Why cooling tower blowdown is the priority stream for a Guarulhos data center
A typical data center in Guarulhos, Brazil, on SABESP-supplied makeup water and operating an evaporative cooling tower near 4 cycles of concentration, will discharge roughly 25–30% of its makeup as cooling tower blowdown (CTBD), making CTBD the single largest recoverable wastewater stream on site (Genesis Water Tech, 2026). For facilities in the 1–10 million-gallon-per-month class, that translates to 2.5–3 million gallons of blowdown per month that the treatment train must handle (Genesis Water Tech, 2026). At the 100 MW class, daily makeup demand for open-loop evaporative cooling can reach ~2 million liters, of which CTBD is the engineered loss point (IDE Tech, 2026).
CTBD chemistry is the reason it sits at the top of the waste hierarchy. Blowdown carries 1,200–6,000 mg/L of total dissolved solids, 4–8× the makeup concentration, plus concentrated calcium, magnesium, silica, and alkalinity, residual biocides, corrosion and scale inhibitors, 10–50 mg/L of suspended solids, and a steady biological load (Genesis Water Tech, 2026). This combination of dissolved salts, scaling minerals, and treatment chemicals is what drives the choice of pretreatment, membrane selection, and downstream polishing, and it is the chemistry any RFQ must address first.
The Alto Tietê basin context makes reuse a non-negotiable engineering question. Sabesp's industrial supply to the São Paulo metropolitan region already serves very large constant demand; adding a hyperscale load onto a stressed watershed changes the permitting and water-stewardship conversation. Macro-basin data must be requested from the local water utility and environmental authority at project stage, because generic water-stress percentages do not translate directly to a specific sub-basin. The relevant comparable case studies for this design pattern are documented in a parallel data-center cooling water sustainability guide and in the broader CTBD engineering literature.
CTBD water balance and reuse targets for a Guarulhos site
The cooling-tower water balance starts with four flows: makeup, evaporation, drift, and blowdown. At 4 cycles of concentration (COC), blowdown is roughly 25–30% of makeup; pushing COC to 6–8 cuts blowdown volume sharply but raises TDS and the scaling load on the cooling water chemistry (Genesis Water Tech, 2026). The water balance table below is the working envelope an engineer should pin down before sizing any unit operation.
| Parameter | Typical range | Source |
|---|---|---|
| Cycles of concentration | 4 (baseline) → 6–8 (high-efficiency mode) | Genesis Water Tech, 2026 |
| Blowdown fraction of makeup | 25–30% at 4 COC | Genesis Water Tech, 2026 |
| CTBD TDS | 1,200–6,000 mg/L | Genesis Water Tech, 2026 |
| CTBD suspended solids | 10–50 mg/L | Genesis Water Tech, 2026 |
| Monthly facility size | 10 million gallons (~37,850 m³) | Genesis Water Tech, 2026 |
| Monthly CTBD design flow | 2.5–3 million gallons (~9,460–11,360 m³) | Genesis Water Tech, 2026 |
| Makeup reuse recovery target | 60–85% of CTBD | Genesis Water Tech, 2026 |
Three reuse pathways apply in Brazil: return to the cooling tower as supplemental makeup, on-site non-potable use (landscape, equipment washdown), or controlled industrial discharge under Sabesp and CONAMA 430/2011 requirements. Cooling-tower makeup reuse is the highest economic value of the three because it directly displaces SABESP-supplied freshwater and avoids discharge surcharges (Genesis Water Tech, 2026). For a 10 million-gallon-per-month facility, the design envelope for any CTBD treatment train is therefore 2.5–3 million gallons per month of blowdown to be characterized, pretreated, and routed to a recovery step. The full logic for mapping these flows against site hydrology is laid out in the data-center cooling water sustainability guide.
Pretreatment train: multimedia filtration, side-stream filtration and UF
Pretreatment is the membrane-protection step, and the source water decides how aggressive it must be. Side-stream filtration at 1–5% of total circulation flow drops suspended solids to manageable levels for downstream membranes and is a prerequisite for stable high-COC operation (Genesis Water Tech, 2026). The capital envelope for a side-stream installation in a typical data-center class is US$50,000–200,000 with minimal operating expense beyond solids disposal (Genesis Water Tech, 2026).
