Master of Science Thesis

KTH School of Industrial Engineering and Management

Energy Technology EGI-2013-0MS! EK"#0
$i%ision of Heat and &o'er Technology
SE-100 (( ST)!KH)*M







+,sor-tion cooling in district heating
net'or./ Tem-erature difference
e0amination in hot 'ater circuit


1u'ardi










-2-




Master of Science Thesis EGI-2013-066MSC
EKV960
Absorption cooling in district heating
network: Temperature difference
examination in hot water circuit



1u'ardi
+--ro%ed
02 +ugust 2013
E0aminer
"i.toria Martin
Su-er%isor
"i.toria Martin
!ommissioner

!ontact -erson

Abstract
Cooling demand occurs in Sweden due to the heating period during summer time. At the moment, only
about 22-26% of the cooling demand in Sweden is satisfied by the district cooling, the rest is satisfied by
individual refrigerators or air conditioning. hus, the opportunity to fulfill the cooling demand by a more
environment friendly way is still widely open in Sweden. !ne approach that can be used considering the
source availability in Sweden is the small scale"decentrali#ed absorption cooling driven by the district
heating. his approach could be considered as a smart strategy since the district heating can directly sell
the e$cessive hot water to produce chilled water instead of decreasing its production.
!ne of the challenges for implementing this approach is to have the certain temperature re%uirements that
have to be fulfilled. Currently, the Swedish district heating supplies the outgoing hot water at temperature
range of &'-(2' C and receives the returning hot water averagely at )& C. he district cooling supplies
chilled water at 6 C and returning water at (2-(6 C. he district heating company desires to have the
returning hot water temperature to be as low as possible, as it give benefits in term of less pump wor*,
and energy recovery for the condensation process at central heating plant. +urthermore, the re%uirement
of 6/(2 C is important to ma$imi#e the amount of cold distributed per unit volume of water pump.
,aving analy#ed the potential of absorption cooling driven by district heating, the main ob-ective of this
thesis is to e$amine the temperature difference .cool-off/ of the hot water circuit at input of &' C in the
decentrali#ed absorption cooling system driven by district heating networ* with outgoing and returning
chilled water respectively at 6 C and (2 C. 0n order to achieve this ob-ective, the main activities are
conducted including literature review and modeling simulation.
he results of the simulation show that the C!1 value decreases as the cool-off increases. he analysis is
done with #ero-order model that is able to predict the minimum heat input temperature needed to drive
the absorption system at nominal C!1. 0t shows that the larger the cool-off, the higher the heat input
temperature needed to avoid the decreasing of C!1. he other important result is the ma$imum lowest
returning water temperature possible is 22 C, with cool-off value of (2 C. his means that the desired
returning hot water temperature at )& C or cool-off value of 23 C is not possible to be achieved. 4ased
on this result, the district heating company has to find the optimum trade-off between the C!1 and cool-
off value of the absorption chiller in real operation.
-3-

Table of Contents
Abstract ........................................................................................................................................................................... 2
5ist of +igures ................................................................................................................................................................ 2
5ist of ables .................................................................................................................................................................. 6
Abbreviation ................................................................................................................................................................... &
6omenclature................................................................................................................................................................. 7
Ac*nowledgment ........................................................................................................................................................... 8
( 0ntroduction ........................................................................................................................................................('
(.( Cooling demand in Sweden .....................................................................................................................('
(.2 Swedish 9istrict ,eating .........................................................................................................................('
(.3 Swedish 9istrict Cooling .........................................................................................................................((
(.) !verview of hermal driven cooling .9C/ technologies ................................................................(2
(.2 9iscussion for establishing 9C driven by district heating ..............................................................(3
(.6 !b-ective ....................................................................................................................................................(3
(.& :ethodology ..............................................................................................................................................()
(.7 Scope and limitation of the thesis ..........................................................................................................(2
2 heoretical bac*ground of single-effect absorption chiller .........................................................................(6
2.( Single-effect 6,3-,2! absorption chiller wor*ing principle ..........................................................(6
2.2 ;ero-order model of absorption cycle...................................................................................................(&
2.3 <or*ing fluids for absorption cycle .......................................................................................................(8
2.) hermodynamic properties of 6,3-,2! solution ............................................................................2(
2.).( Solution composition ......................................................................................................................2(
2.).2 0deal 6,3-,2! solution ...............................................................................................................2(
2.).3 =eal 6,3-,2! solution ................................................................................................................22
2.).) emperature characteristic of real 6,3-,2! solution ............................................................23
2.).2 >nthalpy definition of 6,3-,2! solution in two-phases .......................................................2)
2.2 Absorber .....................................................................................................................................................22
2.6 ?enerator"9esorber ................................................................................................................................26
2.& Condenser ..................................................................................................................................................2&
2.7 hrottling device .......................................................................................................................................2&
2.8 >vaporator .................................................................................................................................................27
2.(' Solution pump ...........................................................................................................................................28
2.(( Solution ,eat >$changer .S,@/ ............................................................................................................28
2.(2 ,eat transfer ..............................................................................................................................................3'
2.(3 Comparison between vapor comparison system and absorption cycle ...........................................3'
3 :odeling of single-effect 6,3-,2! absorption cycle ...............................................................................32
3.( 5iterature review of single-effect ammonia water absorption cycle .................................................32
-)-

3.2 >>S modeling of single-effect ammonia-water absorption cycle .....................................................33
3.2.( :odeling variables ...........................................................................................................................32
3.2.2 :ass A energy balance and heat transfer e%uations in each component ...............................36
3.3 Simulation result A -ustification .............................................................................................................38
3.3.( hermodynamic analysis of the simulation .................................................................................38
3.3.2 Comparison of the simulation result to nominal operation manufacturer data ....................)(
3.3.3 :ain Simulation resultB Cool off at e$ternal hot water side ......................................................))
) Conclusion ...........................................................................................................................................................)7
2 +uture <or* ........................................................................................................................................................2'
4ibliography .................................................................................................................................................................2(
Appendi$ AB >>S code for cool-off simulation at Thot C &' C and Tchilleu, out C 6 C and Tchilleu, in
C (2 C ..........................................................................................................................................................................2)
Appendi$ 4B 6ominal operation condition data by 1in* ?mb, .1in* ?mb,, 2'('a/ .................................28


-2-

ist of !i"#res
+igure (.( Annual cooling demand in Sweden .+-DrrvDrme, 2'((/ .....................................................................('
+igure (.2 9istrict ,eating 6etwor* .9elaby, 2'(2/ ............................................................................................((
+igure (.3 Classification of thermal driven cooling process .6Ene#, 2'('/ ......................................................(2
+igure (.) hesis flow chart ......................................................................................................................................(2
+igure 2.( 4loc* diagram of single-effect absorption chiller .:artin, 2'(2/ .....................................................(6
+igure 2.2 ;ero-order model schematic .,arold, et al., (886F :artin, 2'(2/ ....................................................(7
+igure 2.3 emperature levels at #ero-order model .,arold, et al., (886/ .........................................................(8
+igure 2.) >nthalpy - mass fraction diagram of real and ideal 6,3-,2! solution at constant temperature
.=amgopal, 2'(2a/ .......................................................................................................................................................22
+igure 2.2 1ressure-concentration diagram of real and ideal 6,3-,2! solution at constant temperature
.=amgopal, 2'(2a/ .......................................................................................................................................................23
+igure 2.6 =eal pmix in two-phase real 6,3-,2! solution .=amgopal, 2'(2b/ ............................................23
+igure 2.& >%uilibrium temperature-concentration of 6,3-,2! solution at constant pressure .=amgopal,
2'(2bF ?anesh A Srinivas, 2'('/ ..............................................................................................................................2)
+igure 2.7 >nthalpy of 6,3-,2! solution in two-phases .=amgopal, 2'(2b/ ...............................................2)
+igure 2.8 :i$ing process in the absorber ..............................................................................................................22
+igure 2.(' 9esorption process in generator .left/, enthalpy concentration diagram in the generator .right/
........................................................................................................................................................................................26
+igure 2.(( Condensing process in condenser .left/ and temperature profiles in condenser .right/ .............2&
+igure 2.(2 hrottling process of 6,3-,2! solution .=amgopal, 2'(2b/ ......................................................27
+igure 2.(3 >vaporation process in evaporator .left/ and temperature profiles in the evaporator .right/ ...28
+igure 2.() ,eat transfer process in S,@ ..............................................................................................................28
+igure 2.(2 ?enerator and absorber in absorption chiller replace the mechanical compressor .:artin,
2'(2/ ..............................................................................................................................................................................3(
+igure 3.( Schematic of the single-effect absorption cycle referring to absorption chiller 106G 1C(8 by
106G ?mb, ...............................................................................................................................................................33
+igure 3.2 he 0nfluence of hot inlet temperature variation to C!1 .................................................................38
+igure 3.3 he 0nfluence of heat re-ection inlet temperature variation to C!1 ...............................................)'
+igure 3.) he S,@ efficiency variation influence to C!1 .................................................................................)'
+igure 3.2 0nfluence of the re-cooling inlet temperature towards the HA values in each component .........)(
+igure 3.6 1erformance chart with at various heat input temperature and cold water .5/ at (2"6 C .1in*
?mb,, 2'('a/ .............................................................................................................................................................)2
+igure 3.& +low chart of manufacturer data simulation comparison ..................................................................)2
+igure 3.7 Simulation and manufacturer data at different Iot and Tiecooling ............................................)3
+igure 3.8 4loc* diagram of cool-off simulation at Tiecooling C 22 C and Thot C &' C.........................))
+igure 3.(' Cool-off simulation result with Iot C 72 C ..................................................................................))
+igure 3.(( Cool-off simulation result with Thot C 7' IC ..................................................................................)2
+igure 3.(2 Cool-off simulation result with Thot C &2 IC ..................................................................................)2
+igure 3.(3 Cool-off simulation result with Thot C &' IC ..................................................................................)6
+igure 3.() :inimum heat input temperature at various cool-off and re-cooling temperatures ...................)&


-6-

ist of Tables
able 2-( Comparison between 6,3-,2! and 5i4r-,2! .,arold, et al., (886/ ..........................................2'
able 2-2 Similarities and differences of vapor compression and absorption refrigeration system ..............3'
able 3-( State 1oints in absorption cycle ..............................................................................................................3)
able 3-2 !perational data at nominal condition .1in* ?mb,, 2'('a/ ............................................................)(
able )-( radeoff between cool-off and C!1 priority .......................................................................................)7


-&-

Abbre$iation
:< :ega <att
?<h ?iga <att ,our
C!1 Coefficient of 1erformance
A=S Absorption =efrigeration System
C,1 Combined heat and power
C+C Chlorofluorocarbon
9C hermal driven cooling
5:9 5og mean temperature difference
S,@ Solution ,eat >$changer
=,@ =efrigerant heat e$changer
>>S >ngineering >%uations Solver


-7-

%o&enclat#re
w 1ump wor* in absorption cycle or vapor compression system .*</
@ ?as %uality, the value ranges from ' .(''% li%uid/ to ( .(''% gas/
5i4r 5ithium 4romide
6,3 Ammonia
,2! <ater
n
A
6umber of moles of ammonia
n
w
6umber of moles of water
m
A
:ass of ammonia in solution
m
w
:ass of water in solution
A.inde$/ Ammonia
<.inde$/ <ater
x he ratio of mole fraction of ammonia to the mole fraction of the solution
:ass fraction of ammonia in solution
H
A
he molecular weight of ammonia
H
w
he molecular weight of water

b
?as %uality of two-phase 6,3-,2! solution
: Jolume of li%uid
m :ass flow .*g"s/
H !verall heat transfer coefficient .<"m
2
C/
A ,eat >$changer surface area .m
2
/
C
p
,eat Capacity of water .*K"*g C/
AI emperature difference between incoming and outgoing of e$ternal water .C)
I
hot
0nlet temperature of e$ternal hot water circuit .C)
I
ccooIng
0nlet temperature of e$ternal re-cooling water circuit .C)
t
c],Lu
>vaporator averaged temperature .C)
t
CoId,n
he chilled water temperature coming bac* to the absorption chiller .C)
t
CoId,out
he chilled water temperature going out to the load .C)
AI
mn
he minimal temperature difference limit set in the absorption chiller
AI
u
he variable additional temperature difference in the absorption chiller

-8-

Ac'no(le)"&ent
+ive months have passed since 0 started to wor* in this thesis. he time seems to go very fast. +inally, 0
reach to the completion of this thesis. Although 0 really feel that 0 should have done better, but 0 cannot
achieve this result if there is no support from the others. herefore 0 would li*e to sincerely than* to
those who has supported me throughout this thesisB
(. +irstly, to lord Kesus Christ and mother :ary who have supported me all the time and give me
strength always during my hard time.
2. :y supervisor Ji*toria :artin, who has spent a lot of time to guide me throughout this thesis
process, there is always a new lesson that 0 got in each meeting with her.
3. Se*san Hdomsri, although he is in the vacation, but still has time to have discussion with me and
give me a lot of support. 0 really appreciate it.
). ,ans Alven*rona, who has done a lot for the e$perimental rig improvement of the absorption
chiller and his advice.
2. o all my 0nno>nergy S>5>C friends who did together the thesis with me in G, Cordova,
Ji*tor, Jen*ata, =ui-un and especially to Ale-andro who has spent many of his time to discus
with me about this thesis. han*s also to Alire#a who shares his *nowledge and material that
really helps me a lot for my thesis.
6. 5ast but not least, to my lovely family in 0ndonesia who always pray for my health and gives their
support all the time.

-('-

1 Intro)#ction
his chapter gives a preview to the reader about the bac*ground of this thesis such as the reason of
cooling demand occurrence in a country that is dominated by cold climate such as Sweden, a brief
description of Swedish district heating A cooling, overview of thermal driven cooling technologies, and
the ob-ectives and scope"limitation of this thesis. +urthermore, the discussion to choose the most
appropriate heat driven cooling technologies is also included in this chapter. he analysis for choosing the
most appropriate technology is done by e$amining the latest development of each technology with study
literature and also loo*ing at the suitability of the technology with Swedish district heating potential.
1*1 Coolin" )e&an) in S(e)en
Ambient surface temperature is an important factor that determines the cooling demand of a country or
region. he cooling demand in a country located in tropical region would have higher annual cooling
demand compared to a country that is located in northern region such as Scandinavian countries. Sweden
has a low monthly average ambient temperature of respectively -2 and L(2 IC in Kanuary and Kuly. he
main cause is the overheating period during summer time .=ydstrand, 2'')/. +igure (.( shows the annual
cooling demand in Sweden from (886 up to 2'((. As presented in the figure, the value of Swedish cooling
demand increases throughout the year, from less than ('' ?<h per year in (886 to almost 8'' ?<h per
year in 2'((. he increase has been more than 7''% in -ust (2 years. he main reasons are the growing
amount of well insulated buildings, refrigeration re%uirement, increasing internal heat loads from
computers A other office e%uipment and also high comfortable indoor temperature standard in the
buildings .5indmar*, 2''2F Andersson, 2''2F +-DrrvDrme, 2'(3b/. o conclude, cooling demand has
become an important form of energy delivery even in a cold country such as Sweden.

