Flow Across Banks of Tubes

Heat transfer to or from a bank (or bundle) of tubes in cross flow is relevant to numerous
industrial applications, such as steam generation in a boiler or air cooling in the coil of an
air conditioner. The geometric arrangement is shown schematically
in Figure 7.10. Typically, one fluid moves over the tubes, while a second fluid at a different
temperature passes through the tubes. In this section we are specifically interested in the
convection heat transfer associated with cross flow over the tubes. The tube rows of a bank
are either staggered or aligned in the direction of the fluid velocity V (Figure 7.11). The
configuration is characterized by the tube diameter D and by the transverse pitch ST and
longitudinal pitch SL measured between tube centers. Flow conditions within the bank are
dominated by boundary layer separation effects and by wake interactions, which in turn
influence convection heat transfer.
The heat transfer coefficient associated with a tube is determined by its position in the
bank. The coefficient for a tube in the first row is approximately equal to that for a single
tube in cross flow, whereas larger heat transfer coefficients are associated with tubes of the
inner rows. The tubes of the first few rows act as a turbulence-generating grid, which
increases the heat transfer coefficient for tubes in the following rows. In most
configurations, however, heat transfer conditions stabilize, such that little change occurs in
the convection coefficient for a tube beyond the fourth or fifth row.
Generally, we wish to know the average heat transfer coefficient for the entire tube bundle.
For airflow across tube bundles composed of 10 or more rows (NL
10), Grimison [19] has obtained a correlation of the form
It has become common practice to extend this result to other fluids through insertion

of the factor 1.13Pr1/3, in which case

 (7.60)

All properties appearing in the above equations are evaluated at the film temperature. If
NL , a correction factor may be applied such that

D|(NL<10) = C2D|(NL (7.61)

where C2 is given in Table 7.6. The Reynolds number ReD,max for the foregoing correlations
is based on the maximum fluid velocity occurring within the tube bank. For the aligned
arrangement, Vmax occurs at the transverse plane A1 of Figure 7.11a, and from the mass
conservation requirement for an incompressible fluid
 Vmax=Vo (7.62)

For the staggered configuration, the maximum velocity may occur at either the transverse
plane A1 or the diagonal plane A2 of Figure 7.11b. It will occur at A2 if the rows are spaced
such that

-D) < -D) The factor of 2 results from the bifurcation experienced by the fluid
moving from the A1 to the A2 planes. Hence Vmax occurs at A2 if

in which case it is given by

Vmax=Vo (7.63)
If Vmax occurs at A1 for the staggered configuration, it may again be computed from
Equation 7.62.
Table 7.6 Correction factor C2 of Equation 7.61 for NL <
10. NL 1 2 3 4 5 6 7 8 9
Aligned 0.64 0.80 0.87 0.90 0.92 0.94 0.96 0.98 0.99
Staggered 0.68 0.75 0.83 0.89❑❑0.92 0.95 0.97 0.98 0.99

More recent results have been obtained and Zukauskas [15] has proposed a correlation
of the form
 D=C

where all properties except Prw are evaluated at the arithmetic mean of the fluid inlet and
outlet temperatures and the constants C and m are listed in Table 7.7. The need to
evaluate fluid properties at the arithmetic mean of the inlet (Ti) and outlet (To)
temperatures is dictated by the fact that the fluid temperature will decrease or increase,
respectively, due to heat transfer to or from the tubes. If the fluid temperature change, |
Ti – To|, is large, significant error could result from evaluation of the properties at the
inlet temperature.
If NL < 20, a correction factor of C2 may be applied such that

|(NL < 20) = C2D | (NL (7.65)

where C2 is given in Table 7.8.

Since the fluid may experience a large change in temperature as it moves through the tube
bank, the heat transfer rate could be significantly overpredicted by using ΔT = Tw -T∞ as
the temperature difference in Newton’s law of cooling. As the fluid moves through the
bank, its temperature approaches Tw and Tw -T∞ decreases.
In Chapter 11 the appropriate form of ΔT is shown to be a log-mean temperature
difference, the fluid moves through the bank, its temperature approaches Tw and |ΔT|
decreases.

Δ Tlm =

where Ti and To are temperatures of the fluid as it enters and leaves the bank,

respectively. The outlet temperature, which is needed to determine Δ Tlm ,

may be estimated from
 =exp (7.67)
where N is the total number of tubes in the bank and NT is the number of

tubes in the transverse plane. Once Δ Tlm is known, the heat transfer rate per
unit length of the tubes may be computed from
q’= N(D Δ Tlm) (7.68)
EXAMPLE 7.7
Pressurized water is often available at elevated temperatures and may be used
for space heating or industrial process applications. In such cases it is
customary to use a tube bundle in which the water is passed through the
tubes, while air is passed in cross flow over the tubes. Consider a staggered
arrangement for which the tube out-side diameter is 16.4 mm and the
longitudinal and transverse pitches are SL = 34.3 mm and ST = 31.3 mm. There
are seven rows of tubes in the airflow direction and eight tubes per row. Under
typical operating conditions the cylinder surface temperature is at 70°C, while
the air upstream temperature and velocity are 15°C and 6 m/s, respectively.
Determine the air-side convection coefficient and the rate of heat transfer for
the tube bundle. What is the air-side pressure drop?
Solution :
Known : Geometry and operating codition in the tube bank.
a) Calculate convection film heat transfer coefficient b) Calculate pressure
drop
EXAMPLE 6-10 Heating of Air with In-Line Tube Bank
Air at 1 atm and 10◦C flows across a bank of tubes 15 rows high and 5 rows deep at a
velocity of 7 m/s measured at a point in the flow before the air enters the tube bank. The
surfaces of the tubes are maintained at 65◦C. The diameter of the tubes is 1 in [2.54 cm];
they are arranged in an in-line manner so that the spacing in both the normal and
parallel directions to the flow is 1.5 in [3.81 cm]. Calculate the total heat transfer per
unit length for the tube bank and the exit air temperature.

