Heat Exchangers and Thermal-System Design

Introduction and Context

Heat exchangers are devices that transfer thermal energy between two or more fluids at different temperatures, usually without allowing the fluids to mix. They are central to power plants, refrigeration systems, chemical processing, automobiles, electronics cooling, nuclear systems, desalination plants, and renewable-energy installations. A heat exchanger is not designed in isolation: its performance is coupled to thermodynamics, fluid mechanics, materials, manufacturing, control, safety, maintenance, and economic constraints.

At postgraduate level, heat-exchanger design requires more than calculating a heat-transfer area. The designer must determine the required heat duty, select a suitable configuration, predict heat-transfer coefficients and pressure losses, evaluate fouling and transient operation, assess mechanical integrity, and optimize the complete thermal system. The best design is therefore not necessarily the smallest or most thermally effective exchanger; it is the design that satisfies performance, reliability, safety, environmental, and life-cycle cost requirements.

Learning Objectives

After studying this chapter, the reader should be able to:

1. Explain the thermodynamic and fluid-mechanical principles governing heat exchangers.
2. Distinguish major exchanger configurations and select an appropriate type for a given duty.
3. Apply the logarithmic mean temperature difference and effectiveness–NTU methods.
4. Estimate heat-transfer coefficients, overall thermal resistance, area, and pressure drop.
5. Account for fouling, phase change, temperature-dependent properties, and non-ideal flow.
6. Integrate heat exchangers into broader thermal-system and energy-management decisions.
7. Critically evaluate design trade-offs, failure modes, sustainability, and emerging technologies.

1. Fundamental Energy Balance

For a steady-flow heat exchanger with negligible heat loss to the surroundings, the heat lost by the hot stream equals the heat gained by the cold stream:

Here, Q̇ is the heat-transfer rate in W; ṁ is mass flow rate in kg s⁻¹; c_p is specific heat capacity at constant pressure in J kg⁻¹ K⁻¹; T_i and T_o denote inlet and outlet temperatures; and subscripts h and c refer to hot and cold streams. This relation assumes single-phase streams and approximately constant specific heats. For variable properties, the enthalpy form is more general:

If heat is lost to the environment, an additional term must be included in the overall energy balance. In a phase-change exchanger, enthalpy changes include latent heat. For example, condensation at nearly constant pressure may be represented by Q̇ = ṁ h_fg, where h_fg is the latent heat of vaporization.

The second law imposes an important restriction: heat flows from the hotter stream toward the colder stream, and entropy generation must be non-negative. For an adiabatic exchanger operating at steady state,

where s is specific entropy in J kg⁻¹ K⁻¹. Entropy generation is caused by finite temperature differences, fluid friction, mixing, leakage, and non-ideal phase-change processes. A thermally effective design should also seek low entropy generation, although minimizing entropy generation and minimizing capital cost do not always produce the same geometry.

2. Heat-Exchanger Configurations

2.1 Recuperative and regenerative exchangers

In a recuperative exchanger, the hot and cold fluids remain separated by a wall. Shell-and-tube, plate, double-pipe, and compact exchangers are recuperative devices. In a regenerative exchanger, a solid matrix is alternately heated by the hot stream and cooled by the cold stream. Regenerators are used in gas turbines, cryogenic plants, and some industrial furnaces, but leakage and matrix contamination can be significant concerns.

2.2 Parallel flow, counterflow, and crossflow

In parallel flow, both streams enter at the same end and move in the same direction. The temperature difference is initially large but decreases rapidly. In counterflow, the streams move in opposite directions, generally providing a higher mean temperature driving force and allowing the cold outlet temperature to approach the hot inlet temperature. Crossflow arrangements are common in air heaters and automotive radiators; one or both streams may be mixed or unmixed in the transverse direction.

Counterflow is often preferred when high effectiveness is required. However, configuration selection also depends on pressure drop, cleanability, mechanical constraints, cross-contamination risk, vibration, and manufacturing cost.

