Cooling Tower Calculator

A wet cooling tower cools water mainly by evaporation, not by simple sensible heat transfer to cooler air. Frederick Merkel's 1925 theory captures this by combining heat and mass transfer into a single driving force: the difference between the enthalpy of air saturated at the local water temperature and the actual enthalpy of the air passing through. That leads to the defining integral of cooling tower design:

KaV/L = ∫ cp,w dT / (hw,sat − hair), integrated from cold water temperature to hot water temperature

KaV/L (the "tower characteristic" or number of transfer units) is commonly treated as a dimensionless tower characteristic: K is a mass transfer coefficient, a the packing's surface area per unit volume, V the fill volume, and L the water mass flow. In practice the units bundled into K and a vary between sources, so this ratio is not always strictly dimensionless in every formulation - "dimensionless" here means the conventional way the fill characteristic is reported and compared, not a universal identity. This calculator solves the integral above numerically using the Chebyshev four-point method: a classic technique long associated with cooling tower thermal rating because it needs only four evaluations of the integrand to match full numerical integration to a fraction of a percent for this type of curve. CTI's ATC-105 acceptance test code governs how a tower's performance is measured and verified against guarantee, not which numerical integration algorithm is used to evaluate the Merkel integral - Chebyshev is a widely used, convenient choice for that step, not a method it prescribes.

Two curves, one crossing point. The integral above is the demand: how many transfer units this particular range and approach require. Real fill has a characteristic curve instead — the transfer units it can actually deliver, which manufacturers fit to test data as KaV/L = C·(L/G)-n, falling as the water-to-air mass flow ratio (L/G) rises. Cold water temperature is whatever value makes demand equal characteristic — exactly like a pump curve crossing a system curve. That is why dragging the air flow or fill height sliders changes the approach: you are moving the characteristic curve, and the crossing point moves with it.

The fill characteristic curve above describes a clean, as-new tower. Real towers do not stay that way. Over a service life of years, three related mechanisms erode the transfer coefficient this calculator bundles into KaV/L or, for a dry cooler, into UA:

• Scale. Dissolved minerals in the circulating water - the same ones concentrated by cycles of concentration - precipitate onto fill and heat exchanger surfaces as the water evaporates, building up an insulating layer and narrowing the flow passages between fill sheets.
• Biological growth. Algae, bacteria and biofilm thrive in the warm, wet, sunlit environment inside a wet tower, coating surfaces and, in the worst cases, partially blocking fill passages outright. Water treatment programmes exist mainly to keep this in check.
• Mechanical ageing and debris. Fill sheets sag, warp or become brittle with age and thermal cycling; airborne dust and debris settle on both wet fill and dry-cooler finned coils, and for air-cooled equipment specifically, insect nests and organic debris on the air side are a genuine, recurring maintenance issue.

The fill/coil condition slider applies a single multiplier to the clean-fill characteristic (or the dry cooler's UA) to represent all three together, without modelling them individually. Move it down from 100% and watch approach open up for the exact same fill height, air flow and weather — the same tower, doing measurably less work, is the direct, everyday consequence of deferred cleaning and water treatment. This is precisely why real cooling towers are inspected, dosed with biocides and scale inhibitors, and periodically cleaned: fouling is a genuine capacity loss, not just an aesthetic concern.

Natural draft towers use nothing but buoyancy: air warmed and humidified inside the tower is lighter than the ambient air outside, so it rises through the hyperbolic shell and pulls fresh air in at the base. No fans means no fan power and no moving parts to maintain, but the structure only pays for itself at large scale — which is why you see them almost exclusively at large power stations.

How the natural-draft airflow here is estimated. Unlike a fan-driven tower, where you set the air flow directly, natural-draft air flow is an output of the buoyancy balance. This calculator uses a simplified chimney/stack-draft relation with a single calibration constant tuned so a representative large tower gives a realistic figure — it is an educational estimate, not a predictive design model. A real natural-draft airflow calculation depends on shell and fill pressure losses, rain-zone and drift-eliminator resistance, inlet losses, the internal density/humidity profile, and site effects like wind and recirculation, none of which are resolved here. Treat the reported air flow as an order-of-magnitude estimate that shows the right trends with tower height, diameter and weather, not a number to design against.

