoiststoincreasetheamountofthermalradiationabsorbedbythesheathing. Duetocomplexities

thatwouldariseinthemodelingprocess,theheatexchangebetweenthepipingnetworkandthe

surfacessurroundingthepipeswerecalculatedasaheatfluxratherthanaradiationbody. The

assumptionsthatledtothevaluesusedaresummarizedinTableA1.

AninnovativesolutioninvolvestheuseofSSPCMplatesbelowthesubflooringtostoreextraheatduring

itsphase

change

in

the

form

of

latent

heat.

Modeling

a

material

that

went

through

phase

change

and

incorporatingitslatentheatcapacityinADINAmeantcreatingamaterialwithatemperaturedependent

heatcapacity. Theamountofenergythataunitvolumeofthematerialwouldgainorloseduringphase

changewassplicedinduringoneunitdegreeoftemperaturechangeatthemeltingpoint. Althoughnot

entirelyaccuratethatamaterialwouldcontinuetochangetemperatureduringphasechange,the

amountofenergyisstillaccountedformathematically(EquationA1).


8/29

Theremainderofthematerialschosenintheresearchpossessedconstantconductionandheatcapacity

coefficientsandwouldnotvaryovertimeortemperature.

Theinitialconditionsofthefloorsystemsweresetto10Cunderthescenarioofasystemwakingup

withnopreviousactivation. Thistemperatureiscommoninafloorsystemthathasundergonecooling

overthe

night.

The

optimum

temperature

that

will

be

targeted

to

reach

will

be

ideal

for

active

heating

andforhumancomfort. Therangeofacceptabletemperaturesis24C 32C(Song,2005;Woodson,

1999).

Introducingloadingscenariostothemodelsinvolvedsettingboundaryconditionsonthetopofthe

flooringsystemtosimulateaircurrentswhiletheheatwasbeingintroducedtothesystemsthroughthe

pipingnetwork. Bothweresimulatedasconvectivecurrents. Theheatingconditionswithinthepipes

weresetatatemperatureof50C(Sattari,2006;Song,2005)whileactiveandwereappliedinpulsesof

heatedwaterfor15minutes,releasedfor5minutes,andthenthecyclerepeats. Thispulsingcondition

wasusedtobettertesteachflooringdesignforhowmuchtimewasnecessarytobringthedesignto

optimumtemperatureandtogaugeitsreactionduringvariedheatfluxes.

Theheatedwaterwasgivenatimedependentconvectivecoefficientthatwouldsimulatethepulsesof

thepumpingactionduringthe15minuteintervalofitsactivation. Afterthe15minutepulsewas

ended,theconvectivepropertywaseffectivelyreducedtozerotoindicatethatthewaterwasstill.

Afterthe5minuterechargeperiod,theconvectivecoefficientregaineditsoriginalintensity. These

cycleswouldrepeatthreetimesforthelengthof1hourandthenacooldownperiodofanextrahour

withtheheatingdeactivated.

Thefreestandingairconvectionwasconsideredconstantandwouldnotvaryovertime. Thethermal

interactiononthesurfaceofthefloorhadtobegeneralizedduetothenatureoftheunpredictabilityof

surfacesconditions.

Air

currents,

interior

and

exterior

walls,

furniture,

human

activity,

natural

and

artificiallighting,andapplianceactivitycanallaffectthetransferofheatfromtheheatingsysteminto

thespace. Theexternalinteractionswerechosentorepresentalowermeansofinfluencebyonly

simulatingconvectionduetofreestandingair. Itwasnotedthatthisisaliberalassumptionandareal

scenariowouldbepredictedtoextractmoreheatfromthefloorsystem. Theteststhattheflooring

systemswouldbecompareduponinthisstudywouldbemostlyindependentofexternalinteractions

andwouldratherbeajudgmentoftheinternalthermalinteractionwithinthecrosssection.

Determininghowuniformlyheatwasdistributedacrossthesurfacewasgaugedbycomparingthe

temperatureoftwoparticularnodesonthesurfaceofthecrosssection. Thenodesonthesurface

directly

above

the

pipe

(the

top

point)

and

at

the

midpoint

between

pipes

(the

midpoint)

were

comparedfordifferenceintemperature. Thelowerthedifferencebetweenthenodaltemperatures

wouldrankadesignforhowwellitwoulddistributeheatalongthesurface,creatingbetteruniform

heatingandpreventinghotspots. Thiswillalsomeasureasystemseffectivenessatheatingaspaceand

usinglessenergybymeansofusingatandemthermostatsensorysystemtomeasurethetemperature

intheslabaswellastheairinthespace(Cho,1999).


9/29

Fournodequadrilateralswerethemeshingofchoiceforthesolidmaterialsduetotheirrelative

accuracyfortheamountofnodesrequired(Ho,1995). Meshingwasgivenageneraldensityof0.5

millimetersbutreducedsignificantlysurroundingthepipeopeningsinthecrosssectionforaccuracy.

Twonodeboundaryelementswereusedfortheconvectionelementsonboththesurfaceofthefloor

andthepipeopenings.

ThemethodsoffloorconstructionthatweremodeledinthisstudyaresummarizedinFigures3ae.

Figure3a:TraditionalSlabonGrade(Somesectionsmodeledwithoutinsulation)

Figure3b:SheathingandJoistswithAluminumTrough

Figure3c:SheathingandJoistswithMortarBedandTile


10/29

Figure3d:SheathingandJoistswithMortarBedandHardwoodFinishfloor

Figure3e:SheathingandJoistswithMortarBed,SSPCM,andFinishfloorsupportedbywoodenlathe

ThefollowingpropertiesinTable1weregiventothematerials:

Table1MaterialProperties

HeatConduction,k=(W/mK) HeatCapacity,c=J/((m^3)K) LatentHeatCapacity,L=J/m^3

Concrete 1.7 1,950,000

Wood 0.15 222,600

SSPCM 0.25 1,715,000 156,000,000

Insulation 0.04 325,000

Tile 2.5 2,250,000


11/29

FIGURE4:FiniteElementModelofSSPCMdesign

Results

Themodelresultsrevealedseveralpatternsthatwerecomparedusingthetemperaturedatafromthe

twosurfacenodes. Theserelationshipswere:

1. Thedifferenceintemperaturebetweenthetoppoint(TP)andthemidpoint(MP)2. Thetimethatthemeantemperature(averageofTPandMP)reachedtheoptimalheatingrange3. Therateofcoolingatfloorsurface(Startingatt=4500sec)

DifferenceinTemperature

ByevaluatingthetemperaturepredictionsattheTPandMP,arelationshipwasmadebetweenthe

verticaldistributionsofthepipenetworkintheslabtothelocalizedsurfacetemperatures. Asthe

networkwasplacedfurthertowardsthebottomoftheslab,thesurfacetemperaturedifferencesfrom

thespanofthetwopointswasreduced,diminishingthehotspoteffect. Thecloserthenetworkwas

placedtothesurfaceofthefloor,thegreaterthedifference.

