G-Value, Shear and Short-Circuiting: The Three-Way Tension
Per the source: flocculation neutralizes negatively charged particles with coagulants, then relies on gentle mixing to drive collisions. Success hinges on mixing intensity measured by the velocity gradient (G), on minimizing shear that breaks floc, and - beyond G - on energy distribution and flow dynamics inside the basin, which is why CFD has become the standard tool to evaluate hydraulics, find short-circuiting and model flocculator options.
Comparison frame (per the source): hydraulic baffles, paddle wheels, walking beams, bladed impellers and hyperboloid mixers, judged on maintenance, flexibility, adjustability, energy, redundancy, energy transfer, shear and floc quality.
Engineering implication (analysis): retrofit projects are boxed in by fixed basins - only the impeller head can change, which is precisely the design niche the hyperboloid shape targets.
Hydrofoil-Free Geometry: Radial Flow with VFD Control
Per the source: the hyperboloid body accelerates flow gradually along its ribs until particle velocity matches impeller speed, removing tip-shear step changes by geometry. Installed near the basin floor, it creates a radial flow pattern that sweeps the tank floor, prevents sediment accumulation and keeps floc suspended; variable frequency drives let operators adjust mixing intensity as raw water changes.
Engineering implication (analysis): radial flow plus floor sweeping attacks the two chronic flocculation problems at once - sludge blanketing (later releasing phosphorus and carbon, or punching through filters) and shear-induced floc breakage. VFD turns seasonal G-value retuning from a major change into a setpoint tweak.
Six-Plant Field Data
Per the source: 1) Annapolis, Maryland (2015): bladed impellers to hyperboloids - alum down 30%, less lime, iron removal improved 1.0 to 0.5 ppm, higher capacity in a smaller footprint, minimal maintenance and sediment, CFD-confirmed uniform mixing at all flows; 2) Bellevue, Ohio (2019): paddle wheels replaced - improved turbidity, quieter operation, and PAC dosing enabled at the floc stage; 3) Hickory, North Carolina (2013): one train retrofitted - less sediment buildup, turbidity 0.22 NTU vs 0.31 NTU with blades, better solids suspension per operators; 4) Atlanta, Georgia (Chattahoochee WTP): paddle-to-hyperboloid initially underperformed - CFD revealed basin short-circuiting, and after adding baffles turbidity reached 0.15-0.50 NTU vs 0.55-0.90 NTU in other trains; 5) Houston, Texas (2020): paddle wheels replaced - tank sludge from about 10 feet down to 1 inch (no direct cost comparison due to concurrent chemical changes); 6) Bzenek, Czech Republic (2007-2009): trial then full retrofit - optimal 10 rpm across all mixers, improved iron and manganese removal, lower turbidity and color, 82% energy reduction.
The Atlanta Lesson: CFD First, Baffles Second, Orders Third
The Chattahoochee case is the most instructive (analysis): a better impeller initially performed worse because the problem was basin hydraulics - influent momentum plus the mixer field created short-circuiting and starved floc contact time. CFD located it; baffles fixed it.
Retrofit rule (industry practice): large mechanical flocculation basins must be CFD-verified before ordering hardware - check inlet distribution uniformity, corner dead zones and outlet weir loading. VFD on a mixer is not a fix for flawed flow paths; only baffle geometry corrects those.
Verification cadence (industry practice): Houston's 1-inch vs 10-feet sludge contrast is the metric to copy - re-survey the basin floor three months after commissioning and back-calculate actual flow patterns against the CFD prediction to fine-tune speed staging.
Mapping to Chinese Standards and Adoption Advice
China mapping (analysis): GB 5749-2022 caps finished-water turbidity at 1 NTU (3 NTU where source and treatment constraints apply), with utility internal targets typically 0.3-0.5 NTU - Atlanta's post-retrofit 0.15-0.50 NTU sits inside that band.
Adoption advice (analysis): Chinese plant flocculation basins are mostly baffled channels or paddle-type mechanical basins; retrofits rarely touch civil works. The hyperboloid's floor-level radial flow suits deep basins and heavily blanketed ones. Before retrofit, verify the existing G-value staging per CJJ 58-2009 (industry-typical 20-70 s-1, decreasing stage by stage), and run CFD with local raw-water turbidity and cold-low-turbidity winter conditions - not North American datasets. Pre-order, pin down impeller metallurgy, gearbox brand and shaft-seal/bearing serviceability: these three dominate ten-year maintenance cost on mechanical flocculators.