The Kempink tester widens the effective bandwidth of the PDI (Pellet Durability Index), providing a clearer and more realistic view of pellet quality. The decimal precision becomes less relevant. Desired quality can be defined within a ±5% range, rather than tenths of a percent. Think of it as the difference between measuring with a tape measure versus a micrometer.

There is always a relative relationship between “good” and “poor” performance across all testers. However, there is no scientifically valid conversion factor that applies to all feed types and animal categories. During the transition phase, parallel testing can help build understanding, but a new benchmark will quickly emerge based on the Kempink tester.

Inside a cylindrical chamber with an integrated sieve, a shaft with uniquely shaped blades rotates. These blades generate a controlled airflow in which pellets are centrifuged while being sieved simultaneously. Test duration and rotational speed are precisely controlled via a PLC and frequency inverter, ensuring identical test conditions every time.

Some testers use vacuum motors, while others apply purely mechanical stress to the pellets. Due to factors such as heat generation, wear, and design variability, results between different testers are not comparable. Even repeated tests on the same device can produce deviations.

The laboratory tester operates stand-alone and is not connected to a higher-level system.
The in-line tester, however, offers full modern communication capabilities for exporting test data.

This function is available in the in-line Kempink tester, which measures a fixed volume. After testing, the total weight is determined, providing bulk density as a key quality parameter.
This value is also useful for optimizing bulk truck loading and transport planning.

A major challenge in feed production is the highly variable quality and composition of raw materials. If you always produce the “best possible pellets” from the best possible inputs, product quality (durability) will fluctuate just as much as the input materials.
Would your customers appreciate inconsistent product performance?

To avoid delivering fluctuating quality, it is better to aim for a stable, controlled average quality. This ensures that customers consistently receive what they expect.
At the same time, all available process variables can be optimized to ensure that this average quality is as high as possible.

All factors influencing product quality can be measured and recorded per batch — this is your data. By linking this data to pellet quality (PDI), a learning model is created using machine learning. This model allows operators to predict the required machine settings to achieve the desired product quality under current conditions.

In most cases, achieving a higher PDI requires increased mechanical input from the production line, resulting in higher energy consumption. Producing above the required PDI therefore leads to unnecessary energy use.

Increased machine load not only raises energy consumption but also accelerates wear on motors, bearings, shafts, dies, and other components. This results in: More frequent maintenance, Increased unplanned downtime, Higher spare part costs. Additionally, optimizing machine utilization (e.g. reducing idle running) and improving production planning can increase capacity — without requiring major capital investments.