Cooling Things Down

The Parking Lot

A few years back, I helped a customer replace an aging air-cooled chiller with a centrifugal chiller and a new cooling tower. This was a critical care facility and despite my insistence, they had chosen to do it in the dead heat of summer.

I knew that the transition was critical, so I worked my butt off to make sure that the startup went smoothly. Chemical tanks were staged and filled. The controller was installed and operational. The water chemistry was dialed in.

The day it came online I anxiously monitored the results. I watched the tower climb to 1,200 µS and heard the automated blowdown solenoid click open, right on setpoint. I measured the chemical tracer (PTSA) and it matched the online reading and target dosage exactly. Everything looked perfect.

Which is why I was stunned by the call I received the next day.

“The cooling tower is using way too much water,” the facility manager informed me. “There has to be a problem with your blowdown controller.”

I told him I had checked, monitored, and verified the blowdown, and I was confident it was working correctly.

He responded without missing a beat.

“Then why is it filling up our neighbor's parking lot?”

“Well,” I said, “it shouldn’t be doing that. I’ll come take a look.”

By the time I arrived onsite, the problem was impossible to miss. About 150 yards away, an adjacent parking lot was covered in standing water. Not a puddle. Not a wet spot. A shallow sheet of water spread across the pavement.

I walked over with my conductivity meter, hoping the number would prove the cooling tower innocent.

It did not.

The water in the parking lot read 1,200 µS.

My stomach dropped.

I checked the PTSA tracer next. It matched the tower too.

This was cooling tower water.

A quick call to the mechanical contractor explained why. During installation, the cooling tower blowdown line had been tied into the “most obvious pipe.” Unfortunately, the obvious choice was the wrong one. Instead of discharging to the sanitary sewer, the blowdown had been routed to the neighbor’s parking lot.

That part was a real problem. Cooling tower blowdown is concentrated, chemically treated water. It cannot simply be sent wherever a drain happens to lead. The contractor agreed to correct the issue immediately, and I went to tell the facility manager the good news.

“Great,” he responded. “But that doesn’t answer the real problem. The real problem is that we’re wasting so much water. There’s just no way a cooling tower should be using enough to fill up a whole parking lot.”

I was tempted to reply on the spot, but told him I’d look into it.

The truth was, I knew there was nothing wrong with the cooling tower.

The conductivity sensor was calibrated. The blowdown solenoid opened and closed on command. The basin water level sat below the overflow. The tower was not “wasting” water. It was doing exactly what it was designed to do.

And I could prove it with three numbers.

I made a quick stop at the mechanical room before heading back to my office. The condenser flow was about 3,000 gallons per minute; right on target for a 1,000-ton chiller under full summer load. The condenser water temperature change was 10°F, matching the design spec for the cooling tower. The water chemistry showed the tower was operating at 3 cycles of concentration, which was the practical limit for the treatment program.

I pulled up my cooling tower calculator and plugged in the numbers.

The blowdown (the water flooding the parking lot) came to about 14 gallons per minute.

Running at full load, that was almost 20,000 gallons per day. Enough water to cover a parking lot in a shallow sheet. Enough water to make everyone standing there believe something had to be broken.

But the math was certain.

And it revealed an even bigger number that nobody had asked about.

To keep this building cool, the tower was evaporating about 28 gallons every minute. Twice the blowdown. While everyone stood staring at the flooded parking lot, twice as much water was rising invisibly out of the top of the tower.

Over a six-month cooling season averaging 14 hours of full-load cooling per day, I estimated this tower would evaporate close to 4.3 million gallons of water and blow down just over 2 million more.

I worked up the report and brought it back to the facility manager.

He stared at the numbers in disbelief.

"You're telling me the cooling tower is going to use over six million gallons of water this year. Just to cool the building."

“Yes,” I said. “That’s how it works.”

It took a couple of weeks, and a few more conversations, to convince him the tower was not broken. That every cooling tower does this. It's just that most of the time, the water disappears quietly. Evaporation rises into the atmosphere. Blowdown drains into the sewer. Nobody sees it, so nobody feels it.

The difference was that this tower showed its work.

The mass balance was scrawled across the pavement.

The Problem is Heat

The real lesson from the parking lot wasn’t the volume of water.

It was the problem of heat.

It’s easy to make things hotter. Almost everything generates heat spontaneously: people, lights, pumps, computers, and equipment all shed heat into their surroundings as the unfortunate byproduct of doing anything useful. Above it all, the blistering summer sun beats down on the roof while the ventilation system pulls hot outside air in.

Getting rid of heat is another matter. Heat is intangible; it is the internal energy of matter itself. You can’t scoop it up and throw it away, and you can’t add “cold” to it, because cold is not a substance. The only option is to move heat somewhere else.

And heat is notoriously stubborn about where it goes. Left to its own devices, it moves in only one direction: from hot to cold. Leftover coffee cools to room temperature. Ice cream melts on a hot day. Buildings get warmer.

Heat follows the gradient.

Always downhill. Never back.

A building in summer sits at the bottom of that hill. The rooms are held at 72°F while the afternoon outside pushes 100. Every natural flow of heat runs inward: through the walls, through the windows, through every door that opens. To keep the building cool, heat must be gathered up from every room and forced into air that is already hotter than the rooms it came from.

