Case Studies in Thermal Engineering 12 (2018) 213–222

Contents lists available at ScienceDirect

Case Studies in Thermal Engineering

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

Heat transfer phenomena on waste heat recovery of combustion

T
stack gas with deionized water in helical coiled heat exchanger
⁎
Rithy Konga, Thoranis Deethayatb, , Attakorn Asanakhamb, Tanongkiat Kiatsiriroatb
a
Energy Engineering Program, Faculty of Engineering and Graduate School, Chiang Mai University, Thailand
b
Thermal System Research Laboratory, Department of Mechanical Engineering, Faculty of Engineering, Chiang Mai University, Thailand

A R T IC LE I N F O ABS TRA CT

Keywords: Theoretical and experimental studies on waste heat recovery of combustion stack gas and heat
Waste heat recovery transfer phenomena of a fully developed laminar flow of deionized water in vertical helical coils
Helical coil were carried out with coil dimensions: tube diameter to coil diameter ,di/D = 0.04 − 0.06 and
Heat transfer data pitch to coil diameter, p/D = 0.1 − 0.25. The calculation of heat transfer data was based on
Deionized water
countercurrent flow LMTD method. The result showed that deionized water (DI-water) possessed
better heat transfer than that of normal water. The effect of coil pitch, coil diameter, and coiled
tube diameter on heat transfer phenomena of helical coils had been discussed and a new set of
correlation of heat transfer data was created and it could be found that the results from the
correlation agreed well with the experimental data. In addition, the overall heat transfer coef-
ficient between the hot exhaust gas and the heat transfer fluid in the helical coil was also con-
sidered. Smaller tube diameter gave better overall heat transfer coefficient at low water-side
Reynolds number and when the Reynolds number was over 3500 the bigger tube diameter
showed the advantage. Smaller coil diameter seemed to get better overall heat transfer coefficient
at low water-side Reynolds number.

1. Introduction

Waste heat recovery is a technology in converting waste heat from all kinds of heat sources into useful thermal energy or
electricity generation. In case of power generation, only 20–40% of fossil fuel energy input is converted into electrical power while
other 60–70% is lost in terms of waste heat [1], therefore, the concept on high energy efficiency in energy utilization is also a crucial
aspect and waste heat energy recovery is one main approach to solve the problem. Waste heat could be recovered from some thermal
devices or engines such as boiler, internal combustion engine (ICE), gas turbine, and incinerator, of which the temperature range
varies from 200 ℃ up to 1000 ℃ or greater and the low grade waste heat, below 200 °C, represents the most share of waste heat
due to its abundance [2].
There are many technologies for heat recovery mostly through heat exchanger applications. Various kinds of heat exchangers
have been found in literatures such as shell and tube heat exchanger, double pipe heat exchanger, plate heat exchanger, heat pipes
and helical coiled heat exchanger [3–6]. Helical coiled heat exchanger is one of the promising technologies in extracting heat due to
its compactness in a limit space with higher heat transfer area per unit volume [7]. In addition, helical coiled tubes provide a better
heat transfer coefficient over straight tubes since there is secondary vortex flow generated inside the coil due to the coil curvature
which reduces the thermal resistance of boundary layer [8]. Moreover, the helical coil in the study was vertical type then it was a

⁎
Corresponding author.
E-mail address: thoranis.dee@cmu.ac.th (T. Deethayat).

https://doi.org/10.1016/j.csite.2018.04.010
Received 27 November 2017; Received in revised form 29 March 2018; Accepted 9 April 2018
Available online 10 April 2018
2214-157X/ © 2018 Published by Elsevier Ltd. This is an open access article under the CC BY-NC-ND license
(http://creativecommons.org/licenses/BY-NC-ND/4.0/).
R. Kong et al. Case Studies in Thermal Engineering 12 (2018) 213–222

