How a Horizontal Flow Section Cut Settled Water Turbidity from 6.15 to 1.3 NTU — Full-Scale Tube Settler Retrofit Case Study

A county waterworks built in 1989. Raw water at 0–5°C for five months a year. A treatment train that couldn't break 3 NTU. The plant's engineers made one structural change to their inclined tube settler - adding a horizontal flow rectification section occupying just 1/3 of the tank length - and watched settled water turbidity drop from 6.15 NTU to below 2 NTU within 72 hours of startup. Three years later, the system was still delivering 1.3 NTU average with zero modification to the tank shell. If you're running an inclined tube settler that underperforms in cold water or struggles with shock loads, this case study shows exactly how the fix works - and why it's simpler than you think.

The Challenge: New Standards, Old Plant, Cold Water

When ≤1 NTU Is the Law, and Your Plant Was Built for ≤3 NTU

The county waterworks sits in northern Shaanxi, a region where winter water temperatures routinely drop to 0–5°C and raw water turbidity swings between 15 and 77 NTU depending on the season. The plant was commissioned in 1989 with a conventional process: folded-plate flocculation, inclined tube sedimentation, and gravity valveless filters. Two parallel trains, each rated at 6,000 m³/d, for a total capacity of 12,000 m³/d. For two decades, this was adequate. The finished water sat around 3 NTU - acceptable under the old standard.

Then the regulations changed. The new drinking water sanitation standards raised the bar from ≤3 NTU to ≤1 NTU. The existing process couldn't close that gap. The waterworks faced a decision every aging plant eventually confronts: demolish and rebuild, or retrofit within the existing footprint. The latter was the only realistic option. The treatment structures were housed indoors for thermal insulation - a necessity in northern Shaanxi's climate, where outdoor tanks would freeze. Tearing down walls to expand wasn't on the table. They had to make what they had work better.

The target: increase single-train capacity from 6,000 m³/d to 10,000 m³/d (total plant output: 16,000 m³/d), with settled water turbidity consistently below 3 NTU to give the filters a fighting chance at the ≤1 NTU finished-water requirement.

The Retrofit Strategy: Three Upgrades, One Principle

The engineering team didn't touch the tank shell. They changed the process inside it, converting the old "folded-plate flocculation + inclined tube sedimentation + gravity valveless filter" into "contact flocculation + improved inclined tube settler + gravity valveless filter". Three upgrades drove the transformation:

1. Multi-Tube Mixer Replaces Tubular Static Mixer

The original tubular static mixer did one thing: it mixed. The replacement multi-tube mixer does three things simultaneously: cutting and splitting flow, reverse reflux, and rapid mixing. When water passes through the multi-tube geometry, high-frequency eddy currents form at the pipe wall boundary layer. These eddies create micro-zones of intense shear that distribute coagulant evenly across the entire flow cross-section within seconds. The measured mixing efficiency exceeds 94%, and the head loss is lower than the original static mixer - an unusual combination that means better mixing with less energy.

The practical benefit: 20–30% chemical savings compared to the original mixer, with better adaptability to flow rate and water quality variations. For a plant that sees raw water turbidity swing from 15 to 77 NTU seasonally, that adaptability matters.

2. Grid Flocculation with Star-Shaped Vanes Replaces Folded-Plate Flocculation

Folded-plate flocculation works by forcing water through a serpentine path between parallel plates. It creates turbulence, but not the right kind. The improved design uses star-shaped vanes mounted perpendicular to the flow direction inside grid compartments. These vanes generate high-frequency micro-vortexes - turbulence at a scale that promotes particle collision without shearing apart flocs that have already formed.

The flocculation tank is divided into 11 compartments across two reaction stages. Stage 1: 8 compartments at 1.5 m × 1.5 m each, 8.95-minute retention. Stage 2: 3 compartments at 1.5 m × 2.0 m each, 4.48-minute retention. Total hydraulic retention time: 13.43 minutes at a GT value of 1.7 × 10⁴. The two-stage design creates denser, more shear-resistant flocs than single-stage flocculation - a critical advantage when those flocs later enter the tube settler's distribution zone, where excessive turbulence would break them apart.

