A correct dosage of raw materials is the first step towards a high-quality finished product. The recipe and nutritional value determine the exact raw materials and quantities need to achieve an ideal composition. It is therefore essential to mix them accurately and cleanly, in order to deliver a high-quality product. Correct dosing does not only require a good weigher, but also a suitable dosing instrument and an adequate control system (controller). Only if these three conditions are met can the dosing process be carried out quickly and precisely with guaranteed accuracy.
To determine the exact required dose to make a certain product, we tend to calculate down to the decimal point. But what is the permissible deviation that we accept in this process? This is a major hidden cost in many a factory, because certainty is required the critical components are present in the right quantities in the finished product. In fact, one goes to correct the possible deviation by adding extra product from the ‘’expensive’’ raw materials. To give a guarantee the possible deviation must be included in the dosage. In addition, nutritionists in many companies are also cautious and will often not push the limits for this type of component. This creates a safety barrier in the calculation, in the dosage and in the measurement error of that dosage. Ultimately, this results in a high-quality finished product, but on an annual basis this method of working will cost hundreds of thousands of euros for an average factory.
Fast and accurate dosing requires a controlled flow, a good weigher and a capable controller. The quality of the weigher is determined by both mechanical and electronic aspects. From a mechanical point of view, it is essential for a weigher that the construction is rigid. Even the smallest deflection, for example in an oblong weigher, causes a measurement deviation. In addition, the weighing structure must not be too heavy relative to the products or raw materials to be weighed. Clearly, a weight increase of 100 grams can be determined more accurately on a 5 kg weigher than on a 1.000 kg weigher. This also indicates the weakness of a ‘loss in weight’ weighing system. In this weighing strategy one measures a relatively (very) small decrease of a large weight.
Fast and accurate
A good weigher alone does not guarantee correct dosing; a weigher only determines how much has been dosed. For accurate dosing, the weigher, controller and dosing device must be optimally coordinated. The controller uses the information from the weigher to control the dosing device. The dosing instrument can operate at a fixed or variable speed. With a variable speed a much faster and more accurate dosing can be achieved. The exit point (afterflow) is corrected by default so that the final weight is usually within tolerance limit. A variable dosing rate is only fully utilized if the ‘settings’ (tipping points of dosing rates) are constantly optimized. This is a labor-intensive activity, so in practice the achieved results are often disappointing. Modern software, however, makes it possible to automate this optimization, which greatly improves the quality of the dosing.
Furthermore, in practice it appears that the design of a weighing construction also deserves the necessary attention. For example, it still regularly occurs that not all of the product ends up on the weighing scale, but (partly) on a funnel to the weigher. The product to be weighed should be dosed directly on the scale, and in such a way that no ‘leverage effect’ occurs. A leverage effect occurs, for example, when the product to be weighed lands on the far end of a scale, with the scale exerting a torque moment on a loadcell. And of course, a scale must have sufficient capacity for the product being weighed. This sounds obvious, but it is sometimes forgotten in practice. Without a doubt, the most common error is insufficient ventilation. The air displaced by the product to be weighed must be able to escape without disturbing the weighing process. A flexible cuff of filter cloth (often already already sealed) is insufficient for this venting. For a supply of, say, 50 kg/second of wheat, 75 liters/second = 270m3/hour of air must be vented. Another important design point is the emptying of the weigher; this process must be smooth and complete (non-residual).
The quality of a weigher in terms of electronics depends mainly on the quality of the loadcell and transducer (digitizer or indicator) used. For both, attention must be paid to a sufficiently large weighing range (considering a certain overload) and the distinctive power. This distinctive power determines the smallest possible weighing unit. With a distinctive power of, for example, 3.000 steps, considering 20 % overload, 2.400 steps remain for the actual weighing range. For a 100 kg scale this corresponds to a weighing unit of 42 grams. In practice it will be decided to reduce the number of steps to 2.000, which gives a more workable weighing unit of 50 grams. The impact of this step on the accuracy of a weigher is great, as the weigher will indicate 50 grams at an actual weight of 26-75 grams. Therefore, the potential deviation on the dosage is half a step. This deviation is separate from the dosing deviation that can still occur in the dosing itself and has purely to do with the readout. However, a large number of units of measure (scale divisons) does not say everything about accuracy and can therefore give a false impression of accuracy. The signal coming from the loadcell will also have to be sufficiently stable to make the reading accurate. To read out on, for example, 30.000 scale parts with the ‘’downtime’’ of the weigher jumping a few scale parts and having to average it out makes no sense. This gives false accuracy. It is the combination of the mechanical and electronic properties of the weigher that determines the true accuracy.
Signal delay (latency), the time that elapses between the signal from the loadcell and its processing by the controller, is a severly misunderstood problem in dosing weighers. Signal delay is caused, among other things, by electronic filtering and averaging to improve the stability of the signal. But the delay caused due to the network between weigher and controller should also not be underestimated. A process controller will calculate using outdated data as a result of signal delay. It is therefore more appropriate to speak of ‘afterflow’. For the final determination of a weight, the weight must remain constant long enough. The pitfall here is an electronically created stability that does insufficient justice to reality.
A well-designed weigher does not necessarily guarantee correct weighing. This is because there are also external factors that can adversely affect the weighing result. It is obvious that the scale must be free of disturbing influences as a result of connections with stabilizers and flexible cuffs. Yet it happens that this is in order, and then the weigher is kicked or spilled product interferes with the weigher. Other external influences should also be avoided as much as possible. These include vibrations, sagging floors or pneumatic transport or product movement in connected silos.
The dosing device is the instrument that brings the product to the weigher; for example, a (grid) slide, conveyor screw or vibratory conveyor. Contrary to popular belief, the highest achievable dosing accuracy is usually not dependent on the weigher. In practice, the dosing device is often the weakest link and therefore the limiting factor. This is because accuracy is dependent on the smallest controllable product flow. If the dosing instrument always drops a quantity of 100 grams into the weigher, the guaranteed dosing accuracy will never be better than 50 grams. After all, with 50 grams to go, an additional quantity will not make the deviation smaller. A smaller deviation is only a coincidence and not a guarantee. The smaller the controllable flow, the more accurate the dosing. For the combination of fast and precise dosing, it is necessary that the dosing instrument has a very wide range in dosing speed (flow). The ideal dosing instrument must be able to dose not only very much (quickly) but also controlled very little (precisely). In this respect, the dosing slide has enormous advantages of the dosing screw due to its large range and flexibility. A potential advantage for choosing dosing screws is the possibility for horizontal transport, since dosing slides use gravity and dosing screws also move the product. Moving product does give some other disadvantages such as product damage, higher energy consumption and compaction of product. However, in some situations it is not possible to work only with gravity and a screw can therefore be a solution. Or for example in situations where a weigher would become too side and affect the mechanical stability it can be chosen to work with 2 rows of sliders and 1 row of screws. Combinations of dosing instruments are therefore also a possibility.
The last point to consider is the power supply to the dosing device. A dosing device that is not fully and constantly fed itsef will not be able to provide the weigher with the ideal product flow. It is therefore advisable to choose a dosing instrument that itself influences the product outflow from an overhead silo or container. Here too there is an advantage of the dosing slide over the dosing screw because the movement of the slide activates the product. Depending on the type of product, it may also be necessary to take additional measures such as activation bellows, stirring devices, air pulses or vibrators/knocking devices.