Вы находитесь на странице: 1из 14

Thermal Science and Engineering Progress 5 (2018) 158–171

Contents lists available at ScienceDirect

Thermal Science and Engineering Progress


journal homepage: www.elsevier.com/locate/tsep

Numerical and experimental analysis on shell side thermo-hydraulic T


performance of shell and tube heat exchanger with continuous helical FRP
baffles

Sunil Shinde , Umesh Chavan
Department of Mechanical Engineering, Vishwakarma Institute of Technology, Pune MS-411037, India

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

Keywords: The present work deals with the numerical and experimental investigation of heat and fluid flow in shell and
Shell-and-tube heat exchanger tube heat exchanger with continuous helical baffles on shell side. Seven helix angles, namely 10°, 19°, 21°, 25°,
Helix angle 30°, 38°, and 50° are investigated numerically by modeling a full length continuous helical baffled shell-and-tube
FRP baffle heat exchanger for different mass flow rates and inlet temperature conditions. Results revealed that the larger
Heat transfer coefficient
helix angles (30°, 38° and 50°) adds to lower heat transfer and lower pressure drop, and smaller helix angles (10°,
Pressure drop
19°and 21°) resulted in higher heat transfer as well as higher pressure drop. The experiments were carried out on
shell and tube heat exchanger for helical baffles with 25° helix angle and results were compared with segmental
baffles. New baffle material, FRP (Fiber Reinforced Plastic) is introduced on the shell side of the proposed heat
exchanger. For material comparison between FRP and stainless steel, the deviation of heat transfer coefficient is
8–10%, which can serve as a potential replacement for conventional baffle material, thereby reducing the capital
and operating cost of the tubular heat exchanger.

1. Introduction helical baffles.


Lutcha and Nemcansky [10] in their investigation on heat and fluid
Heat exchangers such as shell-and-tube, plate type and finned tube flow in a helical baffle heat exchanger concluded that appropriate se-
are used in various industries for different applications such as heating, lection of baffle inclination angle will result in enhancement of overall
cooling, condensation or evaporation process. Most of the modern day temperature by the reduction in back mixing. Kral et al. [11] found the
heat exchangers are provided with baffles which help in enhancing the overall performance of the heat exchangers with different helical baffle
heat transfer. Baffles also helps in providing structural rigidity for tubes configurations.
and preventing corrosion and acts as guide ways for the flow across Table 1 shows the research carried out on heat exchangers by
bundles of tube to obtain higher heat transfer rate. In the case of seg- varying baffles angles and implementing different baffle styles.
mental baffles the flow along the shell side is very complex due to back Having realized the advantages of helical baffles, many researchers
mixing, bypass flow and cross flow phenomenon. Also segmental baffles tried to study the heat exchanger for coming up with an optimum de-
effect in a significant pressure drop transversely to the heat exchanger sign solution which will give maximum heat transfer coefficient for a
due to rapid change in direction, expansion, and contraction. To unit pressure drop. Basically, researchers tried to analyze the design of
overcome these difficulties of segmental baffle heat exchanger various the heat exchanger by adopting various approach. Initially after in-
deflectors, rod baffles, sealing strips are used, however drawbacks of vention of the helical baffle, much of the analysis was carried out ex-
the segmental baffle structure still exist [1–7]. In recent years a novel perimentally. However experimental analysis was not always a feasible
type of heat exchanger with helical baffles on the shell side is proposed approach as it was rigid, time consuming and costly method. Therefore,
[8,9]. Fig. 1(a–e) shows the various types of helical baffles used by there was a need for different approach which led to numerical analysis
different researchers. Helical baffles provides enhancement in heat using CFD software. Since then many researchers tried to analyze the
transfer as compared to the conventional heat exchangers. Due to the heat and fluid flow in heat exchangers consisting different baffle con-
significant advantages of the helical baffles over segmental baffles figuration and arrangements.
number of experimental and numerical studies are carried out using Peng et al. [13] using Doppler anemometry determined the flow


Corresponding author.
E-mail address: sunil.shinde@vit.edu (S. Shinde).

https://doi.org/10.1016/j.tsep.2017.11.006
Received 5 September 2017; Received in revised form 16 November 2017; Accepted 18 November 2017
2451-9049/ © 2017 Published by Elsevier Ltd.
S. Shinde, U. Chavan Thermal Science and Engineering Progress 5 (2018) 158–171

Nomenclature V shell side velocity, m/s

A heat transfer area, m2 Greek Symbols


Amin minimum transverse area, m2
B baffle Pitch of 1 cycle β baffle inclination/helix angle, °
Cp specific heat, J/kg K ΔP pressure drop, Pa
D tube diameter, mm K thermal conductivity, W/mK
dit inner diameter of tube, mm µ dynamic viscosity, Pa-s
dot outer diameter of tube, mm ρ density, kg/m3
Dis inner diameter of shell, mm
de characteristics dimension, mm Subscripts
f friction factor
Fs, Ft fouling resistance [m2K/W] In inlet
h heat transfer coefficient, W/m2K Max maximum
ji Colburn j-factor for an ideal tube bank Min minimum
Nu Nusselt number Out outlet
Pt tube pitch, mm
Res shell side Reynolds number Abbreviations
Thi temperature of hot inlet fluid (°C)
Tho temperature of hot outlet fluid (°C) HTC heat transfer coefficient
Tci temperature of cold inlet fluid (°C) FCHB fiber continuous helical baffle
Tco temperature of cold outlet fluid (°C) SB segmental baffle
U overall heat transfer coefficient [W/m2 -K] STHX shell and tube heat exchanger

