Process Integration and Optimization for Sustainability (2020) 4:391–408

https://doi.org/10.1007/s41660-020-00128-5

ORIGINAL RESEARCH PAPER

Effects of Temperature Differences in Optimization of Spiral Plate

Heat Exchangers
A. H. Sabouri Shirazi 1 & M. R. Jafari Nasr 2 & M. Ghodrat 3

Received: 2 April 2020 / Revised: 20 June 2020 / Accepted: 14 July 2020 / Published online: 24 August 2020
# Springer Nature Singapore Pte Ltd. 2020

Abstract
In this research, three aspects of modeling, analyzing, and optimizing spiral plate heat exchangers (SPHEs) are studied. The main
objective of this work is to pave the way for comparing manufacturers’ designed SPHEs with theoretical designed SPHEs
without involving designers in using computational methods. To begin, with assumption of constant overall heat transfer
coefficient and specific heat capacities, a mathematical modeling of SPHE based on energy balance equations is developed to
model the SPHE as a network of series-connected equivalent internal heat exchangers to determine the temperature distribution in
spiral turns. This modeling can facilitate the usage of temperature-enthalpy diagram in SPHEs’ analysis and design. Furthermore,
a new algorithm for thermal design optimization of SPHEs has been proposed. The proposed algorithm is based on maximizing
pressure drops at channels, considering geometric proportion of SPHE and minimizing the total cost simultaneously. To show the
proposed method applicability in analyzing thermal and hydraulic design parameters, a single-phase counter-current SPHE is
assessed and optimized for different design cases with temperature approach variations. Results of comparing manufacturers’/
standard designed SPHEs and research/theoretical designed SPHEs by defining appropriate geometric proportion ranges con-
firmed that temperature approach variations can improve SPHE performance to a higher extent, such as finding temperature
approach ranges for optimized SPHEs with higher compactness to reduce the manufacturing cost. This fact is revealed by
introducing compactness-temperature approach diagram which depicts the geometric optimization of SPHEs and the effects of
temperature differences in SPHE’s optimization.

Keywords Spiral plate heat exchanger . Thermal design optimization . Heat exchangers network . Compactness-temperature
approach diagram . Geometric proportion . Cost estimation

Introduction exchanger (SPHE) has special features to extend its applica-

tions in industry since Rosenblad (1935, 1936) has patented
Heat exchangers are pieces of equipment used widely for ef- this revolutionary type of apparatus. Having high thermal ef-
fective heat transfer between two or more fluids (liquid or gas) fectiveness and low fouling tendency provides an appropriate
in various industries such as power plants and process plants. context to utilize SPHEs in difficult and non-difficult services
Among the various forms of heat exchangers, compact heat ranging from slurry to clean fluids (Shah and Sekulic 2003).
exchangers have been at the centerpiece of recent research by A SPHE is fabricated by rolling a pair of relatively long
virtue of having high performance (Energy Efficiency Office strips of plate with welded spacer studs to form a pair of
2000). As a compact heat exchanger, spiral plate heat uniform spiral passages. By virtue of specific geometry of
channels and flow arrangements in spiral passages, counter-
current flow type of this heat exchanger has good flow distri-
* M. Ghodrat bution, high turbulence at low fluid velocities, and noticeable
m.ghodrat@adfa.edu.au
heat transfer (Kuppan 2013).
1
Mechanical Department, Petrochemical Industries Design and
To review the conducted research on SPHEs, research
Engineering Company (PIDEC), Shiraz, Iran fields can be categorized as follows: (1) designing of
2
Research Institute of Petroleum Industry (RIPI), Tehran, Iran
SPHEs, (2) performance checking and studying temperature
3
distribution in SPHEs, and (3) optimization of SPHEs. These
School of Engineering and Information Technology, UNSW
Canberra, Canberra, ACT 2612, Australia
researches have been carried out by experimental methods,
392 Process Integr Optim Sustain (2020) 4:391–408

analytical methods, numerical methods, or combination of distribution and introduced a dimensionless criterion number
these methods. for performance calculating. Strelow (2000) presented a gen-
Some researchers tried to determine the governing heat eral calculation method for plate heat exchangers mathemati-
transfer and pressure drop equations in SPHEs. The first cally to determine exact temperature profiles and heat flux
comprehensive research on SPHEs was carried out by along the walls of plates without any iteration. In the last
Minton (1970) who reported a set of empirical correlations two decades, Burmeister (2006) and Nguyen and San
of heat transfer coefficient and pressure drop for SPHEs. (2016b) studied analytically the performance of SPHEs. In
Later, Martin (1992) numerically studied the heat transfer other words, most researchers have tended to carry out their
and pressure drop characteristics of a spiral plate heat ex- studies numerically or experimentally in recent years. The first
changer and reported empirical correlations. Rajavel and noticeable numerical research was published by Devois et al.
Saravanan (2008a, 2008b), Saravanan and Rajavel (2008a, (1995). They modeled a two-dimensional SPHE and studied
2008b), and Rajavel (2014) investigated the Martin’s its performance in both steady state and transient state with
Nusselt number expression and proposed a new Nusselt num- different materials. In the last decade, Garcia and Moreles
ber for SPHEs. Other researchers made an attempt to deter- (2012) proposed a numerical method for rating thermal
mine the size and dimensions of SPHEs based on suggested performance and approximated the temperature distribution
correlations. One of the pioneers was Jarzębski (1984) who and overall heat transfer coefficients in SPHEs. Nguyen and
presented simple expressions for calculating approximate di- San (2011) numerically investigated the effects of spiral turns
mensions of SPHEs to minimize the annual cost of heat numbers in the heat transfer performance of a SPHE. Also,
surface and required pumping energy. Later, Dongwu they (2015, 2016a) studied the effects of wall material in NTU
(2003) carried out a thorough research on the geometry of and heat transfer effectiveness of a SPHE. Some other re-
SPHEs and developed the expressions of calculating main searchers made experimental studies and validated the results
geometric factors such as spiral diameter, number of turns, with numerical methods and vice versa. The major policy
and length of plate rounds based on the semicircle model. hired is to study the performance and temperature distribution
Picon Nunez et al. (2006, 2007, 2009) proposed an alternative of SPHEs with different fluids by variations of flow rates or
method to size SPHEs founded on maximizing pressure drops changes in inlet and outlet temperatures. Some important stud-
in channels by varying geometric parameters to minimize di- ies in this realm are Sakariya et al. (2014), Shabiulla and
mensions. This concept was firstly put forth by Polley et al. Sivaprakasam (2014), Khorshidi and Heidari (2016), Manoj
(1990, 1991) who studied the pressure drop considerations in et al. (2016), Kathir Kaman et al. (2017), Kumar et al. (2018),
the retrofit of heat exchangers network and design of compact Tapre and Kaware (2018a, 2018b, 2018c), Metta et al. (2018),
heat exchangers. Later, Guha and Unde (2014) developed a Anil et al. (2018), Memon et al. (2019), to name a few.
mathematical model of SPHEs based on Shah-London empir- Finally, besides researchers worked on design of SPHEs,
ical equation for Nusselt number of flow through a duct and some others implemented optimization methods to find opti-
put forward a straight formula for optimum design of SPHEs mum SPHEs with higher performance, with higher pressure
regarding physical parameters such as channel plate width and drops, and/or with lower cost. Among recent research, follow-
thickness. Last year, Dávalos et al. (2019) worked on Picon ing studies are good samples to show the general idea behind
Nunez studies and presented a procedure, validated by CFD, their attempts. Bidabadi et al. (2013), Vasconcelos Segundo
to design cross-flow SPHEs as a function of the fluid arrange- et al. (2018) and Kumar et al. (2019) applied genetic algo-
ments and government flow by use of Minton’s thermal and rithms, wind-driven optimization method and gradient based
hydraulic performance correlations. search algorithm respectively to increase the overall heat
The second group of researchers worked on performance transfer coefficient and reduce the total cost of SPHEs. A
checking and temperature distribution along the walls of new research by Milovančević et al. (2019) was carried out
plates in SPHEs. A wide range of assumptions were utilized based on entropy generation to study the thermoeconomic
to simplify the mathematical model to derive correlations in analysis of SPHEs. They found that optimal design of
analytical or numerical studies. More or less, these simplify- SPHEs should be calculated based on inlet and outlet temper-
ing assumptions were fewer in experimental studies. The pio- atures and process fluid flow rate.
neer team in analytical and numerical studies on performance With respect to scrutinizing most previous research, it is
and distribution of temperature of SPHEs was Chowdhury worthwhile to mention that none of them does unify the
et al. (1985). Bes and Roetzel (1991, 1992, 1993) developed criteria for comparing a wide range of SPHEs designed based
an analytical solution based on effectiveness, NTU, and on geometric parameters. This study is an effort to show that
number of spiral turns with assumption of constant overall besides thermal, hydraulic, and economic parameters, adding
heat transfer coefficients and heat capacities to describe the a geometric proportion parameter is essential to make design
thermal behavior of SPHEs. Also, they studied the optimization more practical in comparison with available stan-
temperature changes founded on heat flux density dard designed SPHEs by manufacturers. It is crystal clear that
Process Integr Optim Sustain (2020) 4:391–408 393

