Secondary Clarifier Sizing: Four Envelopes, One SVI, and the Plant That Fails on the Fifth Check
Suggested readership: engineers who have just sized an aeration tank and now need the clarifier that keeps the sludge in the system, and utilities retrofitting an undersized clarifier. Reading time: ~6 min. Companion tools: Secondary clarifier · Rectangular secondary · Lamella retrofit
The clarifier is the real test of the aeration tank
Design offices spend weeks optimising aeration volume and then apply a single "surface overflow rate = 24 m³/m²·d" to the clarifier. The failure pattern this produces is so common it deserves a name: the plant that nitrifies beautifully at average flow and washes its sludge out every peak. A secondary clarifier is not a settling tank fed by a pump — it is a solids-handling machine receiving 61,250 kg/d of biomass on our 10,000 m³/d example, and it must return that solids inventory to the aeration tank faster than biology grows it.
That is why there is no single sizing number. There are four envelopes, and the governing one wins:
| Check | Design value | Actual (avg/peak) | Limit | Verdict |
|---|---|---|---|---|
| Surface overflow rate | avg 24 | 18.6 | OK | |
| Surface overflow rate | peak 48 | 27.9 | OK | |
| Solids loading rate | avg 120 | 113.9 | OK | |
| Solids loading rate | peak 240 | 146.5 | OK | |
| RAS concentration achievable | SVI 120 → 8,333 mg/L max | need 8,167 mg/L | 98 % of limit | governing |
The area comes out at 510 m², i.e. two 18.5 m diameter tanks — and it is the solids loading, not the overflow rate, that set it. That is normal for an SRT-20 nitrifying plant: the sludge is the load.
The SVI trap: you can only return what the sludge will settle to
Mass balance says the clarifier must deliver RAS at X_r = (1+R)/R × X = 8,167 mg/L at R = 75 %. The physics says the maximum is X_r,max = 10⁶/SVI — 8,333 mg/L at SVI 120. The design is sitting at 98 % of the physical ceiling. There is no slack.
Now the tropical trap closes. SVI is not a constant — it is a weather report on your biomass. In a 30 °C plant with a low F/M, SVI 120 can quietly become SVI 180 after a rain event or a nutrient shortfall, and X_r,max drops to 5,556 mg/L. The mass balance still demands 8,167. The clarifier cannot deliver it, the RAS concentration falls, the aeration-tank MLSS follows, and the plant de-ratings itself — usually discovered weeks later as a rising effluent ammonia.
This is why the SVI estimate deserves real scrutiny at design stage, and why an MLSS meter on the aeration tank plus a blanket-level sensor in the clarifier are not optional extras: they are the only instruments that show the R-to-SVI mismatch while it is still easy to fix.
Rectangular clarifiers: the velocity check most designers never run
Rectangular secondary clarifiers (chain-and-flight scrapers) are popular in Southeast Asia for shared-wall compactness. On a 40,000 m³/d peak duty, two 42.2 m × 10.5 m tanks deliver 889 m² — again solids-limited, not SOR-limited. But the rectangular geometry adds a check the circular tank does not have: horizontal velocity. Peak flow must not drag settled sludge toward the outlet. Our example runs 0.38 m/min against a 1.5 m/min limit — comfortable — but the check matters because a rectangular tank that fails it fails as a sludge conveyor, not as a settler, and the symptom (a dead zone at the inlet third) is invisible from the effluent weir.
Weir loading is the final rectangular weakness. Long tanks need long weirs: 84 m per tank lands at 237 m³/m·d on peak, just under the 250 limit. Push the same flow through shorter tanks and you will buy a second effluent trough to get back under the line.
When the answer is lamella plates — and when it is not
A tube/plate settler retrofit is the classic cure for an undersized clarifier. On a real case — an existing 250 m² clarifier designed for 10,000 m³/d — adding 60° inclined tubes multiplies effective area by 4.03×, taking theoretical peak capacity to 40,000 m³/d on the same footprint, with the plan-area "SOR" reading an alarming 161 m³/m²·d that is perfectly fine because the effective settling area is what counts (500 m² needed, 124 m² plan → 75 % footprint saving).
The physics is genuinely sound: tube flow at Re ≈ 148 is laminar (limit 500), and the channel velocity of 2.7 mm/s is far below resuspension thresholds. But two warnings belong next to every tube-settler brochure:
- Bulking sludge defeats them. Tube settlers fail when MLSS is above roughly 4.5 g/L or when SVI is climbing — the very conditions tropical plants drift into. They retrofit a hydraulically limited clarifier; they cannot retrofit a biologically failing one.
- They remove storage, not solids. A lamella clarifier holds minutes of sludge inventory, so the RAS/wasting balance must be tighter than in a conventional tank. The blanket sensor stops being an optimisation and becomes a trip wire.
The five-check discipline
Size the clarifier on all four envelopes, then check the fifth number — achievable RAS concentration against the SVI you actually expect at 30 °C. If the fifth check is above ~90 % of 10⁶/SVI, the design is balanced on a knife edge and you should either raise the SRT margin, lower the MLSS, or plan the chemical/lamella insurance now, before commissioning.