Continuous Metering of Particle Size — a Special Optical System Combines Speed and Accuracy

Related Vendors

Frewitt fabrique de machines SA

Elementar Analysensysteme GmbH

Maag Pump Systems AG

Confirming by experiment

The accuracy of the probe was tested on a HammerWitt Lab, a hammer mill used for grinding small volumes of product in the laboratory. It has a maximum speed of 7,000 rpm, and like all hammer mills it relies on impact stress for its grinding effect. The rotor, which is located on one side, can operate either in cutting mode (for fibrous products) or crusher mode (for hard products).

For this test, the HammerWitt Lab was used in crusher mode to grind a mineral salt at rotor speeds of 50–85 m/s. It was fitted with rectangular classification sieves with mesh sizes of 0.5–2.0 mm. A second mill was used to demonstrate the advantages of real-time control of the grinding operation. The ConiWitt-200 used for this test is a conical sieve mill that relies on shear stress. It has a maximum speed of 1,800 rpm, and is suitable for the disagglomeration and fine milling of wet and dry powders. In this case the ConiWitt-200 was used at rotor speeds of 5 m/s and 14 m/s to grind a pharmaceutical placebo. It was fitted with a rotor having a square-profile arm, and a sieve with a circular opening and 0.5 mm mesh size. On both the HammerWitt Lab and the ConiWitt-200, the particle size measuring probe was mounted at the outlet, at right angles to the direction of flow.

Real-time control

A total of nine were carried out using the HammerWitt-Lab with sieves of 0.5–2.0mm mesh. Figure 1 shows values of D₉₀ as calculated by the probe and as measured by laboratory sieve analysis, for each of the tests.

D₉₀ is an important parameter because it provides information about the largest particles, which are often critical to product quality. For example, in Figure 1, test number 1 has a D₉₀ value of around 200 µm, showing that 90 percent of the particles were smaller than 200 µm.

Comparison of the two lines in Figure 1 demonstrate in this case the results from the particle size distribution probe closely match the laboratory sieve analysis. However, unlike with laboratory analysis, the response speed of the inline probe makes it possible to intervene in real time to adjust grinding performance.

Figure 2 shows how such adjustments can benefit product quality. During the test, the ConiWitt-200 conical sieve mill starts up at a rotor speed of 5 m/s. The green line, representing the D₉₀ measurement, clearly shows that the particle size distribution is wide. Increasing the rotor speed to 14 m/s brings the D₉₀, D₅₀ and D₁₀ values (green, red and blue lines respectively) much closer together, indicating a narrower particle size distribution, resulting in a more consistent product.

In summary, inline particle size measurement provides: accurate and reproducible results; rapid response times, allowing real-time control of grinding performance; full documentation of the grinding process; and due to the ability to fine-tune the process, higher product quality without avoidably high speeds and consequentially wasted energy.

* The author is Head of Laboratory Process Analytics with Frewitt SA, Granges-Paccot/Switzerland.

(ID:20772790)