Mechanical draft towers use fans to force the airflow, which is what lets an engineer set air flow independently of tower size. The naming refers to where the fan sits, which is separate from the air/water flow geometry: induced draft puts the fan at the top, pulling air up through the fill, while forced draft puts the fan at the base, pushing air through. Either arrangement can be built as counterflow (air rising against the falling water) or crossflow (air moving horizontally across it) — fan position and flow geometry are independent choices. Induced-draft counterflow is the most thermally effective common arrangement. Forced draft keeps the fan and motor in the cool, dry inlet air rather than the humid, corrosive exhaust, which is easier to maintain, though it is more prone to recirculation of its own moist exhaust back into the inlet.

Hybrid towers run a wet section and a dry section together, trading capital cost for reduced water use and, in some designs, a way to suppress the visible plume. The dry section's weakness is timing: it does least on the hottest days, exactly when the wet section is needed most.

One configuration among several. This simulator represents a simplified parallel water-side hybrid: the circulating water is split between a wet section and a dry section running side by side, and their cooled outlets are blended. That is a legitimate and common arrangement, but it is not the only one — real hybrid towers also use series water routing (water passing through one section then the other), parallel air paths, dedicated dry coils purely for plume abatement, and seasonal wet/dry bypass modes. The results here should be read as representative of the parallel wet/dry concept, not of every hybrid tower on the market.

Dry cooling in this calculator represents a closed-circuit water-to-air fluid cooler: hot circulating liquid rejects heat to ambient air across finned tubes, with no evaporation and no tower makeup-water demand. A direct steam-condensing air-cooled condenser (ACC) requires a different steam-side model and is not simulated here. Air is a far poorer heat transfer medium than water, so a dry cooler needs much more surface area and fan power for the same duty, and critically, its performance tracks dry bulb temperature rather than the usually much lower wet bulb — a real penalty on hot days, visible directly in the comparison table above.

For power plants, fan electricity is not dry cooling's main cost. Because a dry system can only reject heat by raising the temperature (and therefore the pressure) at which it condenses steam, it forces the turbine to exhaust against a higher back pressure than a wet tower would. Higher back pressure means less work extracted per unit of steam — lower turbine output and a higher heat rate (more fuel per unit of electricity). This penalty is worst exactly when it hurts most: on hot afternoons, when the ambient dry bulb is high and electricity is often most valuable. U.S. Department of Energy studies of dry-cooled power plants report material annual efficiency and output penalties versus wet cooling, with the magnitude strongly dependent on climate and plant type — typically modest in cool climates but significant in hot, dry ones. That lost generation, not the fan power, is usually the dominant economic cost of going dry. This calculator sizes and rates the cooling equipment itself; for how condenser temperature and back pressure feed back into turbine output, see the condenser & turbine back-pressure calculator.

The comparison table also carries representative capital cost, operating cost, characteristic structure height, estimated land area, plume and noise figures for each type. These are typical, order-of-magnitude values for comparing the technologies against each other, not a substitute for project-specific costing, noise modelling, or a site visual impact assessment. Land area follows the same logic as fan power: air is a poor heat transfer medium, so dry cooling needs several times the footprint of a wet tower for the same duty, while mechanical draft's purpose-built cells are usually the most compact option. Noise mostly tracks whether a design has fans at all and how many: natural draft (no fans) is typically quietest, and dry cooling (large fan arrays moving huge air volumes) is typically loudest.

Evaporation is the whole reason a wet tower can cool water below the ambient dry bulb temperature: turning a small fraction of the flow to vapour removes a large amount of latent heat per kilogram evaporated, so a little evaporation rejects a lot of heat.