Anotherfactorthatplayedaneffectonthedifferenceofsurfacetemperaturewasthehorizontal

spacingof

the

piping.

Although

the

pattern

of

temperature

difference

remained

the

same

for

those

modelsofthesimilarconstruction,thevalueofthetemperaturedifferencewasamplified. Howeverthe

amountoftimenecessarytoachieveapproximatetemperatureuniformitywasaboutthesame(Figure5

through7).

Alongwiththetemperaturedifferencebetweensystems,arelativepatternformedduetothetypeof

materialsthatwereusedwithineachmodel. Inthecaseofmaterialswithahigherthermalconduction


12/29

factor,suchastheconcreteandtile,thetemperaturedegradationaftertheheatingsupplywas

removedweremuchsteeper(Figure8). However,thesesystemsdiddeliveredheatveryquicklytothe

surfaceenvironment. Theothermodelsthatinvolvedwood,SSPCM,orinsulationhadmuchshallower

temperaturedegradationaftertheheatedwaterwasshutoff,causingtheheattotransferslowertothe

surfaceenvironment(Figure9through11).

Similarpatternsappearedonthejoistandwoodensheathingmodels. Itwasapparentwheretheheat

hadbeenturnedoffhoweverthetemperaturespikeswerelessexaggerated,indicatingamoreuniform

heatingdistributionacrossthecrosssection. Thisisexplainedbythewoodhavingasmallerthermal

conductioncoefficient. Thisslowstheheattransferalongtheinterfacebetweentheconcreteandwood

andcausesthetemperaturecontourstoflattenastheyreachthesurface.

ThefloorconstructionwiththemostnoteworthyperformancepredictionsisthedesignwiththeSSPCM

plates. Thisistheonlyscenariowherethetemperaturedifferencecontinuestoriseaftertheheating

hasbeenshutoff. ThiscanbeexplainedbytheSSPCMasitreleasesitslatentenergyastheinternal

systemcools. Thiseffectcreatesathermalbuffertoprotectthesurfacefromunregulatedandsudden

heatloss.

FIGURE5:Differencebetweentoppoint(TP)andmidpoint(MP)offloorcrosssection.SOG1st

numberrepresentshorizontalspacingofpipe

inmm,2nd

denotesverticalspacingofpipefromsurfaceinmm,_iindicatesifinsulationisplacedalongunderside.



inmm,2nd


0

1

2

3

4

5

6

Temperature(C)

Time(seconds)

100_37.5

100_50

100_75

100_37.5_i

100_50_i

100_75_i

0

1

2

3

4

5

6

7

8

9

10

Temperature(C)

Time(seconds)

150_37.5

150_50

150_75

150_37.5_i

150_50_i

150_75_i


13/29



inmm,2nd


FIGURE8:Differencebetweentoppoint(TP)andmidpoint(MP)offloorcrosssection.TypicalJoistConstructionJrepresentsjoist

construction,tiledenotessurfacematerial,andnumberindicateshorizontalspacingofpipeinmm.

FIGURE9:Differencebetweentoppoint(TP)andmidpoint(MP)offloorcrosssection.TypicalJoistConstructionJrepresentsjoist

constructionandnumberindicateshorizontalspacingofpipeinmm.

0

2

4

6

8

10

12

Tem

perature(C)

Time(seconds)

200_37.5

200_50

200_75

200_37.5_i

200_50_i

200_75_i

0

2

4

6

8

10

12

14

Temperature(C)

Time(seconds)

J_tile_100

J_tile_150

J_tile_200

0

1

2

3

4

5

6

7

8

9

10

Temperature(C)

Time(seconds)

J_100

J_150

J_200


14/29

FIGURE10:Differencebetweentoppoint(TP)andmidpoint(MP)offloorcrosssection.TypicalJoistConstructionwithaluminumtrough

betweenjoistsradrepresentsradiationasmeansofheatflux,numberindicatesamountofpipesaccountedforwithineachtrough.

FIGURE11:Differencebetweentoppoint(TP)andmidpoint(MP)offloorcrosssection.TypicalJoistConstructionwithPCMunderfinishfloor

Jrepresentsjoistconstruction,PCMdenotesPCMisusedunderfinishfloor,numberindicateshorizontalspacingofpipeinmm.

Asexpected,insulationplayslittleeffectindecreasingtheeffectsofhotspotsinatraditionalslabon

gradeassembly. Thisisbecausetheheattransfersupwardsfromthepipeuniformlywhetherornot

thereisinsulationalongthebottom. Insulationplaysagreateffectonincreasingtheoverall

temperaturebybufferingtheslabfromtheground. Iftheaimistoreducethehotspoteffectthena

closerandlowerspacingofpipewithintheslabisencouraged. Awoodencoveringperformswellto

insulateandleveltheheatingcontoursoutacrossthesurfaceastheoptimumtemperaturereachesthe

surface.The

SSPCM

design

further

extends

the

contour

leveling

action.

TargetHeatingTime

Eachsystemrequiredadifferentamountoftimeforthesurfacetoreachthetargetheating

temperature. Thosesystemswithmaterialsthatpossessalargerheatconductioncoefficientwereable

totransferheattothesurfacequickerthanthosesystemswithmaterialshavinglowerheatconduction

coefficients. Itwasfoundthatseveralsystemsdidnotrequireallthreecyclesofheatingbecausethe

0

1

2

3

4

5

6

7

8

Tem

perature(C)

Time(seconds)

1_pipe_rad

2_pipe_rad

3_pipe_rad

0

1

2

3

4

5

6

7

8

Temperature(C)

Time(seconds)

J_PCM_100

J_PCM_150

J_PCM_200


15/29

meantemperaturehadalreadyreachedthecomfortboundaries. Ideally,theheatingsystemwouldbe

shutoffatthatmomentandwouldnotactivateagainuntiltheslabtemperaturedippedbelowthe

optimumheatingtemperature.

Severalsystemsdidnotreachthetargetheatingtemperature. Somesystems,suchasthealuminum

troughdesign,

did

not

produce

the

amount

of

heat

necessary

to

heat

the

material

to

the

preferred

level.

