From high‑efficiency drift eliminators to running towers at the highest safe cycles of concentration, mills are cutting makeup water without touching the cooling load. Some are eyeing treated POME as a future makeup source — but only with serious polishing.
Industry: Palm_Oil | Process: Cooling_Systems
Cooling towers are the palm mill’s biggest water habit. The EPA’s rule‑of‑thumb says about 1.8 gallons of water evaporate per ton‑hour of cooling (EPA WaterSense). That means a 1,000‑ton tower running for an hour flashes off roughly 1,800 gallons. In some industries, towers can represent up to ~95% of process water flow (ProChemTech). The conservation playbook focuses where it can: minimizing drift (unintended droplet carryover) and blowdown (controlled bleed to limit dissolved solids), and — with careful treatment — raising cycles of concentration (COC, the ratio of dissolved solids in recirculating water to makeup).
Alongside those operational levers, some mills are asking whether their own wastewater — treated palm oil mill effluent (POME) — could substitute for fresh makeup. The answer ranges from “promising in principle” to “only after extensive polishing,” given POME’s high organic and nutrient loads.
Evaporative baseline and loss pathways
Evaporation, which drives heat rejection, is locked to the cooling load; conservation targets the smaller losses. High‑efficiency drift eliminators can push drift down to a few parts per million of circulating flow (EPA WaterSense). Uncontrolled drift can be orders of magnitude higher — one estimate cites 0.24–0.36 gallons per ton‑hour (EPA WaterSense). Blowdown scales with COC: higher cycles mean less frequent bleed while still avoiding scale, corrosion, and biofouling.
High‑efficiency drift eliminators (DEs)
Modern DEs can limit drift to ≤0.005% of circulating flow (EPA WaterSense). On a 1,000 gpm system, 0.005% is just 0.05 gpm; compare that with 0.02% (0.2 gpm), a 75% cut in water loss. Upgrading from a 0.002% to a 0.0005% design — now common in advanced towers — similarly cuts drift flow by roughly 75%. Vistech Cooling sums up the mechanism simply: by returning droplets to the tower, DEs “aid in the conservation of water in the cooling system,” and DE efficiency “has a significant part to play in the water savings” (Vistech Cooling).
The arithmetic adds up. If a 1,000‑gpm tower with 0.002% drift (0.02 gpm loss) shifts to 0.0005% (0.005 gpm), annual drift loss falls from ~10,500 to ~2,600 gallons — a saving of ~7,900 gallons per year. Vendors sell DEs rated at 0.001–0.0005%; keeping panels clean and intact matters because even small cracks can spike drift beyond design (Vistech Cooling; EPA WaterSense). Lower drift also reduces off‑tower aerosol dispersion of chemicals such as biocides.
Cycles of concentration and blowdown control
Raising COC cuts blowdown and saves makeup water. In a 1,000‑ton case study using hard city water (~186 mg/L as CaCO₃), the numbers are stark (ProChemTech): at COC 2.2, blowdown ≈8.08 million gal/yr (makeup ≈17.77 M gal/yr); at COC 4.0, blowdown ≈3.23 M gal/yr (makeup 12.92 M gal/yr); at COC 10.0, blowdown ≈1.08 M gal/yr (makeup 10.77 M gal/yr). Moving from 2.2 to 10.0 cut fresh water use by ~39% per year (17.77 to 10.77 M gal) and eliminated ~87% of blowdown (ProChemTech; ProChemTech). The absolute saving was ~7.0×10^6 gallons/year (~26,500 m³/year) and roughly $79,000/year in combined water/effluent costs (ProChemTech). Even a moderate move from ~3 to 6 cycles typically cuts makeup by ~20–30% as the blowdown fraction drops from ~50% to ~20% of evaporative flow.
Controls and chemistry set the ceiling. An automated conductivity meter that triggers blowdown at the maximum allowable setpoint helps “determin[e] the maximum COC… your cooling tower can sustain without risk of scale” (EPA WaterSense). Conventional inhibitor programs (phosphonates plus polymers) typically hold 2–4 cycles on hard water; stronger programs (polycarboxylates, polyacrylates, organophosphonates) may allow 5–8 cycles before scale emerges. Many mills wrap this into a broader cooling‑tower chemical program, with targeted scale inhibitors protecting heat transfer surfaces.