Ultrafiltration ahead of RO is the standard production step for CTBD reuse. UF at 0.01–0.1 μm pore size, 10–30 psi, removes bacteria, colloids, suspended solids, and high-MW organics with 90–95% recovery, producing RO feed compliant with <10–15 micron SDI targets (Genesis Water Tech, 2026). HydropureWater supplies multi-media filters for SDI reduction ahead of RO and a UF pretreatment skid rated 2,000–40,000 L/h on 0.03 μm PVDF hollow-fiber membranes with automatic backwash and air scour, tolerant of up to 300 ppm turbidity on the feed side. Chemical conditioning matters as much as mechanical filtration: antiscalant selection must address silica, CaCO₃, and CaSO₄ together, because high-silica feed favors RO over nanofiltration (Genesis Water Tech, 2026). Without a coherent antiscalant program, recovery ceiling drops before scaling is even measured.
Reverse osmosis: the core recovery step for a Guarulhos CTBD plant
Reverse osmosis is the recovery workhorse. Conventional BWRO on CTBD delivers 50–85% recovery with 95–99% dissolved-solids removal, producing 10–50 mg/L TDS permeate suitable for direct cooling-tower makeup (Genesis Water Tech, 2026). Design pressures of 150–400 psi are required to overcome the osmotic pressure of the concentrate, and antiscalant injection is non-negotiable for any feed above ~2,000 mg/L TDS (Genesis Water Tech, 2026). A 50,000 GPD (≈189 m³/d) blowdown RO installs in the US$250,000–500,000 range, with US$1.50–3.00 per 1,000 gallons of operating cost including energy, chemicals, membrane replacement, and maintenance (Genesis Water Tech, 2026).
Conventional CTBD RO plateaus at 75–80% recovery because silica, calcium carbonate, and calcium sulfate hit scaling thresholds before osmotic limits (IDE Tech, 2026). Pushing past that ceiling requires a different design philosophy: controlled salt precipitation in a fluidized-bed reactor, plus dynamic RO cycling that alternates short production periods with high-velocity flushing to keep supersaturation below the crystallization induction point. In a published industrial installation, this approach ran at around 95% recovery with silica in the permeate near 1 mg/L, then blended the permeate back into the makeup stream (IDE Tech, 2026). The same logic lifts overall CTBD recovery into the 90–98% range without the complexity of multi-stage RO trains. Nanofiltration is a lower-pressure (75–150 psi) option when partial softening is sufficient and hardness is the limiting species rather than total TDS: NF achieves 70–85% recovery with permeate TDS around 30–50% of feed (Genesis Water Tech, 2026).
| RO design parameter | Conventional BWRO | Nanofiltration (NF) | High-recovery / closed-loop |
|---|---|---|---|
| Recovery | 50–85% (plateaus 75–80%) | 70–85% | 90–98% |
| Operating pressure | 150–400 psi | 75–150 psi | 150–400 psi with cycling |
| Permeate TDS | 10–50 mg/L | 30–50% of feed | <10 mg/L (after polishing) |
| Best fit | General CTBD reuse as makeup | Hardness-driven limits, lower energy | Discharge-restricted sites, partial ZLD |
| Cost indicator (50,000 GPD) | US$250,000–500,000 installed; US$1.50–3.00/kgal OPEX (Genesis Water Tech, 2026) | Lower CAPEX, lower energy | Higher complexity, lower blowdown volume |
HydropureWater industrial RO systems reach up to 95% recovery with PLC automation and minimal chemical consumption, and RO membrane elements plus FRP/stainless pressure vessels are stocked locally for replacement cycles. Sizing the RO requires three site-specific numbers that the integrator will request: actual makeup TDS, target COC after reuse, and the local discharge TDS limit.
Beyond RO: polishing, evaporation and partial ZLD for tight discharge sites
Where the cooling-tower chemistry or a downstream liquid-cooling loop requires very low conductivity, RO permeate can be polished with mixed-bed ion exchange, electrodeionization, or capacitive deionization. The supplied research does not enumerate specific CTBD polish targets, so the engineer should request the conductivity and silica limits the cooling chemistry vendor and any AI-server liquid-cooling specification actually require at the project stage.
For sites where discharge volume is the binding constraint, mechanical vapor compression (MVC) on the RO concentrate achieves 95–98% recovery with distillate below 10 mg/L TDS. CAPEX is US$1–3 million for 10,000–30,000 GPD, with energy consumption of 15–25 kWh per 1,000 US gallons of distillate (Genesis Water Tech, 2026). Full ZLD combining RO + evaporative concentration + crystallizer reaches 95–99% overall recovery with less than 1% solid waste; CAPEX US$3–8 million and US$5–15 per 1,000 gallons OPEX (Genesis Water Tech, 2026). For most Guarulhos sites, a partial ZLD that reduces discharge volume by 80–90% captures most reuse benefit at substantially lower cost than full ZLD, with the remaining brine managed by hauling or controlled disposal (Genesis Water Tech, 2026). Brine-handling equipment such as a high-efficiency sedimentation tank and a plate-frame filter press for dewatering the solids stream is typically specified alongside the thermal stage.