Pushed uphill. Against everything heat wants to do. And pushing anything uphill has a cost. Nature delivers heat for free and charges you to haul it away.

That is why the cooling tower flooded a parking lot in a single day. It was not wasting water. It was moving heat.

There are only two ways to pay price of cooling: offer heat a downhill path it is eager to take, or pump it uphill through work. Every cooling system ever built is one of these two moves, or both.

The Two Ways Heat Moves

At an industrial scale, we’re often working with a lot of heat. The number of BTU’s moved through most factories is so enormous that it can start to feel abstract.

But there are only two real ways to move heat efficiently, and you’ve likely been familiar with them your entire life. They are the quiet machinery keeping our homes cool, and they are quite similar to everything we see at the industrial scale.

In the arid Southwest, many homes use evaporative coolers, often called swamp coolers. They work by passing the hot outside air across wet cooling pads, transferring sensible heat out of the air and into the water. As the water recirculates over the pads, evaporation removes latent heat energy from the bulk water. This cools the recirculating water, allowing the system to continuously remove heat from the outside air. It is the downhill path: evaporation carries the heat away almost effortlessly, and the water line running all summer pays for it.

Anyone who has changed those pads knows what the water leaves behind: a white, crusted layer of everything that didn’t evaporate. Changing pads can be a mild maintenance headache, but overall swamp coolers are simple, elegant, and effective. But only when the air is dry.

In more humid climates, evaporative cooling does not work as well. The relative humidity in the surrounding air limits evaporation, which limits the amount of heat removal.

Air conditioners solve this problem with a closed refrigerant loop, which removes heat from the indoor air and rejects it outdoors. The refrigerant absorbs heat from the incoming air as it evaporates inside the fan coils. It then rejects the heat at the AC unit and condenses back into a usable liquid. The compressor is what makes this possible: it supplies the work that lifts heat uphill, out of a cool room and into hotter outdoor air.

These two household systems quietly demonstrate the two halves of industrial cooling:

A swamp cooler is a tiny cooling tower.

A home air conditioner works like a small air-cooled chiller.

One gives heat a downhill path through evaporation. The other pumps heat uphill with compressor work. For small buildings, these two moves are enough. A window unit or a rooftop AC handles the load with margin to spare. And the cost of moving the heat stays invisible: a few gallons through the swamp cooler, a few dollars on the electric bill.

Nobody meters it. Nobody floods a parking lot.

When Cooling Gets Big

At some point, cooling stops being a comfort decision and becomes a structural requirement.

A hospital cannot allow surgical suites to warm up. A data center cannot tolerate a five-degree rise in server inlet temperature. A semiconductor fabrication facility requires cooling water within fractions of a degree, continuously, for months at a time. A pharmaceutical plant cannot let a reactor run hot without destroying the product inside it.

These are not preferences. They are operational constraints.

The cooling loads involved at most of these facilities are enormous. We measure them in tons of refrigeration – a unit that originated in the ice-making industry, where cooling capacity was defined by the amount of ice a machine could freeze in a day. One ton of refrigeration represents 12,000 BTU per hour of continuous heat removal.

Most homes require somewhere in the ballpark of 24,000 to 60,000 BTU of cooling, or about 2 to 5 tons of refrigeration.

A large office building, on the other hand, might require 500 to 2,000 tons. A hospital campus, 3,000 to 5,000. A data center, 5,000 to 20,000 or more. A petrochemical refinery can demand cooling measured in the tens of thousands of tons.

At these scales, the household solutions collapse. And the reason is the transport medium.

Swamp coolers and air conditioners move cooled air to keep our homes cooled, but air is a terrible way to move heat. It’s specific heat is roughly 0.24 BTU/lb·°F (less than a quarter of water’s) and it is nearly 800 times less dense than water at standard conditions. To move the same amount of heat, you must push vastly more volume through vastly larger ducts with vastly more powerful fans. Ductwork grows massive. Fan energy explodes. Rooftop units multiply into a maintenance nightmares. Each with its own refrigerant charge, its own filters, its own failure modes.

Fortunately, we can rely on water, which has none of these problems. In the Hero of Heat Transfer chapter, we explored why that is: water’s high specific heat absorbs large amounts of heat with only modest temperature rise, and its high latent heat of vaporization rejects enormous energy when only a small fraction changes phase. A six-inch chilled water pipe carries as much cooling capacity as a four-foot duct moving air. The pumping energy is a fraction of the fan energy. The piping is compact, durable, and easy to insulate. Redundancy is simple: a second chiller and a set of valves.

Multiple factors determine the breaking point at which cool air will no longer suffice: load size, building size, operating hours, layout, maintenance staff. As a practical rule of thumb, once you’re in the Costco-sized building class (~150,000 sq ft), you’re often in the range where centralized chilled water starts to make sense.

Beyond that point, the logic shifts decisively: you stop trying to move cool air to the problem, and you start moving cool water.

This is why every large cooling system converges on the same architecture. A chiller produces chilled water. That chilled water is pumped to wherever cooling is needed: air handlers, fan coils, process heat exchangers. The water absorbs heat, returns to the chiller, and is cooled again. The chiller does not cool the building directly. It produces a cold fluid that does the work on its behalf.

