Proceedings of the 48th National Conference on Fluid Mechanics and Fluid Power (FMFP)

December 27-29, 2021, BITS Pilani, Pilani Campus, RJ, India.

FMFP2021–03–273

Numerical analysis of generation of Colburn-j factor for the plain rectangular fins of a
compact heat exchanger using CFD approach

P. Raghu Vamsi1, S.M.K Hemanth1 and Chennu Ranganayakulu1

1Department of Mechanical Engineering, BITS Pilani, Pilani-333031, India

ABSTRACT these plate-fin heat exchangers is achieved by employing

A Compact Heat Exchanger is characterized by a high various types of heat transfer surfaces such as plain fins, wavy
heat transfer area per unit volume, which can be achieved by fins, serrated fins, pin fins, etc. These fins shapes are embossed
using very high-density fins. A plain fin is one of the most on a metallic plate and then it is enclosed between two plates.
commonly used fins as it has advantages such as low-pressure Similarly, all the plates are stacked together to form a heat
drop and it is easy to fabricate several types of fins by changing exchanger. The plate heat exchangers are more effective
its geometric parameters such as fin height (h), fin spacing (s), compared to conventional shell and tube heat exchangers, but
and fin thickness (t). In this study, a numerical investigation is there are some limitations such as low operating pressures and
carried out on a plain rectangular fin to establish a correlation high-pressure drops.
for the Colburn-j factor. The analysis is done for a wide range
of Reynolds (Re) numbers covering both the laminar region 2. LITERATURE REVIEW AND OBJECTIVE
(400 ≤ Re ≤ 2500) and turbulent region (3500 ≤ Re ≤ 12000).
There are different types of fins used in Plate-fin heat
A CFD-based numerical study is performed using ANSYS
exchangers such as plain fins, serrated fins, louvered fins, offset
Fluent 21 with air at 300 K as the working fluid. In this
strip fins, wavy fins, etc. The main purpose of the use of fins is
numerical analysis, the Colburn-j factor data is obtained for
to increase the heat transfer rate. Many researchers carried out
different Reynolds numbers as well as for the different non-
an experimental study for different types of fins in a plate-fin
dimensional geometrical parameters such as the fin height, fin
heat exchanger. The type of fin required is selected based on
spacing, and fin thickness values. This data has been validated
requirements such as compactness, manufacturer availability,
with experimental data available in the open literature. The
cost, and thermal performance. Kays and London [1] conducted
correlations of the Colburn-j factor are established for a wide
an experimental study on various plate-fin geometries and
range of Reynolds numbers along with the geometric
established the Colburn-j factor and friction factor f. They
parameters of the plain fins covering the entire operational
carried out experiments on a range of Reynolds numbers
range of the Compact Heat Exchangers for Aerospace as well
ranging from 800 to 12000 on the plain rectangular fin. Ismail
as for other applications.
et al. [2] worked on thermal and hydraulic design of offset fin
and wavy fins. They established variations of j and f in the
Keywords: Compact Heat exchanger, Plain Rectangular Fins,
turbulent region. The friction factor is observed to be high in
Colburn-j factor, Reynolds number, fin height, fin thickness,
offset fins compared to wavy fins.
and fin spacing.
Pallavi and Ranganayakulu [3] have developed Colburn-j
1. INTRODUCTION and friction factor, f correlations on offset fin numerically.
They worked on different offset fin geometries by altering the
A compact heat exchanger is a device that is used for the
geometric parameters of fin height, fin thickness, fin spacing,
transfer of thermal energy between two or more fluids at
and fin lance length and established correlations for a range of
different temperatures. In the plate-fin type of the compact heat
Reynolds number, i.e., 100<Re<1000 for the laminar region
exchangers, the fluids are separated by the separating plates as
and 1000<Re<7500 in the turbulent region. Rui Song et al. [4]
the primary heat transfer surface. Typical applications involve
worked on developing new j and f correlations for offset strip
the heating or the cooling of a fluid stream of concern and
fin based on old data of Manglik et al. [5], Joshi et al. [6], and
evaporation or condensation of two-phase applications in the
Wieting and Allan [7] results. It was observed that the results
aerospace industry. Some of the required qualities for a
obtained are with an average deviation of 13% and are carried
compact heat exchanger in the aerospace industry are good
out in both turbulent and laminar regions. Ramana Murthy et al.
thermal effectiveness, low-pressure drop, and good surface area
[8] worked on establishing j and f correlations for serrated fin
to volume ratio, which can be achieved > 750 m2 per m3 with
using ANSYS Fluent. The numerical study was carried out by
plate-fin construction.
varying geometric parameters of serrated fin such as fin height
Nowadays, the plate-fin compact heat exchanger is one of (h), fin thickness (t), fin spacing (s), and fin length (l). The
the most widely used heat exchangers in sectors such as correlations were established in the laminar region
aviation, pharmaceutical, cryogenic, and automobile industries (100<Re<1000) and turbulent region (1000<Re<15000) with
due to its compactness. The enhancement of heat transfer in water as the fluid medium. There are very few papers on

