Sizing a heat exchanger comes down to calculating how much surface area the two fluids need to exchange the right amount of heat. The core equation is Q = U × A × LMTD, where Q is the heat load, U is the overall heat transfer coefficient, A is the required surface area, and LMTD is the log mean temperature difference between the two fluid streams. Your job is to define or estimate every variable except A, then solve for it.
The Fundamental Sizing Equation
The equation Q = U × A × LMTD is the foundation of nearly all heat exchanger sizing. Rearranged to solve for the surface area you need, it becomes:
A = Q / (U × LMTD)
Each variable represents a piece of the puzzle. Q is the total heat duty: how much thermal energy per unit time you need to transfer from the hot fluid to the cold fluid. U is the overall heat transfer coefficient, which captures how easily heat passes through the wall separating the two fluids (including resistance from fouling and the fluid boundary layers on each side). A is the heat transfer surface area you’re solving for. And LMTD accounts for the fact that the temperature difference between the two fluids changes along the length of the exchanger.
Step 1: Define the Heat Load
Before anything else, figure out how much heat needs to move. If you know the flow rate and temperatures of one fluid stream, the heat load is straightforward:
Q = ṁ × Cp × (T_in − T_out)
Here ṁ is the mass flow rate of the fluid, Cp is its specific heat capacity (how much energy it takes to raise one unit of mass by one degree), and the temperature difference is between the inlet and outlet of that stream. You only need to do this calculation for one side. Conservation of energy means the heat gained by the cold fluid equals the heat lost by the hot fluid, so once you know Q from one stream, you can use it to find the unknown outlet temperature of the other stream.
Getting this number right depends on having accurate fluid properties. You’ll need the specific heat capacity, density, thermal conductivity, and viscosity of both fluids at their expected operating temperatures. These properties shift with temperature, so use values at the average temperature each fluid will experience inside the exchanger.
Step 2: Calculate the Log Mean Temperature Difference
The temperature difference driving heat transfer isn’t constant from one end of the exchanger to the other. At the inlet, the hot fluid might be 150°C and the cold fluid 20°C, giving a 130°C difference. By the outlet, that gap might shrink to 30°C. The LMTD gives you a single representative temperature difference that correctly accounts for this variation:
LMTD = (ΔT₁ − ΔT₂) / ln(ΔT₁ / ΔT₂)
ΔT₁ is the larger temperature difference between the two fluids at one end of the exchanger, and ΔT₂ is the smaller difference at the other end. The “ln” is the natural logarithm. If the two terminal temperature differences happen to be equal, the LMTD simply equals that value.
How you define ΔT₁ and ΔT₂ depends on the flow arrangement. In a counter-flow exchanger (where the two fluids move in opposite directions), you compare hot inlet with cold outlet at one end, and hot outlet with cold inlet at the other. In a parallel-flow exchanger (both fluids entering at the same end), you compare both inlets at one end and both outlets at the other. Counter-flow arrangements produce a higher LMTD for the same set of temperatures, which means less surface area and a smaller exchanger.
The Correction Factor for Multi-Pass Designs
The LMTD formula above assumes pure counter-flow or pure parallel-flow. Real heat exchangers often use shell-and-tube designs with multiple tube passes, crossflow configurations, or other geometries that don’t fit neatly into either category. For these, you multiply the LMTD by a correction factor F:
Effective MTD = F × LMTD
F is always 1.0 or less. It equals 1.0 for true counter-flow, for pure co-current flow, and for situations involving boiling or condensing of pure substances (where one side stays at a constant temperature). For everything else, you look up F on charts specific to your exchanger geometry, using two dimensionless ratios calculated from your four terminal temperatures. If F drops below about 0.75, the design is thermally inefficient and you should consider a different configuration.
Step 3: Estimate the Overall Heat Transfer Coefficient
The overall coefficient U is often the hardest variable to pin down because it depends on the fluid properties, flow velocities, tube geometry, and fouling conditions on both sides. For initial sizing, engineers use published ranges based on the fluid combination:
- Water to water: 800 to 1,500 W/m²·°C
- Steam to water: 1,500 to 4,000 W/m²·°C
- Light oils to water: 350 to 700 W/m²·°C
- Heavy oils to water: 60 to 300 W/m²·°C
These ranges are wide because U depends heavily on flow velocity and turbulence. A water-to-water exchanger with sluggish flow might sit near 800, while one with high velocities and clean surfaces could reach 1,500. For a first pass at sizing, pick a value in the middle of the range, calculate your area, then refine once you’ve selected a specific geometry and can compute U more precisely.