heat exchanger manufacturers are willing to develop an opti-

mized design for mass production of SPHEs to reduce the
manufacturing end prices and to increase the compactness.
One way to consider in achieving higher compactness is to
add a dimensional parameter. This new criterion is the geo-
metric proportion of channel plate width to outside diameter
of SPHE that indicates the compactness of this type of com-
pact heat exchanger intrinsically. Therefore, a new algorithm
for optimization of SPHEs based on maximizing pressure
drops in channels, keeping geometric proportion and minimiz-
ing total cost estimation has been developed in this study to
pave the way for comparing theoretical designed SPHEs with
standard designed ones. In addition to introducing a new op-
timization algorithm, a mathematical model for analyzing
SPHE has been proposed based on energy balance equations
with assumption of constant overall heat transfer coefficient
and specific heat capacities. The aim of the proposed algo-
rithm and modeling is mainly to determine the temperature
Fig. 1 Geometrical features and flow pattern in a SPHE
distribution in spiral turns. Being consistent with proposed
simplified model, temperature-enthalpy diagram is used as a
beneficial tool to facilitate the analysis. Furthermore, temper- where U is the overall heat transfer coefficient, A is the heat
ature approach, difference between outlet temperatures of hot transfer area, FT is the temperature difference correction fac-
side and cold side streams, is set as a key parameter for the tor, and ΔTLM is the logarithmic mean temperature difference.
comparison of thermal and hydraulic design parameters and The overall heat transfer coefficient and heat transfer area are
compactness. At last, a single-phase counter-current SPHE derived from Eqs. 3 and 4 respectively. By virtue of self-
case study has been studied to show the applicability of new cleaning mechanism in channels of SPHEs, it is noticeable
method in standard designed and theoretical designed SPHEs. that fouling factor resistance can be ignorable in Eq. 3.
Moreover, a wide range of performance and geometry param- 1
eters were checked to find optimum ranges of temperature U¼ ð3Þ
1=hh þ p=kw þ 1=hc þ Rfouling
approach for designing optimal SPHEs and show the effects
of temperature differences in optimization of SPHEs. A ¼ 2LH ð4Þ

where h is the heat transfer coefficient, p is the plate thick-

ness, kw is the thermal conductivity of wall plate material,
Fundamentals of Spiral Plate Heat Exchangers Rfouling is the fouling factor resistance, L is the plate length,
and H is the channel plate width.
The same as general governing thermodynamics laws and heat Logarithmic mean temperature difference, ΔTLM, can be
transfer equations in heat exchangers, SPHEs’ governing calculated from Eq. 5, where ΔTmax and ΔTmin are the greater
equations can be developed accurately according to their ge- terminal temperature difference and smaller terminal temper-
ometries. Figure 1 shows general view of geometrical features ature difference for both pure parallel-flow and counter-
and flow pattern in a sample SPHE. current arrangements.
General heat transfer rate between two streams can be cal-
culated by Eq. 1. ΔTmax ΔTmin
ΔTLM ¼ ð5Þ


ln ðΔTmax =ΔTmin Þ
Q ¼ Mh cp;h Th;o Th;i Þ ¼ Mc cp;c Tc;o Tc;i Þ ð1Þ
For flow arrangements other than counter-current arrange-
where Q is the heat transfer rate, M is the mass flow rate, cp ment, ΔTLM has to be corrected by FT. Chowdhury et al.
is the specific heat capacity, and T is the temperature. (1985) presented a straight expression with respect to heat
Subscribes of c and h show cold stream and hot stream, and capacity rate ratios for the determination of the correction
also subscribes of i and o indicate inlet and outlet respectively. factor of SPHEs by Eqs. 6 and 7.
Moreover, Eq. 2 shows the heat transfer rate from a surface  
between two streams of hot and cold in these exchangers N N
FT ¼ tanh C ¼ 1 ð6Þ
NTU NTU
Q ¼ UAFT ΔTLM ð2Þ
394 Process Integr Optim Sustain (2020) 4:391–408
 
1þC where b is the channel plate spacing. Also, the fluid mean
ln 1 þ
1=ε–1 ð7Þ velocity at channel, Vf, and Prandtl number are calculated by
FT ¼ –1 < C ≤0
NTU Eqs. 16 and 17 where ρ is the fluid density and k is the thermal
ð1 þ CÞ
N conductivity of fluid.


Cmin Mcp min
C¼ ¼
 ð8Þ Vf ¼
M
ð16Þ
Cmax Mcp max ρAc
where C (with negative sign for counter-current flow) is the cp μ
Pr ¼ ð17Þ
heat capacity rate ratio and N is the number of spiral turns of k
the stream, based on 2π arc as a circle. Equations 9 and 10 Empirical Eqs. 18, 19, and 20 suggested by Minton are
express the number of heat transfer units, NTU, and thermal dimensional and are applicable for spiral flow with no phase
effectiveness per turn, ε. change in SPHEs.
UA   
NTU ¼ ð9Þ Dh
h ¼ 0:023 1 þ 3:54 cp V f Re–0:2 Pr–2=3 Liquid; Re > Recr
Cmin Ds
1–e–ð1þjCjÞ N
NTU
ð18Þ
ε¼ ð10Þ   
NTU Dh
ð1 þ jCjÞ h ¼ 0:0144 1 þ 3:54 cp V0:8 –0:2
Gas; Re > Recr
N f Dh
Ds
In addition, the total effectiveness of the SPHE can be ð19Þ
calculated from the general definition of heat exchanger ther-  –1=3  –0:14
L μw
mal effectiveness, εtot, as per Eq. 11 (Kuppan 2013). h ¼ 1:86cp V f Re–2=3 Pr–2=3 Liquid; Re < Recr
Dh μb
Q Q ð20Þ
εtot ¼ ¼
ð11Þ
Qmax Cmin Th;i Tc;i Þ
where μb is the fluid bulk viscosity and μw is the fluid
viscosity at the wall temperature. Besides, Minton has approx-
imated the SPHEs outside diameters by Eqs. 21 and 22.
q
ffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffi
Ds ¼ 15:36  Lðbc þ bh þ 2pÞ þ D2i Imperial System ð21Þ
Governing Heat Transfer and Pressure Drop
q
ffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffi
Relationships Ds ¼ 1:28  Lðbc þ bh þ 2pÞ þ D2i Metric System ð22Þ

Minton (1970) has proposed empirical equations in imperial where Di is the core or inside diameter of a SPHE.
system thermal and hydraulic performance of SPHEs to cap- Dongwu (2003) presented a theoretical formula to calculate
ture the flow regimes and flow phases based on average plate the SPHEs outside diameters based on number of spiral turns,
curvature. The major parameters to determine the flow re- N. Equations 23 and 24 calculate the number of spiral turns
gimes are the values of the Reynolds number, Re, and the and the SPHE outside diameter.
critical Reynolds number, Recr, given by Eqs. 12 and 13 re-
  rffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffi
spectively. bh −bc
2 4L
− Di þ þ Di þ bh −b c
þ ðbh þ bc þ 2pÞ
2 2 π
Dh M N¼ ð23Þ
Re ¼ ð12Þ ðbh þ bc þ 2pÞ
μAc
 0:32 Ds ¼ Di þ ðbh þ pÞ þ Nðbh þ bc þ 2pÞ ð24Þ
Dh
Recr ¼ 20000 ð13Þ In addition, Minton suggested following dimensional em-
Ds
pirical expressions for pressure drops in channels through Eqs.
where Dh is the hydraulic diameter, μ is the viscosity, Ac is 25 to 27.
the free flow area, and Ds is the SPHE outer diameter.  2 "   1=3 #
Equations 14 and 15 calculate the hydraulic diameter and L M 1:3μ1=3 H 16
ΔP ¼ þ 1:5 þ Re > Recr
the free flow area of the channels in a SPHE. 1000s Ac b þ 0:125 M L