Figure 1.1 Annual cooling demand in Sweden (Fjrrvrme, 2011)
1*2 S(e)ish +istrict ,eatin"
9istrict heating networ* implies that the heat is supplied by a central heating plant instead of individual
boiler in each house or building. +igure (.2 shows the illustration of a typical district heating networ*. he
heat in form of hot water is produced by the central heating plant and transported among households in
the area by transferring the hot water under high pressure through a system of well insulated pipes
.+-DrrvDrme, 2'(3a/.
0
50
100
150
200
250
300
350
400
0
100
200
300
400
500
600
700
800
900
1000
1996 1998 2000 2002 2004 2006 2008 2010
Delivery,
GWh
Network
e!"th, k#
-((-

Another important parameter for delivering hot water to the customers is the flow rate. 9uring the winter
time, the flow rate of hot water is higher compared to summer time due to the higher heating demand.
his higher flow rate causes the circulating pump to wor* more which implies in more pressure drop and
more energy input re%uired. ,owever, the higher flow rate inside pipes would give less heat loss. hus,
the reduction and balance between heat loss and pump energy consumption is one of the most important
tas*s in maintaining a district heating networ*. 0n each house or building, at least two heat e$changers are
installed in order to utili#e the hot water from central heating plant to heat the radiator for space heating
and also for the hot water tap .+-DrrvDrme, 2'(3a/.

Figure 1.2 District Heating etwor! (Dela"#, 2012)
he temperature difference between outgoing and incoming hot water temperature is also an important
consideration in district heating. he delivered hot water temperature at the customer side ranges between
&' to (2' IC degrees with the average of 7) C in Sweden. !n the other hand, the average hot water
returning temperature is )& C. he reasons to have large temperature difference in a district heating
networ* are less heat loss, larger energy covered in the condensing system at the central heating or C,1
plant, and lower pump power re%uirement. +or instance, at a temperature difference increase of (' C, the
heat loss in district heating networ* is reduced up to 6% and a reduced pump power up to 26%
.Cuadrado, 2''8/.
1*3 S(e)ish +istrict Coolin"
he principle of district cooling is the same as district heating. he difference lies on the temperature of
the delivered water produced by the district cooling plant instead of individual air conditioner at each
houses or buildings. he way to produce chilled water is done by ta*ing cold water from free sources such
as la*e or sea. <hen the la*e or sea water is not cold enough, the chilled water is supplied by other cooling
devices in a central cooling plant. .+ortum Sverige, 2'(2a/. he temperature of water fed is at temperature
of 6 IC while the returning water is in range of (2-(6 IC degrees .+-DrrvDrme, 2'(3a/ to be pumped bac*
to the sea or la*e. he approach of utili#ing either e$cessive hot water or waste heat from heating
production plant to produce cooling demand can be considered as a smart strategy .+-DrrvDrme, 2'(3aF
Andersson, 2''2/. his e$cessive hot water occurs when there is less heating demand on the heating
period of summer time. 0n addition, utili#ing the free heat from waste incineration or from combined heat
and power plant .C,1/ are friendly method to the environment.
he district cooling networ* in Stoc*holm is the largest in the world in term of amount of customer and
the networ* pipe length .+ortum Sverige, 2'(2b/. ,owever, the potential for district cooling is still huge
in Stoc*holm and other cities in Sweden. Currently, the total demand for district cooling in Sweden is
within 2'''-2''' ?<h annually. :eanwhile, the current district cooling system can provide only &''
?<h. 0n 2''&, the district cooling can provide around 22% of Swedish cooling demand and it is predicted
-(2-

to be around 27% in 2'(2. ,aving identified this potential, the analysis and attempt to fulfill the rest of
cooling demand that cannot be reali#ed by the district cooling system is necessary to be put more in effort.
1*- .$er$ie( of Ther&al )ri$en coolin" /T+C0 technolo"ies
he most common technology to fulfill the cooling demand is vapor compression system which accounts
for (' to 2'% of the worldwide electricity consumption .=ydstrand, 2'')/. he main reason is because of
its high reliability and low cost of operation due to its wor*ing principle which is driven by electricity to
operate its mechanical compressor. he amount of vapor compression chiller in the world is also
increasing significantly every year. As reported by the Kapan =efrigeration and Air-conditioning 0ndustry
Association, the amount of global residential air conditioning in 2''& was &,38 million unit and 7,)8
million unit in 2'(2 .K=A0A, 2'(3/. he main issues with vapor compression system are the refrigerant
used .C+C/ that causes o#one depletion and the energy needed to produce the electricity to drive the
system. he increase of commercial"residential vapor compression air conditioner means higher global
electricity consumption which most of it is produced by fossil fuel that contributes to the global warming.
hus, 9C as the alternative technology to fulfill cooling demand is considered since the environmental
issue rose in (87'Ms. 9C has a long history bac* in (82'Ms when the absorption cycle is used for the first
time for air conditioning application .ang, et al., (887/.
9C is the technology to produce cooling with the source of heat instead of electricity. Although this
technology is environmentally friendly but there are some drawbac*s remaining, mainly because of higher
investment cost and bul*y e%uipment .1ilatows*y, et al., 2'((/. +igure (.3 shows the classification of 9C
technologies. >ach of categories has their own advantages and drawbac*s. A category of 9C that has
been well developed and widely used is the absorption cycle. 0t has two most commonly used wor*ing
pairs, 5i4r-,2! and 6,3-,2! .6Ene#, 2'('/.

Figure 1.$ %lassi&ication o& t'ermal driven cooling (rocess ()ne*, 2010)
-(3-

1*1 +isc#ssion for establishin" T+C )ri$en b2 )istrict heatin"
4ased on the review of Swedish district heating and cooling and 9C, it is obvious that the utili#ation of
waste heat or e$cessive hot water to satisfy cooling demand is a feasible. hese are the benefits associated
with this approachB
(. 0t gives solution to the fluctuating Swedish heating demand. <hen there is less heating demand,
the e$cessive hot water produced can be used directly for producing cold water by means of
absorption chiller in the central cooling plant. his is another solution apart from having
seasonal thermal storage.
2. Htili#ation of waste heat is valuable in term of financial, technical, and environmental
perspective. he waste heat can reach temperature of &' C which can be can be considered as a
high grade energy due to its temperature and potential of producing cold water. <hen the waste
heat is not used, it must be processed first before it is thrown away to the natural heat sin* such
as la*e or sea. So, instead of spending money for waste heat processes, the waste heat can be
directly used in thermal driven cooling. 0n addition, the usage of waste heat also means no
additional fossil fuel or any other energy source is needed to produce heat which translates to
another financial saving and no additional C!2 emission produced to the environment.
,aving analy#ed the categories of thermal driven cooling technology in section (.), it is obvious that the
closed cycle by means of absorption or adsorption cycle is more suitable to be integrated with district
heating system compared to open cycle and thermo-mechanical process. he open cycle is not appropriate
since it is not a closed looping process, and the thermo-mechanical process is still in the research phase. 0n
sorption technology, the absorption cycle is more developed in term of technical and financial perspective
in comparison to adsorption cycle .6Ene#, 2'('/. hus, the absorption cooling cycle by absorption chiller
can be considered at the moment as the most appropriate 9C technology to produce cooling with
district heating networ* integration. 0n the conte$t of Sweden, the potential is to develop decentrali#ed
small scale absorption cooling .by means of absorption chiller/ driven by e$cessive hot water production
from district heating networ* .Hdomsri A :artin, 2'(2/. he small scale is chosen because there are many
development and also research conducted that has been improving its technical performance compared to
the large scale absorption chiller. !ne of the large pro-ects that have been conducted is called Polysmart
Project. 0t focuses on the development of small and medium scale absorption chiller integrated with C,1
plant and is followed by 32 participants from various countries in >urope .+ranhofer 0S>, 2'('/.
1*6 .b3ecti$e
,aving reviewed the most appropriate technology of 9C for district heating integration, the challenge is
now lying on how to establish the cooling system that deliver the same chilled water temperature as the
district cooling does, and having the same returning hot temperature as the district heating does by means
of absorption chiller driven by district heating. <hen the hot water is delivered from heating plant at &'
C to drive an absorption chiller, it is e$pected to have a returning temperature around )2 C. he chilled
water produced by the absorption chiller has deliver chilled water at 6 C and returning temperature of
(2C, the main reasons for the (2"6 temperature re%uirement is to ma$imi#e the amount of cold
distributed per unit volume of water pump and the standard temperature re%uirement for fan-coil air
handling unit application .Se*har A an, 2''8/. +urthermore, the reason of having 6 C as the supplied
chilled water temperature is for the dehumidification purpose. hus, the main ob-ective of this thesis is
+,o e-amine t'e cool.o&& o& t'e 'ot water circuit at in(ut o& /0 C in &or an a"sor(tion cooling
s#stem driven "# district 'eating networ! wit' outgoing and returning c'illed water res(ectivel#
at 0 % and 12 %1. Cool off in the hot water circuit can be defined as the temperature difference between
incoming and outgoing hot water. he effect of cool off value changes to the C!1 of the absorption
chiller is observed in this thesis.
-()-

his ob-ective is important for the Swedish district heating company .Svens* +-DrrvDrme/ and the utility
companies. he large cool off implies in the low returning hot water temperature that gives them many
benefits .Cuadrado, 2''8F ;inco, et al., 2''2F <oods, et al., (888/B
(. An increase in the temperature difference results in the decreased mass flow of the pump. hus,
the pumping wor* and pressure drop is reduced. his also means the reduction in operational
cost for the pump.
2. 0n flue-gas condensing systems, the lower return water temperature means more heat that can be
covered. 0n addition, the returning temperature should be below the dew point of the gas.
!therwise, the moisture would not condensate. he effect of this is significant in the central
heating plant that utili#es wood or waste as fuel.
3. 0n cogeneration plants, the lower returning water temperature that comes from the district
heating networ* results in more electricity production since it will e$pand the steam that drives
the bac* pressure steam turbines to a lower pressure.
1*4 Metho)olo"2
0n order to achieve the ob-ective of this thesis, two main activities are conducted during the range of time
between Kanuary to Kune 2'(3B
(. 5iterature review covering the basic theory for thermodynamics and absorption chiller, published
wor*s about simulation and the state of art for single-effect 6,3-,2! absorption chiller. 0n
addition, getting information from the manufacturer of the absorption chiller .106G/ is also very
crucial in the modeling simulation.
2. Simulation modeling of the single-effect ammonia water absorption chiller using >>S. he main
purpose for performing simulation is to get a better understanding of the internal process of
absorption chiller in each component. he method that is used for the modeling is the first law of
thermodynamics which covers the energy and mass balance in each component.
he absorption chiller that is used for the e$periment and modeling reference is 106G 1C(8 single-effect
ammonia water absorption chiller by 1in* ?mb, with the nominal cooling capacity of (8 *<. ,ence, the
modeling approach and theory e$planation are based on this chiller manufacturer data and performance.
0n order to understand how this thesis is done, a flow chart is given in +igure (.).
1*5 Sco6e an) li&itation of the thesis
here are many sub-ects that can be discussed related to the integration of low temperature absorption
chiller driven by district heating. ,owever, the discussion and analysis that are done in this thesis is limited
to the ob-ective of thesis itself. hus, these are the scope and limitat
(. he thesis analysis will be based on the first law of thermodynamics especially in the modeling of
the absorption cycle. he second law or e$ergy analysis
2. !ne of the purposes of doing simula
direction. herefore, the result of simulation will be -ustified
absorption chiller theory and published wor*, and
3. =eview literature about the
cycle, low temperature driven absorption chiller, and
simulation.
). he economic study about financial saving that cou
of hot water is not included in this thesis.
-(2-
Figure 1.2 ,'esis &low c'art
Sco6e an) li&itation of the thesis
be discussed related to the integration of low temperature absorption
chiller driven by district heating. ,owever, the discussion and analysis that are done in this thesis is limited
the ob-ective of thesis itself. hus, these are the scope and limitation of this thesisB
he thesis analysis will be based on the first law of thermodynamics especially in the modeling of
the absorption cycle. he second law or e$ergy analysis in detail is not in the scope of this thesis.
!ne of the purposes of doing simulation analysis is to proof that the simulation is giving the right
e result of simulation will be -ustified by analy#ing it through the available
absorption chiller theory and published wor*, and comparison with the manufacturer data
the absorption chiller is focused on the state of the art of absorption
cycle, low temperature driven absorption chiller, and single-effect absorption chiller modeling
he economic study about financial saving that could be achieved by having optimi#ed cool off
not included in this thesis.

be discussed related to the integration of low temperature absorption
chiller driven by district heating. ,owever, the discussion and analysis that are done in this thesis is limited
he thesis analysis will be based on the first law of thermodynamics especially in the modeling of
is not in the scope of this thesis.
that the simulation is giving the right
analy#ing it through the available
with the manufacturer data.
absorption chiller is focused on the state of the art of absorption
absorption chiller modeling
ld be achieved by having optimi#ed cool off
-(6-

2 Theoretical bac'"ro#n) of sin"le-effect absor6tion
chiller
A good understanding about the internal process in each component of the absorption cycle is necessary
in order to create a modeling simulation. his chapter is started by loo*ing at the overall cyclic process of
a single-effect absorption cycle that is e$plained in a simple way so that the reader could grasp the idea of
how the heat is transformed to cooling effect in a single-effect absorption cycle .9C technology/. After
that, the #ero-order model as the simplest model for absorption cycle is e$plained. ,aving *nown one
e$ample of simple model and basic principle of cycle, the *nowledge of the refrigerant and refrigerant-
absorbent solution is presented since it is necessary especially to build the modeling simulation. he
thermodynamic properties of refrigerant, absorbent and the solution are used for building energy balance
and mass in each component to form the complete cycle. he thermodynamic process in generator,
condenser, evaporator, absorber, throttling device and solution heat e$changer are also discussed in more
detail. +inally, the comparison between absorption and vapor compression cycle is given to see the
difference and similarities between both, and this subsection answers why 9C especially absorption
technology has a good potential although it has lower energy performance compared to the vapor
compression system.
2*1 Sin"le-effect %,3-,2. absor6tion chiller (or'in"
6rinci6le
here are three main flows circulating inside a single-effect absorption chiller namely refrigerant, strong
solution, and wea* solution. <hen ammonia and water are used as the wor*ing pair, the term strong and
wea* implies for the ammonia concentration in the solution .,arold, et al., (886/. +igure 2.( shows the
bloc* diagram of a single-effect absorption chiller that has four main componentsF generator, condenser,
evaporator and absorber. his figure is made on the hori#ontal a$is of temperature and vertical a$is of
pressure to represent the different pressure and temperature values in in each component of a real
absorption chiller.