Nomenclature c heat capacity ratio

ṁ mass flow rate
LMTD log-mean temperature difference Q̇ heat transfer rate
HTF heat transfer fluid q heat flux
d coiled tube diameter h convective heat transfer coefficient
p coil pitch U overall heat transfer coefficient
D coil diameter k thermal conductivity
Re Reynolds number
Pr Prandtl number Subscripts
Dn Dean number, Dn = Re(d/D)1/2
Nu Nusselt number i inside or inlet
R thermal resistance o outside or outlet
T temperature f fluid
Cp specific heat s tube wall surface
C heat capacity

kind of cross flow heat exchanger. Compare to conventional cross flow heat exchanger which was a set of straight tube bank, with the
same surface area, the pressure drop on both sides (inner and outer) should be nearly the same but the inside heat transfer coefficient
of the helical coil should be higher than the normal one which was straight tube.
Janssen et al. [9] had pointed an experimental and numerical study on the fully developed laminar convection heat transfer in
helical coiled tubes under the conditions of constant wall heat flux in a range of 103 − 4.103W/m2 and the working fluid was water-
glycerol mixtures. The experiments were carried out for the specific ratios of tube diameter to coil diameter, 1/100 ≤di /D≤ 1/10;
Reynolds numbers, 20 ≤ Re ≤ 4000; Prandtl numbers, 10 ≤ Pr ≤ 500 and a set of correlations consisted of these
parameters with Nusselt number, Nu = hdi /k, and Dean number, Dn = Re(di /D)1/2 was given as:

Nu = 1.7(Dn2Pr)1/6, Dn< 20 (1)

Nu = 0.9(Re2 Pr)1/6 , 20 < Dn < 1×102 (2)

Nu = 0.7Re0.43 Pr1/6 (di /D)0.07, 1×102< Dn < 8.3×102. (3)

Schmidt by Hardik et al. [10] and Pimenta et al. [11] set experimental study on convective heat transfer and pressure drop inside
the spiral tubes having 1948-5761 mm total length. The experiments were done with the boundary condition of constant wall
temperature and the results were based on log-mean temperature difference LMTD-method of bulk fluid temperatures. The different
geometric parameters of coil-tube diameter ratios, 4.91< D/di < 81.1, were used and three types of fluids; oil, water, and air were
selected as the experimental working fluids. The correlation of average Nusselt number with Reynolds number, Prandtl number and
coil characteristics was presented as:

Nu = 3.65 + 0.08[1 + 0.8(di /D)0.9]Pr1/3Ren . (4)

Where n = 0.5 + 0.2903(di /D)0.194,

100 < Re<Recr . Recr is the critical Reynolds number : Recr = 2300(1 + 8.6(di /D)0.45) .
Xin et al. [12]studied the effects of Prandtl number and different geometric parameters on local and average convective heat
transfer coefficients in helical coils with the number of coil turns lower than 10 turns under the condition of constant heat flux. The
selected fluids were air, water, and ethylene glycol. From the experimental results by LMTD-method, a set of average Nusselt number
correlations using inner tube diameter as the characteristic length, Nu = h. di /k for laminar was defined as:

Nu = (2.153 + 0.318Dn0.643)Pr 0.177. (5)

Where 20 < Dn < 2000, 0.7 < Pr < 175, 0.0267<di /D < 0.0884 .
There are some literatures studying on heat transfer behavior of helical coils such as local heat transfer coefficient in helical coil
[10], forced convective heat transfer in curved pipes [13], mixed convection heat transfer in vertical helical coil [14], experimental
and numerical study of laminar flow in helical coils [15]. All the experimental works in the literatures mostly were conducted with
electrical heaters as heat source for constant heat flux boundary condition.
The transport properties and heat transfer performance of DI-water could be found in experimental works of Godson et al. and
Sohel et al. [16,17] and Sarafraz et al. [18]. It could be noted that the results of thermal conductivity and viscosity of the deionized
water agreed well with the transport properties of pure water calculated in REFPROP, an NIST database. In addition, the heat transfer
coefficient of the deionized water was higher than that of the normal water containing the contaminants and solid particles.
In this present study, hot combustion stack gas was used as heat source and the heat transfer rate was obtained from the heat
extracted by the working fluid in a set of helical coils. The experimental working fluid inside the coils was deionized water (DI water)
in fully developed laminar flow condition. The inside coiled tube convective heat transfer coefficients of DI water and the overall heat
transfer coefficients were determined under the countercurrent flow LMTD-method from the temperature distributions of the hot gas,
the oil wall surface and the DI water. In addition, a new heat transfer correlation was developed for the DI water side.

