Decanter Centrifuge Sludge Dewatering: Design, Sigma Theory and Operation

How does a decanter centrifuge dewater sludge?

A solid-bowl decanter centrifuge is, in essence, a horizontal sedimentation tank wrapped into a rotating cylinder-conical bowl and spun fast enough that centrifugal acceleration replaces gravity as the driving force for settling. Conditioned sludge is pumped through a stationary feed tube into an accelerator at the centre of the machine, where it is brought up to bowl speed and thrown outward against the bowl wall. Denser solids migrate to the wall and form a compacted cake; the clarified liquid (the centrate or centrate) forms an annular pond nearer the axis and overflows weirs at the cylindrical end.

Inside the bowl a helical screw conveyor — the scroll — rotates at a slightly different speed from the bowl. This differential speed (typically 1–30 rpm relative) ploughs the settled cake along the bowl wall, up the tapered beach (the conical section) and out of the solids discharge ports at the narrow end. Because the beach lifts the cake clear of the pond before discharge, a well-tuned decanter drains and compresses the cake as it climbs, producing a drier product than pond conditions alone would suggest. The whole separation is continuous and fully enclosed, which is one of the machine's defining operational advantages for odorous or pathogenic biosolids.

Everything that matters for performance follows from three physical facts: the acceleration field is enormous but finite, the residence time in that field is short, and the scroll simultaneously conveys cake and re-suspends fines. The art of decanter design and operation is balancing those three against one another. For the wider context of where the decanter sits among competing machines, see our overview of sludge dewatering centrifuge technology and the pillar page on sludge dewatering equipment.

What G-force does a decanter apply, and why does it matter?

The separating force is quantified as the relative centrifugal force (RCF), also called the G-force or the number of g's. It is the centripetal acceleration at the bowl wall divided by gravitational acceleration:

Worked example — RCF calculation. Take a bowl of internal diameter 0.5 m (radius r = 0.25 m) turning at N = 3,200 rev/min. First convert to angular velocity: ω = 2πN/60 = 2π × 3,200 / 60 = 335 rad/s. The wall acceleration is ω²r = 335² × 0.25 = 112,225 × 0.25 ≈ 28,060 m/s². Dividing by g: RCF = 28,060 / 9.81 ≈ 2,860 g. So the sludge experiences roughly 2,860 times its own weight pressing it toward the wall.

Why does this matter? Because sedimentation velocity scales directly with the acceleration field. Under gravity a discrete particle settles at its Stokes velocity:

The consequence is dramatic. A 10 µm floc fragment that settles at perhaps 5 × 10^-5 m/s under gravity settles at 2,860 × that — about 0.14 m/s — inside the decanter. A particle that needs hours in a gravity thickener clears the pond in seconds. This is why the centrifuge can capture fine, low-density biological solids that a gravity belt or drying bed would let escape. It also explains the machine's Achilles heel: sub-micron colloids and dispersed fines have such tiny d² that even 3,000 g will not settle them in the available time, which is exactly why polymer conditioning is not optional.

What is sigma (Σ) theory and how is it used to scale up?

RCF tells you the force at the wall, but a centrifuge is not uniform: acceleration varies with radius, and residence time depends on flow and pond geometry. To compare machines of different sizes and speeds on a single hydraulic basis, engineers use the sigma (Σ) concept introduced by Ambler. Sigma is the equivalent settling area — the surface area of an idealised gravity clarifier that would achieve the same separation as the centrifuge at the same feed rate.

