Location via proxy:   [ UP ]  
[Report a bug]   [Manage cookies]                

HE Theory

Download as pdf or txt
Download as pdf or txt
You are on page 1of 6

Unit Operations Lab Heat Exchanger

HEAT EXCHANGER
Heat exchange is an important unit operation that contributes to efficiency and safety of
many processes. In this project you will evaluate performance of three different types of heat
exchangers (tubular, plate, and shell & tube). All these heat exchangers can be operated in both
parallel- and counter-flow configurations. The heat exchange is performed between hot and cold
water.

Concepts to Review
 Heat exchanger configurations; parallel vs. counter-current flow
 LMTD and NTU methods for analysis of heat exchangers
 Convective and conductive heat transfer
 Reynolds, Prandtl, and Nusselt numbers
 Energy balance closure
Degrees of Freedom and Performance Indicators
The following degrees of freedom can be varied in this experiment:
 Flow rate 𝑚̇ℎ and inlet temperature Th,i of hot water,
 Flow rate 𝑚̇𝑐 of cold water,
 Heat exchanger type (tubular, shell & tube, or plate),
 Flow direction (parallel or counter-flow).
You will investigate effects of these variables on the following performance indicators:
 Return temperatures Th,o and Tc,o of the hot and cold water
 Heat flow rate q
 Overall heat transfer coefficient U

Objectives
 Verify the overall energy balance.
 Investigate effects of the control parameters and the heat exchanger configuration on the
rate of the heat transfer, q, and the overall heat transfer coefficient, U.
 Investigate dependence of convective heat transfer coefficients on the Reynolds number
and the flow geometry. To accomplish this goal, it is necessary to experimentally
determine contributions of individual convective heat transfer coefficients to the overall
heat transfer coefficient U.
 Check validity of the theoretical models.

1-1
Unit Operations Lab Heat Exchanger

Theory
Notes
1. The theory presented here is based on Ref. 1. Please refer to this book for more details.
2. Subscripts h and c refer to the hot and cold fluids, respectively, and subscripts i and o refer to
the inlet and outlet of the heat exchanger, respectively. E.g., Th,i denotes temperature of the
hot fluid at the inlet and Tc,o denotes temperature of the cold fluid at the outlet.

Overall Heat Transfer Coefficient U


Consider energy balance in a differential segment of a single-pass heat exchanger shown
schematically in Fig. 1-1. The rate of heat transfer in this segment is
𝑑𝑞(𝑥) = 𝑈∆𝑇(𝑥)𝑑𝐴(𝑥), (1)
where U is the overall heat transfer coefficient, ΔT is the local temperature difference between
the hot and cold fluids, and dA is the contact area in the differential segment.
The overall heat transfer coefficient is inversely proportional to the total resistance Rtot to
the heat flow. The latter is the sum of (1) resistance Rconv,h to convective heat transfer from the
hot fluid to the partition between the fluids, (2) resistance Rp to thermal conduction through the
partition, and (3) resistance Rconv,c to convective heat transfer from the partition to the cold fluid.
Therefore,
1 1 (2)
𝑈= = .
𝑅𝑡𝑜𝑡 𝑅𝑐𝑜𝑛𝑣,ℎ + 𝑅𝑝 + 𝑅𝑐𝑜𝑛𝑣,𝑐

Figure 1-1. Energy balance in a differential element of a single pass heat exchanger operated in the counter-flow
regime. Red, blue, and gray colors represent the hot fluid, cold fluid, and the partition between the fluids,
respectively. The dashed rectangle shows a differential segment corresponding to the energy balance Eq. (1). Three
resistances (Rconv,h, Rp, and Rconv,c) contributing to the total resistance to the heat transfer are indicated schematically.

Conductive heat transfer. The resistance Rp to heat transfer through the partition
depends on the system geometry. In the current experiment, you will need to consider heat

1-2
Unit Operations Lab Heat Exchanger

conduction in the radial direction of a cylindrical tube and heat conduction across a thin plate.
Resistances in both of these cases can be obtained analytically be solving the heat diffusion
equation. Often, Rp can be neglected in comparison with Rconv,c and Rconv,h. If you choose to
neglect Rp in analysis of your experiments, you should provide a justification for this
approximation in your report.
Convective heat transfer. Resistance to the convective heat transfer is inversely
proportional to the convective heat transfer coefficient, h = 1/Rconv. The convective heat transfer
coefficient depends on fluid properties, flow geometry, and the flow rate. It is convenient to
describe this dependence using several dimensionless numbers, namely the Reynold number
𝐿𝑣𝜌 (3)
𝑅𝑒 = ,
𝜇
the Prandltl number
𝑐𝑝 𝜇 (4)
𝑃𝑟 = ,
𝑘
and the Nusselt number
ℎ𝐿 (5)
𝑁𝑢 = .
𝑘

Here, ρ, μ, k, and cp are the density, viscosity, thermal conductivity, and heat capacity of the fluid,
v is the flow velocity, and L is characteristic length. The choice of L depends on the system
geometry. For example, for a flow in a circular pipe, L is the pipe diameter.
Relationship between Re, Pr, and Nu depends on the system geometry and whether the
flow is laminar and turbulent. For example, for a turbulent flow inside a circular pipe,
𝑁𝑢 = 0.027𝑅𝑒 0.8 𝑃𝑟 1/3 (6)
In your report you should provide all correlations relevant to the geometries and the flow
regimes (turbulent and/or laminar) considered in this experiment. Whenever possible, you should
compare these correlation with your experimental data. Since our experimental system allows us
to determine directly only the overall heat transfer coefficient, it is necessary to design a strategy
for extracting individual conductive heat transfer coefficients from your data.
For example, in order to check the Nu-Re correlations for the convective heat transfer in
the cold water, we can vary the cold water flow rate 𝑚̇𝑐 while keeping all other system
parameters constant. Therefore, changes in the overall resistance Rtot observed in this set of
experiments will be exclusively due to changes in the resistance 𝑅𝑐𝑜𝑛𝑣,𝑐 of the cold water
convective transfer. This will allow us to determine how hconv,c depends on 𝑚̇𝑐 and, hence, how
Nuc depends on Rec. A similar analysis can be performed for the convective transfer in hot water.