214
R. Kong et al. Case Studies in Thermal Engineering 12 (2018) 213–222

2. Experimental set-up

A set of copper helical coils with different tube pitches, tube lengths, tube diameters, and coil diameters as shown in Fig. 1 and
Table 1 was mounted inside a stack of 0.3 m in diameter having a flow of LPG combustion gas. Deionized water was used as the
working fluid in the coils to extract heat from the combustion gas and it was circulated by a centrifugal pump between a temperature
controlled storage tank and the tested helical coils. The outlet hot water from the coils was cooled down in a water cooling radiator
before flowing back to the storage tank to maintain the water temperature in the tank at a set value. The water flow rate in the water
loop was regulated by a needle valve and a by-pass flow control. The schematic sketch of the test rig was shown in Fig. 2.
Four helical coils, listed in Table 1, were tested. A set of nine K-type thermocouples was calibrated with a standard mercury
thermometer and connected at three cross sectional levels of the vertical coil, bottom, middle and top. Each sectional level as shown
in Fig. 1 had three thermocouples for measuring temperatures of hot gas in the stack (To) , temperature on the coil wall surface (Ts)
and temperature of water inside the coil (T) i . The coil wall thickness was so thin that the thermal resistance of the coil wall was
neglected (R w = 0) and the wall temperature distributions, inner and outer wall surface, were assumed to be the same (Tis = Tos) .
The coil wall outer surface was considered as overall heat transfer surface.
The flow velocity of hot gas was measured by a hot wire anemometer AM-4234SD at the air inlet before combustion. The
properties of fluids, both water and hot gas, were defined at fluid bulk temperatures. A flow control valve and a water flow meter
were installed to regulate and read the water flow rate. The water flow rate was controlled in a range of 0.25 − 2L/min which was
in laminar regime. The coil inlet water temperature was maintained in a range of 40–50 ℃ using electrical heater in the storage tank
and the inlet temperature of the hot stack gas was maintained in the range of 170 − 230℃ in experimental tests by regulating the flow
rates of air inlet and LPG gas using air flow regulator and LPG needle valve, respectively. In addition, the hot gas inlet temperature
fluctuated in a range of ± 5℃ due to LPG gas burning consumption and also air inlet fluctuation. The average velocity of hot gas in the
stack was 0.66 m/s.
The heat transfer rate was determined from:

Q̇ = m.̇c p. (Two − Twi). (6)

Q̇ = hi . Ai.∆TLm,i. (7)

Q̇ = Us. A o.∆TLm. (8)

Where hi and Us are the inside convective and the overall heat transfer coefficient, respectively. ∆TLm,i and ∆TLm represent the inside
log-mean temperature difference between wall surface and water and the overall log-mean temperature difference between hot gas
and water, respectively in countercurrent flow. The accuracies of the experimental measuring instruments are described in Table 2.
The uncertainties of heat transfer rate δQ̇uncert and overall convective heat transfer coefficient Us,uncert could be found from
uncertainty standard equation as:

2 2 2
̇ ̇ ̇
δQ̇uncert = ⎛⎜ ∂Q .δṁ ⎟⎞ + ⎜⎛ ∂Q .δCp⎟⎞ + ⎜⎛ ∂Q .δ∆T⎟⎞ .
⎝ ∂ṁ ⎠ ⎝ ∂Cp ⎠ ⎝ ∂∆T ⎠ (9)

2 2 2
̇ ̇
δUuncert = ⎛ ∂U .δQ̇⎞ + ⎜⎛ ∂Q .δA s⎟⎞ + ⎜⎛ ∂Q .δ∆TLm⎟⎞ .
⎜ ⎟
̇
⎝ ∂Q ⎠ ⎝ ∂A s ⎠ ⎝ ∂∆TLm ⎠ (10)

The results of the uncertainties were calculated and listed in Table 3. The maximum uncertainty was around 10%.

Fig. 1. Helical coil and heat transfer pattern.

215
R. Kong et al. Case Studies in Thermal Engineering 12 (2018) 213–222

Table 1
Description of experimental sections.
Items Characteristics

Coil di do D p/di N L H
(m) (m) (m) - - (m) (m)

1 0.014675 0.015875 0.245 2 14.5 11.67 0.65

2 0.0115 0.0127 0.245 2.5 14 10.5 0.63
3 0.0115 0.0127 0.245 4 14.5 11.02 0.82
4 0.0115 0.0127 0.19 4 14.5 9.38 0.82

Fig. 2. Experimental setup.