Theothersystems,notablytheSSPCMdesign,internalizedmoreenergythanitreleased. Despitethe

factthateachsystemreceivedthesameamountofenergyastheothersystems,theSSPCMsstorage

capacityforenergywassignificantlylargerthantheothersystems. Ratherthantheappliedthermal

energydiffusingtothesurfaceforheatinguse,theenergywasstoredforlateruse. Althoughthis

propertyprovidesgreatadvantagefortheSSPCMdesign,itfallsshortititsabilitytoquicklytransfer

heattothesurfaceasreadilyastheotherdesigns. Ideally,aSSPCMshouldhavelargelatentheat

capacitywithhighthermalconductance,yettheSSPCMmodeled(paraffin)hasarelativelylowthermal

conductance.

ThefollowingTable2liststhesystemandthetimenecessaryittooktoreachthemeantargetheating

temperature(28C(81F))ifitwasreachedatallduringthecyclingheatingprocess. Thedesignswere

rankedbythespeedinwhichthemeansurfacetemperaturereachedoptimumheatinglevels. Those

thatdidnotmakethedesiredleveldonothaveatimevalue.

Table2:TimenecessarytoreachOptimumHeatingTemperatureforsystems(seefigures511fornamerepresentation)

Type #ofPipe

Horz.Spac.

(mm)

Vert.Spac.

(mm) Insulation? Name

Time

(sec)

MaxTemp

(degC) Rank

SOG 100 37.5 no 100_375 1000 39.922 2

SOG 100 37.5 yes 100_375_i 1000 39.8195 2

SOG

100

50

no

100_50

1840

37.045

6

SOG 100 50 yes 100_50_i 1840 36.861 6

SOG 100 75 no 100_75 3300 31.583 16

SOG 100 75 yes 100_75_i 3340 32.168 17

SOG 150 37.5 no 150_375 1940 32.6765 8

SOG 150 37.5 yes 150_375_i 1940 32.583 8

SOG 150 50 no 150_50 2970 29.9205 12

SOG 150 50 yes 150_50_i 2990 29.733 13

SOG 150 75 no 150_75 26.1665 20

SOG 150 75 yes 150_75_i 26.0565 21

SOG

200

37.5

no

200_375

3040

28.3535

14

SOG 200 37.5 yes 200_375_i 3060 28.28 15

SOG 200 50 no 200_50 23.1925 24

SOG 200 50 yes 200_50_i 23.079 25

SOG 200 75 no 200_75 22.6915 26

SOG 200 75 yes 200_75_i 22.613 27

Trough 1 no RAD_1_pipe 12.0805 30



16/29


Sheathing no J_100 2840 32.2065 11

Sheathing no J_150 2140 30.983 10

Sheathing no J_200 26.895 19

Tile no J_tile_100 610 44.248 1

Tile no J_tile_150 1360 37.353 4

Tile no J_tile_200 1800 32.261 5

PCM no J_PCM_100 3440 30.929 18

PCM no J_PCM_150 25.8905 22

PCM no J_PCM_200 21.673 28

Fromthemodelresults,patternsshowthatdesignswithacloserpipespacing(100mm)allreachedthe

benchmarkforthedesiredheatinglevelunderthesameloadingscenario. Largerspacingofpiping(150

mm,200mm)aswellassystemspossessingmaterialswithlowerthermalconductivitiesandhigh

storagecapacities

did

not

all

reach

the

desired

temperature

level.

Designsthatusemorematerialswithlowerthermalconductivitywouldrequiretheuseofapredictive

activationsystemforitsusetobeofgreaterbenefit. Ifasystemcouldbeactivatedearlierbeforeits

initialusebytheoccupant,thenthesystemwouldreachtheoptimumheatinglevelatthetimethatthe

userwillbeginuseofthespace.

HeatDissipationoverTime

Asthesystemsareremovedfromthesupplyheat,theenergystoredinthematerialscomesintoplay.

Themorethermalenergythatcanbestored,thelesscyclingofthesupplysourceisneeded.

Twoparticulartemperaturesatsettimesweregatheredfromeachmodeltogeneralizethedecayof

temperaturegradientovertimefromasystemafterthesystemwasdeactivated. Everysystem

displayedanearlinearchangeoftemperaturestartingat4500secandcontinuedtotheendofthe

analysis. Thesetrendlineswerecalculatedandsuperimposedoverthegraphofthemeantemperature

readingsfromtime4500to7200sec. Thetimedurationof2700secwassetasaunitlengthoftimeto

deciphertheslopeofeachofthetrendlines. Byfindingaunitslopeforeachsystem,everysystem

couldbecomparedtooneanotherduringthesameperiodofitsheatingcycletoshowwhichsystems

couldretainandtransferheatbetter. ThesecomparisonscanbeseeninFigures12and13.

Somesystemsdidnotholdheatwell,dischargedquickly,andrequiredmorefrequentinterventionfrom

thesupply

heat.

These

systems

were

found

to

be

the

systems

with

materials

with

large

conduction

coefficients. Aswellasthematerialproperties,theverticalspacingbetweentheheatsupplyandthe

floorsurfaceinfluencestherateofheatloss. Thecloseraheatsupplyistothefreesurface,theless

materialbetweenthesurfacestocontainheat. Thisscenariowasnoticedinseveralslabongrade

designsaswellasthetiledesign. Thefurthertheheatingsupplyisfromthesurfaceallowsforagreater

massofmaterialtostoreenergyheatistransferredoffofthetopsurface. Thisscenariowaswitnessed


17/29


18/29

Anysystemthatinvolvedsomesortofinsulatingmaterialnearthetopsurfaceofthecrosssectiondid

betterbycomparisonthanothersystemswithout. Thejoistsystemswithalayerofsubflooringor

sheathingasthetopsurfaceactedwellatretainingheatandslowingitsretreatoutofthesystemwhile

betterretainingapreferabletemperature. ThedesignsthatincludedSSPCM,alongwithalayerof

woodenflooringonthesurfacedidwelltofacilitatetheentrapmentofheatandtocontinuetosupply

heateven

when

the

supply

has

been

deactivated.

Discussion

Byanalyzingthedifferentconstructiontypesundertheseparatetestspresentedabove,several

conclusionscanbeinferredabouteachtype.

RadiationTrough

Theradiationdesignworkswellforconstructionpurposes. Thesystemcanbeinstalledwithoutadding

excessiveextraweighttoanexistingsystem,allowingittobeoneofthemoreversatiledesigns. Forits

easeof

installation

however

the

performance

does

not

measure

up.

After

given

a

full

heating

supply

cycle,noneofthevariantsofthedesignmadeittotheoptimalheatingtemperature,nordiditretain

whatheatitgained. Thevarianceoftemperatureperformancealongitssurfacewasaveragecompared

totheothersystemsbutwithoutbeingabletoreachoptimumheatingtemperature,thevarianceof

temperaturedidlittletoaddheattothespaceinareasonabletimeframe. Theradiationmethodisa

wisemeansofcapturingotherwiselostheat,butasamainsourceofheatforaflooringsystem,itfalls

shortoftheothersystemsinitsabilitytobringtheheatingsystemuptotemperaturewithinthe

timeframeoftheothersystems.