3. The Key Innovation: Horizontal Flow Rectification Section in Front of the Tube Settler

This is where the retrofit departs from standard practice, and where the largest performance gain was achieved. A traditional inclined tube settler receives water directly from the flocculation tank outlet. The transition is abrupt: water goes from a turbulent, mixed-flow regime into a laminar settling regime with no intermediate conditioning. The result is uneven flow distribution across the tube bundle. Some channels see high velocity and flush flocs straight through. Others see low velocity and accumulate sludge. Both conditions degrade effluent quality.

The retrofit added two elements that solve this:

A submerged weir at the flocculation tank outlet. Instead of water dumping directly into the distribution zone, it first flows over a submerged weir that acts as a flow rectifier - smoothing out velocity gradients and dissipating large-scale turbulence before the water enters the settler.

A horizontal flow section occupying 1/3 of the total tank length (24 m²). This is the innovation. After passing over the weir, water enters a horizontal-flow zone before rising through the inclined tubes. Guide walls in both the horizontal section and the inclined tube section control the flow path. The horizontal zone serves three functions simultaneously: it reduces the horizontal flow velocity to 12.73 mm/s (within the 10–18 mm/s window that research shows is critical for stable sedimentation), it provides additional settling time for larger flocs, and it acts as a hydraulic buffer that absorbs flow fluctuations before they reach the tube bundle.

The distribution wall at the end of the transition section uses perforations with wider spacing at the top and narrower spacing at the bottom. This counter-intuitive arrangement is deliberate: it compensates for the natural tendency of water to short-circuit toward the top of the distribution zone, forcing more uniform vertical flow distribution across the full height of the tube bundle.

Why the Horizontal Flow Section Works: The Engineering Rationale

The problem with traditional inclined tube settlers is not the tubes themselves. It's what happens before the water reaches them. Research has identified four parameters that govern flow distribution uniformity in tube settlers: the tank length-to-width ratio (L/B), surface loading rate (q), tube diameter (D), and the height of the distribution zone below the tubes.

Of these, L/B ratio and surface loading (q) have the greatest influence. As the L/B ratio increases, flow distribution becomes more non-uniform, and the critical settling velocity - the minimum velocity at which a particle of a given size settles - rises. Surface loading (q) is roughly proportional to critical settling velocity: push more flow through the same area, and you need a higher settling velocity to capture the same particles. Both effects work against effluent quality.

Traditional tube settlers also suffer from a perforated-wall problem. To achieve uniform distribution, the perforated wall at the inlet needs smaller openings than a horizontal-flow sedimentation tank - which means higher orifice velocities. High-velocity jets through these orifices break apart flocs that the flocculation tank just spent 13 minutes building. Worse, the jets can stir up dead sludge deposited at the bottom of the distribution zone, re-suspending solids that had already settled.

The horizontal flow section solves this chain of problems at its root. By adding 1/3 of the tank length as a horizontal-flow zone before the inclined tubes, the design achieves four things:

Design Parameters: The Numbers Behind the Retrofit

The inclined tube section uses counter-current flow - water rises upward through the tubes while solids slide downward along the 60° incline. The tubes are ethylene-propylene copolymer (EPC) with honeycomb hexagonal cross-sections, 30 mm diameter, 1 m length. EPC was chosen over standard PVC or PP for its superior performance in cold water: it maintains flexibility and impact resistance at temperatures where rigid PVC becomes brittle.

Sludge collection uses large gravity hoppers with a slide angle steeper than conventional small hoppers, ensuring thorough sludge removal with minimal local flow disturbance. The sludge discharge valves are manual-electric quick-opening type - a practical choice that prioritises reliability and ease of maintenance over full automation. Fewer devices, lower failure rates, lower cost.

