What a Zapopan data center is actually discharging
Cooling-tower blowdown is the largest single waste stream a Zapopan data center will produce, and its volume tracks directly with cycles of concentration. At 4 cycles, about 25–30% of makeup water is lost as blowdown; for a 10 MGD facility that is 2.5–3 MGD of waste stream (genesiswatertech.com, 2026-02). Pushing toward higher cycles shrinks the volume but sharpens the quality problem: blowdown TDS runs 4–8× the makeup, typically 1,200–6,000 mg/L, and carries 10–50 mg/L of suspended solids, accumulated calcium, magnesium, silica, and alkalinity, plus treatment chemicals — biocides, scale inhibitors, corrosion inhibitors, and dispersants that the cooling program cannot fully strip out (genesiswatertech.com, 2026-02). Legacy phosphate or chromate chemistries concentrate further and complicate any reuse scheme.
Globally, 75–90% of data centers use water-based cooling, and roughly 57% draw from potable supplies; an estimated 80–90% of that water comes from the same watersheds serving nearby residents (UGA CAES Field Report TP121, 2026-05). That overlap is the operational risk in Zapopan: a Guadalajara dry cycle reduces the public supply the data center shares with the metro, while the discharge stream it sends to drain becomes harder to permit. The supplied research does not include a published Zapopan or Guadalajara tap-water baseline; the buyer should request a current site water analysis covering hardness, alkalinity, chloride, sulfate, and silica before finalizing any design. Treat the local profile as qualitatively hard and mineralized with a seasonal silica swing rather than rely on assumed values.
Why blowdown cannot be dumped to drain in 2026
Discharge economics have changed. Direct discharge fees in water-stressed regions now run $5–15 per 1,000 gallons, and some jurisdictions enforce blowdown TDS limits below 1,500 mg/L — a ceiling that effectively rules out untreated discharge of any cooling-tower blowdown (genesiswatertech.com, 2026-02). Under Mexico's federal framework, the binding local standard is NOM-001-SEMARNAT, which establishes discharge limits for industrial wastewater into municipal sewer or receiving waters; the supplied research confirms the standard's relevance but does not enumerate a current Zapopan-specific numeric table, so a buyer should request the latest NOM-001 limits from the permitting authority rather than rely on quoted values. The permit scope must also cover anti-corrosion chemicals, scale inhibitors, biocides, chlorine compounds, and any glycols carried in the blowdown, with special handling for F-gases and PFAS-bearing coolants that some sites still operate (UGA CAES Field Report TP121, 2026-05).
Chemistry is only half the compliance picture. Thermal discharge from a cooling loop can raise receiving-water temperature and stress aquatic life if sent to a surface-water body, which adds a second compliance vector beyond TDS and metals (UGA CAES Field Report TP121, 2026-05). EPA-style guidance cited in the same source notes that cooling-system blowdown may require wastewater permits before any discharge; the Mexican analogue is the NOM-001 discharge permit. Together those three vectors — chemistry, thermal, and volume — are why a 2026 design for a Zapopan hall treats the train as compliance-driven infrastructure, not optional sustainability.
The 2026 reuse targets a Zapopan cooling loop should hit
Cooling-tower makeup reuse is the highest-value end use: it accepts a controlled envelope, not ultrapure water, and it short-circuits the discharge problem. The reuse envelope is 500–1,500 mg/L TDS, suspended solids held to 10–25 mg/L, hardness and silica controlled to the cooling program, and biological control maintained with low-fouling chemistry (genesiswatertech.com, 2026-02 and 2026-03). Three end-use strategies exist — cooling makeup reuse, process or irrigation reuse, and discharge compliance — and the cooling makeup path typically delivers 60–85% recovery, the highest value per cubic meter (genesiswatertech.com, 2026-02).
The recovery ceiling is set by RO concentrate chemistry. Permeate at 10–50 mg/L TDS is essentially demineralized, so the designer blends permeate with raw makeup to land inside the 500–1,500 mg/L envelope (genesiswatertech.com, 2026-02). Cycles of concentration must be aligned with local makeup hardness and silica to avoid membrane scaling; the supplied research confirms the link between cycles, scaling risk, and recovery but does not give a Zapopan-specific cycles ceiling, so the figure should come from a site water analysis rather than a quoted rule of thumb.
| Parameter | Cooling-tower makeup target | Blowdown feed (typical) | RO permeate |
|---|---|---|---|
| TDS (mg/L) | 500–1,500 | 1,200–6,000 | 10–50 |
| Suspended solids (mg/L) | 10–25 | 10–50 | <1 |
| Hardness / scaling ions | Controlled by program | Concentrated Ca, Mg, silica | 95–99% removed |
| Biological control | Low-fouling chemistry | Biocides, biofilm fragments | UF + chemical handles |
| Recovery vs. blowdown feed | — | — | 50–85% (RO) |
Pretreatment: side-stream filtration, softening, and UF
Skipping pretreatment to save budget is the most common failure mode in blowdown recovery. Self-cleaning spiral side-stream filters at 10–25 μm, sized at 1–5% of circulation flow, drop suspended solids in the blowdown to levels the downstream membranes can handle; CAPEX sits at $50,000–200,000 for typical data center installations, with minimal OPEX beyond solids disposal (genesiswatertech.com, 2026-02). Adding a biological flocculant such as a Zeoturb-style bio-organic polymer improves colloidal capture, which matters when the design targets higher cycles.