Figure 2.1 3loc! diagram o& single.e&&ect a"sor(tion c'iller (4artin, 2012)
he cyclic process of absorption cycle can be continuously established as long as the heat input in the
generator and condensing water are continuously supplied to respectively generator and
condenser"absorber. he general absorption refrigeration cycle process referring to +igure 2.( can be
e$plained by starting from the generator, through condenser, evaporator, absorber and finally return to
generator. +irst thing to be put in mind is that these four components are actually heat e$changer
-(&-

connected with e$ternal water circuit. hese are the following processes occur in the single-effect
absorption chillerB
(. he 6,3-,2! solution comes from the absorber. 0n the generator, the process of ammonia and
water separation .desorption/ is initiated by heat from e$ternal hot water circuit through the heat
e$changer .endothermic process/. After this separation process, there are two flows going out from
the generator. he first one is the ammonia refrigerant to the condenser and the second one is the
wea* solution of 6,3-,2! to the absorber
(
.
2. he ammonia vapor from the generator to the condenser e$periences phase change from
superheated steam becoming li%uid in sub-cooled or saturated state after it passes through the
condenser. ,ere, the trigger for condensation process inside the condenser is the e$ternal re-cooling
water. he e$ternal water ta*es the heat from the condenser and delivers it to the heat
sin*"environment .e$othermic process/.
3. he li%uid ammonia from the absorber passes through the throttling device in order to decrease its
pressure to the evaporator pressure. ,ere, the ammonia refrigerant can have various gas %ualities
with value ranges from ' to (. he point after throttling device is the lowest temperature value in the
absorption chiller.
). After its pressure is decreased through e$pansion device, the ammonia refrigerant enters the
evaporator where the evaporation process occurs. 0n this part, the cooling effect is produced since
the heat is ac%uired from the e$ternal cooling water circuit to evaporate the refrigerant to be
saturated or superheated vapor .endothermic process/. he result is the chilled water coming out
from the e$ternal water circuit that is connected with evaporator heat e$changer.
2. he ammonia vapor coming from the evaporator enters the absorber and the mi$ing between the
wea* solution coming from generator and the ammonia vapor coming from evaporator .absorption/
occurs. he absorption process re-ects heat to the e$ternal re-cooling water circuit .e$othermic
process/ which is connected to the absorber heat e$changer. he result of the mi$ing or absorption
process is called strong solution and it goes to the generator through solution heat e$changer. ,ence,
the cycle of absorption cooling is completed.
2*2 7ero-or)er &o)el of absor6tion c2cle
he factor that determines the performance of a thermodynamic cycle is the losses of thermodynamic
availability or e$ergy, or in the other word, irreversibility. 0n an absorption cycle, the sources of
irreversibility are the heat transfer and internal losses. he #ero-order absorption cycle is based on the
concept of modeling that accounts only for the processes that contribute to the largest irreversibility
which is the heat transfer between the system and its surrounding. hus, the internal losses are ignored in
this model .,arold, et al., (886/. hese are the benefits of using this modelB
(. he ability to predict the performance of cycle .C!1/ is remar*ably well, considering the minimal
effort that is needed to build this simple model
2. 0t is also capable to predict the minimum heat input for the cycle.

(
he wea* and strong solution refers to the ammonia concentration in the solution. 0n other words, the strong
solution has higher ammonia concentration than the wea* solution .AS,=A>, (883/.
-(7-


Figure 2.2 5ero.order model sc'ematic (Harold, et al., 16607 4artin, 2012)
he #ero-order model is built based on the three available temperatures level in an absorption cycle, such
as shown in +igure 2.2. he bloc*ed mar*ed A=S .Absorption =efrigeration System/ represents the
internal wor* of the absorption cycle. Since the internal losses are ignored, this bloc*ed mar*ed is
assumed to be reversible. he #ero-order model emphasi#es on the heat transfer losses. hus, the thermal
resistance for the three temperature levels are presented by R

. he heat transfer calculations in #ero-order

model are presented by e%uation 2-(,2-2, and 2-3.

h
=
I
h
-I
h
R
h

2-(

c
=
I
c
- I
c
R
c

2-2

c
=
I
c
- I
c
R
c

2-3

he overall energy balance in the system is presented by e%uation 2-) with steady state assumption, and
the wor* of pump and internal losses are neglected .,arold, et al., (886/. his e%uation implies that the
sum of the e$othermic reaction .
h
and
c
/ is e%ual to the endothermic reaction .
c
/. 0n addition, the
performance .C!1/ of the cycle is presented by e%uation 2-2.

h
+
c
=
c

2-)
C0P =

h
=
I
c
I
h
_
I
h
-I
c
I
c
- I
c
]
2-2
>%uation 3-32 to 2-2 gives five e%uations and si$ un*nowns. hus, another e%uation is needed in order to
be able to get the outputs of these e%uations. 4y assuming that absorber and condenser are in the same
temperature, e%uation 2-6 can be derived. 0n this e%uation, the temperature difference between the
evaporator and condenser is assumed to be the same as between absorber and generator. 0n addition, this
assumption is only valid for single-effect absorption cycle.
I
h
- I
c
= I
c
- I
c

2-6

-(8-


Figure 2.$ ,em(erature levels at *ero.order model (Harold, et al., 1660)
0t should be noted that heat transfer potential difference can be the same or different, depends on the si#e
and type of heat e$changer in each temperature level. he minimum heat input can be calculated such as
shown by e%uation 2-& referring to +igure 2.3. his e%uation assumes that the temperature difference
.AI
ht
/ is e%ual at the three temperature levels
2
. >%uation 2-7 gives the definition of the temperature lift
which is the difference between the heat re-ection temperature I
c
and the refrigeration temperature I
c

.,arold, et al., (886/.

I
h
= (AI
I]t
+2AI
h
) + 2AI
ht
+ I
c

= AI
I]t
+ 4AI
ht
+ I
c

2-&

AI
I]t
= I
c
- I
c
2-7
he e$ample of the minimum heat input needed can be calculated with inputs
3
I
c
is 6 IC, I
c
C 3( IC,
and the N for the three temperature levels are ( IC. Hsing e%uation 2-&, it is found out that the I
h
C 6'
IC. his means by using the analysis of #ero-order model, &' IC which is the temperature of hot water
coming from district heating is capable to run the single-effect chiller with the re%uired input. he
difference between &' IC and minimum heat input temperature that is calculated by #ero-order model,
can be interpreted as the additional heat input needed to overcome the internal losses that is not
considered by #ero-order model.
2*3 8or'in" fl#i)s for absor6tion c2cle
here are many research and effort to find a new wor*ing pairs to be used for absorption cycle to
overcome the limitations of current wor*ing pairs used in absorption cycle. hese are the following
re%uired characteristics for wor*ing pair in an absorption cycle .=amgopal, 2'(2aF Sun A ;hang, 2'(2/B
(. ,igh solubility of the refrigerant with the solution inside the generator.
2. 5arge difference between refrigerant and absorbent boiling point .more than 2'' C/ to ensure
that only refrigerant that is vapori#ed in the generator. 0n the other word, this is to ensure that
only pure refrigerant circulates through condenser, e$pansion device, evaporator before entering
bac* to absorber.

2
t subscript refers to heat transfer
3
hese inputs values are matched with the 106G 1C(8 absorption chiller and Swedish district cooling re%uirement
-2'-

3. he heat of mi$ing due to the absorption process in the generator should e$hibit small heat of
mi$ing. he purpose of having a small heat of mi$ing is to achieve high C!1. ,owever, the
higher the solubility of the refrigerant with the solution, the heat of mi$ing would be higher also.
his contradicts with the first statement. ,ence, in practice a trade-off is necessary between
solubility and heat of mi$ing.
). he solution should have a high thermal conductivity and low viscosity for high temperature.
2. he solution does not suffer crystalli#ation or solidification inside the system
6. he solution should be safe, chemically stable, non-corrosive, ine$pensive and widely available.
he most commonly used wor*ing pairs at the moment are 6,3-,2! and 5i4r-,2! due to their good
energy performance .Sun A ;hang, 2'(2F <u A >ames, 2'''/, although these wor*ing pairs have also
some drawbac*s according to the re%uired characteristics for absorption cycle wor*ing pair mentioned
above. able 2-( shows the comparison between these two wor*ing pairs in term of refrigerant, absorbent
and solution characteristicB
,a"le 2.1 %om(arison "etween H$.H28 and 9i3r.H28 (Harold, et al., 1660)
:ro(ert# H$.H28 9i3r.H28
;e&rigerant
,igh latent heat ?ood >$cellent
:oderate vapor pressure oo high oo low
5ow free#ing temperature >$cellent 5imited application
5ow viscosity ?ood ?ood
A"sor"ent
5ow vapor pressure 1oor >$cellent
5ow viscosity ?ood ?ood
Solution
6o solid phase >$cellent 5imited application
5ow to$icity 1oor ?ood
,igh affinity between refrigerant
and absorbent
?ood ?ood
As it can be seen in able 2-(, each wor*ing pair has their own advantage and disadvantage. 0n term of
pressure, both wor*ing pairs give too high or too low values. 0n 6,3-,2!, the high pressure can be
more than (' bars so that the temperature of the heat input can be used to separate ammonia from water
at desorber"generator. 0f the high pressure is too low, the heat input temperature re%uired would be very
high. Since the pressure needed is high and also ha#ardous characteristic of ammonia, the absorption cycle
using 6,3-,2! must have a good seal in order to avoid lea*ing from inside the system to the
environment and also warning system if there is any lea*ing occurred.
he main advantage of utili#ing 6,3-,2! compared to 5i4r-,2! is its ability to be used for wide range
of temperature including refrigeration temperature below ' C and also no need to be afraid of
crystalli#ation issue. 0n 6,3-,2!, the fluid that acts as refrigerant is ammonia, which has a very low
free#ing point of -33 C at ( bar pressure .AS,=A>, 2''8/. !n the other hand, the material that acts as
refrigerant in 5i4r-,2! is water. hus, it cannot be used for producing ice since the refrigerant water
itself will get fro#en.
he main drawbac* of utili#ing 6,3-,2! compared to 5i4r-,2! is the small boiling point difference
between the refrigerant and the absorbent which is only (33 C .=amgopal, 2'(2b/. !n the other hand,
boiling point difference between lithium bromide and water is more than (''' C at its operating
absorption cycle temperature .AS,=A>, 2''8/. hus, the conse%uence is the vapor coming out from
generator may contain not only ammonia but also water. hat is why in most of the case, 6,3",2!
-2(-

absorption cycle system has an additional component which is called rectifier. he function of this
component is to reduce the water content in the ammonia vapor coming out from the generator.
106G 1C(8 single-effect absorption chiller utili#es 6,3-,2!. ,owever, it does not have a rectifier after
its generator. he only process used is a short gas cleaning to remove ammonia vapor from wea* solution
coming from generator. he main reason for not having a rectifier component that is claimed by the
manufacturer is that the heat input temperature for nominal operation is only 72 C which is lower than
the boiling point of water
)
. hus, the water is assumed not to vapori#e at that temperature. 0n addition,
the generator and condenser pressure of absorption chiller can reach up to (2 bars. At this pressure, the
water boiling point becomes (77 C .Oengel A 4oles, 2''2/.
2*- Ther&o)2na&ic 6ro6erties of %,3-,2. sol#tion
he four important properties of 6,3-,2! solution in an absorption cycle are the composition, volume,
enthalpy, pressure, and temperature. hese properties are important since they are used in the single-effect
absorption cycle modeling using the software. Although the properties of the solution and the refrigerant
are defined automatically by the software, it is really beneficial to *now the thermodynamic meaning for
each property so that when something thermodynamically wrong or non-sense occurs in the simulation
result, one can notice it immediately.
2*-*1 Sol#tion co&6osition
here are two terms of composition that are important since those are used to obtain the value of the
other properties of the solution such as temperature, pressure and also enthalpy. hose are mass fraction
.$/ and mole"solution fraction .x/. +or 6,3-,2! solution, the mass and mole fraction are defined in
term of ammonia. $ can be defined as the ratio of ammonia mass to the total solution mass, such as
shown by e%uation 2-8B

=
m
A
m
A
+ m
w
2-8
x is defined as the ratio of mole fraction of ammonia to the mole fraction of the solutionB

x =
n
A
n
A
+n
w
2-('
n
A
and n
w
are respectively the number of moles of ammonia and water in solution. hey can be obtained
from the respective masses in solution and molecular weightsB

n
A
=
m
A
H
A
2-((

n
w
=
m
w
H
w
2-(2
H
A
and H
w
are the molecular weight of ammonia and water. heir values are respectively (& *g"*mol and
(7 *g"*mol .=amgopal, 2'(2b/.
2*-*2 I)eal %,3-,2. sol#tion
he ideal ammonia-water solution properties are defined as the mass or mole weighted average of
ammonia and water from e%uation 2-8 and 2-('. >%uation 2-(3 and 2-() show respectively the ideal
solution volume .:
mx
/, and enthalpy .
mx
/ calculation. 0nde$ A and w refer to respectively ammonia
and water. he second term which is .( - / refers to the water component of the solution. his is correct
since the total summation of ammonia and water mass fraction is e%ual (.

)
6ormal boiling point of water at ( bar pressure is ('' C. he physical characteristic of a fluid is when its pressure
is higherF the boiling point value is also increased .Oengel A 4oles, 2''&/.
-22-

:
mx
= . :
A
+ (1 - ). :
w

2-(3

mx
= .
A
+ (1 - ).
w

2-()
he ideal solution pressure .p
mx
/ for li%uid phase follows the =aoltMs law such as shown in e%uation 2-(2.
+or pressure, x is used instead of .=amgopal, 2'(2a/.
p
mx
= x. p
A
+(1 - x). p
w

2-(2
2*-*3 9eal %,3-,2. sol#tion
he real solution deviates from ideal solution due to the real solution contracts or e$pand during mi$ing
process which implies that the real :
mx
is not the same as the weighted mass average of ammonia and
water. hus, e%uation 2-(3 is not valid for real solution. he other reason is mi$ing process is an
e$othermic reaction which means that results in heat of mi$ing .positive A
mx
/. ,ence, e%uation 2-() is
also not valid for a real solution. 0nstead, e%uation 2-(6 is more appropriate for ideal solution enthalpy.
his is also the reason of low C!1 value of an absorption cycle since heat-to-wor* conversion always
produces heat losses generation such as A
mx
.
rcol
mx
= .
A
+(1 - ).
w
+ A
mx

2-(6
he profile of rcol
mx
in accordance with ideal
mx
is shown in +igure 2.). 0t is obvious that due to
A
mx
, the enthalpy of real solution becomes less than its ideal value. <hen the concentration value
approaches ' .pure water/ or ( .pure ammonia/, the real
mx
value is closer to the ideal one.

Figure 2.2 <nt'al(# . mass &raction diagram o& real and ideal H$.H28 solution at constant tem(erature (;amgo(al,
2012a)
9ue to the two reasons mentioned for deviation of real solution from the ideal solution mentioned above,
the pressure value for real solution also deviates. he profile of real p
mx
compared to ideal p
mx
in
various solutions concentration is given by +igure 2.2. he real p
mx
value is lower than the ideal p
mx
.-
ve line/. 0t is obvious also that when the concentration approaches pure component either ' or (, the real
p
mx
is closer to the ideal curve.
-23-


Figure 2.= :ressure.concentration diagram o& real and ideal H$.H28 solution at constant tem(erature (;amgo(al,
2012a)
he solution of 6,3-,2! can have two-phases .li%uid and vapor at the same time/. hus, two new terms
are introduced. hose are
L
as the li%uid phase mass fraction, and
v
as the vapor phase mass fraction.
+igure 2.6 shows the real p
mx
value with
L
and
v
as inputs.

Figure 2.0 ;eal p
m|x
in two.('ase real H$.H28 solution (;amgo(al, 2012")
2*-*- Te&6erat#re characteristic of real %,3-,2. sol#tion
here are two important terms in a real 6,3-,2! solution. hose are the boiling and dew point. 4oiling
temperature I
boI
is defined as the temperature at which the first bubble appears from a 6,3-,2!
li%uid solution. he dew point I
dcw
is defined as the temperature at which the last li%uid droplet
vapori#es. <hen the solution temperature is below I
boI
, it is said that the solution is in sub-cooled state
.both ammonia and water are in li%uid state/. +urthermore, when the solution temperature is above I
dcw
,
it is said that the solution is in superheated state .both ammonia and water are in steam state/. +inally, if
the solution temperature is in between I
boI
and I
dcw
, it is said that the solution is in two-phase state
.ammonia and water have both li%uid and steam form/. +igure 2.& shows the temperature profile of 6,3-
,2! solution with different
L
and
v
at constant pressure. 0t should be noted that the temperature
-2)-

difference
2
between I
boI
and I
dcw
depends on the boiling point difference between the absorbent and
refrigerant .,arold, et al., (886/.

