International Journal of Advanced Nuclear Reactor Design and Technology 3 (2021) 80–96

Contents lists available at ScienceDirect

International Journal of Advanced Nuclear

Reactor Design and Technology
journal homepage: www.sciencedirect.com/journal/international-journal-of-advanced-
nuclear-reactor-design-and-technology

Study on the thermal and geometrical parameters of helical coil

once-through steam generator system
Hao Yao a, Guo Chen b, Kailin Lu a, Haoyu Liao a, Yingwei Wu a, *, Wenxi Tian a, Guanghui Su a,
Suizheng Qiu a
a
School of Nuclear Science and Technology, State Key Laboratory of Multiphase Flow in Power Engineering, Xi’an Jiaotong University, Xi’an, 710049, China
b
Nuclear Power Institute of China, Chengdu, 610226, China

A R T I C L E I N F O A B S T R A C T

Keywords: Due to the special structure, helical coil once-through steam generators (HCOTSGs) can withstand greater
Helical coil once-through steam generator thermal expansion stress, have a larger heat exchange area, and cause the secondary flow phenomenon to
Flow and heat transfer model improve the heat transfer capacity. Thus, it is inevitable to further study the thermal-hydraulic characteristics of
Thermal parameter
the HCOTSG, a vital component of reactors. The optimal physical models were selected for different heat transfer
Geometrical parameter
areas by detailed investigation and comparison, to adopt in the TACS, a self-developed thermal-hydraulic
Code simulation
analysis code that is accurate and widely-used in HCOTSG simulation. After verification with experiments of
Santini, taking the IRIS reactor as the object, the TACS was employed to briefly study the thermal-hydraulic
performances of the HCOTSG system at first, and then discuss the influences of thermal and geometric param­
eters on the HCOTSG. The analysis results showed that: the variation of the primary flow rate greatly affects the
heat transfer coefficient of the primary side, while the maximum value of the heat transfer coefficient of the
secondary side only changes drastically with the secondary flow rate; when the primary side flow drops to 70%,
superheated steam is not generated on the secondary side; when the secondary mass flow rate decreases to a
certain extent, the steam superheat degree increases by a small amplitude; the primary side inlet temperature has
a little influence on the primary temperature difference, but significant influence on the two-phase zone length of
the secondary side; the secondary inlet temperature has very little influence on the HCOTSG system; the inner
diameter and the outer diameter of helical tubes have obvious influences, while the helical diameter has not.

1. Introduction researches on SG systems. The performance and operation mode of

various steam generators are different, so SGs are suitable for different
As the pivotal of reactor coolant loops, the steam generators (SGs) scenarios. Among all of them, helical coil once-through steam generator
are not only the heat exchange equipment but also the pressure (HCOTSG) has been extensively applied in energy and power engi­
boundary. The designed service life of steam generators is usually up to neering, petrochemical engineering and other industries in recent years
40 years. However, previous operating experiences indicate that due to due to its advantages of high thermal efficiency, low thermal stress,
the influence of working mediums flowing in SGs, performances of heat compact structure and easy manufacture. In HCOTSGs, the primary side
transfer tubes and other components will be degraded or even destroyed (PRS) coolant generally flushes the heat transfer tube bundle outside,
by corrosion, vibration, fatigue, sag, fretting and other phenomena, and the secondary side (SES) water driven by a feed pump is heated by
leading to malfunctions after being used for 15–20 years. Nearly 1/4 of the fluid in the flow process along the heat transfer tube wall inside,
unplanned shutdowns happened in pressurized water reactor (PWR) turning into steam and driving the steam turbine to generate electricity.
power plants all around the world are caused by steam generator acci­ The environment and the fluid physical property differences on PRS and
dents [1]. The failure of steam generators seriously affects the safety SES demand high indexes of the device structure, material strength,
characteristics of reactors, and its unanticipated replacement will cost a fatigue resistance, corrosion resistance and water quality involving
lot and require months or even years. Therefore, in order to use nuclear thermal-hydraulics, structure mechanics, material science, water
energy safely and efficiently, it is necessary to carry out comprehensive chemistry, and other disciplines. The thermal-hydraulics is the research

* Corresponding author.
E-mail addresses: [email protected] (H. Yao), [email protected] (Y. Wu).

https://doi.org/10.1016/j.jandt.2021.07.001
Received 15 October 2020; Received in revised form 2 June 2021; Accepted 2 July 2021
Available online 4 July 2021
2468-6050/© 2021 Xi’an Jiaotong University. Published by Elsevier B.V. This is an open access article under the CC BY-NC-ND license
(http://creativecommons.org/licenses/by-nc-nd/4.0/).
H. Yao et al. International Journal of Advanced Nuclear Reactor Design and Technology 3 (2021) 80–96

basis of the others, because the special structure of HCOTSGs makes the HCOTSGs in recent years, but most of them are in the preliminary
fluid in tubes subject to the strong influence of centrifugal force, leading research stage. Relatively, there is a little overall analysis of HCOTSG
the secondary flow phenomenon [2]. Under the effect of this phenom­ using codes among existing studies. The adopted models vary from code
enon, the boundary layer will be destroyed which results in heat ex­ to code, leading to application scope limitations. Through detailed
change capability improvement. Thereby, the heat transfer efficiency of investigation and analysis, this paper selected a set of perfect and ac­
HCOTSGs is higher, and the thermal-hydraulic characteristics of curate mathematical physical models and adopts appropriate calcula­
HCOTSGs are notably different from that of straight-tube and U-tube tion methods to develop a thermal-hydraulic analysis code, named
steam generations, which needs some targeted researches. TACS, which has wide-range applicability for HCOTSG systems. Based
Eustice [3] noticed that the mass flow rate in a bent tube is smaller on the TACS, thermal-hydraulic characteristics analysis of HCOTSG with
than that in a same-diameter straight tube under the same pressure. different thermal and geometric parameters can provide a reference for
Owhadi et al. [4] found that convective heat transfer has a more obvious HCOTSG structural design and performance improvement, which is
effect in most boiling zones and nuclear boiling only plays a role at the conducive to the optimal design and efficiency improvement of nuclear
low steam quality. The secondary flow phenomenon results in the un­ energy system equipment.
even liquid film distribution in the circumferential direction on the tube
wall. The surface fluctuation of the liquid film is also caused to entrain 2. Physical models
steam which will accumulate under the action of the secondary flow.
Through experiments, Styrikovich et al. [5] found that, compared to 2.1. Working principle and basic assumptions
straight tubes, the steam quality at the dry-out point of helical tubes is
higher, the wall temperature jumping in the dry-out zone is more tender, In HCOTSGs, the PRS coolant flows from top to bottom outside the
and the temperature curve is flatter. Austen et al. [6] experimentally helical tube, transferring heat through the tube wall to the SES fluid
studied the effect of helical pitch and found that the secondary flow which flows from bottom to top inside the tube. The PRS fluid keeps its
phenomenon aroused by the centrifugal force are weak at low Reynolds liquid phase, while the SES fluid, being continuously heated in the flow
numbers, and the helical pitch obviously affects the friction coefficient process, goes through the stages of subcooled water, subcooled boiling,
and the Nusselt number, but this effect will weaken by the Reynolds saturated nucleate boiling and forced convection evaporation, liquid
number increasing. Berthoud et al. [7] analyzed a large number of deficiency, and superheated steam sequentially. The flow and the heat
existing studies and compared the effects of system pressure, helical transfer in this process are illustrated in Fig. 1.
diameter, mass flow rate and heat flux on steam quality in the case of Each helical tube in HCOTSGs can basically maintain the same
drying out. Ali et al. [8] used polyethylene tubes and wooden cylinders length through the coordination of geometric parameters such as radial
with different geometrical sizes in their experiment to obtain the change pitch, axial pitch, helix lead angle and number of helical tubes in each
of pressure drop with flow rate under different geometrical conditions. helical coil layer, to ensure uniform heating of each tube. Temperature
Bertani et al. [9] observed flow patterns in the process by plexiglass and pressure drop in each layer of the HCOTSG helical coil have little
helical tubes, and measured pressure differences at different heights differences, and distributions of flow condition and thermal parameters
under different air-water flow combinations by nine differential pres­ are basically the same, which are convenient for modeling. In this
sure sensors. The experiment of Guo et al. [10] showed that the friction simulation, all helical tubes in a HCOTSG were represented by a char­
pressure drop in a small helical diameter tube is larger than that in a acteristic tube whose diameter, hydraulic diameter, length and height
large one, which is more obvious at high Reynolds numbers. Besides, were the same as those of a single helical tube. The mass flow rate and
new relations of friction pressure drops in helical tubes are proposed by the heat exchange area of the fluid flowing through the characteristic
the experiment data. Zhao et al. [11] studied the forced convection tube were the corresponding sums of all helical tubes, and the cross-
boiling heat transfer and the pressure drop characteristics in a horizontal sectional flow area of the characteristic tube is obtained according to
small-scale helical tube, and gave new relations of forced convection the principle of volume conservation. This method is also used in the
boiling heat transfer and two-phase frictional pressure drop. Ye et al. simulation codes of scholars mentioned in the above introduction. The
[12] get the velocity and pressure distribution of the PRS fluid in a schematic diagram of the characteristic tube is shown in Fig. 2.
HCOTSG by ANSYS FLUENT simulation, to study the impacts of different In this study, the flow and the heat transfer of HCOTSGs were
inlet parameters and structural designs on the exchanger performances simulated. Due to the complex structures of HCOTSGs, in order to keep
and flow distributions of the HCOTSG. Mirgolbabaei [13] evaluated the the balance between result accuracy and calculation speed at the same
thermal performances of helical tube exchangers with different PRS time, the following basic assumptions were adopted:
mass flow rates, helical curvature ratios and helical pitches. Castiglia
et al. [14] improved the RELAP5/Mod3.2β code with HCOTSG hydraulic
models and certified its validity of analyzing single-phase and two-phase
flow characteristics of vertical helical tubes under various operating
environments by several experiments. In order to study the hydraulic
phenomenon of incompressible laminar flow in helical tubes, De Amicis
et al. [15] discussed the influence of geometric structure on the change
of axial velocity and the secondary flow phenomenon in helical tubes by
FLUENT, OpenFOAM and COMSOL. Vaidyanathan et al. [16] developed
a one-dimensional code DESOPT with the non-uniform grid (constant
power grid) to predict the performance of the steam generators under
various loads. Colorado et al. [17] established a computational model to
simulate the thermal-hydraulic characteristics of HCOTSG in nuclear
power plants and optimized the relations of the two-phase frictional
pressure drop multiplication factor. Zhang et al. [18] developed the
THAOT code to calculate the heat transfer area of HCOTSG under a
given heat transfer quantity and the transient characteristics of HCOTSG
under given geometric parameters.
Due to its simplicity and convenience, many research institutions
have independently developed thermal-hydraulic analysis codes of Fig. 1. The flow and the heat transfer in a helical tube.

