Heat Rejection: The Design Call You Can't Undo
The heat rejection architecture you pick at schematic design sets your water risk, your permit exposure, and your utility cost trajectory for the next two decades. Here is how to make that call with your eyes open.
Who this is for
- ■Hyperscale and edge data center operators sizing a new site
- ■Industrial manufacturers with high-density process heat
- ■On-site generation hosts with waste heat rejection loads
- ■EV charging depot operators managing thermal loads at scale
Water-cooled, dry-cooled, or hybrid: which architecture matches your climate, your water rights, and your twenty-year cost trajectory?
The heat rejection architecture you choose at schematic design locks in your facility's water risk profile for the life of the facility. You do not correct a bad schematic-phase call with later maintenance discipline. The water risk, the permit exposure, and the utility cost trajectory are all set at design.
Site permits
Water rights
Sunk capital
29%of this guide, read. The rest of it is below.
- 02 The mechanism Wet-bulb, dry-bulb, and why climate decides
Evaporative cooling has dominated for decades because of the physics. When water changes from liquid to vapor it absorbs roughly 970 BTU/lb at atmospheric conditions. That phase change lets an evaporative tower reach a fluid temperature close to wet-bulb, which in most climates sits well below dry-bulb.
System Reference Approach Achievable fluid temp Evaporative tower 24°C 3 to 5 C 27 to 29 C Dry cooler 35°C 5 to 8 C 40 to 43 C That gap forces water-free systems to use larger heat rejection surfaces, higher airflow, and more fan energy to hit comparable capacity. But the size of the gap is climate-dependent. In Phoenix the wet-bulb and dry-bulb spread is large and dry cooling gets penalized hard on energy. In Seattle or Northern Europe the spread collapses and dry cooling approaches evaporative efficiency. You cannot answer this question without a site-specific climate file.
Free cooling hours in temperate climates4,000hoursLow end, annual6,000hoursHigh end, annualIn the Pacific Northwest and Northern Europe, dry coolers can carry the load without compressors for thousands of hours a year, and every one of those hours is compressor work you are not paying for. 203 What it costs you The true cost of water-cooled operationThe ABC Carolinas industry reference cites a figure of roughly 110,000,000 gallons/yr for a medium-sized U.S. data center on open towers. Treat that as a ceiling-case illustration, not a planning number for your site. Water use efficiency varies by an order of magnitude across climates and IT load densities.
Loss breakdownWhere the water goes
Component What it is Share of loss Evaporation Latent heat removal via phase change 70 to 80 percent Drift Water droplets carried out by airflow 0.1 to 0.2 percent Blowdown Controlled discharge to manage dissolved solids 20 to 30 percent Blowdown is not optional. As water evaporates, dissolved minerals concentrate in the basin. Without controlled discharge, scaling and corrosion accelerate and destroy condenser tubes. Managing cycles of concentration is the core operational control variable, and it comes with a stack of costs and liabilities that stick to the facility for its whole life.
Chemical treatment
Biocides, scale inhibitors, and pH control on large facilities run 50,000 $/yr to 150,000 $/yr.Blowdown permits
Monitoring, reporting, and potential surcharges for thermal and chemical discharge to surface waters.Legionella liability
Warm tower basins are a documented Legionella habitat. This is a health liability no operator can afford to treat casually.Tiered water pricing
Municipal utilities in stressed regions are moving to tiered rates that specifically penalize high-volume industrial users. - 03 What it costs you The true cost of water-cooled operation
The ABC Carolinas industry reference cites a figure of roughly 110,000,000 gallons/yr for a medium-sized U.S. data center on open towers. Treat that as a ceiling-case illustration, not a planning number for your site. Water use efficiency varies by an order of magnitude across climates and IT load densities.
Loss breakdownWhere the water goes
Component What it is Share of loss Evaporation Latent heat removal via phase change 70 to 80 percent Drift Water droplets carried out by airflow 0.1 to 0.2 percent Blowdown Controlled discharge to manage dissolved solids 20 to 30 percent Blowdown is not optional. As water evaporates, dissolved minerals concentrate in the basin. Without controlled discharge, scaling and corrosion accelerate and destroy condenser tubes. Managing cycles of concentration is the core operational control variable, and it comes with a stack of costs and liabilities that stick to the facility for its whole life.
Chemical treatment
Biocides, scale inhibitors, and pH control on large facilities run 50,000 $/yr to 150,000 $/yr.Blowdown permits
Monitoring, reporting, and potential surcharges for thermal and chemical discharge to surface waters.Legionella liability
Warm tower basins are a documented Legionella habitat. This is a health liability no operator can afford to treat casually.Tiered water pricing
Municipal utilities in stressed regions are moving to tiered rates that specifically penalize high-volume industrial users. 304 The trap Headline water savings, hidden peak-day penaltyVendors selling water-free or advanced hybrid systems lead with a headline: 90% water reduction versus a conventional tower. That number is achievable in the right climate. The trap is that it does not tell you the full story on your peak day.