2.3 Principal industrial types

3. Logarithmic Mean Temperature Difference Method

The local heat-transfer rate is proportional to the local temperature difference. If the overall conductance is constant, integration along the exchanger gives the logarithmic mean temperature difference (LMTD):

For counterflow operation,

For parallel flow,

The design equation is

where U is the overall heat-transfer coefficient in W m⁻² K⁻¹, A is the selected reference area in m², and F is a correction factor for multipass or crossflow arrangements. For pure counterflow or pure parallel flow, F = 1. The terminal temperature differences must remain physically meaningful; a temperature cross or an incorrect stream arrangement may invalidate a simple LMTD calculation.

Worked Example 1: Counterflow area calculation

Hot oil enters a counterflow exchanger at 150 °C and leaves at 90 °C. Water enters at 25 °C and leaves at 65 °C. The oil flow rate is 2 kg s⁻¹ with c_p = 2.2 kJ kg⁻¹ K⁻¹. The overall coefficient based on outside area is 450 W m⁻² K⁻¹. Estimate the required area, neglecting heat loss.

First, calculate the heat duty:

Q̇ = 2 × 2200 × (150 − 90) = 264000 W

For counterflow,

∆T₁ = 150 − 65 = 85 K and ∆T₂ = 90 − 25 = 65 K.

Thus,

∆T_lm = (85 − 65)/ln(85/65) ≈ 74.6 K.

Since the arrangement is pure counterflow, F = 1. Therefore,

A = Q̇/(U∆T_lm) = 264000/(450 × 74.6) ≈ 7.87 m².

The calculated area is a clean-design value. A practical design would include fouling resistance, uncertainty in U, manufacturing tolerances, and an appropriate design margin.

4. Effectiveness–NTU Method

The LMTD method is convenient when all terminal temperatures are known. In rating problems, however, inlet temperatures and flow rates may be known while outlet temperatures are not. The effectiveness–NTU method is then especially useful.

Heat-exchanger effectiveness is defined as

where the maximum possible heat transfer is

and the heat-capacity rates are

The capacity-rate ratio and number of transfer units are

For a counterflow exchanger with constant properties and negligible heat loss,

For C_r = 1,

For a parallel-flow exchanger,

These relations assume steady operation, negligible axial conduction, constant properties, uniform overall coefficient, and the specified flow arrangement. Crossflow formulas depend on whether each fluid is mixed or unmixed and should be taken from validated heat-transfer references or software.

Interpretation of NTU

A low NTU means that the available conductance is small relative to the stream heat-capacity rate. Increasing area or U increases NTU and generally increases effectiveness, but with diminishing returns. This is a major design insight: doubling area does not generally double useful heat recovery. The marginal thermal benefit must be compared with pressure-drop and capital-cost penalties.

5. Overall Heat-Transfer Coefficient

The overall coefficient combines convection on both sides, wall conduction, fouling, and sometimes contact resistance. For a plane wall, the resistance per unit area can be written conceptually as

where h_h and h_c are hot- and cold-side convection coefficients; R_f is fouling resistance in m² K W⁻¹; δ_w is wall thickness; and k_w is wall thermal conductivity. For cylindrical tubes, the exact expression depends on the area basis. Based on outside area A_o,

Here, h_i and h_o are inside and outside coefficients, r_i and r_o are tube radii, L is tube length, and k_w is wall conductivity. The equation demonstrates why area bases must never be mixed without conversion.

For turbulent internal flow, a common starting point is the Dittus–Boelter correlation:

where Nu = hD/k, Re = ρVD/μ, Pr = μc_p/k, and n is commonly 0.4 when the fluid is being heated and 0.3 when it is being cooled. This correlation is applicable approximately to fully developed turbulent flow in smooth circular tubes with moderate property variation. It is not a universal law and should not be used blindly for laminar flow, strong buoyancy, rough tubes, transitional flow, or large temperature-dependent viscosity changes.

For more general internal-flow design, the Gnielinski correlation is often preferable over an appropriate range of Reynolds number:

where f is the Darcy friction factor. The correlation requires careful attention to its stated range and to whether the friction factor is Darcy or Fanning; confusing these definitions introduces a factor-of-four error.

6. Pressure Drop and Pumping Power

Thermal design cannot be separated from hydraulic design. For a straight tube, the Darcy–Weisbach relation is

where ∆p_f is frictional pressure loss, f is the Darcy friction factor, L is flow length, D is hydraulic diameter, ρ is density, and V is mean velocity. Local losses are represented by

where K is a dimensionless loss coefficient. Total pressure drop includes straight-friction, entrance, exit, bend, nozzle, port, and multipass-turn losses.