Understanding what happens inside the chiller, and why it needs a cooling tower to survive, is the foundation of everything we do in open recirculating water treatment.

What a Chiller Actually Does

A chiller is a heat pump dedicated to cooling.

It moves heat from a chilled-water loop into a condenser-water loop using a refrigerant cycle. The chiller does not create cold. It relocates heat, lifting it from a low temperature where it causes problems to a higher temperature where it can be rejected.

To understand how, we must first return to refrigerants.

In the Hero of Heat Transfer, we explored why water resists phase change so stubbornly. Its hydrogen bonds create an enormous energetic barrier: roughly 970 BTU per pound must be invested to transform liquid water into steam. 

Refrigerants are the opposite.

They are molecules held together by weaker intermolecular forces, which require far less energy to vaporize and condense. Under the right pressure, they can evaporate cold enough to chill water and condense warm enough to reject heat. That pressure-temperature relationship is the key.

R-134a boils at –15°F (–26°C).
Ammonia boils at –28°F (–33°C).

This is why we described refrigerants as “quitters by design.” Their willingness to change phase at low temperatures is precisely what makes refrigeration possible. They absorb heat by evaporating at a useful temperature, and they release it by condensing at a slightly higher one.

But temperature isn’t the only driver of phase changes. Air conditions, chillers, and even refrigerators all rely on a secondary lever to make phase changes happen. The other part of the equation is pressure.

Changing the pressure on a substance changes how tightly packed its atoms or molecules have to be. Pushing hard on a gas, through a compressor, forces the gases widely spread molecules into a smaller space, making them interact like a liquid. Even for high energy gas molecules, raising the pressure provides the opportunity for intermolecular forces to take hold, and for cohesion to win over motion. The impact is an increase in the condensation point of the gas – the temperature at which it can exist as a liquid.

As you might expect, lowering the pressure has the opposite effect. By decreasing how tightly bunched molecules are to one another, they become more inclined to become gases, lowering the boiler point.

Refrigeration relies on precisely this principle to determine when, and where, a refrigerant will be forced to change phase. By changing pressures through compression and expansion, we choose the temperatures at which it absorbs and releases heat through phase changes.

This gives us control over a process that thermodynamics would otherwise forbid: moving heat from cold to hot.

The Four Components of a Chiller

A chiller manages the phase of a refrigerant with four components, each playing a distinct role in the refrigeration cycle.

Evaporator (Chilled Water Side)

The evaporator is a heat exchanger where the refrigerant absorbs heat from the chilled water loop. The refrigerant enters as a cold, low-pressure liquid. Inside the evaporator, pressure is kept low enough that the refrigerant boils at approximately 38–42°F. As it evaporates, it absorbs latent heat from the chilled water flowing over the tubes. The chilled water transfers the building heat into the refrigerant and leaves cold, typically 42–44°F, where it is pumped back into the building.

This is where the chiller does its job. Everything else exists to reset the refrigerant so it can do this again.

Compressor

The compressor takes the low-pressure refrigerant vapor exiting the evaporator and squeezes it. Compression raises both the pressure and the temperature of the vapor, transforming it into a superheated, high-pressure gas. The electricity required to run the compressor is the primary energy input to the system. The compressor is doing the thermodynamic work of lifting heat from a low temperature to a higher one. Nature always demands a cost.

Condenser (Cooling Tower Side)

The condenser is a heat exchanger where the refrigerant releases the heat it absorbed in the evaporator plus the heat added by compression. At the elevated pressure created by the compressor, refrigerants commonly condense around 80-120°F. As it condenses, its latent heat is released into the condenser water flowing over the tubes. The condenser water absorbs this heat and carries it away.

This is the critical handoff. The heat that was in the building is now in the condenser water.

Expansion Valve

The expansion valve throttles the high-pressure liquid refrigerant to a lower pressure liquid/vapor. This pressure drop causes part of the refrigerant to flash and lowers its saturation temperature, returning it to a cold, low-pressure state ready to absorb heat again in the evaporator.

The cycle runs continuously. Evaporate. Compress. Condense. Expand. Each revolution moves heat from the chilled water loop into the condenser water loop.

The Cooling Tower

A chiller doesn’t need a cooling tower to do its job. It has the compressor for that.

Air-cooled chillers rely on the outdoor air to condense the compressed gas back into a usable liquid. The chilled water loop remains the same, but the refrigerant cycle depends on ambient air to accomplish a phase change. This often requires the compressor to work harder, and use more electricity, to increase the pressure of the gas high enough that it will condense on very warm days. The downside is higher electrical consumption, but the benefit is very little water usage.

Water earns that role because it is extraordinarily equipped for the task. It carries heat better than any other common liquid, but the real magic is in how it releases it.

Air-cooled chillers rely on dry-bulb temperature, the standard temperature read by a thermometer. If it’s 100°F outside, it has to push heat into 100°F air, forcing the compressor to work incredibly hard.

Cooling towers, on the other hand, rely on wet-bulb temperature, which is the lowest temperature air can reach through evaporation. Just as sweating cools your body even on a hot summer day, evaporative cooling can remove more heat than the outdoor air might freely allow. The benefits are that cooling towers are capable of handling enormous heat loads, even on hot summer days, and they can do so with less help from the compressor. This makes cooling towers generally more energy efficient than air cooled chillers, but the downside is the amount of water required to do it.