Table 2
Accuracies of the experimental measuring instruments.
Tools Characteristics Accuracy

Thermocouple type K Range: − 200–350 °C ± 0.7 °C

AM-4234SD Hot wire anemometer-SD Card Measuring air flow velocity ± (5%+a)readinga= 0.1 m/s
Range: 0.2–35 m/s
Flow meter Measure water flow rateRange: 0–2 l/mn ± 5%
Water pump QPM60 pompa, SUMOTO ± 2%
Measuring water flow rate
Range: 5–40 l/min
Air blower Model: SB− 30Blowing air into system to generate hot gas ± 5%
Range: 1–7 m3/min

Table 3
Uncertainty of important parameters, heat transfer rate and overall heat transfer coefficient.

δA s δρ δCp δV̇ δṁ δT δ∆T δ∆TLm δQ̇ δU

(m2) (kg/m3) (J/kg. k) (m3/s) (kg/s) (℃) (℃) (℃) (W) (W/ m2 K)

.017 5 1 1.67e-7 1.84e-4 0.7 0.99 1.06 88.17 1.59

(3.8%) (0.51%) (0.02%) (1%) (0.99%) (1.56%) (5.8%) (0.83%) (10.8%) (9.96%)

216
R. Kong et al. Case Studies in Thermal Engineering 12 (2018) 213–222

3. Results and discussions

3.1. Verification on the experimental method

Fig. 3 shows experimental study on heat transfer phenomena of coil 1. The hot gas inlet temperatures were varied from 175 to
230 ℃. The volume flow rate of water varied in a range of 0.25 − 2l/min corresponding to the Reynolds number range of
500 − 7000 . It could be noted from Fig. 3a that the heat rate increased for either the increase of the hot gas temperature or increase of
the water flow rate.
The heat rate increased as the water flow rate increased since higher water-side heat transfer coefficient could be obtained with
higher mass flow rate thus the overall heat transfer coefficient could also be increased. Anyhow, when the water mass flow rate was
too high (over around 1.25 l/min), the heat transfer control was in the gas side therefore the overall heat transfer coefficient in-
creased only slightly with the increase of inside Reynolds number (2000 < Rei < 5000) since the mass flow rate of the hot gas was
rather constant due to that the hot gas velocity and inlet temperature were maintained then the heat transfer rate was slightly
improved. The result of the overall heat transfer coefficient with the water mass flow rate was given in Fig. 3b.
The experimental approach was verified by comparing the experimental values of normal water heat transfer coefficient to the
values from the correlation, Eq. (3) developed by Janssen et al. [9]. It could be seen that both results agreed very well as shown in
Fig. 4 which meant that the testing approach was good enough to study the heat transfer behavior of the helical coils.
In this study, the heat transfer behavior of deionized water (DI-water) was concentrated and it could be found that the viscosity of
DI-water was 0.89 mPa.s compared with 0.92 mPa.s of normal water at fluid temperature of 25 °C (measured by DV-III Ultra
Programmable Rheometer). The increase in viscosity increased the thermal boundary layer of the fluid on the inner wall surface
which resulted in lower heat transfer coefficient. For coil 1, with inlet hot gas temperature around 200 °C, at water Reynolds number
around 1000, the water-side heat transfer coefficient for DI-water was around 750 compared with 510 for normal water and at water
Reynolds number of around 5000, the heat transfer coefficients were around 1950 and 1200, respectively.

3.2. Effect of coiled tube dimensions on water-side heat transfer coefficient for DI-water

Two helical coils (coil1 and coil 2), from Table 1, were tested to find out the effect of tube diameter on water-side heat transfer
coefficient (hi). Coil 1 had bigger tube diameter and both coils had similar coil diameter, close coil pitch and number of coil turns. The

Tgi = 17
75 Tgi = 20
00 Tgi = 230
1.2

1
Heat rate (kw)

0.8

0.6

0.4

0.2

0
0.2
25 0.5 0.75 1 1.25
1 1.5 1.75 2
Waterr flow rate inside coil (L/min)

a. Heat transfer
t rate eextracted from hot gas by water
w (coil 1).