ð25Þ
2bH
Dh ¼ ð14Þ  "     #
bþH L M 1:035μ 0:5
μw 0:17 H 0:5 16
ΔP ¼ þ 1:5 þ 100 < Re < Recr
1000s Ac b þ 0:125 μb M L
Ac ¼ Hb ð15Þ
ð26Þ
Process Integr Optim Sustain (2020) 4:391–408 395
  0:17  
Lsμ μw M analysis, some assumptions were made: (I) The overall heat
ΔP ¼ Re < 100 ð27Þ
3385b2:15 μb H transfer coefficient was constant along the flow passage. (II)
Both specific heat capacities were constant along the flow
where ΔP is the fluid pressure drop and s is the relative passage. (III) The central partition between the hot and cold
density (relative to water at 20 °C). flows was insulated. (IV) The outside surface of the SPHE
Moreover, for a Reynolds number in the range of 400 to was thermally insulated and there was no heat losses to the
30,000, Martin (1992) presented a Nusselt number for SPHEs environment. (V) Conduction in the direction of flow can be
shown in Eq. 28. neglected. (VI) Fluid temperature is uniform in the channel
hDh cross section perpendicular to the flow direction. (VII) Fluid
Nu ¼ ¼ 0:04Re0:74 Pr0:4 ð28Þ properties and plate thermal conductivity were constants
k
along the flow passage. In the light of mentioned assumptions,
This expression shows that the heat transfer coefficients of the mathematical modeling is carried out as follows in five
channels can be estimated without dependency upon the steps.
length of channel plate. In the first step, the SPHE is cutting and unrolling from
illustrated cutting line in Fig. 2 to find an unwound sketch of
the SPHE. Figure 3 shows the unwound sketch of the counter-
current flows SPHE. This figure depicts that the SPHE is
Mathematical Modeling of a SPHE
divided into numbers of streams (cold and hot) as many as
number of spiral turns.
A mathematical model was developed based on mass and
The second step is to determine the heat transferring
energy conservation to analyze temperature distributions of
streams and heat transfer areas. Based on mentioned assump-
the flows in the SPHE. Figure 2 depicts the flow arrangement
tions, there is no heat transfer in entrance and exit parts of this
and the structure of a single-phase counter-current flow
heat exchanger to other streams (from A0) or environment
SPHE. This SPHE is used to show the mathematical modeling
(from AN). The innermost and outermost streams transfer the
in this study.
heat to inner streams only, and the intermediate streams trans-
In mathematical model of the SPHE, the heat conduction
fer the heat to both side streams. According to heat transfer
and convection flows are in the direction normal to the plate.
interaction between streams, the heat transfer areas between
Heat is transferred only through one wall for innermost and
streams can be specified easily. In Fig. 3, A1, Aj, and AN-1
peripheral channel and for internal channels heat is exchanged
show the required heat transfer areas between inner, interme-
through the bilateral sides of the channels. To simplify the
diate, and outer streams respectively.

Fig. 3 Unwound sketch and streams modeling of the counter-current

Fig. 2 Structure and flow arrangement in a counter-current flow SPHE flow SPHE
396 Process Integr Optim Sustain (2020) 4:391–408

The third step is to determine the internal heat transfer rates, capacities, and overall heat transfer coefficient are the
Qj, between streams. Referring to Fig. 3, each stream may same as those of the SPHE. The overall heat transfer is
interact with one side or both side streams. Streams interacted calculated based on a new algorithm of thermal design
with both sides are divided into two sub-streams with two heat optimization of SPHEs by considering maximizing pres-
transfer rates. For instance, the stream between Aj and Aj+1 sure drops in channels and geometric proportion of SPHE
transfers the heat to both sides with heat transfer rates of Qj developed in this study.
and Qj+1. To find the heat transfer rates of streams and sub-
streams, it is important to determine the internal temperatures
at the boundaries (beginning and end) of each stream and sub- Temperature-Enthalpy Diagram of a SPHE
stream. These internal temperatures are also illustrated in Fig.
3 at the boundaries of streams and sub-streams.
Chen et al. (2012) introduced temperature-enthalpy (T-Q) di-
The next step is to derive the governing energy balance agram as a two-dimensional property diagram to obtain the
equation of each heat transfer rate. Thus, each heat transfer influences of heat exchanger area, heat capacity rate, and flow
rate between divided hot and cold streams is modeled as a heat
arrangement on the heat transfer performance during the anal-
transfer rate in an equivalent internal heat exchanger by con- ysis of heat exchangers and their networks. Temperature-
sidering inlet and outlet temperatures of divided streams as the enthalpy diagram formed based on the energy balance equa-
temperature boundaries. This modeling breaks the SPHE into
tion, Q = McpΔT, and it paved the way of analyzing a heat
series of equivalent internal heat exchangers. Figure 4 shows exchangers network more easily. In this study, the mathemat-
the modeling of heat transfer rates of Qj and Qj+1 by two ical modeling of the SPHE shows that the SPHE can be bro-
blocks of equivalent heat exchanger in series.
ken into a series of equivalent internal heat exchangers. That
Equations 29 and 30 and Eqs. 31 and 32 show the is, the SPHE is modeled as a heat exchangers network con-
governing energy balance equations for both equivalent heat nected in series. Moreover, analyzing the T-Q diagram in the
exchangers. All governing energy balance equations of all
SPHE can determine the internal temperature boundaries of
divided streams have to extract the same way to complete each equivalent internal heat exchanger, which in turn deter-
the mathematical modeling.
mines the temperature distribution along the flow passage of


 SPHE.
Mh cp;h T2 j−2 −T2 j−1 þ Mc cp;c T2 j −T2 j−3 ¼ 0 ð29Þ
Figure 5 shows not only the temperature-enthalpy (T-Q)




 T2 j−2 −T2 j−3 − T2 j−1 −T2 j diagram of a single-phase counter-current flow SPHE and its
Mh cp;h T2 j−2 −T2 j−1 þ UA j   ¼ 0 ð30Þ
T2 j−2 −T2 j−3 equivalent internal heat exchangers, but also the heat transfer
ln
T2 j−1 −T2 j rates of sub-streams and their temperature boundaries. In ad-


 dition, Fig. 5 reveals the relations between all governing en-
Mh cp;h T2 j−1 −T2 jþ2 þ Mc cp;c T2 jþ1 −T2 j ¼ 0 ð31Þ ergy balance equations of all divided streams sequence of
solving these equations regarding temperature boundaries.




