Вы находитесь на странице: 1из 55

P.O.

Box 571 | Colville, WA 99114 | USA


T: +1.509.684.2595 | F: +1.509.684.8331
www.colmaccoil.com


DXAMMONIAPIPINGHANDBOOK
BruceI.Nelson,P.E.





TABLEOFCONTENTS










Background







SystemConfiguration 





SystemStability






EvaporatorSelectionandOperation





DT1vsDTMRatings






SensibleHeatRatio,Roomrh%,andEvaporatorRatings 

OptimizingSystemTD 





EffectofTDonExpansionValveOperation



TypesofFrostandSelectionofFinSpacing


CondenserSelectionandOperation




Subcooling







PipingGeneral






LiquidLines







SuctionLines 






HotGasLines 






EffectsofWaterinAmmoniaandItsRemoval 



Separation







DistillationandDisposalofAmmoniumHydroxide


LiquidTransfer 





EffectsofOilonEvaporatorPerformanceandOilSeparation

EstimatingDXEvaporatorRefrigerantChargeInventory 

ColmacSmartHotGasDefrost 




CalculatingtheCostofDefrost 




SmartHotGasSequenceofOperation 



DefrostWaterVolumeandDrainLineSizing



References







APPENDIXADXAmmoniaP&ID





Figure1SingleStage

Figure2SingleStageEconomizedScrew

Figure3TwoStage





































Page

2

4

4

6

6

7

10

15

16

22

23

26

27

28

28

29

32

32

36

37

40

41

42

46

47

50

52

Page1of52
ENG00019544REV02013ColmacCoilManufacturingInc.

P.O. Box 571 | Colville, WA 99114 | USA


T: +1.509.684.2595 | F: +1.509.684.8331
www.colmaccoil.com


I.

Background

Ammoniarefrigerationsystemshavetraditionallyemployedevaporatorssuppliedwith
liquidbyeithergravityflooding(withsurgedrums),orpumpedoverfeed(eitherwith
mechanicalpumpsordischargegasdrivenvessels).Bothofthesedesignstypicallyuse
bottomfeedcoilcircuitingwhichfeedsliquidammoniaatthelowestpointinthecoilcircuit
andcausestheammoniatoflowupwardandpercolatethroughthecoilinascending
passestotheoutletatthetopofthecircuit.Thesecoildesignsalsotypicallyuselarge
diametertubingwhichmeansrelativelylargecoilinternalvolume.Thiscombinationof
refrigerantfeed,circuiting,andtubediameter,resultsinthegreatestevaporatorcharge
inventorypossible.

Endusersofammoniarefrigerationsystemsareincreasinglyinterestedinreducingthe
chargeofammoniainevaporators(andintheoverallsystem)intheinterestofminimizing
therisktoworkersandproductsassociatedwithammonialeaks.Oneveryeffectivewayto
significantlyreduceevaporatorammoniachargeistodesignandoperatetheevaporator
usingdryexpansion(DX)circuitingandcontrols.UsingDXammoniacanreducethe
evaporatorchargebyasmuchas30to50timescomparedtobottomfeedfloodedor
pumpeddesigns.Themagnitudeofthisreductioninammoniachargemayalsomitigate
regulatoryrequirements(PSM,RMP),andpotentiallyreduceinsuranceriskandpremiums.

DXammoniahasbeenusedforsometimeinmediumandhightemperaturesystems
(suctiontemperaturesabove+20degreesF)withsomesuccess.However,inspiteofthe
chargereductionadvantagesmentionedabove,todateDXammoniahasnotbeenapplied
successfullyatfreezertemperatures.Atsuctiontemperaturesbelowabout+20F,the
followingparticularcharacteristicsofammoniaresultinextremelypoorperformanceof
evaporatorsunlessaddressedandmitigated:

1. Separationofliquidandvaporphases.Theveryhighratioofvaportoliquidspecific
volumeofammoniaatlowtemperaturescombinedwithitsveryhighlatentheatof
vaporizationcausesanunavoidableseparationofvaporandliquidphasesinside
evaporatortubes.Thisseparationofphasescausestheliquidammoniapresenttorun
alongtheverybottomofthetubesleavingthetopofthetubescompletelydry.The
resultisextremelypoorevaporatorperformanceandlowerthanexpectedsuction
temperaturesduringoperation.TosolvethisproblemColmachasdeveloped(and
patented)anenhancementtechnique,whichwhenappliedtotheinsideofevaporator
tubes,causestheliquidammoniapresenttocoattheentireinsidesurfaceofthetubes
bycapillaryaction.PerformancewithColmacenhancedtubetechnologyresultsinDX
ammoniaperformanceatlowtemperatureswhichisasgoodorbetterthan
performancewithbottomfeedpumpedammoniacircuiting.
2. Refrigerantdistributortechnology.Traditionallythedistributionofexpandedrefrigerant
tomultipleparallelevaporatorcircuitshasbeendoneusingarefrigerantdistributor
havingafixedorificeplate.Thisdesigndependsonarelativelylargepressuredrop
(approximately4045psi)acrossthefixedorificetothoroughlymixandequally
distributortheliquidandvaporphasesbeforetheyenterthedistributortubesand
evaporatorcircuits.Thisrelativelyhighpressuredropacrossthedistributorreducesthe

Page2of52
ENG00019544REV02013ColmacCoilManufacturingInc.

P.O. Box 571 | Colville, WA 99114 | USA


T: +1.509.684.2595 | F: +1.509.684.8331
www.colmaccoil.com


pressuredropavailablefortheexpansionvalve,andconsequentlylimitshowlow
condensingpressurecanbeallowedtofallduringperiodsoflowambienttemperature.
Theveryhighlatentheatofvaporizationofammoniaresultsinlowrefrigerantmass
flowrateandconsequentlyaverysmallorificediameterforagivencoolingload(the
orificecanbeassmallas1/16diameterinsomecases).Thissmallorificesizeisprone
tofoulingandbeingblockedbyevensmallsizedebris.Otherdisadvantagesofthis
distributordesigninclude:
a. Performanceisverysensitivetoliquidtemperature(subcooling)atthe
expansionvalve.
b. Operatingrangeissmall,atmost50%to125%ofratedcapacity.
c. Theorificeanddistributortubesrestricttheflowofhotgasduringahotgas
defrostcycle.
d. Themaximumnumberofparallelevaporatorcircuitsavailableinasingle
distributorislimitedtoonly15.
ToaddresstheseshortcomingsColmachasdevelopedanew(patentpending)
refrigerantdistributortechnology,theColmacTankDistributor,havingthefollowing
characteristics:
a. RefrigerantpressuredropacrosstheTankDistributorduringoperationisvery
low,only24psi.
b. AnyoilordebrisenteringtheTankDistributoriscapturedinadropleg(whichis
integraltothedesign)beforeitcanenterthecoilandfoultubesurfaces.
c. PerformanceoftheTankDistributoriscompletelyinsensitivetoliquid
temperature(subcooling).
d. Graduatedorificesineachdistributortubeallowequaldistributionof
refrigeranttoallcircuitsoveranextremelywideoperatingrangeof0%to700%
ofratedcapacity.
e. Graduatedorificesandlargediameterdistributortubesallowfullflow(minimal
restriction)ofhotgasduringhotgasdefrost.
f. ThenumberofparallelevaporatorcircuitspossibleinasingleTankDistributor
canbeashighas48.
3. Removalofwaterfromammonia.Asdescribedelsewhere(Nelson2010),evensmall
amountsofwater(13%)intheammoniawillsignificantlypenalizeDXammonia
evaporatorperformance.Watermustbeeffectivelyremovedduringoperation,
particularlyinfreezingsystemswhichoperateatsuctionpressuresbelowone
atmosphere(inavacuum).Currently,theonlyeffectivewaytoremovewaterfrom
ammoniaisinaheateddistillationvessel(anammoniastill).Thisverynegativeeffect
ofsmallamountsofwateronevaporatorperformancehasnotbeenfullyrecognizedin
thepast,butmustbeaddressedduringthedesignoftheDXammoniasystem.Colmac
hasdevelopedaneffectiveammoniadistillationvesseldesignandinstallationstrategy
whichisdescribedwithinthisHandbook.

Colmachasdeveloped,tested,andpatented(Nelson2011)anewLowTemperatureDX
Ammoniasystemwhichcorrectlyaddressesalloftheaboveissuespeculiartoammoniaasa
refrigerantthathaveheretoforepreventeditsuseatlowsuctiontemperatures.Itisnow
possibletosuccessfullyapplyDXammoniaatsuctiontemperaturesdownto50degreesF.

Page3of52
ENG00019544REV02013ColmacCoilManufacturingInc.

P.O. Box 571 | Colville, WA 99114 | USA


T: +1.509.684.2595 | F: +1.509.684.8331
www.colmaccoil.com


ThisPipingHandbookisintendedtoguidethereaderthroughtheprocessofsuccessfully
designingandimplementingDXAmmoniafrom+50Fto50Fandrealizingthebenefitsof:

Dramaticallyreducedammoniacharge
Simplifiedcontrols
Energyefficientdrysuctionline
Reducedlinesizes
Eliminationofammoniarecirculatorpumps

II.

SystemConfiguration

ColmacDXAmmoniacanbeappliedtoanytemperaturelevelandsystemconfiguration.
P&IDdiagramsforvarioustypicalsystemsareshowninAppendixA,simplifiedforpurposes
ofclarity.Selectionandsystempipingdetails(reliefvalves,purgers,isolationvalves,vessel
designs,etc)shouldfollowindustryguidelinesasfoundintheIIARAmmoniaPiping
Handbook(IIAR2004).Thediagramsarenotintendedtopresentanexhaustiverangeof
configurationseveryindustrialrefrigerationsystemwillhaveuniquefeaturesand
requirements.Thisinformationispresentedtoillustratethegeneralsystemfeatures
particulartoasuccessfulDXAmmoniadesign.

a. Figure1SingleStageSingleTemperatureLevel
b. Figure2SingleStage(EconomizedScrew)MultipleTemperatureLevel
c. Figure3TwoStageMultipleTemperatureLevel

III.

SystemStability

Withliquidoverfeedandgravityfloodedsystems,liquidreturntotherecirculatorvesselor
thesurgedrumisnormalandexpectedthroughthewetsuctionline.Therecirculatorvessel
orsurgedrumeffectivelyseparatesreturningliquidfromvaporandinsuresthatthedry
suctionlinecarriesonlyvaporbacktothecompressor.

DXsystems,ontheotherhand,aredesignedtooperatewithadrysuctionlineandareby
definitionmoresensitivetoliquidfloodback.IndustrialDXsystemsshouldincorporatea
suctionaccumulatorvesseltopreventliquidsluggingofthecompressorduringafloodback
event,however,excessivefloodbackfromevaporatorscancausehighlevelalarmingand
systemshutdownuntiltheexcessliquidinthesuctionaccumulatorcanbetransferredback
tothehighpressuresideofthesystem.Stableandsmoothoperationofthesystemandthe
evaporatorexpansionvalve(s)iscriticaltoavoidingliquidfloodback.Instabilitiesand/or
rapidchangesindischargeandsuctionpressuresduringoperationarethetypicalcauseof
unstableoperationofexpansionvalvesandshouldbeconsideredcarefullybythesystem
designerandoperator(s).

Rapidchangesinsystemdischargepressurecancausesysteminstabilitiesinanumberof
ways.Asuddenreductionindischargepressurecanresultinundesirableflashingofliquid
refrigerantinliquidlinesandwillalsobeaccompaniedbyasympathetic,albeitsmaller,
reductioninsuctionpressure.Asuddenincreaseindischargepressurewillbeaccompanied

Page4of52
ENG00019544REV02013ColmacCoilManufacturingInc.

P.O. Box 571 | Colville, WA 99114 | USA


T: +1.509.684.2595 | F: +1.509.684.8331
www.colmaccoil.com


byasympathetic,albeitsmaller,increaseinsuctionpressure.Anincreaseinsuction
pressure,iflargeenoughandrapidenough,willsuppressboilingintheevaporatorswhich
candirectlyleadtoliquidfloodbackfromtheevaporatorstothesuctionaccumulator.

Rapidchangesindischargepressurearenormallycausedby:
a. Condenserfanscyclingonandoff,or
b. Evaporativecondenserpumpscyclingonandoff,or
c. Compressor(s)cyclingonandoff

**NOTE:Designthesystemtolimittherateofchangeincondensingtemperaturetono
morethan5degF/minute.

Rapidchangesinsystemsuctionpressurecanalsoresultinsysteminstabilityandpoor
performance.Itisasuddenincreaseinsuctionpressurethathasthehighestpotentialfor
liquidfloodbackfromDXevaporators.Thissuddenincreaseinsuctionpressureraisesthe
temperatureoftheevaporator,reducestheimposedload,andresultsinliquidrefrigerant
exitingtheevaporatorbeforetheexpansionvalvecanrespondandreducetheflowof
refrigerantenteringtheevaporatoraccordingly.

Rapidchangesinsuctionpressurearenormallycausedby:
a. Compressor(s)cyclingonandoff
b. Multipleliquidfeedsolenoidscyclingonandoff
c. Evaporatorfanscyclingonandoff
d. Evaporatorsstartingorfinishingdefrost
e. Suddenchangesinimposedloadonevaporators

**NOTE:Designthesystemtolimittherateofchangeinsuctiontemperaturetonomore
than2degF/minute.

Followingarerecommendedsystemdesignfeatureswhichwillservetomaximizesystem
pressurestabilityandminimizethepotentialforliquidfloodbackfromevaporators.

1. CondenserFans
a. UseofVFDfanspeedcontrolinsteadoffancyclingforcontrolofheadpressureis
recommended.

2. CondenserPumps
a. Itisalsorecommendedthatevaporativecondensersumpwaterpumpsbeoperated
continuouslyratherthancyclingonandoff,providedambientweatherconditions
allow.

3. CompressorCapacityControl
a. UseofVFDspeedcontrolforcapacitywherepossibleandappropriate.
b. Limitcapacityloading/unloadingsteps(on/off)tonomorethan10%oftotalsystem
capacity.

Page5of52
ENG00019544REV02013ColmacCoilManufacturingInc.

P.O. Box 571 | Colville, WA 99114 | USA


T: +1.509.684.2595 | F: +1.509.684.8331
www.colmaccoil.com

4.

5.

6.

7.

c. Limittherateofchangeofsuctiontemperature(speedofscrewcompressorslide
valvemovement)tonogreaterthan2degF/minute.

EvaporatorDefrost
a. Defrosttheminimumnumberofevaporatorsatonetime.
b. Useableedlinetoequalizepressureslowlyattheendofdefrost.

EvaporatorFans
a. FanspeedandcoolingcapacitycanbecontrolledbyVFD,howeverthefollowing
guidelinesmustbeobservedwhenappliedtoDXevapaorators:
o Rateofchangeinfanspeedmustbegradualandlimitedtoresultinno
morethan2degF/minutechangeinsuctiontemperature.
o Minimumfanspeedmustbesettoproducenolessthan250ft/minface
velocity.
b. Iffansaregoingtobecycledon/offforcapacitycontrol,nomorethan10%ofthe
totalnumberofevaporatorfansshouldbecycledonoroffatthesametime.

