By the time a drawing is released, the pump has already been run through the computer dozens of times.

Two slurry pumps with the same bore, the same duty point and the same pipeline can differ by a factor of two in service life. Materials explain part of it. Hydraulics explains most of the rest: where the blade geometry sits, whether the volute really matches the impeller, and how far the rotational speed can be pulled down before head suffers.

Figure 1 — The hydraulic design loop: six steps per pass, with test-loop data fed back to correct the model.

1. Efficiency and wear life pull against each other

A clear-water pump is designed for one thing: efficiency. Narrow the passages, add a blade, hold the outlet angle small, and hydraulic loss drops. None of that transfers to slurry. Narrow the passage and coarse particles bridge inside it. Add blades and the impeller outlet gets crowded, which makes the particle velocity field less uniform, not more.

So a slurry impeller goes the other way: fewer vanes, wider outlet, thicker sections. In practice:

Vane count is normally 2 to 5, dropping as particle size rises.

A wider outlet lowers the relative velocity at the impeller exit and weakens the jet-and-wake pattern there, which slows wear. Take it too far and the curve develops a hump and the pump runs unstable.

Vane outlet angle is commonly 18° to 40°, taken toward the upper end for coarse particles and lower for fines.

The suction eye is opened up so the mixture is not forced to accelerate into the vane passages. That acceleration separates the two phases, particles start running on their own trajectories, and wear concentrates in one place instead of spreading.

Every one of those choices gives away a little efficiency. What it buys is double the wear life and far fewer blockages, which is not a hard trade to defend on a lifecycle basis.

2. The impeller and the volute are designed as a pair

The volute converts velocity into pressure, so its parameters only mean anything relative to the impeller. The working ranges we hold:

Base circle: 1.05 to 1.1 times the impeller outside diameter. Smaller and recirculation develops at the impeller exit; larger and energy conversion suffers.

Throat area: estimated from F₃ = Qₙ / (K·√(2gHₙ)), with the flow coefficient K between 0.05 and 0.08.

Tongue angle: typically 35° to 45°, widened to 50° where the pump has to work across a broad range of duties.

Tongue clearance: 0.05 to 0.1 times the impeller outside diameter. Tighter raises pressure pulsation; looser increases leakage and recirculation.

One more point that matters in the field. Above roughly 5% solids by volume, a centre-outlet volute is the wrong choice. Go to a side outlet and hold the volute diffusion angle below 8°, otherwise solids settle out in the passage.

3. The software is for iterating, not for producing drawings

The chain runs like this. CFturbo takes flow, head and speed, derives the specific speed, fixes the impeller type and generates a first-pass impeller and volute. The geometry is exported and handed to Fluent for solid-liquid two-phase analysis.

The CFD step is not about producing a colourful streamline image. We run k-ω SST for near-wall accuracy at reasonable cost, pair it with the Schnerr-Sauer model when cavitation is in question, keep y+ below 5 on the first cell layer and use structured meshes in the impeller region. The outputs that actually change the design are the particle trajectory plots and the wall shear stress distribution: they show where the metal is going to be cut through, while it is still a number on a screen rather than a scrapped casting.

Design is therefore a loop, not a line. First pass, simulate, adjust, simulate again, until efficient-range width, wear behaviour and NPSH margin are all satisfied at once. One published CFD-DEM result is worth quoting: increasing the impeller wrap angle from 66° to 96° cut combined impeller and volute wear by roughly 20%, at a small cost in capacity.

4. How the design system at EXCELLENCE PUMP is put together

Two things had to be in place: people and process.

On people, we have built up a technical team in fluid machinery and materials over many years, and we have signed new-product development agreements with the Hydraulic Machinery Research Institute of Tsinghua University and the Fluid Machinery Engineering Research Centre of Jiangsu University, which brings turbomachinery theory, test facilities and real duty data into our development work.

On process, the chain from design to prototype is closed inside the company. New product development uses CAD/CAM and CAPP throughout. Hydraulic design runs on CFD, Fluent and CFturbo; structural verification on ANSYS; 3D modelling and simulated assembly on UG. From hydraulic and wear-resistance prediction, through 3D two-phase flow analysis and structural optimisation, to 3D modelling, simulated assembly and simulated running tests, the whole sequence is completed in-house.

The practical result is that problems are caught before castings are scrapped. That is where development lead time comes from. With design and manufacturing quality held where they are, the development and trial-manufacturing cycle is considerably shorter than it would otherwise be.

The product development department works to three hard targets: high efficiency, long life, low energy consumption. All three have to be met together. Miss one and the design goes back.

5. Give us the duty, get an answer

Whether a hydraulic design suits your application cannot be settled by discussion, and it should not be settled by guesswork. Give us the data and it can be calculated.

What we need before selection: flow rate, head, slurry density and solids concentration Cv, particle size distribution d50 / d85, particle hardness (Mohs or Vickers), pH, temperature, and the pipeline layout including static lift.

EXCELLENCE PUMP provides free selection and scheme design. With a complete data set we will issue a selection review opinion; for unusual duties we can run a dedicated hydraulic model development from the RFQ stage onward.