Figure 2./ <>uili"rium tem(erature.concentration o& H$.H28 solution at constant (ressure (;amgo(al, 2012"7
?anes' @ Srinivas, 2010)
2*-*1 Enthal62 )efinition of %,3-,2. sol#tion in t(o-6hases
+igure 2.7 e$plains the enthalpy of 6,3-,2! in two-phase state. he solution or refrigerant in two-
phases can occur after throttling device.

Figure 2.A <nt'al(# o& H$.H28 solution in two.('ases (;amgo(al, 2012")
A new term is introduced which is the two-phase mass fraction of point b as shown in +igure 2.7. his
term is used to calculate the %uality of two-phase solution at point b, such as presented by e%uation 2-(&.

b
=

b
-
L

v
-
L

2-(&
hus, when
b
is calculated, the
b
can be calculated using e%uation 2-(7B

2
0n some literature, the difference between I
boI
and I
dcw
is called temperature glide
-22-


b
=
b
.
A
+(1 -
b
).
w

2-(7

2*1 Absorber
he mi$ing process between the absorbent and refrigerant vapor occurs in the absorber. +igure 2.8 shows
the two entering streams coming to the absorber and becoming one stream after passing through the
absorber. 0n actual absorber, stream 2 is the wea* solution coming from the generator. 0n 1in* 1C(8
absorption chiller, the ammonia vapor mi$ed with the wea* solution is removed by a short gas cleaning
process before entering the absorber. he purpose of doing so is to enable the wea* solution to absorb
more refrigerant vapor coming from the evaporator. he other incoming stream .stream (/ is the
ammonia refrigerant vapor. he result of refrigerant absorption by the absorbent is called strong solution
since it contains more ammonia compared to the wea* solution.

Figure 2.6 4i-ing (rocess in t'e a"sor"er
=eferring to +igure 2.8, the important e%uations in the absorber can be defined. hose are two mass
balances and one energy balance, which are represented respectively by e%uation 2-(8, 2-2', and 2-2(
6
B
m
1
+m
2
= m
3

2-(8
m
1
x
1
+ m
2
x
2
= m
3
x
3

2-2'

u
= m
1

1
+m
2

2
- m
3

3

2-2(
>%uation 2-(8 and 2-2' are the mass balance e%uation that does not care whether the solution is
considered as ideal or not. <hen assuming that the solution is ideal,
3
can be calculated using e%uation
2-22. he value of
3
in an ideal solution will lay in a tie line between
1
and
2
- ,owever, in the heat of
mi$ing is considered in a real solution as presented in +igure 2.). hus, instead of using e%uation 2-22,
3

is obtained by using enthalpy concentration diagram for 6,3-,2! solution that is available in many
refrigeration references. +urthermore, the e$ternal re-cooling water circuit connected to the absorber is
shown by I
cxtcnuI1
and I
cxtcnuI2
in +igure 2.8. herefore, I
cxtcnuI2
must have a higher temperature
compared to I
cxtcnuI1
. he reason is the heat of mi$ing re-ected by the absorption process raises the
temperature of I
cxtcnuI1
becoming I
cxtcnuI2
.

1
x
1
+
2
x
2
=
3

2-22

6
he energy balance in each component of the absorption chiller is arranged so it gives positive value regarding it is
e$othermic or endothermic reaction
-26-

he e$ternal water circuit in the absorber is termed as re-cooling water circuit since it acts as the heat sin*
for the absorption cycle where the heat is re-ected. +urthermore, the heat sin* for the e$othermic process
in absorber and condenser usually has the same re-cooling water circuit .Same temperature/. he
configuration can be in parallel or series. he parallel configuration with mass flow rate in each
component e%ual to (''% of series flow rate has the highest C!1 compared to other configuration
.,arold, et al., (886/. ,owever, series configuration implies in more simple installation and control
system. hus, the choice between parallel or series depends on oneMs preference.
2*6 Generator:+esorber
he desorption process occurs in generator where the generation of vapor from the condensed li%uid
phase of the solution. +igure 2.(' left shows the one stream in the generator separated becoming two
streams with their properties. 0n actual generator, stream ( is the wea* solution coming from the absorber.
Stream 2 is the strong solution going to the absorber, and stream 3 is the refrigerant ammonia vapor.

Figure 2.10 Desor(tion (rocess in generator (le&t), ent'al(# concentration diagram in t'e generator (rig't)
=eferring to +igure 2.(' .left/, the important e%uations in the generator can be defined. hose are two
mass balances and one energy balance, which are represented by e%uation 2-23, 2-2), and 2-22B
m
2
+m
3
= m
1

2-23
m
2
x
2
+ m
3
x
3
= m
1
x
1

2-2)

g
= m
2

2
+ m
3

3
-m
1

1
2-22
he enthalpy value at each stream and also the heat input re%uirement to establish desorption process in
the generator is given in +igure 2.(' .right/. A similar graphical way can be also used to describe how
much heat is re-ected in the absorber.
he strong solution .Stream 3/ is assumed to be always in the saturated condition while the wea* solution
.Stream (/ can be in sub-cooled state or saturated and the refrigerant vapor can be in saturated or
superheated state. +urthermore, from +igure 2.(' right, there are four heat input possible valuesB
(. 5ine (M P 2MB heat input for generating superheated refrigerant vapor by a sub-cooled wea*
solution.
2. 5ine (M P 2B heat input for generating saturated refrigerant vapor by a sub-cooled wea* solution.
-2&-

3. 5ine ( P 2MB heat input for generating superheated refrigerant vapor by a saturated wea* solution.
). 5ine ( - 2B heat input for generating saturated refrigerant vapor by a saturated wea* solution.
2*4 Con)enser
=eferring to +igure 2.(( left, two important e%uations in the condenser are the mass and energy balance
e%uation, such as shown in e%uation 2-26 and 2-2&B
m
in
= m
out
= m

2-26

c
= m(
in
-
out
)

2-2&
here are three temperature profiles at the refrigerant side such as shown in +igure 2.(( right. hose are
sensible temperature change from superheated to saturated state, latent constant temperature from dew to
boiling point, and sensible temperature change from saturated to sub-cooled state.

Figure 2.11 %ondensing (rocess in condenser (le&t) and tem(erature (ro&iles in condenser (rig't)
he three temperatures available in the condenser would ma*e the calculation becoming complicated for
heat transfer calculation. hus, the common method used is to assume that there is an average constant
temperature throughout the condensation process .=amgopal, 2'(2cF ,arold, et al., (886/, shown by the
red line in +igure 2.(( right. 4y using this assumption, the 5:9 in the absorber can be defined, such as
shown in e%uation 2-27.

IHI
c
=
I
cxt,out
- I
cxt,n
ln (
I
c
-I
cxt,out
I
c
-I
cxt,n
)

2-27
2*5 Throttlin" )e$ice
he most important assumption of throttling device is that it is ta*en as an adiabatic process. his
assumption results in the constant enthalpy process, such as shown by e%uation 2-28. he mass flow of
refrigerant through throttling device is the same as the mass flow through condenser and evaporator.

1
=
2

2-28

-27-

he difference between incoming and outgoing stream is the temperature, pressure and also its gas %uality.
he outlet state of throttling device could be sub-cooled or some portion of the li%uid can vapori#e as it
passes through the throttle .wo-phase/. he more sub-cooled the incoming stream from the condenser,
the fewer vapors will be contained in the outgoing stream and the better C!1 value can be obtained
.,arold, et al., (886/. +igure 2.(2 shows the incoming and outgoing stream at the throttling device and the
enthalpy fraction diagram of throttling process.

Figure 2.12 ,'rottling (rocess o& H$.H28 solution (;amgo(al, 2012")
2*9 E$a6orator
he mass balance in the evaporator is the same as in the condenser so does the energy e%uation which is
the multiplication of the mass flow flowing through the evaporator with incoming and outgoing enthalpy
difference. he three temperature profiles also occurs in the evaporator since the sub-cooled stream
coming from the e$pansion valve is vapori#ed to saturated or superheated vapor by the evaporator. +igure
2.(3 left shows the incoming and outgoing stream through the evaporator with its properties .left/ and the
constant average temperature for the evaporator. =eferring to +igure 2.(3 right, the 5:9 for evaporator
can also be obtained, such as shown in e%uation 2-3'.

IHI
c
=
I
cxt,n
-I
cxt,out
ln (
I
cxt,n
-I
c
I
cxt,out
-I
c
)

2-3'

-28-


Figure 2.1$ <va(oration (rocess in eva(orator (le&t) and tem(erature (ro&iles in t'e eva(orator (rig't)
2*10 Sol#tion 6#&6
he pump in the solution circuit has two functions, to circulate the wea* solution through solution heat
e$changers and to increase the pressure from absorber .p
Iow)
/to the generator pressure .p
hgh
/. he
pump wor* is given in e%uation 2-3( .,arold, et al., (886/B

w
pump
= (p
hgh
- p
Iow)
:. m
p
p
2-3(
: is the volume of li%uid and p
p
is the pump efficiency.
2*11 Sol#tion ,eat E;chan"er /S,<0
5ocated between absorber and generator, the S,@ has a function to transfer the heat energy contained in
the wea* solution from the generator to the strong solution from the absorber. his component is very
important for the absorption cycle performance since it reduces the heat input re%uired for the generator
by preheating the strong solution. +igure 2.() shows the process of heat transfer between the wea* and
strong solution.

Figure 2.12 Heat trans&er (rocess in SHB
-3'-

2*12 ,eat transfer
,eat transfer between the e$ternal water circuit with the internal component of the absorption chiller and
the heat transfer in the solution heat e$changer are important to determine the performance of the
absorption chiller. hese are the main e%uation to be used for heat transfer in this thesisMs absorption
chillerB
= uA IHI
2-32

= m C
p
AI 2-33
HA is the overall heat transfer coefficient multiplied by the heat e$changer area, C
p
is heat capacity of
water and AI is the temperature difference between incoming and outgoing e$ternal water circuit.
2*13 Co&6arison bet(een $a6or co&6arison s2ste& an)
absor6tion c2cle
Since both vapor compression system and absorption system are closed cycles for refrigeration purpose,
there are several similar characteristics between both methods. able 2-2 presents the similarities and
differences of these two cycles .4hatia, 2'(3/.
,a"le 2.2 Similarities and di&&erences o& va(or com(ression and a"sor(tion re&rigeration s#stem
Similarities 9ifferences
(. =efrigerant circulates inside the system passing
through condenser, e$pansion device and
evaporator.
2. 4oth cycles have a device to increase the
refrigerant pressure and also e$pansion devices
to reduce the pressure.
3. =efrigerant vapor is condensed at high
pressure and temperature. his process re-ects
heat to the environment .e$othermic reaction/
). =efrigerant vapor is vapori#ed at low pressure
and temperature creating endothermic process.
his process ta*es heat from the environment
.endothermic reaction/
(. he main driver for vapor compression
system is the electricity to operate the
mechanical compressor. !n the other hand,
the driver for absorption refrigeration system
is thermal energy.
2. he absorption refrigeration system uses
refrigerant that does not create environmental
ha#ard as the vapor compression system does
with its C+C.
3. he only moving part inside the absorption
chiller is the solution pump, while the
mechanical compressor in the vapor
compression system is the main moving part.

Another main difference between both cycles is the way to raise the pressure. 0n vapor compression
refrigeration system, the vapor is compressed mechanically by its compressor such as shown by +igure
2.(2. 0n absorption chiller, the vapor is absorbed in the absorbent to form li%uid solution before it is
pumped to high pressure by solution pump. hus, for the same pressure difference, the wor* input
re%uired to pump a li%uid .solution/ is significantly less than the wor* re%uired for compressing vapor due
to the small specific volume of li%uid such as shown by e%uation 2-3(.
able 2-2 shows that there are many advantages of an absorption cycle compared to a vapor compression
system. ,owever, in term of technical performance, the absorption system has much less C!1 compared
to the vapor compression system mainly because the absorption device receives heat instead of wor* for
its power input. herefore, it must perform the heat-to-wor* conversion internally that generates entropy
or heat loss during its conversion .?ordon A 6g, (882F :yat, et al., 2'((/. 0n addition, a safety
precaution should be addressed carefully when having an absorption cycle with 6,3-,2! as the wor*ing
pair. his is because of its ha#ardous characteristic of ammonia water solution that can be to$ic by
inhalation, creates burn in-ury, and also very to$ic to a%uatic organism .1in* ?mb,, 2'((/.
-3(-


Figure 2.1= ?enerator and a"sor"er in a"sor(tion c'iller re(lace t'e mec'anical com(ressor (4artin, 2012)
Considering the potential of district heating in Sweden, the absorption system is clearly a more
appropriate technology than the vapor compression system to satisfy the cooling demand that cannot be
fulfilled by the district cooling. Htili#ing the e$cessive hot water production from district heating or waste
heat from C,1 plant to drive absorption system instead of producing additional electricity to operate
vapor compression system would be smart strategy for a long term plan to deal with the increasing cooling
demand in a country such as Sweden. his strategy saves money and in term of environmental perspective
since it does not contribute to global warming.

-32-

3 Mo)elin" of sin"le-effect %,3-,2. absor6tion c2cle
he importance of having a literature review before performing the modeling simulation is to get the
*nowledge and e$perience from the published wor* about single-effect 6,3-,2! absorption chiller, so it
will result in time saving for performing the modeling for this thesis. 4efore presenting the modeling
simulation section, the published wor* review about state-of-the-art of the sorption refrigeration
technology and single-effect 6,3-,2! simulation are given.
3*1 iterat#re re$ie( of sin"le-effect a&&onia (ater
absor6tion c2cle
Absorption chiller has various types and si#es since it has a long history of development. he current
development of absorption chiller is mainly aimed at having a better C!1, lifetime, simpler construction
and lower heat source temperature .hira*omen, 2''8F 1ilatows*y, et al., 2'((/. 9ouble and triple effect
absorption chiller has higher C!1 compared to single-effect absorption chiller. A double effect
absorption chiller has the C!1 as defined in e%uation 3-32 .Sri*hirin, et al., 2''(/. ,owever, when
considering integration with district heating that has hot water temperature of &' C, single-effect would
be the most appropriate to be chosen since double and triple stage re%uires higher input temperature of
appro$imately more than ('' C .+igueredo, et al., 2''7/%
C0P
oubIc
= C0P
SngIc
+ (C0P
SngIc
)
2
3-(
Ker*o 5abus .5abus, 2'((/ investigated five modeling methods for 106G 1C(2 absorption chiller with the
capacity of (' *<. he only method out of those five that fully describes the internal processes by
physical and thermodynamic principle was the law of thermodynamic model, while the other models were
semi-empirical or completely empirical. +urthermore, assuming HA value as constant in the modeling
would not result in a good agreement with the e$periment result. he reason was that the HA value
changed with temperature and mass flow.
Several published wor*s stated that the manufacturer data is important to improve the modeling to have
right agreement with the e$perimental data. =onald :uhumu#a and Kerco 5abus .:uhumu#a, 2'('F
5abus, 2'((/ suggested that having the manufacturer performance data would be valuable especially when
there is no e$periment comparison available. Gayna*li and Gilic .Gayna*li A Gilic, 2''6/ investigated the
effect of different operating conditions on the performance of absorption refrigeration cycle such as
variation of condenser and absorber temperature, generator and evaporator temperature, effectiveness of
S,@ and refrigerant heat e$changer .=,@/. he investigation result showed that the performance of the
thermal load in condenser and evaporator were not affected by the absorber temperature, if the absorber
and condenser have their own e$ternal water circuit. +urthermore, the C!1 value of the system is largely
affected by the e$ternal water circuit temperature profile. 0ncreasing the incoming hot water temperature
to the generator and cold water to the evaporator generally has the tendency to increase the C!1 of the
system. !n the other hand, the increasing of absorber and condenser incoming water temperature will
decrease the C!1. 0t is concluded also that the =,@ effect is very small compared to S,@ since it only
gives 2% of C!1 increase, while the presence of S,@ could increase the C!1 up to ))%.
he second law of thermodynamic has been discussed in many published wor*s to understand the
behavior of the absorption cycle .:yat, et al., 2'((F Aphornratana A >ames, (882/. his law is used to
identify and %uantify the performance degradation and the components responsible for it. 0t is found out
that the generator contributes to the largest entropy generation appro$imately &8% of the total internal
entropy generation. he thermal energy that is dissipated by the heat transfer losses at the generator
causes the reduction in the generator heat re%uirement in the system. 0n addition, the e$periment result
lays about &'-7'% of the calculated values. his implies that the heat loss from the generator has a strong
effect on the C!1 .Gong, et al., 2'('/.
-33-