TileFinish

Thejoist

and

tile

finish

designs

best

attribute

is

its

ability

to

respond

quickly

to

user

input.

The

amount

oftimenecessaryforthesystemtoheatupwasunmatchedaslittleasonly10minutes. Thiscomes

withthedisadvantagesofrequiringheatmoreoftenfromtheheatsupplyanditsinabilitytodistribute

heatevenlyoverthesurfaceofthefloorcausingseverehotspots. Ifasystemlikethisweretobe

installed,closerpipespacingwouldberequiredduetothisdesignsinabilitytodistributeheatuniformly

overitssurface. Thiswouldrequiremorepipingmaterialaswellasenergytosupplythepipingwith

heatandtopumpthefluid.

SlabonGrade

The

slab

on

grade

was

the

most

widely

researched

design

in

this

study.

With

many

different

layouts,

it

becameclearwhichlayoutsperformedmostadvantageously. Slabongradedesignsaresimilartothe

tiledesigninregardstoitsabilitytotransferheattothesurface. However,theopportunitygainedwith

theslabongradedesignistheabilitytoplacetheheatingsupplypipesatvariouslocations,therefore

placedtooptimizethemagnitudeofpositiveeffects. Thehotspotandthermalvariabilityissuecanbe

resolvedbyplacingthepipingnetworklowerintheslabtogiveagreatervolumefortheheatto

dispersethroughbeforeitreachesthesurfaceofthefloor. Reactiontimetoreachtheoptimal


19/29

temperaturecanbeaccountedforbylayingpipeclosertogetherhorizontallyandclosertothesurface

vertically. Itcouldevenbedevisedthatthenetworkofpipecouldvaryandstaggerlocation,oreven

twonetworkscouldbeinterlaidtocombinethepositiveeffects. Thiswouldhavetobeexploredin

futureresearch. Insulationisawiseintroductiontoaslabongradedesignbecausetheheatdistribution

ismoreuniformthelowerthepipingnetworkisplacedintheslab. Themaineffectsoftheinsulation

areseen

from

the

ability

to

keep

the

slab

protected

from

the

cooler

effects

of

the

ground,

and

raising

theoveralltemperaturewhilehavinglittleeffectonthesurfacetemperaturevariance.

Disadvantagesintroducedincludethestructuralrequirementsthedesignhasonanexistingprojectand

itsinabilitytomaintainanoptimumsurfaceheatingtemperature. Theamountofthermalmassing

requiredtomaintainthetargetlevelofheatingquicklyaddsuptomorethanasimpleretrofitsolution.

Thissystemcontributesconsiderabledeadweighttoastructuraldesignandmanyhomescannot

accommodatethissystemotherthanonthegroundfloor. Ifthesystemistobeappliedtosuspended

levels,athinnerslabwillberequiredtomeetstructuralcapabilitiesofanexistingfloortherebylosing

theadvantageofpositioningpipinginoptimalheightswithintheslabheight.

Likeothersystemswithlowerconductioncoefficients,theheattransfersoutofthesystemrather

quicklyandthereforerequiringagreaterheatsupplytochargethesystem. Thischaracteristicofslab

ongradedesignisshownbythedecreaseoftemperatureaftertheheatingsystemhasbeenshutoff

(Figure12).

Woodenfinish

Thesesystemsperformedwellinseveralregards. Althoughthesurfacetemperaturedidnotreachthe

optimalheatingtemperatureasquicklyasothersystems,theconstructionofwoodenfinishand

subfloorsheathingsandwichingtheencapsulatedpipelayerprovidedthermalbufferingandhelped

regulateheat

loss.

The

thermal

buffering

created

a

more

uniform

distribution

of

heat

on

the

surface

of

thefloor. Hotspotsandsurfacedifferentiationwerereducedduringtheheatdispersionthroughthe

woodprovidingamoreuniformsurfacetemperature. Systemswiththeseconstructionattributesalso

showedalowerrateofheatlossafterthesystemhadbeendeactivatedofferingopportunitiesfora

lowerrechargedemand.

Thissystemalsoworkswellforretrofitsolutionsaswell. Theloadingontheexistingstructuralsystemis

lessthanthatofatilefinishoptionanditretainsheatlonger,reducingenergycostsbyrequiringless

heatingrecharge. Althoughthesystemwillrespondslowertoheatingdemands,anintegrated

activationsystemwillbeabletochargethesystemtobringthesystemuptotemperaturebeforethe

use

of

the

space.

PhaseChangeMaterialbufferlayer

TheSSPCMsystemprovidedsomeofthemostinterestingresults. Itwasoneoftheslowersystemsto

chargeduetothenatureofprovidingadequateenergytotheSSPCMtofullymeltthematerial. Someof

thedesignsofthesystemdidnotreachthetargetheatingtemperatureshowingthatasignificantinput


20/29

ofheatisnecessarytobringthesystemuptotherequiredlevel. Butassoonasthelevelofheathad

beenobtained,thebenefitsthesystemprovidesbecomeapparent.

Auniformandstablesurfacetemperatureisprovidedbythesystemwithsmallersurfacedifferences

thantheothersystems. Thehotspoteffectisreducedsignificantlyincontrastwiththeothersystems,in

somecases

by

a

factor

of

2

during

the

greatest

difference

in

temperature

with

similar

pipe

space

(Figure

8&Figure11). Aftertheheatingsystemhasbeendeactivated,theheatgainedbythelatentheat

capacityoftheSSPCMbecomesanewsourceofheatandreleasestheenergygainedtothesurfaceof

thefloorsystemasitcools. Thisaddstothebuffereffectbecauseeventhoughthereisafluxofenergy

fromphasechange,thesurroundingtemperatureremainsstablewithsmalldeviation(Figure13).

FutureRecommendations

Woodenmaterialscancontributeapowerfulmeansofbufferingheatwithinathermalmassduringthe

heatingprocess. Ifawoodencompositematerialcouldbemanufacturedtosandwichthermalmassing,

orevenbethethermalmassingitselfprovideditcanallowfortheheatedpipetorunthroughit,the

gentletransfer

of

heat

through

the

floor

system

would

be

employed.