A cooling tower, or condenser water system, leaves the chiller carrying heat that has been absorbed from the building, lifted by the compressor, and transferred across the tubes. It provides the final release through evaporation.

Warm condenser water is pumped to the top of the cooling tower and distributed across a large surface area of fill media. As air is drawn across the wetted fill, a small fraction of the water evaporates. Each pound of water that transitions from liquid to vapor absorbs roughly 970 BTU of latent heat, cooling the remaining liquid water. 

The general rule of thumb bears repeating: evaporating 1% of the recirculating water cools the bulk water by approximately 10°F. This is an extraordinary amount of heat rejection accomplished by a very small amount of water loss.

The cooled water collects in the tower basin and is pumped back to the chiller condenser, where it absorbs heat again. The cycle repeats.

When the Cycle Breaks Down

If the condenser cannot transfer heat effectively, the consequences cascade upstream.

Condenser tubes are scaled or fouled → heat transfer is impeded → refrigerant condensing temperature and pressure rise → the compressor must work across a larger pressure differential, a condition called increased lift.

Energy consumption rises. Cooling capacity falls. If the imbalance is severe enough, the chiller protects itself by unloading or shutting down entirely.

The tower does not do the thermodynamic work of the chiller. But the chiller cannot function without the tower’s cooperation.

Every threat we have studied in this book converges at the cooling tower. Scale insulates heat-transfer surfaces and reduces the tower’s ability to wet its fill. Corrosion weakens the structural steel and piping that hold the system together. Biofilm fouls surfaces, restricts heat transfer, creates health hazards, and shelters organisms from biocides. Suspended solids settle in basins and condenser tubes, amplifying everything else.

We do not treat cooling towers because we care about towers. We treat them because the chiller, the building, and every process downstream depends on the tower doing its job.

The Cooling Tower Mass Balance

A cooling tower is an open water system. Water enters as makeup. Water leaves as evaporation, blowdown, drift, and leaks. The balance between these flows determines the chemistry, the efficiency, and the cost of the entire operation.

Understanding the mass balance is the foundation of cooling tower management, and must answer each of the four questions:

1. What enters?

2. What leaves?

3. What accumulates?

4. What changes form?

What Enters and What Leaves

The overall water mass balance is simple: what comes in must equal what goes out.

Makeup = Evaporation + Blowdown + Drift + Leaks

Evaporation

Evaporation is the tower’s purpose. It is how the system rejects heat. 

We derived this relationship in the Hero of Heat Transfer Chapter: evaporating water absorbs approximately 1,000 BTU/lb of latent heat. If the tower must reject a given heat load, a corresponding amount of water must evaporate. The Engineering Notes provide two methods for calculating evaporation rate in the field.

This is the first key idea in the tower mass balance: evaporation is set by load.

Blowdown

Evaporation leaves dissolved minerals behind. As pure water leaves as vapor, the remaining water becomes more concentrated. Calcium rises. Alkalinity rises. Silica rises. Chloride rises. Everything that does not evaporate becomes more crowded, including microbes and suspended solids.

Blowdown is the controlled release of that concentrated water. Tower water lost to blowdown is replaced by fresh makeup, which reduces the chemical concentration and prevents the tower from reaching the point of scale, corrosion, or severe fouling. Controlled blowdown can generally be metered while the rest of water loss can only be calculated.

This is the second key idea in the tower mass balance: blowdown is the main controllable loss term.

Drift and Leaks

Drift is the small quantity of liquid water that escapes the tower as fine droplets carried by the exhaust air. Modern drift eliminators reduce this loss to roughly 0.001–0.006% of the recirculation rate. It is typically a minor term in the water mass balance, but it is not zero.

Leaks are unpredictable but must be accounted for. Both drift and leaks represent unmetered water leaving the system. They carry the full concentration of the tower water with them.

This is the third key idea in the tower mass balance: unmetered losses act as uncontrolled blowdown, and they change the mass balance.

Makeup

Makeup is the fresh water added to replace everything the tower loses:

Makeup = Evaporation + Blowdown + Drift + Leaks

Every gallon of water the tower consumes traces back to these four terms.

What Accumulates

So far we have accounted for the water that enters and leaves the cooling tower. Now we must account for what the water carries along the way.

Makeup water introduces a fresh, relatively low baseline of dissolved ions. Some (calcium, phosphate, silica) are likely to form scale as they enter the cooling tower system. Others (chloride and sulfate) are likely to contribute to corrosion inside the recirculating system. Makeup can also carry suspended solids and biological matter that lead to fouling and biofilm. Effective pretreatment minimizes these last contaminants, but the tower collects its own: during operation, a cooling tower acts as a giant air scrubber, pulling massive amounts of outdoor air across the fill and mixing airborne dust, pollen, and microbes into the recirculating water.

Everything the water carries, the tower concentrates.