81
H. Yao et al. International Journal of Advanced Nuclear Reactor Design and Technology 3 (2021) 80–96

h − hl
x= (4)
hg − hl

1
α= ρg (5)
1 + 1−x x ρl

ρ = αρg + (1 − α)ρl (6)

ρh = αρg hg + (1 − α)ρl hl (7)

Vg = V l (8)

G = αVg ρg + (1 − α)Vl ρl (9)

/( )
x 1− x
μ=1 + (10)
μg μl

2.2.2. Enthalpy equation and pressure equation

Because density depends on pressure and enthalpy, it can be written
as:
∂ρ ∂ρ ∂P ∂ρ ∂h
= + (11)
∂t ∂P ∂t ∂h ∂t
Substituting this equation into Eq. (1), we can get a relational
expression with ∂∂Pt and ∂∂ht . It is noted that Eq. (2) is also an equation with
Fig. 2. The schematic diagram of characteristic tube. these two terms, so they can be simultaneously solved:
( ) ( )
W ∂h ∂ρ
(1) The fluid flow is one-dimensional. Within an axial section, the − qU + A ∂z ∂h
− ρ ∂ W
∂z A
∂P A

thermal-hydraulic parameters of the PRS and SES fluids remain = (12)

∂t ρ ∂∂ρP + ∂ρ
unchanged.
∂h

(2) The PRS fluid does not undergo the phase transition. ( )
∂ρ
( )
qU W ∂h ∂ W
(3) The external heat dissipation of HCOTSGs is ignored. The heat ∂h A
− A ∂z ∂P − ∂z A

capacity of all structures except the wall of helical tubes is not = (13)
∂t ρ ∂∂ρP + ∂ρ
taken into consideration.
∂h

(4) In the radial direction, the temperature gradient of the PRS and
2.2.3. Heat conduction equation
SES fluids is large, so the axial heat conduction of tube walls can
According to the heat transfer theory, the heat conduction equation
be neglected compared with that of the radial direction.
of the tube wall is:
( )
2.2. Basic models ∂Tw hp Ap Tp − Tw1 − hs As (Tw2 − Ts )
= (14)
∂t mcpw
2.2.1. Conservation equations
Because the heat flux of a tube wall is continuous, the energy bal­
At present, there are three kinds of two-phase flow mathematical
ances of both sides of tube walls under the combined action of heat
models for numerical simulations of steam generators, namely the ho­
conduction and convection can be respectively described as:
mogeneous flow model, the drift flux model and the two-fluid model.
The Ref. [19] indicates that the overall slip ratio of helical tubes is small, ( ) Tw1 − Tw
hp Tp − Tw1 = λ (15)
and the thermodynamics is basically balanced. Therefore, according to δ/2
the characteristics of two-phase flow mathematical models, the devel­
oped code adopts the homogeneous flow model, which has been suc­ Tw − Tw2
λ = hs (Tw2 − Ts ) (16)
cessfully applied in the two-phase flow analysis of HCOTSGs by Yoon δ/2
[20], Ma [21], Zhang [18], Xu [22]. Its convenient solution method and
high calculation accuracy have also simply proved.
The mass, energy and momentum conservation equations for the 2.3. Heat transfer models
single-phase are:
( ) 2.3.1. Criteria for judging heat transfer zone
∂ρ ∂ W The Saha-Zuber formula, adopted in program developments for
+ =0 (1)
∂t ∂z A HCOTSGs by Yoon [20], Xia [23] et al., is generally used to determine
the starting point of bubble separating from the wall.
∂h W ∂h qU ∂P
ρ + = + (2) The helical tubes used in the experimental device of Guo [24] were
∂t A ∂z A ∂t all small spiral diameters (0.256 m and 0.3 m), while the helical di­
( ) ameters in HCOTSGs are generally large, so the Guo relation is limited in
∂W ∂ W 2 ∂P fW|W|
+ =− A − − ρgA (3) applicability. Santini [25] built an experimental device of full-size he­
∂t ∂z ρA ∂z 2De ρA
lical tubes to simulate the HCOTSG of the International Reactor Inno­
In a homogeneous flow, the conservation equations for the two- vative and Secure (IRIS), and the experimental cases covered the
phase will have the same forms as those for the single-phase if two- operating conditions of the IRIS reactor steam generator. Thus, the
phase parameters can be expressed as follows: relation of steam quality at the dry-out point fitted by experimental data

