There are two ways to analyze a heat exchanger, and the choice between them comes down to one question: do you already know the outlet temperatures? The familiar log-mean-temperature-difference method assumes you do. But in the most common design situation — you have the inlet conditions, the flow rates, and a candidate exchanger, and you want to know what it will deliver — the outlets are exactly what you are trying to find. Use LMTD there and you are stuck iterating. The effectiveness-NTU method was built for precisely this case.
This article explains the NTU method, works a counterflow example end to end, and flags the mistakes that quietly corrupt the result.
Why this calculation matters
A heat exchanger is rarely designed from a blank sheet. More often you are rating one: checking whether a particular unit, with a known surface area and overall heat transfer coefficient, can do the duty in front of it. The inlets and flow rates are fixed by the process. The outlets are unknown.
The LMTD method needs the outlet temperatures to compute the mean temperature difference, so applying it to a rating problem forces a guess-and-correct loop. The NTU method removes the loop. It works directly from inlet conditions and exchanger properties to give the heat duty and both outlet temperatures in a single pass. For rating work — and for any design where the outlets are not pinned down — it is simply the better tool.
The core method
The method rests on three quantities.
Heat capacity rates. For each stream, the heat capacity rate is the mass flow times the specific heat:
C = m_dot * cp [W/K]
Identify the smaller of the two as C_min and the larger as C_max. Their ratio is Cr = C_min / C_max.
NTU — the Number of Transfer Units — is a dimensionless measure of the exchanger's thermal size:
NTU = U A / C_min
A large NTU means a large surface area relative to the flow it must heat or cool — a thermally "big" exchanger.
Effectiveness is the ratio of the heat actually transferred to the maximum that is thermodynamically possible:
effectiveness = Q / Q_max, Q_max = C_min * (T_hot_in - T_cold_in)
Effectiveness depends only on NTU, Cr, and the flow arrangement. For a counterflow exchanger:
e = (1 - exp(-NTU (1 - Cr))) / (1 - Cr * exp(-NTU (1 - Cr)))
Once you have the effectiveness, the heat duty is Q = effectiveness × Q_max, and the outlet temperatures follow directly from energy balances on each stream.
A worked example
A counterflow heat exchanger cools hot water with cold water.
- Hot stream: flow 1.0 kg/s, cp = 4180 J/kg·K, inlet 80 °C
- Cold stream: flow 2.0 kg/s, cp = 4180 J/kg·K, inlet 20 °C
- Overall conductance: UA = 8000 W/K
Step 1 — heat capacity rates.
C_hot = 1.0 x 4180 = 4180 W/K
C_cold = 2.0 x 4180 = 8360 W/K
C_min = 4180 (hot side), C_max = 8360, Cr = 0.50
Step 2 — NTU.
NTU = UA / C_min = 8000 / 4180 = 1.91
Step 3 — effectiveness (counterflow, Cr = 0.50):
exponent = -1.91 x (1 - 0.50) = -0.957
e = (1 - e^-0.957) / (1 - 0.50 x e^-0.957)
e = (1 - 0.384) / (1 - 0.192) = 0.616 / 0.808 = 0.762
Step 4 — heat duty.
Q_max = C_min x (80 - 20) = 4180 x 60 = 250,800 W
Q = 0.762 x 250,800 = 191,200 W = 191 kW
Step 5 — outlet temperatures, from an energy balance on each stream:
T_hot_out = 80 - Q / C_hot = 80 - 191200/4180 = 34.3 C
T_cold_out = 20 + Q / C_cold = 20 + 191200/8360 = 42.9 C
The exchanger transfers 191 kW and delivers the hot stream out at 34.3 °C and the cold stream out at 42.9 °C — all from inlet data alone, with no iteration. Note that the cold outlet (42.9 °C) ends up warmer than the hot outlet (34.3 °C). That temperature cross is perfectly normal for a counterflow exchanger and is one reason counterflow extracts more heat than the parallel-flow arrangement.
Common mistakes
Picking the wrong stream as C_min. NTU and Q_max are both defined on C_min. The smaller heat capacity rate is the one that limits the exchange — choose the wrong stream and every downstream number is off.
Using the wrong effectiveness relation. Counterflow, parallel flow, crossflow, and shell-and-tube each have their own effectiveness-NTU equation. They can differ by tens of percent at the same NTU. Match the formula to the real flow arrangement.
Assuming constant properties. Specific heat, viscosity, and U all drift with temperature. For a large temperature swing, evaluate properties at a representative mean rather than at one inlet.
Forgetting fouling. UA in service is not UA when clean. Scale and deposits add thermal resistance over time, lowering U and effectiveness. A design that ignores a fouling allowance is optimistic from the first week of operation.
Try the interactive NovaSolver calculator
The effectiveness equations are compact but easy to slip up on, especially the exponentials. The heat exchanger effectiveness-NTU calculator on NovaSolver does that bookkeeping for the counter-, parallel-, and cross-flow arrangements: enter the two heat capacity rates C_h and C_c, the conductance UA, and the inlet temperature difference, and it returns NTU, the counter- and parallel-flow effectiveness, and the heat duty Q. From Q, the outlet temperatures follow directly from the energy balances shown above.
Related calculators
- Log-mean temperature difference — the companion method for when the outlet temperatures are already known.
- Shell-and-tube heat exchanger — for the industrial multi-pass geometry and its correction factor.
- Double-pipe heat exchanger — the simplest concentric-tube arrangement, ideal for learning the fundamentals.
The full set is in the heat exchanger tools hub.
Closing note
The effectiveness-NTU method is the right tool whenever the outlet temperatures are unknown — which describes most real heat exchanger work. The mental model is clean: NTU measures how thermally large the exchanger is, effectiveness measures how close it gets to the thermodynamic ideal, and the two are linked by one equation per flow arrangement. Get C_min right, match the effectiveness relation to the geometry, and the duty and outlet temperatures fall out in a single calculation.
Top comments (0)