Evaporation removes pure water and nothing else. The dissolved solids stay behind, accumulating in the recirculating water. How far they concentrate, on average, is measured by cycles of concentration (COC). Strictly speaking, COC is defined as the flow ratio between makeup and all non-evaporative water loss, which is commonly expressed as blowdown:

COC = $\frac{Makeup\ Flow\ Rate\ }{Blowdown\ Flow\ Rate}$

Because not every tower flow is metered, we rely on proxies to estimate what is actually occurring. At lower blowdown rates, everything the makeup delivers accumulates: conductivity, TDS, chloride, calcium, silica, tracers. The higher the cycles, the more accumulation. The chemistry itself becomes the meter:

COC ≈ $\frac{Concentration\ in\ Blowdown\ (Tower\ Water)\ }{Concentration\ in\ Makeup}$

COC is the single most important variable governing water usage, chemical consumption, and scaling risk in an open recirculating system. We introduced it in the Scale chapter as a measure of how crowded the solution has become. Here, it becomes the operational control point for the entire mass balance.

The relationship between cycles and blowdown follows a curve of diminishing returns. Moving from 2 cycles to 3 cycles cuts blowdown in half. Moving from 3 to 4 cuts it again, but by only 33%. Moving from 4 to 5 saves 25% more. The biggest gains come from the first improvements. After roughly 6 cycles, the water savings flatten.

But the risk does not flatten.

Every makeup water has a ceiling.

Concentrate any water far enough and something falls out of solution. The only question is what falls first. That first constituent is the limiting factor, and it sets the maximum cycles for the tower. It is different for every water. 

Sometimes the limit is scale. Calcium, alkalinity, silica, sulfate, pH, and temperature combine until the water can no longer hold the minerals in solution. Sometimes the limit is corrosion. Chloride and sulfate concentrate until the water becomes too aggressive for the metals in the system. Sometimes the limit is microbiology or suspended solids. Higher cycles mean less dilution, more nutrients, more dirt, and more demand on the treatment program. 

Treatment chemistry extends each of these limits. No chemistry removes them.

Finding a tower's ceiling is a calculation, not a guess. This is where the Pillar Chapters return. They allow us to assess the risk of each pillar of water treatment (corrosion, scale, microbiological, suspended solids) and determine the limiting factor.

The question is not, “How high can I set the controller?” The real question is ““Which pillar fails first?” 

That determination is what set the "practical limit for the treatment program" back in the parking lot. That tower's makeup hit its ceiling at 3 cycles. The blowdown that flooded the lot was not a malfunction, and it was not waste. It was the requirement of water chemistry.

Each additional cycle multiplies every dissolved species in the water. Calcium, alkalinity, silica, sulfate, chloride, they all climb in lockstep. The scaling indices worsen. The demand on inhibitors increases. This is the fundamental tradeoff of cooling tower management: water efficiency versus chemical risk.

The Engineering Notes derive the exact relationships between evaporation, blowdown, makeup, and cycles, and include a water savings table that shows the diminishing returns in hard numbers.

What Changes Form

In practice, COC is usually estimated with conductivity. This works because most dissolved solids contribute to conductivity, making it a practical real-time surrogate for concentration. But it is not a perfect representation.

Conductivity can tell you that the water is concentrating, but it does not account for transformation within the system. This brings us to the final mass balance question: what changes form?

Past a mineral’s solubility limit, scale and deposition begin to occur. When calcium carbonate precipitates out of solution, its contribution to conductivity disappears, but the system does not know the difference. Makeup water continues to enter. The conductivity controller continues to hold its setpoint. The system makes up the shortfall, and the true concentration of scaling species drifts away from what conductivity reports.

This is why the best practice is to compare the COC between a soluble tracer ion (one that does not scale or deposit) against a scaling-risk ion (one that is likely to scale). Chloride is often used as the standard tracer, but only when chloride-containing oxidizers (like NaOCl) are not in use. Calcium is often the standard scaling indicator. If the chloride-based COC and the calcium-based COC diverge significantly, minerals are leaving the mass balance as scale.

This comparison is a diagnostic tool, not a routine measurement. But it is the only way to verify that the mass balance is truly in balance. That what we think is accumulating in solution has not quietly changed form and left the water as a deposit on a surface.

Where the Leverage Sits

Evaporation is the price of rejecting heat. Blowdown is the price of controlling concentration.

That distinction is the practical heart of cooling tower treatment. The water treater cannot negotiate with the heat load. If the building has heat to reject, the tower must evaporate water to reject it.

But blowdown is different.

Blowdown is controlled by how far we allow the water to concentrate before letting it go. Too little blowdown and the tower becomes a scale machine. Too much blowdown and the system wastes water, chemical, and money.

This is why cycles of concentration matter. They are not just a number on a controller. They are the operating compromise between water conservation and chemical risk.

The goal is not to run the highest possible cycles. The goal is to run the highest responsible cycles: high enough to conserve water, low enough to protect heat-transfer surfaces, and stable enough to keep corrosion, scale, biology, and suspended solids under control.

The flooded parking lot made the water balance visible.

The job of the water treater is to control that balance before it announces itself.

The Complete Picture

Follow the heat one last time.

It began in the building: shed by people, lights, and machines that never stop making it. The chilled water gathered it. The refrigerant absorbed it, and the compressor lifted it. The condenser water carried it to the tower. And at the fill, the water did the only thing that could finish the job: a few gallons left as vapor, carrying the heat of thousands with them.

Every step of that journey ran on the same molecule.