82
H. Yao et al. International Journal of Advanced Nuclear Reactor Design and Technology 3 (2021) 80–96

can reflect the operating characteristics of the HCOTSG more practi­ 2.3.3. Critical Reynolds number relations
cally. The steam quality relations at the dry-out point proposed by As most relational expressions proposed by scholars, such as Ito
Ruffell [26] include two parts, the inside tube part and the outside tube [27]，Kutateladze [28]，Kubair [29]，Schmidt [30]，EI-Genk [31], are
part. This one-dimensional simulation is not easy to use Ruffell relations. based on their experimental data, their applicability is limited to other
Compared the experiment data with the calculated value of Ruffell conditions. Ei-Genk et al. [31] proposed a critical Reynolds number
relation, Santini found that the average deviation was 35.1%, the relational expression that can apply to a wide range of helical tube sizes
root-mean-square deviation is as high as 45.6%, and Ruffell relations by sorting out data from previous experimental studies. In this relation,
calculation results are even obviously wrong in the low mass flow rate the modified helical curvature ratio (λm) is adopted. Compared with
(about 200 kg m− 2 s− 1). As a consequence, the Santini relation is others, it considers the influences of the helical diameter, the tube
adopted in the code to calculate the critical steam quality at the dry-out diameter and the helical pitch on the conversion from laminar flow to
point. So far, the criteria adopted in this study can judge the heat turbulent flow. For the whole experimental data set, the calculation
transfer zones of SES fluid in helical tubes, as shown in Fig. 3. deviation of this relation is less than 10%, so the code adopted this
EI-Genk relation to estimate the critical Reynolds number.
2.3.2. Fluid heat transfer model
The total transfer heat (Q) from the PRS to the SES is: 2.3.4. Heat transfer relations in tube side
The mechanism of the flow and heat transfer in helical tubes is
Q = H⋅A⋅ΔT (17)
complex, especially in two-phase zones. The relations are quite different
In the heat transfer process, there are four thermal resistances: in predicting the thermal-hydraulic characteristics of helical tubes,
convective heat resistance between the primary fluid and the outer tube which need to be selected carefully according to the actual situation of
wall, heat conduction resistance of the tube wall, convective heat HCOTSGs and the applicable scope of models.
resistance between the SES fluid and the inner tube wall and fouling The code developed in this paper used the Schmiat relation [30] to
thermal resistance caused by impurities deposited on the inner tube calculate the single-phase heat transfer inside tubes, which has a wide
wall. If the inner side area of tubes is selected as the calculation area, the range of applications and basically covers all Reynolds number ranges
total heat (Hi) transferring from the PRS fluid to the SES fluid is: during the operation of HCOTSGs. For the two-phase zones, the heat
transfer performance of helical tubes is similar but not the same to that
1
Hi = (18) of straight tubes, so the heat transfer relation in two-phase zones is
1 di di
+ 2λ ln dd0i + h1s + Rx
hp do usually obtained by modifying boiling heat transfer relation in straight
tubes. The most widely modified straight tube relation is the Chen for­
where Rx is referred to the fouling resistance. It is about 2.6 × 10− 3-5.2 mula. In this simulation, the two-phase region is divided into two re­
× 10− 3 m2 K W− 1 for heat transfer tubes. gions in detail. The Zhao relation [32] is used for the subcooled boiling
zone, and the Chen relation [33] is used for the saturated nucleate
boiling and forced convection evaporation zone. Although Owhadi [4]
also proposed a relation for heat transfer in the saturated nucleate
boiling and forced convection evaporation zone of helical tubes based on
the Chen formula, the influence of helical diameters is not taken into
consideration. In order to avoid the discontinuity, the value of S in both
relations was modified from 0.1 to 0.0797.
In the liquid deficiency zone, the heat transfer may simultaneously
include convection, radiation and direct contact conduction, which has
a complex heat transfer mechanism. The Miropolskiy relation [34] was
selected for calculating the heat transfer coefficient of the fluid defi­
ciency zone in the simulation, which has been widely used by Xia [23],
Lian [35] and so on.

2.3.5. Heat transfer relation in shell side

In HCOTSGs, helix lead angles of tubes are usually only a few de­
grees, so the PRS fluid in shell sides scours tube bundles almost from a
transverse angle. The Zhukauskas heat transfer relation [36] is usually
chosen for the single-phase fluid scouring external bundles.

2.3.6. Frictional resistance coefficient relations in tube side

The Darcy equation, generally adopted to calculate the frictional
pressure drop in a single-phase flow, is as follows:

L ρV 2
ΔPf = f (19)
De 2

where f is the frictional resistance coefficient and L is the flow channel

length, m. The Darcy frictional resistance coefficient is needed before
the calculation of single-phase frictional pressure drop. Studies on
single-phase flow in helical tubes have been carried out in depth, and
these relations proposed by experts and scholars can predict the fric­
tional resistance of single-phase zones well. Colombo [37] compared
relations of the single-phase frictional coefficient with the data obtained
from the experiment on IRIS full-size helical tube and found that all the
relations are in good accord with the experimental data in the laminar
Fig. 3. The flow diagram of heat transfer zone judgment.