field in helical baffle heat exchanger and concluded that heat transfer to be easy and cost effective.
enhancement depends upon Reynolds number as well as helix angle. The detail process of manufacturing the FRP Helical Baffle is as
Zhang et al. [7] experimentally studied effect of petal shaped helical follows:
baffles on pressure drop and heat transfer. Material used: 20% Glass filled FRP with 5 mm thickness. (Fiber
Numerical analysis on the continuous helical baffles in shell-and- reinforced plastic)
tube heat exchanger is carried out by J-F Yang et al. [20] they con- Density of FRP: 1800 kg/m3
cluded that CSSP-STHX provides better heat transfer performance. Operating Temp: up to 90 0C
Prithviraj and Andrews [21] computed heat transfer and fluid flow in Step – 1) Creating the CAD model and draft drawing of helical
shell-and-tube heat exchangers using the distributed resistance method. baffle.
Numerical investigation on shell-and-tube heat exchanger by Lei et al. Step – 2) Preparation of the wooden pattern from drawing.
[3] showed good agreement with experimental results. They used Step – 3) Making the mould from wooden pattern.
concept on distributed resistance with porous medium. Table 2 gives
research carried out on different baffle types numerically. • Alternate layer Glass wool and adhesive is applied on the wooden
From literature it is observed that only few investigators have car- pattern till the required thickness is achieved. Then it is kept for
ried out numerical analysis of heat and fluid flow in heat exchanger in soaking for 24 h.
shell and tube heat exchanger using full-length model. Also, no major • After complete soaking the mould is removed from the pattern.
efforts have been carried out to investigate the effect of baffle material • Before applying the glass wool, thin layer of separator solution is
on the performance of heat exchanger. applied on the surface of the pattern which does not allows the glass
The present study examines the thermo-hydraulic performance of a wool to stick to the pattern.
shell and tube heat exchanger with single helical baffles using numer- • Again the alternate layers of glass wool and adhesives are applied on
ical and experimental methods for seven different helix angles ranging the mould till the required thickness is achieved.
from 10° to 50°. The comparisons of the performance of seven heat • Then it is kept for the soaking and after complete soaking it is re-
exchangers with diverse baffle inclination angle are presented based on moved from the mould.
numerical results. Also the effect of baffle material is discussed. • Here also before applying the glass wool thin layer of separator
solution is applied on the surface of the moulds which does not al-
lows the glass wool to stick to the pattern.
1.1. Fiber reinforced plastic (FRP) baffles
• Here the adhesive used for making the pattern is of very high quality
compared to the adhesive used for making the final product.
Baffles are used to create the turbulence and hence enhance the heat
transfer. In case of the helical baffle heat exchanger, the baffles used are • Thus the final product i.e. one helical baffle (one cycle) is produced
from the mould.
of helicoids type which is three-dimensional geometry which makes its
manufacturing a challenging task. FRP is a composite material con- • Depending on length of heat exchanger and pitch of helical baffle,
several helical baffles are linked together to form a continuous he-
sisting a polymer (the resin) and a ceramic. In FRP, the strength and
lical baffles.
stiffness is obtained due to presence of glass fibers while resin matrix
provided the compressive strength, impact resistance and corrosive
With this method, fabrication and mounting of helical baffle (any
resistance. The manufacturing of the helical baffle with sheet metal
material) without the central tube is much simpler and cost effective.
(Steel or Aluminum) requires very complex die design. Whereas the
Considering all the above facts, an attempt is made to fabricate a
manufacturing of the FRP baffle is much simpler and cost-effective with
helical baffle of FRP material and consequently investigate its perfor-
the help of a wooden pattern. The process of fabricating helical baffle
mance numerically as well as experimentally.
using the wooden pattern is comparatively easier and cheaper. With
large scale implementation of helical baffle, this process would turn out

159
S. Shinde, U. Chavan Thermal Science and Engineering Progress 5 (2018) 158–171

(a) Tube and disk type baffles (b) Helical Baffles


(i) Single Helix (ii) Double Helix [9]
(i) Disk-and-Donut type baffles (ii) Tube in a rod
baffle exchanger supported by four rods[9]

(c) Continuous Helical Baffle [2]


(d) Non Continuous Helical Baffle [1]

(d)Four pieces middle-overlapped helical baffle


arrangement [4]
(e) One Cycle Continuous Helical Baffle [15]

Fig. 1. Different baffle schemes used by researchers.

2. Numerical study study, B, d, and Dis kept constant, while the helix inclination angle
ranges from 10° to 50°. The range of geometrical parameters used for
2.1. Numerical model and grid distribution present study is given in Table 3. Schematic diagram of geometrical
parameters of baffles and tube is shown in Fig. 3.
To acquire the comprehensive properties of helix angle full-length Variation of helix cycles for different baffle helix angle is shown in
model of length 1123 mm along with seven helix angles namely, 10°, Fig. 4.
19°, 21°, 25°, 30°, 38° and 50° were considered. The numerical domain The domain is meshed with help of unstructured layered polyhedral
of 25° helix angle is shown in Fig. 2. mesh, generated by the software, STAR CCM+. The meshes on the
The various geometrical parameters considered are: baffle pitch B, outer wall and side view of geometry are shown in Figs. 5 and 6.
tube diameter d, shell diameter D, and helix inclination angle β. For this The total number of nodes ranges from 6,913,913 to 2,272,077. The