E = flatent · Q / hfg

Not quite all of the rejected heat leaves as evaporation, though. A wet tower rejects heat by two mechanisms at once: latent heat (water evaporating) and sensible heat (the air simply leaving warmer than it arrived). Only the latent part actually consumes water, so this calculator multiplies the duty by a representative latent fraction flatent ≈ 0.8 before dividing by the latent heat of vaporisation. Assuming the entire duty were latent (the common textbook shortcut E = Q/hfg) would overstate water loss by roughly 20%. The real split shifts with conditions — the sensible share grows in hot, dry air and shrinks in cool, humid air — and a full treatment would track it through the air-side enthalpy and humidity balance; the fixed fraction here is the honest level of detail for a screening estimate. With it, the predicted evaporation lands near the standard field figure of roughly 1% of circulating flow for every 5–7 °C (about 10–13 °F) of cooling range.

Drift is fine liquid droplets carried out of the tower by the airflow without evaporating — pure water loss (plus whatever is dissolved in it) with no cooling benefit. Modern drift eliminators keep this to a typical 0.001–0.02% of circulating flow, adjustable here.

Blowdown exists because evaporation leaves dissolved solids behind, concentrating them in the remaining water. Left unchecked, this leads to scale and corrosion, so a fraction of the circulating water is deliberately discharged and replaced with fresh makeup, controlled by the cycles of concentration you choose to run at:

B = E / (Cycles of concentration − 1) − D

The displayed shorthand is often just B = E/(COC−1), but drift D already removes concentrated water from the system, so it counts toward holding the cycles of concentration and is subtracted from the blowdown the tower still needs — which is exactly what this calculator computes (and it never lets blowdown go below zero, for the case where drift alone would over-satisfy the target). Running at a higher COC needs less blowdown (and therefore less total makeup water) for the same evaporation, at the cost of a more concentrated, more scale-prone water chemistry — the trade-off the cycles-of-concentration slider lets you explore directly.

Seasonal variation. Makeup is not constant through the year. Evaporation is set by the heat rejected and the latent heat at the operating water temperature, both of which move with the weather: warm, dry conditions push the water temperatures up and increase evaporative loss, while cool, humid conditions reduce it. Change the ambient dry bulb and humidity above (or the built-in seasonal scenarios) and the makeup figure moves with them — peak makeup demand is a summer number, not an annual average.

A deliberate simplification: makeup temperature. Fresh makeup water enters at its source temperature — a river or reservoir in winter can be far colder than the recirculating water — so strictly it carries a small sensible heat term into the basin that shifts slightly with season. This calculator does not model that term, because it is genuinely second-order: at a typical makeup rate near 1–2% of the circulating flow, even a 10 °C makeup-to-basin temperature difference is under about 2% of the cooling duty, and it does not change the tower sizing in any meaningful way. It is called out here so the omission is explicit rather than hidden — a full plant heat-and-mass balance would include it, but it would not move the numbers this tool reports.

Fan power is simply the work done moving air against the tower's internal resistance, divided by how efficiently the fan and motor do that work:

P = ΔP · Air flow / efficiency

Why this matters: for a fixed fan and tower geometry, resistance (the pressure the fan works against) rises with the square of flow along the system's resistance curve, so power - pressure times flow - rises with the cube. Increasing airflow improves thermal performance but also increases fan power roughly with the cube of fan speed for geometrically similar fans: doubling airflow on the same fan takes roughly eight times the power, not twice. The pressure-rise slider above is anchored to this tower's own design flow, so dragging the air flow slider away from the auto-scaled default is exactly where you can see that cube-law penalty appear directly in the fan power readout.

This is genuine parasitic load — electricity the plant generates but never sells — so real designs treat fan power as a cost to be traded against thermal performance, not a free lever. The resulting fan demand is a direct auxiliary electrical load. Its practical significance depends on the plant output, operating schedule, electricity price and the rest of the cooling-system auxiliaries. One caution on that figure: it is fan power as a percentage of heat rejected (a thermal quantity), not as a percentage of the plant's electrical output. Those are very different denominators — for a power plant, heat rejected is roughly two to three times gross electrical output, so a fan power of, say, 1% of heat rejected can be on the order of 2–3% of electrical output. The label is written as "fan power / heat rejected" here specifically to avoid that confusion.