 T2 j−1 −T2 j − T2 jþ2 −T2 j−1 It is worth mentioning that 1/(Mcp) is the slope of hot or
Mc cp;c T2 jþ1 −T2 j −UA jþ1   ¼0
T2 j−1 −T2 j cold flow in T-Q diagram according to rearrangement in heat
ln
T2 jþ2 −T2 j−1
ð32Þ

The fifth step is to solve the energy balance equations

and find unknown parameters based on aforementioned
assumptions. The unknown parameters are temperature
boundaries of sub-streams and heat transfer areas related
to spiral turns. The mass flow rates, specific heat

Fig. 4 Modeling of internal heat transfer rates of sub-streams in the Fig. 5 Temperature-enthalpy (T-Q) diagram of modeled single-phase
counter-current flow SPHE with series of equivalent heat exchangers counter-current flow SPHE
Process Integr Optim Sustain (2020) 4:391–408 397

transfer energy balance (ΔT = Q/Mcp). Indeed, these slopes

ny Ce Hw ðMh ΔPh =ρh þ Mc ΔPc =ρc Þ=η
Cod ¼ ∑ ð35Þ
indicate temperature difference, a heat transfer driving force, n ð1 þ i Þn
between hot and cold fluid. To study the temperature differ-
where Ce is the energy cost (Ce = 0.15 USD/KW hr), η is
ences due to variations of mass flow rates and/or specific heat
the pumping efficiency (η = 0.75), Hw is the amount of hours
capacities, three possible cases take place in temperature dis-
of work (Hw = 8000 h/yr), i is the annual discount rate (i =
tributions of a single-phase counter-current flow SPHE.
0.1), and ny is the equipment life (ny = 15 yr).
Figure 6 shows these possible cases of temperature distribu-
tions, and graphical definitions of normal temperature, tem-
perature meet, and temperature cross distributions.
Temperature approach defined as the temperature difference
Algorithm of Thermal Design Optimization
between outlet temperatures of hot side and cold side streams
of SPHEs
is set as a key parameter of study the three mentioned temper-
ature distributions. That is, temperature approach is positive (a
Having a brief review on previous available design methods
positive driving force), zero (a zero driving force), and nega-
of SPHEs (by Minton (1970), Picon Nunez et al. (2006, 2007,
tive (a negative driving force) in normal temperature, temper-
2009), Guha and Unde (2014), Dávalos et al. (2019)) or opti-
ature meet, and temperature cross distributions respectively.
mization algorithms (by Bidabadi et al. (2013), Vasconcelos
Segundo et al. (2018) and Kumar et al. (2019)), researchers
attempted to find new optimum SPHEs with higher overall
heat transfer coefficient and/or lower cost. Except Minton’s
Cost Estimation Methodology empirical-based procedure applicable for standard SPHEs,
manufacturers have no tendency to apply suggested methods
The total cost estimation of each heat exchanger is composed
because lack of a geometric criterion prevents them from de-
of capital cost at purchasing time and operating cost during
veloping and tabulating mass production of SPHEs to reduce
lifetime of the heat exchanger. Equation 33 shows the cost
prices (capital cost) and increase the compactness. Although
estimation of a SPHE.
Minton (1970) tried to find the best design of SPHE with
Ctot ¼ Ci þ Cod ð33Þ approximately equal channel plate width and outside
diameter, or Picon Nunez et al. (2006, 2007, 2009) or
where Ctot, Ci, and Cod indicate the total, capital, and oper- Dávalos et al. (2019) made attempts to maximize pressure
ational costs of a SPHE based on USD respectively. drops at channels by change of geometry parameters, none
To estimate the capital cost of SPHEs, Hall et al. (1990) of their methods considered both targets simultaneously to
suggested Eq. 34 based on heat transfer area for stainless steel make the design procedure more applicable in industries.
SPHEs. One way to include the geometric aspect ratio and compact-
ness in the optimization procedure of SPHEs is to define and
Ci ¼ 19687A0:59 ð34Þ
add a geometric proportion criterion. Moreover, this geomet-
Estimation of operational costs, the second term in Eq. 33, ric proportion will give the permission to manufacturers to
depends on the costs involving overcoming the pressure loss compare available manufacturers’/standard designed SPHEs
caused by friction, evaluated using Eq. 35. with research/theoretical designed SPHEs. The geometric pro-
portion, GP, is defined as the ratio of channel plate width to

Fig. 6 a Principles of temperature

distributions in a single-phase
counter-current flow SPHE. b
Principles of temperature ap-
proach in a single-phase counter-
current flow SPHE
398 Process Integr Optim Sustain (2020) 4:391–408

outside diameter of SPHE expressed in Eq. 36. That is, this

dimensionless parameter indicates the compactness feature of
this type of compact heat exchanger intrinsically. Appropriate
value of GP occurs when the channel plate width equals to
outside diameter of SPHE.
H
GP ¼ ð36Þ
Ds

Also, the compactness factor or the surface density, β, can

be defined as the ratio of the heat transfer area to total volume
of the SPHE by Eq. 37.
A
β¼ ð37Þ
V
To describe the new algorithm generally, two major con-
cepts are deemed. The first concept is to maximize pressure
drops at both channels to have higher heat transfer coefficient,
and the second one is to find SPHEs with approximately equal
channel plate width and outside diameter of SPHE. Figure 7
shows clear and concise flow chart of the algorithm.
The first step of algorithm is to select an initial geometry,
flow arrangement and assume the overall heat transfer coeffi-
cient based on specified design data. The next step is to cal-
culate heat transfer area and check the pressure drops with
allowable pressure drops. Two main geometric parameters,
channel plate spacing and channel plate width, play significant
role in maximizing pressure drops. At third step, by changing
channel plate spacing, pressure drops are raised to allowable
pressure drops. The next stage is to check preliminary de-
signed SPHE for acceptable geometric proportion. In this al-
gorithm, a range of GP is defined with ± 2% deviation from
targeted GP (0.98 ≤ GP ≤ 1.02). It is noticeable that selecting
the GP range depends on designer consideration. The selected
range of GP contributes to finding cheaper designed SPHE by
designer. By variation of channel plate width, specified range Fig. 7 Flow chart of new thermal design optimization algorithm and
of GP is checked to obtain minimum total cost in the fifth step. temperature distribution of the SPHE
In the last step, optimized SPHE is mathematically modeled
with a network of heat exchangers connected in series to find The suggested algorithm was implemented on a excel
temperature distribution of streams and temperature-enthalpy sheet. It is remarkable that the study divided into two stages.
diagram of the SPHE. In the first stage, to reveal the importance of geometric
proportion and maximizing pressure drops at both channels,
the results of optimization SPHEs based on new algorithm
were compared with suggested methods by Minton (1970)
Case Study and Picon Nunez et al. (2006, 2007, 2009). In the next stage,
a comprehensive investigation has been carried out on two
The new algorithm is demonstrated using the case study re- aspects with constant cold side inlet and outlet temperature:
ported by Minton (1970) for a single-phase counter-current (1) changes of inlet and outlet temperatures of hot stream
flow SPHE. The operating conditions and average physical provided that cp,h is constant and (2) changes of cp,h with
properties for the case study are shown in Table 1 and constant outlet temperature and variable inlet temperature of
correspond to a light organic compound. In addition, Minton hot stream. The following parameters based on temperature
(1970) suggested a geometry standard for SPHEs, Table 2, to approach variations are checked in the second stage to show
standardize the manufactures’ products. the comparison of standard designed SPHEs and optimized
Process Integr Optim Sustain (2020) 4:391–408 399

Table 1 Case study operating conditions (Minton 1970) Results and Discussion
Conditions Hot side Cold side
In the first part of this section, the application of the suggested
Mass flow rate (kg/s) 0.7843 0.7465 algorithm to optimize SPHEs is demonstrated. As mentioned
Molecular weight 200.4 200.4 above, designed SPHE is compared with previous design
Specific heat capacity (J/kg K) 2972.628 2763.288 methods to reveal the importance of geometric proportion
Thermal conductivity of fluid (W/m K) 0.34789 0.32339 and maximizing pressure drops at both channels. Table 3
Inlet temperature (°C) 200 60 shows the results of various designs of case study by Minton
Outlet temperature (°C) 120 150.4 (1970), Picon Nunez et al. (2006, 2007, 2009), and the new
Allowable pressure drop (KPa) 6.8948 6.8948 algorithm.
Density (kg/m3) 843 843 Comparing the results from different methods of design in
Viscosity (kg/m s) 0.00335 0.008 Table 3 shows that considering geometric proportion and
Material of construction Stainless steel maximizing pressure drops not only has noticeable effects
Thermal conductivity of material of construction 17.3073 on increasing the Reynolds numbers in channels and overall
(W/m K) heat transfer coefficient, but also decreases the heat transfer
area remarkably. To analyze the results of new algorithm more
in detail, the increment in Reynolds numbers is a result of
SPHEs in both aspects: (I) hot side and cold side pressure decrements in channel plate spacing and channel plate width
drops, (II) hot side and cold side plates spacing, (III) channel (by combining Eqs. 12, 14, and 15). Higher Reynolds num-
plates width, (IV) SPHEs outside diameter, (V) hot side and bers in channels lead to higher heat transfer coefficients in
cold side Reynolds numbers, (VI) hot side and cold side hy- channels and higher overall heat transfer in the SPHE.
draulic diameters, (VII) heat transfer area, (VIII) overall heat Furthermore, the decreasing in channel plate spacing results
transfer coefficient, (IX) LMTD, (X) number of heat transfer in pressure drop escalating in channels to allowable pressure
units, (XI) effectiveness, (XII) total cost of SPHEs, (XIII) drop. Finally, higher overall heat transfer coefficient causes a
compactness and geometric proportion. reduction in required heat transfer area. In the light of men-
In this research, the standard designed SPHEs are designed tioned result analysis and comparison of designed SPHE com-
by developed Minton’s algorithm which is one of the contem- pactness, it is obvious that proposed algorithm of designing
porary methods in industries used by manufacturers to design SPHEs in this study has acceptable results.
SPHEs. In this method, after finalizing the channel plate In addition, based on proposed mathematical modeling of a
width, channel plate spacing is targeted to maximize pressure SPHE to a network of heat exchangers connected in series, the
drops by checking standard dimensions for channel plate results of equivalent internal heat exchangers including heat
spacing in Table 2. Figure 8 shows the simplified flow chart transfer rates, heat transfer areas, and temperature boundaries
of developed Minton’s algorithm. To uses mentioned flow are depicted in Table 4. To complete the study, temperature
charts, manufacturers tend to select dimensions from stan- distribution of hot and cold streams along the channels and
dardized tables such as Table 2.