© The Author(s), under exclusive license to Springer Nature Singapore Pte Ltd. 2023 339
S. Bhattacharyya et al. (eds.), Fluid Mechanics and Fluid Power (Vol. 3),
Lecture Notes in Mechanical Engineering, https://doi.org/10.1007/978-981-19-6270-7_57
340 P. Raghu Vamsi et al.

numerical analysis of compact plain fins available in the open forces, they are taken as zero. The energy equation for the flow
literature for designer's use. However, Bala Sundar Rao is given below:
Ramisetty et al. [9-10] worked on j and f correlations for the ∂
plain fins with air as fluid medium and it is found that these (ρ𝐸) + ∇. [𝑢(ρE + p )] = ∇. (𝐾𝑒𝑓𝑓 ∇𝑇 + (𝜏𝑒𝑓𝑓 . 𝑢)) + 𝑆ℎ (9)
∂t
correlations underpredict both j and f for plain fins. Hence,
there is a need to generate accurate correlations for the same. where E is the total energy of the fluid, 𝐾𝑒𝑓𝑓 is the effective heat
The complete Compact Heat Exchanger design methods, transfer coefficient, 𝜏𝑒𝑓𝑓 is the viscous stress and p is the
numerical analysis, and testing methods were presented by pressure. The equations 5-10 are taken as reference from
Ranganayakulu and Seetharamu. [11]. Anderson et al. [12].
The numerical procedure adopted in this paper is similar to
3. METHODOLOGY the procedure adopted by Patankar [13]. Different turbulence
3.1. Data reduction: models are taken into consideration to know which model gives
a better result. The analysis is conducted using six turbulence
The Colburn factor (j) for the plate-fin is calculated using models as shown in Table 1 for Re=6000 on a plain rectangular
the empirical formula given below: fin with a fin density of 6.2 per inch taken from Kays and
London [1] and this experimental data is used for validation of
𝑗 = 𝑆𝑡𝑃𝑟 2/3 (1) CFD results.
Where St is the Stanton number and Pr is the Prandtl number Table 1: Comparison of different turbulence models
𝑁𝑢
𝑆𝑡 = (2)
𝑅𝑒𝑃𝑟
S. Turbulence j value by j- Difference
where Nu is the Nusselt number, Re is the Reynolds number. No model CFD experimental (%)
Using equation (1) & equation (2), Colburn-j factor can be value ([1])
written as,
1 SST k 0.003923 0.00325 20.70
𝑗=
𝑁𝑢
(3) omega
𝑅𝑒𝑃𝑟 1/3
2 Standard k 0.003859 0.00325 18.74
The Colburn-j factor equation established in Rui Song et al [4]
epsilon
is taken as reference for the numerical study, which is given
with
below:
standard
𝑗=
𝐷ℎ
𝑃𝑟 2/3 𝑙𝑛 (
𝑇𝑤 −𝑇𝑖𝑛
) (4) wall
4𝑙 𝑇𝑤 −𝑇𝑜𝑢𝑡 function
where, 𝐷ℎ is the hydraulic diameter of the fin, 𝑇𝑤 is the wall 3 Standard k 0.003628 0.00325 11.64
temperature, 𝑇𝑖𝑛 is the inlet temperature of fluid and 𝑇𝑜𝑢𝑡 is the epsilon
outlet temperature of the fluid. with
scalable
The above equations taken from the Rui Song et al. [4] can be wall
calculated using the properties of the fluid, hydraulic diameter, function
geometric parameters of the fin, and temperatures obtained 4 Standard k 0.004039 0.00325 24.28
from the numerical study. omega
3.2. Governing Equations:
5 Realizable 0.003369 0.00325 4.21
The governing equations for the flow are taken according to k epsilon
Anderson et al. [12]. The continuity equation for the flow is with
given below: scalable
∂ρ ∂(ρux) ∂(ρuy ) ∂(ρuz )
wall
+ + + =0 (5) function
∂t ∂x ∂y ∂z
6 RSM with 0.003455 0.00325 6.31
where 𝑢𝑥 , 𝑢𝑦 and 𝑢𝑧 are velocity components along x, y, and z scalable
which are the directions of the flow. The momentum equations wall
for the flow are given below: function
∂ 𝜕𝑃
(ρux ) + ∇. (ρux u ) = − + ∇. (μ∇ux ) + 𝑆𝑥 (6) Based on the comparative study from Table 1, it is
∂t 𝜕𝑥
∂ 𝜕𝑃 observed that the Realizable k-epsilon with scalable wall
(ρuy ) + ∇. (ρuy u ) = − + ∇. (μ∇uy ) + 𝑆𝑦 (7) function model gives the lowest difference among the six
∂t 𝜕𝑦
∂ 𝜕𝑃
models studied when compared with experimental data [1].
(ρuz ) + ∇. (ρuz u ) = − + ∇. (μ∇uz ) + 𝑆𝑧 (8) Hence, this model is used for turbulence modelling, for
∂t 𝜕𝑧
Reynolds numbers ranging from 3500 to 12000, and the laminar
where 𝑆𝑥 , 𝑆𝑦 and 𝑆𝑧 are source terms that include body forces model is used for Reynolds numbers ranging from 400 to 2500.
and other surface forces present in the flow. In absence of those