In the Hero of Heat Transfer chapter, we called water's hydrogen bonds a gift. They are the reason a single pound of vapor can haul a thousand BTU into the sky. The cooling tower is where the gift presents its bill. The bonds that make evaporation powerful enough to cool a hospital are the bonds that demand millions of gallons to do it. The gift and the cost were never two things.

They are the same property, read from opposite sides of the meter.

The facility manager looked at that flooded parking lot and saw a malfunction. You can now read what was actually written there: a heat load the building could not keep, the evaporation that heat demanded, the ceiling the water chemistry allowed, and the blowdown the weakest pillar required. Every gallon accounted for. Every gallon obeying the same chemistry this book began with.

That is the mass balance. Most cooling towers hide it – vapor into the sky, blowdown into a sewer, arithmetic no one ever sees.

One of them showed its work.

Engineering Notes: Cooling Things Down

"If you are reading this straight through, you can skip this section and lose nothing essential to the story. These notes are for the operators, engineers, and technicians who need to do the math."

Why This Matters in the Field

Cooling tower management is fundamentally a water balance problem. If you cannot calculate evaporation, blowdown, and makeup, you cannot set a conductivity target, size a blowdown valve, justify a treatment budget, or explain to a facility manager why their water bill changed.

These calculations are not complicated. But skipping them is how people end up wasting millions of gallons per year, or scaling a condenser in a single summer.

Core Tools & Constants

Tower Performance: Range and Approach

Range is the temperature drop of the water across the tower:

Range = T_(return) − T_(supply)

Typical range: 10–15°F for comfort cooling. Set by heat load and flow rate.

Approach is how close the cold water temperature gets to the ambient wet-bulb temperature:

Approach = T_(supply) − T_{(wet bulb)}

Typical approach: 5–10°F.

A smaller approach means a more effective tower, but usually a larger tower with more fill, more fan energy, and higher cost.

Wet-bulb temperature is the lowest temperature achievable solely through evaporative cooling. It is measured using a sling-psychrometer: a thermometer wrapped in a water-soaked cloth that is spun through the air to maximize evaporation. As water evaporates from the cloth, the measured temperature cools.

When relative humidity is 100%, the wet-bulb temperature equals the actual air temperature (dry-bulb). At lower humidity, the wet-bulb temperature is lower than the air temperature because evaporation proceeds faster allowing more cooling.

This defines the limit of evaporative cooling. As the water approaches the wet-bulb temperature, the air's remaining capacity to accept vapor approaches zero (and along with it, the driving force for evaporation).

This is the same reason swamp coolers work well in dry climates and poorly in humid ones. In dry air, water evaporates readily and the temperature can move significantly downward toward the wet-bulb. In humid air, the evaporation driving force is weaker because the air is already closer to saturation.

A useful way to think about it:

• Dry-bulb temperature tells you how hot the air is

• Wet-bulb temperature tells you how much evaporative cooling potential the air still has

• Approach tells you how effectively the tower is using that remaining potential

Refrigeration Tons vs. Cooling Tower Tons

This distinction matters for every calculation that follows.

A standard refrigeration ton is defined as 12,000 BTU/hr·ton. This is the amount of heat the chiller removes from the building. A 1,000-ton chiller at full capacity removes 12,000,000 BTU/hr from the building.

But the cooling tower must reject more than just the building's heat. The chiller compressor adds heat to the system as it does work. The cooling tower ton is commonly treated as 15,000 BTU/hr·ton_CT, a field convention that accounts for the building heat plus compressor heat and design margin.

The total heat load at the cooling tower for a 1,000-ton chiller is therefore roughly 15,000,000 BTU/hr.

When calculating evaporation, blowdown, or sizing a tower, always use the cooling tower heat load, not the chiller tonnage alone.

The Cooling Tower Mass Balance Equations

Overall Mass Balance

What enters must equal what leaves:

Makeup = Evaporation + Blowdown + Drift + Leaks

Drift and leaks are generally small. For calculation purposes, we simplify to:

M = E + B

Evaporation Rate

Method 1: Known heat rejection and applied evaporation factor:

E (lb/hr) = $\frac{\mathbf{Q}\ (\mathbf{BTU}/\mathbf{hr})\ \times \ \mathbf{f}}{\mathbf{h}_{\mathbf{fg}}\ (\mathbf{BTU}/\mathbf{lb})}$

Where:

• Q = heat load (BTU/hr)

• f = fraction of heat rejected through evaporation, commonly estimated around 0.85 for field calculations.

• h_fg = latent heat (~1,000 BTU/lb in cooling towers)

Example: For a heat load of 6,000,000 BTU/hr

E = $\frac{6,000,000\mathbf{\ }(BTU/hr)\ \mathbf{\times \ }0.85}{1,000\ (BTU/lb)}$ = 5,100 lb/hr

Converting to (GPM):

E = $\frac{5,100\ lb/hr\ }{8.34\ lb/gal\ \times \ 60\ \min/hr}$ = 10.2 gpm

Method 2: From Recirculation Rate and Range

In practice, the heat load is often unknown. The evaporation rate can be estimated from operating conditions and an evaporation factor:

E (gpm) ≈ $\frac{\mathbf{Recirc}\ \mathbf{Rate}\ (\mathbf{gpm})\ \times \ \mathbf{\Delta T}\ \times \ \mathbf{f}\ }{\mathbf{1},\mathbf{000}}$

Where:

• Recirc Rate is the water flow through the tower system in GPM. It can be estimated from nominal cooling tower tonnage: Recirc Rate = CT Tonnage × 3 GPM/ton. The nominal tonnage is typically found in the tower's engineering specifications.