LiquidFeedSolenoids
a. Avoidcyclingmultipleliquidfeedsolenoidsallatthesametime.i.e.Liquidfeed
solenoidsshouldbecycledsequentially.

SuddenchangesinloadonEvaporators
a. Avoidlocatingevaporatordirectlyabovedoorways.
b. Mitigateintermittentprocessloadslocatedclosetoevaporators.


IV.

EvaporatorSelectionandOperation

1. DT1vsDTMratings

Asexplainedindetailelsewhere(Nelson2012(a))evaporatormanufacturerstypically
presenttheircapacityratingsusingoneoftwodefinitionsoftemperaturedifference,
DT1orDTM.SomemanufacturerspublishratingsbasedonbothDT1andDTMand
allowthedesignertochoosethepreferreddefinition:

DT1=AirOnTemperatureEvaporatorTemperature
DTM=Average(Room)AirTemperatureEvaporatorTemperature

Figure1belowgraphicallyillustratesthesetwodefinitionsoftemperaturedifferencefor
thesameevaporatorandtheireffectonLMTD(LogMeanTemperatureDifference),and
henceratedcapacity.Inthisexample,thesameevaporatorhavinga20degF
evaporatingtemperatureratedusingDTMproduces33.3%(DTMLMTDof9.6degF
versusDT1LMTDof7.2degF)morecapacitythanthesameevaporatorratedusing
DT1!


Page6of52
ENG00019544REV02013ColmacCoilManufacturingInc.

P.O. Box 571 | Colville, WA 99114 | USA


T: +1.509.684.2595 | F: +1.509.684.8331
www.colmaccoil.com


Inshort,byusingtheDTMratingmethodamanufacturercanshowcoolingcapacities
thataremuchhigher(30to40%higher),andsoofferalowercostevaporatorwith
muchlesssurfaceareathanthemanufacturerusingtheDT1ratingmethod.

Unfortunately,onecannotgetsomethingfornothing.Eventhoughevaporators
selectedusingDTMratingswillbecheaperinitiallybecausetheyhavelesssurfacearea,
theywillcausethesystemtorunatalowersuctionpressurewithhigheroperatingcosts
thanevaporatorsselectedusingDT1ratings.Thisdifferenceinoperatingcostbetween
DTMandDT1evaporatorshasbeencalculatedandtheincrementalreturnon
investmentshowntodramaticallyfavorselectingevaporatorsusingDT1ratings(Nelson
2012(b)).Additionally,inthesamearticletheauthorshowsthatthebasicDTM
assumptionthattheaverageairtemperaturewithintheevaporatorequalstheaverage
roomtemperatureisafundamentallyflawedandfalseassumptionbecauseofair
entrainmentandmixingintheroom.

FIGURE1
TemperatureProfilesforDT1vsDTM

(a) DT1=10F(AirOn)TempDifference(b)DTM=10F(Average)TempDifference

DT1 LMTD = 7.2 F

DTM LMTD = 9.6 F

Inconclusion,ColmachighlyrecommendsthatevaporatorsbeselectedusingDT1ratings
ratherthanDTM.

2. SensibleHeatRatio,RoomRelativeHumidity(rh%),andEvaporatorRatings

Accuratepredictionoftherefrigerationload,bothsensibleandlatentcomponents,is
importanttoproperrefrigerationsystemequipmentselectionandsuccessfuloperation
Page7of52
ENG00019544REV02013ColmacCoilManufacturingInc.

P.O. Box 571 | Colville, WA 99114 | USA


T: +1.509.684.2595 | F: +1.509.684.8331
www.colmaccoil.com


(Nelson2012(a)).Varioustypesofsensiblecoolingloadsmustbeanticipatedand
includedinthecalculation,suchas:lighting,electricmotors,forklifts,product
cooling/freezing,transmissionofheatthroughwalls,ceilings,andfloors,andcoolingof
infiltrationair.Latentcoolingloadsarepresentwhenevermoistureisaddedtotheairin
therefrigeratedspace.Sourcesofintroducedmoisturetypicallyinclude:infiltrationair,
respiringfoodproducts,surfacemoistureonproducts,packagingandotherobjects
enteringthespace,residualwaterleftonfloorsafterwashdown(processrooms),
humanrespiration,andhumidificationequipment(abovefreezing).

Roomrelativehumidity(rh%),whichistheindicationofhownearlytheairinthe
refrigeratedspaceissaturatedwithwatervapor,willbetheequilibriumcondition
resultingfromthebalanceofmoistureintroducedintothespacewiththemoisture
removedfromspacebytheevaporatorcoils(Cleland2012).

Wheneverevaporatorsurfacesoperateattemperaturesbelowthedewpointoftheair
beingcooled,watervaporintheairstreamiscondensedtoliquid(attemperatures
above32F(0C))ordepositedtoformfrost(below32F(0C)).Thecoolingeffect
associatedwiththisdehumidificationoftheairstreamistermedlatentcooling.The
sumofthesensiblecoolingloadandlatentcoolingloadistermedthetotalload.The
ratioofthesensiblecoolingloaddividedbythetotalcoolingloadiscalledtheSensible
HeatRatio(SHR)anddefinestheslopeoftheairprocesslineonapsychrometricchart.










 







(1)

Refrigeratedspaceswithproductbeingtransferringinandoutthroughdoorwayswill
verytypicallyhavearelativehumidityinthe85to95%rangeduetoinfiltrationand
othersourcesofmoisture.Lowerroomrelativehumiditymaybefoundinsome
exceptionalcaseswheretrafficthroughdoorwaysisverylight,productistightly
packaged,dehumidificationequipmentisusedatdoorways,etc.Therelationship
betweenroomrelativehumidityandSHRisshowninTable1below:

TABLE1
SHRFORDT1=10degFATVARIOUSTEMPERATURESANDROOMRH%


SensibleHeatRatio,SHR

RoomTemperature,F(C)
65%rh
75%rh
85%rh
95%rh

45(7.2)
1.0
0.84
0.67
0.56

32(0)
0.98
0.84
0.73
0.64

10(12.2)
0.98
0.92
0.87
0.83

0(17.8)
0.98
0.95
0.92
0.89

10(23.3)
0.99
0.97
0.95
0.93

30(34.4)
0.99
0.99
0.98
0.97

TheroomrelativehumidityandresultingSHRcanhavealargeeffectonevaporator
coolingcapacity,especiallyathigherroomtemperatures.
Page8of52
ENG00019544REV02013ColmacCoilManufacturingInc.

P.O. Box 571 | Colville, WA 99114 | USA


T: +1.509.684.2595 | F: +1.509.684.8331
www.colmaccoil.com

CapacityFactor,Qtotal/Qsensonly


Usingacomputermodeldevelopedtoaccuratelycalculatefinefficiencyandsurface
effectivenessforbothsensibleandcombinedsensibleandlatentheattransfer,a
predictionoftheincreaseinevaporatorcoilperformanceasafunctionofSHRhasbeen
made(Nelson2012(a)).ResultsofthepredictedcapacityincreaseasafunctionofSHR
foranammoniarefrigerationevaporatorcoiloperatingoverawiderangeofroom
temperatures(+45Fto30F)andhavingtypicalfinspacingsandgeometrywithDT1=
10FareshowninFigure2below.

FIGURE2


TotalCoolingCapacityFactorvsSHR

1.5

1.45

1.4

1.35


1.3

1.25

1.2

1.15

1.1

1.05

1

0.6
0.65
0.7
0.75
0.8
0.85
0.9
0.95
1


SensibleHeatRatio,SHR

Tomakethingsabitmorecomplicated,someevaporatormanufacturersincludethe
effectofroomrh%intheirratings,othersdonot.AsshowninFigure2,thelowerthe
SHRthegreaterthetotalcoolingcapacityoftheevaporator.Amanufacturerwhoshows
theirevaporatorratingsasallsensible(SHR=1)willbemoreconservative(havemore
surfacearea)thanthemanufacturerwhoshowstheirratingsat85or95%rh.

Selectingevaporatorsusing85to95%rhratingswillresultinevaporatorshavingless
surfaceareaandlowerfirstcostcomparedtoevaporatorsselectedusingallsensible
ratings.Theriskinthisapproachisundersizingtheevaporatorsinthecasewherethe
actualoperatingroomrh%islessthantherh%usedduringtheselectionprocess.

Conclusion:Thelatentloadshouldalwaysbeestimatedandincludedinthetotal
calculatedrefrigerationload.Sizeevaporatorsforthedesigntotalcalculated
refrigerationloadattheestimatedroomrelativehumidity.Ifroomrelativehumidityis
difficulttoestimateorcannotbeestimated,thenaconservationapproachistoselect
evaporatorsbasedonalowroomrelativehumidity(i.e.65to75%rh)orusingsensible
onlyratings.

3. OptimizingSystemTD
Page9of52
ENG00019544REV02013ColmacCoilManufacturingInc.

P.O. Box 571 | Colville, WA 99114 | USA


T: +1.509.684.2595 | F: +1.509.684.8331
www.colmaccoil.com



Theproductbeingstoredorprocessednormallydeterminestheroomairtemperature
inarefrigeratedfacility.Appropriatetemperaturesforstoringandprocessingvarious
foodsandfoodproductscanbefoundelsewhere(ASHRAE2009).

Oncetheroomtemperatureisdetermined,theevaporatortemperaturemustbe
decideduponbythedesigner.Compressorpowerandenergyconsumptionisastrong
functionofthesuctionpressureandtemperature.Thehigherthesuctionpressurethe
moreefficientlythecompressorwillrunandthelesspowerwillbeconsumed.Energy
efficiencycanbecharacterizedbyaratiotermedCoefficientofPerformance(COP),
definedas:


 




 



(2)


Inthecaseofarefrigerationcompressor,









 


(3)



Figure3belowshowstypicalammoniascrewcompressorCOPvsSST(SaturatedSuction
Temperature).Thefigureassumes2Stagecompressionisusedbelowasuction
temperatureof20degF.


FIGURE3
























R717ScrewCompressorCOPvsSST
(SCT=85DegF)
8

CompressorCOP

1
50

40

30

20

10

10

20

30

40

SST,DegF


Page10of52
ENG00019544REV02013ColmacCoilManufacturingInc.

P.O. Box 571 | Colville, WA 99114 | USA


T: +1.509.684.2595 | F: +1.509.684.8331
www.colmaccoil.com



ItwouldappearfromFigure3thatasmallerTD(TD=RoomTemperatureEvaporator
Temperature)wouldalwaysbedesirablefromanenergyconsumptionstandpointsince
thesmallertheTD,thehighertheevaporator(SST)temperatureandcompressorCOP.
This,however,isnotthecase.

Heatistransferredfromtheroomviatheaircirculatedbytheevaporators.Thecooling
capacityofanevaporatorcanbecharacterizedbytheNTUeffectivenessequation.This
equationindicatesthatforaconstantcoolingcapacityandevaporatoreffectiveness(an
expressionofhowcloselytheleavingairtemperatureapproachestheevaporating
temperature),theflowrateoftheairwillbeinverselyproportionaltotheTD.





    


(4)


Where:



 
 
Evaporatoreffectivenessis,infact,verynearlyconstantoverthetypicalnarrow
operatingrangeofarefrigerationevaporator.Theeffectivenessequationshowsthatas
TDbecomessmaller,theairflowratemustbecomelargerinthesameproportionfora
givencoolingcapacity.

Fanpowercanbecalculatedusingasimpleequationasfollows:








  



(5)



Where:


 


Theairpressuredropthroughtheevaporatorcoil,andthereforefanpower,willbe
affectedby:
1. Thecoilfacevelocity,
2. Tubediameter,spacing,andpattern,
3. Numberofcoilrowsdeep,
4. Finspacingandpattern
5. Frostthickness


Page11of52
ENG00019544REV02013ColmacCoilManufacturingInc.

P.O. Box 571 | Colville, WA 99114 | USA


T: +1.509.684.2595 | F: +1.509.684.8331
www.colmaccoil.com

COP


TherelationshipsaboveindicatethatcompressorCOPwilldecreasewithincreasingTD
whileFanCOPwillincreasewithincreasingTD.Figure4showstheserelationshipsforan
exampleevaporatorcoilhaving8rowsdeepand3FPIfinspacing.


FIGURE4


COPvsTD

8Row3FPI,R717,AirOnTemp=10DegF

7


6


5

4
Fan(500FPM)x10^1

Fan(750FPM)x10^1
3
Fan(1000FPM)x10^1

2
Compressor

1

0
5
10
15
20
25
30
35
0
TD=AirOnTemp SuctionTemp,DegF





Figure4impliestherewillbesomemaximumcombinedCOPforcompressorandfans
whichwillrepresenttheoptimumoperatingTDintermsofenergyefficiency.This
combinedCOPisshownbelowinFigures5,6,and7,foratypicalammoniaevaporator
coilhavingthefollowingcharacteristics:

Tubing:7/8ODAluminum
TubePattern:2.25Staggered
Fins:ConfiguredAluminumPlateType
RowsDeep:8
FaceVelocity:500,750,and1000FPM
FinSpacing:2,3,and4FPI
AirOnTemperature:10degF
FrostThickness:0mm




Page12of52
ENG00019544REV02013ColmacCoilManufacturingInc.

P.O. Box 571 | Colville, WA 99114 | USA


T: +1.509.684.2595 | F: +1.509.684.8331
www.colmaccoil.com


FIGURE5

Compressor+FanCOPvsTD
8Row4FPI,R717,AirOnTemp=10DegF
2.4
2.2
2
1.8
1.6
500FPM
1.4

750FPM

1.2

1000FPM

1
0.8
0.6
0

10

15

20

25

30

35

TD=AirOnTemp SuctionTemp,DegF

FIGURE6

Compressor+FanCOPvsTD
8Row3FPI,R717,AirOnTemp=10DegF
2.6
2.4
2.2

CombinedCOP




































CombinedCOP














2
500FPM

1.8

750FPM
1.6

1000FPM

1.4
1.2
1
0

10

15

20

25

30

35

TD=AirOnTemp SuctionTemp,DegF

Page13of52
ENG00019544REV02013ColmacCoilManufacturingInc.

P.O. Box 571 | Colville, WA 99114 | USA


T: +1.509.684.2595 | F: +1.509.684.8331
www.colmaccoil.com


FIGURE7

Compressor+FanCOPvsTD
8Row2FPI,R717,AirOnTemp=10DegF
2.6
2.4
2.2

CombinedCOP


























2
500FPM

1.8

750FPM
1.6

1000FPM

1.4
1.2
1
0

10

15

20

25

30

35

TD=AirOnTemp SuctionTemp,DegF

ThefollowingisobservedfromFigures5through7:
a. CombinedCOPisaverystrongfunctionofcoilfacevelocity.COPat500FPMis
approximately10%higherthanCOPat750FPMand20%higherthanCOPat
1000FPM.
b. CombinedCOPincreasesasthedistancebetweenfinsisincreased.Coilswith
2FPIspacingwillhavehighercombinedCOPthancoilswith3FPI,whichwillhave
higherCOPthan4FPI.
c. Theoptimum(maximum)TDincreaseswithincreasingfacevelocity.
d. Inallcases,combinedCOPdecreasesveryrapidlybelowabout7degFTD.

