RAS Culture Tank Hydraulics and Dual-Drain Tanks

How circular RAS tanks clean themselves, how Cornell-type dual drains split flow between bottom and sidewall outlets, and how exchange time, rotation and water velocity are set for fish.

The culture tank is the first unit process in a recirculating aquaculture system (RAS), and its hydraulics decide how fast waste reaches the treatment loop. A tank that concentrates and flushes solids quickly protects the downstream filters; a tank that lets solids settle and decay adds oxygen demand and ammonia to the system.

Why circular tanks clean themselves

In a circular tank, or a square tank with cut corners, the whole water column rotates around the centre. The FAO guide to recirculation aquaculture notes that this pattern gives organic particles a residence time of only a few minutes, depending on tank size. SRAC Publication 453 explains why: even weak circulation carries solids towards the centre, where radial velocities are lowest, so a tank with a centre drain is naturally good at solids removal. Removal can be improved further with a centre-sloping bottom or a centred dual-drain system, together with an optimised depth-to-diameter ratio. FAO adds that bottom slope has little effect on self-cleaning itself but makes complete draining easier.

Shape involves trade-offs. SRAC 453 states that rectangular tanks use floor space about 20% more efficiently and are easier to harvest, but move solids poorly; FAO describes octagonal tanks as a compromise that keeps circular-tank hydraulics while using space better, and D-ended raceways as a hybrid that is rarely used in practice. Because the water column is constantly mixed, oxygen is nearly uniform in a circular tank, and FAO recommends placing the oxygen probe near the outlet, away from oxygen injection points.

The Cornell-type dual drain

A Cornell-type dual-drain tank has two outlets: a bottom-centre drain and an elevated sidewall drain. The small bottom flow carries concentrated settleable solids to a compact settling unit, while most of the flow, now low in settleable solids, leaves through the sidewall drain to the main treatment loop. In the 150 m³ tank studied by Davidson and Summerfelt (Aquacultural Engineering, 2005), about 92–93% of the flow left through the sidewall drain and 7–8% through the bottom-centre drain. A computational study of octagonal tanks with this drain type (Computers and Electronics in Agriculture, 2018) used a base case with 45% of flow through the centre drain and examined how the flow split changes velocity, uniformity and vorticity, which shows that the split is a design variable rather than a fixed rule.

Bottom-drain flow and solids flushing

Davidson and Summerfelt (Aquacultural Engineering, 2004) measured 10 m³ and 150 m³ Cornell-type tanks holding Arctic char or rainbow trout at commercial densities. How fast settleable solids were flushed depended strongly on the bottom-centre drain flow and on the rotation period of the water. A bottom-drain flow of at least 5–6 L/min per square metre of tank plan area flushed settleable solids within 1–2 minutes in the 10 m³ tank and 3–6 minutes in the 150 m³ tank, and a rotation period of 1.3–1.7 minutes gave optimal velocities for flushing.

Hydraulic exchange time

Hydraulic retention (exchange) time is the time needed for the flow to replace one tank volume, that is, tank volume divided by flow rate. In the 2004 study, relatively uniform mixing was achieved at high fish densities of 90–98 kg/m³ with exchange rates of one tank volume every 20–32 minutes. Under these conditions, with inlet dissolved oxygen of 16–18 mg/L, dissolved oxygen across the tank cross-section ranged from 10.0 to 10.6 mg/L in the 10 m³ tank and from 9.0 to 11.2 mg/L in the 150 m³ tank. Mixing in that study was not affected by bottom-drain flow over the tested range, but it was affected by the orientation of the inlet nozzles.

Setting rotational velocity

Rotational velocity in a dual-drain tank is highest at the perimeter and rose almost linearly with distance from the centre in the 2004 measurements, so fish can choose the current they prefer. Two simple controls were confirmed: a larger bottom-drain fraction (12% compared with 6% or 0%) produced higher rotational velocities, and redirecting inlet nozzles changed velocity directly; reversing one of six 45° nozzles lowered the perimeter velocity of the 10 m³ tank from 17.8 to 13.4 cm/s. FAO describes a vertical inlet pipe with horizontally adjustable outlets as an efficient way to control the current. A 2025 experimental study in the journal Water found that inlet flow rate, inlet distance from the wall, angle and pipe design all affected average velocity, velocity uniformity and dissolved oxygen layering in a circular tank.

Water velocity for the fish

Velocity must suit the fish as well as the solids. The Ontario Ministry of Agriculture, Food and Agribusiness expresses current for salmonids in body lengths per second (BL/s): velocity in cm/s divided by fish length in cm. Its guidance targets about 1.0 BL/s and reports that this level reduces aggression and stress hormones and improves growth and size uniformity in several trout, char and salmon species. Larger fish perform better at lower relative velocities. In land-based systems, velocity can be raised by increasing inflow, lowering the water level or speeding up rotation in circular tanks. FAO notes that species differ: flatfish such as turbot and sole need more bottom area and can be held at lower depth and current, whereas pelagic species such as salmonids benefit from larger volumes and higher water speeds.