• ΔT is the temperature change across the cooling tower (range) in °F. For most comfort cooling calculations it is typically around 10-15°F, but should be verified in the field.

• f is the evaporation factor: the percentage of cooling that is being accomplished through latent heat transfer (evaporation) vs sensible heat transfer. It is commonly estimated at 0.85, meaning that 85% of the heat transfer is due to evaporation, while 15% is sensible heat transfer. This value is relatively stable for mechanical-draft towers in typical comfort cooling applications, but can shift in extreme humidity or with very high-efficiency fill.

• 1,000 in the denominator is the rounded latent heat value, chosen for ease of field use.

Example: For a 400-ton cooling tower

The nominal cooling tower tons can be found in the design specifications, and shown to be equivalent to the previous example:

400 CT Tons × 15,000 BTU/hr·ton_CT = 6,000,000 BTU/hr Heat Load

Based on the CT Tons:

Recirc Rate (gpm) = CT Tonnage × 3 = 1,200 gpm

E ≈ $\frac{1,200\ (gpm)\ \times \ 10{^\circ}F\ \times \ 0.85\ }{1,000}$ = 10.2 gpm

Reconciling the Two Methods:

Method 1 calculates the evaporation rate based on heat rejection. The result is tied directly to the underlying thermodynamics, but the value for heat rejection is not always known. Method 2 calculates the evaporation rate based on design specifications, which are often easier to find. The results show that both methods can be used to arrive at the same answer. 

Whichever method is used, incorporating the evaporation factor is more accurate for real systems and should be used for field calculations. For field calculations, 0.85 is a practical default, but it varies with air conditions, tower design, and operating load. The story at the beginning of the chapter uses rounded field values, which fall within the expected range for a large tower operating in a dry climate.

The Solids Mass Balance and Cycles of Concentration

The overall mass balance accounts for water flows, but evaporation removes only pure water. Dissolved solids stay behind and accumulate in the tower. To account for this accumulation, we need a solids mass balance.

At steady state, the mass of solids entering the tower must equal the mass of solids leaving:

M × C_Makeup = B × C_Tower

Where C represents the concentration of dissolved solids, and the concentrations in the cooling tower and blowdown are equal.

Rearranging this formula:

M = B × $\frac{C\ (Tower)\ }{C\ (Makeup)}$

Where Cycles of Concentration (COC) = C(tower) ÷ C(makeup):

M = B × COC

This is the fundamental relationship linking the water balance to the solids balance.

Blowdown Rate

Substituting M = E + B into M = B × COC:

E + B = B × COC

E = B × COC − B

E = B × (COC − 1)

Solving for blowdown:

B = $\frac{\mathbf{E}\ }{(\mathbf{COC}\ - \ \mathbf{1})}$

Example**:** At 4 cycles with an evaporation rate of 10.2 gpm:

B = $\frac{10.2\ gpm\ }{(4\ - \ 1)}$ = 3.4 gpm

Makeup Rate

Substituting the blowdown equation back into M = E + B:

M = E + $(\frac{E\ }{(COC\ - \ 1)})$

M = E × (1 + $\frac{1\ }{(COC\ - \ 1)}$)

M = E × $(\frac{(COC\ - 1)\ + \ 1\ }{(COC\ - \ 1)})$

M = E × $(\frac{\mathbf{COC}\ }{(\mathbf{COC}\ - \ \mathbf{1})})$

Example: At 4 cycles with an evaporation rate of 10.2 gpm:

M = 10.2 × $\mathbf{(}\frac{4\ }{(4\ - \ 1)}\mathbf{)}$ = 13.6 gpm

Check: M = E + B = 10.2 + 3.4 = 13.6 gpm ✓

Conductivity and Cycles

In the field, COC is monitored through conductivity:

COC ≈ $\frac{Tower\ Conductivity}{Makeup\ Conductivity}$

This works because most dissolved solids contribute to conductivity. A conductivity controller on the blowdown valve maintains the target COC by opening the valve when tower conductivity rises above the setpoint.

The limitation: Conductivity does not account for transformation. If calcium carbonate precipitates out of solution as scale, its contribution to conductivity disappears. But the controller does not know this. Makeup continues, the conductivity setpoint holds, and the true concentration of scaling species drifts from what conductivity reports.

Verifying the Balance: Tracer Ion Comparison

The diagnostic for catching this is to compare cycles calculated from a soluble tracer ion against cycles calculated from a scaling-risk ion:

COC (tracer) = $\frac{\lbrack Cl\rbrack\ (Cooling\ Tower)\ }{\lbrack Cl\rbrack\ (Makeup)}$

COC (scaling) = $\frac{\lbrack Ca\rbrack\ (Cooling\ Tower)\ }{\lbrack Ca\rbrack\ (Makeup)}$

If COC(tracer) ≈ COC(scaling): the mass balance is intact. Solids are accumulating in solution as expected.

If COC(tracer) > COC(scaling): calcium is leaving the water and depositing as scale. The divergence quantifies the loss.