The pumping or fan power is approximately

where V̇ is volumetric flow rate and η is the pump or fan efficiency. For a complete system, the energy penalty may be more important than the exchanger’s thermal effectiveness. Excessive velocity may increase h but also increases pressure drop, erosion, vibration, and power consumption.

7. Fouling, Scaling, and Degradation

Fouling is the accumulation of unwanted material on heat-transfer surfaces. Mechanisms include particulate deposition, crystallization, corrosion products, biological growth, chemical reaction, and freezing. Fouling increases thermal resistance and may reduce flow area, thereby increasing pressure drop.

A fouling resistance is commonly included in the overall resistance model. However, fouling is time-dependent rather than a fixed material property. A design based only on a nominal fouling factor can be unreliable when the fluid chemistry, velocity, wall temperature, or cleaning schedule changes.

Best practice includes:

• controlling fluid cleanliness and chemistry;
• maintaining velocities high enough to limit deposition but low enough to avoid erosion;
• avoiding dead zones and severe temperature hot spots;
• selecting corrosion-resistant materials;
• providing inspection, isolation, drainage, and cleaning access;
• using online performance monitoring to detect degradation.

The thermal performance can be tracked through an inferred U-value. A falling U-value at comparable operating conditions may indicate fouling, flow maldistribution, sensor error, or a changing fluid property.

8. Phase-Change Heat Exchangers

Condensers and evaporators frequently dominate refrigeration and power-cycle performance. During boiling or condensation, the fluid temperature may remain approximately constant over a substantial region, but the local coefficient depends strongly on quality, mass flux, pressure, surface condition, orientation, and flow regime.

In a condenser, the condensing-side resistance may be small compared with the liquid-side or wall resistance, but non-condensable gases can create a major additional resistance. In an evaporator, dryout or critical heat flux can sharply reduce heat transfer and damage surfaces. Pressure drop changes saturation temperature and therefore affects the thermodynamic driving force.

The simple relation Q̇ = ṁ h_fg applies only when the phase change dominates and sensible enthalpy changes are negligible. In a real evaporator, inlet subcooling, two-phase evaporation, superheating, and pressure losses must be treated separately.

9. Thermal-System Design Methodology

A robust design proceeds iteratively rather than through a single formula.

Step 1: Define the duty and constraints

Specify fluid identities, mass flow rates, inlet and target outlet states, allowable pressure drops, operating range, startup and shutdown conditions, fouling environment, materials, codes, and required availability.

Step 2: Perform thermodynamic feasibility checks

Use enthalpy balances to calculate duty. Check that the requested outlet temperatures do not violate the second law. In a counterflow exchanger, the cold outlet cannot exceed the hot inlet temperature in a physically realizable finite-area exchanger. A very small terminal temperature difference may imply a disproportionately large area.

Step 3: Select the configuration

Consider phase change, pressure and temperature, fluid cleanliness, leakage consequences, maintenance, compactness, thermal stress, and cross-contamination. A plate exchanger may be excellent for clean liquid-to-liquid service but inappropriate for high-pressure toxic gas service.

Step 4: Estimate heat-transfer coefficients and area

Use validated correlations consistent with geometry and flow regime. Include fouling and wall resistance. For multipass equipment, apply an LMTD correction factor or use a segmented numerical model.

Step 5: Check hydraulics and mechanics

Calculate pressure drops, pump or fan power, tube vibration risk, thermal expansion, wall stress, gasket limitations, and relief requirements. A design meeting the thermal duty but exceeding allowable pressure loss is not acceptable.

Step 6: Evaluate off-design operation

Model changes in flow rate, ambient temperature, fluid properties, fouling, control-valve position, and partial load. Thermal systems often spend more time at off-design conditions than at the nominal design point.

Step 7: Optimize life-cycle performance

Compare capital cost, energy cost, cleaning cost, downtime, replacement cost, emissions, and residual value. A larger exchanger may reduce pumping power and improve reliability, while a smaller exchanger may reduce capital expense but increase operating cost and failure risk.

