Plate Heat Exchanger Sizing: Flow, Approach Temperature and Pressure Drop
A 60 kW label does not define a heat exchanger. Build a selection brief that checks both water circuits, the temperature gap between them and the pressure each pump can afford.
Published by MEPFlow · Methods and limitations · Primary references
Use the method while you read
Follow the worked examples, check your inputs and use the related MEPFlow tool to explore your own design.
In this guide: jump to a step
A plate heat exchanger transfers heat through a wall between two fluid circuits. Selecting one starts with a duty and operating conditions, not a pipe connection size or a catalogue capacity taken out of context. Two selections can both transfer 60 kW yet need different plate arrangements because the available temperatures and pressure allowances differ.
This guide follows one original water-to-water heating example from the temperature schedule to a supplier review. It covers steady, single-phase counterflow. It does not select a model, predict seasonal performance or cover refrigerant evaporation and condensation. Every numerical case below is hypothetical; the temperatures are design inputs, not recommended heat-pump setpoints.
For the complete layout, explore MEPFlow’s hydronic design and pipe sizing software and review equipment duty, water flow, pipe sizes and circuit assumptions together.
1. Write the four temperatures before choosing a model
Name the hot-side inlet and outlet separately from the cold-side inlet and outlet. In a heating application, the cold-side outlet usually supplies the heating distribution. Calling both outlets “return” makes a calculation easy to misread, particularly when the exchanger separates a plant loop from a building loop.
For this example, the building needs 60 kW. The hot stream enters at 55°C and leaves at 45°C. The cold stream enters at 35°C and leaves at 42°C. Keep those four numbers together on the schedule, with arrows showing direction. Later changes to any one of them require another balance check.
First decide why a separating wall is needed. Different fluid compositions or independently pressurized circuits may justify fluid separation. A low-loss header or closely spaced tees connect shared water and answer a different question; see the hydraulic separation guide. A heat exchanger also introduces a thermal temperature difference and resistance in each circuit.
| Quantity | Hot side | Cold side |
|---|---|---|
| Duty transferred | 60 kW released | 60 kW absorbed |
| Inlet temperature | 55°C | 35°C |
| Outlet temperature | 45°C | 42°C |
| Fluid temperature change | 10 K decrease | 7 K increase |
| Fluid assumptions | Water; density 1,000 kg/m³ | Water; density 1,000 kg/m³ |
| Specific heat assumption | 4.18 kJ/(kg·K) | 4.18 kJ/(kg·K) |
Scope of the example
Steady operation, negligible heat loss to the room, no phase change and constant approximate water properties. Real selections need fluid properties at their operating temperatures. This example is not a potable-water, pressure-vessel or local-code approval.
2. Balance the heat, then calculate each flow
Use the temperature change along one fluid stream to calculate its flow. Do not use the temperature gap between the two streams in this equation. With heat in kW, specific heat in kJ/(kg·K) and temperature change in K, the result is mass flow in kg/s. Divide by density to obtain m³/s, then multiply by 1,000 for L/s.
The hot-side mass flow is 60 ÷ (4.18 × 10) = 1.4354 kg/s. At the assumed density that is 1.4354 L/s, or 5.1675 m³/h. The cold side needs 60 ÷ (4.18 × 7) = 2.0506 kg/s: 2.0506 L/s, or 7.3821 m³/h. The flows differ because the prescribed temperature changes differ.
Check backwards before sending the schedule: 4.18 × 1.4354 × 10 is approximately 60 kW; 4.18 × 2.0506 × 7 is also approximately 60 kW. Keep unrounded values in calculations and round only the displayed schedule. Equal heat rates do not require equal volumetric flows.
This balance checks the requested operating point. It does not prove that a proposed exchanger can achieve it. The water duty and flow calculator supports the water-side arithmetic; use the hydronic flow guide for the next pipe-sizing step. If using GPM, identify US gallons rather than silently substituting Imperial gallons.
For a glycol circuit, replace both density and specific heat with values for the actual product, concentration and temperature. Record whether concentration is by mass or volume. Do not copy the water flow to the glycol side or reuse a water-only pressure loss; the glycol guide explains that separate check.
Q = ṁ × cp × ΔT
Q is heat rate in kW; ṁ is mass flow in kg/s; cp is specific heat in kJ/(kg·K); ΔT is the inlet-to-outlet temperature change of that same stream in K. A temperature difference of 1°C equals 1 K.
Take the next step on your own floor plan
Create a MEPFlow account to work through loads, ductwork, domestic water or hydronic layouts in your browser.
Start free trial3. Check the approach at both ends
An approach temperature is a gap between the two fluids at an end of the exchanger. State which end you mean. In counterflow, the hot inlet sits opposite the cold outlet, while the hot outlet sits opposite the cold inlet. The two terminal differences in our example are 55 − 42 = 13 K and 45 − 35 = 10 K. The smaller terminal difference is 10 K.