Table 2 Standard geometries for SPHEs (Minton 1970)

H (mm) Maximum Ds (mm) Di (mm) b (mm) SS plate (mm) CS plate (mm)

101.6 812.8 203.2 4.8, 6.4 2.0, 2.4, 2.8, 3.2, 3.6, 4.0, 4.4, 4.8, 5.2, 5.6, 6.0, 6.4 3.2, 4.8, 6.4, 7.9
152.4 812.8 203.2
304.8 812.8 203.2
304.8 1473.2 304.8
457.2 812.8 203.2 6.4, 7.9, 9.5, 12.7, 15.9, 19.1, 25.4
457.2 1473.2 304.8
609.6 812.8 203.2
609.6 1473.2 304.8
762.0 1473.2 304.8
914.4 1473.2 304.8
1219.2 1473.2 304.8
1524.0 1473.2 304.8 7.9, 9.5, 12.7, 15.9, 19.1, 25.4
1828.8 1473.2 304.8
400 Process Integr Optim Sustain (2020) 4:391–408

temperature of the hot side, founded on energy balance equa-

tion of hot side and cold side, provided that the hot side inlet
temperature is fixed. Thus, these variations lead to changes in
temperature approach. In addition, LMTD is calculated from
energy balance equation and its trend is increasing with grow-
ing rate of temperature approach. Constant specific heat ca-
pacities cause to having linear LMTD (LMTD–ΔT–Std and
LMTD–ΔT–Opt), but variable cp,h leads to having a non-
linear LMTD diagram. That is, with an increasing range of
temperature approach, LMTD rises. To recapitulate, Fig. 10
shows the basis of analyzing methods in two aforementioned
steps in other key thermal and geometric parameters discussed
through Figs. 11, 12, 13, 14, 15, 16, and 17.
Figure 11 compares the overall heat transfer coefficients
(solid lines) and heat transfer area (dash lines) in the two
aforementioned algorithms, new optimization algorithm, and
developed Minton’s algorithm, versus a range of temperature
approach. This figure shows a gradual rise in overall heat
transfer coefficient when the temperature approach goes up.
It is noticeable that hot side specific heat capacity variations in
both algorithms is more effective in overall heat transfer co-
Fig. 8 Simplified flow chart of developed Minton’s algorithm efficient growing (U–Δcp–Std and U–Δcp–Opt) than having
constant specific heat capacities (U–ΔT–Std and U–ΔT–
Opt). This result can be easily analyzed by temperature-
temperature-enthalpy diagram of optimized SPHE is illustrat- enthalpy diagram. The larger specific heat capacity, the higher
ed in Fig. 9. Results in Table 4 and Fig. 9 are in lined with hot side outlet temperature. This higher temperature brings
predicted mathematical molding given in Fig. 5, and they about more heat transfer driving force, which results in higher
confirm the linear temperature distribution of hot and cold overall heat transfer coefficients. Moreover, the presented re-
streams along the channel length of designed SPHE and in sults in Fig. 11 indicate that the rate of increment in overall
each equivalent internal heat exchanger. The specific points heat transfer coefficients in reaching maximum pressure drop
on the hot and cold streams in Fig. 9a and b show the temper- (U–Δcp–Opt) is higher than the overall heat transfer coeffi-
ature boundaries in the heat exchangers network. cients in standard design (U–Δcp–Std). That is to say, the
In the second part of this section, the new suggested opti- higher specific heat transfer coefficient and pressure drop
mization algorithm for SPHEs and developed Minton’s algo- leads to a higher overall heat transfer coefficient at lower
rithm are compared. To compare the results of both algo- temperature approach.
rithms, a key parameter of temperature approach is set as the Furthermore, Fig. 11 depicts that in all cases, the heat trans-
basis of comprehensive investigation in this study. This study fer area decreases when temperature approach increases.
has been carried out in two following conditions, Fig. 10, with Since temperature approach acts as a driving force of heat
assumptions of constant cold side inlet and outlet temperature: transfer, larger driving force reduces the heat transfer area in
(1) changes of inlet and outlet temperatures of hot stream a larger extent. In addition, dash line graphs show new opti-
provided that cp,h is constant and (2) changes of cp,h with mization algorithm (A–ΔT–Opt and A–Δcp–Opt) finds
constant outlet temperature and variable inlet temperature of smaller heat transfer areas in lieu of developed Minton’s algo-
hot stream. rithm (A–ΔT–Std and A–Δcp–Std) by virtue of maximizing
Figure 10 depicts that LMTD variations (solid lines) and pressure drops. Finally, in a wider range of temperature ap-
hot side specific heat capacity changes (dash lines) versus proach, heat transfer area is not sensitive to the specific heat
temperature approach variations. Based on the assumptions coefficient changes whether it is constant or not.
in the study, hot side specific heat capacity, cp,h, is constant The given data in Fig. 12 shows that both channel plate
in the first step (Hot cp–ΔT–Std and Hot cp–ΔT–Opt) and is widths and outside diameters of SPHEs decrease as the tem-
increasing in a selected domain in the second step (Hot perature approach rises. This trend is a result of incremental
cp–Δcp–Std and Hot cp–Δcp–Opt). Based on that, tempera- rate in heat transfer driving force. As the main target is to find
ture approach increases due to having either constant or incre- a SPHE with identical channel plate width and outside diam-
mental rate of hot side specific heat coefficients. The variable eter, results in cited diagram are in lined with this goal. It is
hot side specific heat capacity directly affects the outlet remarkable that the variation of channel plate widths in
Process Integr Optim Sustain (2020) 4:391–408 401

Table 3 Comparison design results of SPHE for different algorithms

Parameter Minton (1970) Picon Nunez Picon Nunez Picon Nunez Picon Nunez New algorithm
et al. (2006) et al. (2007)a et al. (2009)b et al. (2009)c