160
S. Shinde, U. Chavan Thermal Science and Engineering Progress 5 (2018) 158–171

region adjacent to the tubes and baffle walls are meshed finer with a ∂ ∂ ∂ ⎛ ∂ε ⎞ ε ε2
growth rate of 1.1. A grid independence test is done with help of shell (ρε ) + (ρε ui) = ⎜αk μeff ⎟ + C1 G k −C 2ε ρ
∂t ∂xi ∂xj ⎝ ∂xj ⎠ k k (7)
domain cell count since our focal point is on the performance of shell
side only. For grid independence test, the computational domain of 25° where
helix angle has been chosen, as flow pattern can be completely devel-
μeff = μ + μt , μt = ρCμ k2/ ε, Gk = 2μt Sij Sij, η = 2Sij Sij
oped in presence of 5 helical cycles. Simulations were carried out with
five grid densities 1,256,790, 1,542,509, 2,079,880, 2,944,506 and k/ ε, Sij = 1/2((∂ui / ∂xj) + (∂uj/ ∂xj))
4,358,125. In view of convergent time, resolution, precision solution
and low computational time, the grid system of 2,079,880 was selected The empirical constants are assigned the following values:
for the final computational model. Fig. 7 shows a variation of HTC with
Cμ = 0.0845, C1ε = 1.42, C2ε = 1.68, β = 0.012, η0 = 4.38, α
shell domain cell count.
= 1, αε = 1.3
2.2. Governing equations and boundary conditions The heat transfer coefficient was calculated by using the coupled
method which is given as
Commercial CFD code STAR CCM+ was used for numerical analysis
of computational domains of seven helical baffle heat exchangers. The ∂t ⎞
tw| b= tf |b, −k ⎛ = ho(tw−tf)
governing equations (1)(7) used for numerical analysis are as follows: ⎝ ∂eta ⎠ (8)
Continuity Equations For the present numerical study following assumptions are made:
∂ρ ∂ (ρ u) ∂ (ρ v) ∂ (ρ w)
+ + + =0 1. All the simulations are carried by considering steady-state.
∂t ∂x ∂y ∂z (1)
2. The thermo-physical properties of working fluid are constant.
Momentum Equations: 3. Heat loss to surrounding is negligible.
∂Tyx 4. Leakages between baffles and tubes are negligible.
∂ (ρ u) ∂ρ ∂ ∂
x−momentum + div(ρ uU) = − + Txx + + Tzx + Fx
∂t ∂x ∂x ∂y ∂z
For the numerical analysis full length model of heat exchanger was
(2) considered with convection boundary condition on both sides of heat
∂ (ρ v) ∂ρ ∂Txy ∂Tyy ∂Tzy exchanger. All the solid surfaces were given no slip boundary condition,
y−momentum + div(ρ vU) = − + + + + Fy while adiabatic condition was applied on shell side wall. Ambient water
∂t ∂y ∂x ∂y ∂z
and hot water are considered as working fluid in shell side and tube side
(3)
respectively. Coupled wall function with thin wall model was used for
∂ (ρ w) ∂ρ ∂ ∂Tyz ∂ the heat conduction through baffles. SIMPLE method was used for
z−momentum + div(ρ wU) = − + Txz + + Tzz + Fz
∂t ∂z ∂x ∂y ∂z pressure–velocity coupling with second order upwind scheme for dis-
cretization of convective terms. The convergence criterion of 10−5 and
(4)
10−8 is kept for residuals of flow and energy equations respectively.
Energy Equation: The thermophysical properties of the working fluid and FRP material
∂ (ρ w) λ are listed in Table 4. All the boundary conditions used for analysis are
+ div(ρUT) = div ⎛ grad T⎞ + S T
⎜ ⎟
listed in Table 5.
∂t ⎝ CP ⎠ (5)
Turbulent kinetic energy is given by: 3. Computational results

∂ ∂ ∂ ⎛ ∂k ⎞ 3.1. Model validation


(ρk) + (ρkui) = ⎜αk μeff ⎟ + G k + ρε
∂t ∂xi ∂xj ⎝ ∂xj ⎠ (6)
Since the analysis is carried out for turbulent flow it is essential to
Turbulent dissipation energy is given by:
select the appropriate turbulence model. Four different turbulence
models are used to check results for the smooth tube. The models in-
Table 1
Studies on heat exchangers by varying the baffle angles. clude Realizable k-ε, Standard k-ε and SST k-ω. The results obtained by
these models were compared with experimental results of B. Peng et al.
Author Baffle style Baffle angle Di/do [25]. Simulations were carried out for three flow rates to determine the
(degree) heat transfer coefficient. It can be observed from Fig. 8 that the k-ε
Andrews and Non Continuous 10, 25, 40 0.787/ Standard and k-ω SST turbulence models under predicts the experi-
Master [12] 0.0254 mental results while k-ε realizable turbulence model showed good
B. Peng et al. [13] Continuous 40 0.027/0.018 agreement with the experimental results and the average percent de-
Lei et al. [3] Continuous 15, 20, 30, 40, 0.11/0.019 viation between the two results was found to be in the range of 8%.
45, 50
Huang et al. [14] Continuous NA NA
Hence all the further simulations were carried out using k-ε Realizable
Lei et al. [4] Non Continuous 20 0.313/0.019 turbulence model. Also k-ε realizable turbulence model gives superior
Wang et al. [15] Continuous 25 0.25, 0.125/ performance for the flows involving rotation, strong pressure gradients
0.019 and recirculation’s. In short realizable k-ε model has better tendency to
Zhang et al. [5] Non Continuous 40 0.211/0.019
capture the complex flow structures which is present in heat exchanger.
Zhang et al. [6] Non Continuous, 30, 40, 50, 40 0.211/0.019
Continuous
Chen et al. [16] Continuous 7.25 0.200, 0.1/ 3.2. Effect on heat transfer
0.012
Shui et al. [17] Continuous DSP-36, SSP- 0.207/0.019 Fig. 9 illustrate the variation of HTC against different helix angles
20
for different mass flow rates. Following points are noted from the graph:
Chen et al. [18] Non Continuous, Combined 8 0.207/0.01
Continuous, Continuous
Taher et al. [19] Non Continuous 40 0.87/0.0254 1. The HTC as expected goes on increasing as the mass flow rate

161
S. Shinde, U. Chavan Thermal Science and Engineering Progress 5 (2018) 158–171

Table 2
CFD studies on different baffle configurations.

Researchers Baffle type Fluid Turbulence model Schematic

Z. Duan et al. [22] Non-continuous baffles with continuous connection and middle overlapped Oil Realizable k-ε
between two sections

C. Dong et al. [23] Trisection, quadrant, quadrant end-to-end and continuous helical baffles Water RNG k- ε

M. Bahiraei [24] Elliptical quarters with inclined arrangement Nano-fluid RNG k- ε

J.F. Yang et al. [20] Continuous and Discontinuous baffles with 40° helix angle. Oil RNG k- ε

Y.P. Chen et al. [25] Circumferential overlap trisection helical baffle Water RNG k-Є

it is about 31% for 21° helix angle.


It can be observed in Fig. 10 that due to the temperature is almost
even due to the presence of helical baffles, apart from for very minor
regions, due to the small dead regions formed behind baffles and tubes.
These dead zones are formed due to the minimum flow in the certain
areas of the heat exchanger. The minimum flow produces the poor heat
transfer which leads to the fouling and damage to the tubes of heat
exchanger. Centrifugal flow which is induced due to the helical flow
pushes the fluid on the edge of the inner side of shell wall. Further due
to the radial pressure difference the centripetal force acts on the fluid
Fig. 2. Schematic of computational domain. which makes it to flow back towards the center of baffle wall where
smaller fluid velocity weakens the centrifugal force. This alternate
convergence and divergence induces the turbulence which enhances
Table 3
the heat transfer.
Range of geometrical parameters.