25 Tgi=175 Tg
gi=200 Tgi=230
Overall heat transfer coefficient (Us)

20

15

10

5
0 1000 2000
2 3000 4000 5000 60000
Reynolds num
mber (Rei)

b. Av
verage overall heat transfer coefficients with ynolds number (coil 1).
w water Rey

Fig. 3. Heat transfer behaviors at various water flow rates and hot gas temperatures (coil 1). The gas velocity was 0.66 m/s.

217
R. Kong et al. Case Studies in Thermal Engineering 12 (2018) 213–222

Average inside convective heat transfer coefficient (hi)

this study: Tgi= 175 C this study: Tgi = 200 C
this study: Tgi = 230 Janssen et al.

Covective heat transfer coefficients (W/m2.K)

2000

1500

1000

500

0
0 1000 2000 3000 4000 5000 6000
Reynolds number (Rei)

Fig. 4. Convective heat transfer coefficients on water side for coil 1 with various gas inlet temperatures (175, 200, and 230 °C) and gas velocity of
0.66 m/s.

experimental results showed that the helical coil with smaller tube diameter (coil 2) got better water-side heat transfer coefficient at
the same water-side Reynolds number due to a higher fluid velocity which reduced the thermal boundary layer. The results were
given in Fig. 5a. The experimental heat transfer coefficients of coil 2 agreed very well with those calculated from correlations
presented by Schmidt et al. [10,11], Eq. (4), and slightly deviated from Xin et al. [12], Eq. (5), but those of coil 1 differed from the
correlations significantly due to the coil scale was not in the range defined in the correlations.
Effect of coil pitch on heat transfer coefficient could be seen from coil 2 and coil 3 in Table 1. The results in Fig. 5b revealed that
the coil pitch did not affect the heat transfer coefficient. Again, the heat transfer coefficient data agreed very well with those
evaluated by the correlation of Schmidt et al. [10,11] and slightly differed from those evaluated by correlation of Xin et al. [12].
Effect of coil diameter on heat transfer coefficient was studied from coil 3 and coil 4. The results indicated that decrease of coil
diameter or increase of coil curvature (d/D) significantly increased the heat transfer coefficients inside the coiled tube, as shown in
Fig. 5c. Coil curvature caused the secondary flow in the fluid flow inside the coil and bigger coil curvature could reduce the thermal
boundary layer and higher heat transfer coefficient could be obtained. In the Figure, coil 4 which had smaller coil diameter gave
better heat transfer coefficient than that of coil 3 which had bigger coil diameter. The results evaluated by Schmidt et al. [10,11]
seemed to be slightly better than those of Xin et al. [12].

3.3. Average Nusselt number for helical coils for DI-water

The average Nusselt number of DI-water in helical coil could be determined by

hi . di
Nui = .
ki (11)

From the experimental data, the new correlation of Nusselt number for deionized water with Reynolds number, Prandtl number,
and coil could be performed as

Nu di = 3.66 + 0.464(0.102 + 1.273(di /D)0.657 + 0.012(p/D)0.023).Pr1/3.Re n . (12)

Where n = 0.5156 + 0.998(di /D)0.776 .

This new experimental correlation is applicable for fully developed laminar flow within the ranges of Reynolds number of
d
6.102≤Rei ≤ 6.103, Prandtl number of 2≤Pri ≤ 5, and coil characteristics of 0.04< Di < 0.06; 0.1 < p/D < 0.25. The results from the
new correlation were found to be fitted the experimental data than those from Schmidt et al. [10,11] and Xin et al. [12], as shown in
Fig. 6.

3.4. Average overall heat transfer coefficient (Us) of helical coils with DI-water

The overall heat transfer coefficient (Us) could be calculated from Eq. (8) as Us = Q̇/(A o.∆TLm) and the results for all tested coils
when the inlet hot gas temperature at 200 °C and the average gas velocity at 0.66 m/s were shown in Fig. 7. The gas side heat transfer
coefficient controlled the overall heat transfer coefficient of the heat exchangers therefore at high water-side Reynolds number (Rei
≥ 3500) even the water side velocity or the Rei increased, the overall heat transfer coefficient increased only slightly. At low Rei (Rei
≤ 3500), the water-side heat transfer coefficient was also low thus increase of Rei also gave better overall heat transfer coefficient.
The unit with a bigger coil tube diameter (coil 1) gave higher gas-side heat transfer coefficient compared to a smaller coil tube