83
H. Yao et al. International Journal of Advanced Nuclear Reactor Design and Technology 3 (2021) 80–96

flow zone. For the turbulent zone, Ito relation [27] accords best with 3.1. Verification of two-phase frictional pressure drop calculation
experimental results. Its average error and maximum error are less than
4% and 6% respectively. Moreover, this relationship also has high pre­ In order to verify TACS’s ability to calculate the two-phase friction
cision in the transition zone of laminar and turbulent conversion. Since resistance in helical tubes, the code was used to simulate the two-phase
the Ito relation has corresponding forms in laminar and turbulent flow flow experiment conducted by Santini [43], and the calculated values
and its calculated values can reflect the real experimental results, the Ito were compared with the experimental value.
relation is employed for calculating the single-phase frictional resistance The experiment device of Santini is shown in Fig. 4. A helical tube
coefficient. In the low Reynolds number range which is out of the with 12.53 mm inner diameter, 1 m helix diameter, 0.8 m helical pitch
applicable range of Ito relation, Gupta relation [38] is used. And in the and 32 m length is arranged vertically, which can directly reproduce the
vacant range between these two relations, the frictional resistance co­ flow in the full-size HCOTSG tube. In this experiment, the pressure was
efficient is estimated by linear interpolation. 1.1–6.3 MPa, the mass flow rate was 192–824 kg m− 2 s− 1, and the heat
It is much more difficult to predict the frictional pressure drop of flux was 50–200 kW m− 2. The flow processes at 200 and 400 kg m− 2 s− 1
two-phase flows in a helical tube due to the phenomena of gas-liquid inlet mass flow rates were simulated by the TACS. Considering the
stratification and mixture, which is also researched by many scholars. experiment uncertainty and the calculation deviation of the model, 20%
Most of them used the two-phase frictional multiplier to calculate the error bars were set for the experimental data.
liquid phase frictional pressure drop gradient and then to gain the The variation trends of frictional pressure drop per unit length with
pressure drop. The Colombo relation [39] based on the multiplication steam quality at 200 kg m− 2⋅s− 1mass flow rate were calculated by the
factor of the frictional pressure drop in the liquid phase was adopted for TACS. Among them, the results under 2 MPa, 4 MPa and 6 MPa pressure
this simulation. Colombo started from a two-phase frictional resistance are compared with the experimental data in Fig. 5. When P = 2 MPa, the
coefficient for straight tubes, used the comparison of physical explana­ deviation is less than 20%, and becomes a little larger at 0.2–0.35 low
tion and experiment results to introduce an appropriate correction, and steam quality and 0.8–1.0 high steam quality. Although the applicability
finally got this relation. The effects of physical properties, mass flow of Colombo relation is relatively weak in these ranges, the maximum
rate, helical diameter and some other factors on local frictional resis­ deviation is not more than 40% and the average deviation is 16.4%. As
tance can be reflected in this relation. That is the reason why it can be for P = 4 MPa, the calculation values conform to the experiment with
applied in wide ranges of geometric parameters and working conditions. 14.0% maximum deviation and 5.2% average deviation. At 6 MPa
pressure, all the deviations are less than 20% except for a 21.8% devi­
2.3.7. Frictional resistance coefficient relation in shell side ation point, and the average deviation is 8.6%.
In shell sides, the PRS fluid scours tubes almost from a transverse The comparison at G = 400 kg m− 2 s− 1 is also drawn in Fig. 6. When
direction with no phase transition, and thus researches on its frictional P = 2 MPa, the average deviation is 10.1%. The deviation at the high
pressure drop are much less than that in tube sides. Gilli relation [40], steam quality (more than 0.9) is larger but still less than 33%. The de­
considering the influence of helix lead angle and tube row number, is viation of the rest range is no more than 20%. The code calculated values
applicable to both liquids and gases whose Pr>0.1. This relation, having fit the experimental data well with 10.1% average deviation and 17.2%
been used in HCOTSG simulations by Zhang [18], Lian [35], Ding [41] maximum deviation at 4 MPa pressure. As for 6 MPa, the average de­
et al., is also adopted in the TACS. viation and the maximum deviation are 7.6% and 20.5%.
To sum up, the processes of flow and heat transfer on both sides of In these two comparisons, the pressure drop part of the TACS is used
HCOTSG helical tubes can be calculated according to these relations. All to simulate the experiment of Santini. The calculated frictional pressure
the relations adopted in the TACS are summarized in Table 1. variation trend is in accord well with the experiment. The deviations
between them are almost less than 20%, which proves that the adopted
3. Code and verification model could simulate the complicated two-phase friction phenomenon
in helical tubes.
The fixed boundary method was adopted in the code to keep the grid
number and the boundary position in calculation processes. The sys­ 3.2. Verification of two-phase heat transfer calculation
tematic Thermal-hydraulic Analysis Code for helical coil once-through
Steam generator (TACS) developed in this paper employed the finite Another two-phase boiling heat transfer experiment conducted by
difference method to coupled solve the above HCOTSG physical models Santini [44] is also simulated by the heat transfer calculation part of the
and auxiliary models, the GEAR algorithm as the numerical solution TACS code to verify TACS’s calculation capacity of two-phase heat
method, and the Fortran language to compile. In the TACS, the outlet transfer in helical tubes. Generally speaking, the experimental section of
temperature of the PRS fluid and the inlet pressure of the SES fluid are the two-phase heat transfer experiment is relatively short. Based on the
assumed according to the given boundary conditions and thermal- experiment section mentioned above, Santini carried out a boiling heat
hydraulic parameters. The loop calculation, starting from the lowest transfer experiment of modular reactor helical tubes. In this experiment
control volume, runs upward until the results of power and outlet steam section, the first 24 m part was evenly heated by a direct-current power
pressure meet the error requirements. supply, modeling the boiling phenomenon before dry-out.
In recent decades, scholars have done a lot of researches on helical The comparison of the heat transfer coefficient between the tube-
tubes, of which researches on the single-phase flow heat transfer are inner fluid and the tube wall under 6 MPa pressure is shown in Fig. 7,
more thorough. Through comparison, scholars have found that the where the experimental data were also depicted with 20% error bars.
prediction ability of each single-phase model has little difference, and When G = 200 kg m− 2 s− 1 and q = 46 kW m− 2, the average deviation is
most of them are in good accord with the experimental data set. How­ 10.1% and only one point is of 20.5% deviation. Under the other two
ever, its two-phase characteristics are complicated. Although scholars conditions, the deviations of all the calculated values are less than 20%.
have made great progress and proposed a large number of relations of The maximum deviations and the average deviations of them are 16.6%,
two-phase heat transfer and friction pressure drop. These empirical or 6.8% and 18.2%, 6.5% in sequence. Therefore, the method and the
semi-empirical relations differ greatly, which is reflected by that they models adopted in TACS can nicely reflect the two-phase heat transfer
conform well within their own scopes but may not be suitable for other characteristics in helical tubes before dry-out.
conditions and geometric parameters. Therefore, two typical experi­ In the liquid deficiency zone after dry-out, the fluid is subject to the
ments are used to analyze the applicability of two-phase calculation by comprehensive interaction of gravity, centrifugal force and secondary
the TACS. flow phenomena, and the heat transfer phenomenon is complex with
many influencing factors, so it is difficult to predict accurately. Due to

84
 H. Yao et al.
Table 1
The adopted relations of flow, heat transfer and zone judgement.
Region Zone and transition point Heat transfer relation Frictional pressure drop coefficient relation
[ ] ⎧
Tube Subcooled Single- Schmiat ⎧ ( )0.9
di Ito relation [27], ⎪
⎪ 21.5De
( )0.5
Dc
a 1/3 ⎨ fs
side water zone phase zone relation [30] ⎪ 3.65 + 0.08 1 + 0.8 Re Pr , Re < Recr Gupta relation [38], , 13.5 < Re < Recr
⎪
⎪
⎪
⎪ Dc fc = {1.56 + log10 De} 5.73 di
⎪
⎪
⎪ [ ] and linear ⎪
⎪
⎩ − 0.25 − 0.5
⎪
⎨ (
di
)( )
di 1.3 interpolation 0.304Re + 0.029(Dc /di ) , Re > Recr
Nu = 0.023 1 + 14.8 1 + Reb Pr1/3 , Recr < Re < Re1 ( )2
⎪ Dc Dc Dc
⎪
⎪
⎪ 5 < Dc /di < 370 fc = fs , Re < 0.034
⎪
⎪
⎪ [ ( )( )0.8 ] di
⎩ 0.023 1 + 3.6 1 − di di
⎪
⎪
Re0.8 Pr1/3 , Re1 < Re < Re2
Dc Dc
( )0.194 ( )0.1
di di
where a = 0.5 + 0.2903 , b = 0.8 − 0.22 , Re1 = 22000 and Re2 =
Dc Dc
150000
Critical EI-Genk 0.2 di Dc
Recr = 2300[1 + 5.164 × 104 λ1.575
m ] where λm = and Rc =
Reynolds relation [31] 2Rc 2[1 + (b/πDc )2 ]
number
⎛
Bubble disengagement Saha-Zuber qdi Colombo relation ( ) ( ) (
dp dp C
starting point relation [42] ⎜ Tsat − 0.0022 k , Pe ≤ 70000 [39] = φ2l where φ2l = 0.0986 1 + +
TFDB = ⎜
⎜ l dz tp dz l χ
⎝ q
Tsat − 154 , Pe > 70000 )[ ( )0.5 ]0.19 ( )− 0.4
Gcpl 1 di ρm
Subcooled boiling zone Zhao relation
( ) ( ) Rel
Gm di 0.85 0.4 kl di 0.1 χ2 Dc ρl
[32] h = 0.023 Prl + 0.00122×
μl di Dc
( 0.79 0.45 0.49 ) where S =
kl cpl ρl Tw − Tsat
85