InordertomakethefinaldecisionaboutselectingtheoptimumTD,thecostofpower
aswellasinstalledcostofthecompressor(s)andevaporatorsmustbeknown(or
estimated).Thesevariablescanthenbecombinedtocalculatetheincrementalreturn
oninvestmentcomparingdifferentevaporatordesigns(facevelocityandfinspacing)in
termsoffirstcostvsoperatingcost.

Sincethesecostsarehighlyvariable,thefinalreturnoninvestmentcalculationmustbe
madeonacasebycasebasisandpresentedtotheclientinawaywhichallowsthefinal
decisiontobemadegiventheprojectfinancialconstraintsandrequirements.



Page14of52
ENG00019544REV02013ColmacCoilManufacturingInc.

P.O. Box 571 | Colville, WA 99114 | USA


T: +1.509.684.2595 | F: +1.509.684.8331
www.colmaccoil.com


Conclusions:
1. ForhighestsystemCOP/energyefficiency,selectevaporatorsforthelowestface
velocityandwidestfinspacingfinanciallypractical.Colmacrecommendsmaximum
facevelocityof600FPMandfinspacingof3FPIorwider(lowerFPI).
2. Forcoilfacevelocitiesbetween500and750FPMadesignTDbetween10degFand
15degFisrecommended.
3. FinaloptimizedevaporatordesignandTDmustbedeterminedbasedonspecific
projectfinancialconstraintsandacceptablereturnoninvestment.

4. EffectofTDonExpansionValve(EV)Operation

Withdirectexpansion(DX)evaporatorstheflowofrefrigeranttotheevaporatoris
meteredbyanautomaticexpansionvalveinresponsetoacontrolsignalmeasuredat
theevaporatoroutlet.Thecontrolsignalisnormallytheamountofsuperheatinthe
refrigerantsuctiongas.Thetheoreticalmaximumamountofsuctiongassuperheatthat
canbegeneratedisequaltotheoperatingTD(TD=AirOnTemperatureEvaporating
Temperature).

Theamountofsuperheatrequiredforstableoperation(modulation)oftheexpansion
valvevarieswiththetypeofvalveemployed.Twobasictypesofexpansionvalvesare
currentlyavailableonthemarket,Thermostatic(TEV)andElectronic(EEV).Bothuse
superheatinthesuctiongasasthecontrolsignal.

Thermostaticexpansionvalvesmeasureandmechanicallycalculatesuperheatbymeans
ofatemperaturesensingbulbandpressureequalizingline.Thesevalvesandtheir
operationaredescribedindetailbythevalvemanufacturers.Theadvantageofthistype
ofvalveistheirlowcostandcompactness.Withthistypeofvalve,temperaturesensing
isaccomplishedbyarefrigerantfilledbulbstrappedtotheoutsideofthecoilsuction
connection.Adisadvantageofthissystemistheadditionalsuperheatrequiredto
overcomethethermalresistanceofthepipewall.Thisadditionalsuperheatforcesthe
operatingTDtobeapproximately5degFgreaterthanforanelectronicexpansionvalve
thatusesatemperaturetransducertomeasuretemperaturedirectly.

Electronicexpansionvalvesoperatebasedonasignalreceivedfromasuperheat
controllerwhichreadssuctiongastemperatureandpressurefromacombinationof
sensors.Theexpansionvalveitselfmayoperatebasedonanopen/close(pulsing)
principleoronamotorizedpositioningprinciple.Advantagesofthistypeofvalve
includemoreaccurateandresponsivesensingofsuperheatwhichallowsstable
operationatsmallerTDthanthermostatictypevalves.PIDcontrolparameterscanalso
beadjustedinthecontrollertofinetuneoperationoverawiderangeofconditions.
Theprimarydisadvantageofelectronicexpansionvalvesisthehigherfirstcost
comparedtothermostaticvalves.This,however,ischangingasvalvemanufacturersare
findinglowercostsolutionsandbeginningtooffercostcompetitiveelectronicvalvesto
themarket.


Page15of52
ENG00019544REV02013ColmacCoilManufacturingInc.

P.O. Box 571 | Colville, WA 99114 | USA


T: +1.509.684.2595 | F: +1.509.684.8331
www.colmaccoil.com


MinimumrecommendedTDandsuperheatsettingsforbothtypesofexpansionvalves
areshowninTable2below:

TABLE2
MINIMUMRECOMMENDEDDXAMMONIATDANDSUPERHEATSETTING

ExpansionValveType MinimumRecommended RecommendedSuperheat
TD,degF
Setting,degF
Thermostatic
15
12
Electronic
10
8

Note:Colmacoffersfactorysuppliedandmountedexpansionvalvesandcontrollers,
boththermostaticandelectronictype.

5. TypesofFrostandSelectionofFinSpacing

Frostcanaccumulateonevaporatorcoilfinsbyoneoftwomechanisms:

1. Bydeposition,and/or
2. Asairborneicecrystals

Designingevaporatorstoproperlyhandlethesetwotypesoffrostisdescribedinthis
section.

1. Deposition:

Wheneverthetemperatureoftheevaporatorcoilsurfaceisbelowthedewpoint
temperatureoftheroomair,moisturewillcondenseandbedepositedonthesurface
eitherasliquidwater(abovefreezing)orasfrost(belowfreezing).Thismasstransfer
process,whenrelatedtotheformationoffrostiscalleddeposition,andisdrivenbythe
differenceinwatervaporpressurebetweentheairandthesurfaceofthecoil.The
amountofheatassociatedwiththismasstransferprocessistermedlatentheatandis
quantifiedbytheSHR(seepreviousdefinition).WhenevertheSHRislessthan1.0,the
depositionoffrostwilltakeplace.Therateatwhichfrostwillbedepositedonthecoil
surfacescaneasilybecalculatedasafunctionofthetotalcoolingload,theSHR,andthe
surfaceareaoftheevaporator.

Thesurfaceeffectivenessofarefrigerationevaporatorisrelativelyhigh(usuallygreater
than80%)duetothetypicallysmallTDandlowheatfluxcomparedtoairconditioning
andprocessevaporators.Thishighsurfaceeffectivenessresultsinamoreorless
constantsurfacetemperatureanduniformdepositionoffrostovertheentiresurfaceof
theevaporator.Thisassumptionofuniformfrostdepositionismadeinthefollowing
equation:



Page16of52
ENG00019544REV02013ColmacCoilManufacturingInc.

P.O. Box 571 | Colville, WA 99114 | USA


T: +1.509.684.2595 | F: +1.509.684.8331
www.colmaccoil.com








 



(6)

 



Where:
 
 

 
 
 

EXAMPLE:
Anevaporatorhaving8rowsdeepandfinspacingof3FPIisoperatingwitha10degF
TD(DT1)ina+10degF/85%rhroom.Theevaporatorhasacoolingcapacityof240,000
Btu/h(20TR)andoutsidesurfaceareaof4100ft2.Whatwillbetherateoffrost
deposition?

Answer:
FromTable1theexpectedSHRatthisroomairtemperatureandrh%willbe0.87.







Asfrostisdepositedontheevaporatorcoilsurfacesthelocalairvelocitybetweenfins
willincreaseandresultinincreasedairpressuredropacrossthecoil.Theincreaseinair
pressuredropduetoaccumulationoffrostcanbeapproximatedbythefollowing
equation:




 





(7)

 





Where:
 

 
 
 

Page17of52
ENG00019544REV02013ColmacCoilManufacturingInc.

P.O. Box 571 | Colville, WA 99114 | USA


T: +1.509.684.2595 | F: +1.509.684.8331
www.colmaccoil.com


Therateofblockageofthecoilwithfrostandassociatedpressuredropwillresultina
reductioninairflowandcoolingcapacity.Thisreductionincoolingcapacitywill
ultimatelydeterminedefrostfrequencyandefficiency,andoverallsystemenergy
efficiencyandpowerconsumption.
Coolingcapacityoftheevaporatorcanbecharacterizedasfunctionsoffacevelocity,
SHR,andfrostthickness.Knowinghowcapacitychangeswiththeseparameters,
combinedwiththerelationshipsshowninequations5,6,and7above,allowsthe
constructionofasimplemodelwhichwillpredictthechangeincoilcapacityovertime.
Simplifyingassumptionsinthemodelinclude:

o Frostisdepositeduniformlyoverthesurfaceofthecoil
o Frostdensityisuniformandofafixedvalue
o Suctiontemperatureremainsconstant

Normally,refrigerationloadsarecalculatedbasedon16to18hoursofruntimeperday.
Itisnotclearwhenorwhythisruleofthumbcameintopractice.Perhapsitisacorollary
tothe2to1ruleforhotgasdefrosting.Thatis,twocoilsmustbeinoperationwhile
thethirdcoilisinhotgasdefrostinordertoprovideasufficientquantityofhotgasfor
thedefrostcycle.Perhapsitissimplyanadditionalcatchallsafetyfactor.One
industryhistorianmentionedthatthe1618runhoursrulecamefromsplitsystem
applicationswherecapacitydropsoutduringdefrostcomparedtoacentralsystem
whichwouldrebalancetheTDontheremainingevaporators(Welch2013).Itmakesthe
mostsensetotheauthorthattheruntimeadjustmenttothedesignrefrigerationload
shouldbeusedtoaccountforthedegradationincoilperformanceovertimedueto
accumulationoffrost.

Inalargerefrigerationsystemhavingaconstantrefrigerationloadandcompressor
unloadingcapability,thecompressorswillunloadtomaintainaconstantsuction
temperatureasthecoilcapacityfallsoffduetofrosting.Asthecompressorsunloadto
maintainsystemsuctionpressure,theywillrunlongertomaintainroomtemperature.
Thisimpliesthatinordertomaintainroomtemperature,defrostingshouldbeinitiated
whenevaporatorcapacityfallstoalevelequaltothedesignruntimeratio(design
runtimedividedby24).Table3belowshowsthemaximumreductioninevaporator
capacityduetofrostingthatshouldbeallowedbeforedefrostisinitiated.Notethatthis
tableobviouslydoesnotapplytoevaporatorsoperatingabovefreezing.

TABLE3
MINIMUMEVAPORATORCAPACITYTOINITIATEDEFROSTATVARIOUSDESIGN
RUNTIMES

DesignRuntime,h/day
MinimumEvaporatorCapacityat

InitiationofDefrost

14
58%

16
67%

18
75%

20
83%

Page18of52
ENG00019544REV02013ColmacCoilManufacturingInc.

P.O. Box 571 | Colville, WA 99114 | USA


T: +1.509.684.2595 | F: +1.509.684.8331
www.colmaccoil.com

CoolingCapacity,%

CoolingCapacity,%

Theevaporatorperformancemodeldescribedabovewasusedtoexaminetheeffectof
finspacingandSHRonlossofcoolingcapacityduetofrostaccumulation.SeeFigures8,
9,10,and11below.

FromTable1itisclearthatthehighestfrostload(lowestSHR)willoccurinhigh
temperature(+32F)roomswithhighrelativehumidity.Thelowestfrostloads(highest
SHR)occuratfreezertemperatures,evenwhenrelativehumidityishigh.

FIGURE8

CapacityvsTimeforVariousFinSpacings


7/8StaggPattern,0.65SHR

100%

90%

80%

70%

60%

RBR(Stagg)8R2F

50%
RBR(Stagg)8R3F

40%
RBR(Stagg)8R4F

30%

20%

10%

0%

0
0.5
1
1.5
2
2.5
3
3.5
4

Time,h


FIGURE9

CapacityvsTimeforVariousFinSpacings

7/8StaggPattern,0.75SHR

100%


90%

80%

70%

60%

RBR(Stagg)8R2F
50%

RBR(Stagg)8R3F
40%

RBR(Stagg)8R4F

30%

20%

10%

0%

0
1
2
3
4
5
6
7

Time,h
Page19of52
ENG00019544REV02013ColmacCoilManufacturingInc.

P.O. Box 571 | Colville, WA 99114 | USA


T: +1.509.684.2595 | F: +1.509.684.8331
www.colmaccoil.com



FIGURE10

CapacityvsTimeforVariousFinSpacings
7/8StaggPattern,0.85SHR
100%
90%
80%
70%

CoolingCapacity,%






















60%
RBR(Stagg)8R2F

50%

RBR(Stagg)8R3F

40%

RBR(Stagg)8R4F
30%
20%
10%
0%
0

10

12

Time,h

CoolingCapacity,%

FIGURE11

CapacityvsTimeforVariousFinSpacings


7/8StaggPattern,0.95SHR

100%

90%

80%

70%

60%

RBR(Stagg)8R2F
50%

RBR(Stagg)8R3F

40%
RBR(Stagg)8R4F

30%

20%

10%

0%

0
5
10
15
20
25
30
35
40

Time,h


AnumberofobservationscanbemadewhenconsideringFigures811,Table1,and
Table3:


Page20of52
ENG00019544REV02013ColmacCoilManufacturingInc.

P.O. Box 571 | Colville, WA 99114 | USA


T: +1.509.684.2595 | F: +1.509.684.8331
www.colmaccoil.com


1. Therateatwhichfrostaccumulatesonanevaporatorincreasesastheroom
temperatureincreasesforagivenroomrelativehumidity.i.e.Frostonan
evaporatoroperatinginaroomat+32Fand85%rhwillaccumulatemuchfaster
thanonthesameevaporatoroperatingat10Fand85%rh.Thisisduetothehigher
watervaporpressureinairathighertemperaturesandtheresultinglowerSHR.
2. Foragivenreductioninevaporatorcapacity,widerfinspacingalwaysresultsin
longeractualruntimebetweendefrosts.
3. Asdesignruntimeisincreased,thenumberofdefrostsperdayrequiredincreases.
Inthecaseofveryhighfrostload(SHRlessthan0.75)usingdesignruntimegreater
than14h/daymayresultinaninabilityoftherefrigerationsystemtomaintainroom
temperature.
4. Figures8through11canbeusedtoestimatedefrostfrequencywhenroomSHR,
designruntime,andcoilfinspacingareknown.
Example:
Anevaporatorhasbeenselectedfora+10F/85%rhroombasedondesignruntime
of16h/day.Finspacingselectedis3FPI.Estimatethedefrostfrequencyusing
Tables1and3,andFigures8through11.
FromTable1:SHR=0.87
FromTable3:CoolingCapacityatTimeofDefrost=67%
FromFigure10:Timebetweendefrosts=5.2h
Therefore,estimateddefrostfrequency=24h/day/5.2h=5defrostsperday

Conclusion:Table4belowshowssuggestedmaximumfinspacing,designruntime,and
defrostfrequencyforvariousvaluesofSHR.Thistableisintendedtobeusedasageneral
guidelineinconjunctionwithTable1.Notethatroomrelativehumidity,andtherefore
SHR,willlikelychangethroughouttheyeardependingonlocationandclimate
conditions.Thisimpliesthatoptimumdefrostfrequencymaybedifferentinsummer
monthsvswintermonths.