Adisadvantageofwoodenmaterialsistherelativelylowthermalstoragecapacity.Acombinationof

materialscouldovercomethisdeficiency,wherethethermalmassingmaterial(concreteormortar)

wouldstoretheheatandthewoodwouldallowforitsslowreleaseoutofthesystem. Tilewouldnotbe

awisefinishmaterialunlessalayerofwoodorinsulatingmaterialwastobuffertheheatreleasefrom

thethermalmassingconcreteormortarbecausethetilewouldallowtheheattoescapetooquickly.

ShapeStabilizedPhaseChangeMaterialswouldbeanexcellentadditiontoaradiantfloorheating

systemforgreatlyincreasingtheamountofthermalstorageinasystemwithoutaddingexcessive

weight.

Placing

a

layer

of

SSPCM

before

the

surface

layer

would

act

as

a

thermal

buffer

to

allow

heat

transferatrelativelyconstanttemperaturegradient. TheinternaltemperaturesbelowtheSSPCMmay

havegreaterrangethanthoseabove,buttheSSPCMwillactasthedistributionandsupplyheatforthe

surfaceratherthanthepipingnetworkdirectly. Thisgreatlyincreasesuniformityandcomfort. A

monitoringsystemwillbenecessarywithintheslabtoindicatewhentheheatsupplyneedstoreactive.

Thisisduetothefactthatifthemonitoringsystemwerecompletelyexternaltothesystem,bythetime

themonitoristrippedtoinitiateheating,theamountoftimenecessarytoreheatthesystemwillcause

anuncomfortableenvironmentandunnecessaryenergyinefficiency. Havingmultiplestagesofsensors

atdifferentlayersoftheslabwillprovevaluabletotheperformanceofthesystemsenergyefficiency.

SelectingthebestSSPCMforthedesignwillinvolvefindingaSSPCMthatpossessesamelttemperature

closeorwithintheboundsoftheoptimumheatingrange. Havingknowntheboundariesunderideal

conditions(24C 32C),itwillbeimportanttofindthetemperatureforactualconditions.

Materialsthattransferandstoreheatwellwillproveimportanttodevelopingadesigntosuitthermal

needsbutstructuralpropertiesneedtobeconsideredaswell. Asthesystemheatsandexpands,itwill

experiencenonuniformstressalongthecrosssection. Fromtheaestheticsperspective,iftheradiant

floorheatingsystemisdirectlyexposedtothesurfaceofthespaceanditundergoescrackingdueto


21/29

heatexpansion,itwillhavefailed. Findingaheatrangeformaterialaestheticswillbeimportantwhen

selectingmaterialsandhowtoincorporatethemintothedesign.

SummaryandConclusions

Radiantfloorheatinghasshowntobeaneffectivemeansofdeliveringheattoanenvironmentthat

promoteshealthbenefitsandenergyefficiency. Fromthesimulations,itwasfoundthat

1. Designsthatfeaturedsurfacematerialswithlargerheatconductioncoefficientswarmedthefloorsurfacemuchquickerthanthosesystemswithsmallercoefficients. Thiscameatapriceby

havinglesssurfacetemperatureuniformityandquickerheatlossaftersystemdeactivation.

2. Designsfeaturingsurfacematerialswithlowerheatconductivecoefficientsrequiredlongerlengthsoftimewiththeheatingsystemontoreachthetargetheatingrange. However,when

theheatingsystemwasdeactivated,thesurfacetemperatureremainedstableforlonger. Heat

varianceswereminimizedwhencontrastedtotheotherheatingdesigns.

3. ThedesignwithSSPCMincludedwasfoundtosupportthebestreductionofsurfacetemperature

variance

and

after

the

heating

system

was

deactivated,

the

design

possessed

the

slowestdegradationofsurfacetemperature. Althoughitdidrequiremoretimetoreachthe

targettemperature,thesurfacetemperatureswerethemostuniformofanyofthesystems

modeled.

Thevalueofusingmodelingasatooltodiscoverinnovativedesignswasdemonstrated. Byrunning

simulationsofcurrentandfutureradiantfloorheatingdesigns,crosssectionscanbeoptimizedfor

heatinguniformityandenergystorage. Currentwoodenmaterialshaveimportantattributesthatcan

benefitradiantfloorheatingsystems. Combiningwoodwithamaterialwithahigherheatcapacity,such

asconcreteormortar,cansignificantlyincreasethefloorsabilitytowarmaspacebycontrollingthe

releaseof

the

generated

heat.

ByintroducingaSSPCMintothefloordesign,theeffectsofintroducingwoodenmaterialsisaddedupon

andtheoverallheatingstorageamplifies. Thisstudydemonstratedpossibilitiesforintroducingthermal

storagecapacitiesintoabuiltenvironmentthatdonotnecessaryrequireheavythermalmassingwhich

increasestheoverallstructuraldemand. Opportunitiesareavailablethatprovidethenecessarycomfort

appeartobetechnicallyfeasible.


22/29

References

S.Sattari,(2006),Aparametricstudyonradiantfloorheatingsystemperformance,RenewableEnergy,

V.31pg.16171626

N.A.Neeper,(2000),ThermalDynamicsofWallboardwithLatentHeatStorage,SolarEnergy,V.68

pg.393403

S.Y.Ho,(1995),Simulationofthedynamicbehaviorofahydronicfloorheatingsystem,HeatRecovery

Systems&CHP,V.15,pg505519

A.Khudhair,(2004),Areviewonenergyconservationinbuildingapplicationswiththermalstorageby

latentheatusingphasechangematerials,EnergyConservationandManagement,V.45,pg268275

M.Farid,(2004),Areviewonphasechangeenergystorage:materialsandapplications,Energy

ConservationandManagement,V.45,pg15971615

Y.P.Zhang,

(2006),

Preparation,

thermal

performance

and

application

of

shape

stabilized

phase

change

materialsinenergyefficientbuildings,EnergyandBuildings,V.38,pg12621269

G.Song,(2005),ButtockresponsestocontactwithfinishingmaterialsovertheONDOLfloorheating

systeminKorea,EnergyandBuildings,V.37,6575

S.H.Cho,(1999),AnexperimentalStudyofMultipleParameterSwitchingControlforRadiantFloor

HeatingSystems,Energy,V.24,433444

M.ZaheerUddin,(1997),OptimalOperationofanEmbeddedPipingFloorHeatingSystemwithControl

InputConstraints,EnergyConservationManagement,V.38,713725

D.Song,(2008),PerformanceEvaluationofaRadiantFloorCoolingSystemIntegratedwithDehumidified

Ventilation,AppliedThermalEngineering,V.28,12991311

S.H.Cho,(2003),PredictiveControlofIntermittentlyOperatedRadiantFloorHeatingSystems,Energy

ConservationandManagement,V.44,13331342

R.D.Woodson,(1999),CompleteConstruction,RadiantFloorHeating,McGrawHillPublishing,ISBN0

071347860

D.B.Crawley,(2005),ContrastingtheCapabilitiesofBuildingEnergyPerformanceSimulationPrograms,

Buildingand

Environment,

V.43,

661

673


23/29

AppendixA(Calculations)

EquationA1:LatentHeatAccommodationinADINA

HeatCapacity=

=

LatentHeat=

=

=

HeatCapacity+LatentHeatper1degree=

TableA1:RadiationCalculationswithuseofExcel

This

is

the

general

equation

for

finding

the

amount

of

energy

is

emitted

by

a

heated

body.