The precise calculation of U combines three resistances in series: the convective heat transfer on the hot side, conduction through the tube wall, and convective heat transfer on the cold side. Each convective coefficient depends on the Reynolds number (which tells you whether flow is laminar or turbulent) and the Prandtl number (which captures how the fluid’s thermal properties compare to its flow behavior). Turbulent flow dramatically improves heat transfer, so exchanger designs aim for Reynolds numbers above about 10,000 on both sides when possible.
Step 4: Account for Fouling
Over time, deposits build up on heat transfer surfaces. Scale, corrosion products, biological growth, and sediment all add thermal resistance and reduce performance. If you size your exchanger based only on clean conditions, it will underperform within months of operation.
Fouling is handled by adding fouling resistances to the overall coefficient. The industry standard is to use values published by TEMA (Tubular Exchanger Manufacturers Association), which provides recommended fouling factors for different fluids and services. These resistances increase the effective thermal resistance, which lowers U and increases the surface area you need. Typical fouling allowances add 10% to 30% to the required area depending on how dirty the fluids are. Clean treated water needs a modest allowance; heavy fuel oil or untreated cooling water needs significantly more.
Step 5: Solve for Area and Select a Configuration
With Q, U (including fouling), and the effective LMTD (with correction factor if needed) in hand, solve A = Q / (U × LMTD). This gives you the total heat transfer surface area in square meters or square feet.
That area translates into a physical exchanger through the geometry you choose. For a shell-and-tube exchanger, the area determines how many tubes of a given diameter and length you need. For a plate exchanger, it determines the number and size of plates. A compact plate exchanger packs far more surface area into a small volume than a shell-and-tube unit, which is why plate exchangers dominate when both fluids are liquids and pressures are moderate.
After selecting a geometry, go back and verify your assumptions. Recalculate U using the actual tube diameters, flow velocities, and baffle spacing. Check that the pressure drop on both sides is acceptable. If U changes significantly from your initial estimate, recalculate A and iterate until the design converges.
When Outlet Temperatures Are Unknown
The LMTD method works cleanly when you know all four terminal temperatures (hot in, hot out, cold in, cold out). In many real problems, you only know the inlet temperatures and want to find what outlet temperatures a given exchanger will produce. You can’t calculate the LMTD without knowing the outlets, so the method requires iteration.
The alternative is the Effectiveness-NTU method. Effectiveness (ε) is defined as the actual heat transferred divided by the maximum heat that could possibly be transferred. That maximum occurs if one fluid changes temperature by the full difference between the two inlet temperatures. The formula is:
Q = ε × C_min × (T_hot,in − T_cold,in)
C_min is the heat capacity rate (mass flow rate times specific heat) of whichever fluid stream has the lower value. The effectiveness itself is a function of NTU (Number of Transfer Units, equal to UA/C_min) and the ratio of C_min to C_max, with specific relationships published for each exchanger configuration. This method lets you solve directly for heat duty or outlet temperatures without iteration, making it the better choice for rating an existing exchanger or for problems where only inlet conditions are known.
Choosing Counter-Flow vs. Parallel-Flow
Counter-flow arrangements, where the hot and cold fluids move in opposite directions, are almost always preferred. They produce a higher effective temperature difference for the same inlet and outlet conditions, which means less surface area. They also allow the cold fluid outlet temperature to exceed the hot fluid outlet temperature, something that’s physically impossible in parallel flow. This makes counter-flow exchangers capable of higher thermal recovery.
Parallel flow has niche uses: when you need to limit the maximum temperature the cold fluid sees (both fluids approach the same intermediate temperature at the outlet), or when rapid initial heat transfer at the inlet end is beneficial. But for general sizing, assume counter-flow unless you have a specific reason not to.
Common Sizing Mistakes
Using fluid properties at the wrong temperature is one of the most frequent errors. Viscosity in particular changes dramatically with temperature, and it directly affects the convective coefficients. Always evaluate properties at the mean bulk temperature of each stream.
Ignoring the correction factor F for multi-pass exchangers leads to undersized designs. A two-pass shell-and-tube exchanger might have F values of 0.8 to 0.9, which means 10% to 20% more area than a pure counter-flow calculation would suggest.
Neglecting fouling is the classic mistake that looks fine on paper and fails in the field. Even a modest fouling layer adds measurable thermal resistance. Build it into your U-value from the start rather than treating it as an afterthought.