Another important aspect in a modeling simulation is to choose the appropriate software. !ne of the
main factors to decide which software to be used is the ob-ective itself. 0n this thesis, the detail internal
process of a single-effect absorption chiller is observed through the first law of thermodynamic. hus,
>ngineering >%uation Solver .>>S/ is chosen to simulate the single-effect absorption chiller of 106G 1C
(8. his software has been widely used for absorption chiller simulation in many published wor*s since
the properties of wor*ing fluids for the absorption cycle including 6,3-,2! and 5i4r-,2! are
available, easy coding format and also its reputation to have a good agreement with the e$perimental result
.Somers, 2''8/.
3*2 EES &o)elin" of sin"le-effect a&&onia-(ater absor6tion
c2cle
0n order to -ustify the simulation, two analysis methods are performed. he first one is analysis with the
thermodynamic theory. his analysis consists of the influence of the heat input and re-cooling
temperature to the C!1, influence of Solution heat e$changer .S,@/ to the C!1 and HA value analysis.
he second analysis is comparison with the manufacturer data at the nominal operation. After the
simulation -ustification has been performed to prove that the simulation is in the right direction, the main
result of the simulation which is the cool-off of hot water circuit is given.
he modeling for single-effect absorption chiller in this thesis is modified from an e$isting model of
single-effect absorption chiller model in chapter 8 of QAbsorption chillers and heat pumpsR .,arold, et al.,
(886/. ,owever, the reference model is a standalone single-effect 6,3-,2! absorption chiller system
which means that the energy inputs for each component is given by a value directly, instead of defining
the e$ternal water circuit connected to generator, condenser, evaporator and absorber. hus, some
modifications are made from this e$isting modelB
( =emoving the e$isting rectifier between generator and condenser.
2 Addition of the e$ternal water flow circuits to the generator, condenser, evaporator and
absorber. he re-cooling water circuit is installed parallel for the condenser and absorber. his
means that the condenser and absorber get the same e$ternal water temperature value.
3 he refrigerant heat e$changer .=,@/ located between throttling device and evaporator is not
used since this component does not e$ist in 106G 1C(8 absorption chiller.

Figure $.1 Sc'ematic o& t'e single.e&&ect a"sor(tion c#cle re&erring to a"sor(tion c'iller :CD :%16 "# :CD ?m"H
-3)-

+igure 3.( shows the simulation model schematic for this thesis. he inde$ number of component coming
out of the generator in the figure is 8 instead of &. his is the conse%uence due to the rectifier removal
from the e$isting model since number & and 7 are used as the inde$ of the components inside the rectifier.
his model has a thermodynamic state at each point. 0n real condition, it will change depends on the
temperature, pressure and concentration of the solution or refrigerant. ,owever, for the sa*e of
simplification, several assumptions and considerations are derivedB
(. he simulation runs on steady state condition.
2. here are only two internal pressures that are considered. hose are high pressure at the
generator A condenser and low pressure at evaporator A absorber. +urthermore, the pressure
drops across the chillerMs components are negligible.
3. Although there is no rectifier, the concentration of ammonia coming out from the generator is
considered to be almost pure or 88,8%.
). =efrigerant and solution throttling valve are considered to be isentropic .>nthalpy value before
and after entering throttling valve do not change/.
2. he solution pump is assumed to be isentropic. his means that the enthalpy change across the
pump is defined only by the volume multiplied with the pressure change .,arold, et al., (886/.
6. he solution heat e$changer efficiency value is ',8 as suggested by 9avid ,annl in its published
wor* about 106G 1C absorption chiller performance .,annl A =ieberer , 2'((/.
&. he refrigerant vapor leaving the generator .State point 8/ is assumed to be superheated at its
respective temperature.
7. he condensed refrigerant leaving the condenser .State point ('/ and the vapor leaving the
evaporator .State point (3/ are assumed to be saturated at their respective saturation
temperatures.
8. he sub-cooled heat e$changer between condenser and evaporator is deactivated by setting its
efficiency value at '.
('. he strong solution leaving the absorber .State point (/ and the wea* solution leaving the
generator .State point )/ are assumed to be saturated at their respective temperatures.
able 3-( shows the corresponding thermodynamic state points in each internal point of the absorption
cycleB
,a"le $.1 State :oints in a"sor(tion c#cle
1oint State
( Saturated vapor-li%uid solution
2 Sub-cooled li%uid solution
3 Saturated or sub-cooled li%uid solution
) Saturated li%uid solution
2 Sub-cooled li%uid solution
6 Saturated li%uid-vapor solution .two-phases/
8 Superheated refrigerant vapor
(' Saturated refrigerant li%uid
(( Saturated li%uid-vapor refrigerant .two-phases/
(2 Japor-li%uid refrigerant
(3 Saturated refrigerant vapor
() Saturated refrigerant vapor

he relation between e$ternal water and internal wor*ing fluid temperature are established from the data
from manufacturer. 0n the nominal condition stated by the manufacturer, the internal temperatures are
given as .,almdienst, 2'(2/B
-32-

(. ?enerator temperature at wea* solution side .S)T/B && IC.
2. Condenser temperature .S('T/B 3( IC.
3. Absorber internal temperature is not given in the manufacturer data, however it is assumed to be
the same as the condenser temperature, since it is supplied by the same e$ternal water source.
). >vaporator temperature is given by e%uation 3-2 as suggested by .,annl A =ieberer , 2'((/. his
semi empirical e%uation represents the control algorithm in the evaporator that depends on input
parameters that can be entered in the control panel system of the absorption chiller.
t
c],Lu
=
(t
CoId,n
+t
CoId,out
)
2
-(AI
mn
+AI
u
) -

c

L,Mux

3-2
<here,
t
c],Lu
is the evaporator averaged temperature which is S(3T
t
CoId,n
is the chilled water temperature coming bac* to the absorption chiller which is S22T
t
CoId,out
is the chilled water temperature going out to the load which is S2(T
AI
mn
is the minimal temperature difference limit set in the absorption chiller
AI
u
is the variable additional temperature difference in the absorption chiller
+urthermore, there are only three mass flows in this absorption cycleB
1. =efrigerant mass flow rate coming out from generator to condenser, e$pansion valve, and
evaporator, and goes bac* to absorberB m|9] = m|1u] = m|11] = m|12] = m|1S] =
m|14].
2. Strong solution flowing from absorber to the generatorB m|1] = m|2] = m|S].
S. <ea* solution flowing from generator to absorberB m|4] = m|S] = m|6].
3*2*1 Mo)elin" $ariables
he variables for modeling can be defined in 6 categoriesB
(. :ass flowB three e$ternal water circuit and one internal refrigerant
2. emperatureB four e$ternal inlet water circuit
3. ?as %ualityB two at generator, one at condenser and evaporator, and two at condenser
). >fficiencyB solution pump, S,@ and =,@
2. Control parameterB AI
mn
and AI
u

6. ,eat capacity of e$ternal water
9epending on the ob-ective of the simulations performed, some of the variables mentioned in this section
could be a fi$ed or varied input. All the inde$ given in the following variables refer to +igure 3.(.
3.2.1.1 Generator & External hot water circuit
S(2T ,ot water incoming temperature .C/
m
hot
:ass flow rate of hot water circuit .*g"s/
%S)T ?as %uality of the solution leaving generator to absorber
%S8T ?as %uality of the refrigerant leaving the generator to condenser
$S8T Ammonia fraction of the refrigerant leaving the generator to condenser
3.2.1.2 Condenser & External re-cooling water circuit
S(8T =e-cooling water incoming temperature .C/
-36-

m
ccooIng
:ass flow rate of re-cooling water circuit .*g"s/
%S('T ?as %uality of the condensed refrigerant leaving condenser
$S('T Ammonia fraction of the condensed refrigerant leaving condenser
3.2.1.3 Evaporator & External chilled water circuit
S22T Chilled water outgoing temperature .C/
m
ccooIng
:ass flow rate of re-cooling water circuit .*g"s/
%S(3T ?as %uality of the refrigerant vapor leaving evaporator to absorber
$S(3T Ammonia fraction of the refrigerant vapor leaving evaporator to absorber
3.2.1.4 Absorber &External re-cooling water circuit
S(&T ,ot water incoming temperature .C/
m
coId
:ass flow rate of hot water circuit, same as in the condenser .*g"s/
%S(T ?as %uality of the solution leaving absorber to generator
%S6T ?as %uality of the solution entering absorber from generator
3.2.1. !ther co"ponents
Cp ,eat capacity of e$ternal water circuit .*K"*g C/
mS(T :ass flow of refrigerant .*g"s/
etap 0sentropic efficiency of solution pump
esh$ Solution heat e$changer efficiency
esc >ffectiveness of condensate sub-cooled heat e$changer
1Ue$t 1ressure of e$ternal water circuit .( atm/
3*2*2 Mass = ener"2 balance an) heat transfer e>#ations in each
co&6onent
0n this section, the theory e$plained in chapter 2 is applied. 0n generator, condenser and absorber, there is
a temperature relation e%uation which defines the temperature difference between internal component
and e$ternal water temperature which value is (. hese e%uations are derived from the manufacturer data
design at nominal operation design with an assumption that the temperature difference of ( applied to all
condition regardless different heat input or re-cooling water temperature given to the system .,almdienst,
2'(2/. All the inde$ numbers in this section refer to +igure 3.(.
3.2.2.1 Generator
here are 6 e%uations in the generators which consist respectively of 2 mass balance, 2 heat transfer, (
temperature, and ( 5:9 definition.
m|S] = (m|9] + m|4]) 3-3
m|S] x|S] = (m|9] x|9] + m|4] x|4]) 3-)
g = (m|9] |9] + m|4] |4] - m|S] |S]) 3-2
g = m
ot
Cp (I|1S] - I|16]) 3-6
I|4] = I|16] - 1 3-&
-3&-


IHI
u
=
(I|1S] -I|4]) - (I|16] -I|9])
ln
(I|1S] -I|4])
(I|16] -I|9])

3-7
At nominal operation, the manufacturer data design states that S)T is && IC when S(6T is &7 IC
.endothermic/. >%uation 3-& is derived from this statement and it is assumed that the ( IC difference can
be used for all condition of absorption chiller operation.
3.2.2.2 Condenser
here are four e%uations in the condenser which consist respectively of ( mass balance, 2 heat transfers, (
temperature relation, and ( 5:9 definition.
m|9] = m|1u] 3-8
c = m|9] (|9] - |1u]) 3-('
c = m
rccooling
Cp (I|2u] - I|19]) 3-((
I|1u] = I|2u] + 1 3-(2

IHI
C
=
(I|1u] - I|2u]) - (I|1u] - I|19])
ln
(I|1u] -I|2u])
(I|1u] -I|19])

3-(3
At nominal operation, the manufacturer data design states that S('T is 3( IC when S2'T is 3' IC
.e$othermic reaction/. >%uation 3-(2 is derived from this statement and it is also assumed that the ( IC
difference can be used for all condition of absorption chiller operation.
3.2.2.3 Evaporator
here are four e%uations for the evaporator which consist respectively of 2 heat transfers, ( temperature
relation, and ( 5:9 definition. he temperature relation for evaporator has been given in e%uation 3-2.
hus, there are only three e%uations that are described here.
c = m|1S] (|1S] - |12]) 3-()
c = m
colJ
Cp (I|21] - I|22]) 3-(2

IHI
L
=
(I|21] - I|1S]) - (I|22] - I|1S])
ln
(I|21] -I|1S])
(I|22] -I|1S])

3-(6

3.2.2.4 Absorber
here are five e%uations for the absorber which consist of respectively 2 mass balance, ( energy balance, (
heat transfer, and ( temperature relation. >%uation 3-2( implies that the internal temperature of absorber
and condenser is assumed to be the same.
m|14] + m|6] = m|1] 3-(&
m|1] x|1] = m|14] x|14] + m|6] x|6] 3-(7
o = (m|14] |14]) + (m|6] |6]) - (m|1] |1]) 3-(8
o = m
rccooling
Cp (I|18] - I|17]) 3-2'
I|17] = I|19] 3-2(

-37-

3.2.2. #olution $eat Exchanger %#$&'
he purpose of e%uation at S,@ is to obtain the enthalpy of solution coming out from the S,@ to
generator and absorber, respectively |S] and |S]. he e%uation describes the heat transfer from the part
which has more %uantity of heat to the less part.
ls = m|1] (|Sc] -|2]) 3-22
ls stands for the heat %uantity at the left side or the rich solution side that comes from the absorber.
|Sc] is the enthalpy at point 3 that is *nown from the property of temperature of generator S)T, high
pressure pS3T, and strong solution $S3T.
rs = m|4] (|Sc] -|4]) 3-23
rs stands for the heat %uantity at right side or the wea* solution side coming from the generator. |Sc]
is the enthalpy at point three that is *nown from the property of temperature of solution after the solution
pump I|2], low pressure p|S], and wea* solution x|S]. hus, the heat transfer rate can be e$pressed by
e%uation 3-2) that firstly e$amines and choose which side has the lower heat %uantity by function of :06.
After that, |S] value can be obtained by adding heat %uantity multiplied with the efficiency of heat
e$changer csx to |2] .e%uation 3-22/. Simultaneously, , |S] can be obtained by subtracting |4] with
the heat %uantity transferred multiplied with efficiency of SEX .e%uation 3-26/.
min = HIN(ls; rs) 3-2)
|S] = |2] + sx (minm|1]) 3-22
|S] = |4] - sx (minm|4]) 3-26
0n e%uation 3-22 and 3-26, min is divided by the mass flow rate, since it will be added or subtracted to
the enthalpy that has unit of k[kg .enthalpy per mass unit/.
3.2.2.( #olution )u"p
0n some modeling simulation, since the pump consumes a very small %uantity of wor*, it is usually
neglected and |1] would be the same as S2T. ,owever, in this simulation, the wor* of pump is also
considered as e$pressed by e%uation 3-2&. 0n addition, the enthalpy change due to pump wor* is
presented by e%uation 3-28. :|1] is the volumetric density of the solution at temperature I|1], low
pressure P|1], and ?as %uality |1] with unit of m
3
kg. ctop is the isentropic efficiency of the pump.
wp = m|1] (|2] -|1]) 3-2&
|2] = |1] +swp 3-27
swp = :|1] (p|2] - p|1])ctop 3-28

3.2.2.* !verall c+cle e,uation
0n most of available modeling simulation, the coefficient of performance .C!1/ is defined as evaporator
heat %uantity divided by heat input needed in the generator. ,owever, since the pump wor* is also
calculated, the C!1 is e$pressed in more detail as shown in e%uation 3-3'. +urthermore, in order to chec*
whether all the heat %uantity in four main components plus the pump are in e%uilibrium, e%uation 3-3( is
used. he value has to be very close to #ero.
C0P = c(wp + g) 3-3'
ccck = g + c + wp - (c + o) 3-3(

-38-

3*3 Si&#lation res#lt = 3#stification
4efore giving the main result of this thesis which is the cool-off simulation result, the -ustifications based
on thermodynamic theory A literature review, and comparison to manufacturer data are given in the
following section.
3*3*1 Ther&o)2na&ic anal2sis of the si&#lation
+irst analysis in this section is to see the effect of the heat input and re-cooling water temperature
variation to the performance of the chiller .C!1/ as represented respectively by +igure 3.2 and +igure 3.3.
0n +igure 3.2, the roughly flat trend that is shown by C!1 curve can be e$plained by the #ero-order model
presented in chapter 2. Since absorption chiller is basically a three level temperature device, the
temperature increase in the hot side would also increase the generator temperature and reduce the
evaporator temperature. hese two temperature changes cause both Vg and Ve to increase .Compensate
each other/, that is why the C!1 change is not obvious. ,owever, the C!1 seems to slightly decrease
when the inlet hot temperature is very high, this is mainly caused by the heat transfer irreversibility that
overwhelms the increasing effect of Vg and Ve .,arold, et al., (886/.
+igure 3.3 shows that increasing the re-cooling water temperature decrease the C!1 value. his is related
to the process happen in the condenser and also in the absorberB
(. he higher the re-cooling temperature available in the condenser, the more vapor flash instead of
sub-cooled refrigerant li%uid that comes out from the e$pansion valve after the condenser.
,ence, less cooling effect would be produced in the evaporator.
2. 0n the absorber, the higher re-cooling temperature means that less refrigerant vapor coming from
the evaporator that can be absorbed by the wea* solution. hus, this affects the whole system
performance, because the less refrigerant amount available in the strong solution, the less also
refrigerant will be separated from the solution in the generator.