218
R. Kong et al. Case Studies in Thermal Engineering 12 (2018) 213–222

this study : Coil 1 Xin et al. : Coil 1 Schmidt et al. : Coil 1

this study: Coil 2 Xin et al. : Coil 2 Schmidt et al. : Coil 2

Convective heat transfer coefficient (W/m2.K)

3000

2500

2000

1500

1000

500

0
0 2000 4000 6000 8000
Reynolds number (Rei)

a. Effect of tube diameter on water-side convective heat transfer coefficients (coil 1 and coil 2).

this study : Coil 2 Xin et al. : Coil 2 Schmidt et al. : Coil 2

this study : Coil 3 Xin et al. : Coil 3 Schmidt et al. : Coil 3
Convective heat transfer coefficient (W/m2.K)

3000

2500

2000

1500

1000

500

0
0 2000 4000 6000 8000
Reynolds number (Rei)

b. Effect of coil pitch on water-side heat transfer coefficient (coil 2, p/di=2.5 and coil 3, p/di=4).

this study: Coil 3 Xin et al. : Coil 3 Schmidt et al. : Coil 3

this study: Coil 4 Xin et al. : Coil 4 Schmidt et al. : Coil 4

3500
Convective heat transfer coefficient (W/m2.K)

3000

2500

2000

1500

1000

500

0
0 2000 4000 6000 8000
Reynolds number (Rei)

c. Effect of coil diameter on water-side heat transfer coefficient (coil 3, D=23 cm, coil 4, D = 18 cm).
Fig. 5. Effects of Coil dimensions on water-side heat transfer coefficients of DI-water. The inlet hot gas temperature was at 200 °C and the average
gas velocity was 0.66 m/s.

diameter (coil 2) having similar pitch due to its higher exposed heat transfer surface to the hot gas which reduced thermal boundary
layer thickness around the tube. This phenomenon was similar to a flow over tube bundle and the gas-side heat transfer coefficient
increased with the smaller pitch/diameter ratio [19,20]. However, when Reynolds number was less than 3500, opposite results were
found since the water heat transfer coefficient was also lower due to the lower water velocity thus the overall heat transfer coefficient
of the bigger diameter was poorer than that of the smaller one.
The effect of coil pitch on the overall heat transfer coefficient was not significant (coil 2 vs. coil 3) while the small coil diameter

219
R. Kong et al. Case Studies in Thermal Engineering 12 (2018) 213–222

Coil 1 Coil 2
this study Xin et al. this study Xin et al.
Schmidt et al. Present correlation Schmidt et al. Present correlation
60 60

50
Nusselt number (Nui)
50

Nusselt number (Nui)

40
40
30
30
20

20
10

0 10
0 2000 4000 6000 0 2000 4000 6000 8000
Reynolds number (Rei) Reynolds number (Rei)

Coil 3 Coil 4
this study Xin et al. this study Xin et al.
Schmidt et al. Present correlation Schmidt et al. Present correlation
60 60

Nusselt number (Nui) 50

50
Nusselt number (Nui)

40
40
30
30
20

20 10

10 0
0 2000 4000 6000 8000 0 2000 4000 6000 8000
Reynolds number (Rei) Reynolds number (Rei)

Fig. 6. Comparison of average inside Nusselt number evaluated by the new correlation with those by the correlations presented by Schmidt [10,11]
and Xin et al. [12]. The inlet hot gas temperature at 200 °C and average gas velocity of 0.66 m/s.

Present : Coil 1 Present : Coil 2 Present : Coil 3 Present : Coil 4

40
Overall heat transfer coefficient (W/m2.K)

30

20

10

0
0 1000 2000 3000 4000 5000 6000 7000
Reynolds number (Rei)

Fig. 7. Average overall heat transfer coefficients with water-side Reynolds number. The inlet hot gas temperature at 200 °C and average gas velocity
of 0.66 m/s.

showed significant advantage when the water Reynolds number was less than 3500 (coil 3 vs. coil 4). For the latter case, the smaller
coil diameter unit was installed closer to the center of the stack thus the gas velocity passing over the coil was higher than that of the
unit installed close to the stack wall then the overall heat transfer coefficient of coil 4 was higher than that of coil 3. This meant that
installation of helical coil around the stack axis was better than that around the stack wall surface.