TABLE4
SUGGESTEDFINSPACING,RUNTIME,ANDDEFROSTFREQUENCYVSSHR







SHR

SuggestedMaximum
FinSpacing,FPI
2
3
4
4

RecommendedMaximum
DesignRuntime,h/day
14
16
16
18

SuggestedDefrostFrequency,
No.Defrosts/day
9
8
6
2

0.65
0.75
0.85
0.95


Theabovediscussionandrecommendationsarebasedonfrostaccumulationby
depositionanddonotincludetheeffectsofairborneicecrystalsonfinspacingand
defrostfrequency.Theeffectsofairborneicecrystalsarediscussedinthefollowing
section.


2. AirBorneIceCrystals:

Page21of52
ENG00019544REV02013ColmacCoilManufacturingInc.

P.O. Box 571 | Colville, WA 99114 | USA


T: +1.509.684.2595 | F: +1.509.684.8331
www.colmaccoil.com


Thistypeoffrostisformedquitedifferentlyfromthefrostformedbydepositionas
explainedabove.Itaccumulatesonevaporatorsurfacesbyadifferentmechanism,and
ismoredifficulttoquantifyandpredict.

Airborneicecrystalsasatypeoffrostthatcanbedepositedoncoilsurfaceshasbeen
recognizedanddiscussedforsometime(Cleland2002,Stoecker1988).Theseice
particulatesformwheninfiltrationairmixeswithrefrigeratedairtoproducea
supersaturatedcondition.Onapsychrometricchart,asupersaturatedconditionis
indicatedwhenthemixedairconditionfallstotheleftofthesaturation(100%rh)line
(thinkoffogthathasfrozeninmidair).

Ratherthanaccumulaterelativelyuniformlyovertheentirecoilsurfaceasisthecase
withfrostformedbydeposition,airborneicecrystalsaccumulateontheleadingedges
ofthecoilfinsandhavetheprimaryeffectofrestrictingairflow.Thistypeoffrostis
moredifficulttopredictsinceitsformationdependsonnotonlytheconditionoftheair
outsidetherefrigeratedspace,butalsoontheconditionofdoorwaysandhowtheyare
operated.

Whenevaporatorsarelocateddirectlyabovedoorwayswhereairborneicecrystalsare
formedthistypeoffrostcanaccumulateveryquicklyandhaveseriousconsequencesin
termsofdegradedperformanceandinabilitytodefrosteffectivelyduetoexcessive
accumulationofhoarfrostandice.Inoneparticularcaseobservedbytheauthor,two
identicalevaporatorswereinstalledinthesamerefrigeratedspace(a10degFfreezer)
alongthesamewall,onedirectlyoverthedoorwayandthesecondoffsetbetween
doorways.Theevaporatordirectlyoverthedoorwayhadchronicproblemswithrapid,
heavyaccumulationoffrost,andwithdefrostissuesrelatedtoaccumulationoficeon
theunitcabinetandinthedrainpan.Theevaporatorlocatedonly20feetawaybetween
doorways,operatedwithoutaccumulatingiceonthecabinetandorinthepanand
defrostednormallyandeffectively.Itisthereforerecommendedthatevaporatorsnotbe
locateddirectlyabovedoorwayswheneverpossible.

Ifitisknownthattheevaporatorwillbeexposedtothistypeoffrost,variablefin
spacingisrecommended.Thatis,afinspacingarrangementwhichhasfinsonthefirst
onetotworowsontheairenteringsideofthecoilspacedwiderthanintheremaining
rows.Typicalarrangementsare1/2fpi(finsperinch),1.5/3fpi,and2/4fpi.

V.

CondenserSelectionandOperation

Anumberofdifferenttypesofcondensersareavailableforusewithammonia.

WaterCooled
AirCooled
Evaporative
Hybrid(Adiabatic)AirEvaporative


Page22of52
ENG00019544REV02013ColmacCoilManufacturingInc.

P.O. Box 571 | Colville, WA 99114 | USA


T: +1.509.684.2595 | F: +1.509.684.8331
www.colmaccoil.com

VI.

Incertaincasesthetypeofcompressionequipment(screwvsreciprocating)selectedand
theexpectedmaximumambienttemperaturewilldeterminewhetherornotaircooled
condensingwillbepossible.Inothercasestheavailability(orunavailability)ofwatermay
requiretheuseofaircooledcondensing.ThegoodnewsisthatDXammoniaiscompatible
withalltypesofcondensingsystems!

Properselectionandoperationofammoniacondensingequipmentisoutlinedinthe
condensermanufacturersliterature.

Itisrecommendedthatthesystemdesignercarefullyconsiderthefollowingpointswhen
selecting/designingcondensingequipment:
Energyefficiency
Partloadoperation
Lowambientoperation
Internalvolumeandammoniacharge
Gasinletandliquidoutletpiping
Purgingofnoncondensiblegases
VFDcondenserfancontrol(highlyrecommended)

Subcooling

Refrigerantliquidleavingthecondenseristypicallyatornearsaturationtemperatureand
pressure.Iftheliquidhasnotbeensubcooledbeforeitenterstheliquidline,anydropin
pressure,and/oranyheatinput,willcausetheliquidtoboilandflashgaswillbeformed.
Becauseoftheverylargevolumeoccupiedbyvaporcomparedtoliquid,theflashgas
increasestherefrigerantvelocityandcausesanexcessivepressuredropintheliquidline,
Thisreducesthecapacityandinterfereswiththeoperationoftheexpansionvalve,and
consequentlywillreducesystemcapacity.Adequatesubcoolingoftheliquidwillpreventthe
formationofflashgasinliquidlines.


Subcoolingtheliquidafteritleavesthereceiveristhereforeanecessityforpropersystem
operation.Notethatanysubcoolingdonewithinthecondenserorbetweenthecondenser
andthereceiverwillbeeliminatedinthereceiverduetotheequalizerline.Theamountof
subcoolingrequiredcorrespondstotheliquidlinepressuredropandheatgain.Thepressure
dropisthesumof1)thelossinpressureduetoelevationgainintheliquidline,2)liquidline
pressuredropduetofriction,and3)pressuredropthroughserviceandcontrolvalves.

Table5showsthepressuredropinliquidlinesproducedbyelevationgainbetweenthe
receiverandevaporatorswithammonia.







Page23of52
ENG00019544REV02013ColmacCoilManufacturingInc.

P.O. Box 571 | Colville, WA 99114 | USA


T: +1.509.684.2595 | F: +1.509.684.8331
www.colmaccoil.com














TABLE5

PressureDropinAmmoniaLiquidLinesDuetoElevationGain
ElevationGain
PressureDrop
ft
m
psi
kPa
1
0.3
0.3
1.9
5
1.5
1.4
9.3
10
3.0
2.7
18.7
15
4.6
4.1
28.0
20
6.1
5.4
37.3
25
7.6
6.8
46.7
30
9.1
8.1
56.0
35
10.7
9.5
65.4
40
12.2
10.8
74.7
45
13.7
12.2
84.0
50
15.2
13.5
93.4

Oncethetotalliquidlinepressuredrop(thesumofelevationpressuredropplusfrictional
pressuredroppluspressuredropthroughvalves)iscalculated,therequiredamountof
subcoolingtopreventflashgasinthelinecanbedeterminedfromTable6.Notethatthe
amountofsubcoolingrequiredforagivenpressuredropincreasesascondensing
temperaturedecreases.

TABLE6




















TotalLiquidLine
PressureDrop
psi
kPa
1
6.9
4
27.6
6
41.4
8
55.2
10
68.9
12
82.7
14
96.5
16
110.3
18
124.1
20
137.9
25
172.4
30
206.8
35
241.3
275.8
 40
 45
310.3
 50
344.7

RequiredAmountofSubcooling(Ammonia)
120F(49C)SCT
95F(35C)SCT
65F(18C)SCT
degF
degC
degF
degC
degF
degC
0.2
0.1
0.3
0.2
0.5
0.3
1.0
0.5
1.3
0.7
1.9
1.0
1.4
0.8
1.9
1.1
2.8
1.6
1.9
1.1
2.6
1.4
3.8
2.1
2.4
1.3
3.2
1.8
4.7
2.6
2.9
1.6
3.8
2.1
5.6
3.1
3.4
1.9
4.5
2.5
6.6
3.7
3.8
2.1
5.1
2.8
7.5
4.2
4.3
2.4
5.8
3.2
8.5
4.7
4.8
2.7
6.4
3.6
9.4
5.2
6.0
3.3
8.0
4.4
11.8
6.5
7.2
4.0
9.6
5.3
14.1
7.8
8.4
4.7
11.2
6.2
16.5
9.2
9.6
5.3
12.8
7.1
18.8
10.5
10.8
6.0
14.4
8.0
21.2
11.8
12.0
6.7
16.0
8.9
23.5
13.1

Page24of52
ENG00019544REV02013ColmacCoilManufacturingInc.

P.O. Box 571 | Colville, WA 99114 | USA


T: +1.509.684.2595 | F: +1.509.684.8331
www.colmaccoil.com




AcommonlyusedmethodofsubcoolingliquidrefrigerantistermedMechanicalSubcooling.This
istheCOLMACRECOMMENDEDmethodofliquidsubcoolingandreferstousingaportionof
liquidrefrigerantfromtheuncooledliquidlinetoevaporateandcooltheremainingliquid.A
heatexchanger(typicallyaplatetypeexchanger)isinstalledintheliquidlineinsuchawayasto
cooltheliquidrefrigerantononesideoftheexchangerbyevaporatingarelativelysmallamount
oftherefrigerantontheothersideoftheexchanger.Theevaporatingsiderefrigerantis
meteredbyaTXVormotorizedvalveinresponsetoliquidlinetemperatureandtheevaporated
refrigerantthenreturnedtothesuctionline.Thismethodofsubcoolingproducespredictable
resultsunderallconditions,andisrequiredtoinsureproperoperationofColmacDXAmmonia
evaporatorcontrols.Withmechanicalsubcoolingthereisnonetlossofrefrigeratingeffector
systemenergyefficiency.

Alternatemethodsforsubcoolingrefrigerantliquidcanbeapplied,buthavevariousdrawbacks:

AmbientSubcooling.Thisinvolvesusingaseparatecircuitwithinthecondensertoroute
liquidrefrigerantfromthereceivertothesystemcausingtherefrigeranttoapproachthe
ambientairtemperature.Thisisarelativelysimpledesign,howevertheamountof
subcoolingwillbelimitedtothecondenserTD.Thismaynotbeasufficientamountof
subcoolingtoavoidformationofflashgasduringcertaintimesoftheyear.Therefore,this
methodofsubcoolingisNOTrecommended.

LiquidPumping.Herealiquidpumpisinstalledattheexitofthereceivertopressurizethe
liquidlinesufficientlytoovercomethetotalpressuredropduetofrictionandelevationgain.
Whileeffectiveateliminatingflashgasregardlessofoperatingconditions,thismethodadds
complexityandwillcausetheliquidlinetooperateatapressurewhichishigherthan
condensingpressure.Aswithambientsubcooling,thismethodisNOTrecommended.

NOTE:ReferringtoFigures13,liquidtemperatureleavingthemechanicalsubcoolerisshownas
40degF.Thisliquidtemperatureisconservativeandshouldpreventtheformationofflashgas
inliquidlinesinmostifnotallcases.

SubcoolerPiping:

Figure12belowillustratestypicalmechanicalsubcoolerheatexchangerpiping.










Page25of52
ENG00019544REV02013ColmacCoilManufacturingInc.

P.O. Box 571 | Colville, WA 99114 | USA


T: +1.509.684.2595 | F: +1.509.684.8331
www.colmaccoil.com



FIGURE12

SUBCOOLER KIT

SUCTION

HOT LIQUID IN


N.C.


SUBCOOLED LIQUID OUT

TEMPERATURE

TRANSDUCER

T


TEMPERATURE
CONTROLLER


SOLENOID VALVE
GLOBE VALVE
PRV VALVE


ANGLE VALVE
STRAINER
T TEMPERATURE GAUGE


Toinsureeffectiveliquidsubcooling,besuretoobservethefollowingrules:
1. Sizepipingandvalvesforthemaximumrefrigerantflowconditionanticipated,i.e.
lowestheadpressure/highestsuctionpressure.Thisconditiontypicallyoccursduring
wintermonths.
2. Alwaysinsulateliquidlinestopreventheatgainandlossofsubcooling.
3. Locatesubcoolerheatexchangerdownstreamofthereceiverattheentrancetothe
liquidline,NOTbetweenthecondenserandreceiver.SeeP&IDexamplesabove.
4. Usegoodpipingpractice,ascanbefoundintheIIARAmmoniaRefrigerationPiping
Handbook(IIAR2004).

MechanicalSubcoolerSelection:

ColmacofferspreengineeredMechanicalSubcoolerswhicharefactorypipedandpackaged
inafreestandingframe,andincludethefollowingcomponents:
Stainlesssteelplatetypesubcoolingheatexchanger
Electronicexpansionvalveandtemperaturecontroller
Service(isolation)valves
Pressurereducingvalveforcontrolledleavingliquidlinepressure
ULlistedandwiredcontrolpanel

SeeseparateEngineeringBulletinforsubcoolerselectionandspecificationdetails.

VII.

PipingGeneral

1. Cleanliness.Thesmallinternalpassagesfoundinexpansionvalves(andothercontrol
valves)inDXammoniasystemsareparticularlysensitivetofoulingandpluggingwith

Page26of52
ENG00019544REV02013ColmacCoilManufacturingInc.

P.O. Box 571 | Colville, WA 99114 | USA


T: +1.509.684.2595 | F: +1.509.684.8331
www.colmaccoil.com

VIII.

relativelysmallamountsofdirtanddebris.Forthisreason,particularcareneedstobe
takenduringtheinstallationofsystempipingtoinsurecleanlinessandtominimizethe
introductionofweldscaleanddust,andothertypesofdirtanddebris.

2. EvacuationPriortoChargingtheSystemwithAmmonia.BecausetheperformanceofDX
ammoniaevaporatorsisdramaticallyaffectedbyevensmallamountsofwater,itisvery
importanttofollowgoodpressuretestingandevacuationprocedurespriortocharging
thesystemwithammonia.Recommendedevacuationprocedurecanbefoundinthe
separateColmacEngineeringBulletinonthistopic.

