Technical Results

The experimental campaign focused on a two‑stage multicyclone geometry that combines a primary cyclone with a secondary, smaller cyclone to capture fine particles that escape the first stage. A full‑scale test rig was built, allowing the measurement of dust concentration, pressure drop, and particle size distribution for a wide range of operating conditions. The data set was used to validate CFD models and to derive empirical correlations for separation efficiency as a function of Reynolds number, particle size, and shape.

Numerical simulations revealed that the secondary cyclone reduces the axial velocity component of the recirculating flow, thereby increasing the residence time of sub‑micron particles. The pressure loss of the combined system was found to be 20 % lower than that of a conventional single‑cyclone of equivalent dust‑capture capacity. Moreover, the multicyclone achieved a separation efficiency above 95 % for particles larger than 5 µm, while maintaining a pressure drop below 1.5 kPa at a volumetric flow rate of 10 m³ min⁻¹.

The influence of particle shape was quantified by testing spherical, cylindrical, and irregular particles. The results show that non‑spherical particles exhibit a 10 % higher capture probability in the primary stage, but a larger fraction of them is recovered in the secondary stage due to their lower settling velocity. This shape‑dependent behaviour was incorporated into a reduced‑order model that predicts the overall dust‑capture performance with an error below 5 % compared to the experimental data.

• Two‑stage geometry reduces pressure drop by ~20 % compared to single cyclone.
• Separation efficiency exceeds 95 % for particles >5 µm.
• Secondary cyclone captures fine particles that escape the primary stage.
• Particle shape influences capture probability; non‑spherical particles are better retained.
• Reduced‑order model predicts performance within 5 % of measurements.

Collaboration

The project was a joint effort between the private engineering firm Ambros Schmelzer & Sohn GmbH & Co. KG, which supplied industrial expertise and pilot‑scale equipment, and the Institute for Mechanical Process Engineering at the University of Stuttgart, which provided academic research capabilities and advanced simulation tools. Funding was provided by the Federal Ministry of Economic Affairs and Energy (BMWi) under reference 03ET1250B. The partnership enabled a rapid translation of laboratory findings into a scalable industrial solution.