Industrial softeners protect the RO membranes from Ca/Mg scaling and let the system push higher recovery. A twin-tank industrial softener at the 1–45 T/h class is the standard pairing with antiscalant chemistry ahead of the RO. Dissolved air flotation is the right tool when the incoming wastewater carries fats, oils, grease, colloids, or oil-in-water emulsions — feeds that show up in mixed industrial-park streams feeding a co-located makeup line; a DAF unit handles 4–300 m³/h with automatic skimming.
UF then becomes the membrane-protecting barrier. Operating at 0.01–0.1 μm and 10–30 psi, a UF system removes bacteria, viruses, and high-MW organics at 90–95% recovery, with permeate backwash every 1–3 months (genesiswatertech.com, 2026-02). For a Zapopan build the sequence is non-negotiable: side-stream filtration first to drop particulates, softening to drop hardness, then a UF system as the final membrane shield, with PLC-controlled antiscalant dosing trimmed to the site's silica and alkalinity profile.
The heart of the train: RO (and where NF earns its place)
RO is the workhorse for a 2026 Zapopan build. It removes 95–99% of dissolved solids, hardness, silica, and most treatment chemicals at 150–400 psi with 50–85% recovery. A 50,000 GPD blowdown RO skid lands at $250,000–500,000 installed, with OPEX of $1.50–3.00 per 1,000 gallons including energy, chemicals, membrane replacement, and maintenance (genesiswatertech.com, 2026-02). Recovery is bounded by scaling potential as concentrate TDS climbs, which is why antiscalant choice and pretreatment discipline drive the actual number a site will hold.
NF earns its place when hardness and sulfate are the limiting ions rather than total TDS — typical of a partial-softening reuse scheme. NF runs at 75–150 psi, recovers 70–85%, and produces permeate at roughly 30–50% of feed TDS, which lets the designer hold higher cycles without committing to a full RO energy load (genesiswatertech.com, 2026-02). UF alone is acceptable only when the objective is biological and particulate control at moderate cycles; RO or NF should be reserved for higher cycles or stricter reuse. Membrane protection is non-negotiable: feed must be filtered below 10–15 μm, pH-adjusted, and dosed with antiscalant, with catalytic or electrochemical pretreatment such as GCAT-style approaches available to lower chemical consumption while protecting the industrial RO system.
| Technology | Pressure | Recovery | Permeate quality | Best fit |
|---|---|---|---|---|
| UF | 10–30 psi | 90–95% | Particulate-free, dissolved salts pass | Pretreatment to RO; standalone for biological control |
| NF | 75–150 psi | 70–85% | 30–50% of feed TDS | Partial softening, hardness-limited feeds |
| RO | 150–400 psi | 50–85% | 10–50 mg/L TDS | High-silica feeds, full reuse envelope |
When Zapopan should go further: evaporative concentration and ZLD
Once the membrane train hits its osmotic or scaling ceiling, evaporative concentration takes over. Mechanical vapor compression compresses vapor to drive evaporation, producing distillate below 10 mg/L TDS at 95–98% recovery of the RO concentrate; CAPEX is $1–3M for 10,000–30,000 GPD, with OPEX of 15–25 kWh per 1,000 gallons of distillate (genesiswatertech.com, 2026-02). A Zapopan-class train typically pairs RO at 50–75% recovery with MVC on the concentrate at 95% recovery of that stream, reaching 85–95% overall recovery with minimal liquid discharge.
Full ZLD adds a crystallizer that converts the 20–30% TDS MVC brine to solid salt cake. ZLD CAPEX is $3–8M and OPEX $5–15 per 1,000 gallons, justified only when alternative supply is unavailable or discharge is forbidden (genesiswatertech.com, 2026-02). Partial ZLD — 80–90% volume reduction with the remaining concentrate deep-well injected, hauled, or discharged under a special permit — captures most of the reuse economics at a fraction of full ZLD cost. Evaporation ponds are a Zapopan-relevant option because Guadalajara's dry season delivers high solar evaporation rates and industrial-park land is available; the choice is site-specific and requires geotechnical and stormwater review before committing.