Figure $.2 ,'e Cn&luence o& 'ot inlet tem(erature variation to %8:
0
0%1
0%2
0%3
0%4
0%5
0%6
0%7
0%8
0%9
1
58 64 70 76 82 88 94 100 106 112 118
C
O
P
Hot Inlet temperature (C)
-)'-


Figure $.$ ,'e Cn&luence o& 'eat rejection inlet tem(erature variation to %8:
he second analysis is about the S,@ performance to the absorption cycle performance such as shown by
+igure 3.). As it can be seen, the efficiency of S,@ has an important role to the whole cycle performance.
he main reason is that there is more heat transferred from wea* to strong solution in the S,@ when the
efficiency is higher. his transferred heat decreases the heat re%uirement in the generator significantly. 0n
addition, this process also cools down the wea* solution coming out from the generator that improves the
performance of absorption process in the absorber.

Figure $.2 ,'e SHB e&&icienc# variation in&luence to %8:
he third analysis that is done is about the HA value as shown in +igure 3.2. 0t can be seen that with the
re-cooling temperature change
&
, the HA values in all components change, e$cept at the evaporator .HAe/.
his line is constant as the conse%uence of e%uation 3-2 which is a semi empirical approach in the
modeling. he changes of HA values are because of the characteristic of the H value itself that depends on
the temperature
7
and mass flow
8
.0ncropera, et al., 2'((/. ,ence, this concludes that the constant HA
cannot be used for a simulation since it will give a result that is not in good agreement with the real
situation.

&
emperature change in one component will affect the otherMs parameter in an absorption cycle, as this is also the
case with the hot temperature input variation analysis previously.
7
+or H value calculation, the effect of temperature is presented by 6usselt number
8
+or H value calculation, the effect of mass flow is presented by =eynolds number
0%1
0%2
0%3
0%4
0%5
0%6
0%7
0%8
22 24 26 28 30 32
C
O
P
Re-cooling Inlet temperature
&hot ' 70
&hot ' 75
&hot ' 80
&hot ' 85
0
0%1
0%2
0%3
0%4
0%5
0%6
0%7
0%8
0%9
0 0%1 0%2 0%3 0%4 0%5 0%6 0%7 0%8 0%9 1 1%1
C
O
P
SHX Efficiency
-)(-


Figure $.= Cn&luence o& t'e re.cooling inlet tem(erature towards t'e EA values in eac' com(onent
3*3*2 Co&6arison of the si&#lation res#lt to no&inal o6eration
&an#fact#rer )ata
able 3-2 gives the performance data of the absorption chiller at nominal condition .he full nominal
operation data is given in appendi$ 4/B
,a"le $.2 8(erational data at nominal condition (:in! ?m"H, 2010a)
1erformance data Hnit +an-coils =e-cooling System
Cold water circuit 1ower *< (7,6
emp in A out C (2"6
+low rate :
3
"h 2,&
,ot water circuit 1ower *< 3'
emp in A out C 72"&7
+low rate :
3
"h 3,6
,eat =e-ection
Circuit .re-cooling/
1ower *< 3'
emp in A out C 72"&7
+low rate :
3
"h 3,6

0n nominal condition as mentioned in able 3-2, the cooling capacity produced is (7,6 *<. 0n addition,
for the inputs given above the nominal condition, the cooling capacity could reach 2( *< .,annl A
=ieberer , 2'((/. ,owever, it will not be discussed further since the aim of this thesis is to investigate the
chiller at lower heat inputs given than the nominal condition.
+igure 3.6 shows the performance chart at the same temperature in and out at each circuit while the heat
input and re-cooling input temperature is variedB
0
1
2
3
4
5
6
23 24 25 26 27 28 29 30 31
U
A

a
l
u
e

(
!
"
m
#
c
)
Incoming Inlet Recooling $emperature (C) at %ot temp & '( C
()*
()+
()e
()"
Figure $.0 :er&ormance c'art wit' at various 'eat in(ut
he ob-ective of this simulation -ustification is to
with the same input and compare it with the performance cha
3.& shows the simulation flow process for this ob-ectiveB
Figure $./ Flow c'art o& manu&acturer data simulation com(arison
Since there are no data for the mass flows
inputs of mass flow given in the simulation
0
2%5
5
7%5
10
12%5
15
17%5
20
22%5
22 24 26
)
e

(
*
!
)
Re
-)2-
at various 'eat in(ut tem(erature and cold water (9,) at 12F0 %
2010a)
he ob-ective of this simulation -ustification is to recreate the manufacturer performance chart using >>S
with the same input and compare it with the performance chart given by the manufacturer,
process for this ob-ectiveB
Flow c'art o& manu&acturer data simulation com(arison
e no data for the mass flows of each hot input temperature from the manufacturer
given in the simulation is adapted to match with the temperature difference
28 30 32 34 36 38
Re-cooling Inlet $emp (C)
&hot ' 85 C
&hot ' 80 C
&hot ' 75 C

% (:in! ?m"H,
performance chart using >>S
the manufacturer, 106G. +igure

from the manufacturer, the
to match with the temperature difference
&hot ' 85 C
&hot ' 80 C
&hot ' 75 C
-)3-

re%uirement as mentioned in able 3-2 for each circuit .rial and error method/. ,ence, +igure 3.& shows
the comparison of manufacturer data and the simulation result.

Figure $.A Simulation and manu&acturer data at di&&erent Iot and T|
As it can be seen from +igure 3.7, the simulation result and manufacturer data shows a good agreement
although there is a slight difference. hese are the possible causes of the difference between the
manufacturer data and the simulation resultB
(. he simulation does not account for the internal pressure drop of the refrigerant. his factor is
not included because there is no way currently available to *now the internal refrigerant mass flow
and also its function of pressure loss.
2. he heat e$changer efficiency is assumed to be (''% in the simulation while in the reality this
value depends on the type and si#e of the heat e$changer, and the pressure and temperature of
the stream passing through the heat e$changer.
3. he linear assumption of internal and e$ternal temperature in the simulation as shown in e%uation
3-2, 3-&, and 3-(2.
Analy#ing point per point in each curve, it is found out that the biggest value difference is only &,&%
which is located at the I
hot
C 72 C curve in +igure 3.7. his means that the assumption and linear
temperature relation in the simulation is valid to be used to recreate the manufacturer performance chart.
+urthermore, although the manufacturer data for I
hot
C &' C is not available, the simulation can give the
cooling capacity that is valid referring to the other three curves that have a good agreement with the
manufacturer curves.

0
2%5
5
7%5
10
12%5
15
17%5
20
22%5
22 24 26 28 30 32 34 36 38
)
e

(
*
!
)
Re-cooling Inlet $emp (C)
,*!-.*+t-rer &hot ' 85 C
,*!-.+*t-rer &hot ' 80 C
#*!-.*+t-rer &hot ' 75 C
/i#-l*tio! &hot ' 85 C
/i#-l*tio! &hot ' 80 C
/i#-l*tio! &hot ' 75 C
/i#-l*tio! &hot ' 70 C
3*3*3 Main Si&#lation res#lt?
,aving done the simulation -ustification by
this section presents the main result of simulation which is the hot temperature cool off
absorption cycleMs C!1. +igure 3.8 shows the
this ob-ective at I
hot
C &' C and I
e%uations used in the modeling. >%uations (.( to (.(' in the bo$ are the functions in >>S that re%uires
three inputs combination to get five outputs. +or e$ample, e%uation (.( at point ( in the absorber is
defined by giving input of t .temperature/, p .pressure/, and % .gas %uality/
concentration/, h .enthalpy/, v .specific volume/, and s .entropy/
different I
hot
at 72, 7', &2 and &' C
effect to the C!1. he way to vary the cool
circuit. he result of this simulation is four
different re-cooling inlet temperature respectively 22, 2), 26, 27, 3'
Figure $.6 3loc! diagram
Figure $.10
0%58
0%6
0%62
0%64
0%66
0%68
0%7
0%72
0 5 10
C
O
P
Hot temperature cool
-))-
Main Si&#lation res#lt? Cool off at e;ternal hot (ater si)e
aving done the simulation -ustification by the thermodynamic analysis and manufacturer data simulation,
this section presents the main result of simulation which is the hot temperature cool off influence
shows the e$ample of bloc* diagram of the simulation process with
I
ccooIng
C 22 C. he bo$ in the middle of +igure
in the modeling. >%uations (.( to (.(' in the bo$ are the functions in >>S that re%uires
to get five outputs. +or e$ample, e%uation (.( at point ( in the absorber is
defined by giving input of t .temperature/, p .pressure/, and % .gas %uality/. he outputs are
ecific volume/, and s .entropy/. he simulation is performed on four
C. 0n each I
hot
temperature, the cool-off variation is
he way to vary the cool-off is to change the mass flow of the hot water e$ternal
he result of this simulation is four graphs of different I
hot
with each graph have
inlet temperature respectively 22, 2), 26, 27, 3' C.
3loc! diagram o& cool.o&& simulation at T
|
G 22 % and T

G /0 %
10 %ool.o&& simulation result wit' I II I o oo ot tt t G A= %
15 20 25 30 35
Hot temperature cool-off (C)
&re+ool ' 22 C
&re+ool ' 24 C
&re+ool ' 26 C
&re+ool ' 28 C
&re+ool ' 30 C
&hot ' 85 C
e;ternal hot (ater si)e
thermodynamic analysis and manufacturer data simulation,
influence to the
of the simulation process with
+igure 3.8 gives the
in the modeling. >%uations (.( to (.(' in the bo$ are the functions in >>S that re%uires
to get five outputs. +or e$ample, e%uation (.( at point ( in the absorber is
he outputs are $.ammonia
imulation is performed on four
off variation is done to see its
off is to change the mass flow of the hot water e$ternal
have 2 curves at

%

&re+ool ' 22 C
&re+ool ' 24 C
&re+ool ' 26 C
&re+ool ' 28 C
&re+ool ' 30 C
-)2-

Figure $.11 %ool.o&& simulation result wit' T

G A0 H%

Figure $.12 %ool.o&& simulation result wit' T

G /= H%
0%6
0%62
0%64
0%66
0%68
0%7
0%72
0 5 10 15 20 25 30
C
O
P
Hot temperature cool-off (C)
&re+ool ' 22 C
&re+ool ' 24 C
&re+ool ' 26 C
&re+ool ' 28 C
&re+ool ' 30 C
&hot ' 80 C
0%58
0%6
0%62
0%64
0%66
0%68
0%7
0%72
0 5 10 15 20
C
O
P
Hot temperature cool-off (C)
&re+ool ' 22 C
&re+ool ' 24 C
&re+ool ' 26 C
&re+ool ' 28 C
&re+ool ' 30 C
Thot 3 42 5!
-)6-


Figure $.1$ %ool.o&& simulation result wit' T

G /0 H%
+igure 3.(', +igure 3.((, +igure 3.(2, and +igure 3.(3 present the results of the simulation at various heat
input temperature. he curve shows that the C!1 decreases with increasing cool-off value. >%uation 2-33
states that lowering the mass flow of e$ternal hot water circuit would result in higher temperature
difference that can be obtained in the heat transfer process. ,owever, this also brings a conse%uence
which is higher minimum heat input re%uirement. his can be e$plained through #ero-order model to get
a value of minimum heat input for a certain condition, such as described in section 2.2. ,owever, for
cool-off in hot temperature side analysis effect to the C!1, e%uation 2-& has to be modified since it
assumes that the driving temperature difference in all temperature levels .hot, re-cooling, and cold/ is the
same. >%uation 3-32 shows the modified e%uation to calculate the minimum heat input re%uirementB

I
h
= AI
I]t
+AI
hot
+ AI
ccooIng
+AI
coId
+I
c
3-32
hese are the e$planations of each variable in e%uation 3-32B
(. I
h
is the heat input minimum re%uirement .C/
2. AI
I]t
is the difference between incoming re-cooling .S(&T or S(8T/ and outgoing chilled water
temperature .S22T/
3. AI
hot
is the temperature difference between the system and environment at hot water side
). AI
ccooIng
is the temperature difference between the system and environment at re-cooling
water side
2. AI
coId
is the temperature difference between the system and environment at chilled water side
6. I
c
is the re-ection water temperature .re-cooling/
AI
hot
is the parameter that will change since the difference between incoming and outgoing water
temperature at hot side .cool-off/ is varied to see the effect to the absorption chillerMs C!1. =eferring to
+igure 3.(, AI
hot
is defined as the difference between S)T and S(2T. he value of S)T will always be
lower by (C than the outgoing hot water S(6T such as e$plained in e%uation 3-&. hus, by putting the $-
a$is as the hot water cool-off and I
h
as the result of e%uation 3-32 in the y-a$is, +igure 3.() can be
obtained. he inputs given refer to the e%uations for the modeling such as described in section 3.2.2.
AI
ccooIng
is 2C .(C for absorber and (C for condenser/, AI
coId
is 2 C, AI
hot
is varied from 2.2 to
3( C, and I
c
are 22,2),26,27,3' C.
0%6
0%62
0%64
0%66
0%68
0%7
0%72
0 2 4 6 8 10 12 14 16
C
O
P
Hot temperature cool-off (C)
&re+ool ' 22 C
&re+ool ' 24 C
&re+ool ' 26 C
&re+ool ' 28 C
Thot 3 40 5!
-)&-


Figure $.12 4inimum 'eat in(ut tem(erature at various cool.o&& and re.cooling tem(eratures
he orange hori#ontal line is the baseline of hot water temperature which is &' C. he points above it
mean that &' C is not enough as heat input temperature to drive the absorption chiller at that certain
value of cool-off. he result due to this insufficient heat input is the C!1 decreases significantly, as
suggested by =amgopal .=amgopal, 2'(2d/. 0t is also shown that the higher the temperature of re-cooling
input water, the more points are located above the baseline I
hot
C &' C. hat is why at higher re-cooling
water temperature, the C!1 decreases rapidly starting at lower cool-off value such as shown in +igure
3.(' P 3.(3. +urthermore, it should be noted that #ero-order model analysis only accounts for the heat
transfer losses, but not the internal losses of the system. his means that the actual heat input re%uirement
would be higher for each value of cool-off than those shown in +igure 3.(). +urthermore, to show that
+igure 3.(' P 3.(3 and +igure 3.() agrees each other, an analysis can be done on the line of C!1 at
I
ccooI
C 3' C. 0n +igure 3.(3, there is no curve I
ccooI
C 3' C because as it can be seen in +igure 3.()
that the heat input re%uirement is always more than &' C. hat is why the simulation brea*s down
('
since
cool-off value of 2,2 C at I
hot
C &' C and I
ccooI
C 3' C.
he simulation also shows consistent results with the thermodynamic analysis as the I
hot
variation does
not change the C!1 value while the I
ccooIng
increase causes the decrease of the C!1 value. I
ccooIng

is a very important parameter and it should be as reasonably low as possible if higher C!1 to be obtained
while maintaining big cool-off value. +or e$ample, at I
hot
&' C .in +igure 3.(3/, the C!1s are ',&( and
',6( at respectively 22 C and 27 C .at cool-off value of 2 C/.