Annual operating cost here is fan electricity + pump electricity + water + chemical treatment, each a direct multiplication of an engineering result (fan power, circulating pump power, makeup water volume) by a price you set. Pump power was added specifically because a circulating water pump is real, continuous electricity consumption that earlier versions of this calculator omitted entirely - it uses a representative total pumping head per tower type (static lift to the distribution point plus fill or coil friction), not a full piping and fitting model. Dry cooling's head is set higher than a wet tower's for two real reasons: its finned coil bundles sit on an elevated support structure, and finned tube coils have a genuinely higher pressure drop per pass than open fill.

What this total does not include, and why it matters more than anything above: for a power generation application, the cooling tower's performance changes the turbine's own condenser backpressure. A tower with a worse approach runs a hotter condenser, which measurably reduces turbine output for the exact same fuel or reactor heat input - electricity the plant simply never generates, which typically costs far more than the direct fan, pump, water and chemical costs combined for any tower type performing meaningfully worse than the alternatives. This calculator does not model that turbine-side effect at all; it is a genuinely different calculation, covered by the Condenser Cooling & Circulating Water Calculator linked above. Treat the operating cost total here as the direct, controllable cost of running the cooling system itself - not as the full economic picture for comparing tower types on a power plant, where the lost-generation effect of a poorer approach will often dominate every number on this page.

Prices are also genuinely variable by region and by year - electricity, industrial water, and water-treatment chemical costs can each easily differ by two to three times what the default sliders show, and the water price in particular is what usually decides which tower type looks cheapest here. Adjust all three to your own site before drawing a conclusion from the comparison table.

Tower exhaust air is warm and close to saturated with moisture. Whether that produces a visible plume depends on what happens when it mixes with the surrounding air: if the exhaust's dew point still exceeds the ambient temperature after mixing, moisture condenses back out as visible droplets — the white "steam" plume familiar from cooling towers on cold mornings. This calculator uses a simplified screening check on that condition, not a full plume dispersion model, so treat the result as indicative rather than a prediction for a specific site. The qualitative pattern it reproduces is the real one: plumes are far more common on cold, humid days than on hot, dry ones.

Genuine Merkel theory, solved as a real characteristic-vs-demand crossing rather than a canned lookup table, is the technical core of this tool and the reason it responds correctly to fill height, air flow, weather and fill type all independently.

Representative, not site-specific. Fill characteristic constants, drift and fan efficiency defaults are typical published figures, not a specific manufacturer's test data. Use this to build intuition about the trade-offs, not to size or accept a real tower.

No wind effects, no tower internal recirculation, no icing. Wind speed is not modelled here; in reality, cross-wind can recirculate warm, humid exhaust back into a tower's own air intake and measurably hurt performance, especially on natural draft towers.

Height profile uses a simplifying air-side process line. The water temperature profile up the tower is a direct result of the Merkel solve and is robust. The air temperature and humidity profile shown alongside it assumes a smooth, physically reasonable humidification path rather than a cell-by-cell simulation, so treat its exact shape as illustrative.

Rough benchmarks for judging whether a design or an operating condition is running well, poorly, or somewhere in between. These are general industry rules of thumb, not a substitute for a manufacturer's guarantee or a specific project's design basis.

Worth noting how these interact: pushing approach toward "excellent" needs more fill and airflow (see Required Tower Sizing above), which tends to push fan parasitic load the wrong way at the same time - these two rows pull against each other, not independently. Cycles of concentration sits in its own trade-off with water treatment, covered in Water balance above.

Wet fill characteristic. In the heat-capacity-consistent Merkel formulation used here, film fill uses KaV/L = 1.25(L/G)−0.60 per metre of fill and splash fill uses 0.85(L/G)−0.55 per metre. These are representative educational curves calibrated for plausible teaching examples, not certified data for a specific commercial fill.

Natural draft. Airflow is estimated from tower height, diameter and an empirically calibrated buoyancy coefficient. It demonstrates trends but is not a project-specific aerodynamic tower model.