Decision framework for a Zapopan 2026 build
The choice collapses into four buckets keyed to cycles of concentration and the reuse target. Bucket 1 — Reuse as cooling makeup only — uses side-stream filtration, a twin-tank industrial softener, UF, and RO, targets 60–85% blowdown recovery, and delivers the lowest CAPEX and the fastest payback in water-cost savings. Bucket 2 — Reuse plus partial ZLD — adds MVC on the RO concentrate for 85–95% overall recovery, sensible when discharge permits are restrictive or the site wants to future-proof against tighter NOM-001 enforcement. Bucket 3 — Full ZLD — combines RO, MVC, and crystallizer for 95–99% recovery, justified only when alternative supply is unavailable or haul-away cost exceeds $5–15/kgal OPEX. Bucket 4 — Makeup substitution — co-locates with a Guadalajara industrial park and accepts treated municipal or industrial wastewater (typically 600–1,200 mg/L TDS after secondary treatment) as makeup, leveraging the 30–50% cost gap versus potable supply documented in mature reuse markets (genesiswatertech.com, 2026-03; Open Engineering Inc., 2026).
The cost lens is consistent across the buckets: a 50,000 GPD RO skid at $250,000–500,000 installed and $1.50–3.00/kgal OPEX; ZLD escalating to $3–8M and $5–15/kgal (genesiswatertech.com, 2026-02). These are benchmarks, not quotes — the supplied research does not include a Zapopan labor, energy, or haulage rate, so a buyer should request a site-specific number before locking CAPEX. For a parallel view of the same question in a different water-stress context, the data center blowdown treatment in Khulna guide walks through a comparable train in Bangladesh, and the Osaka data center blowdown guide shows a higher-purity reuse envelope; the Guadalajara industrial wastewater 2026 guide covers the local permitting ground for the Bucket 4 option. A multistage filter ahead of the RO and a UV sterilizer on the reuse line are common add-ons in the tighter reuse envelopes.
| Bucket | Train | Overall recovery | CAPEX (50,000 GPD class) | OPEX ($/kgal) | Trigger |
|---|---|---|---|---|---|
| 1. Reuse as cooling makeup | Side-stream + softener + UF + RO | 60–85% | $300,000–700,000 | 1.50–3.00 | Default for new Zapopan hall on municipal supply |
| 2. Reuse + partial ZLD | RO + MVC on concentrate | 85–95% | $1.3–3.7M | 2.50–5.00 | Restrictive permit, dry-cycle risk |
| 3. Full ZLD | RO + MVC + crystallizer | 95–99% | $3–8M | 5.00–15.00 | Discharge forbidden, no alternative supply |
| 4. Makeup substitution | UF + RO on reclaimed water | 70–90% (on-site) | Site-dependent | 30–50% below potable | Industrial-park co-location |
Frequently Asked Questions
What budget envelope should a 5–10 MW Zapopan data center plan for a reuse-capable blowdown train in 2026?
Budget against the published RO benchmarks: a 50,000 GPD blowdown RO skid sits at $250,000–500,000 installed with OPEX of $1.50–3.00 per 1,000 gallons (genesiswatertech.com, 2026-02). Add $50,000–200,000 for side-stream filtration and the softener–UF–antiscalant dosing stack, and escalate only into the $1–3M MVC range when the Bucket 2 trigger is met. The supplied research does not include a Zapopan labor, energy, or haulage rate, so the buyer should request a site-specific installation quote rather than rely on a generic percentage markup.
How long does a Zapopan blowdown treatment train take to build, and what should a supplier be asked to prove?
Lead time is dominated by the membrane skids and the MVC or crystallizer packages, not the civil work. A buyer should require the supplier to document RO recovery under the site's actual cycles-of-concentration target, OPEX per 1,000 gallons at that recovery, and a reference list of comparable blowdown installations. For the SLA framing on similar builds in other water-stress markets, the packaged vs cast-in-place blowdown STP comparison is a useful procurement lens.
What cycles of concentration should a Zapopan cooling loop target to make reuse pay back?
Hold 4–6 cycles as the working band once the side-stream filter and softener are online; below 4 cycles, blowdown volume and discharge fees dominate the OPEX, and above 6 cycles the silica and hardness scaling risk on the RO membranes rises without antiscalant tuning (genesiswatertech.com, 2026-02). The supplied research does not give a Zapopan-specific cycles ceiling, so the upper bound should be set from a site water analysis covering silica, calcium, magnesium, and alkalinity.
Does NOM-001-SEMARNAT allow treated blowdown to be discharged to the Zapopan municipal sewer?
NOM-001-SEMARNAT is the binding federal discharge standard for industrial wastewater in Mexico, including cooling-tower blowdown with its biocides, scale inhibitors, and corrosion chemicals (UGA CAES Field Report TP121, 2026-05). The supplied research confirms NOM-001's relevance but does not enumerate the current numeric limits; the buyer should request the latest permissible discharge table from the Zapopan permitting authority and confirm the site's specific discharge permit conditions before sizing the Bucket 1 reuse line.
Related Equipment
- UF system — specifications, capacity range, and technical data
- industrial RO system — specifications, capacity range, and technical data
- twin-tank industrial softener — specifications, capacity range, and technical data
- PLC-controlled antiscalant dosing — specifications, capacity range, and technical data