For

the

purposesofmodelingtheheatemittedintheADINAmodels,thesetableswerecreated. Itassumes

that:

1. Halfoftheenergyemittedinitiallygoestothewoodandhalfgoestothetrough2. Allenergyiseitherabsorbedbythematerialorreflectedbacktowardtheothersurface3. Theairdoesnotgainanyheatwithinthetrough

Itisimportanttonotethatthetableneededtoberuntwice:oncefortheheatedpiperadiationatfull

temperature(323K)andonceheatedpiperadiationatnormaltemperature(298K). Thesetwo

calculationsneededtobedonesothemodelwouldaccuratelyportraytheeffectswhenthesystemwas

shutoff.

It

is

important

to

note

that

the

material

surrounding

the

trough

would

increase

in

temperature

overtimebecauseitwasabsorbingmoreenergy,causingtheheatfluxtodiminish. Thiswasneglected

forthesimplicityofthemodelandtoshowthateventhoughaliberalassumptionthatextraheatwas

gained,themodelstilldidnotheathashopedtoreachtheoptimumtemperaturerange. Thesefluxes

wereappliedusingthesametimefunctionthatrepresentedtheconvectioncoefficientsforthewater.

3pipepertroughexample(323K):

Emissivity

(Cu)

Stefan

Boltzmann T(body)

T

(surroundings) Area #ofpipe Heatflux 1/2ofheat

0.87 5.67E08 323 283 0.0628 3 41.54723354 20.77361677

Energy

Available Absorbtivity

Energy

Absorbed

Energy

Reflected

Energy


Energy

Absorbed

Energy

Reflected

20.77361677 0.65 13.5028509 7.27076587 28.04438264 0.09 2.523994438 25.5203882

25.5203882 0.65 16.58825233 8.932135871 8.932135871 0.09 0.803892228 8.128243643

8.128243643 0.65 5.283358368 2.844885275 2.844885275 0.09 0.256039675 2.5888456

2.5888456 0.65 1.68274964 0.90609596 0.90609596 0.09 0.081548636 0.824547324

0.824547324 0.65 0.53595576 0.288591563 0.288591563 0.09 0.025973241 0.262618323

0.262618323 0.65 0.17070191 0.091916413 0.091916413 0.09 0.008272477 0.083643936


24/29

0.083643936 0.65 0.054368558 0.029275378 0.029275378 0.09 0.002634784 0.026640594

0.026640594 0.65 0.017316386 0.009324208 0.009324208 0.09 0.000839179 0.008485029

0.008485029 0.65 0.005515269 0.00296976 0.00296976 0.09 0.000267278 0.002702482

0.002702482 0.65 0.001756613 0.000945869 0.000945869 0.09 8.51282E05 0.00086074

0.00086074 0.65 0.000559481 0.000301259 0.000301259 0.09 2.71133E05 0.000274146

0.000274146 0.65 0.000178195 9.5951E05 9.5951E05 0.09 8.63559E06 8.73154E05

8.73154E05 0.65 5.6755E05 3.05604E05 3.05604E05 0.09 2.75044E06 2.781E05

2.781E05 0.65 1.80765E05 9.73349E06 9.73349E06 0.09 8.76014E07 8.85748E06

8.85748E06 0.65 5.75736E06 3.10012E06 3.10012E06 0.09 2.7901E07 2.82111E06

2.82111E06 0.65 1.83372E06 9.87387E07 9.87387E07 0.09 8.88648E08 8.98522E07

8.98522E07 0.65 5.84039E07 3.14483E07 3.14483E07 0.09 2.83035E08 2.86179E07

2.86179E07 0.65 1.86017E07 1.00163E07 1.00163E07 0.09 9.01465E09 9.11481E08

9.11481E08 0.65 5.92463E08 3.19018E08 3.19018E08 0.09 2.87117E09 2.90307E08

TotalEnergy/secAbsorbedbywood: TotalEnergy/secAbsorbedbytrough:

37.84364667 3.703586848

AreaoverHeatapplied 0.36 m^2 AreaoverHeatapplied 0.36 m^2

HeatFluxonArea 105.1212407 HeatFluxonArea 10.28774124

3pipepertroughexample(298K):

Emissivity

(Cu)

Stefan

Boltzmann T(body)

T

(surroundings) Area #ofpipe Heatflux 1/2ofheat

0.87 5.67E08 298 283 0.0628 3 13.67997266 6.839986329

Energy


Energy

Absorbed

Energy

Reflected

Energy


Energy

Absorbed

Energy

Reflected

6.839986329 0.65 4.445991114 2.393995215 9.233981544 0.09 0.831058339 8.402923205

8.402923205 0.65 5.461900083 2.941023122 2.941023122 0.09 0.264692081 2.676331041

2.676331041 0.65 1.739615177 0.936715864 0.936715864 0.09 0.084304428 0.852411437

0.852411437

0.65

0.554067434

0.298344003

0.298344003

0.09

0.02685096

0.271493043

0.271493043 0.65 0.176470478 0.095022565 0.095022565 0.09 0.008552031 0.086470534

0.086470534 0.65 0.056205847 0.030264687 0.030264687 0.09 0.002723822 0.027540865

0.027540865 0.65 0.017901562 0.009639303 0.009639303 0.09 0.000867537 0.008771766

0.008771766 0.65 0.005701648 0.003070118 0.003070118 0.09 0.000276311 0.002793807