LiquidLines

Industryacceptedmethodsandpracticeforpropersizingandarrangementofliquidlines
canbefoundintheIIARAmmoniaRefrigerationPipingHandbook(IIAR2004).Additionally,
followtheguidelinesexplainedbelow.

1. Designmassflowrate.Liquidlinesmustbesizedappropriatelyforthetypeofline
(condensertoreceiver,receivertoexpansionvalve,etc.)andtheexpectedmaximum
massflowratecondition.Themaximummassflowrateconditionwilloccurwhen
dischargepressureisatitsminimum,suctionpressureisatitsmaximum,and
compressorsarerunningfullyloaded.Typicallythiswouldoccurwithfloatinghead
pressuresystemsduringwintermonths.Designingliquidlinesforthehottestdayofthe
year(commonlytakenasthedesignpoint)willlikelyleadtoundersizedliquidlines
andhigherthanexpectedpressuredropwiththepotentialofformingflashgasinthe
liquidline.
2. Insulation.Insulationofliquidlinesdownstreamoftheliquidsubcoolerbecomes
criticallyimportantinDXammoniasystemstoavoidheatgainandthepotentialfor
developingflashgasintheliquidlineupstreamoftheexpansionvalves.Usegood
qualityinsulationsystemswithadequateinsulationvalueandprotectionagainst
physicalandweatherdamage.
3. TypeofExpansionValve.ThreetypesofexpansionvalvesarecommonlyusedinDX
systems:1)Thermostatic,2)Motorized,and3)PulsewidthModulating.Thermostatic
andmotorizedvalvesmodulateinresponsetotheimposedloadonthecoilandsoliquid
linesshouldbesizedforthemaximumexpecteddesignmassflowrate(seeparagraph
V.1.above).Pulsewidthmodulating(PWM)expansionvalves,ontheotherhand,
alternatebetweenwideopenandfullyclosedataratewhichcorrespondstotheduty
calledforbytheelectroniccontroller.Becausethemassflowrateofrefrigerantwillbe
determinedbythewideopencapacityofthePWMvalve,thelocalliquidlinefromthe
liquidsupplymaintotheindividualevaporatormustbesizedtohandlethemaximum
capacityofthevalve.WhenPWMvalvesareused,theliquidsupplymainlinemustbe
sizedtohandlethiswideopencapacitybyusingadiversityfactorbasedonthe
numberofevaporatorsexpectedtobeoperatingatthesametimedividedbythetotal
numberofevaporators.
4. PressureRegulatingValve.AsshowninFigures13,theliquidlinepressureleavingthe
subcoolerassemblyismaintainedat75psigbyapressureregulatingvalve.When
defrosthotgaspressureisregulatedtomaintain90psig(alsoshowninFigures13)this

Page27of52
ENG00019544REV02013ColmacCoilManufacturingInc.

P.O. Box 571 | Colville, WA 99114 | USA


T: +1.509.684.2595 | F: +1.509.684.8331
www.colmaccoil.com


pressuredifferentialallowsdefrostcondensateleavingtheevaporatorsduringdefrost
tobefeddirectlybackintotheliquidlineandsenttootheroperatingevaporators.
5. Pipematerialspecifications.RefertotheIIARAmmoniaRefrigerationPipingHandbook
(IIAR2004)andANSI/IIARStandard22008(IIAR2008)fordetailedpipematerial
specificationrequirementsforammonialiquidpiping.

IX.

SuctionLines

X.

Unlikepumpedammoniasystems,nowetsuctionlinesareneededforDXammonia.
Althoughtheyshouldbepitchedandtrappedtoaccommodatetheoccasionalpresenceof
liquid,suctionlinepressuredropshouldbecalculatedtoreflectdryoperation.Refertothe
IIARAmmoniaRefrigerationPipingHandbook(IIAR2004)forpropersizingandarrangement
ofdrysuctionlines.Additionally,followtheguidelinesexplainedbelow.

1. Designmassflowrate.Aswithliquidlines,drysuctionlinesshouldbesizedforthe
expectedmaximummassflowratecondition.Again,themaximummassflowrate
conditionwilloccurwhendischargepressureisatitsminimum,suctionpressureisatits
maximum,andcompressorsarerunningfullyloaded.
2. Trappedverticalrisers.Suctionlineswithverticalupflow(suctionrisers)mustbe
installedwithaptrapatthebottom(entrance)oftheriseranddischargeintothetopof
theoverheadsuctionmainpipe.Whenvaryingloadsontheevaporatorareexpected,a
doubleriserdesignshouldbeused.RefertotheIIARAmmoniaRefrigerationPiping
Handbook(IIAR2004)forexamplesofdoublesuctionriserdesigns.
3. Pitchedsuctionlines.Suctionlinesmustbepitchedaminimumof1/8perfoottoward
thesuctionaccumulatortofacilitategooddrainageofanyliquidrefrigerantand/oroil
thatentersthesuctionline.
4. Pipematerialspecifications.Particularattentionmustbepaidtocarbonsteelpipe
materialspecificationsinlowtemperature(suctiontemperaturesbelow20degF),
whichmayrequireimpacttesting.RefertotheIIARAmmoniaRefrigerationPiping
Handbook(IIAR2004)andANSI/IIARStandard2(IIAR2008)fordetailedpipematerial
specificationsandrequirements.

HotGasLines

Industryacceptedmethodsandpracticeforpropersizingandarrangementofhotgaslines
canbefoundintheIIARAmmoniaRefrigerationPipingHandbook(IIAR2004).Additionally,
followtheguidelinesexplainedbelow.

1. Designmassflowrate.Hotgas(defrost)linesshouldbesizedforthemassflowrate
correspondingtothemaximumnumberandsizeofevaporatorsexpectedtodefrostat
thesametime.Conventionalwisdommaintainsthateachindividualevaporatorrequires
aflowofhotgasequalto2xtimestheflowrequiredduringcooling,andsothiswould
limitthenumberofevaporatorsbeingdefrostedatthesametimetoamaximumof1/3
thetotalnumberofevaporatorsinthefacility(thetwotoonerule).However,
evaporatorsequippedwithColmacSmartHotGascontrolscaneffectivelydefrostan
evaporatorwithhotgasflowingtotheevaporatorforonly8to10minutes.Withan

Page28of52
ENG00019544REV02013ColmacCoilManufacturingInc.

P.O. Box 571 | Colville, WA 99114 | USA


T: +1.509.684.2595 | F: +1.509.684.8331
www.colmaccoil.com

2.

3.

4.

5.

6.

effectivebuildingmanagementcontrolsystem,anddependingonthefrostloadand
frequencyofdefrosting,itispossibletolimittheamountofdefrosthotgasflowingat
anygiventimetoonlythatrequiredforthelargestsingleevaporatorinthefacility.This
approachobviouslyhasthepotentialtoreducethehotgaslineandPRVsizeandcost.
Expectedmassflowrateofhotgasfordefrostofagivensizedevaporatorcanbe
calculatedusingthemethoddescribedbelowintheHotGasDefrostsection.
Insulation.Insulationofhotgaslinesiscriticallyimportanttoinsurefastdefrosting.Use
goodqualityinsulationsystemswithadequateinsulationvalueandprotectionagainst
physicalandweatherdamage.
PressureRegulatingValve.AsshowninFigures13,thehotgaslinecomingfromthe
compressordischargelineismaintainedat90psigbyapressureregulatingvalve.When
defrosthotgaspressureisregulatedtomaintain90psigandtheliquidlineis
maintainedat75psig(alsoshowninFigures13)thispressuredifferentialallowsdefrost
condensateleavingtheevaporatorsduringdefrosttobefeddirectlybackintotheliquid
lineandsenttootheroperatingevaporators.Maintainingthehotgaslinepressureat
thereduced90psigalsominimizesheatlosstothesurroundingambient.
Pitchedhotgaslinesanddriplegs.Hotgaslinesmustbepitchedaminimumof1/8per
foottowardtheevaporatorstofacilitategooddrainageofanycondensedrefrigerant
(condensate)todriplegsinstalledaheadoftheevaporatorcontrolvalvegroup(s).
Liquiddrainers.Ashotgasfordefrosttravelsfromtheengineroomtotheevaporators
someofitsenergywillbereleasedtoheatupthepipingitself,andsomereleaseddue
toheatlossthroughinsulation.Condensatewillthereforeforminthehotgaspiping
whichmustthenbeeffectivelytrappedanddrainedbeforeitreachestheevaporators.
Unlessitiseffectivelyremoved,accumulatingcondensedliquidupstreamofhotgas
solenoidvalveswillcausecavitationontheseatsofthesolenoidvalveswhenthevalve
isclosed(Jensen2013).Condensatewillcollectindriplegs(describedabove)andmust
bereturnedtoeitheranearbysuctionline,oracondensatereturnline.Usealiquid
draineroranappropriatelysizedsteamtraptoallowonlyliquidtoleavethedripleg.
Usingliquiddrainersalsoeffectivelykeepshotgaslinescontinuallyheatedandreadyto
supplyfullflowofhotgastoevaporatorsimmediatelyondemandfordefrosting.
Pipematerialspecifications.RefertotheIIARAmmoniaRefrigerationPipingHandbook
(IIAR2004)andANSI/IIARStandard22008(IIAR2008)fordetailedpipematerial
specificationrequirementsforammoniahotgaspiping.


XI.

EffectsofWaterinAmmoniaandItsRemoval

Asexplainedindetailelsewhere(Nelson2010),thepresenceofevensmallamountsof
waterinammoniahasasignificantnegativeeffectonDXevaporatorperformance.
Unfortunately,waterisdifficulttoentirelykeepoutofindustrialammoniarefrigeration
systemsforanumberofreasons:Residualwaterinpressurevesselsleftfromhydrotesting,
incompleteevacuationofthesystempriortostartup,leaksinpartsofthesystemwhich
normallyoperateinavacuum,etc.

Thisresidualwatergoesintosolutionwiththeammoniaandincreasesandtheboilingpoint
(bubblepoint)temperature.Ataconcentrationof20%(bymass)waterinammonia,the

Page29of52
ENG00019544REV02013ColmacCoilManufacturingInc.

P.O. Box 571 | Colville, WA 99114 | USA


T: +1.509.684.2595 | F: +1.509.684.8331
www.colmaccoil.com

IncreaseinBubblePointTemperature,F

boilingpointrisestoapproximately10degFabovetheboilingpointofpureammoniaatthe
samepressure.SeeFigure13below.

Astheammoniawaterliquidenterstheevaporatorcircuititbeginstoboil.Becauseofthe
largedifferenceinvaporpressuresofammoniaandwater,onlyammoniavaporis
generatedduringtheevaporationprocess,leavingthewaterbehindintheremainingliquid.
Sotheevaporationprocessresultsinanincreaseinwaterconcentrationanda
correspondingincreaseintheboilingpointoftherefrigerantasitpassesthroughthecoil
circuit.Inthecaseofanevaporatoroperatingwitha10degFTD,therefrigerantwillstop
boilingoncethewaterconcentrationreachesabout20%sincetheboilingpointwillhave
risenby10degF.Thiscessationofboilingwilloccuratsomepointalongthelengthofthe
evaporatorcircuit,thepointatwhichboilingstopsdependingontheinitialconcentrationof
waterandsuctionpressure.Atthepointwheretheincreaseinthewaterconcentrationhas
causedanincreaseintheboilingpointequaltothecoilTD,liquidrefrigerantwillexitthe
evaporatorandenterthesuctionline.

Figure13belowshowstheincreaseinboilingpoint(bubblepoint)forvariousinitialwater
concentrationinammoniaatvariouspressures.

FIGURE13

IncreaseinBubblePointTemperaturevsComposition

AmmoniaWater

40


35

48.2 psia(3.3bar)
30.4psia(2.1bar)

18.3psia(1.3bar)
30
10.4psia(0.7bar)


25


20

15


10


5


0

0
0.1
0.2
0.3
0.4
0.5
Composition(MassFractionWater)



Anexampleoftheincreaseinbubblepointtemperatureoverthecircuitlengthofan
evaporator,representedbythechangeinvaporquality,isshowninFigure14belowforan
initialwaterconcentrationinammoniaof3%atapressureof10.4psia(40degFSST).In

Page30of52
ENG00019544REV02013ColmacCoilManufacturingInc.

P.O. Box 571 | Colville, WA 99114 | USA


T: +1.509.684.2595 | F: +1.509.684.8331
www.colmaccoil.com

Increase in Tbub, F

thisexamplethebubblepoint(Tbub)hasincreasedby10degFatavaporqualityofapprox.
0.89.
FIGURE14

Incr in Bubble Point vs NH3 Vap Quality, 3.00% Water in NH3,

Tsat = -40.0F

50


45

40


35

30

25


20

15


10

5


0
0
0.2
0.4
0.6
0.8
1

NH3 Vapor Mass Fraction (Quality)


Thisincreaseinbubblepointsignificantlyreducesthemeantemperaturedifferenceand
thereforethecoolingcapacityoftheevaporatorisreducedasillustratedinFigure15.


FIGURE15

















Page31of52
ENG00019544REV02013ColmacCoilManufacturingInc.

P.O. Box 571 | Colville, WA 99114 | USA


T: +1.509.684.2595 | F: +1.509.684.8331
www.colmaccoil.com



Inadditiontotheperformancepenaltyseenwhenrelativelysmallamountsofwaterare
presentintheammonia,thisalsomeansthatthemassfraction(10.89)=0.11,or11%of
themassofrefrigerantexitingtheevaporatorasliquidwillhavetobecaptureddownstream
inthesuctionaccumulator.

Knowingthatammoniawaterliquidofapproximately20%waterconcentrationwill
unavoidablyleavetheevaporatorswheneverevensmallamountsofwaterarepresentin
theammoniaisimportantforthedesignertounderstand.Thesuctionaccumulatormust
thereforebeproperlydesignedtoperformthefollowingfunctions:

i. Separateliquidandvaporrefrigerantandallowonlyvaportoreturntothe
compressor,
ii. Captureanddistill(byheating)ammoniawaterliquidtoaconcentrationthat
cansafelyberemovedfromthesystemfordisposal.
iii. Transferexcesstrappedliquidtothehighpressurereceiver,orintothereduced
pressureliquidline.

1. Separation

Liquidvaporseparationinsuctionaccumulatorvesselsiswellunderstoodanddesign
methodswelldocumented.Refertorecognizedpublishedsizinganddesignmethods
(Stoecker1988,Wiencke2002).

Colmacoffersarangeofpreengineeredfactoryassembledsuctionaccumulator
packagesspecificallysuitedtooperationwithDXammonia.SeeseparateEngineering
Bulletinforselectionandspecificationdetails.

2. DistillationandDisposalofAmmoniaWaterSolution(AmmoniumHydroxide)

Distillation:

AmmoniaishighlysolubleinwaterduetothepolarityofNH3moleculesandtheir
abilitytoformverystronghydrogenbonds(Nelson2010).Thishighsolubilitymakes
ammoniawateragoodworkingfluidpairinabsorptionrefrigerationmachines,taking
advantageofthelargevaporpressuredifferencesbetweentheammoniavaporand
weaksolution.However,thissamebehaviormakeswaterremovalfromammonia
refrigerationsystemssomewhatchallenging.