The power of the sigma concept is scale-up. Because Q/Σ is (to first order) a machine-independent property of the sludge and its target cut point, a separation proven on a pilot decanter can be transferred to a full-scale one by matching Q/Σ:

Worked example — sigma scale-up. Suppose a pilot decanter with Σ₁ = 1,200 m² cleanly dewaters a digested sludge at Q₁ = 3 m³/h with acceptable centrate. A production machine offered for the duty has Σ₂ = 9,000 m². Matching Q/Σ, the expected hydraulic capacity is Q₂ = 3 × (9,000 / 1,200) = 22.5 m³/h at the same separation efficiency. In practice you apply an efficiency de-rating factor (0.5–0.9) because real decanters deviate from ideal plug flow, so a prudent design figure might be ~15–20 m³/h — but sigma gives the defensible starting point. Note that sigma scales with ω²: doubling speed roughly quadruples Σ and hence hydraulic capacity for the same cut, which is why high-G machines are compact for their throughput.

How do differential speed and pond depth control cake dryness?

If RCF and sigma set what the machine can separate, the differential speed and pond depth are the operating levers that set what it actually delivers — and they force a genuine three-way compromise between cake dryness, solids capture and throughput.

Differential speed (ΔN) is the difference between bowl and scroll rotational speeds. It sets the conveying rate and therefore the cake residence time on the beach:

• Low ΔN conveys cake slowly. The cake dwells longer in the compaction and drainage zone, so it comes out drier and solids capture rises (fines get more time to settle). But the solids inventory in the bowl grows, torque climbs, and if pushed too far the bowl chokes and the machine trips.
• High ΔN clears cake quickly, protecting against overload and allowing higher feed rates, but at the cost of a wetter cake and more fines carried into the centrate (lower capture) because material is scraped off before it has fully drained.

Pond depth is set by the position of the overflow (dam) weirs, which fix the liquid surface radius r₁. It governs the split between the clarification zone (in the pond) and the drying/beach zone (above the pond):

• Deep pond (small r₁, large liquid volume) maximises clarification residence time and hence centrate clarity and solids capture, but shortens the exposed beach where the cake drains, giving a wetter cake.
• Shallow pond (large r₁) lengthens the drying beach for a drier cake, but reduces clarification volume, so fine solids are more likely to escape in the centrate.

Worked example — the trade-off in numbers. A digested sludge fed at 4.5% dry solids (DS) is dewatered at 20 m³/h. At a baseline setting (moderate pond, ΔN = 12 rpm) the machine gives 24% DS cake at 95% solids capture. Tightening to a shallow pond and ΔN = 6 rpm lifts cake to 27% DS, but the centrate solids rise so capture falls to ~90% and the operator must drop feed to 16 m³/h to avoid choking. The mass balance shows the cost: at 20 m³/h × 45 kg DS/m³ = 900 kg DS/h fed, 95% capture recovers 855 kg DS/h into cake; at 90% capture and 16 m³/h the recovered load is only 0.90 × 16 × 45 = 648 kg DS/h, and the extra 5% of fines returns to the works in the centrate, loading the inlet works. Drier cake is not automatically cheaper cake once you count the recycled solids and the disposal saving from the lower cake mass.

Why is polymer conditioning essential, and how much do you dose?

As the Stokes relation makes clear, settling velocity scales with the square of particle diameter. Individual biological and colloidal particles in sludge are so small — often 1–20 µm, with a significant sub-micron colloidal fraction — that even 3,000 g cannot capture them in the sub-minute residence time. Polymer conditioning solves this by aggregating fine particles into large, fast-settling flocs, effectively increasing the operative d in the settling equation by one to two orders of magnitude.

High-molecular-weight cationic polyacrylamides are the norm for dewatering negatively charged biosolids. They work by two mechanisms: charge neutralisation of the particle surface (compressing the electrical double layer) and bridging, where a single long polymer chain adsorbs onto several particles and physically links them into a floc. The floc must be strong enough to survive the intense shear of the feed accelerator and the scroll, yet the dose must not be so high that unreacted polymer blinds the cake and holds water. There is a clear optimum.