Sr. No. Parameter Values


3.3. Effect on pressure drop
1 Helix Angle (β) 10°, 19°, 21°, 25°, 30°, 38°, 50°
2 Shell Diameter (Dis) (mm) 153
3 Length (mm) 1123 Fig. 11 shows the variation of pressure drop against different helix
4 Baffle Thickness (mm) 3 angles for different mass flow rates. Pressure drop is gradual for higher
5 Helix Pitch (mm) 85, 160, 187, 225, 281, 374, 562 helix angles i.e. 25°–50°. However, pressure drop tends to increase ra-
6 Helix Cycles 13, 7, 6, 5, 4, 3, 2 pidly for smaller helix angles i.e. 10°–21°. This is because when helix
7 Number of Tubes 24
8 do/di 1.33
angle is small, the helical cycles are more as well as baffle pitch. Due to
9 Tube Pitch (mm) 22.5 smaller baffle pitch, cross flow velocity dominates the spiral flow of
shell side fluid, which results in rapid increase in pressure drop.
Baffle pitch also affects the pressure drop significantly. For a helix
increase for every decrement in helix angle. angle of 21°, there is decrease in pressure drop from 2423 Pa to ap-
2. At lower helix angle, HTC is quite higher than at the higher helix proximately 2315 Pa, with the increase in baffle pitch from 187 mm to
angle because of the fact that as we go on increasing the helix angle 225 mm while decrease in pressure drop is only 2206–2177 Pa for helix
the residence time of fluid goes on decreasing and the flow contact angle of 50° when baffle pitch is increased from 374 mm to 562 mm.
thereby reduces with leads to decrease in HTC with helix angle. Pressure drop is observed to increase by 40% for mass flow rate of
3. Also after helix angle of 20°, the decrease in HTC with helix angle is 0.5 kg/s when baffle helix angle increases from 19° to 25°, while there is
quite marginal due to the fact that the effect of helix angle on HTC only 10% increase in pressure drop when baffle inclination angle in-
goes on decreasing. This phenomenon is due to cross flow behavior creases from 38° to 50°. Also it is noted that effect of baffle inclination
of the fluid which experiences intense mixing due to lower helix on pressure drop is less for low Reynolds number as compared to higher
angle. Reynolds number.
Pressure distribution is demonstrated in Fig. 12 for a flow rate of
It can be observed that the values of HTC are greater for lower helix 1.25 kg/s for different baffle inclination angles ranging from 10° to 50°.
angle from 25° to 10° for pitches of 225–86 mm. Effect of baffle pitch on It is seen that pressure drop tends to increase with increase decrease in
heat transfer coefficient for lower helix angle of 10° is nearly 43% while helix angle.

162
S. Shinde, U. Chavan Thermal Science and Engineering Progress 5 (2018) 158–171

Fig. 3. Geometrical parameters of baffle and tube.

(a)

(b)

3.4. Fluid flow characteristics the cross flow velocity is dominant and as it strikes the baffle surface,
this velocity is converted into radial, axial and angular components, due
The shell-side velocity vector distribution for seven helix angle is to baffle geometry. At the entry of outlet nozzle, the velocity suddenly
shown in Fig. 13. It is observed that the fluid flows along the spiral increases due to sudden contraction.
passage due to the presence of helical baffles. Also there are almost no Higher velocity gradients are observed near the tube due to free
stagnation zones in the shell-side. At the inlet nozzle entry, the velocity vortices created by the tangential component of velocity and helical
is high due to pumping action and the velocity suddenly decreases at flow along the shell radius. Fig. 14 shows the velocity streamlines for
the inlet nozzle exit due to sudden expansion. At the inlet nozzle exit different helix angles. The flow pattern inside the shell of the heat

Fig. 4. Different configurations of helical cycles.

163
S. Shinde, U. Chavan Thermal Science and Engineering Progress 5 (2018) 158–171

ShellIN

TubeIN

TubeOUT

ShellOUT

Fig. 5. Outer boundary grid.

Table 5
Boundary conditions.

Boundary Boundary type Values


Tube
Baffle Inlet shell Mass flow inlet 0.50–2.00 kg/s & 25 °C
Outlet shell Pressure outlet 0 Pa (Gauge pr.)
Inlet tube Mass flow inlet 0.5 kg/s and 55 °C
Outlet tube Pressure outlet 0 Pa (Gauge pr.)
Shell wall Wall, Stationary, no slip, adiabatic Adiabatic
FRP Baffle Wall Conducting

converted into the radial component, which leads to swirling action of


the fluid.
Fig. 6. Side view grid.

50
3.5. Performance evaluation factor

45 The overall performance thermal and hydraulic performance of heat


42
HTC W/m2K

exchanger is evaluated by considering the ratio of heat transfer coef-


41
43 ficient to pressure drop (HTC/ΔP).
40 40
Fig. 15 shows variation of performance evaluation factor against
38 different helix angles for different flow rates. The heat transfer coeffi-
35 cient for the same pressure drop increases with the decrease in baffle
inclination angle i.e. when β > 21°, while decreases with the increase
in baffle inclination angle i.e. for β < 21°. This is due to the fact that,
30
with the increase of baffle inclination angle, the pressure drop dom-
inates the heat transfer coefficient thus reducing the overall perfor-
25 mance ratio. However, further increase in baffle inclination angle the
0 0.5 1 1.5 2 2.5 3 3.5 4 4.5 5 pressure drop tends to decrease and also the heat transfer coefficient
Shell Domain Cell Count (Million) even decreases for large helix angle. The maximum value of perfor-
mance factor is observed for helix angle of 25° and lies in the range of
Fig. 7. Variation in HTC with Shell domain cell count.
4.05–0.58 and thus is used for the experimental study.
It can be also observed that at higher Reynolds number the per-
Table 4 formance evaluation factor decreases due to the fact that, as the mass
Thermo-physical properties.
flow rate goes on increasing and with increase in helix angle, the per-
Property Hot Water (Hot Service Water FRP SS formance evaluation factor goes on decreasing as the residence time
fluid) (Cold fluid) (contact time) goes on decreasing, due to which heat transfer coeffi-
cient is less as compared lower mass flow rates. At higher helix angles,
Density, kg/m3 985 997 1660 7870
the flow pattern tends to be very smooth due to which the pressure drop
Viscosity, Pa·s 0.000504 0.00089 – –
Specific Heat, J/kg·K 4181 4182 1880 510
is minimal. Due to impact of inclination of helix angles, even though at
Thermal Conductivity, W/ 0.6493 0.62027 0.51 16.2 higher flow rates, performance evaluation factor is small due to lower
mK values of heat transfer coefficient and pressure drop at higher helix
angle region.