220
R. Kong et al. Case Studies in Thermal Engineering 12 (2018) 213–222

4. Conclusion

In this research paper, the experiments of four copper helical coiled heat exchangers, listed in Table 1, were conducted under the
conditions of fully developed laminar flow of deionized water during the heat recovery from the hot stack gas at around 200 °C. The
results on the heat transfer characteristics of the heat recovery could be noted as:

- Deionized water HTF showed better heat transfer over normal water since the latter fluid contained more contaminants which
generated more friction and resulting in greater thermal boundary thickness.
- The helical coil with smaller tube diameter showed better water-side heat transfer coefficient due to higher fluid velocity.
- The coil pitch did not affect the heat transfer coefficient
- Decrease of coil diameter or increase of coil curvature (d/D) increased the water-side heat transfer coefficient. Coil curvature
caused the secondary flow in the fluid flow inside the coil and bigger coil curvature could reduce the thermal boundary layer and
higher heat transfer coefficient could be obtained.
- A new set of correlation from experimental results for deionized water in helical coiled heat exchangers was correlated as:
Nu di = 3. 66 + 0. 464 (0. 102 + 1. 273 (di / D) 0 . 657 + 0. 012 (p/ D) 0 . 023). Pr 1 / 3. Re n ,
Where n = 0.5156 + 0.998(di /D)0.776 .
di
6.102≤Rei ≤ 6.103, 2≤Pri≤5; 0.04< < 0.06; 0.1 < p/D < 0.25.
D

- The gas side heat transfer coefficient controlled the overall heat transfer coefficient of the heat exchangers at high water-side
Reynolds number (Rei) ≥ 3500) even the water side velocity or the Rei increased, the overall heat transfer coefficient increased
only slightly. At low Rei (Rei ≤ 3500), the water-side heat transfer coefficient was also low thus increase of Rei also gave better
overall heat transfer coefficient.
- The unit with a bigger coil tube diameter (coil 1) gave higher gas-side heat transfer coefficient compared to a smaller coil tube
diameter (coil 2) due to its higher exposed heat transfer surface to the hot gas which reduced thermal boundary layer thickness
around the tube. However, when Reynolds number was less than 3500, opposite results were found since the water heat transfer
coefficient was also lower due to the lower water velocity thus the overall heat transfer coefficient of the bigger diameter was
poorer than that of the smaller one.
- The effect of coil pitch on the overall heat transfer coefficient was not significant (coil 2 vs. coil 3) while the small coil diameter
showed significant advantage when the water Reynolds number was less than 3500 (coil 3 vs. coil 4). For the latter case, the
smaller coil diameter unit was installed closer to the center of the stack thus the gas velocity passing over the coil was higher than
that of the unit installed close to the stack wall then the overall heat transfer coefficient of the smaller coil diameter was higher
than that of the bigger coil diameter. This meant that installation of helical coil around the stack axis was better than that around
the stack wall surface.

Acknowledgements

The authors are thankful to Department of Mechanical Engineering, Faculty of Engineering and Center of Excellence for
Renewable Energy, Chiang Mai University for supporting testing facilities and financial support.