Dry cooling. The model is a closed-circuit water-to-air fluid cooler using effectiveness–NTU. Design airflow is estimated from the selected dry-air temperature rise; it is not a direct steam-condensing ACC model.

Fan and pump assumptions. Fan pressure rise is entered by the user. Representative pump heads are 12 m natural draft, 13 m induced draft, 11 m forced draft, 15 m hybrid and 16 m dry; pump efficiency is 75%.

Water balance. Evaporation uses a fixed 80% latent-heat fraction. The exact latent/sensible split varies with weather and tower operation, so water-consumption results are screening estimates.

Hybrid arrangement. The hybrid model assumes ideal hydraulic balancing between parallel wet and dry water-side sections. Series arrangements, bypass valves and seasonal routing controls are not simulated.

Why do nuclear plants use cooling towers?
A nuclear plant's turbine condenser has to reject roughly two-thirds of the reactor's thermal output as waste heat - only the remainder becomes electricity, the same as any steam-cycle power station. That heat has to go somewhere. A cooling tower rejects it to the atmosphere instead of continuously drawing from and discharging into a river, lake or sea, which is why towers are the default choice wherever a large enough natural water body isn't available, or where thermal discharge limits rule out once-through cooling. The reactor itself doesn't care how the heat is removed; the cooling system is a balance-of-plant choice, not a nuclear safety one.

What is Merkel theory?
Merkel theory, published by Frederick Merkel in 1925, is the foundational method for analysing wet cooling tower performance. Its key move is combining heat and mass transfer into a single enthalpy-based driving force, since a wet tower cools mainly through evaporation, not simple sensible heat transfer to cooler air. That produces the defining integral this calculator solves - see "How this calculator works" above for the full equation and how it is evaluated numerically.

Why does cooling tower efficiency decrease in summer?
A wet tower's real ceiling is wet bulb temperature, not dry bulb - and wet bulb rises with both temperature and humidity. A hot, humid summer day pushes wet bulb up more than a hot, dry day does, shrinking the usable driving force for evaporative cooling and pushing cold water temperature up with it. This is the seasonal pattern the weather presets and calculated approach are intended to demonstrate: summer generally means a higher approach for the same tower, not because anything in the tower changed, but because the atmosphere itself became a worse heat sink.

What is approach?
Approach is the difference between the cold water leaving the tower and the ambient wet bulb temperature (or dry bulb, for a dry cooler) — the coldest the surrounding air could possibly make that water. It is the single number that best summarises how close a tower is running to the physical limit set by the weather.

Why does a lower approach need a bigger tower?
The Merkel demand integral grows sharply as cold water temperature approaches the wet bulb, because the driving force between water and air shrinks toward zero right at that limit. Squeezing out the last few degrees of approach needs disproportionately more transfer units — more fill height, more air flow, or both — for rapidly diminishing returns. Reducing approach from 6 °C to 4 °C (about 10.8 °F to 7.2 °F on a delta scale) at otherwise unchanged conditions often requires around 35–40% more fill volume, not the roughly 33% you might expect from the temperature change alone — try it in the Required Tower Sizing panel above. That is exactly why real towers are usually designed to a specific approach, balancing capital cost and fan power against the diminishing thermal return, rather than chasing the smallest possible number.

Why does humidity hurt performance even when temperature doesn't change?
Wet bulb temperature, not dry bulb, is the real ceiling on how cold a wet tower can get the water. At a fixed dry bulb, raising relative humidity raises the wet bulb temperature toward the dry bulb, shrinking the usable driving force for evaporative cooling — which is why the same 35 °C (95 °F) day can be a good cooling day (dry) or a terrible one (humid).

Why is evaporation the main loss, but blowdown the one you control?
Evaporation is fixed by the heat you need to reject — it is the mechanism doing the cooling, not a waste to be minimised. Blowdown, by contrast, is a choice: how concentrated you are willing to let the water chemistry become before discharging some of it, set directly by the cycles-of-concentration slider.