exchangers with a continuous helical baffle is observed to be rotational


and swirling. Because of helix angle, the cross flow component is

164
S. Shinde, U. Chavan Thermal Science and Engineering Progress 5 (2018) 158–171

3600 Fig. 8. Validation of turbulence and numerical model.

k-İ Realizable
3200
k-İ Standard
2800 k-Ȧ SST
HTC (W/m2K)

2400 B.Peng et.al

2000

1600

1200

800

400

0
1000 1200 1400 1600 1800 2000 2200 2400 2600
Re

8000 18000
0.50 kg/s
0.5 kg/s 16000
7000 0.75 kg/s
0.75 kg/s
14000 1.00 kg/s
6000 1.00 kg/s
1.25 kg/s
1.25 kg/s
¨P (Pa)

12000
HTC (W/m2 .K )

1.50 kg/s
5000 1.50 kg/s
10000 2.00 kg/s
2.00 kg/s
4000
8000

3000 6000

2000 4000

2000
1000
0
0 5.000 15.000 25.000 35.000 45.000 55.000 65.000
5.000 15.000 25.000 35.000 45.000 55.000
Helix Angle (deg)
Helix Angle (deg)
Fig. 11. Variation of shell side pressure drop against helix angles at different flow rates.
Fig. 9. Variation of shell side HTC against helix angles at different flow rates.

Fig. 10. Temperature distribution for different helix angles at flow rate of 1.25 kg/s.

165
S. Shinde, U. Chavan Thermal Science and Engineering Progress 5 (2018) 158–171

Fig. 12. Pressure distribution for different helix angles at flow rate of 1.25 kg/s.

4. Effect of baffle material 5. Experimental study

A review of the existing literature suggests that no major work is In the present study, the heat transfer and pressure drop of heat
carried out on the baffle material. The most commonly baffle material exchangers with single helical baffle with 25° helix angle and
used in the industry is stainless steel. In present work, FRP baffle ma- Segmental heat exchanger is studied experimentally.
terial is used to study the effect on heat transfer and pressure drop of
the shell and tube heat exchanger. Figs. 16 and 17 show the effect of
5.1. Experimental setup
stainless steel (SS) and FRP baffle material on heat transfer coefficient
and pressure drop for heat exchanger with 25° baffle helix angle.
The experimental setup used for the present study is similar to that
The heat transfer coefficient of SS material is more than FRP ma-
used by Shinde and Chavan [26] as shown in Fig. 18.
terial due to its conducting properties. It is observed that the average
Fig. 19 shows the line diagram of the experimental setup. The setup
percentage deviation between the SS and FRP material is less than 6%.
consists of two separate loops i.e. cooling loop and a heating loop.
Hence it can be stated that there is very less variation in heat transfer
Heating loop consists of heater coils to heat inlet water at preset tem-
coefficient and pressure drop when SS material is replaced by FRP
perature. The cooling loop comprises of a centrifugal pump, volumetric
material. Therefore FRP can serve as good substitute material in in-
flow meter, and a heat exchanger. The required flow rate to the shell
dustrial applications to save maintenance cost with prolong service life
side is adjusted through bypass valve 2. The tube side conditions were
and also reducing the overall weight of heat exchanger.
kept at constant inlet temperature and flow rate for all the flow rate
variation on the shell side. The segmental and helical baffle heat ex-
changers are mounted on the same structure. The inlet and outlet piping
of the 2 heat exchangers is kept common. The hydraulic valves are

Fig. 13. Velocity vectors for different helix angles at flow rate of 1.25 kg/s.

166
S. Shinde, U. Chavan Thermal Science and Engineering Progress 5 (2018) 158–171

Fig. 14. Streamlines of various helix angles at flow rate of 1.25 kg/s.

6.2 12000
5.8 0.50 kg/s
0.75 kg/s 10000
5.2 1.00 kg/s
4.8 1.25 kg/s 8000
1.50 kg/s
4.2
¨P ( Pa)

2.00 kg/s 6000


3.8
HTC/ǻP

3.2
4000 SS Baffle
2.8
2000 FRP Baffle
2.2
1.8
0
0.00 0.50 1.00 1.50 2.00 2.50
1.2
0.8
Flow Rate ( kg/s)
Fig. 17. Effect of baffle material on pressure drop for 25° baffle helix angle.
0.2
5 15 25 35 45 55
Helix Angle (deg) 100 temperature sensors each are installed on tube and shell side (Inlet
and outlet). Pressure transducers (least count-1 Pa) are installed at shell
Fig. 15. Variation of performance ratio (HTC/ΔP) against helix angles at different flow
rates.
inlet and outlet respectively.
It was also ensured that there exists energy balance before an ex-
perimental run. The pressure transducers were used to measure pres-
4000 sure of the shell side flow rate. Flow rate was varied in the range of
3500 30–120 lpm.
3000 The details of the equipment used for measuring the flow rate and
HTC (W/m2K)

pressure are given in Table 6below.