1-3
Unit Operations Lab Heat Exchanger

Relationships between Overall Heat Transfer Coefficient and the Inlet


and Outlet Fluid Temperatures
In the previous section we discussed heat transfer in a differential segment of the heat
exchanger. Here, we analyze the heat transfer in the entire heat exchanger. Heat exchangers are
usually analyzed using either the Logarithmic Mean Temperature Difference (LMTD) or the
Effectiveness – Number of Transfer Units (ε-NTU) methods. The LMTD method is convenient
for determining the overall heat transfer coefficient based on the measured inlet and outlet fluid
temperatures. The ε-NTU method is more convenient for prediction of the outlet fluid
temperatures if the heat transfer coefficient and the inlet temperatures are known.
The analysis presented below assumes that
1. There is no energy loss to the environment
2. Heat exchanger is at a steady-state
3. There are no phase changes in the fluids
4. Heat capacities of the fluids are independent of temperature
5. Overall heat transfer coefficient is independent of the fluid temperature and position
within the heat exchanger.
Since all heat exchangers considered in this experiment have a single pass for both the hot and
cold fluids, the discussion below is limited to single-pass heat exchangers. Qualitative
dependence of the fluid temperature on position inside a single-pass heat exchanger is shown in
Fig. 1-2.

Figure 1-2. Temperature profiles in (a) counter-flow and (b) parallel flow single pass heat exchangers. Note that in a
counter-flow heat exchanger the outlet temperature of the cold fluid can exceed the outlet temperature of the hot
fluid but this cannot happen in a parallel flow system.

1-4
Unit Operations Lab Heat Exchanger

Logarithmic Mean Temperature Difference (LMTD) Method


The total heat transfer rate is
2 (7)
𝑞 = ∫1 𝑑𝑞 ,

where dq is given by Eq. (1) and points 1 and 2 are defined in Fig. 1-2. This integral cannot be
computed by a direct integration of Eq. (1), since ΔT typically varies with position in the heat
exchanger, as shown in Fig. 1-2.
However, it is possible to obtain q by combining Eq. (1) with energy balances in
differential segments of the heat exchanger,
𝑑𝑞 = −𝐶ℎ 𝑑𝑇ℎ = 𝐶𝑐 𝑑𝑇𝑐 (8)
Here, dTk is the temperature change of fluid k (k = c or h) in the interval under consideration, and
Ck is the heat capacity rate of fluid k,
𝐶𝑘 = 𝑚̇𝑘 𝑐𝑘 , 𝑘 = 𝑐 or ℎ, (9)
where 𝑚̇𝑘 and ck are the mass flow rate and heat capacity of fluid k, respectively. This analysis
yields
𝑞 = 𝑈𝐴 ∆𝑇𝑙𝑚 , (10)
where A is the total contact area and ΔTlm is the logarithmic mean temperature difference
(LMTD),
∆𝑇1 – ∆𝑇2 (11)
∆𝑇𝑙𝑚 =
ln(∆𝑇1 /∆𝑇2 )
Here, ΔTk refers to temperature difference between the hot and cold fluids at point k (k = 1 or 2),
i.e.
∆𝑇1 = 𝑇ℎ,𝑖 − 𝑇𝑐,𝑜 and ∆𝑇2 = 𝑇ℎ,𝑜 − 𝑇𝑐,𝑖 (12)
for the counter-current flow and
∆𝑇1 = 𝑇ℎ,𝑖 − 𝑇𝑐,𝑖 and ∆𝑇2 = 𝑇ℎ,𝑜 − 𝑇𝑐,𝑜 (13)
for the parallel flow.

Notes
1. If heat capacity rates of the cold and hot fluids are the same and the heat exchanger is
operated in the counter-flow regime then ΔT is independent of position in the heat exchanger.
In this case Eqs. (10), (11) are not applicable and the total heat transfer rate q should be
obtained by direct integration of Eq. (1).

1-5
Unit Operations Lab Heat Exchanger

2. This result holds for single pass heat exchangers only. However, the LMTD method can be
extended to more complex heat-exchanger designs (e.g., multi-pass and cross-flow systems)
using a correction factor (see Ref. 1).
3. The LMTD formulas can be rearranged in a form more suitable to predict outlet temperatures
if the overall heat transfer coefficient U is known. The resulting formulas constitute the ε-
NTU method. A common mistake made is to use the ε-NTU to “verify” U obtained with the
LMTD method by computing the outlet temperatures using this U and comparing the
obtained temperatures with the experimentally measured ones. However, since U is
computed using the experimental values of outlet temperatures and the ε-NTU method
directly follows from the LMTD method, the “predictions” for the outlet temperatures are
guaranteed to be the same as the experimental values. I.e., this approach cannot be used to
validate the heat transfer coefficient values.

References
1. F. P. Incropera, D. P. DeWitt, T. L. Bergman, and A. S. Lavine, “Fundamentals of Heat and
Mass Transfer”, 6th ed. (Wiley, 2007).

1-6

You might also like