10. Pinch Analysis and Heat-Exchanger Networks

In process plants, individual exchangers should be considered as part of a heat-exchanger network. Pinch analysis identifies thermodynamically constrained regions and opportunities for process-to-process heat recovery. Composite curves represent the available hot and cold enthalpy streams after temperature shifting by a chosen minimum approach temperature, ∆T_min.

Reducing ∆T_min generally increases heat recovery but requires greater exchanger area and capital cost. Increasing ∆T_min reduces area but demands more external heating and cooling. The economically optimal value is determined by the trade-off among energy prices, capital costs, maintenance, and operating flexibility.

Heat integration should not compromise controllability. A network that is thermodynamically efficient but highly sensitive to disturbances may be inferior to a slightly less integrated arrangement with bypasses, trim heaters, and independent control authority.

11. Illustrative Case: Waste-Heat Recovery from an Industrial Furnace

Background

Consider a fictional medium-sized process plant in which hot exhaust gas leaves a furnace at 420 °C. The plant uses pressurized water at 30 °C to produce hot water for a cleaning process. The exhaust contains dust and corrosive species, and the plant operates 7,000 h per year.

Problem

The plant currently rejects useful heat through a stack. The proposed recovery exchanger must deliver 500 kW, maintain an exhaust outlet above the acid-dew-point limit, and operate with a water-side pressure drop below 60 kPa. Dust deposition and corrosion are major risks.

Analysis

A direct gas-to-water exchanger is thermodynamically feasible, but the gas-side coefficient is relatively low and the dirty gas creates fouling. A compact high-area exchanger would reduce volume but might be difficult to clean. The exhaust outlet temperature must be selected with chemical composition and dew-point data, not merely with an assumed temperature approach.

The design team compares a finned-tube economizer with a remotely located plate exchanger supplied by an intermediate clean-water loop. The finned-tube unit has lower capital cost and higher direct effectiveness, but its gas-side fouling is more difficult to manage. The intermediate loop adds a pump and a second temperature difference but isolates the clean process water from contaminated exhaust surfaces.

Approach and solution

The selected design is a modular finned-tube economizer with removable access panels, soot-blowing provisions, corrosion-resistant tube material, a bypass damper, differential-pressure monitoring, and a control strategy that maintains the stack above the minimum safe temperature. The water circuit includes a three-way valve and a trim heater for startup and low-load operation.

The design is rated using enthalpy balances and a segmented model because gas properties and heat-transfer coefficients vary substantially with temperature. Pressure-drop calculations include fin-side friction and dust-loading allowance. The economic evaluation includes avoided fuel use, fan power, cleaning downtime, and expected corrosion replacement.

Outcome

The illustrative design meets the 500 kW duty at clean conditions and retains acceptable performance after the specified fouling interval. Annual fuel savings are significant, but the project is approved only after confirming that the additional fan power and periodic cleaning do not eliminate the expected life-cycle benefit.

Learning points

1. Heat recovery must be designed around the complete system, not only the exchanger core.
2. Fouling and corrosion can dominate long-term performance.
3. Temperature limits may be dictated by chemistry and safety rather than heat-transfer convenience.
4. Bypasses, monitoring, and maintainability are essential for industrial availability.
5. A thermally superior design may not be economically or operationally superior.

12. Advanced Design Considerations

Uncertainty and robust design

Inputs such as flow rate, fouling resistance, material conductivity, and heat-transfer correlations are uncertain. A robust design evaluates ranges rather than a single deterministic point. Sensitivity analysis can identify whether uncertainty in the gas-side coefficient, fouling, or flow distribution dominates the required area.

Thermal stress and fatigue

Large temperature differences and rapid transients produce thermal gradients. Differential expansion between tubes, shells, plates, and welds can cause fatigue. Startup procedures, sliding supports, expansion joints, flexible connections, and appropriate material selection are therefore part of thermal design.

Control and instrumentation

Important measurements include inlet and outlet temperatures, flow rates, pressure drops, differential pressure, wall temperature where necessary, and fluid quality indicators. Control systems may regulate bypass flow, fan speed, coolant flow, or phase-change pressure. Sensor placement must reflect mixing quality; a poorly located temperature sensor can make a healthy exchanger appear to malfunction.