2500 Figs. 20 and 21 shows the photograph of helical baffles with 25°
2000 helix angle and segmental baffle used for the experimental study.
1500 FRP Baffle The shell side and baffle geometrical parameters of the 25° helix
angle baffle and segmental baffles used for the experimental study are
1000
SS Baffle given below in Table 7.
500
0 5.2. Equations for calculations
0.00 0.50 1.00 1.50 2.00 2.50
Flow Rate (kg/s) The Reynolds number and friction factor on the shell side are cal-
culated by using following equations:
Fig. 16. Effect of baffle material on heat transfer coefficient for 25° baffle helix angle.
ρ · uc · d o
Res =
μ (9)
provided to operate the heat exchanger at the desired flow rates. . Two
separate Electromagnetic flow meters (least count-1 lpm) are used to ΔP·do
capture accurate flow measurement on the tube and shell side. Two PT f= 1
2
·ρ ·u2·l (10)

167
S. Shinde, U. Chavan Thermal Science and Engineering Progress 5 (2018) 158–171

1 d
Amin = ·B·Dis ⎛1− o ⎞
⎜ ⎟

2 ⎝ Pt ⎠ (12)
The heat transfer coefficient and Nusselt number are defined by Eqs.
(5) and (6)
Q
h=
AΔT (13)

h·de
Nu =
k (14)
The uncertainties associated with the experimental setup are esti-
mated by the method proposed by Kline and McClintock [27]. Un-
certainty in the experimental analysis of various parameters such as the
Nusselt number and friction factor are within ± 4.6% and ± 8.2% re-
spectively.

5.3. Experimental setup validation

The experimental setup is validated using the Kern method [28]


with help of Segmental baffle heat exchanger results. The results are
compared and the average percent deviation between the two results is
about 7% as shown in Fig. 22.
The total rate of heat transfer used in the calculation is the average
of the shell-side and water-side as follows:

(a) The shell-side heat transfer rate is given as:

Fig. 18. Photograph of experimental setup. (ṁ s× Cp × ΔT)cold (15)

qm (b) The tube-side heat transfer rate is given as:


uc =
Amin (11)
(ṁ t× Cp × ΔT)hot (16)
A min is the minimum transverse area which given by

Fig. 19. Schematic diagram of experimental setup.

168
S. Shinde, U. Chavan Thermal Science and Engineering Progress 5 (2018) 158–171

Table 6
Details of measuring equipments.

Sr. No. Instrument Make/Model No Range Least Count

1. Electromagnetic Flow Meter Micro Mag Tube side-0–60 lpm Tube side-1 lpm
Shell side-0–100 lpm Shell side-1 lpm
2. PT100 Temperature sensors A-class element 0–100 °C 0.1 °C
3. Pressure Transducers Huba Control 5436 Wurenios Switzerland Type- PT 0–9999 Pa 1 Pa

(c) The logarithmic-mean temperature difference ΔT is determined by


(17) (b) Hence the shell-side pressure drop is determined by,
Δ T= F× ΔT m
f×Gs2 (Nb + 1)×Dis
ΔPs =
where F is the correction factor and ΔT m is the counter-flow 2 × ρ × De × φs (25)
logarithmic-mean temperature difference which is given by:

⎧ (Thi−Tco)−(Tho−Tci) ⎫
ΔTm = 6. Results and discussions


ln ( Thi − Tco
Tho − Tci ) ⎬
⎭ (18)
The experiments were conducted for two heat exchangers i.e. helical
baffle heat exchanger with 25° helix angle and segmental baffle heat
(d) The overall heat transfer coefficient U can be determined from: exchanger and their results were compared with computational results
Q= U× A× ΔTm (19) of the helical baffle heat exchanger with 25° helix angle.

6.1. Effect on heat transfer coefficient


(e) Correction factor is taken as one, as heat exchanger is new one.
(f) Now the baffle spacing is to be calculated as, The variation of heat transfer coefficient with the mass flow rate in
B= Π × Dis × tanβ (20) the shell side of the heat exchanger is shown in Fig. 23 for the 3 cases. It
is observed that heat transfer coefficient increases with increase in mass
flow rate in all cases. The increase is gradual in the helical heat ex-
(g) Then the cross-flow area at the shell diameter is calculated as, changer, while there is a steep rise in pressure drop for Segmental Heat
A s = (Dis−Nt × dot) × B (21) exchanger. The numerical and experimental results of helical baffle
heat exchangers indicate the similar trend and the deviation are less
than 10%, which indicates good agreement between the results. In
(h) The equivalent diameter is determined from, comparison with experimental results of the helical heat exchanger,
Πd 2 heat transfer coefficient of Segmental heat exchanger rises in the range
⎡ P 2T− 4ot ⎤ of 31–65%, due to zigzag shaped fluid flow creating turbulence and in
De = 4 ⎢ ⎥
⎢ Πd ot ⎥ helical, due to the absence of dead zones, as the flow is rotational.
⎣ ⎦ (22)

6.2. Effect on pressure drop


(i) Thus the shell-side heat transfer coefficient is calculated with using
base of above relations & “ji” Colburn’s factor as, From Fig. 24, it can be seen that pressure drop is very gradual in the
2/3 0.14 flow rate range of 0.5–1.25 kg/s. As the flow rate increases from
hs = ji × Cp × ⎛
ṁ s ⎞ ⎛ Ks ⎞
×⎜ ⎟
⎛ μs ⎟⎞
⎜ 1.25 kg/s, the pressures drop increases tremendously.
⎝ As ⎠ ⎝ μs ×Cps ⎠ ⎝ μw ⎠ (23) The deviation between the numerical and experimental results of
the helical heat exchanger are less than 7%, which indicates that there
is good agreement between both the results.
The shell-side pressure drop (ΔPs ) is calculated as follows: In comparison with experimental results of the helical heat ex-
changer with, the pressure drop of Segmental heat exchanger rises in
(a) The friction factor is calculated as: the range of 19–62%.The reason for this phenomenon is due to zig-zag
flow pattern on shell side whereas, in helical baffle heat exchanger, the
f= exp(0.576−0.19 × lnRe) (24) primary flow pattern is rotational and does not change dramatically.

Fig. 20. Helical baffle with 25° helix angle.

169
S. Shinde, U. Chavan Thermal Science and Engineering Progress 5 (2018) 158–171

Fig. 21. Segmental baffle.