0.002793807 0.65 0.001815975 0.000977833 0.000977833 0.09 8.80049E05 0.000889828

0.000889828 0.65 0.000578388 0.00031144 0.00031144 0.09 2.80296E05 0.00028341

0.00028341 0.65 0.000184217 9.91935E05 9.91935E05 0.09 8.92742E06 9.02661E05

9.02661E05 0.65 5.8673E05 3.15931E05 3.15931E05 0.09 2.84338E06 2.87498E05

2.87498E05 0.65 1.86873E05 1.00624E05 1.00624E05 0.09 9.05617E07 9.1568E06

9.1568E06 0.65 5.95192E06 3.20488E06 3.20488E06 0.09 2.88439E07 2.91644E06

2.91644E06 0.65 1.89569E06 1.02075E06 1.02075E06 0.09 9.18679E08 9.28886E07

9.28886E07 0.65 6.03776E07 3.2511E07 3.2511E07 0.09 2.92599E08 2.9585E07

2.9585E07 0.65 1.92303E07 1.03548E07 1.03548E07 0.09 9.31928E09 9.42283E08

9.42283E08

0.65

6.12484E

08

3.29799E

08

3.29799E

08

0.09

2.96819E

09

3.00117E

08

3.00117E08 0.65 1.95076E08 1.05041E08 1.05041E08 0.09 9.45369E10 9.55873E09

TotalEnergy/secAbsorbedbywood: TotalEnergy/secAbsorbedbytrough:

12.46051801 1.219454642

AreaoverHeatapplied 0.36 m^2 AreaoverHeatapplied 0.36 m^2

HeatFluxonArea 34.61255002 HeatFluxonArea 3.387374006


25/29

AppendixB ReferenceSummary

S.Sattari,(2006),Aparametricstudyonradiantfloorheatingsystemperformance,RenewableEnergy,

V.31pg.16171626

Anotherwayoftestingradiantfloorheatingsabilitytotransferheatisbydevelopingafiniteelement

model. Modelingatwodimensionalradiantfloorheatingsystemistheaimofthestudy. Thisreference

willbeofgreathelpbecauseitoutlinesveryspecificallythematerials,initialaswellasboundary

conditions,andthenecessarymathematicalequationsthatareinvolvedinthemodelwhichwillbe

comparedwithinthisstudy. However,thedifferencewiththismodelisthatthereisnophasechange

materialsintroducedandthesystemisinitiatedfromstart(t=0)insteadofadiurnalanalysis,which

fluctuatesasthedayprogresses.

Thisstudydoesprovehelpfulbecauseitconductedmanytrialsandvariedcertainconditions,suchas

thepipediameter,material,thicknessofthefloormaterialabovethepipingnetwork,thefloortype,and

thefrequencyofpipesalongtheflooringsystem. Thestudyreportswhichattributescreatedthe

greatestchange

in

heat

transfer

and

will

be

helpful

to

eliminate

potentially

limiting

designs

and

guide

developingefficientmodelswithacombinationofadvantageousattributes.

N.A.Neeper,(2000),ThermalDynamicsofWallboardwithLatentHeatStorage,SolarEnergy,V.68

pg.393403

Understandingthenecessaryphysicalandengineeringconstructsofheattransferisimportanttothe

studyofradiantfloorheating. Thisstudyexploresthemathematicalmodelingofwallboardwithinfused

phasechangematerials. Specifically,thestudyexplainsthestepsindevelopingtimelageffectsofheat

gainand

exchange

with

other

materials

and

the

room.

By

developing

a

mathematical

modeling

system,

severaltrialswereconductedproducingresultsfordifferenteffects. Bycomparingthemodelswiththe

phasechangematerialinfusedwallboardtothemodelswithout,adirectcorrelationcanbemade

betweenthemodels,outliningthebenefitsofthermalstabilitybyinstallingthewallboard.

S.Y.Ho,(1995),Simulationofthedynamicbehaviorofahydronicfloorheatingsystem,HeatRecovery

Systems&CHP,V.15,pg505519

Developingafiniteelementmodelwillbeacarefulprocedureandthisstudygoestodescribethe

changesthat

were

made

in

the

defining

of

elemental

meshing

in

the

model

instead

of

varying

the

radiantfloorheatingsystemsnetworkforpotentialbenefits. Themodelwaschosenasatypical

residentialconstructionscheme,boundaryconditionsaswellasinitialconditions,andmaterialselection

arediscussed. Aswellashowthetemperaturevariesthroughthesystem.


26/29

A.Khudhair,(2004),Areviewonenergyconservationinbuildingapplicationswiththermalstorageby

latentheatusingphasechangematerials,EnergyConservationandManagement,V.45,pg268275

Applyingphasechangematerialsintotheradiantfloorheatingsystemwillrequireacompleteinquiry

intowhatconstructionmaterialshavebeenusedalreadyandtheireffectsoftransferringandstoring

latentheat.

Although

this

study

is

more

of

a

general

review

of

the

benefits

of

radiant

floor

heating

and

phasechangematerialapplications,ithasastronglistingofreferencesthathaveallcompleted

thoroughinvestigationsofaparticularapplicationoflatentheatstoragesystems.

M.Farid,(2004),Areviewonphasechangeenergystorage:materialsandapplications,Energy

ConservationandManagement,V.45,pg15971615

Thisstudyhasacompletelistingofphasechangematerialsthatcanbeusedforcommonpracticeas

wellastablesoftechnicaldataforthesematerials. Withinthedatathedifferentcompoundsofphase

changematerials

are

separated

by

inorganic,

organic,

and

commercial

paraffin

wax.

Their

properties

canbeevaluatedsidebysideandcompoundscanbechosenfortheirlatentheatpotentialandmelting

pointsdependingontheapplicationofthematerials. Iffurtherresearchisconductedordiscovered,

perhapsanenvironmentalfocusonthedifferentcompoundscouldbediscoveredanddocumented.

Aswellasprovidingageneralintroductionofbuildingapplicationsforphasechangematerials,the

processofimpregnatingthematerialintocommonbuildingmaterialsisalsodiscussed.

Y.P.Zhang,(2006),Preparation,thermalperformanceandapplicationofshapestabilizedphasechange

materialsin

energy

efficient

buildings,

Energy

and

Buildings,

V.38,

pg

1262

1269

Aphysicalexperimentalroomwasconstructedforthisstudyusingstrategicallyplacedphasechange

materialencapsulatedinplatesbeneaththeflooringsystemtoprovideacomfortableheatingscenario.

Theyareclearintheirconstructiontype,method,andresultsfordifferentexternaltemperaturesaswell

aswiththeabsenceandpresenceofthephasechangematerialplates. Severalotherdiscoverieswere

madeaboutthemostefficientwaytotransferheattotheplatesandhowtoincreasetheefficiencyof

deliveringthatheattotheneededlivingspaceoveradiurnaltimeperiod. Suchcharacteristicsinclude

platethickness,airgappresence,flooringmaterial,andthepreferredconductivityofthephasechange

materials.

