What Happens When a Feeder Is Sized on Press Tonnage?

The Order That Started With a Tonnage Number

Tonnage describes the press, not the strip. It says how hard the press can push, and nothing about what the feeder has to pull.

The plant's job mix at the time of purchase was 2.0 mm and 2.5 mm mild steel, 900 to 1,200 mm wide. Any feeder in the catalogue would have run that work.

The quotation was written against a 200-ton frame, which produced a feeder with generous roll diameter, a wide body and a motor sized for the heaviest strip that frame could handle.

On the original job mix the machine was, if anything, overbuilt. It fed accurately, it never slipped, and the maintenance log stayed empty.

Month One: It Ran and Nobody Complained

Commissioning took a day and a half. Feed length calibration was straightforward and the first production run held 0.08 mm over 300 strokes.

The operator liked it because the rolls gripped hard and nothing moved. Roll pressure was set high at commissioning and never touched again.

For four months the feeder ran two shifts on 2.0 mm and 2.5 mm work with no quality escapes. It became the machine nobody thought about, which is the best thing you can say about a feeder.

Month Four: Short Feeds on the Thin Gauge Job

The plant won a contract for 0.6 mm galvanised panels, 700 mm wide, at 180 SPM. First shift produced 40 good parts, then a run of short feeds averaging 0.4 mm.

The die survived because the pilot pins were doing their job. The scrap bin did not.

The obvious suspects were ruled out one by one: encoder coupling was tight, feed length was correct for the die, and the drive reported no following error.

Roll pressure was still set for 2.5 mm material. On 0.6 mm strip that pressure flattened the strip section under the roll, and the roll recovered before the strip did.

The material was being extruded slightly at every stroke, and the accumulated effect over 200 strokes showed up as drift rather than as a single bad feed.

Reading the Data Backwards

The plant pulled 200 strokes of feed length data and plotted it as cumulative error rather than absolute error. The picture changed immediately.

Absolute error per stroke looked random and small, well inside tolerance. Cumulative error climbed steadily and crossed the die's pilot clearance at around stroke 140.

That pattern points at either roll slip or material deformation, and both were happening: high pressure on thin strip causes the second, which causes the first.

A single-stroke check would have passed this machine every day of the week. Only the accumulated run exposed it.

What the Sizing Sheet Should Have Asked

Tonnage appears nowhere on that list. It is a useful number for the press foundation and a misleading one for the feeder.

The Fix and What It Cost

The fix was a pressure regulator with a gauge readable at a glance, plus a pressure table for each gauge in the job mix.

Total parts cost was under USD 900. The commissioning engineer spent two hours setting four pressure values and writing them on a laminated card taped to the guard.

The scrap and rework from four months of intermittent short feeds was never fully calculated, but the quality engineer put the known figure above USD 30,000.

Nothing was replaced. The machine was correctly built and wrongly set, which is the usual shape of this particular failure.

The table is lopsided on purpose. Almost all the money sits in the row that bought nothing at all.

That is the shape of most sizing errors. The correction is cheap and the exposure before correction is not, which makes the case for a proper sizing sheet financial rather than technical.

The plant now writes a thickness range on every feeder enquiry, including repeat orders for machines they already own and understand.

A second machine bought on the same tonnage logic would have repeated the same four months on a different press.

Sizing Rules That Survive a Job Mix

For reference, the press feeder range publishes roll diameter, feed length and motor rating per model, which is exactly the comparison a sizing sheet needs.

Reading that sheet against a job mix takes an afternoon. Buying the wrong size takes a year to notice and four months to admit.

• Size on the lightest strip in the range, not the heaviest. Light strip exposes grip and pressure limits that heavy strip hides.
• State the full thickness range with yield strength, and treat any future material as a new sizing exercise.
• Ask for a pressure-versus-gauge table at commissioning, and require it to be posted at the machine.
• Test at the fastest speed and thinnest material together, because that combination is the worst case and also the one nobody tests.
• Log cumulative feed error, not single-stroke error. The die cares about accumulation.

Where Tonnage Is Still a Useful Number

Tonnage belongs in the press specification, the foundation design and the die load calculation. It is a legitimate input in all three.

It also correlates loosely with feeder size in practice, because bigger presses tend to run wider and heavier strip. That correlation is not a sizing method.

What tonnage does tell you is die weight and shut height, both of which constrain where the feeder can sit.

It also sizes the press foundation, which matters when a line has to fit an existing bay without rebuilding the floor.

Those are real constraints. They are simply not the ones that decide whether strip slips between two rolls at 180 SPM.

A sizing sheet that starts at tonnage and ends at roll pressure is working backwards through the problem, and it will produce a machine that runs until the job mix moves.

The failure mode above is what happens when a correlation gets used as a specification. It works until the job mix changes, and job mixes always change.