('
Simulation brea* down in >>S means that there is one or more variable that is not in the logical or reasonable
value. +or e$ample, entropy and mass flow values cannot be negative.
40
45
50
55
60
65
70
75
80
85
90
95
100
0 5 10 15 20 25 30 35
+
i
n
i
m
u
m

H
o
t

I
n
p
u
t

$
e
m
p
e
r
a
t
u
r
e

(

C
)
Hot temperature Cool-off (C)
&re+ool ' 22 C
&re+ool ' 24 C
&re+ool ' 26 C
&re+ool ' 28 C
&re+ool ' 30 C
&hot ' 70 C
-)7-

- Concl#sion
he obligatory elements of a competitive district heating"cooling system are a suitable cheap heat source,
demands from the heat"cooling mar*et, and pipes that connect the demand and the source. hose three
elements should be all local in order to minimi#e the capital investment in the distribution networ*
.Andrews, et al., 2'(2/. hus, the small scale "decentrali#ed absorption cooling driven by district heating
networ* can be considered to be very potential to be developed considering the resources availability in
Sweden, especially by utili#ing the e$cessive hot water during heating period in summer .Hdomsri A
:artin, 2'(2/. he main benefits obtained by having this approach are the central heating plant does not
have to decrease its production which means high efficiency can still be maintained and also the e$cessive
hot water can be directly sold to the utility company that will produce the chilled water by absorption
chiller. ,owever, the ideal temperature re%uirement by implementing this approach is to have the
returning temperature as low as possible with the temperature input of &' C and 6"(2 C temperature
re%uirement for the chilled water.
:odeling simulation is the main activity in this thesis to achieve the ob-ective. he modeling simulation is
started by loo*ing at the most similar available published simulation using 6,3-,2! which is found in
=adermacher absorption chiller boo* .,arold, et al., (886/. ,owever, after reviewing many literatures, the
simulation of 6,3-,2! single-effect absorption chiller without rectifier is uncommon in literatures about
absorption cycle. hus, the main challenge is to create the simulation of a 6,3-,2! single-effect
absorption chiller without rectifier that ma*es sense in term of thermodynamic theory and also behaves
closely to the chiller used. !ne of the most important points in ma*ing the modeling after the literature
review is to obtain the information from the manufacturer as much as possible since each type of
absorption chiller from different manufacturer has their own characteristic although generally it follows
the thermodynamic law.
he modeling simulation result shows that the cool-off at e$ternal hot water side really affects the
performance of the absorption cycle after a certain temperature depends on the I
hot
and I
ccooI
value.
here is a tradeoff between having large cool-off and also high C!1 such as shown in able )-(.
,a"le 2.1 ,radeo&& "etween cool.o&& and %8: (riorit#
:rioriti*ing t'e large cool.o&& (lower returning
district 'eating tem(erature) over %8:
:rioriti*ing t'e %8: over t'e large cool.o&&
(lower returning district 'eating tem(erature)
AdvantageB
5ess pump wor*
:ore heat is recovered in the condensation
process at central plant
:ore electricity is produced in the C,1 plant
AdvantageB
Smaller absorption chiller unit needed
5ess heat input needs to be supplied from the
central plant
he advantages of each priority provided in able )-( are made by assuming that the cooling demands
and the heat re-ection temperature are the same. 1rioriti#ing the large cool-off over the C!1 .but still in a
reasonable value/ seems to be a better choice compared to the other one. he main reason is that the
larger cool-off would provide the closer condition to the current district heating which has the supply
temperature at &'-72 C and returning temperature averagely around )& C. Another important factor in
the real operation of the absorption chiller driven by district heating is the heat re-ection"re-cooling
temperature. 4y *eeping the heat re-ection temperature reasonably low, the higher C!1 could be
achieved. +or illustration, based on the simulation result, at 2 C cool-off with hot input temperature of &'
C, the C!1 are respectively around ',&( and ',6( when the heat re-ection temperatures are respectively
22 and 27 C. his should not be a problem in Sweden when the re-ection heat water can be provided
from a free source such as sea or la*e since the average sea temperature in, for e$ample, Wstermalm area,
Sweden in the hottest season .August/ is (7.7 C .<orld Sea emperature, 2'(3/.
-)8-

he simulation also gives an important result which is the ma$imum cool-off value that can be achieved is
(2 C when the heat input temperature is &' C. his means that the returning hot water temperature is 22
C and the desired returning water temperature at )& C is not possible to be achieved. he ne$t
important step is to have an e$perimental activity to confirm this result. After having both simulation and
e$perimental result confirming each other, the district heating and the utility company can use the result it
to set the plan for implementing small scale"decentrali#ed absorption cooling driven by district heating.
+urthermore, temperature is only one parameter that needs to be considered for having the small scale
"decentrali#ed absorption cooling driven by district heating. here are still more challenges that have to
be put into account for implementing this approach .+ranhofer 0S>, 2'('/B
(. he development of optimi#ed hydraulic schemes
2. he si#ing of thermal storages which might be needed .heat storage, cold storage/
3. he development of the overall system control

-2'-

1 !#t#re 8or'
he most important wor* to do in the future is to have e$periments to investigate real cool-off value
effect to the absorption cycle C!1. hus, these are the several recommendations for the future
e$periment activityB
(. ?et information from the manufacturer if there is any change or improvement of 106G 1C (8
absorption chiller
2. Although a thermostat has been installed at the cooling load in the e$perimental rig, it would be
better if a thermometer is installed also there so that (2 C can be seen directly.
3. Adding the capacity of the heater for the hot water tan* to be able to deliver 72 C to the chiller.
Currently, it is only able to give up to 7' C.
). he safety and operation procedure for operating the absorption chiller has to be completed
before performing the e$perimental activity.
2. A safety improvement that could be done is by having a portable ammonia detector when
conducting an e$periment.
he improvement on the modeling simulation can also be performed by adding the ammonia vapor
cleaning process from the wea* solution in the absorber and also the ammonia vessel component. 0n
addition, the information about the type and si#e of the heat e$changer of each component of the chiller
would be valuable also to build the simulation with the definition of H value. +inally, the study at the
district heating side can also be done by loo*ing at the effect of the cool-off value variation to the central
heating or C,1 plant performance.
-2(-

@iblio"ra6h2
Andersson, K., 2''2. Hot and Cool Heat-pumping technologies for a modern society, 4rommaB Swedish >nergy
Agency.
Andrews, 9. et al., 2'(2. Background report on EU-27 istrict Heating and Cooling Potentials, Barriers, Best Practise
and !easures of promotion, 1ettenB >uropean CommisionB K=C Scientific and 1olicy =eports.
Aphornratana, S. A >ames, 0. <., (882. hermodynamic analysis of absorption refrigeration cycles using
the second law of thermodynamics method. "nternational #ournal of $efrigeration, (7.)/, pp. 2))-222.
AS,=A>, (883. %&H$%E Hand'ook( AtlantaB American Society of ,eating, =efrigerating and Air-
Conditioning >ngineers, 0nc.
AS,=A>, 2''8. Chapter 28 =efrigerants. 0nB %&H$%E Hand'ook( AtlantaB American Society of ,eating,
=efrigerating and Air-Conditioning >ngineers, 0nc, p. 28.2.
Atmaca, 0., Xigit, A. A Gilic, :., 2''2. he effect of input temperatures on the absorber parameters.
"nternational Comm( heat !ass )ransfer, 28.7/, pp. ((&&-((76.
4hatia, A., 2'(3. *+er+ie, of -apor %'sorption Cooling &ystems( S!nlineT
Available atB httpB""www.cedengineering.com"upload"Japor%2'Absorption%2':achines.pdf
Oengel, X. A. A 4oles, :. A., 2''2. Appendi$ 4B hermodynamics ables - :etric Hnits. 0nB
)hermodynamics. %n engineering %pproach( 4ostonB :cgraw-,ill.
Oengel, X. A. A 4oles, :. A., 2''&. )hermodynamics an engineering approach( 4ostonB :c?raw-,ill .
Cuadrado, =. S., 2''8. $eturn temperature influence of a district heating net,ork on the CHP plant production costs ,
?DvleB 9epartement of technology and built environment, Hniversity of ?Dvle.
9elaby, X., 2'(2. istrict Heating / Cooling, 4russelsB 9,CL echnology 1latform.
+igueredo, ?. =., 4ourouis, :. A Coronas, A., 2''7. hermodynamic modelling of a two-stage
absorption chiller driven at two-temperature levels. %pplied thermal engineering, 27.2-3/, pp. 2((-2(&.
+-DrrvDrme, S., 2'((. 0j1rrkyla, le+eranser och n1tl1ngd 2334-2522( S!nlineT
Available atB httpB""www.svens*f-arrvarme.se"Statisti*--1ris"+-arr*yla"
+-DrrvDrme, S., 2'(3a. istrict Heating in &,eden( S!nlineT
Available atB httpB""www.svens*f-arrvarme.se"0n->nglish"9istrict-,eating-in-Sweden"
+-DrrvDrme, S., 2'(3b. *m fj1rrkyla( S!nlineT
Available atB httpB""www.svens*f-arrvarme.se"+-arr*yla"
+ortum Sverige, 2'(2a. &6 fungerar fj1rrkyla( S!nlineT
Available atB httpB""www.fortum.com"countries"se"foretag"f-arr*yla"sa-fungerar-
f-arr*yla"pages"default.asp$
SAccessed 2' Kune 2'(3T.
+ortum Sverige, 2'(2b. 0j1rrkyla i &tockholm( S!nlineT
Available atB httpB""www.fortum.com"countries"se"foretag"f-arr*yla"f-arr*yla-i-
stoc*holm"pages"default.asp$
SAccessed 2' Kune 2'(3T.
+ortum Sverige, 2'(3a. Pris fj1rrkyla( S!nlineT
Available atB httpB""www.fortum.com"countries"se"foretag"f-arr*yla"pris-f-arr*yla"pages"default.asp$
+ranhofer 0S>, 2'('. Polysmart. Polygeneration ,ith ad+anced small and medium scale thermally dri+en air-conditioning
and refrigeration technology( S!nlineT
Available atB
-22-

httpB""www.polysmart.org"cms"upload"publications"1olyS:A=U1ublishableU+inalUActivityU=eport.p
df
?anesh, 6. S. A Srinivas, ., 2'('. >valuation of thermodynamic properties of ammonia-water mi$ture
up to ('' bar for power application system. #ournal of mechanical engineering research, Kanuary, 3.(/, pp. 22-38.
?ordon, K. :. A 6g, G. C., (882. A general thermodynamic model for absorption chillersB heory and
e$periment. Heat $eco+ery &ystems / CHP, (2.(/, pp. &3-73.
,almdienst, C., 2'(2. %uslegungsdaten f7r die 8H9:H2* %'sorptionsk1ltemaschine mit 25 k;C, WsterreichB
1in* ?mb,.
,annl, 9. A =ieberer , . =., 2'((. %nalyse einer %mmoniak:;asser %'sorptionsk1lteanlage - Betrie'serfahrungen
und &imulationsmodell'ildung, ?ra#B 0nstitut fYr <Drmetechni* .0</, echnische HniversitDt ?ra#.
,arold, G., =adermacher, =. A Glein, S. A., (886. %'sorption chillers and Heat Pumps( +loridaB C=C 1ress,
0nc.
0ncropera, +. 1., 9ewitt, 9. 1., 4ergman, . 5. A 5avine, A. S., 2'((. 0undamentals of heat and mass transfer(
&th ed. KeffersonB Kohn <iley A Sons.
K=A0A, 2'(3. !onthly &hipments of %ir Conditioners / $esidential Heat Pump ;ater Heaters, GobeB Kapan
=efrigeration and Airc-conditioning 0ndustry Association.
Gayna*li, !. A Gilic, :., 2''6. heoretical study on the effect of operating conditions on performance of
absorption refrigeration system. Energy Con+ersion and !anagement, Jolume )7, pp. 288-6'&.
Gong, 9., 5iu, K., ;hang, 5. A +ang, ;., 2'('. 0nvestigation of a ammonia-water absorption chiller
performance. Po,er and Energy Engineering Conference <%PPEEC=, pp. (-).
5abus, K., 2'((. !odelling of small capacity a'sorption chillers dri+en 'y solar thermal energy or ,aste heat, aragonaB
9epartement of mechanical engineering, Hniversitat =ovira 0 Jirgili.
5indmar*, S., 2''2. )he role of a'sorption cooling for reaching sustaina'le enegry systems, Stoc*holmB 9ept. of
Chemical >ngineering and echnology, G,- =oyal 0nstitute of echnology.
:artin, J., 2'(2. Heat ri+en Cooling( Stoc*holmB 9epartement of >nergy echnology, Gungliga e*nis*a
,Zgs*olan.
:uhumu#a, =., 2'('. !odelling, "mplementation and &imulation of a &ingle-effect %'sorption Chiller in !E$"),
?lasgowB 9epartement of :echanical >ngineering, Hniversity of Stratchclyde.
:yat, A. et al., 2'((. A second law analysis and entropy generation minimi#ation of an absorption chiller.
%pplied thermal engineering, pp. 2)'2-2)(3.
6Ene#, ., 2'('. hermally driven coolingB echnologies, 9evelopments and Applications. #ournal of
&ustaina'le Energy, 9ecember, 0.)/, pp. (6-2).
1ilatows*y, 0. et al., 2'((. State of the art of Sorption =efrigeration Systems. 0nB :orelos, ed. Cogeneration
0uel Cell-&orption %ir Conditioning &ystems( s.l.BSpringer, pp. 22-&3.
1in* ?mb,, 2'('a. %'sorption Chiller PinkChiller PC23, 5angenwangB 1in* >nwergy A Speichertechni*.
1in* ?mb,, 2'('b. PinkChiller PC23 ruck+erlustkur+en : Pressure-drop cur+es, 5angenwangB 106G >nergie
A Speichertechni*.
1in* ?mb,, 2'((. PinkChiller PC23 23 k;c %s'orptionsk1ltemaschine %r'eitsl>sung ;asser:%mmoniak
okumentation:Betrie'sanleitung, 5angenwangB 1in* ?mb,.
=amgopal, :., 2'(2a. ?esson 2@ -apour %'sorption $efrigeration systems( S!nlineT
Available atB httpB""nptel.iitm.ac.in"courses"<ebcourse-
contents"00%2'Gharagpur"=ef%2'and%2'Air%2'Cond"pdf"=AC%2'5ecture%2'().pdf
-23-