Solution
❑❑
The constants for use with Equation (6-17) may be obtained from Table 6-4, using
SL/D=38.1 mm/25.4 mm=ST/D=1.5 so that
C =0.278 and n=0.620.
The properties of air are evaluated at the film temperature, which at entrance to the
tube bank is
= (65+10)/2=37.5 ◦C=310.5 K
ρf =MP/RT=(29 kg/kmol)(101325 Pa)/(8314 J/kmol.K x310.5 K) =1.137 kg/m3

μf =1.894×10−5 kg/m ·s

kf = 0.027 W/m · ◦C [0.0156 Btu/h · ft · ◦F]

Cp = 1007 J/kg K Pr = 0.706

To calculate the maximum velocity, we must determine the minimum flow area. From
Figure 6-14 we find that the ratio of the minimum flow area to the total frontal area is
(ST − D)/ST =
The maximum velocity is thus
Vmax =Vo =(7 m/s) = 21 m/s; Vmax =21 m/s

where Vo is the incoming velocity before entrance to the tube bank. The Reynolds number
is computed by using the maximum velocity.
=DVmax/μ= (0.0254 m)(21 m/s) (1.137 kg/m3)/ 1.894×10−5 kg/m.s=32 020
=32020
=C
=0.278(32020)0.62(0.706)1/3=153.82
=153.82=D/k; .k/D=153.82(0.027 W/m.K)/0.0254 m=163.5
=163.5 W/m2.K
This is the heat-transfer coefficient that would be obtained if there were 10 rows of tubes
in the direction of the flow. Because there are only 5 rows, this value must be multiplied
by the factor 0.92, as determined from Table 6-5.
The total surface area for heat transfer, considering unit length of tubes, is
A=N(DL)=(15x5)(DL)=75x x0.0254m (1 m)= 5.985 m2/m

where N is the total number of tubes.

Before calculating the heat transfer, we must recognize that the air temperature increases
as the air flows through the tube bank. Therefore, this must be taken into account when
using
q=h A(Tw −Tb)
=C

where the constants C and n are tabulated in Table 6-2. Properties for use with Equation
(6-17) are evaluated at the film temperature as indicated by the subscript f.
Table 6-4 Modified correlation of Grimison for heat transfer in tube banks of 10 rows or
more, from Reference 12, for use with Equation (6-17).
FREE CONVECTION
In preceding chapters we considered convection transfer in fluid flows that originate from
an external forcing condition. For example, fluid motion may be induced by a fan or a pump,
or it may result from propulsion of a solid through the fluid. In the presence of a
temperature gradient, forced convection heat transfer will occur. Now we consider
situations for which there is no forced velocity, yet convection currents exist within the
fluid. Such situations are referred to as free or natural convection, and they originate when
a body force acts on a fluid in which there are density gradients. The net effect is a
buoyancy force, which induces free convection currents.
In the most common case, the density gradient is due to a temperature gradient, and the
body force is due to the gravitational field. Since free convection flow velocities are
generally much smaller than those associated with forced convection, the corresponding
convection transfer rates are also smaller. It is perhaps tempting to therefore attach less
significance to free convection processes. This temptation should be resisted. In many
systems involving multimode heat transfer effects, free convection provides the largest
resistance to heat transfer and therefore plays an important role in the design or
performance of the system. Moreover, when it is desirable to minimize heat transfer rates
or to minimize operating cost,
free convection is often preferred to forced convection.
the operating temperatures of power generating and electronic devices. It plays a major
role in a vast array of thermal manufacturing applications. Free convection is important in
establishing temperature distributions within buildings and in determining heat losses or
heat loads for heating, ventilating, and air conditioning systems. Free convection
distributes the poisonous products of combustion during fires and is relevant to the
environmental sciences, where it drives oceanic and atmospheric motions, as well as the
related heat transfer and mass transfer processes.
In free convection fluid motion is due to buoyancy forces within the fluid, while in
forced convection it is externally imposed. Buoyancy is due to the combined presence of a
fluid density gradient and a body force that is proportional to density. In practice, the
body force is usually gravitational, although it may be a centrifugal force in rotating fluid
machinery or a Coriolis force in atmospheric and oceanic rotational motions. There are
also several ways in which a mass density gradient may arise in a fluid, but for the most
common situation it is due to the presence of a temperature gradient. We know that the
density of gases and liquids depends on temperature, generally decreasing (due to fluid
expansion) with increasing temperature ( ⵼/ T ⵼ 0).