Glycol in Chilled-Water Systems: Correcting Flow, Pressure Drop and Chiller Capacity
Glycol changes the design basis
Adding glycol to a chilled-water circuit changes more than its freezing point. A water-only selection no longer establishes the required flow, circuit resistance or available refrigeration capacity. The practical task is to reconcile those three results using the same fluid, concentration and temperatures.
This matters when an outdoor chiller is added to an existing building, an exposed branch needs protection, or a process requires colder fluid. A note reading “30% glycol” is incomplete: it must identify glycol type, commercial formulation and whether concentration is by mass or volume. The equipment schedules, fluid specification and commissioning records should agree.
Establish what must remain protected
Begin with the lowest credible fluid temperature during operation, shutdown and restart. Distinguish freeze protection, which addresses ice formation, from burst protection, which can permit a partly frozen mixture that cannot circulate normally. A burst-protected outdoor loop may still be unsuitable for an immediate cold start. Caleffi's discussion of antifreeze protection explains this distinction and why excess concentration carries a hydraulic penalty.
Ask the fluid supplier and equipment manufacturer to agree on the concentration and required margin. Check their minimum concentration for inhibitor effectiveness as well as the maximum allowed by the chiller. Do not select a percentage from an unrelated product's freeze chart. A premixed product, pure glycol and an inhibited concentrate are different purchasing descriptions.
Where only a small outdoor circuit needs protection, compare a separated glycol loop against filling the entire installation. Include the intervening heat exchanger's temperature approach, additional pumping and maintenance in that comparison. The smaller protected volume alone does not settle the decision.
Build one fluid-property record
For the selected formulation, obtain density ρ, specific heat cp, dynamic viscosity μ and thermal conductivity at the relevant temperatures. Use mean operating temperature for the preliminary heat balance, then inspect the colder supply and cold-start conditions separately. For large temperature spans, use the supplier's enthalpy data or an appropriate temperature-dependent calculation.
Dow's FLUIDFILE resources provide a route to product-specific thermal and hydraulic properties. Save the input concentration, temperature and units with the calculation, rather than copying an unexplained correction factor into a spreadsheet.
Density and specific heat determine how much cooling a litre can carry. Viscosity affects resistance to flow and the operating point. Thermal conductivity and fluid-side convection affect heat-exchanger performance. These are related effects, but they require different calculations. Caleffi's hydronic fluid-property guidance confirms that glycol mixtures are more viscous than water and that viscosity changes with temperature.
Worked example 1: correct flow, then resistance
Assume a 500 kW load and a 5 K fluid temperature rise. For illustration only, take a mixture at its mean design temperature to have ρ = 1,040 kg/m³ and cp = 3.80 kJ/(kg·K). These assumed properties are not a concentration recommendation or a product rating. Neglect pipe heat gain.
The heat balance is:
Q = ρ × V̇ × cp × ΔT
With Q in kW, V̇ in m³/s and cp in kJ/(kg·K):
V̇ = 500 ÷ (1,040 × 3.80 × 5) = 0.0253 m³/s = 25.3 L/s
Using the simplified water properties 1,000 kg/m³ and 4.18 kJ/(kg·K), the original water duty would require:
V̇water = 500 ÷ (1,000 × 4.18 × 5) = 23.9 L/s
The mixture therefore needs approximately 5.8% more volume flow for this duty. Keeping the original 23.9 L/s would transport about 473 kW at the same 5 K rise. That is a fluid heat-balance result; it says nothing yet about how much cooling the chiller can produce.
Now examine one unchanged pipe section. For Darcy-Weisbach friction, Δp is proportional to fρv² when length and diameter stay fixed. Suppose a separate Reynolds-number and roughness calculation gives an illustrative Darcy friction-factor ratio fmixture/fwater = 1.25 at the revised operating point. Then:
Δpmixture/Δpwater = 1.25 × 1.04 × (25.3/23.9)² ≈ 1.46
A water-only friction loss of 40 kPa becomes approximately 58 kPa for that section under these assumptions. The 1.25 ratio is an assumed input for demonstrating the calculation, not a general glycol multiplier. Obtain the actual friction factor from viscosity, velocity, diameter and roughness at the intended operating condition.
For that section alone, hydraulic power is Δp × V̇. It rises from approximately 0.96 kW to 1.47 kW. Electrical input requires pump and drive efficiency as well. This shows why a modest flow increase can accompany a much larger resistance increase. It does not justify multiplying the whole system head by 1.46: valves, coils and evaporators need their own applicable data.
Illustrative fluid properties show why a glycol solution needs its own thermal-flow and hydraulic calculations. These values do not specify a glycol concentration.
Obtain a complete glycol-rated chiller selection
The fluid heat balance determines required flow. Chiller selection determines whether the refrigeration machine can deliver the duty at that flow and leaving-fluid temperature. Request a selection that explicitly states fluid type, concentration basis, entering and leaving temperatures, ambient or condenser conditions, cooling capacity, input power, evaporator pressure drop and permitted flow range.
Manufacturer correction procedures are equipment-specific. The Daikin Applied chiller installation manual illustrates separate adjustments for capacity, flow, pressure drop and power. Its factors are not transferable to another evaporator or model. Also check whether a quoted pressure-drop correction uses the original water flow or the corrected mixture flow; applying a second flow correction to an already corrected result can double-count the effect.
In a design review, put the final glycol-rated selection beside the original water-rated selection. Trace each changed number to a documented input. A one-line statement that the unit is “suitable for glycol” is insufficient evidence of capacity at the project's conditions.
Worked example 2: reconcile the available capacity
Consider a different project with a 950 kW required duty. Assume the manufacturer's project-specific selection gives a capacity factor of 0.96 for the chosen fluid and conditions, applied to a water-rated capacity of 1,000 kW. This factor is hypothetical and must not be used for procurement.
Available mixture capacity = 1,000 × 0.96 = 960 kW
The remaining margin is 10 kW, or about 1.1% of the required duty. If the same documented factor applied over the small selection range, the minimum equivalent water rating would be:
Required water rating = 950 ÷ 0.96 = 989.6 kW
That calculation is a screening check. Obtain the final software selection because the factor may change with model, temperature or flow. Calculate the operating mixture flow from the actual 950 kW duty and design temperature difference, then confirm the evaporator's permitted flow limits. Do not assume that increasing flow recovers the refrigeration capacity removed by the glycol correction.
Make the site condition match the calculation
Before acceptance, record the supplied formulation and test the circulating mixture after it is thoroughly mixed. Use a test method and concentration scale appropriate to the product. A concentration reading and an inhibitor-condition assessment answer different questions; agree on both with the fluid supplier.
The handover should retain the approved property sheet, equipment selections and measured concentration together. Record fluid temperature during flow and pressure-drop measurements so the readings can be compared with the correct selection condition. Require revised calculations when the concentration is changed, including after repeated water top-ups or an incorrect refill.
A useful final review asks four questions: Is the protection adequate? Can the loop circulate at its coldest required condition? Can the selected equipment deliver the duty? Does the commissioned fluid match the calculation? Answering all four prevents a frost-protection decision from quietly becoming a capacity problem.
For the surrounding plant design workflow, the Chilled Water System Design Handbook is a related resource. Keep the glycol supplier's property data and the equipment manufacturer's final selection alongside it as project-specific evidence.
Comments