Inline dosing of the last ingredients in a beverage or liquid food product just before packaging is in numerous cases the best process solution because of its high flexibility and cost effectiveness.
The flexibility comes from the possibility to tune a final product starting from a common basic product where at least instant the extra wanted ingredients are added to get a final product. If the dosing is executed just before conditioning, the shift from one product to another can happen at any moment after any quantity of the previous product. No need to point out what cost-savings can be generated from this way of working.
In line dosing is commonly seen as un reliable. The difficulty to ensure no concentration gradients at stopping and starting because of intermittent machines like fillers, centrifuges,... Has to be countered with a number of measures not only in control but also in process concept itself.
Thanks to our customers we got the opportunity to solve a lot of issues on concentration gradients.
And the solution is as in most cases a mix of several measures taken.
- A model based control algorithm
- A physical buffering to damp minor concentration changes. The buffer is however accounted for in the control model.
- Redundancy of the measurement used for dosing
- The use of appropriate dosing equipment like pumps and control valves for each specific ingredient to dose.
- A total service needed to integrate an adapt the dosing unit to existing process at the customers house.
- By means of iterative tracking the dosing is monitored on line so that product that does not meat specifications cannot pass
- A recipe based operation of the dosing. Where in fact operating interface is based on quality parameters like concentration. This measure makes it possible for operators to handle easily product switches. Since the more abstract concentration set points are contributed to the recipe of a certain product they are loaded automatically by recipe selection. Or in other words the operator only has to select product names instead of entering abstract set points of concentration.