Accumulation on Slat Chain Conveyors: How Much Buffer a Filling Line Actually Needs

A filler runs at 400 containers a minute. The labeller, twenty metres downstream, jams twice an hour and takes forty seconds to clear. Between those two facts sits a decision that usually gets made on a floor plan, by drawing a longer table and hoping. How many seconds of product the line stores, and where it stores them. Get it right and the filler does not notice the labeller. Get it wrong and the filler stops twenty-four times a shift for forty seconds — sixteen minutes of production that does not show up in a report, because it does not appear as one event.

1. What a Buffer Is Actually Buying

2. The Arithmetic Nobody Runs

3. Where the Buffer Sits on the Line

4. Three Ways to Accumulate on a Slat Chain

5. What Backpressure Does to a Filled Container

6. Changeover and Cleaning

7. The Numbers to Bring

8. A Closing Line for the Floor

A buffer does not add a single container of capacity. What it buys is a window: for as long as the window lasts, a stoppage downstream is invisible upstream.

That framing matters because it changes what the buffer is measured in. Not metres, not table area — seconds. A line with 45 seconds of accumulation absorbs a 40-second jam and does not notice. A line with 25 seconds absorbs the first 25 and then stops, and the remaining 15 are paid by the filler and by everyone behind it.

The window is consumed in one direction and refilled in the other, and those two rates are not symmetrical. That asymmetry is where most buffer sizing goes wrong.

Take a 330 ml can, 66 mm diameter, on a slat chain with 6 mm between containers. Call it 72 mm of pitch. At 400 containers a minute the chain surface runs at 0.48 m/s, and one minute of production occupies 28.8 metres of single-file conveyor.

A 12-metre accumulation table holds 166 containers, which is 25 seconds — a film splice, a label reel change, an operator reaching across the line.

The second number is the one that gets missed. A buffer only refills at the difference between the two machines' rates. With a packer running 5% faster than the filler, the net fill rate is 5% of line rate, so thirty seconds of accumulation consumed at 400 a minute takes ten minutes to rebuild. A line with a 30-second buffer and a stoppage every six minutes does not return to full; it runs the shift in a partially emptied state, and the twenty-fifth stoppage stops the filler even though the first twenty-four did not.

Two design consequences follow. The buffer has to cover the long stoppages, not the typical ones — sizing on the mean guarantees the mean and nothing else. And the refill rate sets how many stoppages per hour the line can absorb.

Position is not a matter of taste. A buffer between the filler and the capper protects the filler from capper faults but exposes filled, open containers to standing time, and an open container in a queue is a spill waiting for a nudge. A buffer between the capper and the labeller protects both, and everything in it is sealed.

The rule that survives contact with real lines: the buffer goes after the least reliable machine, not after the most expensive one. Reliability data usually exists, in maintenance logs and the shift book, and it is more honest than the layout drawing.

On a filling line there are two natural positions — after the capper or sealer and ahead of the labeller or date coder, and again ahead of the case packer. The second is frequently left out, which is why a line with a perfectly sized first buffer still stops when the case packer runs out of blanks.

Zero-pressure accumulation. The chain stops under each full zone, so containers do not touch. Zones are created by sensors at fixed intervals. This is the kindest arrangement for the container and the least kind to the chain: every zone start is a chain pull, every stop is a wear event on the wear strips, and a line with fifty zones starts and stops thousands of times a shift.

Low-pressure accumulation. The chain runs continuously and containers touch, with a controlled amount of backpressure. Simpler, quieter, cheaper, and it puts a continuous load on the head of the queue.

A long table, single file or multi-lane. Crude, and often the most effective. Width buys seconds faster than length does, at the price of a lane divider and a combiner at the far end.

Where the floor is already full, a Spiral Conveyor stores seconds without buying ground — it trades horizontal metres for vertical ones at roughly the density in the table. It also adds two transfers and a surface that is awkward to wash, which is the price of the footprint. The third lever is cheaper than either: running the packer slightly faster than the filler shortens refill without adding a metre, and it is the one most often left out.

A queue on a running chain is a stack of containers pushing on each other. Each contact carries a few newtons; a full table of four hundred carries a few hundred, and the container at the discharge carries all of it.

· Closures back off. A screw cap tightened to a specified torque does not hold that torque against a sustained axial push on the container body. The failure shows up as a leaker at the palletiser, three stations from the cause.

· Labels scuff. Wet-glue and pressure-sensitive labels abrade against the next container. Paper labels on a wet bottle after four minutes in a queue look handled, because they have been.

· Can seams work. The double seam is a mechanical fold, and repeated lateral load at the seam shoulder is the load it likes least.

· PET panels creep. Lightweighted bottles deform under sustained axial load, and the deformation is permanent once the container cools.

None of this argues against accumulation. It argues for knowing what the queue is doing while it stands still, and for putting the longest queue where the containers are sealed and the load is lowest.

An accumulation table is the largest flat surface in a wet part of the building, and it collects what the line sheds: label adhesive, rinse water, broken glass, the odd cap. Cleaning follows the water — detergent at the discharge end, worked back against the direction of travel so loosened material does not cross a cleaned surface, rinse in the same direction. The chain is lifted at the discharge, the wear strips are checked for the black paste that signals abrasion, and the return way is flushed separately: the dirtiest part of any slat chain line is the return, because nothing on it is visible during production. On a line with changeovers, the table is also where the old product hides, usually in a zone behind a sensor.

Six numbers decide the buffer, and all six are known in the plant before anyone draws a layout:

· Filler rate at the label, and the downstream machine's rate, in containers per minute.

· Stoppage distribution downstream: how often, how long, and the eightieth percentile rather than the average.

· Container diameter, mass and pitch as it actually runs on the chain.

· Closure type and application torque, and whether the container is glass, tinplate or PET.

· Available floor length and width, including the aisle that cannot be blocked.

· Washdown regime: chemistry, temperature, and how often the table is opened.

A layout quoted with those answers can be checked on paper. Without them, the buffer becomes whatever length fits, and the price of finding out is one figure on the drawing, ten after fabrication, a hundred once the line is standing and the first long stoppage arrives at 2 a.m.

A line buffered to the average stoppage stops every shift, because averages are made of a few long events and a great many short ones. The seconds cost more than the metres, and the cheapest seconds are the ones bought by running the packer slightly faster than the filler.

Orange Zhou, Technical Sales Manager · orange@xstlconveyor.com