Hydronic Heat Rate: Match Water Density to the Flow-Meter Location

The familiar water-side calculation Q = 4.18 × flow in L/s × temperature difference gives a useful quick estimate. Its simplicity hides two property assumptions: water density is treated as 1,000 kg/m³ and specific heat as 4.18 kJ/(kg·K). At a different temperature, the errors from those assumptions need not cancel.
A more careful calculation also asks where the volume flow was measured. Density converts that local volume flow into mass flow. The heat carried per kilogram comes from the temperature interval across the equipment. Using a single “average water property” for both jobs can put the calculation on the wrong basis.
Begin with mass flow and enthalpy difference
For a steady single-phase stream exchanging heat, neglecting shaft work and changes in kinetic and potential energy across the chosen equipment boundary, the thermal-power magnitude is Q = ṁ × |hout − hin|. Use mass flow ṁ in kg/s and specific enthalpy h in kJ/kg to obtain kW. For heating-water delivery, the hotter inlet has the larger enthalpy; for cooling-water heat pickup, the warmer outlet does.
When pressure effects are negligible for the required accuracy, write the enthalpy difference as c̄p × |Tout − Tin|. Here c̄p is the interval-average specific heat, equal to Δh/ΔT at the common pressure used for the property calculation. A value at the mean temperature can approximate that average, but it is not its definition.
For an actual-volume flow reading V̇, mass flow is ρmeter × V̇. Use the density at the flow sensor’s local fluid temperature and pressure. With V̇ in L/s, Q = (ρmeter/1,000) × V̇ × Δh. If the instrument already reports mass flow or compensated reference-volume flow, first identify that output basis rather than applying another conversion.
Work one heating-water example from published properties
Assume clean liquid water enters a heat exchanger at 80°C and leaves at 60°C. A volume meter on the 80°C supply reads 5.000 L/s. For this property comparison, evaluate both states at 0.50 MPa absolute and neglect the pressure difference across the exchanger. Assume steady operation, no bypass mixing, no leakage and no heat storage within the boundary.
NIST’s compressed-water table, Table 3 on printed page 32, gives the following rounded values at 0.50 MPa: at 80°C, density 971.97 kg/m³ and enthalpy 335.37 kJ/kg; at 60°C, density 983.37 kg/m³ and enthalpy 251.58 kJ/kg. These are water properties, not mixture values or equipment ratings.
The enthalpy drop is 335.37 − 251.58 = 83.79 kJ/kg. Over 20 K, the implied average specific heat is 83.79/20 = 4.1895 kJ/(kg·K). Supply mass flow is 971.97 × 5.000/1,000 = 4.85985 kg/s. The heat-transfer rate calculated from these rounded properties is 4.85985 × 83.79 = 407.21 kW.
The fixed-property shortcut gives 4.18 × 5.000 × 20 = 418.00 kW. Relative to the property-based result, that is 10.79 kW or approximately 2.65% higher. This is an error comparison for the stated example, not a universal correction factor. The measured temperatures, pressure, fluid composition and meter basis determine the appropriate calculation.
Why correcting specific heat alone can move the wrong way
Replacing 4.18 by the interval-average 4.1895 while retaining 1,000 kg/m³ gives 418.95 kW. The result moves farther from 407.21 kW because this case’s reduced density has the larger effect. A temperature-dependent specific heat does not compensate for using an incompatible density.
Likewise, the table’s 70°C density is 977.94 kg/m³. Multiplying that mean-temperature density by the measured 80°C volume flow gives 409.71 kW, about 0.61% high. A midpoint property may be convenient, but density’s reference is the volume measurement. The averaged specific heat and the meter-location density answer different questions.
ONICON’s discussion of thermal-energy calculation identifies both specific-heat and density corrections among the contributors that a measurement system must address. Correcting only one does not establish the performance of the complete measurement chain.
The return volume flow is different, while mass flow is the same
For the same steady water stream, return volume flow is 4.85985/983.37 × 1,000 = 4.94204 L/s at 60°C. Using that return volume with the return density recovers the same mass flow and the same 407.21 kW. Supply and return actual-volume readings therefore need not be numerically identical even without a leak.

Original steady-state example using rounded NIST water properties at a common 0.50 MPa absolute reference. Pressure-difference effects, leakage, bypass mixing and heat storage are neglected.
If 5.000 L/s were instead a measurement at the 60°C return, it would describe a slightly different mass flow. Using return density with the original supply reading would overstate mass flow, and therefore heat rate, by about 1.17%. Keep the location attached to the reading when values are copied between a schedule, field sheet and control-system calculation.
This configuration detail appears in Belimo’s 22PE-1UHH thermal-energy-meter datasheet: supply or return installation is selected during commissioning. Follow the selected meter’s actual instructions and fluid settings; do not assume all devices infer the location automatically.
A chilled-water comparison has a smaller difference here
Now assume 5.000 L/s measured at 5°C, with the water leaving the equipment at 15°C. Use the same 0.50 MPa property reference and simplified boundary. NIST lists density 1,000.16 kg/m³ at 5°C, with enthalpies of 21.52 and 63.46 kJ/kg at 5°C and 15°C respectively.
Mass flow is 5.0008 kg/s and enthalpy rise is 41.94 kJ/kg, giving 209.73 kW. The shortcut gives 209.00 kW, approximately 0.35% lower. The shortcut performs much more closely in this example than in the 80/60°C case. That does not make 0.35% a meter-accuracy claim: sensor, flow and installation uncertainties remain separate.
Choose the level of detail the decision needs
Record actual-volume, reference-volume or mass-flow units and the meter’s supply/return location.
Identify the fluid and concentration basis; use supplier properties for glycol mixtures.
Document temperatures, pressure basis, property source and any interpolation.
Compare the fixed-property approximation with the property-based result before judging its significance.
Keep raw measurement uncertainty and property-model approximation separate in the review.
For a detailed energy balance with meaningful pressure differences, evaluate enthalpy at the actual inlet and outlet states and retain any other energy terms that matter. For contractual metering, use the approved meter configuration and applicable measurement requirements. Extra decimal places from property tables cannot correct a poorly measured temperature difference.
To connect load, flow, piping and commissioning checks in one reference, explore the paid Chilled Water System Design Handbook. Its published contents include worked calculations, a complete design example and commissioning coverage. Keep the fluid-property basis alongside the design inputs so later performance comparisons use the same definitions.



Comments