If COC(tracer) < COC(scaling): calcium is entering the water from somewhere other than the makeup. This can occur if calcium scale is being dissolved.

Example: If chloride-based COC is 5.0 but calcium-based COC is only 3.8, roughly 24% of the calcium entering the system is precipitating as scale rather than remaining in solution. This is direct evidence that the mass balance is not in balance, and the system is scaling.

Water Savings vs. Cycles

Impact of cycles on blowdown makeup for a tower evaporating 25 GPM:

(Assumes 8,760 hr/yr operation and negligible drift. Annual values rounded.)

Key insight: moving from 3 to 6 cycles saves nearly 4 million gallons per year on a single tower. Moving from 6 to 10 saves only 1.2 million more. The biggest gains come first.

Condenser Performance and Approach Temperature

The condenser approach temperature is one of the most important ways to gain insight into the heat transfer efficiency of the condenser. It calculates the difference between the Leaving Condenser Water Temperature (LCWT) and the Condenser Saturated Refrigerant Temperature (CSRT):

Condenser Approach (°F) = CSRT (°F) – LCWT (°F)

Condenser approach tells you how close the refrigerant condensing temperature is to the leaving condenser water temperature. Under comparable load and flow conditions, a rising approach means the condenser is losing heat-transfer efficiency, usually from scale, biofilm, fouling, low flow, refrigerant-side issues, or non-condensables.

The best practice is to monitor the approach at full load, where a normal range for a clean condenser is typically 1–3°F approach. This can change dramatically if the heat transfer surface becomes fouled, reducing the heat transfer ability of the system.

As fouling builds, the compressor must work harder to provide a suitable gaseous refrigerant to the condenser. This is reflected in a rough field rule:

1°F rise in approach ≈ 1.5–2% increase in chiller energy consumption

Example: A 1,000-ton chiller running at full load for 4,000 hours/year where fouling has increased the approach temperature +5°F, assuming $0.10/kWh.

Baseline Electrical Cost Estimate:

1,000 tons × 0.60 kW/ton = 600 kW

600 kW × 4,000 hr/yr = 2,400,000 kWh/yr

2,400,000 kWh/yr × $0.10/kWh = $240,000/yr

Fouling Penalty from 5°F increase to Approach (upper end):

600 kW × (5 × 2%) = 60 kW excess power draw

660 kW × 4,000 hr/yr × $0.10/kWh = $264,000/yr

The fouled condenser requires ~$24,000 more per year to operate, which is why scale and biofilm on condenser tubes are not just water problems, they’re energy problems too.

Worked Cooling Tower Mass Balance

Scenario: You are servicing a cooling tower connected to a 500-ton chiller. Makeup water conductivity is 400 µS/cm. You want to operate at 5 cycles of concentration. What are your targets?

Step 1: Determine Cooling Tower Heat Load

The 500-ton chiller rating is in standard refrigeration tons. The cooling tower must reject the building load plus compressor heat:

Q = 500 tons × 15,000 BTU/hr·ton_CT = 7,500,000 BTU/hr

Step 2: Estimate Recirculation Rate

Recirc Rate = 500 CT tons × 3 GPM/ton = 1,500 gpm

Step 3: Estimate Evaporation Rate

Assuming 10°F range across the tower and f = 0.85:

E = $\frac{Recirc\ Rate\ (gpm)\ \times \ \Delta T\ \times \ f\ }{1,000}$ = $\frac{1,500\ (gpm)\ \times \ 10{^\circ}F\ \times \ 0.85\ }{1,000}$ = 12.75 gpm

Step 4: Calculate Blowdown at 5 Cycles

B = $\frac{E\ }{(COC\ - \ 1)}$= $\frac{12.75\ gpm\ }{(5\ - \ 1)}$ = 3.19 gpm

Step 5: Calculate Makeup

M = E × $(\frac{COC\ }{(COC\ - \ 1)})$ = 12.75 × $\mathbf{(}\frac{5}{(5\ - \ 1)}\mathbf{)}$ = 15.94 gpm

Check: M = E + B = 12.75 + 3.19 = 15.94 gpm ✓

Step 6: Set Conductivity Target

Target tower conductivity = Makeup conductivity × COC = 400 × 5 = 2,000 µS/cm

Set the conductivity controller to open the blowdown valve at 2,000 µS/cm.

Step 7: Verify with Tracer Ion

At your next water analysis, compare:

• Chloride COC = Cl⁻ (tower) ÷ Cl⁻ (makeup)

• Calcium COC = Ca²⁺ (tower) ÷ Ca²⁺ (makeup)

If both are near 5.0, the balance is holding. If calcium COC is significantly lower, scale is forming and the treatment program needs adjustment.

Annual Water Budget:

At 5 cycles, running 8,760 hr/yr:

• Evaporation: 12.75 gpm × 60 min/hr × 8,760 hr/yr = 6.70 Mgal/yr

• Blowdown: 3.19 gpm × 60 min/hr × 8,760 hr/yr = 1.68 Mgal/yr

• Total makeup: 8.38 Mgal/yr

If this tower were running at 3 cycles instead of 5, blowdown would be 6.38 gpm and total makeup would be 10.07 Mgal/yr.

The difference: 1.69 million gallons per year saved by running 5 cycles instead of 3.