Results: 4 NTU Lower in 3 Days, 1.3 NTU After 3 Years

72-Hour Trial Run: Immediate Performance Gap

A three-day continuous trial was conducted from October 15–17, 2011, with the influent flow controlled at 750 m³/h. The plant sampled raw water, effluent from the unmodified traditional tube settler (parallel train), and effluent from the improved tube settler simultaneously throughout the trial.

The results were unambiguous. The improved settler's effluent turbidity was more than 4 NTU lower than the traditional design across all sampling points. The gap was consistent and immediate - no break-in period, no tuning required. The effluent quality met design expectations from the first day, substantially reducing the filter load downstream.

TOC Removal: 6.08% Improvement Over the Traditional Design

Turbidity and organic matter are closely correlated in surface water - the same particles that scatter light also carry adsorbed natural organic matter (NOM). During the trial, the plant measured TOC in the effluent from both sedimentation processes. The raw water TOC was stable at 2.5–3.0 mg/L.

A 6% improvement in TOC removal may seem modest, but in drinking water treatment, organic matter removal in the sedimentation stage directly reduces disinfection by-product (DBP) formation potential downstream. Every milligram of TOC removed before chlorination is a milligram that won't form trihalomethanes (THMs) or haloacetic acids (HAAs) in the distribution system.

Three-Year Verification: The Numbers Hold

The retrofit was commissioned on April 15, 2012. Over three years of continuous operation, the plant regularly sampled settled water turbidity from both the retrofitted train and the unmodified parallel train. The data tells a clear story of sustained performance:

The 1.3 NTU settled water average is the number that matters most. By delivering water at 1.3 NTU to the filters instead of 2.3 NTU, the improved settler reduced the filter's solids load by approximately 43%. This means longer filter runs between backwashes, less backwash water consumption, and - critically - finished water that can reliably meet the ≤1 NTU standard. A filter receiving 2.3 NTU settled water has very little margin to achieve 1 NTU finished water. A filter receiving 1.3 NTU settled water has room to work.

What This Means for Your Plant

This case study demonstrates that an underperforming inclined tube settler does not need to be replaced. The problem is rarely the tubes themselves - it's what happens in the distribution zone before the water reaches them. The retrofit principles are transferable:

1. If your settled water turbidity is inconsistent, check your flow distribution first. Uneven flow across the tube bundle is the most common cause of performance variation in inclined tube settlers. A submerged weir at the flocculation outlet and perforated distribution wall with graduated hole spacing are low-cost fixes that address this at the source.

2. If you treat cold water (<5°C), add a horizontal flow section. Cold water produces weak flocs that cannot survive the turbulence of a traditional tube settler inlet. The horizontal flow section provides a low-energy transition zone where these fragile flocs can settle without being sheared apart.

3. If your raw water quality swings seasonally, increase the distribution zone buffer. The horizontal section's ability to absorb shock loads is not a nice-to-have - it's the difference between stable effluent and seasonal water quality complaints. The 1/3 tank length allocation used in this project is a good starting point for design.

4. Tube material matters in cold climates. The EPC tubes chosen for this project maintain flexibility at temperatures where rigid PVC becomes brittle. If your raw water drops below 5°C for extended periods, discuss tube material selection with your supplier. Standard PVC and PP have different low-temperature performance envelopes.

5. A 6% improvement in TOC removal is worth more than it looks. In drinking water treatment, organic matter removed before disinfection is DBP formation potential eliminated. The 6.08% TOC improvement at the sedimentation stage translates to measurably lower THM and HAA formation downstream - a regulatory compliance advantage that compounds over the life of the plant.

The retrofit described here increased treatment capacity by 33% (from 12,000 to 16,000 m³/d) within the same tank footprint, while simultaneously improving effluent quality. That combination - more water, better water, same building - is what makes the horizontal flow rectification concept worth evaluating for any aging water treatment plant facing tighter standards.