Heat transfer area (m2) 15.044 14.772 11.24 14.48 18.50 12.325
Hot side Re 760 4390* 5208* 773 759 870
Cold side Re 303 1750* 2073* 307 301 346
Heat transfer rate (W) 186,486 186,500 186,500 186,300 186,174 186,486
Hot side heat transfer coefficient (W/m2 K) – 455.17 604.07 771.9 573.1 844.24
Cold side heat transfer coefficient (W/m2 K) – 467.15 619.93 390.1 304.6 450.49
Overall heat transfer coefficient (W/m2 K) 226.9 230.53 305.7 247.4 192 276.91
Plate thickness (m) 0.003175 0.003175 0.003175 0.003175 0.003175 0.003175
Channel plate width (m) 0.61 0.5334 0.45 0.6 0.61 0.5334
Inside diameter of SPHE (m) 0.203 0.203 0.203 0.203 0.203 0.2032
Outer diameter of SPHE (m) 0.59 0.61468 0.57 0.56 0.6 0.5419
Hot side pressure drop (KPa) 4.08 4.895 6.89 6.89 3.651 6.399
Cold side pressure drop (KPa) 5.09 6.89 6.89 6.89 4.249 6.440
Hot side channel plate spacing (m) 0.00635 0.00635 0.00508 0.00477 0.00635 0.00476
Cold side channel plate spacing (m) 0.00635 0.00635 0.00635 0.00627 0.007938 0.00595
Number of spiral turns – – – – – 9.659
Operating cost (USD) – – – – – 141.86
Capital cost (USD) – – – – – 86,643.60
Total cost (USD) – – – – – 86,785.46
Geometric proportiond 1.034 0.868 0.789 1.071 1.017 0.984
2 3 e
Compactness (m /m ) 90.207 93.325 97.885 97.983 107.263 100.186
a
Design with maximizing both pressure drops
b
Design with full pressure utilization on both streams
c
Design with standard dimensions
d
This parameter is calculated for all methods based on Eq. 36
e
This parameter is calculated based on the ratio of heat transfer area to SPHE volume
*This Reynolds number is irrationally higher than calculation of Eq. 12 or in comparison with next researches by Picon Nunez et al. (2009)

standard design (H–ΔT–Std and H–Δcp–Std) is based on Hot b–Δcp–Std, Cold b–ΔT–Std, and Cold b–Δcp–Std) is
dimensions cited in Table 2, but those of in optimum design curbed by standard dimensions mentioned in Table 2.
(H–ΔT–Opt and H–Δcp–Opt) is a continuous change. However, in new optimization algorithm, standard limita-
Indeed, observing some fluctuations in the trend of developed tions are not considered (Hot b–ΔT–Opt, Hot b–Δcp–Opt,
Minton’s algorithm (standard design) is caused by the dimen- Cold b–ΔT–Opt, and Cold b–Δcp–Opt). Then again,
sion selections from Table 2. Moreover, results in Fig. 12 reaching the maximum pressure drop at each side can reduce
illustrate that outside diameters of SPHEs are smaller in new channel plate spacing as much as possible. Lastly, in case of
optimization algorithm (Ds–ΔT–Opt and Ds–Δcp–Opt) than having variable specific heat capacity (Hot b–Δcp–Opt and
developed Minton’s algorithm (Ds–ΔT–Std and Ds–Δcp– Cold b–Δcp–Opt), by increasing in temperature approach,
Std) because of keeping the geometric proportion in specified channel plate spacing is larger than constantan specific heat
ranges, maximizing the pressure drops and having no limita- capacity at both sides (Hot b–ΔT–Opt and Cold b–ΔT–Opt).
tions in channel plate spacing. This significant result is due to growing specific heat capacity
Figure 13 illustrates the channel plate spacing (hot side and by incremental rate of temperature approach. Figure 13b
cold side channels) and their relevant hydraulic diameters ver- shows the channel hydraulic diameter at both hot and cold
sus temperature approach. Both hot side and cold side channel side. The patterns of hydraulic diameters are the same as chan-
plate spacing, reduces with a rising rate of temperature ap- nel plate spacing at both sides because of the hydraulic diam-
proach (Fig. 13a). This reduction results in maximizing the eter definition, Eq. 14, in this case study.
pressure drop at each side and increasing the overall heat The presented data in Fig. 14 depicts that channel pressure
transfer coefficient. It is noticeable that in the developed drop and channel Reynolds number at both hot side and cold
Minton’s algorithm, channel plate spacing (Hot b–ΔT–Std, side versus temperature approach variations. The study of
402 Process Integr Optim Sustain (2020) 4:391–408

Table 4 Results of equivalent internal heat exchangers of optimized This occurs due to a decrease in the channel plate spacing
SPHE
and an increase in fluid velocity in hot and cold side channel.
Spiral Qj: Heat Aj: Heat Lj: Heat Tj: Hot side Tj: Cold side Figure 15 illustrates two dimensionless parameters of NTU
turn transfer transfer transfer temperature temperature and total effectiveness of SPHE versus changes of temperature
rate (W) area (m2) length (m) (°C) (°C) approach. Results show that both NTU and total effectiveness
0 0 0 0 120.000 60.000
are equal in developed Minton’s algorithm and new optimiza-
1 12,843.098 0.777448 0.728766 125.524 66.226
tion algorithm, and they decrease when temperature approach
goes up. That is, the larger SPHEs, in terms of lower temper-
2 14,543.821 0.891837 0.835993 131.762 73.276
ature approach, had higher NTU and total effectiveness. The
3 16,170.181 1.006226 0.943219 138.698 81.115
trend of reduction in these two dimensionless parameters is
4 17,714.702 1.120615 1.050445 146.296 89.702
higher when temperature differences aspect is deemed
5 19,170.595 1.235004 1.157672 154.518 98.995
(NTU–ΔT–Std, NTU–ΔT–Opt, Effec.–ΔT–Std, and
6 20,531.809 1.349394 1.264898 163.324 108.948
Effec.–ΔT–Opt) because of constant value of C min .
7 21,793.072 1.463783 1.372125 172.671 119.512
However, with variation of the specific heat capacities in sec-
8 22,949.919 1.578172 1.479351 182.514 130.637
ond studying condition, the reduction trend in NTU and total
9 23,998.712 1.692561 1.586578 192.807 142.271
effectiveness is less sensitive to temperature approach incre-
9.659 16,770.066 1.209607 1.133865 200.000 150.400
ments (NTU–Δcp–Std, NTU–Δcp–Opt, Effec.–Δcp–Std,
and Effec.–Δcp–Opt). This trend is a result of changes in
specific heat capacities to calculate Cmin. That is, this graph
channel pressure drop, Fig. 14a, shows that in new optimiza- shows the importance of specific heat capacities in SPHE
tion algorithm, pressure drops at both sides (Hot ΔP–ΔT– design. It is worth mentioning that increasing in specific heat
Opt, Hot ΔP–Δcp–Opt, Cold ΔP–ΔT–Opt, and Cold capacities leads to decrease in NTU and total effectiveness.
ΔP–Δcp–Opt) can approach near allowable pressure drop Figure 16 shows the compactness and total cost of designed
(6.8948 KPa), because of having no limitation on minimum SPHEs versus temperature approach variations. According to
channel plate spacing, whereas in developed Minton’s algo- solid line graphs, compactness of SPHEs in new optimization
rithm, both sides pressure drops (Hot ΔP–ΔT–Std, Hot algorithm (Comp.–ΔT–Opt and Comp.–Δcp–Opt) is notice-
ΔP–Δcp–Std, Cold ΔP–ΔT–Std, and Cold ΔP–Δcp–Std) ably larger than that of in developed Minton’s algorithm
experience fluctuations and fall off by increasing in tempera- (Comp.–ΔT–Std and Comp.–Δcp–Std). This difference oc-
ture approach. This trend is caused by limitations on channel curred due to the limitations on geometry and channel plate
plate spacing selection from Table 2 and is normal. Therefore, spacing of SPHEs designed by developed Minton’s algorithm
increasing the actual pressure drops below the allowable pres- referring to Table 2. More details and applicability of
sure drops in such SPHEs is impossible. In addition, Fig. 13b compactness-temperature approach diagram will be discussed
presents the Reynolds numbers of both sides in both algo- in Fig. 17.
rithms. It is noticeable that although Reynolds numbers are According to graphs of TCost–ΔT–Std, TCost–Δcp–Std,
limited in developed Minton’s algorithm (Hot Re–ΔT–Std, TCost–ΔT–Std, and TCost–Δcp–Std in Fig. 16, estimated
Hot Re–Δcp–Std, Cold Re–ΔT–Std and Cold Re–Δcp– total cost of SPHEs decreases as temperature approach in-
Std), because of applying mentioned limited dimensions in creases. It is noticeable that total cost of designed SPHEs by
Table 2, Reynolds numbers are increasing and become larger new optimization algorithm is lower than that of by developed
in new optimization algorithm (Hot Re–ΔT–Opt, Hot Minton’s algorithm. Although energy cost is higher in new
Re–Δcp–Opt, Cold Re–ΔT–Opt, and Cold Re–Δcp–Opt). optimization algorithm, due to having higher pressure drops,