Digital monitoring and artificial intelligence

Digital twins and machine-learning models can estimate fouling resistance, detect leaks, predict cleaning requirements, and identify abnormal pressure-drop trends. Their use should supplement, not replace, physical models and engineering judgment. Training data must represent relevant operating conditions, and alarm logic must remain interpretable enough for safety-critical decisions. Cybersecurity, data integrity, model drift, and fail-safe operation are governance requirements.

Environmental and regulatory considerations

Designers must consider refrigerant regulation, water consumption, chemical cleaning waste, hazardous-fluid containment, pressure-vessel codes, occupational exposure, and end-of-life recycling. Preventing cross-contamination is especially important in food, pharmaceutical, nuclear, and potable-water applications. Materials should be selected for compatibility with both the fluid and the cleaning procedure.

Challenges and Limitations

No single correlation or design method is valid for every exchanger. LMTD calculations become awkward when properties vary strongly or when many zones and phase changes exist. Effectiveness–NTU relations depend on idealized flow arrangements. Computational fluid dynamics can reveal maldistribution and local hot spots but requires suitable turbulence, multiphase, and wall models and may be expensive to validate. Vendor performance data are valuable, but the basis, uncertainty, fouling condition, and test standard must be examined.

Common design errors include confusing gauge and absolute pressure, using inconsistent area bases, applying an inappropriate correlation, ignoring nozzle and return losses, neglecting startup conditions, assuming clean surfaces indefinitely, and optimizing thermal duty while overlooking maintenance or control requirements.

Best-Practice Checklist

• Formulate the enthalpy balance before selecting hardware.
• Check thermodynamic feasibility and temperature cross limits.
• Define the reference area whenever U is reported.
• Use correlations within their validated ranges.
• Include fouling, uncertainty, and off-design conditions.
• Treat pressure drop and pumping power as design variables.
• Provide access for inspection and cleaning.
• Evaluate thermal stress, vibration, corrosion, and leakage.
• Integrate the exchanger with control, safety, and utility systems.
• Compare alternatives using life-cycle cost and environmental impact.

Chapter Summary

Heat exchangers transfer energy between streams and form essential links within thermal systems. Their design begins with a steady-flow energy balance and must satisfy the second law. The LMTD method is effective when terminal temperatures are known, while the effectiveness–NTU method is useful for rating exchangers when outlet temperatures must be predicted. Overall performance depends on convection, wall conduction, fouling, flow arrangement, and surface area.

Thermal design must be coupled with hydraulic analysis because higher velocity can improve heat transfer while increasing pressure loss, power consumption, erosion, and vibration. Phase change, temperature-dependent properties, maldistribution, and fouling require more advanced or segmented analysis. At system level, pinch analysis and heat-exchanger networks reveal opportunities for heat recovery, but economic value, controllability, maintainability, safety, and environmental performance must also be considered.

Modern practice increasingly combines validated physical models, sensors, digital monitoring, and data analytics. Nevertheless, robust engineering judgment remains essential, particularly for uncertain fouling, transient operation, materials compatibility, regulatory compliance, and failure prevention.

Key Terms

Heat exchanger: Device that transfers heat between fluids or between a fluid and a solid surface.

Recuperator: Heat exchanger in which hot and cold streams are separated by a wall continuously.

Regenerator: Device using a matrix that is alternately heated and cooled by different streams.

LMTD: Logarithmic mean temperature difference used as an integrated temperature-driving force.

Effectiveness: Actual heat transfer divided by the maximum thermodynamically possible heat transfer.

NTU: Number of transfer units, defined as UA divided by the minimum heat-capacity rate.

Heat-capacity rate: Product of mass flow rate and specific heat, ṁc_p.

Overall heat-transfer coefficient: Combined conductance coefficient accounting for convection, conduction, and fouling resistances.

Fouling resistance: Added thermal resistance caused by deposits or surface contamination.

Counterflow: Arrangement in which streams move in opposite directions.

Parallel flow: Arrangement in which streams move in the same direction.

Crossflow: Arrangement in which streams move approximately perpendicular to one another.

Thermal effectiveness: Ratio of actual to maximum possible heat transfer.

Pressure drop: Reduction in fluid pressure caused by friction and local flow disturbances.

Pinch analysis: Method for optimizing heat recovery and utility consumption in process systems.

Temperature cross: Condition in which an outlet temperature approaches or exceeds the opposing inlet temperature, requiring careful feasibility analysis.