The 7 K cold-side rise is not the approach, and neither is the 10 K hot-side drop. The repeated number 10 is coincidental here. Label every subtraction rather than relying on a bare “delta T” field. That distinction matters when a supplier asks whether a quoted difference is the load-side rise or the thermal gap across the plates.
A counterflow cold outlet can be warmer than the hot outlet without violating heat transfer direction: those outlets are at opposite ends. Compare temperatures that actually face each other. For the ideal passive, single-phase case here, both terminal differences must remain positive for a finite conventional selection carrying positive duty. A zero or negative terminal gap is a reason to stop and revisit the requested temperatures, not take the absolute value and continue.
The diagram is an original temperature map. It shows endpoints and opposing directions, not manufacturer port assignments or an installation arrangement. Use the supplied connection drawing when piping a selected unit.
Original counterflow temperature map for the 60 kW example. The streams remain separate. Scroll horizontally on a small screen; the endpoint pairs are also written below.
Left: hot inlet 55°C minus cold outlet 42°C = 13 K. Right: hot outlet 45°C minus cold inlet 35°C = 10 K. Endpoint map only; no internal temperature profile, port assignment or installation detail is implied.
4. Use LMTD to test how demanding the brief is
The driving temperature difference varies through the exchanger. For this ideal counterflow check, combine the two positive terminal differences using the log mean temperature difference, or LMTD. The natural logarithm is required. In the base case, (13 − 10) ÷ ln(13 ÷ 10) = 11.4345 K.
Dividing the 60 kW duty by that driving difference gives a required overall conductance UA of 5.2473 kW/K. Here U represents overall heat-transfer coefficient and A represents effective area. This is a duty requirement, not a plate count. A guessed U value cannot account for a real channel geometry, fluid viscosity, flow distribution, fouling allowance and manufacturing arrangement.
When both terminal differences are equal, the LMTD equals either difference. Do not evaluate a zero divided by zero expression. For a 50/40°C hot stream against the same 35/42°C cold stream, the terminal gaps are 8 K and 5 K; LMTD is 6.3829 K and required UA is 9.4001 kW/K.
The table holds 60 kW and both stream temperature changes fixed while lowering the prescribed hot-side temperatures. At 47/37°C, the gaps narrow to 5 K and 2 K: required UA rises to 18.3258 kW/K. That is 3.49 times the base requirement. It is not a claim that buying 3.49 times as many plates will work; U and pressure losses change with the actual selection.
These are separate design briefs. They do not predict the output of one fixed exchanger as a heat pump changes temperature. To answer that question, ask the manufacturer to rate the selected model at the new inlet temperatures and flows. Avoid attaching an efficiency saving to these temperatures without a separate heat-source and system analysis.
LMTD = (ΔT₁ − ΔT₂) ÷ ln(ΔT₁ / ΔT₂)
For counterflow: ΔT₁ = hot inlet − cold outlet; ΔT₂ = hot outlet − cold inlet. Both are positive here. Required UA = Q ÷ LMTD. Equal terminal differences use their common value.
| Hot inlet → outlet | End gaps | LMTD | Required UA |
|---|---|---|---|
| 55 → 45°C | 13 K / 10 K | 11.4345 K | 5.2473 kW/K |
| 50 → 40°C | 8 K / 5 K | 6.3829 K | 9.4001 kW/K |
| 47 → 37°C | 5 K / 2 K | 3.2741 K | 18.3258 kW/K |
A hand check is not a manufacturer selection
The equations describe the stated ideal counterflow duty. Other flow arrangements may require a correction or a different analysis. The supplier must verify the selected model, passes, usable area, thermal performance and hydraulic limits.
5. Give each side its own pressure-drop allowance
Pressure drop is resistance while fluid flows. Pressure rating is the allowable containment condition of the selected equipment. A 25 kPa flow-loss allowance says nothing about whether a unit can contain the system’s static or maximum operating pressure. Record these as separate fields.
Here is an original pressure-budget illustration, independent of any exchanger catalogue. Suppose the hot-side critical circuit has 70 kPa available at design flow and its other components use 45 kPa. The remaining allowance is 25 kPa. Suppose the cold-side circuit has 85 kPa available and other losses total 55 kPa. Its allowance is 30 kPa. A proposed selection must be checked against each limit at its own flow and fluid condition.
Do not add 25 and 30 kPa and assign 55 kPa to either pump. The separating wall keeps the fluid paths distinct. Include only the losses traversed by the circuit being evaluated, as described in the critical-circuit pump head guide. Confirm that strainers, control valves, connectors and intended control margins are actually included in the “other losses” allowance.
If a supplier returns a thermally adequate model with excessive pressure loss, revise the selection or the reviewed circuit budget. Raising pump head is a system design change, not an automatic correction. Also ask what the quoted loss includes: plate passages only, or ports and connections too. Keep the answer with the selection sheet so it is not counted twice or omitted.