Table 7 7000
Shell side and baffle parameters used for experimental study.
Segemental Baffle Expt.
6000
Item Dimensions Helical Baffle Expt.
5000
Helical Baffle CFD

HTC ( W/m2K)
Shell side parameters
Do/Di (m) 0.167/0.153
4000
Tube side parameter
Dos/Dis 0.012/0.0095 3000
Effective length (m) 1.123
Number 24 2000
Layout pattern (degree) 90°
Tube pitch (m) 0.0225
1000
Baffle parameters
Helix angle (degree) 25° 0
Helix pitch (m) 0.224 0.00 0.50 1.00 1.50 2.00 2.50
Baffle thickness (m) 0.003
Number 5 Mass Flow Rate ( kg/s)

Nozzle parameters Fig. 23. Comparative results of HTC for helical and segmental baffle heat exchanger.
Inner diameter (m) 0.032

25000
8000 30000 Segemental Baffle Expt.
HTC by Kern method
7000 20000 Helical Baffle Expt.
HTC by Experiment 25000
6000 Helical Baffle CFD
ǻP by Kern Method
HTC, W/m2k

15000
ǻP by Experiment 20000
5000
ǻP,pa

¨P (Pa)

4000 15000 10000

3000
10000 5000
2000
5000
1000 0
0.00 0.50 1.00 1.50 2.00 2.50
0 0
0 0.5 1 1.5 2 2.5 Mass Flow Rate (kg/s)

Mass flow rate, kg/s Fig. 24. Comparative results of ΔP for helical and segmental baffle heat exchanger.

Fig. 22. Validation of HTC and pressure drop for segmental baffle heat exchanger.
performance ratio is observed to be more in case of helical baffle heat
exchanger than segmental baffle heat exchanger.
6.3. Effect on performance ratio (HTC/ΔP)
7. Conclusion
Fig. 25 shows the comparison of performance ratio (HTC/ΔP) by
experimental results for heat exchanger with helical baffle and seg-
In the present numerical and experimental study, the heat transfer
mental baffle with numerical data for helical baffle heat exchanger. The
and pressure drop performance of the shell and tube heat exchanger
flow in the shell side of the heat exchanger with continuous helical
with helical baffles on shell side was presented. Subsequently, an ex-
baffles is smooth, rotational and helical due to the shape of the baffle,
perimental study has been conducted on 25°FCHB-STHX and SB-STHX
which indicates better performance at flow rates higher than 1 kg/s.
to judge the performance of heat exchanger. Finally, a comparative
The numerical results of helical baffle heat exchanger are more than
study between numerical study and experimental study has been done
both cases, as the bypass leakages have not been taken into account.
to evaluate the performance. The conclusions are drawn as the fol-
For experimental results of helical baffle heat exchanger and seg-
lowing:
mental heat exchanger, it can be noted that the pressure drop is com-
paratively less for lower flow rates as compared to heat transfer,
1. From the Numerical & experimental results it is confirmed that the
whereas it is exactly opposite in the high flow rate region. The

170
S. Shinde, U. Chavan Thermal Science and Engineering Progress 5 (2018) 158–171

4.7 05/2009. The authors thank Vishwakarma Institute of Technology,


4.2 Pune, India for providing Experimental space.
Segemental Baffle Expt.
3.7
Appendix A. Supplementary data
Helical Baffle Expt.
HTC /¨P (W/m2K)

