Refineries Are Turning Towers Into Water Savers

In the world of refinery utilities, the cooling tower is the biggest drinker on site. Industry guides note that in some sectors up to 95% of intake water passes through the tower (www.prochemtech.com), making it the prime target for conservation—especially in water‑stressed regions (e.g., Indonesia) where costs and regulations are rising.

The losses are well known: evaporation (≈1% of flow per 10°F cooling), bleed/blowdown to control dissolved solids, and drift (droplets carried out in exhaust air). Without modern eliminators, drift can run 0.05–0.2% of flow—about 0.24–0.36 gallons per ton‑hour of cooling (nepis.epa.gov). Even modest cuts matter: the EPA’s WaterSense guidance estimates that boosting cycles of concentration from 3 to 6 alone trims makeup demand by roughly 20% (nepis.epa.gov), and one study found that going from 6.5→9 cycles would save more than 1.1×10^6 m³ per year for the same duty (www.researchgate.net).

Cooling tower water balance

Evaporation, blowdown, and drift dominate tower water loss. Evaporation scales with heat rejection (≈1% of circulating flow per 10°F), while blowdown is intentionally discharged to cap dissolved solids. Drift—mechanical carry‑over—can be 0.05–0.2% of flow without effective eliminators (≈0.24–0.36 gal per ton‑hr) (nepis.epa.gov). Because these losses also waste treatment chemistry, refineries are pursuing interventions with measurable payback.

Simple setpoint changes can deliver outsized gains. EPA’s WaterSense guide shows that moving from 3 to 6 cycles of concentration cuts makeup by ~20% (nepis.epa.gov). In parallel, modern chemistry programs—such as targeted scale inhibitors—support higher cycles without scale.

High‑efficiency drift eliminators

Drift eliminators are structured baffles that force air and entrained droplets through turns so water falls back to the basin. High‑efficiency designs reduce drift to ≪0.005% of circulation flow (nepis.epa.gov), often cutting drift by 90–99% versus legacy parts.

One practical illustration: in a 200 m³/hr tower, an older eliminator at 0.002% drift loses ~4 L/h, while a high‑efficiency unit at 0.0005% loses ~1 L/h—about 26 m³ per year saved per tower in this example (www.vistechcooling.co.uk). Reducing drift also prevents loss of chemicals and minimizes environmental discharge, aiding compliance where regulators and legionella safety guidelines limit drift emissions (sources as above).

Operating at higher cycles

Cycles of concentration (COC) measure how many times salts in the circulating water are concentrated relative to makeup; higher COC reduce blowdown and cut freshwater use. EPA calculations show ~17–20% water savings by doubling cycles, with 3→6 cycles reducing makeup by ~20% (nepis.epa.gov).

Field data back this up. In a 1,000‑ton cooling tower, increasing COC from 2.2 to 10 cut makeup from 17.77×10^6 to 10.77×10^6 gallons per year—about a 39% reduction (www.prochemtech.com). Under ideal treatment, exceeding 10 COC is possible: a patented program (softening plus specialized inhibitors) routinely ran >10 COC with mild steel corrosion <0.5 mil/yr (www.prochemtech.com).

Pushing COC requires tight scale control. Calcium and other ions precipitate as scale when concentrated, so integrated treatment—phosphonate or polyelectrolyte inhibitors, pH control, and ion exchange softening—comes to the fore. A study that raised COC from 6.5 to 9 (saving 1.1×10^6 m³/yr) succeeded by lowering pH and adding Zn/mercaptobenzothiazole inhibitors to prevent orthophosphate/calcium precipitation (www.researchgate.net). Real‑time meters for conductivity and pH, plus incremental monitoring of hardness and Langelier Saturation Index (LSI), are advised (source as above). For stable control, refinery programs commonly deploy precise dosing pumps for pH and inhibitors and, where needed, softening systems.

Alternative makeup sources

Beyond efficiency, switching the water itself moves the needle. Refineries increasingly use non‑potable sources—treated effluent, stormwater, condensate, and internal waste streams—to replace freshwater makeup. Many reclaimed waters (e.g., municipal secondary effluent) carry higher hardness, alkalinity, phosphate, ammonia, or organics, raising scale and corrosion risk (www.powermag.com). Advanced treatments or chemicals can mitigate this; in one power‑plant study, high‑phosphate reclaimed water (13 ppm PO₄) ran at high cycles using a novel terpolymer inhibitor, avoiding costly softening or polishing (www.powermag.com). For pretreatment, utilities often turn to ultrafiltration ahead of RO trains when reclaiming effluents.

Cooling tower blowdown itself can be recycled. Pilot tests show treated blowdown reuse reducing net water footprint by about 13% (www.sciencedirect.com). A 750‑MW plant in Chile achieved 96% recovery of CT blowdown—versus ~55–60% for conventional RO—by pairing RO with a salt‑precipitation reactor (www.watertechonline.com). This approach nearly eliminates blowdown discharge: achieving 96% recovery means only 4% of makeup is fresh water, drastically cutting overall water use (sources as above). When blowdown is recycled as makeup, tower cycles can be pushed to chemical limits (since no discharge limits apply) (same sources). High‑recovery systems often integrate membrane packages matched with tower chemistry.

Internal refinery streams are in play too. Stripped sour water—process water with H₂S/NH₃ removed in a stripper—still contains some ammonia, organics, and salts, but can displace freshwater in lower‑grade services. At Saudi Aramco’s Ras Tanura refinery, stripped sour water quality was adequate to replace boiler‑feed water for equipment “wash” duties; a phased 25%→50% implementation delivered about 33% reuse of SSW and roughly $1 million in first‑year savings (www.watertechonline.com). Ras Tanura plans to use the remaining SSW as cooling water (projected 50% reuse) once ancillary issues are solved (www.watertechonline.com). Programs like these typically complement tower programs with cooling‑tower chemical control and targeted corrosion inhibitors.

Other sources include HVAC condensate (essentially distilled water) and rainwater. ASHRAE notes that air‑handler condensate can meet ~10–40% of tower makeup in humid climates and that roofs yield ~1 liter per mm of rain per m² of area (handbook.ashrae.org) (handbook.ashrae.org). Any alternative makeup should be screened for particulates and paired with biocide addition; many operators standardize on MBR‑quality reuse water or add biocides to maintain microbial control.

Outcomes and implementation

DOE/EPA guidance projects 20%–60% reductions in cooling‑tower makeup by lifting COC into the 6–10 range, with reliable drift eliminators slashing droplet losses to negligible levels (nepis.epa.gov) (nepis.epa.gov). Case studies align: the Ras Tanura refinery achieved ~33% reuse of sour water (with ~$1M savings/yr) (www.watertechonline.com) (www.watertechonline.com), and a power‑sector demo cut blowdown discharge by ~91% via 96% reuse (www.watertechonline.com).

Execution basics matter: meter all water flows, apply real‑time conductivity and pH control, and involve water‑treatment specialists. When blending or switching makeups, account for chemistry differences (LSI, silica, chlorides) to avoid hidden corrosion. The payoff—often millions of gallons and tens‑of‑thousands of dollars saved annually—makes the combination of high‑efficiency eliminators, aggressive COC operation, and reclaimed/process‑water makeup a data‑backed route to cutting freshwater demand while staying reliable (nepis.epa.gov) (www.watertechonline.com).