Fig. 9 a Temperature distribution

of streams in the SPHE. b
Temperature-enthalpy diagram of
the SPHE
Process Integr Optim Sustain (2020) 4:391–408 403

Fig. 10 Relations of hot side

specific heat capacity,
temperature approach, and
LMTD

than developed Minton’s algorithm, the capital cost is notice- incremental rate of hot side specific heat coefficients. By in-
ably decreased by the reduction in heat transfer area. In fact, creasing the temperature approach, compactness grows and
the total cost estimation in optimum design is lower than that experiences peak ranges, and then falls off. Figure 17 not only
of standard design. shows appropriate ranges of temperature approach for de-
To complete the analysis of results in compactness- signing SPHEs with higher compactness, but also con-
temperature approach diagram of Fig. 16, Fig. 17 shows the firms the priority of selecting the new proposed optimiza-
results and fitting curves (polynomial-order: 4) for all de- tion algorithm for designers and manufacturer to reduce
signed cases by new optimization algorithm (Comp.–ΔT– end prices by designing more compact heat exchangers.
Opt and Comp.–Δcp–Opt) and developed Minton’s algorithm To summarize, this diagram, compactness-temperature ap-
(Comp.–ΔT–Std and Comp.–Δcp–Std). The trends of com- proach diagram, shows a new method of geometric opti-
pactness in both algorithms for all cases are the same when mization of SPHEs and the effects of temperature differ-
temperature approach rises due to having either constant or ence in optimization of SPHEs.

Fig. 11 Relations of overall heat

transfer coefficient and heat
transfer area versus temperature
approach
404 Process Integr Optim Sustain (2020) 4:391–408

Fig. 12 a Relations of channel

plate width versus temperature
approach. b Relations of outer
diameter of SPHE versus
temperature approach

Conclusion capacities. Secondly, the proposed mathematical model was

used to analyze the internal heat transfer interactions between
This work presented three major aspects of modeling, analyz- fluids in spiral turns and to find the temperature distributions
ing, and optimizing SPHEs. Firstly, a mathematical model of streams along the channels. Results from modeling of a
was proposed based on energy balance equation assuming single-phase counter-current SPHE with a series of connected
constant overall heat transfer coefficient and specific heat equivalent internal heat exchangers showed that the predicted

Fig. 13 a Relations of channel plate spacing versus temperature approach. b Relations of channel hydraulic diameter versus temperature approach
Process Integr Optim Sustain (2020) 4:391–408 405

Fig. 14 a Relations of channel pressure drop versus temperature approach. b Relations of channel Reynolds number versus temperature approach

linear temperature distribution of temperature-enthalpy dia- to reach the maximum pressure drops in channels and to keep
gram was valid, and they confirmed the applicability of T-Q appropriate geometric proportion in SPHEs considering eco-
diagram in SPHE analysis. Thirdly, a newly developed ther- nomic viewpoint. Comparing new algorithm results with pre-
mal design optimization algorithm of SPHEs was introduced vious methods showed an increase in overall heat transfer

Fig. 15 Relations of NTU and

total effectiveness versus
temperature approach
406 Process Integr Optim Sustain (2020) 4:391–408

Fig. 16 Relations of compactness

and total cost of SPHE versus
temperature approach

coefficient and a decrease in heat transfer area. Moreover, the and changes of cp,h with constant outlet temperature and var-
new optimization algorithm was employed to study SPHEs iable inlet hot stream temperature.
design parameters thoroughly, and the results were compared With assumption of constant heat transfer rate and flow
with developed Minton’s algorithm used extensively in man- rates in a SPHE design, and setting approach temperature as
ufacturers’ design. The study was conducted for different de- a key parameter to analyze the SPHE, results depicted that (I)
sign cases with assumptions of constant cold side inlet and an increase in specific heat capacity led to a higher rate of
outlet temperature in two conditions of changes of inlet and rising in overall heat transfer coefficient and decreasing in heat
outlet temperatures of hot stream provided that cp,h is constant, transfer area and total cost in new optimization algorithm; (II)

Fig. 17 Fitted curves for relations

of compactness versus
temperature approach of SPHE
Process Integr Optim Sustain (2020) 4:391–408 407

the rate of reduction in channel plate spacing, channel plate References

width, hydraulic diameter and outer diameter of SPHE in new
optimization algorithm is higher than those of in developed Anil M, Prasad KL, Kumar AN (2018) Performance study of spiral plate
heat exchanger. Int J Res Eng Appl Manag 4(4):176–180
Minton’s algorithm; (III) maximizing pressure drops in chan-
Bes T, Roetzel W (1991) Approximate theory of spiral heat exchanger.
nels is highly likely by keeping geometric proportion and Design and Operation of Heat Exchangers, Proceedings of the
economic consideration in new optimization algorithm; (IV) Eurotherm Seminar No. 18, Hamburg, Germany (Edited by
new optimization algorithm designed more compact SPHEs Roetzel W, Heggs PJ, Butterworth D (1991) springer–Verlag,
than developed Minton’s algorithm; and (V) the variation of Berlin: 223–232)
Bes T, Roetzel W (1992) Distribution of heat flux density in spiral heat
temperature approach parameter showed a peak range of exchangers. Int J Heat Mass Transf 35(6):1331–1347
SPHE compactness for designers to design more compact heat Bes T, Roetzel W (1993) Thermal theory for spiral heat exchanger. Int J
exchangers. The diagram of compactness-temperature ap- Heat Mass Transf 36(3):765–773
proach showed a range of temperature approach with higher Bidabadi M, Sadaghiani AK, Vahdat Azad A (2013) Spiral heat exchang-
er optimization using genetic algorithm. Sci Iran B 20(5):1445–
compactness designed SPHEs, which in turn showed a geo-
1454
metric optimization method. This diagram accentuates the im- Burmeister LC (2006) Effectiveness of a spiral-plate heat exchanger with
portance of temperature approach parameter and the effects of equal capacitance rates. J Heat Transf 128(3):295–301
temperature differences in designing optimized SPHEs with Chen Q, Xu YC, Guo ZY (2012) The property diagram in heat transfer
higher compactness to reduce the manufacturing prices. and its applications. Chin Sci Bull 57(35):4646–4652
Chowdhury K, Linkmeyer H, Bassiouny MK, Martin H (1985)
Analytical studies on the temperature distribution in spiral plate heat
Greek Letters β Compactness factor or surface density (m2/m3)
exchangers: straightforward design formulae for efficiency and
ΔP Pressure drop (KPa)
mean temperature difference. Chem Eng Process 19:183–190
ΔTLM Logarithmic mean temperature difference-LMTD (K)
Dávalos LC, Luna EM, Angeles MAR, Delgado VJC (2019) Designing
ΔTmax Greater terminal temperature difference (K)
spiral plate heat exchangers to extend its service and enhance the
ΔTmin Smaller terminal temperature difference (K)
thermal and hydraulic performance. IntechOpen. https://doi.org/10.
ε Thermal effectiveness per turn
5772/intechopen.85345
εtot Thermal effectiveness of heat exchanger
η Pumping efficiency Devois JF, Durastanti JF, Martin B (1995) Numerical modelling of the
μ Viscosity (kg/m s) spiral plate heat exchanger. J Therm Anal 44:305–312
μb Fluid bulk viscosity (kg/m s) Dongwu W (2003) Geometric calculations of the spiral heat exchanger.
μw Fluid viscosity at the wall temperature (kg/m s) Chem Eng Technol 26(5):592–598
ρ Density (kg/m3) Energy Efficiency Office (2000) Compact heat exchangers: a training
package for engineers. Energy Efficiency Office, London
Garcia MM, Moreles MA (2012) A numerical method for rating thermal
Compliance with Ethical Standards performance in spiral heat exchangers. Mod Appl Sci 6(6):54–63
Guha P, Unde V (2014) Mathematical modeling of spiral heat exchanger.
Conflict of Interest The authors declare that there is no conflict of Int J Eng Res 3(4):226–229
interest. Hall SG, Ahmad S, Smith R (1990) Capital cost targets for heat exchang-
er networks comprising mixed materials of construction, pressure
Nomenclature A , Heat transfer area (m2); Ac , Free flow area (m2); b , ratings and exchanger types. Comput Chem Eng 14(3):319–335
Channel plate spacing (m); cp , Specific heat capacity (J/kg K); C , Heat Jarzębski AB (1984) Dimensioning of spiral heat exchangers to give
capacity rate ratio; Ce , Energy cost (USD/KW hr); Ci , Capital cost minimum costs. J Heat Transf 106(Aug.):633–637
(USD); Cod , Operational cost (USD); Ctot , Total cost (USD); CS , Kathir Kaman MD, Sathishkumar A, Balasuthagar C, Ponsankar (2017)
Carbone steel; Dh , Hydraulic diameter (m); Di , Core or Inside diameter Design and analysis of spiral plate heat exchanger for cooling appli-
(m); Ds , Outer diameter of SPHE (m); FT , Temperature difference cations. J Chem Pharm Sci 10(1):511–514
correction factor; GP , Geometric proportion; h , Heat transfer coefficient Khorshidi J, Heidari S (2016) Design and construction of a spiral heat
(W/m2 K); H , Channel plate width (m); Hw , Work hours (hr/yr); HX , exchanger. Adv Chem Engineer Sci 6:201–208
Heat exchanger; i , Annual discount rate; k , Thermal conductivity Kumar KPM, Vijayan V, Kumar BS, Vivek CM, Dinesh S (2018)
(W/m K); L , Plate length (m); M , Mass flow rate (kg/s); Mweight , Computational analysis and optimization of spiral plate heat ex-
Molecular weight; n , Year (yr); N , Number of spiral turns of the stream; changer. J Appl Fluid Mech 11:121–128
NTU , Number of heat transfer units; ny , Equipment life year (yr); p , Kumar TA, Sharma N, Mohammad MN, Pradeep BT, Saichand U,
Plate thickness (m); P , Pressure (KPa); Pr , Prandtl number; Q , Heat Vamsi NM (2019) Optimization of spiral plate heat exchanger by
transfer rate (W); Rfouling , Fouling factor resistance (m2 K/W); Re, gradient based optimizer. Int J Innov Technol Explor Eng 8(6):
Reynolds number; Recr , Critical Reynolds number; s , Relative density 1819–1823
(relative to water at 20 °C); SPHE , Spiral plate heat exchanger; SS ,
Kuppan T (2013) Heat exchanger design handbook, 2nd edn. CRC Press
Stainless steel; T , Temperature (°C); U , Overall heat transfer coefficient
– Taylor and Francis Group, New York
(W/m2 K); V , Total volume of SPHE (m3); Vf , Fluid mean velocity
Manoj V, Gopal P, Senthilkumar T (2016) Heat transfer enhancement by
(m/s)Subscripts b , Bulk fluid properties; c , Cold stream; cr , Critical;
using nanofluid in spiral plate heat exchanger. Int J Eng Res Technol
f , Fluid; h , Hot stream; HX , Heat exchanger; i , Inlet; j , Counter; l ,
5(6):101–105
Liquid; max , Maximum; min , Minimum; o , Outlet; s , Scale or fouling
material; w , Wall plate material or at wall temperature Martin H (1992) Heat exchangers. CRC Press – Taylor and Francis
Group, Washington DC
Memon S, Gadhe P, Kulkarni S (2019) Design and testing of a spiral plate
heat exchanger for textile industry. Int J Sci Eng Res 10(7):149–157
408 Process Integr Optim Sustain (2020) 4:391–408