Asmentionedabove,ammoniawatersolutionconcentratedtoapproximately20%
waterwillreturnfromevaporatorsviathesuctionlinetobetrappedinthesuction
accumulator.Thisaqueousammoniasolution,calledAmmoniumHydroxide,ata
concentrationof80%ammonia(20%water)wouldbeverydifficulttosafelyremove
fromthesystemfordisposal.Furtherdistillationofthesolutionisneededtobringthe
ammoniaconcentrationinthesolutiondowntothepracticalminimumbeforeitis
removed.
Page32of52
ENG00019544REV02013ColmacCoilManufacturingInc.

P.O. Box 571 | Colville, WA 99114 | USA


T: +1.509.684.2595 | F: +1.509.684.8331
www.colmaccoil.com



TheonlypracticalwaytodistilltheAmmoniumHydroxideisbyheatinginaseparate
distillationvessel,calledastill.AmmoniumHydroxidetrappedinthesuction
accumulatordrainsbygravityintothestillwhereitisheatedtoatemperature
correspondingtothepointonaPhaseEquilibriumdiagramwheretheslopeofthedew
pointlinechangesrapidlyfromnearlyverticaltomorenearlyflat.Thispointisshownon
Figure16asPointA.Belowthistemperature(between100and120degF),nearlypure
ammoniavaporwillleavethestillandtravelthroughtheventlinebacktothesuction
accumulatorwhereitwillthenbetakenbacktothecompressor.Abovethis
temperature,watervaporwillbegintoleavetheAmmoniumHydroxidesolutionand
exitthestillventlinewhereitwillgobackintosolutionwithanyammonialiquid
presentinthesuctionaccumulator.Basedonthis,theheatingelementinthestillmust
becontrolledtobringthesolutiontemperatureuptoamaximumof100to120degF,
atwhichpointitisreadytoberemovedsafelytoastoragecontainerforfurther
processingand/ordisposal.

FIGURE16
























Figure16alsoshowsthatthemaximumwaterconcentrationintheAmmonium
Hydroxidesolutionheatedto100degFvarieswithsuctionpressure.Themaximum
waterconcentrationspossibleareshownasPointsB,C,D,andE,inFigure16,andare
listedinTable7below.Itisapparentfromthefigureandthetablethatthestillisableto
achievehigherwaterconcentrationsatlowersuctionpressures.
Page33of52
ENG00019544REV02013ColmacCoilManufacturingInc.

P.O. Box 571 | Colville, WA 99114 | USA


T: +1.509.684.2595 | F: +1.509.684.8331
www.colmaccoil.com



TABLE7
AMMONIAWATERCONCENTRATIONS@100degFvsSUCTIONPRESSURE
SuctionPressure,psia
WaterConcentration, AmmoniaConcentration,
(SaturationTemp,F)
%bymass
%bymass
48.2(+20F)
57.5
42.5
30.4(0F)
65.0
35.0
18.3(20F)
72.5
27.5
10.4(40F)
80.0
20.0

Usingtheaboveinformation,anestimatecannowbemadeofthevolumeof
AmmoniumHydroxidesolutionthatwillbegeneratedbythestillforagivensystem
ammoniachargehavingagiveninitialwatercontent.Table8belowshowsthe
expectedvolumeofAmmoniumHydroxidesolutionperpoundofinitialammonia
chargethatwillhavetoberemovedbythestill(anddisposedof)inorderto
completelyremovethewaterfromrefrigerationsystem.

TABLE8
EXPECTEDVOLUMEOFDISTILLEDSOLUTIONAMMONIUMHYDROXIDE@100F
(GALLONSperPOUNDINITIALAMMONIACHARGE)

 InitialWater
SaturatedSuctionPressure,psia(SaturatedTemp,F)

Content,%
48.2(+20F)
30.4(0F)
18.3(20F)
10.4(40F)

0
0.00000
0.00000
0.00000
0.00000

1
0.00246
0.00211
0.00183
0.00161

3
0.00738
0.00633
0.00550
0.00484

5
0.01230
0.01055
0.00917
0.00807

10
0.02460
0.02109
0.01835
0.01614

20
0.04920
0.04218
0.03669
0.03229 


EXAMPLE:
Asystemhasaninitialammoniachargeof5,000lbswith3%watercontent.Thestill
isinstalledonthe20degFsuctionaccumulator.Whatwillbethetotalvolumeof
distilledAmmoniumHydroxidesolutiondrainedfromthestill?

Answer:
FinalDistilledSolutionAmmoniaConcentrationforDisposal(fromTable7):27.5%
ExpectedVolumeofDistilledSolutionperPound(fromTable4):0.0055gal/lbs

TotalExpectedVolumeof27.5%AmmoniumHydroxideforDisposal:
5,000lbsx0.0055gal/lbs=27.5gallons

StorageandDisposal:

Page34of52
ENG00019544REV02013ColmacCoilManufacturingInc.

P.O. Box 571 | Colville, WA 99114 | USA


T: +1.509.684.2595 | F: +1.509.684.8331
www.colmaccoil.com


AmmoniumHydroxideislistedasahazardoussubstanceunderCWA(40CFR
1164.40CFR117.3ReportableQuantityCategoryC.1000lbs/454kg).Assuch,itis
importanttocomplywithalllocalandnationalregulationsforsafehandlingand
disposalofthesolutionremovedfromthesystemstill(Smith2010).

ItisinterestingtonotethatsuitablydilutedAmmoniumHydroxidemaybedisposed
ofonagriculturallandasfertilizer.However,thematerialshouldbekeptfrom
enteringstreamsandlakesasitisharmfultoaquaticlifeandcancause
environmentaldamage.

AmmoniumHydroxidewillreactexothermically(heatisgenerated)withacids,and
soneutralizingofthesolutionbyunqualifiedpersonnelisnotrecommended.

ItisimportanttopreventcontactoftheAmmoniumHydroxidesolutionwith
chemicalssuchasmercury,chlorine,iodine,bromine,silveroxide,and
hypochlorites,astheycanformexplosivecompounds.Contactwithchlorineforms
chloraminegaswhichisaprimaryskinirritantandsensitizer.

Figure16incombinationwithTable7canbeusedtopredictthetemperatureabove
whichammoniavaporwillbegeneratedwhenthedistilledsolutionisstoredinan
opencontainer.Thisvaporneutraltemperatureisfoundusingtheammonia
concentrationsshowninTable7forvarioussuctionpressures,intersectingalineof
constantbubblepointtemperature(Tbub)atatmosphericpressure(14.7psia)on
Figure16.Table9belowshowstheambient(storage)temperaturesbelowwhich
ammoniavaporwillnotbegeneratedfromtheAmmoniumHydroxidesolution
dischargedfromthestill.

TABLE9
RECOMMENDEDMAXIMUMAMMONIUMHYDROXIDESTORAGETEMPERATURES
SuctionPressure,psia AmmoniaConcentration, StorageTemperature,
(SuctionTemp,F)
%bymass
DegF
48.2(+20F)
42.5
42
30.4(0F)
35.0
64
18.3(20F)
27.5
90
10.4(40F)
20.0
108

AmmoniumHydroxidesolutionhasacorrosivereactionwiththefollowingmaterials
whichshouldnotbeusedtostorethedistilledAmmoniumHydroxidesolution
(LaRocheIndustries1987):
x Galvanized(zinccoated)surfaces
x Copper
x Brassandbronzealloys
x Certaintypesofelastomers

ThedistilledAmmoniumHydroxidesolutioncanbesafelystoredincontainersmade
ofthefollowingmaterials:
Page35of52
ENG00019544REV02013ColmacCoilManufacturingInc.

P.O. Box 571 | Colville, WA 99114 | USA


T: +1.509.684.2595 | F: +1.509.684.8331
www.colmaccoil.com


x
x
x
x

Carbonsteel
Stainlesssteel
Aluminum
CastIron


Generallyspeaking,aluminumalloysarenotrecommendedforexposuretoaqueous
solutionshavingapHgreaterthan9.0duetoacceleratedcorrosionandmetalloss.
AmmoniumHydroxidehowever,eveninhighconcentrations,isanexceptiontothis
rule.(Davis1999).

Inconclusion,AmmoniumHydroxidesolutioncollectedfromthestillshouldbestored
inanappropriatelyconstructedcontainerlocatedinacoolspaceoutofdirect
sunlight.Itisrecommendedthatthedistilledsolutionbedisposedofusingalocal
qualifiedwastedisposalvendor.

MoredetailedhandlingandsafetyinformationcanbefoundonMSDSsheets
publishedbysuppliersofAmmoniumHydroxide(TannerIndustries2000,LaRoche
Industries1998).

3. LiquidTransfer

Theanticipatedvolumeofammoniawaterliquidleavingtheevaporator(s)basedon
anaverage20%waterconcentrationattheevaporatorexithasbeencalculatedand
showninTable10below.Multiplythevalueshowninthetablebythetotalcapacity
ofthesystemintons(TR)todeterminethevolumeofammoniawaterliquid
returningtothesuctionaccumulator.

TABLE10
VolumeofAmmoniaWaterLiquidLeavingDXEvaporators

InitialWaterContentin
VolumetricFlowrateofAmmoniaWater
Ammonia,%
(20%waterconcentration)LeavingDX
Evaporators,ft3/h/TR
0.5
0.01
1.0
0.02
3.0
0.07
5.0
0.12
10.0
0.24

Example:
Ithasbeendeterminedthattheammoniachargeina500ton(TR)ammoniasystem
hasawatercontentof3%.Iftheevaporatorsareoperatedasdirectexpansion(DX),
howmuchammoniawaterliquidisexpectedtoreturnfromtheevaporatorstothe
suctionaccumulator?
Answer:
500TRx0.07ft3/h/TR=35ft3/h=4.4gal/min
Page36of52
ENG00019544REV02013ColmacCoilManufacturingInc.

P.O. Box 571 | Colville, WA 99114 | USA


T: +1.509.684.2595 | F: +1.509.684.8331
www.colmaccoil.com

InsideFoulingFactor,ft2Rh/Btu

XII.


Thisvolumeofliquidreturningtothesuctionaccumulatorduetowaterinthe
ammonia(atleastinitially)mustbeaddedtothevolumeofliquidrefrigerant
floodbacknormallyconsideredwhendesigningthesuctionaccumulatortransfer
vessel.

AliquidtransfervesselisincorporatedintotheColmacPreEngineeredSuction
AccumulatorSkidpackage(seeseparateEngineeringBulletin).Thetransfervesselis
designedtohandleupto5%initialwatercontentinthesystemammoniacharge.
NotethatIndustrialRefrigerationgradeanhydrousammoniaiscertifiedtobe
99.95%pureammonia.

EffectsofOilonEvaporatorPerformanceandOilSeparation

ImmisciblelubricantsarerecommendedovermisciblelubricantsforlargeindustrialDX
ammoniarefrigerationsystemsforanumberofreasons:
o Lowercost
o Easeofseparation
o Relativeinsensitivitytocontaminants(water,dirt)

Therefore,immiscibleoilispreferredovermiscible,however,DXevaporator
performanceisseverelypenalizedifimmiscibleoilisallowedtoreachevaporatorsand
coatinternaltubesurfaces.Evenathinlayerofoildepositedonevaporatortubeswill
resultinarelativelylargefoulingfactorasshowninFigure17below.

FIGURE17

InsideFoulingFactorvsOilFilmThickness

0.006


0.005


0.004


0.003


0.002


0.001


0

0
0.001
0.002
0.003
0.004
0.005

OilFilmThickness,inches


InatypicalDXammoniaevaporator,thisfoulingfactorcausesasignificantreductionin
coolingcapacityasisshownbelowinFigure18.

Page37of52
ENG00019544REV02013ColmacCoilManufacturingInc.

P.O. Box 571 | Colville, WA 99114 | USA


T: +1.509.684.2595 | F: +1.509.684.8331
www.colmaccoil.com



FIGURE18

CapacityReductionvsOilFilmThickness
AmmoniaEvaporator,15FSST,5FAirOn
1

0.95

CapacityReductionRatio




















0.9

0.85

0.8

0.75
0

0.0005

0.001

0.0015

0.002

0.0025

0.003

0.0035

0.004

OilFilmThickness,inches



ItisapparentfromFigures17and18abovethatitishighlydesirablefromanenergy
efficiencystandpointtopreventcompressorlubricatingoilfromreachingtheevaporators.
Toachievethis,thefollowingshouldbecarefullyconsideredandspecifiedinthesystem
design:

a. Typeofcompressorlubricatingoil
b. Compressoroilseparatordesignandefficiency
c. Oilcaptureinthesystemreceiverandaccumulator(s)
d. Oilcaptureattheevaporator

TypeofOil:

Dependingonthetypeofcompressorused(reciprocatingorrotaryscrew),varyingamounts
oflubricatingoilwillunavoidablybedischargedwiththeammoniavapor.Oilwillleavethe
compressorbothinliquiddropletformandasoilvapor.Theliquiddropletscanbecaptured
mechanicallyintheoilseparatorvesselbycontrollingvelocityandbyincorporating
coalescingelements.Theoilwhichiscombinedwiththeammoniainvaporformismore
difficulttocapture.Generallyspeaking,asvolatilityandsolubilityoftheoilincrease,
separationbecomesmoredifficult.Theamountofoilwhichisnotcapturedintheseparator
andreturnedtothecompressorisreferredtoasoilcarryover.


Page38of52
ENG00019544REV02013ColmacCoilManufacturingInc.

P.O. Box 571 | Colville, WA 99114 | USA


T: +1.509.684.2595 | F: +1.509.684.8331
www.colmaccoil.com


Alllubricatingoilsusedintheammoniarefrigerationindustryareblendsofabasefluid(s)
withadditives(Wierbosch2010).Thebasefluidcontrolsvolatilityandsolubiltyoftheoil.
Today,commonlyusedbasefluidsinclude:
Napthenic
SolventRefinedParaffinic
AlkylBenzene
2StageHydrocracked
PAO/AB

Thearomaticcontentofthebasefluidhasalargeeffectonvaporpressure(volatility)and
solubility.Thehigherthevaporpressureoftheoil,themoreoilvaporwillleavethe
compressorwiththeammoniainthedischargegas(Briley1984).Sincethisoilvaporis
difficulttocaptureintheseparator,itisdesirabletoselectanoilwhichhasthelowestvapor
pressurepossible.AlkylBenzeneandNaphthenicbaseshavethehighestaromaticcontent,
vaporpressure,andsolubility.2StageHydrocrackedbaseshavelowestaromaticcontent,
vaporpressure,andthelowestsolubility.

Itisthereforerecommendedthatammoniarefrigerationoilhavinga2StageHydrocracked
basefluidbeusedintheDXammoniacompressionsystemdesign.2StageHydrocracked
mineraloilmanufacturedbyCPI(CPI100868)isrecommendedforapplicationin
reciprocatingandscrewtypeammoniacompressorsfortemperaturesabove40.