Worked example — polymer demand. For the 900 kg DS/h duty above, at a dose of 11 kg active polymer per tonne DS, the polymer requirement is 900 × 10^-3 t/h × 11 kg/t = 9.9 kg active polymer/h. Supplied as a 0.4% w/w make-up solution from neat emulsion, that is roughly 9.9 / 0.004 ≈ 2,475 kg/h ≈ 2.5 m³/h of dosing solution — a figure that sizes the make-up and dosing skid. Because polymer is often the single largest running cost after power, jar and bench-scale optimisation genuinely pays. We cover selection and dosing in depth in the guide to sludge conditioning with polymer.

How much power does a decanter draw, and where does it go?

Decanters are energy-intensive relative to press-type dewatering, and understanding the power split helps both specification and operation. The total shaft power draw has three main components:

• Feed acceleration power — the kinetic energy needed to spin the incoming feed up from rest to bowl tangential velocity. This is unavoidable and grows with feed rate and the square of speed.
• Conveying (scroll torque) power — the work done by the back-drive/gearbox to plough compacted cake up the beach against the centrifugal force. This rises as differential speed falls and as cake load increases.
• Windage and friction — air drag on the spinning bowl and bearing/seal losses, significant at high RCF.

Worked example — acceleration power and specific energy. For Q = 20 m³/h = 5.56 × 10^-3 m³/s of sludge at ρ ≈ 1,010 kg/m³, the mass flow is ṁ = 1,010 × 5.56 × 10^-3 ≈ 5.6 kg/s. With ω = 335 rad/s and r = 0.25 m, the tangential velocity is ωr = 83.8 m/s, so v² ≈ 7,020 m²/s². Then P_acc ≈ ½ × 5.6 × 7,020 ≈ 19.7 kW just to accelerate the feed. Add conveying torque and windage and a realistic total main-drive draw for this duty is 30–45 kW. Over the 20 m³/h feed that is roughly 1.5–2.3 kWh/m³, and against 900 kg DS/h it is about 33–50 kWh/t DS — squarely in the expected band.

Wear is the other cost of the high-G environment. Abrasive grit and the constant scraping of cake erode the scroll flight tips, the feed accelerator and the solids-discharge ports. These are protected with replaceable tungsten-carbide tiles or hard-facing, but they are consumables with real cost and downtime. High RCF and abrasive feeds accelerate wear, so the drive for maximum dryness is tempered by maintenance economics.

Decanter vs belt press vs screw press: how does it compare?

The decanter is one of three dominant mechanical dewatering routes for municipal and industrial sludge. Each occupies a different point on the dryness / energy / footprint / attention spectrum, and the right choice is duty-specific.

The headline is that the decanter wins on cake dryness, capture, footprint and enclosure — the reasons it dominates large municipal and digested-sludge duties — but pays for it in energy and wear. The screw press is the low-energy, low-attention counterpoint suited to small and remote sites; the belt press sits between them but needs wash water and hands-on belt management. For a detailed head-to-head of the two press technologies, see our comparison of the screw press versus belt press. As a rule of thumb, choose a decanter when the drivers are the driest possible cake (to minimise disposal tonnage), a fully contained process for odour or pathogen control, or a large throughput in a small building — and accept the higher power bill and wear-parts budget that come with 3,000 g.

What operating variables should you adjust, and what do they do?

Commissioning and day-to-day optimisation of a decanter comes down to a handful of levers. The table below summarises the primary control variables and the direction each pushes the key outputs, holding the others constant — invaluable when tuning by trial.

A sensible commissioning sequence is: fix bowl speed near the machine's rated RCF; establish a stable polymer dose from bench tests and confirm on the machine by watching centrate clarity; then trim differential speed and pond depth together to reach the target cake dryness without losing capture or tripping on torque. Record the torque as your safety margin — a rising torque trend at fixed settings usually signals feed solids climbing or the cake building, and it is the earliest warning of an impending choke.

Because the decanter is the last step that sets the mass and dryness of the material leaving site, it should be sized and tuned alongside the upstream thickening and conditioning, not in isolation. If you are selecting between machines or scaling a proven pilot result to a production duty, MCBA can help translate sigma and torque data into a defensible specification — start with our sludge treatment and dewatering support pages.