What the pitch says What the operator needs to know Water-free is always less efficient Annualized performance is often competitive in temperate climates once free cooling hours are counted. Peak summer is a different question than annual energy. Water-free is too expensive up front Eliminating tower basins, makeup water lines, chemical dosing, and blowdown handling offsets a real portion of the coil and fan premium. Run the actual capex, not the headline. Hot climates make water-free impractical Multi-row coils, appropriate approach temperatures, and limited hybrid assistance make it engineerable. It costs more, and you have to price that in. Water-free means maintenance-free It shifts the work from water chemistry to fan drives, coil cleaning, and refrigerant systems. Do not delete the O&M line from your model. This is where our episode on demand charges, Episode 1, becomes relevant to a cooling decision. When your dry coolers kick into high gear during a summer peak, that parasitic load does not just cost you in energy, it hits your demand charge too. Build the model with both.
- 04 The trap Headline water savings, hidden peak-day penalty
Vendors selling water-free or advanced hybrid systems lead with a headline: 90% water reduction versus a conventional tower. That number is achievable in the right climate. The trap is that it does not tell you the full story on your peak day.
What the pitch says What the operator needs to know Water-free is always less efficient Annualized performance is often competitive in temperate climates once free cooling hours are counted. Peak summer is a different question than annual energy. Water-free is too expensive up front Eliminating tower basins, makeup water lines, chemical dosing, and blowdown handling offsets a real portion of the coil and fan premium. Run the actual capex, not the headline. Hot climates make water-free impractical Multi-row coils, appropriate approach temperatures, and limited hybrid assistance make it engineerable. It costs more, and you have to price that in. Water-free means maintenance-free It shifts the work from water chemistry to fan drives, coil cleaning, and refrigerant systems. Do not delete the O&M line from your model. This is where our episode on demand charges, Episode 1, becomes relevant to a cooling decision. When your dry coolers kick into high gear during a summer peak, that parasitic load does not just cost you in energy, it hits your demand charge too. Build the model with both.
405 Your leverage Lifecycle math and the heat recovery layerThe lifecycle math is where the decision consolidates. The savings drivers are water and sewer charges, chemical treatment programs, blowdown permit compliance costs, and the maintenance labor open tower systems require. The number swings hard on your local water tariff, your climate file, and your chemical program cost. Build the model with your actual numbers before you accept anyone's lifecycle claim.
Ten megawatt data center: towers to dry coolersThis is an illustrative case from industry sources, not a planning number for your site. Use it to see the shape of the trade, then rebuild it with your own tariff and climate. Net annual savings on that illustrative case land between 135,000 $/yr and 275,000 $/yr. Over a lifecycle of 10 years to 20 years, cumulative savings of $2,000,000 to $5,000,000 can justify a CAPEX premium of $1,000,000 to $3,000,000.
Heat recoveryThe second revenue vector
Warm-water liquid cooling operates roughly between 104°F and 113°F. At those temperatures, recovered heat can serve domestic hot water, building heat, campus or district loads. A heat recovery chiller can produce chilled water for the facility while simultaneously raising recovered heat to a useful temperature for adjacent buildings. Thermal energy storage bridges the mismatch between continuous heat generation and variable heating demand.
We did a full episode on turning waste heat into watts using ORC systems and heat exchangers, Episode 31. This is the other side of that coin: what happens when you cannot recover the heat and must reject it. And in Episode 19 on CHP, we covered why on-site generation makes this rejection question larger, not smaller.
505 Your leverage Lifecycle math and the heat recovery layerThe lifecycle math is where the decision consolidates. The savings drivers are water and sewer charges, chemical treatment programs, blowdown permit compliance costs, and the maintenance labor open tower systems require. The number swings hard on your local water tariff, your climate file, and your chemical program cost. Build the model with your actual numbers before you accept anyone's lifecycle claim.
Ten megawatt data center: towers to dry coolersThis is an illustrative case from industry sources, not a planning number for your site. Use it to see the shape of the trade, then rebuild it with your own tariff and climate. Net annual savings on that illustrative case land between 135,000 $/yr and 275,000 $/yr. Over a lifecycle of 10 years to 20 years, cumulative savings of $2,000,000 to $5,000,000 can justify a CAPEX premium of $1,000,000 to $3,000,000.
Heat recoveryThe second revenue vector
Warm-water liquid cooling operates roughly between 104°F and 113°F. At those temperatures, recovered heat can serve domestic hot water, building heat, campus or district loads. A heat recovery chiller can produce chilled water for the facility while simultaneously raising recovered heat to a useful temperature for adjacent buildings. Thermal energy storage bridges the mismatch between continuous heat generation and variable heating demand.
We did a full episode on turning waste heat into watts using ORC systems and heat exchangers, Episode 31. This is the other side of that coin: what happens when you cannot recover the heat and must reject it. And in Episode 19 on CHP, we covered why on-site generation makes this rejection question larger, not smaller.