0.24
S 0.5 0.24 0.24 0.29 ΔTsat ΔP0.75
sat
σ hfg ρg μl Tw − Tl
⎧

International Journal of Advanced Nuclear Reactor Design and Technology 3 (2021) 80–96
⎨ 1/[1 + 0.12(ReTP )1.14 ], ReTP < 32.5 (
Gm di
)
1/[1 + 0.42(Re TP )0.78
], 32.5 ≤ Re TP < 70.0 and ReTP = × 10− 4
⎩ μl
0.1, ReTP [ ≥ 70.0 ]0.85 ( )0.1
Saturated nucleate boiling Chen relation Gm (1 − x)di kl di
and forced convection [33] h = 0.023F Prl 0.4 +
μl di Dc
evaporation zone ( 0.79 0.45 0.49 ) where F =
kl cpl ρl 0.24
0.00122S 0.5 0.24 0.24 0.29 ΔTsat ΔP0.75
sat
σ hfg ρg μl
⎧
⎨ 1.0, X− 1 ≤ 0.1 ( x )0.9 ( ρ )0.5 (μ )0.1
tt l g
, X− 1 = ,S =
⎩ 2.35(X− 1 + 0.213)0.736 , X− 1 > 0.1 tt 1− x ρg μl
tt tt
⎧
⎨ 1/[1 + 0.12(ReTP )1.14 ], ReTP < 32.5 [ ]
Gm (1 − x)di 1.25
0.78 , and ReTP = F × 10− 4
⎩ 1/[1 + 0.42(ReTP ) ], 32.5 ≤ ReTP < 70.0 μl
0.1, ReTP ≥ 70.0
Dry-out point Santini xDO = (0.44 − 0.006p)G0.114 G: 200–800 kg m− 2 s− 1; q: 50–200 kW m− 2; P: 2.0–6.0 MPa
relation [25]
( )0.85 [ ]0.85 ( )0.01
Liquid deficiency zone Miropolskiy Gdi ρg di
relation [34] Nu = 0.023Y x+ (1 − x) Pr0.8
w where Y = 1 −
μg ρl Dc
( )0.4
ρ
0.1 l − 1 (1 − x)0.4
ρg
( )0.25
Shell Single-phase zone Zhukauskas Gilli relation [40] nG2
0.36 Prl where Ci = (cos β)− 1.8 (cos ϕ)1.355 , Cn =
Δp = 0.334feff Ci Cn
side relation [36] Nu = CRem
l Prl
Prw 2ρ
0.375 ( α)
1+ ,β =α 1− , and ϕ = α + β
n 90
H. Yao et al. International Journal of Advanced Nuclear Reactor Design and Technology 3 (2021) 80–96

Fig. 4. The experiment section of Santini [43].

2 1
Fig. 5. The comparison of frictional pressure drop per unit length at G = 200 kg m− s−

86
H. Yao et al. International Journal of Advanced Nuclear Reactor Design and Technology 3 (2021) 80–96

2 1
Fig. 6. The comparison of frictional pressure drop per unit length at G = 400 kg m− s−

Fig. 7. The comparison of heat transfer coefficient at 6 MPa pressure.

87
H. Yao et al. International Journal of Advanced Nuclear Reactor Design and Technology 3 (2021) 80–96

the lack of public experimental data on the heat transfer phenomenon in Table 2
the liquid deficiency zone, Ref. [45] was used to analyze the Santini The key design parameters of HCOTSGs of the IRIS.
relation of steam quality at the dry-out point. It found that compared Items Values
with other relations, Santini relation can better predict the experimental
Helical structure parameters
data set, and can be used in the thermal-hydraulic analysis of HCOTSGs Number of tubes 655
in small-medium modular reactors (SMRs). Material of tubes Inconel-690
Inner diameter of tubes/m 0.01324
Outer diameter of tubes/m 0.01746
4. Thermal-hydraulic analyses
Length of tubes/m 32
Coil number of tubes 20
The International Reactor Innovative and Secure (IRIS) is a typical Radial pitch/m 0.025
SMR, as shown in Fig. 8. Its vital parameters for designing HCOTSG are Axial pitch/m 0.025
listed in Table 2. Taking the steam generator in the IRIS as the analysis Innermost helical diameter/m 0.64
Outermost helical diameter/m 1.59
object, the TACS was used to study the thermal and hydraulic charac­ Operation parameters
teristics of a HCOTSG system. Power/MWt 125
The grid independence analysis decided to use 150 grids in this Inlet flow rate of PRS fluid/kg⋅s− 1 589.0
simulation. The TACS results and the design parameters of the IRIS are Inlet flow rate of SES fluid/kg⋅s− 1 62.85
Inlet temperature of PRS fluid/K 601.55
listed in Table 3. It can be seen that the overall results fit well. Except for
Inlet pressure of PRS fluid/MPa 15.5
the PRS pressure drop, the absolute values of the deviation of these key Inlet temperature of SES fluid/K 497.05
parameters are all less than 0.4%. For 15.5 MPa PRS pressure of the Outlet pressure of SES steam/MPa 5.8
steam generator, 0.072 MPa pressure drop is relatively very small and
not much different with the value of 0.068 MPa calculated by the TACS.
Table 3
4.1. Thermal-hydraulic calculation results at full load The result comparison of IRIS stream generator.
IRIS operation parameters Design TACS
In addition to the parameters above, TACS can also provide spatial values
Calculation Deviation/
distributions of some physical quantities in HCOTSGs, such as SES heat results %
transfer zone, heat transfer coefficient, thermal resistance, temperature, Power/MW 125 124.97 − 0.02
SES steam quality and void fraction, heat flux, pressure, pressure drop Heat transfer area of SES/m2 872 872.70 0.08
et al., which can be used to predict and understand thermal-hydraulic Outlet temperature of PRS 565.15 564.98 − 0.03
characteristics of HCOTSGs. Some key calculated results are pictured fluid/K
Outlet temperature of SES 590.15 588.33 − 0.31
in Fig. 9. Its horizontal coordinate of 0–32 m is the tube length from
fluid/K
bottom to top, which is the same direction of the SES fluid flow and the Pressure drop of PRS fluid/ 0.072 0.068 − 5.56
opposite direction of the PRS fluid flow. As is shown in Fig. 9 (b), the MPa
thermal resistances of the PRS and the tube wall change little in the Pressure drop of SES fluid/ 0.296 0.295 − 0.34
whole flow process, while the SES thermal resistance varies constantly MPa

and determines the changing trend of total thermal resistance. Before

the liquid deficiency zone, the wall thermal resistance is the largest, resistance increases dramatically and becomes the main thermal resis­
which has a great impact on the heat transfer performance of the tance. It needs to be mentioned that the frictional pressure drop relation
HCOTSG. After the heat transfer deteriorates, the secondary thermal of helical tubes has included the influence of local resistance caused by
tube bends, so the local pressure drop is treated as a part of the frictional
pressure drop.

4.2. Influences of thermal parameters on thermal-hydraulic

characteristics

On the basis of design parameters in Table 2, the mass flow rate of

PRS fluid (Wpr), mass flow rate of SES fluid (Wse), the inlet temperature
of PRS fluid (Tinpr) and the inlet temperature of SES fluid (Tinse) were
altered in order to study the influences of these key factors on thermal-
hydraulic systematic characteristics of IRIS.

4.2.1. Influence of primary mass flow rate

The mass flow rate of the PRS inlet changed to 85% and 70% load
respectively while other parameters remained unchanged. The TACS
calculated results are depicted as Fig. 10.

(a) When the mass flow rate of PRS fluid decreases, the heat transfer
performance is weakened, and the PRS heat transfer coefficient
reduces. However, no phase transition occurs on the PRS, and the
heat transfer coefficient on the PRS remains almost unchanged
with the tube length. For the SES, the change of heat transfer
coefficient between the single-phase zone and the liquid-deficient
zone is not obvious. In the subcooled boiling zone, the bubble
perturbation is weakened and the heat transfer coefficient falls
Fig. 8. The overall layout of IRIS [46]. due to the decrease of the heat supplied for the SES fluid. The

88
H. Yao et al. International Journal of Advanced Nuclear Reactor Design and Technology 3 (2021) 80–96

Fig. 9. The key thermal-hydraulic results of IRIS.