=amgopal, :., 2'(2b. ?esson 24 -apour a'sorption refrigeration systems 'ased on ammonia:,ater pair( S!nlineT
Available atB httpB""nptel.iitm.ac.in"courses"<ebcourse-
contents"00%2'Gharagpur"=ef%2'and%2'Air%2'Cond"pdf"=AAC%2'5ecture%2'(6.pdf
=amgopal, :., 2'(2c. ?esson 22. Condensers / E+aporators( S!nlineT
Available atB httpB""nptel.iitm.ac.in"courses"<ebcourse-
contents"00%2'Gharagpur"=ef%2'and%2'Air%2'Cond"pdf"=AAC%2'5ecture%2'22.pdf
=amgopal, :., 2'(2d. ?esson 2A -apour %'sorption $efrigeration &ystems Based *n ;ater-?ithium Bromide Pair(
S!nlineT
Available atB httpB""nptel.iitm.ac.in"courses"<ebcourse-
contents"00%2'Gharagpur"=ef%2'and%2'Air%2'Cond"pdf"=AC%2'5ecture%2'(2.pdf
=ydstrand, :., 2''). Heat-ri+en cooling in district energy systems, Stoc*holmB 9epartement of Chemical
>ngineering and echnology"1rocesses, G,.
Se*har, S. C. A an, 5. ., 2''8. !ptimi#ation of cooling coil performance during operation stages for
improved humidity control. Energy and Buildings, 0ssue )(, pp. 228-233.
Somers, C. :., 2''8. &imulation of a'sorption cycles for integration into refining processes, :arylandB +aculty of the
?raduate School of the Hniversity of :aryland, College 1ar*..
Sri*hirin, 1., Aphornratarna, S. A Chungpaibulpatana, S., 2''(. A review of absorption refrigeration
technologies. $ene,a'le and &ustaina'le Energy $e+ie,s, 0ssue 2, pp. 3)3-3&2.
Sun, K. A ;hang, S., 2'(2. A review of wor*ing fluids of absorption cycles. $ene,a'le and &ustaina'le $e+ie,s,
0ssue (6, pp. (788-(8'6.
ang, ., Jillarreal, 5. A ?reen, K., (887. Buideline %'sorption Chillers. %d+anced esign Buideline &eries,
CaliforniaB 6ew 4uildings 0nstitute.
hira*omen, G., 2''8. Co-generation and the ne, era of %'sorption Chiller, <angtonglangB American Society of
,eating, =efrigerating and Air-Conditioning >ngineers hailand Chapter.
Hdomsri, S. A :artin, J., 2'(2. 0j1rr+1rmeanpassad %'sorptionskyla, Stoc*holmB Svens* +-DrrvDrme A4.
<oods, 1. et al., (888. *ptimiCation of operating temperatures and an appraisal of the 'enefits of lo, temperature district
heating( S!nlineT
Available atB httpB""old.iea-dhc.org"scannedUbrochures"79,C88-'(5=.pdf
<orld Sea emperature, 2'(3. Blo'al &ea )emperature( S!nlineT
Available atB httpB""www.seatemperature.org"europe"sweden"oestermalm.htm
<u, S. A >ames, 0. <., 2'''. 0nnovations in vapour-absorption cycles. %pplied Energy, 0ssue 66, pp. 22(-
266.
;iegler, +., 2''2. State of art in sorption heat pumping and cooling technologies. "nternational #ournal of
$efrigeration, Jolume 22, pp. )2'-)28.
;inco, ,. et al., 2''2. "mpro+ement of operational temperature differences in district heating systems, 6y*ZpingB
0nternational >nergy Agency.


-2)-

A66en)i; A? EES co)e for cool-off si&#lation at T

A 40
C an) T
|,
A 6 C an) T
|,|
A 12 C
"*******************************************************************************************************************
The following procedures are used to convert units to SI
*******************************************************************************************************************"
PROCEDURE TPQ(TC;P;Q:x;h;v;s)
"given T (temperature), P (pressure) and Q (quality) in SI units,
this procedures returns x, h, v and s in SI units"
TK=TC+273,15
Pbar=P*,01
CALL NH3H2O(128;TK;Pbar;Q:TK;Pbar;x;h;s;u;v;Q)
END

PROCEDURE TPX(TC;P;x:Q;h;v;s)
"given T (temperature), P (pressure) and x (ammonia wt fraction) in SI
units, this procedure returns Q, h, v and s in SI units"
TK=TC+273,15
Pbar=P*,01
CALL NH3H2O(123;TK;Pbar;x:TK;Pbar;x;h;s;u;v;Q)
END

PROCEDURE PXQ(P;x;Q:TC;h;v;s)
"given P (pressure in bar), x (ammonia wt fraction) and Q, this procedure
returns T, h, v and s in SI units"
Pbar=P*,01
CALL NH3H2O(238;Pbar;x;Q:TK;Pbar;x;h;s;u;v;Q)
TC=TK-273,15
END

PROCEDURE PHX(P;h;x:TC;Q;v;s)
"given P (pressure in bar), enthalpy (J/g), and x (ammonia wt
fraction), this procedure returns T, Q, v and s in SI units"
Pbar=P*,01
CALL NH3H2O(234;Pbar;x;h:TK;Pbar;x;h;s;u;v;Q)
TC=TK-273,15
END

PROCEDURE TXQ(TC;x;Q:P;h;v;s)
"given T (temperature in C), Q (quality) and x (ammonia wt fraction) in SI
units, this procedure returns P, h, v and s in SI units"
TK=TC+273,15
CALL NH3H2O(138;TK;x;Q:TK;Pbar;x;h;s;u;v;Q)
P=Pbar*100
END

PROCEDURE PXS(P;x;s:TC;h;v;Q)
"given P (pressure in bar), x (ammonia wt fraction) and entropy s, this procedure
returns T, h, v and Q in SI units"
Pbar=P*,01
CALL NH3H2O(235;Pbar;x;s:TK;Pbar;x;h;s;u;v;Q)
TC=TK-273,15
END

"*******************************************************************************************************************
This procedure calculates the amount of heat that is exchanged in the solution heat exchanger
*******************************************************************************************************************"
PROCEDURE SHX(eshx;mls;mrs;ha;hb;hc;hd:hcc;hdd;qshx)
qls=mls*(hd-ha) "amount of heat on 'left side', rich solution"
-22-

qrs=mrs*(hb-hc) "amount of heat on 'right side', poor solution"
qmin=MIN(qls;qrs)
qshx=eshx*qmin "eshx is the sol. hx. effectiveness"
hcc=hb-eshx*qmin/mrs "calculation of outlet enthalpies"
hdd=ha+eshx*qmin/mls
END

"*******************************************************************************************************************
Efficiencies of respectively circulation pump, Solution heat exchanger and Refrigerant heat exchanger
*******************************************************************************************************************"
etap=0,50 "isentropic efficency of pump"
eshx=0,90 "solution heat exchanger efectiveness"
esc=0,0 "effectiveness of condensate precooler, if esc=0, no
subcooler is used"

"*******************************************************************************************************************
Solution circulation ratio and weak and strong ammonia concentration difference
*******************************************************************************************************************"
f=(x[9]-x[4])/(x[3]-x[4]) "solution circulation ratio"
deltax = x[3] - x[4]

"*******************************************************************************************************************
Input parameters in the absorption chiller panel for determining evaporator internal temperature
*******************************************************************************************************************"
DeltaT_min = 4,1
DeltaT_var = 4,1
Qe_max = 18,6 [kW]

"*******************************************************************************************************************
Internal and external water circuit temperature relation
*******************************************************************************************************************"
t[10]= T[20]+1 "Condenser"
t[1] = t[10] "Condenser and Absorber internal temperature is assumed the same"
T[13] = ((T[21]+T[22])/2) - ((DeltaT_min+DeltaT_var)*Qe/Qe_max) "Evaporator"
T[4] = T[16] -1 "Generator"
T_re-cooling_out = (T[20]+T[18])/2

"*******************************************************************************************************************
Mass Fraction of ammonia coming out from the generator and mass flow of strong solution
*******************************************************************************************************************"
x[9]=0,999634
m[1]= 0,170

"*******************************************************************************************************************
Gas quality (-0,001 = subcooled; 0 - 1 = saturated; 1,001 = superheated
*******************************************************************************************************************"
Q[13]=0,998
Q[10]=0,00
Q[9]=1,001
Q[1]=0
Q[4]=0

"*******************************************************************************************************************
External water circuit heat capacity definition and mass flow of extenal water circuit
*******************************************************************************************************************"
P_ext = 1 {bar}
cp_hot=Cp(Water;P=P_ext;x=0)
cp_re-cooling=Cp(Water;P=P_ext;x=0)
cp_cold=Cp(Water;P=P_ext;x=0)
m_Re-cooling = 1,05
-26-

{m_Hot = 0,408} "This variable is varied"
{m_Cold = 0,645} "This variable is varied"

"*******************************************************************************************************************
Heat transfer definition by UA values and LMTD
*******************************************************************************************************************
Qa = UAa*LMTDa
Qc = UAc*LMTDc
Qg = UAg*LMTDg
Qe = UAe*LMTDe

"*******************************************************************************************************************
LMTD Definition
*******************************************************************************************************************"
LMTDa=(T[6]-T[18]-T[1]+T[17])/ln(Arga)
Arga = (T[6]-T[18])/(Downa)
Downa = (T[1]-T[17])

LMTDc = (T[10] - T[20] ) - (T[10] - T[19] )/ ln (Argc)
Argc = (T[10] - T[20] )/Downc
Downc = (T[10] - T[19] )

LMTDe = (T[21] - T[13] ) - (T[22] - T[13] )/ ln (Arge)
Arge = (T[21] - T[13] )/Downe
Downe = (T[22] - T[13] )

LMTDg = (T[15] - T[4] ) - (T[16] - T[9] )/ ln (Argg)
Argg = (T[15] - T[4] )/Downg
Downg = (T[16] - T[9] )

"*******************************************************************************************************************
Sensible heat transfer definition and external water circuit temperature difference definition
*******************************************************************************************************************"
Qa =m_Re-cooling*Cp_re-cooling*(T[18] - T[17])
Qc = m_Re-cooling*Cp_re-cooling*(T[20] - T[19])
Qg = m_Hot*Cp_hot*(T[15] - T[16])
Qe = m_Cold*Cp_cold*(T[21] - T[22])

delta_T_evaporator = (T[21] - T[22])
delta_T_condenser = (T[20] - T[19])
delta_T_generator = (T[15] - T[16])
delta_T_absorber = (T[18] - T[17])
delta_T_re-cooling = (Tmix_re-cooling - T[17])

"*******************************************************************************************************************
Temperature inputs in each component
*******************************************************************************************************************"
"Evaporator"
T[21] = 12
T[22] = 6

"Generator"
T[15] = 70
{T[16] = 60}

"Condenser"
T[19] = 22
{T[20] = T[19] + 4}

"Absorber"
-2&-

T[17] = T[19]

Tmix_re-cooling = (T[20] + T[18]) / 2 "Parallel connection between absorber and condenser external
water line"

"*******************************************************************************************************************
Main equations in each components (Mass and energy balance, work of pump, enthalpy definition and
overall equations)
*******************************************************************************************************************"
"ABSORBER"
m[14]+m[6]=m[1]
m[14]*x[14]+m[6]*x[6]=m[1]*x[1]
(m[14]*h[14]+m[6]*h[6] - m[1]*h[1])=Qa

CALL pxq(P[3];x[3];Q3s:T3f;h3f;v3f;s3f)
Q3s = 0
T[9] = T3f "Definition for the T[9] from the weak solution"

"DESORBER"
balm=m[3]-(m[9]+m[4])
bala=m[3]*x[3]-(m[9]*x[9]+m[4]*x[4])
Qg=(m[9]*h[9]+m[4]*h[4] - m[3]*h[3])

"CONDENSER"
Qc=m[9]*(h[9]-h[10])

"EVAPORATOR"
Qe=m[13]*(h[13]-h[12])

"PUMP"
swp=v[1]*(p[2]-p[1])/etap
h[2]=h[1]+swp
Wp=m[1]*(h[2]-h[1])

"EXPANSION VALVE"
h[5]=h[6]

"SHX"
CALL tpx(T[2];P[5];x[5]:Q5e;h5e;v5e;s5e)
CALL tpx(T[4];P[3];x[3]:Q3e;h3e;v3e;s3e)
CALL shx(eshx;m[2];m[4];h[2];h[4];h5e;h3e:h[5];h[3];Qshx)

"EXPANSION VALVE"
h[11]=h[12]
"C-E HX"
CALL tpx(t[10];P[14];x[14]:Q14e;h14e;v14e;s14e)
CALL tpx(t[13];P[11];x[11]:Q11e;h11e;v11e;s11e)
CALL shx(esc;m[13];m[10];h[13];h[10];h11e;h14e:h[11];h[14];QCE)

"OVERALL"
COP=Qe/(Wp+Qg)
checkQ=Qg+Qe+Wp-(Qc+Qa)
pratio=phigh/plow
"*******************************************************************************************************************
Pressure definition as high or low
*******************************************************************************************************************"
p[1]=plow
p[2]=phigh
P[3]=phigh
p[4]=phigh
-27-

P[5]=phigh
P[6]=plow
p[9]=phigh
p[10]=phigh
p[11]=plow
p[12]=plow
P[13]=plow
P[14]=plow

"*******************************************************************************************************************
Trivial mass balances
*******************************************************************************************************************"
m[1]=m[2]
m[2]=m[3]
m[4]=m[5]
m[5]=m[6]
m[9]=m[10]
m[10]=m[11]
m[11]=m[12]
m[12]=m[13]
m[13]=m[14]

"*******************************************************************************************************************
Trivial NH3 balances
*******************************************************************************************************************"
x[1]=x[2]
x[2]=x[3]
x[4]=x[5]
x[5]=x[6]
x[9]=x[10]
x[10]=x[11]
x[11]=x[12]
x[12]=x[13]
x[13]=x[14]

"*******************************************************************************************************************
State points definition (Input 3 variables)
*******************************************************************************************************************"
CALL tpq(t[1];p[1];Q[1]:x[1];h[1];v[1];s[1])
CALL phx(p[2];h[2];x[2]:T[2];Q[2];v[2];s[2])
CALL phx(P[3];h[3];x[3]:T[3];Q[3];v[3];s[3])
CALL tpq(t[4];p[4];Q[4]:x[4];h[4];v4;s4)
CALL phx(P[5];h[5];x[5]:T[5];Q[5];v[5];s[5])
CALL phx(P[6];h[6];x[6]:T[6];Q[6];v[6];s[6])
CALL txq(t[9];x[9];Q[9]:p9;h[9];v[9];s[9])
CALL txq(t[10];x[10];Q[10]:p[10];h[10];v[10];s[10])
CALL phx(p[11];h[11];x[11]:T[11];Q[11];v[11];s[11])
CALL phx(p[12];h[12];x[12]:T[12] ;q[12];v[12];s[12])
CALL txq(t[13];x[13];q[13]:p[13];h[13];v[13];s[13])
CALL phx(P[14];h[14];x[14]:T[14];Q[14];v[14];s[14])
-28-

A66en)i; @? %o&inal o6eration con)ition )ata b2 Bin'
G&b, /Bin' G&b,C 2010a0



-6'-