References

[1] R.F. Boehm, H. Yang, J. Yan Handbook of Clean Energy Systems. n.d.
[2] D. Ziviani, A. Beyene, M. Venturini, Advances and challenges in ORC systems modeling for low grade thermal energy recovery, Appl. Energy 121 (2014) 79–95,
http://dx.doi.org/10.1016/j.apenergy.2014.01.074.
[3] B. Farajollahi, S.G. Etemad, M. Hojjat, Heat transfer of nanofluids in a shell and tube heat exchanger, Int. J. Heat. Mass Transf. 53 (2010) 12–17, http://dx.doi.
org/10.1016/j.ijheatmasstransfer.2009.10.019.
[4] P.K. Swamee, N. Aggarwal, V. Aggarwal, Optimum design of double pipe heat exchanger, Int. J. Heat. Mass Transf. 51 (2008) 2260–2266, http://dx.doi.org/10.
1016/J.IJHEATMASSTRANSFER.2007.10.028.
[5] A.K. Tiwari, P. Ghosh, J. Sarkar, Performance comparison of the plate heat exchanger using different nanofluids, Exp. Therm. Fluid Sci. 49 (2013) 141–151,
http://dx.doi.org/10.1016/j.expthermflusci.2013.04.012.
[6] R. Sureshkumar, S.T. Mohideen, N. Nethaji, Heat transfer characteristics of nanofluids in heat pipes: a review, Renew. Sustain Energy Rev. 20 (2013) 397–410,
http://dx.doi.org/10.1016/j.rser.2012.11.044.
[7] M.E. Ali, Experimental investigation of natural convection from vertical helical coiled tubes, Int. J. Heat. Mass Transf. 37 (1994) 665–671, http://dx.doi.org/10.
1016/0017-9310(94)90138-4.
[8] M. Tayde, J. Wankhade, S. Channapattana, Heat transfer analysis of a helically coiled heat exchanger (ISSN 2278 – 0211), Int J. Innov. Res. Dev. (2014) 3.
[9] L.A.M. Janssen, C.J. Hoogendoorn, Laminar convective heat transfer in helical coiled tubes, Int J. Heat. Mass Transf. 21 (1978) 1197–1206, http://dx.doi.org/
10.1016/0017-9310(78)90138-2.
[10] B.K. Hardik, P.K. Baburajan, S.V. Prabhu, Local heat transfer coefficient in helical coils with single phase flow, Int J. Heat. Mass Transf. 89 (2015) 522–538,
http://dx.doi.org/10.1016/j.ijheatmasstransfer.2015.05.069.
[11] T.A. Pimenta, J.B.L.M. Campos, Heat transfer coefficients from Newtonian and non-Newtonian fluids flowing in laminar regime in a helical coil, Int J. Heat. Mass
Transf. 58 (2013) 676–690, http://dx.doi.org/10.1016/j.ijheatmasstransfer.2012.10.078.
[12] R.C. Xin, M.A. Ebadian, The effects of Prandtl numbers on local and average convective heat transfer characteristics in helical pipes, J. Heat. Transf. 119 (1997)
467, http://dx.doi.org/10.1115/1.2824120.

221
R. Kong et al. Case Studies in Thermal Engineering 12 (2018) 213–222

[13] Y. Mori, W. Nakayama, Study on forced convective heat transfer in curved pipes, Int J. Heat. Mass Transf. 10 (1967) 681–695, http://dx.doi.org/10.1016/0017-
9310(67)90113-5.
[14] N. Ghorbani, H. Taherian, M. Gorji, H. Mirgolbabaei, Experimental study of mixed convection heat transfer in vertical helically coiled tube heat exchangers, Exp.
Therm. Fluid Sci. 34 (2010) 900–905, http://dx.doi.org/10.1016/j.expthermflusci.2010.02.004.
[15] J. De Amicis, A. Cammi, L.P.M. Colombo, M. Colombo, M.E. Ricotti, Experimental and numerical study of the laminar flow in helically coiled pipes, Prog. Nucl.
Energy 76 (2014) 206–215, http://dx.doi.org/10.1016/j.pnucene.2014.05.019.
[16] L. Godson, B. Raja, D.M. Lal, S. Wongwises, Experimental investigation on the thermal conductivity and viscosity of silver-deionized water nanofluid, Exp. Heat.
Transf. 23 (2010) 317–332, http://dx.doi.org/10.1080/08916150903564796.
[17] S.M. Sohel Murshed, S.-H. Tan, N.-T. Nguyen, Temperature dependence of interfacial properties and viscosity of nanofluids for droplet-based microfluidics, J.
Phys. D. Appl. Phys. 41 (2008) 85502, http://dx.doi.org/10.1088/0022-3727/41/8/085502.
[18] M.M. Sarafraz, F. Hormozi, Experimental studies on the effect of water contaminants in convective boiling heat transfer, Ain Shams Eng. J. 5 (2014) 553–568,
http://dx.doi.org/10.1016/J.ASEJ.2013.11.006.
[19] W.A. Khan, J.R. Culham, M.M. Yovanovich, Convection heat transfer from tube banks in crossflow: analytical approach, Int J. Heat. Mass Transf. 49 (2006)
4831–4838, http://dx.doi.org/10.1016/j.ijheatmasstransfer.2006.05.042.
[20] A. Safwat Wilson, M. Khalil Bassiouny, Modeling of heat transfer for flow across tube banks, Chem. Eng. Process Process Intensif. 39 (2000) 1–14, http://dx.doi.
org/10.1016/S0255-2701(99)00069-0.

222