3.2
2.7 Supplementary data associated with this article can be found, in the
Helical Baffle CFD online version, at http://dx.doi.org/10.1016/j.tsep.2017.11.006.
2.2
1.7 References
1.2
[1] R.K. Shah, D.P. Sekulic, Fundamentals of Heat Exchanger Design, John Wiley & Sons,
0.7 2003.
[2] C. Zhang, G.N. Xie, B. Peng, Q.W. Wang, L.Q. Luo, Q.Y. Chen, M. Zeng, An experimental
0.2 study of shell-and-tube heat exchangers with continuous helical baffles, J. Heat Transfer
0.00 0.50 1.00 1.50 2.00 2.50 129 (10) (2007) 1425–1431.
[3] Yong-Gang Lei, Ya-Ling He, Rui Li, Ya.-Fu. Gao, Effects of baffle inclination angle on flow
Mass Flow Rate (kg/s) and heat transfer of a heat exchanger with helical baffles, Chem. Eng. Process. 47 (12)
(2008) 2336–2345.
Fig. 25. Comparative results of HTC/ΔP for Helical baffle and Segmental baffle heat [4] Yong-Gang Lei, Ya-Ling He, Pan Chu, Rui Li, Design and optimization of heat exchangers
with helical baffles, Chem. Eng. Sci. 63 (17) (2008) 4386–4395.
exchanger.
[5] Jian-Fei Zhang, Ya-Ling He, Wen-Quan Tao, 3D numerical simulation on shell-and-tube
heat exchangers with middle-overlapped helical baffles and continuous baffles–Part I:
numerical model and results of whole heat exchanger with middle-overlapped helical
performance of tubular heat exchanger can be improved by helical baffles, Int. J. Heat Mass Transfer 52 (23) (2009) 5371–5380.
baffles instead of conventional segmental baffles. [6] Jian-Fei Zhang, Ya-Ling He, Wen-Quan Tao, 3D numerical simulation on shell-and-tube
2. The fabrication cost of producing helical baffles is simple, easy and heat exchangers with middle-overlapped helical baffles and continuous baffles–Part II:
simulation results of periodic model and comparison between continuous and non-
cost effective with the help of wooden pattern. continuous helical baffles, Int. J. Heat Mass Transfer 52 (23) (2009) 5381–5389.
3. The variation of heat transfer coefficient is large for the lower helix [7] Jian-Fei Zhang, Bin Li, Wen-Jiang Huang, Yong-Gang Lei, Ya-Ling He, Wen-Quan Tao,
Experimental performance comparison of shell-side heat transfer for shell-and-tube heat
angles (β < 21°). However, the effect of baffle inclination angle on exchangers with middle-overlapped helical baffles and segmental baffles, Chem. Eng. Sci.
heat transfer coefficient is small for higher helix angles (β > 21°). 64 (8) (2009) 1643–1653.
4. Pressure drop increases with increase in baffle inclination angle. [8] Jian-Fei Zhang, Ya-Ling He, Wen-Quan Tao, A design and rating method for shell-and-
tube heat exchangers with helical baffles, J. Heat Transfer 132 (10) (2010) 051802-(1-8).
Pressure drop is less in the region of low helix (β < 21°) and it is [9] Shui Ji, Du. Wen-jing, Peng Wang, Lin Cheng, Numerical investigation on double shell-
large in the high helix angle region (β > 21°). pass shell-and-tube heat exchanger with continuous helical baffles, J. Thermodyn. 2011
(2011) 1–7.
5. The effectiveness (integrated performance) of helical baffle heat
[10] J. Lutcha, J. Nemcansky, Performance improvement of tubular heat exchangers by helical
exchanger is large in low helix angle region, and it is small in high baffles, Chem. Eng. Res. Des. 68 (3) (1990) 263–270.
helix angle region. Maximum effectiveness is observed for helix [11] D. Kral, P. Stehlik, H.J. Van Der Ploeg, B.I. Master, Helical baffles in shell-and-tube heat
exchangers, Part I: experimental verification, Heat Transfer Eng. 17 (1) (1996) 93–101.
angle of 21°. [12] M.J. Andrews, B.I. Master, Three-dimensional modeling of a Helixchanger® heat ex-
6. Variation of pressure drop and heat transfer coefficient for Stainless changer using CFD, Heat Transfer Eng. 26 (6) (2005) 22–31.
Steel material as compared to FRP material is in the range of 3–5% [13] B. Peng, Q.W. Wang, C. Zhang, G.N. Xie, L.Q. Luo, Q.Y. Chen, M. Zeng, An experimental
study of shell-and-tube heat exchangers with continuous helical baffles, J. Heat Transfer
and 10–12%.As the deviation is small, FRP material can serve as a 129 (10) (2007) 1425–1431.
good potential replacement for the existing material, which will [14] Yu-qi Huang, Yu. Xiao-li, Lu. Guo-dong, Numerical simulation and optimization design of
the EGR cooler in vehicle, J. Zhejiang Univ. Sc. A 9 (9) (2008) 1270–1276.
tremendously reduce the initial cost and operation cost of the heat [15] Q. Wang, Q. Chen, G. Chen, M. Zeng, Numerical investigation on combined multiple
exchanger. shell-pass shell-and-tube heat exchanger with continuous helical baffles, Int. J. Heat Mass
7. For prediction of thermal performance of FCHB-STHX, numerical Transfer 52 (5) (2009) 1214–1222.
[16] Gui-Dong Chen, Min Zeng, Wang Qiuwang, Experimental and numerical studies on shell-
and experimental results are compared for 25° helix angle. It was side performance of three different shell-and-tube heat exchangers with helical baffles, J.
found that numerical results were in good agreement with experi- Enhanced Heat Transfer 18 (5) (2011) 449–463.
[17] Shui Ji, Du Wen-Jing, Peng Wang, Lin Cheng, Numerical investigation on double shell-
mental results and the deviation was in the range of 6–11%.
pass shell-and-tube heat exchanger with continuous helical baffles, J. Thermodyn. 2011
8. For experimental results of FCHB-STHX and SB-STHX, pressure (2011) 1–7.
drop is comparatively low as compared to heat transfer coefficient [18] Gui-dong Chen, Min Zeng, Qiu-wang Wang, Shi-ze Qi, Numerical studies on combined
parallel multiple shell-pass shell-and-tube heat exchangers with continuous helical baf-
at lower flow rates, whereas it is exactly opposite in the high flow fles, Chem. Eng. 21 (2010) 229–234.
rate region. The effectiveness is more for SB-STHX till 1 kg/s and [19] Farhad Nemati Taher, Kazem Razmi, Sirous Zeyninejad Movassag, Reza Tasouji Azar,
thereafter effectiveness of FCHB-STHX is more than SB-STHX. Baffle space impact on the performance of helical baffle shell and tube heat exchangers,
Appl. Therm. Eng. 44 (2012) 143–149.
9. Even though the initial and installation cost of the helical baffles is [20] J.F. Yang, M. Zeng, Q.W. Wang, Effects of sealing strips on shell-side flow and heat
higher than segmental baffles, it can greatly reduce the main- transfer performance of a heat exchanger with helical baffles, Appl. Therm. Eng. 64 (1)
(2014) 117–128.
tenance and operating costs by almost 20–40%. Energy cost can
[21] M. Prithiviraj, M.J. Andrews, Three dimensional numerical simulation of shell-and-tube
amount to saving of 15–20% on the usage of continuous helical heat exchangers. Part I: foundation and fluid mechanics, Numer. Heat Transfer, Part A 33
baffle heat exchangers in the industry. (8) (1998) 799–816.
[22] Z. Duan, F. Shen, X. Cao, J. Zhang, Comprehensive effects of baffle configuration on the
10. Use of helical baffles in heat exchanger reduces shell side pressure performance of heat exchanger with helical baffles, Nucl. Eng. Des. 300 (2016) 349–357.
drop, pumping cost, size, weight, fouling etc. as compare to seg- [23] C. Dong, Y.P. Chen, J.F. Wu, Influence of baffle configurations on flow and heat transfer
mental baffle for new installations. The helix baffle type heat ex- characteristics of trisection helical baffle heat exchangers, Energy Convers. Manage. 88
(2014) 251–258.
changers can save capital cost as well as operating and maintenance [24] M. Bahiraei, M. Hangi, M. Saeedan, A novel application for energy efficiency improve-
cost and thus improves the reliability and availability of process ment using nanofluid in shell and tube heat exchanger equipped with helical baffles,
Energy 93 (2015) 2229–2240.
plant in a cost effective way. [25] Y.P. Chen, Y.J. Sheng, C. Dong, J.F. Wu, Numerical simulation on flow field in cir-
cumferential overlap trisection helical baffle heat exchanger, Appl. Therm. Eng. 50 (1)
(2013) 1035–1043.
[26] S. Shinde, U. Chavan, Performance of turbulence models on heat transfer and pressure
Acknowledgement drop with a 25° continuous helical baffled heat exchanger, Int. Rev. Mech. Eng. 11 (1)
(2017) 69–76.
This research work has been funded by Board of College and [27] S.A. Kline, F.A. McClintock, Describing uncertainties in single-sample experiments, ASME
Mech. Eng. 75 (1953) 3–8.
University Development (BCUD), under S. P. Pune University, under the [28] T. Kuppan, Heat Exchanger Design Handbook, Mercel Dekker Inc., New York, 2002, pp.
University research grant scheme. Grant No. BCUD/OSD/184 dated 11/ 69–73.

171

Вам также может понравиться