variation of the maximum heat transfer coefficient is not obvious, primary flow rate, the variation amplitude is very small, so the
but the position of the dry-out point appears backer with a less SES saturation temperature is almost the same.
mass flow rate of PRS fluid.
(b) As the mass flow rate of PRS fluid reduces, HCOTSG heat transfer 4.2.2. Influence of secondary mass flow rate
quantity and the heating degree on the SES decrease, leading to In this turn, the mass flows of SES fluid are 100%, 85% and 70% of
flow velocity reducing and pressure drop falling. Since HCOTSG full load mass flow respectively. The key results obtained by TACS are
adopts a constant steam pressure scheme, the SES inlet pressure shown in Fig. 11.
needs to be reduced to keep the same outlet pressure.
(c) As the mass flow rate of PRS fluid decreases, the heat capacity of (a) The reduction of secondary mass flow rate has little effect on PRS
the PRS fluid shrinks and the cooling degree of the PRS fluid in­ properties, so the PRS heat transfer coefficient is almost un­
creases, so the temperature drop rate is accelerated and the outlet changed. However, the secondary velocity drops, the Reynolds
temperature is lessened. number becomes smaller, and the heat transfer coefficient in the
(d) The SES liquid is heated to saturated water and then to be single-phase zone drops. As the heat capacity of the SES fluid
completely vaporized. The heat required for the SES fluid in this decreases, the temperature rise rate is accelerated, and the bubble
process is unchanged if the primary flow rate is cut down. The perturbation in the subcooled boiling zone is strengthened, the
SES heat absorption rate slows down, resulting in the extension of heat transfer coefficient enhances. The heat required for total
the single-phase water and two-phase zone and the outlet tem­ evaporation is reduced, the two-phase region is signally short­
perature reduction. Even if the primary flow rate drops to 70%, ened, and the dry-out point occurs earlier. The heat transfer co­
the fluid is still saturated state and fails to become steam. efficient falls down greatly in the saturated nucleate boiling and
Although the SES pressure drop reduces with the decrease of the

89
H. Yao et al. International Journal of Advanced Nuclear Reactor Design and Technology 3 (2021) 80–96

Fig. 10. The calculated results of primary flow rate variation.

Fig. 11. The calculated results of secondary flow rate variation.

90
H. Yao et al. International Journal of Advanced Nuclear Reactor Design and Technology 3 (2021) 80–96

forced convection evaporation zone, so the maximum heat the second side is enhanced, and the single-phase liquid zone and
transfer coefficient in the whole heating process decreases. two-phase zone of the second side are shortened. Furthermore,
(b) The reduction of secondary mass flow rate leads to a smaller SES the bubble perturbation increases the SES heat transfer coeffi­
pressure drop. In addition to the reduction of heat transfer cient in the subcooled boiling zone, while the maximum coeffi­
quantity, the decrease amplitude of SES pressure drop is larger cient increases only a little.
than that of the case of the primary mass flow rate decrease. (b) The greater the PRS inlet temperature is, the greater the heat
(c) As the cooling degree of the PRS fluid lowers when the SES mass exchange quantity between the PRS and SES of HCOTSG will be.
flow rate of the SES fluid decreases, the temperature decline rate Under this condition, the heating degree of the SES fluid is higher,
slows down, so the outlet temperature of the PRS fluid rises. the SES pressure drop rate becomes faster, and the pressure drop
(d) Even if the mass flow rate of the SES fluid decreases, the mass becomes greater.
flow rate and the temperature of the PRS fluid do not change. The (c) The PRS mass flow rate is large, about 9 times that on the SES, so
temperature rise of the SES fluid is accelerated, and the heat the PRS temperature difference is much smaller than that of the
required for the heating and evaporation process of all liquid SES. Therefore, the PRS inlet temperature has little influence. On
being converted into steam is reduced. The length of the single- the primary side, the variation trend of primary outlet tempera­
phase liquid and the two-phase zone is shortened, while the ture is positively correlated with that of inlet temperature.
single-phase steam zone is lengthened with a phenomenon of (d) The PRS inlet temperature will obviously change the length of the
outlet temperature rise. The SES saturation temperature reduces SES two-phase zone. When the PRS inlet temperature increases
slightly because of the dramatic decline of the SES pressure drop. by 15K, the SES two-phase zone is shortened by nearly 36% and
When the mass flow rate of the SES fluid drops from 85% to 70%, the superheated steam zone is prolonged by about 9 m, resulting
the SES outlet temperature only increases by 1.5K. It can be seen in the outlet steam temperature rising by 27.5k. In the case of the
that the steam superheat degree goes up a little with the decrease PRS inlet temperature 15K lower, the SES heat absorption rate
of the SES mass flow rate. slows down, and fails to generate superheated steam.

4.2.3. Inlet temperature of primary side fluid 4.2.4. Inlet temperature of secondary side fluid
The PRS inlet temperature is 601.55K at full load and the cases with The SES inlet temperature is 497.05K at full load and the cases with
15K lower and 15K higher ones are simulated by the TACS respectively. 15K lower and 15K higher ones are also simulated by the TACS
The calculated results are also pictured as Fig. 12. respectively. The calculated results are also drawn as Fig. 13. The
change of SES inlet temperature has very little influence on the PRS heat
(a) When the PRS inlet temperature rises, the PRS heat transfer co­ transfer coefficient. Although it will change the Prandtl number of the
efficient also increases, but the phase transition of the PRS fluid SES fluid, the centrifugal force and the secondary flow phenomenon play
does not happen. Thereby, the physical properties and the heat a role to enhance the heat transfer performance. In sum, the inlet tem­
transfer coefficient of the PRS vary very little. The temperature perature of the SES fluid also has little influence on the SES heat transfer
difference of both sides expands, the heat absorption capacity of coefficient. The rise of secondary inlet temperature leads to the decrease

Fig. 12. The calculated results of primary inlet temperature variation.

91
H. Yao et al. International Journal of Advanced Nuclear Reactor Design and Technology 3 (2021) 80–96

Fig. 13. The calculated results of secondary inlet temperature variation.

Fig. 14. The calculated results of inner diameter variation.

92
H. Yao et al. International Journal of Advanced Nuclear Reactor Design and Technology 3 (2021) 80–96

of inlet subcooled degree, the shortening of the single-phase liquid zone enhancement. Therefore, with the increase of internal pipe diameter, the
and the earlier appearance of the two-phase zone, so the superheated SES pressure drop decreases, as shown in Fig. 14 (b), which also leads to
steam zone becomes longer and the SES pressure drop increases slightly. a slight decrease in the SES saturation temperature in Fig. 14 (d).
The temperature difference between both sides shrinks and the heat
transfer weakens, which results in diminutions of the temperature drops 4.3.2. Outer diameter
of both sides. In general, the secondary inlet temperature has less in­ The outer diameters were taken as 17.46 mm (design value), 15.96
fluence on HCOTSG thermal-hydraulic characteristics than other ther­ mm (1.5 mm smaller) and 18.96 mm (1.5 mm larger) when other pa­
mal parameters mentioned above. rameters were not changed. The key calculated results are also pictured
as Fig. 15.
4.3. Influences of geometric parameters on thermal-hydraulic With the increase of outer diameter, the PRS heat transfer coefficient
characteristics decreases but very slightly. In this case, a thicker tube wall leads to the
increase of tube wall thermal resistance. By the way, it has been
This section studies the influences of these key geometric parameters mentioned that the tube wall thermal resistance dominates the heat
on the thermal-hydraulic characteristics of IRIS HCOTSGs by changing transfer process before the dry-out point in Fig. 9 (b). Even if the PRS
the inner diameter (di), the outer diameter (do) and the helical diameter heat transfer area expands, the heat transfer quantity still decreases with
(Dc) of the helical tube on Table 2. the increase of the outer diameter. The single-phase liquid zone and the
two-phase zone on the SES are lengthened, and the increase rate of heat
4.3.1. Inner diameter transfer coefficient in the two-phase zone slows down. The slight decline
The inner diameters were taken as 13.24 mm (design value), 11.74 of the SES pressure drop, as shown in Fig. 15 (b), almost does not affect
mm (1.5 mm smaller) and 14.74 mm (1.5 mm larger) when other pa­ SES saturation temperature.
rameters were not changed. The key calculated results are depicted as
Fig. 14. 4.3.3. Helical diameter
The variation of inner diameter has little influence on the PRS heat The 1.115 m, the average value of the helical diameter of IRIS
transfer coefficient, but it can obviously affect the distribution of the SES HCOTSGs, was taken as the basis simulation case. Some additional cases
heat transfer coefficient. The larger inner diameter leads to the retard of with 0.25 m helical diameter differences were also calculated by the
the SES fluid velocity, and the fall of the heat transfer coefficient in the TACS. The key results of these cases are drawn as Fig. 16.
single-phase zone and the liquid deficiency zone. However, the thermal The PRS heat transfer coefficient is almost unchanged with different
resistance of the tube wall diminishes and the heat transfer area ex­ helical diameters. In order to avoid confusion, only the heat transfer
pands, which are the dominant factors. As a consequence, the heat coefficient distribution on the SES is given in Fig. 16 (a). With the in­
transfer is promoted, and the single-phase liquid zone and the two-phase crease of helical diameter, the SES heat transfer coefficient does not
zone are shortened. The temperature drops of the PRS and the SES in­ change much in the single-phase zone, but decreases slightly in the two-
crease as shown in Fig. 14 (c) and (d), but the variation amplitude phase zone, especially in the portion of high steam quality. This is
former is relatively small due to the large mass flow rate of PRS fluid. because the SES fluid is more affected by centrifugal force which be­
The flow velocity slowing down makes the SES pressure drop reduce, comes smaller with a larger helical diameter. The helical diameter
and this effect is greater than the pressure drop caused by heat transfer enlargement slightly weakens the HCOTSG heat transfer capacity, as