Anotherfactoraffectingoilvaporpressureistheoiltemperature.Thehighertheoil
temperature,thehigherthevaporpressure.Reducingthedischargegas(andoilvapor)
temperaturebeforeitenterstheseparatorwillthereforereduceoveralloilcarryoverand
increasetheefficiencyoftheseparator.Forexample,desuperheatingthedischargegas
from80degCto35degCreducestheoilvaporpressure,andthereforecarryoverofoil
vapor,byapproximately85%(Wiencke2012).

OilSeparator;

ForsuccessfulDXammoniasystemoperation,anoilseparatorwithcoalescingelements
capableofguaranteeing57ppmcarryovershouldbespecifiedandinstalled.

Forreasonsmentionedabove,itisalsorecommendedthatthedischargegasbe
desuperheatedasmuchaspracticalpriortoenteringtheoilseparator.

NOTE:MeshPadoilseparatorsasfoundonolderscrewcompressorpackagesand
reciprocatingcompressorswillnothavetherequiredseparationefficiencyandarenot
recommended!

OilCaptureattheEvaporator:

Evenwhen2StageHydrocrackedoilisusedwithahighlyefficientoilseparator,itispossible
forsomeverysmallamountofoiltoreachtheentrancetotheevaporators.

Page39of52
ENG00019544REV02013ColmacCoilManufacturingInc.

P.O. Box 571 | Colville, WA 99114 | USA


T: +1.509.684.2595 | F: +1.509.684.8331
www.colmaccoil.com


Toseparateandcapturethisoilbeforeitreachestheevaporator,Colmachasdesigneda
proprietaryDXammoniadistributor.TheColmacTankDistributor(patentpending)
incorporatesadroplegintothebodyofthedistributortankwhichservestocollectoiland
debriswhereitcanbeperiodicallydrainedandremovedfromthesystemattheevaporator.

Figure19belowshowsacrosssectionoftheColmacTankDistributorwithitsintegraldrop
legfeatureforcapturingandremovingoil.

FIGURE19
COLMACTANKDISTRIBUTOR(CROSSSECTION)




















OilCapturefromtheSystem:

Colmacpreengineeredsuctionaccumulatorandintercoolervesselskidpackagesincludean
integraloilpotforcollectionandperiodicremovalofoil.

Itisalsorecommendedthathighpressurereceiversbedesignedwithanoilsumpanddip
tubeforcollectionandperiodicremovalofoil.

XIII.
EstimatingDXEvaporatorRefrigerantChargeInventory

Inordertoproperlysizethevolumeofthesystemvessels(highpressurereceiverand
lowpressureaccumulator),anestimateoftherefrigerantchargeheldinthe
evaporatorsmustbemade.Designersnormallycalculatetheevaporatorchargeasa
percentageoftheevaporatorinternalvolumetimestheliquiddensityofammonia.


Page40of52
ENG00019544REV02013ColmacCoilManufacturingInc.

P.O. Box 571 | Colville, WA 99114 | USA


T: +1.509.684.2595 | F: +1.509.684.8331
www.colmaccoil.com

XIV.

OneofthemainadvantagesofDXoperationisthesignificantlyreducedevaporator
ammoniachargecomparedtopumpedammonia.Manysystemdesignersestimate
evaporatorchargeforbottomfeedpumpedammoniaevaporatorstobeasmuchas
80%oftheinternalvolumetimestheliquiddensitytoaccountforlowloadandidle
conditions.DXammoniaevaporatorchargecanbeestimatedusingtwophasevoid
fractionequations.Table11belowshowsDXammoniaevaporatorchargeasa
percentageofinternalvolume.

TABLE11
COLMACDXAMMONIACHARGEINVENTORY
SuctionPressure,psia
ColmacDXAmmonia
(SaturationTemp,F)
EvaporatorChargeInventory,
lbs/ft3ofInternalVolume
48.2(+20F)
1.01
30.4(0F)
0.83
18.3(20F)
0.63
10.4(40F)
0.52

Thissignificantlyreducedsystemchargenotonlyreducestherequiredsizeofthe
receiver(and/orothersystemvessels),italsogreatlyreducespumpouttimeforthe
evaporatorspriortodefrosting.Thisservestoshortentotaldefrosttimeandincrease
theeffectivenessofhotgasdefrosting,reducingenergyconsumptionandoperating
costs.

EXAMPLE:
AColmacDXammoniaevaporatoroperatingatasuctiontemperatureof20degFhas
aninternalvolumeof12ft3.WhatistheexpectedDXammoniaoperatingcharge?What
wouldtheoperatingchargebeforpumpedammoniaoperationusingthe80%rule?

Answer:
ColmacDXammoniacharge=12ft3x0.63=7.6lbs
Pumpedammoniacharge=12ft3x0.8x42.2lbs/ft3=405lbs

ColmacSmartHotGasDefrost


Theenergyefficiencyofhotgasdefrostingevaporatorsdependsonthefollowing
(Nelson2011(1)):

1. Minimizingconvectiveheatloss.
Uselowestpracticaldefrostregulatorsetting.75to90psig(50to60F)
shouldbeadequate.Note:Ifhigherpressuresareneeded,lookfor
problemselsewhere.
2. Shortendefrostduration.
UsetopfeedorDX(directexpansion)evaporatorfeedtoreducetime
requiredforpumpout.
Openthehotgassolenoidonlylongenoughtoclearcoil(68minutes).
Page41of52
ENG00019544REV02013ColmacCoilManufacturingInc.

P.O. Box 571 | Colville, WA 99114 | USA


T: +1.509.684.2595 | F: +1.509.684.8331
www.colmaccoil.com

Installaseparatehotgassolenoidanddefrostregulatorforpreand
postheatingofthepanloop.Alternately,installelectricresistance
drainpanheating.
3. Reducethenumberofdefrostsperday.
Reducethenumberofdefrostsperdaytomatchthefrostload.
Chooseevaporatorswithwidefinspacing(3fpiinsteadof4fpi)to
maximizefrostcarryingcapacity.
Mitigateinfiltrationofhumidairintotherefrigeratedspaceby:
x Properdesignandoperationofdoorways,and
x Keeploadingdocksatthelowestpracticaldewpoint
temperature.

CalculatingtheCostofDefrost

Asmentionedabovedefrostefficiencycanbesignificantlyimprovedbyreducingthe
amountofenergylosttotheroombyconvectionduringdefrost.Theoperatingcost
savingsduetoareductionindefrostdurationhasbeencalculatedandpresentedbelow
basedon:

1. Reducingdefrostdurationfrom30minutesto10minutes,and
2. Increasingfrostthicknessfrom1mmto2mm(reducingthenumberofdefrosts
perdaybyhalf).

Thecalculationsassume:

Evaporatorcapacity:100TR
Compressorruntime:16h/day
CostofElectricity:$0.10/kWh

Table12showscalculatedcostsavingsforfourdifferentroomtemperatures.














Page42of52
ENG00019544REV02013ColmacCoilManufacturingInc.

P.O. Box 571 | Colville, WA 99114 | USA


T: +1.509.684.2595 | F: +1.509.684.8331
www.colmaccoil.com


TABLE12
CALCULATEDCOSTSAVINGS($/y/100TR)FOROPTIMIZEDVSCONVENTIONALDEFROST

















SHR
System COP:
Frost Removed, kg/day:
Frost Removed, kg/y:
I. Baseline (30 min, 1 mm)
Defrost Efficiency, %
Defrost Convective Losses, %:
Defrost Convective Losses, kWh/y:
Baseline Cost of Defrost (Convective), $/y:
II. Optimized (10 min, 2 mm)
Defrost Efficiency, %
Defrost Convective Losses, %:
Defrost Convective Losses, kWh/y:
Optimized Cost of Defrost (Convective), $/y:

SmartHotGasDefrostPiping

Savings
 Optimized vs Baseline, $/y:

Room Temp, C (F)


0 (+32)
-18 (0)
-23 (-10)
0.66
0.89
0.93
3.2
2.5
2.2
2,778
899
572
1,014,096 328,090 208,784

-34 (-30)
0.97
2
245
89,479

32%
18%
46%
61%
1,012,438 753,334
$31,639 $30,133

17%
63%
545,922
$24,815

14%
65%
283,071
$14,154

61%
15%
168,740
$5,273

46%
26%
125,556
$5,022

43%
27%
90,987
$4,136

40%
30%
47,178
$2,359

$26,366

$25,111

$20,679

$11,795



Conventionalammoniaevaporatorsaretypicallyarrangedforbottomfeedwiththehot
gaspanlooppipedinserieswiththecoil.TheColmacSmartHotGasDefrostsystem
(Nelson2011(2))resultsinthehighestpossibledefrostefficiencyandlowestoperating
costbyutilizingtopfeedDXcircuitingwiththehotgaspanlooppipedseparatelyfrom
thecoil.Thisresultsin:

Pumpoutperiodshortenedto5minutes
Defrostduration(timecoilhotgassolenoidisopen)ofonly68minutes

ColmacSmartHotGasDefrostcontrolvalvegroupsareshowninFigures20and21
below.















Page43of52
ENG00019544REV02013ColmacCoilManufacturingInc.

P.O. Box 571 | Colville, WA 99114 | USA


T: +1.509.684.2595 | F: +1.509.684.8331
www.colmaccoil.com



FIGURE20
COLMACDXSMARTHOTGASDEFROSTCONTROLVALVEGROUP
HIGHPRESSURELIQUID(HPL)FEED











































DC

GLOBE VALVE

ANGLE VALVE

SOLENOID VALVE

ELECTRONIC EXPANSION VALVE

STRAINER

HAND EXPANSION VALVE

SUCTION STOP VALVE

CHECK VALVE

TANK DISTRIBUTOR

AUTO VENT

SIGHT GLASS

Page44of52
ENG00019544REV02013ColmacCoilManufacturingInc.

P.O. Box 571 | Colville, WA 99114 | USA


T: +1.509.684.2595 | F: +1.509.684.8331
www.colmaccoil.com



FIGURE21
COLMACDXSMARTHOTGASDEFROSTCONTROLVALVEGROUP
INTERMEDIATEPRESSURELIQUID(FROMINTERCOOLER)FEED
































GLOBE VALVE


SOLENOID VALVE


STRAINER


SUCTION STOP VALVE


DC

ANGLE VALVE

CHECK VALVE

ELECTRONIC EXPANSION VALVE

SIGHT GLASS

HAND EXPANSION VALVE

PRV VALVE

TANK DISTRIBUTOR




Page45of52
ENG00019544REV02013ColmacCoilManufacturingInc.

P.O. Box 571 | Colville, WA 99114 | USA


T: +1.509.684.2595 | F: +1.509.684.8331
www.colmaccoil.com



Withaconventionalbottomfeedandhotgasdefrostpipingarrangement,hotgasisfirst
sentthroughthedrainpanloopandtheninseriesthroughthecoilblock.Thiscommonly
usedarrangementiseffectiveandsimple,however,itrequiresthatthehotgassolenoid
remainsopentokeepthedrainpanheatedlongenoughforallwatertocompletely
drainandexitthroughthedrainpiping.Convectiveheatlosstotheroomcontinuesafter
thecoilisclearoffrostwhilethepanisdraining.

Amoreefficientarrangementistocontrolhotgastothecoilblockandtothedrainpan
loopseparatelythroughtwoseparatelytimedhotgassolenoidvalves.Thisarrangement
shortenstheamountoftimehotgasisflowingthroughthecoilblock,minimizingthe
convectiveheatlossandmaximizingdefrostefficiency.

AproperlysizedcontrolvalvegroupfortheColmacSmartHotGasDefrostpiping
arrangementislessexpensivethanaconventionalbottomfeedhotgasdefrostpiping
arrangementwithdefrostregulator.

ColmacDXSmartHotGasDefrostSequenceofOperation

1. Defrostisinitiated.
2. LiquidLineSolenoid(LLS)closes
3. Briefpumpoutperiod(5minutes)
4. Fan(s)stop
5. Panloopsolenoidenergizedfortimedpanpreheat(23minutes)
6. Coilhotgassolenoidandpilotsolenoid(closessuctionstopvalve)open
7. Timeddefrost(68minutes)
8. Coilhotgassolenoidcloses
9. Aftercoilpressureisequalizedtosuctionpressure(35minutes),SuctionStop
Valveopens
10. OpenLLS
11. Panloopsolenoiddeenergized
12. After5minutecooldowndelayfansrestart


DiversityandDefrostTiming

Aswithallhotgasdefrostsystems,thetwotoonerulemustbeobservedinthe
executionofdefrosts.Thatis,aminimumoftwoevaporatorsinthesametemperature
zonemustberunning(liquidlinesolenoidsopen)atthesametimeoneevaporatoris
defrosting.Thisstrategyisneededtoprovideenoughloadtobalanceevaporatingto
condensing(defrosting)capacityintherefrigerationsystem.

HotGasFlowrateandValveSelection

Colmacprovidestoitsrepresentativesandselectedcustomersacalculationtoolfor
estimatinghotgasflowrategivenoperatingtemperaturesandevaporatordimensional
data.
Page46of52
ENG00019544REV02013ColmacCoilManufacturingInc.

P.O. Box 571 | Colville, WA 99114 | USA


T: +1.509.684.2595 | F: +1.509.684.8331
www.colmaccoil.com

XV.


ThecompleteColmacDXSmartHotGasDefrostevaporatorcontrolvalveisalsooffered
fromColmacasakitorfactorymountedontheevaporator.

DefrostWaterVolumeandDrainLineSizing


Followingisasimplemethodtocalculatetheamountofmoistureremovedbytheair
coolersfromairintherefrigeratedspacesinordertodetermine:

A) Totalseweragerequirementsforthefacility,and
B) Properdrainpipingsizestohandlepeakflowratesduringdefrost.

Thissectionwillpresenttwosimplecalculationmethodsfordeterminingthese
importantdesignparameters(Nelson2008).

DeterminingTotalVolumeofMoistureRemoved:

Inordertoestimatethevolumeofwatergeneratedfromdefrosting(orwetfin)air
coolers,thehoursperdaythecooler(s)operatealongwiththeSensibleHeatRatio
(SHR)mustbeknown.Assumingaroomrelativehumidityof90%,theSHRforanair
cooleroperatingatvarioustemperatureswillbeasshowninTable13below:

Table13
SHRfor90%rhAiratVariousTemperatures

RoomTemp,F
AirSHR
45
0.59
32
0.70
10
0.85
10
0.93
30
0.98


Theamountofmoistureaccumulatedonthesurfacesoftheaircooler(s)thatwillbedrained
ascondensedwaterinhightemproomsorasmeltedfrostinmediumandlowtemprooms,
canbeestimatedusingthefollowingformula:

 Gal / day

1.35t 1  SHR Q 

(8)


where:
t=OperatingTime,hours/day
SHR=AirSensibleHeatRatio
Q=SystemCoolingCapacity,tons(note:1ton=12,000Btuh)
1.35(constant)=12,000Btuh/ton/(8.33lbs/galx1,068Btu/lbs)
8.33lbs/gal=liquiddensityofwater
1,068Btu/lbs=latentheatofvaporizationofwater

Page47of52
ENG00019544REV02013ColmacCoilManufacturingInc.

