What is the Truth Definition?

"CHECKWEIGHER ACCURACY DEFINITION:
The most crucial parameters used to calculate the accuracy of a scale are Linearity and Repeatability in the case of a CheckWeigher. Linearity in a CheckWeigher expresses how close the measured weight of a test load is to its actual weight when weighed on the scale each time. The difference between the actual weight and the displayed weight is expressed as an error. Less error signifies a more linear system. The average error is defined as the average difference between the actual weight of a package and the displayed weight. There are differences between the actual linearity of the LoadCell and the linearity parameter we use. Repeatability is calculated with standard deviation. The standard deviation of a CheckWeigher is defined as the weight variation calculated from multiple weighing processes of a specific product. A smaller standard deviation indicates better repeatability or precision for the CheckWeigher. Graph 0-3 compares scale results with a target, where each point symbolizes the weighing of a specific product. Imagine testing four different CheckWeighers, each weighing a product five times. The diagram below represents the static values of measurements taken on a calibrated scale. CheckWeigher number one is neither repeatable nor linear, with widely scattered weight values. The second CheckWeigher is linear but not repeatable, with weight values clustered around the target but significantly different from each other. The third CheckWeigher is repeatable but not linear, with all weights close to each other but distant from the target. The fourth CheckWeigher is both linear and repeatable, as seen in the target, with weights clustered around the target and close to each other. Graph 0–2 represents accuracies of trials toward the target.

CHECKWEIGHER ACCURACY: What CheckWeigher manufacturers generally refer to as accuracy is often repeatability rather than linearity. CheckWeighers easily eliminate linear errors. How do we determine CheckWeigher accuracy? CheckWeigher control compares whether each product falls within the weight limits of the current region (Zone) and target values. If the CheckWeigher can precisely separate products above the region limits at each boundary and target value, it is ideally functioning. Unfortunately, achieving this is difficult due to environmental and machine factors that will be identified later. Instead of defining a small weight region, upper and lower limits need to be established for the classification of the package. The uncertain areas are defined in graph 0-4. The accuracy of the CheckWeigher is critical at the boundaries of weight regions. Consider the passage of a 110g product through the CheckWeigher, as shown in graph 0–4. The black region represents the normal distribution curve of the CheckWeigher for a 110g product. The graph demonstrates that 95% of the 110g product is classified between 109.8g and 110.2g. The rate of change in 95% of the products (2 standard deviations) is ±0.2g. 99.7% of the products range from 109.7g to 110.3g. The accuracy or variation in a 110g product is classified and shown in the 2nd and 3rd regions. CheckWeigher accuracy is defined as the standard deviation of values obtained by repeatedly passing a single product through the CheckWeigher. This represents the variability or sensitivity of the CheckWeigher.

A CLOSE LOOK AT CHECKWEIGHER ACCURACY: The measurement value of a CheckWeigher will show small variations in repeated weighing processes of a single package. The CheckWeigher's own variation is independent of the change occurring in the actual weight from one product to another. For example, in graph 0-4, the variation in CheckWeigher measurements for a 110g product is plotted as a frequency histogram. The actual weight can vary from 100g to 130g, but the measured value from the CheckWeigher can be up to only 0.6g for any product. Unlike static scales, CheckWeighers experience dynamic forces from various variables in packages and scales. Due to these forces, achieving accuracy in a CheckWeigher is not possible as it is in a highly precise balance. Consider the environmental effects around the CheckWeigher. Hundreds of packages enter and exit the scale continuously at a rapid pace within a minute. Imagine being weighed on a scale in your bathroom. Step onto the scale, read the value, and then run around the scale and move over it. You will notice that the value on the scale changes, right?
CheckWeigher load cells have a very short return time to zero, but since packages never stop moving, the scale never reaches zero. Most manufacturers offer products with accuracy ranges that suit your needs. Accuracy comes at a price. To obtain better accuracy from a CheckWeigher, the flexibility of machine application or the determination of the physical characteristics of the machine is necessary. A higher precision system may be more expensive, but in the long run, it can save you more by reducing product, part, and labor losses. A CheckWeigher with smaller precision can lead to more losses in long-term operation.

TESTING CHECKWEIGHER ACCURACY: The easiest way to test the accuracy of a CheckWeigher is to first weigh a specific product on a stable scale and then pass it through the CheckWeigher several times, recording the values. Calculate the average and standard deviation using these values. CheckWeigher accuracy can be defined as ±1.2 or 3 standard deviations (Sigma) from the average. In this book, we will express accuracy as 2 standard deviations from the average for the same CheckWeigher in 6 different situations. The average error of a product is simply equal to the exact difference between the average weight value calculated by the CheckWeigher and the actual weight value.

WEIGHT AND CALCULATION PRINCIPLES: The principles of weights and measurements are based on the resolution of scales or the number of partitions they can detect. This is expressed as the smallest amount of change that can be shown by the indicator. In dynamic CheckWeighers, there is no correlation between resolution and accuracy. For example, a scale with 10,000 partitions can weigh a 500g package in 0.05g increments, and the maximum allowable error would be 0.125g. Under reasonable static conditions, results are obtained after several weighings of the same load."