Reuse, discharge or ZLD: the Brazil-specific decision framework
The decision framework for a Guarulhos site collapses to a comparison of three value streams: water saved, discharge cost avoided, and CAPEX/OPEX amortized. Cooling-tower makeup reuse at 60–85% CTBD recovery is the highest-value path and aligns with water-stewardship expectations; landscape or equipment washdown reuse is the next tier when cooling chemistry cannot accept the recovered water (Genesis Water Tech, 2026). Discharge-only treatment is justified primarily on avoided disposal fees. Direct discharge in water-stressed regions globally is reported in the US$5–15 per 1,000 gallons range (Genesis Water Tech, 2026); the current Sabesp industrial tariff and CETESB permit conditions for the specific discharge point must be requested at the project stage rather than assumed from this benchmark. Some jurisdictions enforce a discharge TDS limit below 1,500 mg/L, which effectively prohibits untreated CTBD disposal and forces at least TDS reduction or full reuse (Genesis Water Tech, 2026). CONAMA Resolution 430/2011 and CETESB Decision 846/2020 govern industrial effluent in São Paulo state and should anchor any discharge permit conversation; the supplied research does not enumerate numeric limits, so these are items to confirm with the local regulator.
| Pathway | Recovery | When it pays | Key dependencies |
|---|---|---|---|
| Cooling-tower makeup reuse | 60–85% of CTBD | Default for new builds; displaces SABESP makeup and avoids discharge fees | CTBD chemistry compatible with tower, antiscalant program, RO sizing |
| On-site non-potable reuse | Site-dependent | Where cooling chemistry rejects recovered water | Local reuse rules, irrigation demand profile |
| Discharge-only treatment | Volume reduction only | Where reuse is not feasible and discharge fees are the binding cost | CETESB/Sabesp permit, TDS and contaminant limits |
| Partial ZLD | 80–90% volume reduction | Where discharge volume is constrained but full ZLD economics fail | Brine hauling or controlled disposal route |
| Full ZLD | 95–99% | Where discharge is prohibited or freshwater supply is physically constrained | US$3–8 M CAPEX, US$5–15/kgal OPEX (Genesis Water Tech, 2026) |
Process support equipment, such as PLC-controlled antiscalant dosing and a UV sterilizer for biological control on the reuse loop, should be specified in the same RFQ package as the RO and any downstream thermal stage.
Frequently Asked Questions
What CTBD treatment train should a Guarulhos data center put in an RFQ?
Specify pretreatment with side-stream filtration and a multi-media filter for SDI reduction, followed by a UF pretreatment skid at 0.01–0.1 μm, then BWRO designed for 75–80% conventional recovery (with optional high-recovery architecture to reach 90–98%), and reserve MVC or partial ZLD for sites where discharge volume is binding. The reference design logic for each unit operation is documented in the industrial RO system specification.
What is the realistic budget envelope for a CTBD treatment train?
For a 50,000 GPD (≈189 m³/d) blowdown RO, installed CAPEX is US$250,000–500,000 with US$1.50–3.00 per 1,000 gallons OPEX (Genesis Water Tech, 2026). MVC for 10,000–30,000 GPD adds US$1–3 million CAPEX, and full ZLD adds US$3–8 million CAPEX and US$5–15 per 1,000 gallons OPEX (Genesis Water Tech, 2026). For a Guarulhos project, the buyer must also request the current Sabesp industrial tariff and any CETESB-specific effluent charges to complete the operating-cost picture; these are not published as a generic figure in the research.
Which Brazilian permits and standards govern CTBD discharge in Guarulhos?
CONAMA Resolution 430/2011 and CETESB Decision 846/2020 govern industrial effluent in São Paulo state, and Sabesp industrial discharge rules apply at the collection point. The supplied research does not enumerate the numeric effluent limits that apply to a specific discharge, so the buyer must confirm the current TDS, total suspended solids, biocide residual, and temperature limits with CETESB and Sabesp before sizing the polishing or ZLD stage.
How do I select a Brazilian integrator versus importing equipment?
Compare integrators on three inputs that the research does not provide as a benchmark: local references on RO or ZLD systems in the Alto Tietê basin, lead time on FRP/stainless pressure vessels and membrane elements, and demonstrated experience obtaining CETESB operational permits. Request a list of installed CTBD or ZLD systems in São Paulo state within the last 24 months, plus documented recovery and discharge results, before committing the RFQ. Local stocking of RO membrane elements and consumables is a measurable proxy for ongoing service capability.