G.Song,(2005),ButtockresponsestocontactwithfinishingmaterialsovertheONDOLfloorheating

systeminKorea,EnergyandBuildings,V.37,6575

Understandingtheeffectsonthehumanbodybyradiantfloorheatingwillbeimportantbecausethe

systemthatisderivedmustbeabletosatisfytheclientofthesysteminallaspectsofwhatitcan


27/29

provide. Thestudybuiltanexperimentalroomwitharadiantfloorheatingsystemandbroughtpeople

tobesubjectstotestthefloorsabilitytowarmthem. Subjectvariedinamultitudeofcharacteristicsto

developabroadrangeofresultsandconclusions. Bychangingtheflooringcoverandinput

temperature,thesubjectswereaskedtorespondtotheircomfortlevel. Fromthesubjectremarkson

thecombinationofcharacteristics,thebettercombinationswererevealed.

S.H.Cho,(1999),AnexperimentalStudyofMultipleParameterSwitchingControlforRadiantFloor

HeatingSystems,Energy,V.24,433444

Comparedtoconventionalheatingsystems,radiantfloorheatingmethodshaveshowntoreliably

deliverabetterdistributionofheatingtoaspace. Studieshavegonefurthertomakeradiantfloor

heatingsystemsevenmoreefficient. Thisstudyworksonthesensoryequipmentthatinformsthe

systemtoactivateandbyhowmuch. Mostsystemsrelyonthetemperaturegagewithintheroom,

whilethisstudyaddsasensorintheslabitselftochecktheslabtemperature. Thehypothesiswasto

discoverif

by

increasing

the

systems

sensory

with

a

switching

algorithm

would

decrease

the

amount

of

energyusedtoheataspace. Thealgorithmwouldcheckforanacceptablerangeforairtemperatureas

wellasslabtemperature. Ifanyparametersfelloutsideofthedesignatedlimits,thesystemwouldreact

tocorrectthesituation.

Itwasfoundthatthetwoparameterswitchingcontrolallowedforabettermanagementofenergyand

reducedtheamountoftimeandenergythesystemneededtoregainthedesignatedparameters.

M.ZaheerUddin,(1997),OptimalOperationofanEmbeddedPipingFloorHeatingSystemwithControl

InputConstraints,

Energy

Conservation

Management,

V.38,

713

725

UsingatraditionalONDOLradiantflooringsystemandaconventionalboilersupply,everyaspectofthe

systemwasmathematicallymodeled. Givencertainexpectedboundaryconditionssuchasoutdoorand

indoorairtemperature,theoptimalcharacteristicsofadesignlayoutcouldbedetermined. Such

characteristicsincludelengthofpipe,loopsofpiping,distancefromboilertoroom,watertemperature,

andtimeheatingwouldremainactive. Althoughthemathematicalmodelingallowedforseveral

assumptions,themodelwasextremelythoroughandcouldbeusedtomodelanyothersystemthat

wouldbedeveloped.

D.Song,(2008),PerformanceEvaluationofaRadiantFloorCoolingSystemIntegratedwithDehumidified

Ventilation,AppliedThermalEngineering,V.28,12991311

Traditionally,radiantfloorsystemshavebeenusedtoheatoccupiedspacesbutthisstudyinvestigated

theabilityofthesamesystemtocoolaspace. Foreseeingtheissueofcondensationduetohumidity

uponthecooledsurfaces,adehumidifiersystemisproposedtoreducethecondensationandthe


28/29

damageitcouldcausetosurfacecoverings. Theexperimentsconductedcomparedthecoolingsystem

withandwithoutthehumidificationsystem. Itwasfoundthatthedehumidificationsystemhelpedtwo

foldbyeliminatingcondensationbyloweringthedewpointandbycirculatingtheairintheroom

infiltratingareaswithcoolair.

S.H.Cho,(2003),PredictiveControlofIntermittentlyOperatedRadiantFloorHeatingSystems,Energy

ConservationandManagement,V.44,13331342

Indeterminingbetterwaystomakeradiantfloorheatingmoreefficient,thisstudyinvestigatesthe

timingofheatingcyclesandtofindmoreeffectivewaysofschedulingheatingcycles. Traditionally,

whenoutdoortemperaturesreachcertainlevel,theheatingsystemisencouragedtoturnonfor

specifiedlengthsoftimeduringstrategictimesofthedayspecifiedbypeoplesactivity. Althoughthe

conventionalheatingcyclesareeffective,theycanbemoreefficientgiventheconditionsoftheday.

Thestudyproposesamathematicalmodelfortimingtheheatingcycleslengthandpositionintheday

determinedby

the

high

and

low

temperatures

and

their

occurrences

during

the

day.

By

downloading

theinformationfrommetrologicaldatafromyearspastandcurrentdata,theinvestigationputthe

modeltothetestandwasabletoreduceenergyconsumptionupto20%fromtheconventionalheating

cycles.

R.D.Woodson,(1999),CompleteConstruction,RadiantFloorHeating,McGrawHillPublishing,ISBN0

071347860

Fortheeyesandmindsofthecontractor,thisbookfeaturesindepthconstructionmethodsanddetails

thathelp

the

novice

understand

the

working

of

radiant

floor

heating

and

how

to

construct

it

properly.

Manyofthedesignsoffloorswerefoundandconfirmedfromthisliterature. Potentialhealthbenefits

andcostsavingswerealsodiscussedandcomparedwithotherconventionalheatingsystemsused

already. Potentialpitfallsanderrorstobeavoidedwereshownandwereabletoguidethe

developmentofthemodels.

D.B.Crawley,(2005),ContrastingtheCapabilitiesofBuildingEnergyPerformanceSimulationPrograms,

BuildingandEnvironment,V.43,661673

Thisstudy

is

a

thorough

investigation

into

energy

performance

software.

By

setting

specific

objectives

thatadesignerofabuildingorspacewouldwanttomodel,eachsoftwarepackageisanalyzedforits

capabilityandaccuracyonaparticulartopic. Eachsystemisgivenanoverviewandgeneralperformance

recommendationsareincluded.


29/29

AppendixCModelTitleGuide

Name

Horizontal

Spacingof

Pipe

Vertical

Spacingof

Pipe

Pipesper

trough Insulation?

Surface

Material SSPCM?

100_375 100mm 37.5mm no SOG no

100_375_i 100mm 37.5mm yes SOG no

100_50 100mm 50mm no SOG no

100_50_i 100mm 50mm yes SOG no



150_375 150mm 37.5mm no SOG no

150_375_i 150mm 37.5mm yes SOG no





200_375 200mm 37.5mm no SOG no

200_375_i 200mm 37.5mm yes SOG no





1_pipe_rad 1 no plywood no



J_100

100mm

centroid

no

plywood

no

J_150 150mm centroid no plywood no

J_200 200mm centroid no plywood no

J_tile_100 100mm centroid no tile no



J_SSPCM_100 100mm centroid no plywood yes



Documents

Radiant Floor With Phase Change