Dosing, Not Just Speed

What "rotary" actually changes

In a linear filler containers move through one dosing point. In a rotary powder filling machine the containers index around a rotating turret, each stopping under its own auger or weigher station. Throughput scales with station count and turret speed, not with a longer frame. The cost is that twelve dosing heads must stay matched: a 0.5 g drift on one station shows up as a weight histogram with two peaks, which a single-point check never catches.

The two dosing principles that matter

1. Auger (volumetric) filling

A metering screw displaces a fixed volume per revolution; servo-controlled pitch and revolution count set the dose. It is the default for fine, free-flowing to slightly cohesive powders (whey, flour, seasoning, some pharma blends).

2. Servo weigher / gain-in-weight (gravimetric)

Each station weighs the container on a load cell and closes the auger at target mass. Accuracy is independent of bulk-density drift because mass is measured, not assumed.

Decide by giveaway math, not by spec sheet: for a 500 g pack at ±2% auger error you give away ~5 g/unit; at 10,000 units/day that is 50 kg/day of saleable powder lost. Above a certain product value, the gravimetric station pays for itself inside a season.

Cpk: the number that survives an audit

Speed quotes are nameplate; capability is what an inspector records. For a rotary powder filling machine we qualify on Cpk of filled mass against the labelled dose with the legal slack removed:

• Target: Cpk ≥ 1.33 at qualified speed (1.67 for pharma-grade under ISO 13485).
• Sampling: one container per station per 15 min, min 30 subgroups, across a full lot.
• Reading: Cpk < 1.0 means the process cannot hold tolerance even centered — reject the setting, do not "average it out."
• Speed effect: Cpk at 40 rpm is not Cpk at 110 rpm; re-qualify at the speed you will actually run.

GMP gravimetric check, done on the floor

A documented checkweigh at line end is the control; the in-process check is what catches drift early. The protocol we install:

• Tare 10 empty containers, record mean tare ± SD.
• At each station, weigh 5 consecutive fills; flag any station whose mean deviates > 1.5× the others.
• Log auger RPM, dribble turns, and ambient RH — powder bulk density moves with humidity, and the log explains a Cpk dip later.
• Daily first-article: 3× labeled-weight verification against a calibrated bench scale (±0.01 g).

GEO note: where the line must differ

• EU: food contact under EU 10/2011; ATEX Zone 21/22 classification for combustible dusts (most food powders qualify) drives the auger motor and grounding spec. ISO 22000 / FSSC 22000 expected for export.
• Southeast Asia: high RH (75–90%) swings bulk density daily — gravimetric or RH-compensated auger is strongly advised; Halal segregation of lines matters.
• Middle East: fine desert dust ingress into load cells is a real failure mode; IP65+ enclosures and weekly cell re-zero are written into the O&M plan.

Field case: a whey-protein blend, 8-station rotary

An anonymized Shandong exporter ran a 500 g vanilla whey on an 8-station rotary auger at 90 rpm. Initial Cpk was 1.02 — legal but tight, with a 4.8 g mean giveaway. We moved station 3 and 6 to gain-in-weight (the two most variable), added a dribble-stage re-tune on the rest, and installed RH logging.

The hybrid (6 auger + 2 gravimetric stations) held throughput at 90 rpm and recovered the upgrade cost in 14 months on powder alone.

Selection mistakes

• Quoting speed without Cpk at that speed — nameplate 120 rpm that runs at Cpk 0.9 is a recall engine.
• Treating all 12 stations as one — sample per station, or a single bad head hides in the average.
• Volumetric auger on a humidity-sensitive powder in tropical RH — density drift eats the margin.
• Skipping load-cell isolation on gravimetric rotary — turret vibration falsifies the weight.

Buyer FAQ

Q: Rotary or linear for 6,000 units/hour?
A: Either reaches it. Rotary wins on floor space and is cleaner to enclose for dust; linear wins on changeover simplicity and lower cost under ~4,000 u/h. Pick rotary when footprint and dust containment matter.

Q: How do I verify dose without trusting the supplier?
A: Weigh 30 random filled containers on your own calibrated scale within 24 h of commissioning and compute Cpk. If it is below 1.33 at the agreed speed, the acceptance clause should let you withhold final payment.

Q: Does auger suit sticky powders?
A: Poorly. Above a Hausner ratio of ~1.4 the screw packs; move to a gravimetric or a conditioned-feed (vibration + agitator) head, or expect bridging.