Dosing systems: accuracy drifts with viscosity, safety overfill

Dosing accuracy changes with viscosity and temperature

What happens. The dosing system is set up with one product at one temperature. When the material gets warmer, colder or from another lot, its viscosity changes and the delivered amount drifts. Operators notice from the checks and recalibrate by hand.

What it costs. Material given away or, on the other side, parts that are under-dosed and rejected. Each manual recalibration also stops the line for a short time and relies on whoever knows the machine.

Why the usual fix fails. Calibration tables per product are long and never cover every temperature or lot. A tighter calibration interval costs production time. Heating the material to a fixed temperature helps only where the material allows it.

How a fixed-price remote diagnosis checks it. From dosing records, temperatures, pressures or pump data and the checks that follow, I look at how much of the error follows viscosity and temperature and how much comes from other causes. The fixed-price remote diagnosis tells you whether an adaptive correction can hold the amount without hand recalibration, and what it needs from the machine.

Background: Control that adapts to drift and wear

Fillers overfill to be safe

What happens. Every pack must contain at least the declared amount. Because the filler scatters, the target is set above the declared value with a safety margin, and the margin is rarely revisited.

What it costs. Product given away in every pack, which adds up quickly at filling speed. The cost is invisible on the shop floor and shows only in the material balance.

Why the usual fix fails. Lowering the target by hand works until a batch falls short and gets rejected, and then the margin returns. A better scale does not help when the scatter comes from the process, not the measurement.

How a fixed-price remote diagnosis checks it. From weight records, filler settings and product data, I check how large the scatter is, how it behaves over time and which part is predictable. The fixed-price remote diagnosis tells you whether adaptive filling can bring the average closer to the declared amount without more rejects, and what it would take.

Checkweigher feedback reaches the filler too late

What happens. The checkweigher sits well after the filler. By the time it reports a trend, many packs have already been filled with the old setting, and the correction then comes in steps or too strongly.

What it costs. Variation in fill weight, and with it either giveaway or rejects. The more the product changes during a run, the worse the delay hurts.

Why the usual fix fails. Making the feedback faster is limited by the transport delay between the two machines. A strong correction on a noisy weight makes the filler hunt, and a weak one reacts too slowly.

How a fixed-price remote diagnosis checks it. From weight series with time stamps, filler commands and line speeds, I work out the delay and noise of the loop and test how a feedback with prediction would have behaved on your own data. The fixed-price remote diagnosis answers whether predictive feedback would reduce the variation, and how much hardware change it would need.