- Decision matrix
When water-free or hybrid pencils, and when it does not
✓ Lean water-free or advanced hybrid- Site is in a water-stressed region with tiered municipal pricing and permitting scrutiny
- Climate file shows a temperate profile with thousands of viable free cooling hours
- You have or can secure an adjacent heat load for a recovery layer
- Your ESG or client reporting requires demonstrable water reduction at facility level
- Fifteen to twenty year lifecycle model shows water and chemical savings offsetting energy delta
✗ Stay with evaporative, for now- Hot-humid climate with a large wet-bulb to dry-bulb spread and cheap, abundant water
- Load profile is dominated by summer peaks and site is on a punishing demand charge tariff
- No local water tariff pressure, no permitting risk, no ESG reporting requirement
- Retrofit into an existing site where basin, piping, and permit are already sunk and working
- Vendor has not produced a site-specific climate file or a design-day parasitic comparison
Questions for your morning huddle- Have we pulled a site-specific climate file and compared wet-bulb and dry-bulb profiles across all seasons, or did the engineering firm default to a system selection without that analysis?
- Do we know our all-in annual cost for chemical treatment, blowdown permit compliance, and water and sewer charges, and have we modeled what those costs look like at year ten under tiered pricing?
- Have we quantified the design-day parasitic fan load penalty, specifically the extra kilowatts hitting our coincident peak when dry cooling replaces evaporative capacity, and what that does to demand charge exposure?
- Have we evaluated whether a heat recovery chiller, thermal storage, or a campus distribution connection is viable at our site before committing to reject all this heat by default?
The one thing to rememberBottom line: your heat rejection architecture is set at schematic design and it sets your water, permit, and utility cost trajectory for the life of the facility. Do not accept a system selection that was not built on your site-specific climate file and your actual water tariff.
Before you sign off on the mechanical schematic, require your engineer to deliver three artifacts: an hourly climate file for the site, an all-in fifteen year lifecycle model using your actual water and chemical rates, and a design-day parasitic comparison showing peak fan load and demand charge impact under each architecture option.
6The Energy Decision BlueprintKnow if the numbers actually pencil out before you sign anything.
A written second opinion on the project in front of you, whether that is a rate change, new equipment, or a renewable installation.
- 01A short call, to figure out quickly whether we can actually be helpful. If we can't, we'll say so on the spot.
- 02We pull the data, your bills, your rate structure, vendor proposals, project specs.
- 03You get the verdict in writing: whether the payback will materialize, and the opportunities or risks nobody has raised.
Get a Blueprint at blueprint.tac-nrg.com Free for Indiana-based operations spending five figures or more a month on electricity. No obligation. You keep the write-up either way. - The one thing to remember
Bottom line: your heat rejection architecture is set at schematic design and it sets your water, permit, and utility cost trajectory for the life of the facility. Do not accept a system selection that was not built on your site-specific climate file and your actual water tariff.
Before you sign off on the mechanical schematic, require your engineer to deliver three artifacts: an hourly climate file for the site, an all-in fifteen year lifecycle model using your actual water and chemical rates, and a design-day parasitic comparison showing peak fan load and demand charge impact under each architecture option.
The Energy Decision BlueprintKnow if the numbers actually pencil out before you sign anything.
A written second opinion on the project in front of you, whether that is a rate change, new equipment, or a renewable installation.
- 01A short call, to figure out quickly whether we can actually be helpful. If we can't, we'll say so on the spot.
- 02We pull the data, your bills, your rate structure, vendor proposals, project specs.
- 03You get the verdict in writing: whether the payback will materialize, and the opportunities or risks nobody has raised.
Get a Blueprint at blueprint.tac-nrg.com Free for Indiana-based operations spending five figures or more a month on electricity. No obligation. You keep the write-up either way. 7Glossary- Heat rejection architecture
- The system that removes heat from a facility and moves it to the environment. Options include water-cooled towers, dry coolers, and hybrid designs that combine both.
- Approach temperature
- How close the leaving fluid temperature gets to the reference air temperature. A smaller approach means a colder fluid but a larger, more expensive coil.
- Wet-bulb temperature
- The temperature air would reach if cooled by evaporation to saturation. It is the reference an evaporative tower approaches, and it sits below dry-bulb in most climates.
- Free cooling hours
- Annual hours when ambient air is cool enough that the dry cooler can meet the load without mechanical refrigeration. Extended in temperate climates.
- Blowdown
- Controlled discharge of tower water to prevent dissolved minerals from concentrating to the point where scaling and corrosion damage condenser tubes and tower internals.
- Cycles of concentration
- The ratio of dissolved solids in circulating water to make-up water. Higher cycles mean more reuse and less water use, but greater scaling and fouling risk if unmanaged.
- Hybrid cooling system
- A system that switches between evaporative and dry modes based on ambient conditions, load, water cost, and energy price. Advanced designs operate mostly dry with rare wet assistance.
- Thermosyphon
- A passive refrigerant loop that circulates by density difference between vapor and liquid, reducing or eliminating compressor work in cooler climates.
- Heat recovery chiller
- A chiller configured to deliver chilled water for the facility while simultaneously raising recovered heat to a useful temperature for adjacent building, campus, or district heating loads.