P.O. Box 571 | Colville, WA 99114 | USA


T: +1.509.684.2595 | F: +1.509.684.8331
www.colmaccoil.com


Example1:
 RoomTemp:45F
 OperatingTime:12hours/day
 RoomSHR(fromTable1):0.59
 SystemCoolingCapacity:50tons
 CondensedWaterVolume=1.35x12x(1.59)x50=332gal/day

Example2:
 RoomTemp:10F
 OperatingTime:16hours/day
 RoomSHR(fromTable1):0.93
 SystemCoolingCapacity:100tons
 DefrostWaterVolume=1.35x16x(10.93)x100=151gal/day

DeterminingPeakDefrostWaterFlowrate

Todeterminethepeakdefrostwaterflowrateleavingafrostedcoilsurface,firstcalculate
thevolumeofwateryieldedbyacoolingcoilduringdefrostusingthefollowingequation:










 










 t fin

S fin

 0.0937
 u Asurf
 u 
 u H
2

Vdef

 :
where

(9)

Vdef
Asurf
S fin
t fin

H
Note:Thisequationassumesfrosthasaveragedensityof150kg/m3(Besant1999),approx.
1/6ththatofliquidwater.

Example1:
 TotalSurfaceArea=4,500sqft
 FinSpacing=4finsperinch
 FinThickness=0.012inches
 FractionofFrostBlockage=0.5
 VolumeofDefrostWater=0.0937x4,500x(1/40.012)/2x0.5=25gallons

Page48of52
ENG00019544REV02013ColmacCoilManufacturingInc.

P.O. Box 571 | Colville, WA 99114 | USA


T: +1.509.684.2595 | F: +1.509.684.8331
www.colmaccoil.com



Inordertothendeterminethepeakflowrate,anestimateofthelengthofdefrosttime
mustbemade.Forhotgasdefrosting,themajorityofdefrostwaterflowstothedrainina
relativelyshortperiodoftime.Toestimatepeakflowrateofdefrostwateranestimated
durationofdefrostof5minutescanreasonablybemade(Stoecker1983).

Tocalculatepeakflowrate,simplydividethevolumeofdefrostwaterbytheestimated
durationofdefrost.Fortheexample:

Estimatedpeakdefrostflowrate=25gal/5min=5gpm

Drainlinescannowbesizedbasedonthecalculatedpeakdefrostflowrate.Themaximum
peakflowrateforafacilitywillbethecombinedflowratesforthemaximumnumberof
coolingcoilsexpectedtodefrostsimultaneously.

Colmacprovidestoitsrepresentativesandselectedcustomersacalculationtoolfor
estimatingdefrostflowrategivenoperatingtemperaturesandevaporatordimensional
data.

SizingSlopingDrainLines

TheAmericanSocietyofPlumbingEngineers(ASPE)publishessizingmethodsforvertical
andslopingdrains(ASPE1999).ThefollowingtableistakenfromtheASPEDataBook
Volume2,page8.

Table14
ApproximateDischargeRatesandVelocitiesinSlopingDrains,n=0.015*

ActualInsideDiameter FullFlowDischargeRateandVelocity
ofPipe,inches
Basedoninch/ftSlope
Discharge,gpm
Velocity,fps
13/8
3.13
1.34
1
3.91
1.42
15/8
4.81
1.50
2
8.42
1.72
2
15.3
1.99
3
24.8
2.25
4
53.4
2.73
5
96.6
3.16
6
157
3.57
8
340
4.34
10
616
5.04
12
999
5.67


*n=Manningcoefficient,whichvarieswiththeroughnessofthepipe.

Page49of52
ENG00019544REV02013ColmacCoilManufacturingInc.

P.O. Box 571 | Colville, WA 99114 | USA


T: +1.509.684.2595 | F: +1.509.684.8331
www.colmaccoil.com


Horizontaldrainlinesmustbepitchedatleastperfoottoinsurepositivedrainage.

Drainlinesrunningthroughfreezingspacesshouldbeactivelyheatedwithheattracecable
andthenwellinsulated.

Drainlinesshouldalsohaveptrapsinstalledjustoutsidetherefrigeratedspacetoprevent
backflowofwarmhumidambientairthroughthedrainlineintotherefrigeratedspace.


XVI.

References

Briley,G.C.1984.Lubricant(Oil)Separation.InternationalInstituteofAmmonia
Refrigeration,Alexandria,VA.ProceedingsoftheIIAR1984AnnualMeeting,pp107F131F.

ClelandD.J.,OHaganA.N.2002.PerformanceofanAirCoolingCoilUnderFrosting
Conditions.AmericanSocietyofHeating,RefrigeratingandAirConditioningEngineers.
ASHRAETransactions2002V.109,Pt.1

Cleland,D.J.2012.TheEffectofWaterVapouronFoodRefrigerationSystems.The
InstituteofRefrigeration.London,UK.Proc.Inst.R.201112.5.

Davis,J.R.,1999,CorrosionofAluminumandAluminumAlloys,ASMInternaional,
MaterialsPark,OH.Chap2,pp38.

IIAR2004,AmmoniaRefrigerationPipingHandbook.InternationalInstituteofAmmonia
Refrigeration.Alexandria,VA.

IIAR2008,ANSI/IIAR22008Equipment,Design,andInstallationofClosedCircuitAmmonia
MechanicalRefrigeratingSystems,InternationalInstituteofAmmoniaRefrigeration.
Alexandria,VA.

Jensen,S.2013.Personalcorrespondence.

LaRocheIndustries,1987,AquaAmmoniaInformationManual,LaRocheIndustriesInc.
Concord,NC.

LaRocheIndustries,1998,MaterialSafetyDataSheet#4003(AmmoniumHydroxide),
LaRocheIndustriesInc.Concord,NC.

Nelson,B.I.,2008,DeterminingDefrostWaterVolume.TechnicalBulletin.ColmacCoil
Manufacturing,Inc.Colville,WA.

Nelson,B.I.,2010,ThermodynamicEffectsofWaterinAmmoniaonEvaporator
Performance,InternationalInstituteofAmmoniaRefrigeration,Alexandria,VA.
ProceedingsoftheIIAR2010AnnualMeeting,pp201236.


Page50of52
ENG00019544REV02013ColmacCoilManufacturingInc.

P.O. Box 571 | Colville, WA 99114 | USA


T: +1.509.684.2595 | F: +1.509.684.8331
www.colmaccoil.com


Nelson,B.I.,2011,U.S.Patent7,958,738,DirectExpansionAmmoniaRefrigerationSystem
andaMethodofDirectExpansionAmmoniaRefrigeration.

Nelson,B.I.,2011(1),OptimizingHotGasDefrost.TechnicalBulletin.ColmacCoil
Manufacturing,Inc.Colville,WA.

Nelson,B.I.,2011(2),ColmacIntroducesDXAmmoniawithSmartHotGasSHGDefrost.
TechnicalBulletin.ColmacCoilManufacturing,Inc.Colville,WA.

Nelson,B.I.,2012(a),ComparingAirCoolerRatingsPart1:NotAllRatingMethodsare
CreatedEqual.TechnicalBulletin.ColmacCoilManufacturing,Inc.Colville,WA.

Nelson,B.I.,2012(b),ComparingAirCoolerRatingsPart2:WhyDTMRatingsCostYou
Money.TechnicalBulletin.ColmacCoilManufacturing,Inc.Colville,WA.

Smith,E.2010,TechnicalNote.InternationalInstituteofAmmoniaRefrigeration(IIAR).
Alexandria,VA.CondenserMagazineAugust2010.

Stoecker,W.F.1988,IndustrialRefrigerationHandbook,BusinessNewsPublishing
Company,Michigan.

TannerIndustries,2000,MaterialSafetyDataSheet(AmmoniumHydroxide),Tanner
Industries,Inc.Southampton,PA.

Welch,J.2013.Personalcorrespondence.

Wiencke,B.2002,SizingandDesignofGravityLiquidSeparatorsinIndustrial
Refrigeration,InternationalInstituteofAmmoniaRefrigeration,Alexandria,VA.
ProceedingsoftheIIAR2002AnnualMeeting,pp63133.

Wiencke,B.2012.Personalcorrespondence.

Wierbosch,M.andSandler,M.2010,BaseFluidEffectonPerformanceinanAmmonia
RefrigerationSystem.RMSupportBV.Henglelo,NL.












Page51of52
ENG00019544REV02013ColmacCoilManufacturingInc.

P.O. Box 571 | Colville, WA 99114 | USA


T: +1.509.684.2595 | F: +1.509.684.8331
www.colmaccoil.com


XVII.


APPENDIXA

ThreeP&IDdiagramsareshownrepresenting:

Figure1SingleStageSingleTemperatureLevel
Figure2SingleStage(EconomizedScrew)MultipleTemperatureLevel
Figure3TwoStageMultipleTemperatureLevel



































Page52of52
ENG00019544REV02013ColmacCoilManufacturingInc.

LP
HLCO

SH

OPTIONAL
BPRV

DX EVAPORATOR
AIR DEFROST

SH
P

DX EVAPORATOR
HOT GAS COIL ONLY

P
SH

FIGURE1
COLMACDXAMMONIA
SINGLESTAGESINGLETEMPERATURELEVEL










DRY
SUCTION
OPERATION
TRAP





OIL POT

T
HTR

110F
WATER STILL
TRANSFER







e

SOLENOID VALVE
ELECTRONIC EXPANSION VALVE

 VALVE
DRAIN VALVE
3-WAY

RELIEF VALVE
GLOBE VALVE

AUTO VENT
CHECK
 VALVE

TANK DISTRIBUTOR
PUMP

PRESSURE GAUGE
REFLEX LEVEL EYE


SUPERHEAT TRANSDUCER
TEMPERATURE
TRANSDUCER
T





EXPANSION
TANK
EVAPORATIVE
CONDENSER

90 psig
60F

GLYCOL OIL
COOLING LOOP

COMPRESSOR
LEVEL
RANGE

HIGH PRESSURE
RECEIVER

MOTOR

COALESCING OIL
SEPARATOR

SUBCOOLER

N.C.

GLYCOL
OIL COOLER
e
SUBCOOLER SUCTION

75 psig
(50.3F SAT)
T
SUBCOOLED
to 40F

SUBCOOLED REDUCED PRESSURE LIQUID

PRV VALVE W/ ELECTRIC SHUT OFF


HAND EXPANSION VALVE
ANGLE VALVE
STRAINER
DOUBLE ACTING RELIEF VALVE
LP LEVEL PROBE

Copyright2013ColmacCoilManufacturingInc.

FIGURE2
COLMACDXAMMONIA
SINGLESTAGE(ECONOMIZEDSCREW)MULTIPLETEMPERATURELEVEL


90 psig
60F

70 psig

(47F)


LP

HLCO
LEVEL
FLASH COOLER/

RANGE
ECONOMIZER
RECEIVER


SUBCOOLER

N.C.

OIL POT

SUBCOOLER
75 psig
e
(50.3F SAT)

N.C.
T

SUBCOOLED

SUBCOOLED
to 40F
e
by 10F

ECONOMIZER
T



SUCTION

LP

HLCO

SUCTION
DRY
TRAP
OPERATION

MOTOR


COMPRESSOR

COALESCING OIL
SEPARATOR

OIL POT
T
HTR
110F
WATER STILL
TRANSFER
GLYCOL

OIL COOLER


e
SOLENOID VALVE
ELECTRONIC EXPANSION
VALVE
PRV VALVE W/ ELECTRIC SHUT OFF


DRAIN VALVE
3-WAY VALVE
HAND EXPANSION VALVE

RELIEF VALVE
GLOBE VALVE
ANGLE VALVE

LIQUID DRAINER
CHECK VALVE
STRAINER

TANK DISTRIBUTOR
DOUBLE ACTING RELIEF VALVE
PUMP


AUTO VENT
PIPE CAP (CLEANOUT)
LP LEVEL PROBE

SH
P

HOT GAS PAN LOOP


DX EVAPORATOR
SMART HOT GAS COIL & PAN
MEDIUM TEMPERATURE

SH
P

HOT GAS PAN LOOP


DX EVAPORATOR
SMART HOT GAS COIL & PAN
LOW TEMPERATURE

PRESSURE GAUGE

SH SUPERHEAT TRANSDUCER

REFLEX LEVEL EYE

TEMPERATURE TRANSDUCER
PRV VALVE

EXPANSION
TANK
EVAPORATIVE
CONDENSER

GLYCOL OIL
COOLING LOOP

Copyright2013ColmacCoilManufacturingInc.

FIGURE3
COLMACDXAMMONIA
TWOSTAGEMULTIPLETEMPERATURELEVEL

90 psig

60F

70 psig
(47F)


LP

HLCO

LEVEL
INTERCOOLER/
RANGE
ECONOMIZER

RECEIVER

HIGH STAGE
COMPRESSOR
SUBCOOLER

MOTOR

N.C.

OIL POT

SUBCOOLER
75 psig
e
(50.3F SAT)

COALESCING OIL
T
SEPARATOR
N.C.

SUBCOOLED

to 40F
e
SUBCOOLED
GLYCOL

by 10F
OIL COOLER
T




LP

BOOSTER

COMPRESSOR
HLCO
SUCTION
DRY
OPERATION
TRAP 
MOTOR



COALESCING OIL

SEPARATOR
OIL POT
T

HTR
110F

WATER STILL
TRANSFER
GLYCOL
OIL COOLER


e

SOLENOID VALVE
ELECTRONIC EXPANSION VALVE
PRV VALVE W/ ELECTRIC SHUT OFF

DRAIN VALVE
3-WAY VALVE 
HAND EXPANSION VALVE
RELIEF VALVE
GLOBE VALVE
ANGLE VALVE

LIQUID DRAINER
CHECK VALVE
STRAINER

TANK DISTRIBUTOR
DOUBLE ACTING RELIEF VALVE
PUMP


SH
P

HOT GAS PAN LOOP


DX EVAPORATOR
SMART HOT GAS COIL & PAN
MEDIUM TEMPERATURE

SH
P

HOT GAS PAN LOOP


DX EVAPORATOR
SMART HOT GAS COIL & PAN
LOW TEMPERATURE

AUTO VENT

PIPE CAP (CLEANOUT)

PRESSURE GAUGE

REFLEX LEVEL EYE

SH SUPERHEAT TRANSDUCER

LP LEVEL PROBE
T

TEMPERATURE TRANSDUCER
PRV VALVE

EXPANSION
TANK
EVAPORATIVE
CONDENSER

GLYCOL OIL
COOLING LOOP

Copyright2013ColmacCoilManufacturingInc.