Hardness removal and pretreatment options
Removing hardness is the most direct enabler. EPA‑cited data note that softening “allows COC to be increased without the potential for scale formation” (ProChemTech). In the 1,000‑ton example, an appropriately sized softener of ~41 gpm (3,370 lb resin) enabled stable operation at 10 cycles (ProChemTech; ProChemTech). Payback was rapid — a ~4‑month ROI on a capital of ≈$26,000, followed by ~$78,800/year lower operating cost (ProChemTech).
For higher purity makeup, reverse osmosis and ultrafiltration can produce near‑pure water (often used for boiler feed), enabling very high COC (e.g., >10) without scale. EPA guidance explicitly suggests considering RO or water conditioners on makeup to raise cycles (EPA WaterSense). In palm mills that pursue this path, brackish‑water RO and upstream ultrafiltration for pretreatment are the common station stops.
Treated POME as makeup water
POME is abundant. Typical mills generate ~1–2.5 m³ of POME per tonne of crude palm oil (Water Practice & Technology). One Ghanaian study found 62–89% of process water exited as POME (Water Practice & Technology). But raw POME is not cooling‑ready: COD often runs 50,000–150,000 mg/L, oil/grease ~2,000–5,000 mg/L, and suspended solids ~20,000–60,000 mg/L (Membranes). Its ammonia and nutrient content fuels intense microbial growth if untreated.
Experience with “graywater” is instructive: a Water Technology review notes that municipal effluent with “double‑digit ppm” ammonia/organics can trigger “explosive growth of organisms” in cooling systems unless thoroughly treated (Water Technology). EPA guidance, by contrast, highlights low‑TDS sources like condensate and rainwater as ideal internal makeups (EPA WaterSense; EPA WaterSense).
Treatment train and reuse feasibility
Where mills deploy advanced treatment, staged reuse is plausible. Tertiary bioreactors or membrane bioreactors (MBRs) can drive COD/BOD down to the low hundreds of mg/L and remove ammonia; this aligns with research that has targeted high‑end, membrane‑based polish to boiler‑feed standards (Membranes). In practice, that points to membrane bioreactors followed by granular filtration (e.g., sand filtration) and disinfection via ultraviolet systems. One Western case study, not palm‑specific, even considered reusing treated blowdown or effluent in towers and estimated ~13% reduction in water footprint (ScienceDirect).
Until such polishing is in place, operators typically avoid direct POME use in towers. Instead, POME is often tapped for onsite energy (biogas) or irrigation, while tower conservation comes from drift and COC gains. EPA’s reuse discussion underscores the preference for low‑nutrient internal sources first (EPA WaterSense; EPA WaterSense).
Regulatory and economic context
Indonesia’s framework (e.g., PP 82/2001) encourages lower fresh water intake and reuse where possible. But tertiary POME treatment — MBR, RO, or advanced oxidation — is capital‑ and operations‑intensive; an MBR system might cost several dollars per cubic meter treated, with added energy and chemical use. Saving even 1,000 m³/day of fresh water by reusing POME would likely require such systems. In many cases, the first tranche of savings is achieved more economically through drift control and higher COC — measures that have demonstrated multi‑year ROI (ProChemTech).
Quantified outcomes and sequencing
Facilities that sequence upgrades report clear gains: implement high‑efficiency DEs and maximize cycles within scaling limits first, then evaluate POME recycling as a longer‑term project. Globe‑scale data indicate that each 1% drift reduction or one additional cycle can save hundreds of cubic meters of water per year per MW of cooling. In the 1,000‑ton example, raising COC to 10 saved ~7.0×10^6 gallons/year (~26,500 m³/year) (ProChemTech; ProChemTech), while also trimming energy and chemical use (less pumping, less biocide).
If a large mill generates 2,000 m³/day of POME and can treat and reuse half, that’s ≈1,000 m³/day of fresh water offset (~365,000 m³/year) — but untreated POME is unfit for direct use because of fouling risks (Water Technology). Any reuse scheme must therefore invest in filtration, biological treatment, and disinfection — potentially including ultrafiltration and RO polishing if the target is high COC makeup (Membranes).
Vistech Cooling’s point about drift — that returning droplets “has a significant part to play in the water savings” — captures the spirit of these programs (Vistech Cooling). The rest is disciplined control of COC, with instrumentation (EPA WaterSense), inhibitors, and — where justified — pretreatment that removes hardness and organics.