Fig. 15. The calculated results of outer diameter variation.

93
H. Yao et al. International Journal of Advanced Nuclear Reactor Design and Technology 3 (2021) 80–96

Fig. 16. The calculated results of helical diameter variation.

shown in Fig. 16 (b). The fluid physical properties in Fig. 16 (c) and (d) HCOTSG tubes is taken as the characteristic helical diameter.
only change a little. All of these show that thermal-hydraulic charac­
teristics are not drastically sensitive to the helical diameter, and thus it is Declaration of competing interest
reasonable for simulation calculations to adopt the method that the
average helical diameter of all the HCOTSG tubes is taken as the char­ The authors declare that they have no known competing financial
acteristic helical diameter. interests or personal relationships that could have appeared to influence
the work reported in this paper.
5. Conclusions
Nomenclature
The optimal physical models were selected for different heat transfer
areas by detailed investigation and comparison, to adopt in the TACS, a A sectional area of flow, m2
self-developed thermal-hydraulic analysis code that is accurate and Ap heat transfer area of primary fluid in tube wall, m2
widely-used in HCOTSG simulation. After verification with experiments As heat transfer area of secondary fluid in tube wall, m2
of Santini, taking the IRIS reactor as the object, the TACS was used to cpw constant pressure specific heat capacity of tube, J⋅kg− 1⋅K− 1
briefly study the thermal-hydraulic characteristics of the HCOTSG sys­ De equivalent hydraulic diameter, m
tem at first, and then discuss the influences of thermal and geometric di inner diameter of tube, m
parameters on the HCOTSG. The analysis results of thermal parameters do outer diameter of tube, m
showed that: the variation of the primary flow rate has the greatest in­ f coefficient of frictional resistance
fluence on the PRS heat transfer coefficient, while the maximum value of G mass velocity, kg⋅m− 2⋅s− 1
the SES heat transfer coefficient only changes greatly when the sec­ g gravitational acceleration, m⋅s− 2
ondary flow rate changes; when the PRS flow drops to 70%, superheated Hi total heat transfer coefficient, W⋅m− 2⋅K− 1
steam is not generated on the SES; when the secondary flow rate de­ h specific enthalpy, J⋅kg− 1
creases to a certain extent, the steam superheat degree increases by a hg specific enthalpy of saturated steam, J⋅kg− 1
small amplitude; the PRS inlet temperature has a little influence on the hl specific enthalpy of saturated liquid, J⋅kg− 1
primary temperature difference, but a significant influence on the two- hp heat transfer coefficient between primary side fluid and tube
phase zone length of the SES (the two-phase zone is shortened by nearly wall, W⋅m− 2⋅K− 1
36% when the temperature rises by 15K, and the secondary outlet fluid hs heat transfer coefficient between secondary side fluid and
is still saturated water when the temperature drops by 15K); compared tube wall, W⋅m− 2⋅K− 1
with the above three parameters, the secondary inlet temperature has L length of flow channel, m
very little influence on the thermal-hydraulic characteristics of HCOTSG m mass of tube, kg
system. The analysis results of geometric parameters showed that: the P pressure, Pa
inner diameter and the outer diameter of helical tubes have obvious Pe Peclet number, Pe = Gdicpl/kl
influences, while the helical diameter has not, which proves that the q heat flux, W⋅m− 2
reasonability of the method that the average helical diameter of all the Rx fouling resistance, m2⋅K⋅W− 1