Exchanger allowance = available differential − other circuit losses
Use a consistent design operating point and one fluid circuit at a time. These example numbers are assumed budgets; actual exchanger pressure drops require manufacturer data.
| Circuit | Available differential | Other circuit losses | Exchanger allowance |
|---|---|---|---|
| Hot water: 1.4354 L/s | 70 kPa | 45 kPa | 25 kPa |
| Cold water: 2.0506 L/s | 85 kPa | 55 kPa | 30 kPa |
6. Turn the calculation into a reviewable selection request
A useful request lets the supplier distinguish fixed requirements from quantities they may solve. Mark the 60 kW duty and required building supply temperature explicitly. If the source temperature is still under study, submit separate named cases rather than a range that could be interpreted in several ways.
Alfa Laval’s calculation guidance identifies temperatures, flows and pressure limits as selection inputs. SWEP’s single-phase documentation also distinguishes design from performance and rating calculations. Use that distinction in the request: “select equipment for these conditions” and “check this existing model at changed conditions” are different tasks.
Download the worked example CSV to reproduce the three thermal cases. The selection review CSV is a blank handoff record for the project team. Both contain fixed values or blank fields, without calculation formulas or macros. Fill in project-specific limits rather than treating the illustration as a default specification.
MEPFlow’s hydronic design workflow can keep the surrounding layout, duty and circuit assumptions together.
1.State the operating cases
Provide duty, four temperatures, flows, fluid names and concentration basis. Identify fixed inputs, calculated values and any unresolved assumption.
2.Set hydraulic and containment limits
Give each side’s allowable flow loss, maximum operating pressure and temperature. Identify installation jurisdiction and required approvals for supplier review.
3.Ask for the selection evidence
Request model and plate arrangement, calculated temperatures and losses, thermal margin basis, connection drawing and applicable materials. Do not silently add a second fouling margin to an already derated selection.
4.Review operation and maintenance
Resolve low-load and start-up cases, freeze exposure, water quality, filtration, cleanability, isolation and service access with the manufacturer and project designer.
7. Commission against the same boundary conditions
A commissioning record should allow someone else to repeat the heat balance. Measure both flows and all four fluid temperatures during a stable operating period. Record fluid composition, sensor locations and instrument accuracy alongside the values. Measure differential pressure across the same boundaries used in the selection report.
Calculate the heat released by the hot stream and absorbed by the cold stream separately. If they disagree, investigate measurement uncertainty, changing conditions, heat loss and fluid-property assumptions before declaring the exchanger undersized. A single supply-temperature reading cannot establish transferred duty.
For example, a measured 7 K cold-side rise has an approximately ±0.4 K worst-case difference uncertainty if each temperature reading is independently specified only within ±0.2 K. That alone is about ±5.7% of the temperature difference before flow and property errors are included. This simple bound is not a statistical confidence interval. Use the actual instruments and the project acceptance method.
Keep the first accepted readings as a baseline. A later change in pressure loss or thermal performance warrants comparison at matched flows and inlet conditions. Alfa Laval’s service guidance connects performance checks with actual operating conditions, filtration and fouling review. A worse approach during a transient does not by itself identify fouling.
Close the review with the approved selection sheet, operating cases, setpoints, instrument records and outstanding limits. Preserve the distinction between a checked energy balance, a manufacturer-rated exchanger and an accepted installed system. These are three different pieces of evidence.
Frequently asked questions
Is approach temperature the same as water delta T?
No. A stream’s delta T is its inlet-to-outlet change. Approach compares the two fluids at an exchanger end. State the terminal pair; the example has a 10 K hot-side drop, a 7 K cold-side rise and terminal gaps of 13 K and 10 K.
Can the cold outlet be warmer than the hot outlet?
Yes, in counterflow those outlets are at opposite ends. Check hot inlet against cold outlet and hot outlet against cold inlet. Positive local gaps are what matters for the ideal passive single-phase case.
Can I size a plate heat exchanger from kW alone?
No. The same duty at a tighter temperature approach needs more overall conductance. Supply the temperatures, fluids, flows and separate pressure limits, then obtain a manufacturer selection.
Do the two sides need the same water flow?
No. They must balance heat under the stated steady, negligible-loss assumptions. Different stream temperature changes or fluid properties lead to different required flows.
Does the UA calculation give a plate count?
No. It gives required overall conductance at the stated ideal operating point. Real area, channel geometry, heat-transfer coefficient and pressure loss must be resolved together in the equipment selection.
Can the worksheet be used for refrigerant or domestic hot water?
It is a single-phase hydronic teaching worksheet. It does not cover refrigerant phase change or establish potable-water safety, separation or code compliance. Those applications require an appropriate selection method and project-specific review.
Continue the engineering workflow
Put this guide into practice with MEPFlow
Move from the calculation method into a connected browser-based design workflow while keeping the assumptions visible for engineering review.
Start free trialPrimary references
Sources and further reading for the methods discussed in this guide.