Metta VR, Konijeti R, Dasore A (2018) Thermal design of spiral plate Rajavel R, Saravanan K (2008b) Heat transfer studies on spiral plate heat
heat exchanger through numerical modeling. Int J Mech Eng exchanger. Therm Sci 12(3):85–90
Technol 9(7):736–745 Rosenblad CF (1935) Heat exchanger apparatus made of sheet metal.
Milovančević UM, Jaćimović BM, Genić SB, El-Sagier F, Otović MM, Patent No US2011201A
Stevanovic SM (2019) Thermoeconomic analysis of spiral heat ex- Rosenblad CF (1936) Heat exchanger apparatus. Patent No US2060440A
changer with constant wall temperature. Therm Sci 23(1):401–410 Sakariya PJ, Jhavar PM, Gujarati RD (2014) Analysis of heat transfer in
Minton PE (1970) Designing spiral-plate heat exchangers. Chem Eng spiral plate heat exchanger using experimental and CFD. Int J Sci
77(May):103–112 Res Dev 2(7):446–451
Nguyen DK, San JY (2011) Heat transfer performance of a spiral heat Saravanan K, Rajavel R (2008a) Analysis of heat transfer enhancement in
exchanger. Conference: Proceedings of the 28th National spiral plate heat exchanger. Mod Appl Sci 2(4):68–75
Conference on Mechanical Engineering of CSME A01–008,
Saravanan K, Rajavel R (2008b) An experimental investigation of heat
Taiwan
transfer coefficients for spiral plate heat exchanger. Mod Appl Sci
Nguyen DK, San JY (2015) Effect of solid heat conduction on heat
2(5):14–20
transfer performance of a spiral heat exchanger. Appl Therm Eng
76:400–409 Shabiulla AM, Sivaprakasam S (2014) Optimization of process parame-
Nguyen DK, San JY (2016a) Decrement in heat transfer effectiveness due ters for water–sea water (3%) system in a spiral plate heat exchanger
to solid heat conduction for a counter–current spiral heat exchanger. (SHE) using response surface methodology (RSM). Int J Dev Res
Appl Therm Eng 103:821–831 4(5):1020–1026
Nguyen DK, San JY (2016b) Heat transfer and exergy analysis of a spiral Shah RK, Sekulic DP (2003) Fundamentals of heat exchanger designing.
heat exchanger. Heat Transf Eng 37(12):1521–0537 John Wiley & Sons Inc., Hoboken, New Jersey
Picon Nunez M, Canizalez Davalos L, Morales Fuentes A (2006) Strelow O (2000) A general calculation method for plate heat exchangers.
Alternative design approach for spiral plate heat exchangers. Heat Int J Therm Sci 39:645–658
Transf Eng 27(6):12–21 Tapre RW, Kaware JP (2018a) Experimental study of the effect of the
Picon Nunez M, Canizalez Davalos L, Martinez Rodriguez G, Polley GT Reynolds number on overall heat transfer coefficient in spiral heat
(2007) Shortcut design approach for spiral heat exchangers. Food exchanger for acetic acid – water system. SSRG Int J Chem Eng Res
Bioprod Process 85(4):322–327 5(2–May to Aug):1–4
Picon Nunez M, Canizalez Davalos L, Medina Flores JM (2009) Tapre RW, Kaware JP (2018b) Experimental analysis of spiral heat ex-
Alternative sizing methodology for compact heat exchangers of changer: evaluation of Reynolds number and Nusselt number for
the spiral type. Heat Transf Eng 30(9):744–750 acetic acid – water system. Int J Sci Res Sci Technol 4(9):1–4
Polley GT, Panjeh Shahi MH, Jegede FO (1990) Pressure drop consider- Tapre RW, Kaware JP (2018c) Comparison of theoretical and experimen-
ations in the retrofit of heat exchanger networks. Chem Eng Res Des tal overall heat transfer coefficient for acetic acid–water system in
68(May):211–220 spiral heat exchanger. SSRG Int J Therm Eng 4(3–Sep to Dec):1–4
Polley GT, Panjeh Shahi MH, Picon Nunez M (1991) Rapid design Vasconcelos Segundo EH, Mariani VC, dos Santos CL (2018) Design of
algorithms for shell and tube and compact heat exchanger. Trans I spiral heat exchanger from economic and thermal point of view
Chem E 69(A):435–444 using a tuned wind-driven optimizer. J Braz Soc Mech Sci Eng
Rajavel R (2014) Experimental and numerical studies of a spiral-plate 40(212):1–14
heat exchanger. Therm Sci 18(4):1355–1360
Rajavel R, Saravanan K (2008a) An experimental study of spiral plate
Publisher’s Note Springer Nature remains neutral with regard to jurisdic-
heat exchanger for electrolytes. J Univ Chem Technol Metall 43(2):
tional claims in published maps and institutional affiliations.
225–260