Thermal stress: Stress caused by non-uniform temperature or constrained thermal expansion.

Heat-exchanger network: Integrated arrangement of exchangers connecting multiple hot and cold process streams.

Life-cycle cost: Total cost including capital, energy, maintenance, downtime, and disposal over the system life.

Review Questions

Short Answer Questions

1. State the steady-flow energy balance for a single-phase, adiabatic heat exchanger.
2. Why is counterflow generally more thermally effective than parallel flow?
3. Define heat-exchanger effectiveness and NTU.
4. Identify four contributions to overall thermal resistance.
5. Why must pressure drop be considered during thermal design?

Long Answer Questions

1. Derive the LMTD relation conceptually from a differential heat balance and explain its assumptions.
2. Compare shell-and-tube, plate, air-cooled, and compact heat exchangers for industrial service.
3. Explain how fouling affects heat-transfer performance, pressure drop, maintenance, and economics.
4. Discuss the use of the effectiveness–NTU method for rating a counterflow exchanger.
5. Explain how pinch analysis supports the design of heat-exchanger networks.

Analytical / Application Questions

1. A counterflow exchanger heats water from 20 °C to 70 °C using oil that cools from 140 °C to 90 °C. Develop the heat balance and determine the two terminal temperature differences. Explain whether the arrangement is thermodynamically feasible.
2. A proposed exchanger meets its heat duty but exceeds the allowable pressure drop by 40%. Evaluate at least three redesign options and discuss their thermal, economic, and operational consequences.
3. Design an investigation plan for an exchanger whose measured U-value has fallen steadily over six months. Include required measurements, possible causes, diagnostic calculations, and corrective actions.

Multiple Choice Questions with Answers

Question 1

Which method is most appropriate when exchanger inlet temperatures and flow rates are known but outlet temperatures are unknown?

A. Effectiveness–NTU method

B. Only the parallel-flow LMTD method

C. Fourier series method

D. Bernoulli equation alone

Correct Answer: A. Effectiveness–NTU method

Question 2

For an adiabatic single-phase heat exchanger, which expression represents the heat-capacity rate of a stream?

A. ṁ/c_p

B. ṁc_p

C. c_p/ṁ

D. ṁρc_p

Correct Answer: B. ṁc_p

Question 3

In a pure counterflow exchanger, the LMTD correction factor is normally:

A. 0

B. 0.5

C. 1

D. Dependent only on fouling

Correct Answer: C. 1

Question 4

Which change generally increases both internal convection coefficient and pressure drop?

A. Increasing mean fluid velocity

B. Decreasing fluid velocity to zero

C. Removing all wall area

D. Reducing density without changing flow

Correct Answer: A. Increasing mean fluid velocity

Question 5

The primary effect of fouling on a heat exchanger is to:

A. Decrease thermal resistance

B. Increase thermal resistance and often increase pressure drop

C. Eliminate the need for cleaning

D. Make the second law inapplicable

Correct Answer: B. Increase thermal resistance and often increase pressure drop

Question 6

For a counterflow exchanger with equal heat-capacity rates, the effectiveness relation is:

A. ε = NTU/(1 + NTU)

B. ε = 1 + NTU

C. ε = NTU²

D. ε = exp(NTU)

Correct Answer: A. ε = NTU/(1 + NTU)

Question 7

The Darcy–Weisbach equation is primarily used to estimate:

A. Latent heat

B. Fouling thickness only

C. Frictional pressure loss

D. Entropy at equilibrium only

Correct Answer: C. Frictional pressure loss

Question 8

Which statement about heat-exchanger area is correct?

A. The reference area used for U must be stated

B. Inside and outside areas are always identical

C. Area has no effect on NTU

D. Area is irrelevant when fouling exists

Correct Answer: A. The reference area used for U must be stated

Question 9

Pinch analysis is mainly used to:

A. Determine molecular mass

B. Optimize heat recovery and utility use

C. Measure tube roughness directly

D. Replace all process controls

Correct Answer: B. Optimize heat recovery and utility use

Question 10

A non-negative entropy-generation rate in a heat exchanger follows from the:

A. First law alone

B. Second law of thermodynamics

C. Ideal-gas law alone

D. Continuity equation alone

Correct Answer: B. Second law of thermodynamics