94
H. Yao et al. International Journal of Advanced Nuclear Reactor Design and Technology 3 (2021) 80–96

t time, s [13] H. Mirgolbabaei, Numerical investigation of vertical helically coiled tube heat
exchangers thermal performance, Appl. Therm. Eng. 136 (2018) 252–259.
TFDB fluid temperature at the point of bubbles starting leaving wall,
[14] F. Castiglia, M. Giardina, G. Morana, et al., Analyses of single- and two-phase flow
K pressure drops in helical pipes using a modified RELAP5 code, Nucl. Eng. Des. 250
TINpr inlet temperature of primary side fluid, K (2012) 585–591.
TINse inlet temperature of secondary side fluid, K [15] J. De Amicis, A. Cammi, L.P.M. Colombo, et al., Experimental and numerical study
of the laminar flow in helically coiled pipes, Prog. Nucl. Energy 76 (2014)
Tpr temperature of primary side fluid, K 206–215.
Tse temperature of secondary side fluid, K [16] G. Vaidyanathan, A.L. Kothandaraman, L.S.S. Kumar, et al., Development of one-
Tsat saturation temperature of fluid under current pressure, K dimensional computer code DESOPT for thermal hydraulic design of sodium-
heated once through steam generators, Int. J. Nucl. Energy Sci. Technol. 5 (2)
Tw average temperature of tube wall, K (2010).
Tw1 primary side temperature of tube wall, K [17] D. Colorado, D. Papini, J.A. Hernández, et al., Development and experimental
Tw2 secondary side temperature of tube wall, K validation of a computational model for a helically coiled steam generator, Int. J.
Therm. Sci. 50 (4) (2011) 569–580.
U perimeter of heating, m [18] Y. Zhang, D. Wang, J. Lin, et al., Development of a computer code for
Vg velocity of gas, m⋅s− 1 thermal–hydraulic design and analysis of helically coiled tube once-through steam
Vl velocity of liquid, m⋅s− 1 generator, Nucl. Eng.Technol. 49 (7) (2017) 1388–1395.
[19] X. Yu, B. Sun, Y. Li, et al., Numerical investigation of thermal-hydraulic parameter
W mass flow of fluid, kg⋅s− 1 distribution characteristics during dryout evolution in the helically coiled once-
Wpr mass flow of primary side fluid, kg⋅s− 1 through steam generator, Int. J. Heat Mass Tran. 139 (2019) 373–385.
Wse mass flow of secondary side fluid, kg⋅s− 1 [20] J. Yoon, J.-P. Kim, H.-Y. Kim, et al., Development of a computer code, ONCESG, for
the thermal-hydraulic design of a once-through steam generator, J. Nucl. Sci.
x steam quality
Technol. 37 (5) (2012) 445–454.
z coordinate in the direction of fluid flow, m [21] Y. Ma, X. Li, X. Wu, Thermal–hydraulic characteristics and flow instability analysis
of an HTGR helical tube steam generator, Ann. Nucl. Energy 73 (2014) 484–495.
[22] R. Xu, D. Zhang, W. Tian, et al., Thermal-hydraulic analysis code development for
Greek symbols sodium heated once-through steam generator, Ann. Nucl. Energy 127 (2019)
385–394.
α void fraction
[23] G. Xia, Y. Yuan, M. Peng, et al., Numerical studies of a helical coil once-through
δ tube thickness, m steam generator[J], Ann. Nucl. Energy 109 (2017) 52–60.
λ thermal conductivity of tube wall, W⋅m− 1⋅K− 1
[24] L. Guo, X. Chen, C. Xu, et al., Forced convection boiling heat transfer and dryout
λm modified helical curvature ratio characteristics in helical coiled tubes with various axial angles[C], Heat Transfer
Conference (1998).
μ dynamic viscosity, Pa⋅s− 1 [25] L. Santini, A. Cioncolini, C. Lombardi, et al., Dryout Occurrence in a Helically
μg dynamic viscosity of saturated steam, Pa⋅s− 1 Coiled Steam Generator for Nuclear Power application[C], EPJ Web of
μl dynamic viscosity of saturated liquid, Pa⋅s− 1 Conferences, 2014.
[26] A.E. Ruffell, The application of heat transfer and pressure drop data to the design of
ρ density of fluid, kg⋅m− 3 helical coil once-through boilers[C], IChemE. Symp. Ser. (1974).
ρg density of saturated steam, kg⋅m− 3 [27] H. Ito, Friction factors for turbulent flow in curved pipes, J. Basic Eng. 81 (2)
ρl density of saturated liquid, kg⋅m− 3 (1959) 123–132.
[28] S.S. Kutateladze, V.M. Borishanskiĭ, A Concise Encyclopedia of Heat transfer[M],
Pergamon, 1966.
[29] V. Kubair, N.R. Kuloor, Heat transfer to Newtonian fluids in coiled pipes in laminar
Acronyms flow, Int. J. Heat Mass Tran. 9 (1) (1966) 63–75.
FDB fully developed nucleate boiling [30] E.F. Schmidt, Wärmeübergang und Druckverlust in Rohrschlangen, Chem. Ing.
HCOTSG helical coil once-through steam generator Tech. 39 (1967) 781–789.
[31] M.S. Ei-Genk, T.M. Schriener, A review and correlations for convection heat
IRIS International Reactor Innovative and Secure
transfer and pressure losses in toroidal and helically coiled tubes, Heat Tran. Eng.
PRS primary side 38 (5) (2017) 447–474.
SES secondary side [32] Z. Zhao, R. Zhu, Reactor Thermo-hydraulics[M], Tsinghua University Press,
Beijing, 1992.
SMR small-medium modular reactor
[33] J.C. Chen, Correlation for boiling heat transfer to saturated fluids in convective
flow, Ind. Eng. Chem. Process Des. Dev. 5 (3) (1966) 322–329.
References [34] Z.L. Miropolskiy, Heat transfer in film boiling of a steam-water mixture in steam-
generator tubes, Teplonergetika 10 (1963) 49–52.
[35] Q. Lian, W. Tian, X. Gao, et al., Code improvement, separate-effect validation, and
[1] P.E. MacDonald, V.N. Shah, L.W. Ward, et al., Steam Generator Tube failures[R],
benchmark calculation for thermal-hydraulic analysis of helical coil once-through
Nuclear Regulatory Commission, Washington, DC, US, 1996.
steam generator, Ann. Nucl. Energy (2020) 141.
[2] H. Zhu, Z. Li, X. Yang, et al., Flow regime identification for upward two-phase flow
[36] A. Žukauskas, Heat Transfer from Tubes in Crossflow[M], 1972, pp. 93–160.
in helically coiled tubes, Chem. Eng. J. 308 (2017) 606–618.
[37] M. Colombo, Experimental Investigation and Numerical Simulation of the Two-
[3] J. Eustice, Flow of water in curved pipes, in: Proceedings of the Royal Society of
phase Flow in the Helical Coil Steam Generator[D], Politecnico Di Milano, 2013.
London. Series A, Containing Papers of a Mathematical and Physical Character 84,
[38] R. Gupta, R. Wanchoo, A.T. Jafar, Laminar flow in helical coils: a parametric study,
1910, pp. 107–118, 568.
Ind. Eng. Chem. Res. 50 (2) (2011) 1150–1157.
[4] A. Owhadi, K.J. Bell, B. Crain Jr., Forced convection boiling inside helically coiled
[39] M. Colombo, L.P.M. Colombo, A. Cammi, et al., A scheme of correlation for
tubes, Int. J. Heat Mass Tran. 11 (12) (1968) 1779–1793.
frictional pressure drop in steam–water two-phase flow in helicoidal tubes[J],
[5] M.A. Styrikovich, V.S. Polonsky, V.V. Reshetov, Experimental investigation of the
Chem. Eng. Sci. 123 (2015) 460–473.
critical heat flux and post-dryout temperature regime of helical coils, Int. J. Heat
[40] P.V. Gilli, Heat transfer and pressure drop for cross flow through banks of
Mass Tran. 27 (8) (1984) 1245–1250.
multistart helical tubes with uniform inclinations and uniform longitudinal pitches,
[6] D. Austen, H. Soliman, Laminar flow and heat transfer in helically coiled tubes with
Nucl. Sci. Eng. 22 (3) (1965) 298–314.
substantial pitch, Exp. Therm. Fluid Sci. 1 (2) (1988) 183–194.
[41] H. Ding, Y. Zhang, K. Ye, et al., Development of a model for thermal-hydraulic
[7] G. Berthoud, S. Jayanti, Characterization of dryout in helical coils, Int. J. Heat Mass
analysis of helically coiled tube once-through steam generator based on Modelica,
Tran. 33 (7) (1990) 1451–1463.
Ann. Nucl. Energy (2020) 137.
[8] S. Ali, Pressure drop correlations for flow through regular helical coil tubes, Fluid
[42] Y. Changqi, Gas-liquid Two-phase flow[M], Harbin Engineering University Press,
Dynam. Res. 28 (4) (2001) 295–310.
Harbin, 1995.
[9] C. Bertani, M. De Salve, M. Malandrone, et al., in: Studio sperimentale del deflusso
[43] L. Santini, A. Cioncolini, C. Lombardi, et al., Two-phase pressure drops in a
bifase aria-acqua in un tubo elicoidale, XXV, Congresso Nazionale UIT sulla
helically coiled steam generator, Int. J. Heat Mass Tran. 51 (19–20) (2008)
Trasmissione del Calore, Trieste, Trieste, 2007, pp. 18–20.
4926–4939.
[10] L. Guo, F. Ziping, X. Chen, An experimental investigation of the frictional pressure
drop of steam–water two-phase flow in helical coils, Int. J. Heat Mass Tran. 44 (14)
(2001) 2601–2610.
[11] L. Zhao, L. Guo, B. Bai, et al., Convective boiling heat transfer and two-phase flow
characteristics inside a small horizontal helically coiled tubing once-through steam
generator, Int. J. Heat Mass Tran. 46 (25) (2003) 4779–4788.
[12] K. Ye, Y. Zhang, X. Sheng, et al., Numerical analysis of the flow behavior in a
helically coiled once through steam generator, Nucl. Eng. Des. 330 (2018)
187–198.

95
H. Yao et al. International Journal of Advanced Nuclear Reactor Design and Technology 3 (2021) 80–96

[44] L. Santini, A. Cioncolini, M.T. Butel, et al., Flow boiling heat transfer in a helically [46] M. Carelli, L. Conway, M. Dzodzo, et al., The SPES3 Experimental Facility Design
coiled steam generator for nuclear power applications, Int. J. Heat Mass Tran. 92 for the IRIS Reactor Simulation vol. 2009, Science and Technology of Nuclear
(2016) 91–99. Installations, 2009, pp. 1–12.
[45] J. Gou, H. Ma, Z. Yang, et al., An assessment of heat transfer models of water flow
in helically coiled tubes based on selected experimental datasets, Ann. Nucl.
Energy 110 (2017) 648–667.

96