An Expert in Decoiler Feeder & Metal Coil Processing

Home / All / NC Servo Straightener Feeder Guide / Can You Set a 3-in-1 From a Sample Part?

Can You Set a 3-in-1 From a Sample Part?

Nov 20,2026

A good sample part carries the roll settings that made it, if you know which four measurements to take off it.

The alternative is starting from a blank setting sheet on a machine you have never run, which usually costs 40 to 80 parts before the first acceptable piece.

Here is the sequence we teach operators on an NC straightener feeder, in the order it has to happen.

Why a Sample Part Beats a Setting Sheet

A setting sheet records what someone wrote down. A sample part records what the machine actually did to the material in front of it.

Sheets drift. A die gets re-ground, a coil lot changes hardness, an operator adjusts a gap to suit a batch and never updates the paper.

The part in your hand was made by the same rolls, the same gap and the same penetration that you are trying to reproduce, and it has the evidence on its surface.

That evidence is worth reading before you touch a handwheel, because it tells you which roll pair was working and which was not.

Step One: Prove the Part Came Off This Machine

Start with the edge. A sheared edge from a die has a rollover, a burnished band and a fracture zone, and the proportions differ between tools.

If the sample came from another plant, the edge will not match the die sitting in your press, and every setting you derive from it will be wrong by the thickness of that difference.

Check the width against the strip the line is threaded with. A part 2 mm narrower than the incoming strip means the sample came off a different width, which changes the guide setting as well as the gap.

Look for the feed marks on the part. Progressive dies leave a witness line where the pilot entered, and the distance from that line to the part edge tells you where the strip sat in the guides.

Two minutes here saves an hour later. A sample from the wrong machine is worse than no sample at all.

Step Two: Measure the Strip, Not the Part

Take thickness at five points across the width and three along the length, with a micrometer reading to 0.001 mm.

A 1.5 mm strip that reads 1.52 mm at the centre and 1.49 mm at the edges has 0.03 mm of crown. Set the gap from the centre reading and the edges will be under-rolled.

Set from the average instead, and open the gap 0.01 mm above nominal to allow for the thickest point. The rolls work on the average and the crown passes through.

Width matters for a different reason. A strip 0.5 mm over nominal will bind in the guides, and the resulting side load shows up as a diagonal bow that looks like a roll problem.

Record both numbers on the setting sheet before you change anything. The next operator will need them, and so will you when the coil lot changes.

Step Three: Read the Surface for Roll History

Every work roll leaves a track. Measure the spacing between repeated marks along the strip direction and you can name the roll that made them.

A mark every 322 mm comes from a roll with a 322 mm circumference, which is a 102 mm diameter roll. A mark every 251 mm comes from an 80 mm roll.

On a five-roll straightener the upper rolls are usually the larger diameter and the lower rolls smaller, so the spacing identifies the position as well as the part.

Polish lines running along the strip rather than across it are different. They come from a roll with a damaged edge or a burr on the strip itself, and they are not a setting problem.

Marks that appear only on one side of the strip point to a roll that is not parallel to the other side, which is a levelling issue rather than a gap issue.

Write down the spacing before you change the gap. Once you start adjusting, the old marks are still there and they will confuse the next reading.

Operator measuring strip thickness before setting an NC straightener feeder

Step Four: Turn Curvature Into Penetration

Lay the sample on a surface plate and measure the gap under its centre with a feeler gauge. That gap is the sagitta, and it is the only number you need to start.

A 300 mm part with 1.2 mm of bow has a radius of about 9,400 mm. That is the curvature the straightener failed to remove, and it tells you the last setting was short on penetration.

Penetration is set as a percentage of the roll gap, and the useful range for mild steel is 60 to 75 percent. Below 60 the strip passes through and keeps its bow; above 80 the surface starts to mark.

Start at 65 percent on a sample with visible bow, run three parts, and measure again. Move in 5 percent steps rather than 15, because the curve flattens quickly near the right setting.

For a 1.5 mm strip on a 102 mm roll set, that is a first move of about 0.05 mm on the penetration screws. Small numbers, large effect.

MeasurementWhat it tells youWhere it misleads
Thickness at five pointsGap setting and crown allowanceA single reading at the edge understates a crowned coil by 0.03 mm
Mark spacing along the stripWhich roll is marking, by circumferenceTwo rolls of similar diameter give almost the same spacing
Sagitta over a 300 mm chordResidual curvature and penetration directionA part that is flat on the plate can still be bowed across its width
Edge condition of the sampleWhether the part came off this dieA re-ground die keeps the same edge look while changing the strip position

Step Five: Three Parts and a Straight Edge

Run three parts and check each one on the plate with a straight edge and a 0.05 mm feeler.

The feeler should not enter anywhere along a 300 mm length. If it enters at the ends only, the strip is still bowing upward and needs more penetration.

If it enters in the middle, the strip is over-rolled and the bow has reversed. Back the penetration off by half the last adjustment, not all of it.

Check the width direction as well, with the part standing on edge. A part that is flat along its length and dished across its width needs a different roll pair, not more pressure.

Then run ten parts at production speed. Settings that hold at 20 strokes per minute can move at 180, because the strip spends less time in the rolls and the back tension changes.

Only after the ten-part run is clean should the setting go on the sheet.

Stamped strip parts checked on a surface plate after straightener setup

Where the Sample Stops Being Evidence

A sample part proves one thing: this material, on this machine, at this setting, produced an acceptable piece. It does not prove the setting transfers.

Hardness is the biggest gap. A 1.5 mm SPCC at 180 MPa yield and a 1.5 mm SPCC at 240 MPa need different penetration for the same flatness, and the part in your hand cannot tell you which one it came from.

Position in the coil matters too. The head of a coil carries more residual curvature than the body, so a part cut from the first 20 m describes a condition you will not see again for 400 m.

Surface marks are also lost downstream. A part that has been through a wash and a paint line has no roll history left, and a galvanised sample hides the marks under the coating.

Finally, part flatness is not strip flatness. A die can clamp a bowed strip flat for the length of the part and release it bowed again, and the part will look acceptable while the strip is not.

Use the sample to get close, then verify on the strip with a straight edge held against the running material before the first production stroke.

  • Sample matched to the die and strip width before any measurement
  • Thickness read at five points, average used for the gap
  • Mark spacing recorded and the roll named by circumference
  • Penetration started at 65 percent and moved in 5 percent steps
  • Ten parts at production speed before the setting is written down

Reading a part back to a setting is the fastest route on a machine with no history.

Reading a strip back to a flatness spec is the durable one, and it is the subject of the 3-in-1 decoiler straightener feeder guide on roll count, pitch and penetration.

Roll diameters, roll counts and gap ranges by frame size sit in the straightener data library, the drawing set to open before a pressure is specified.

How close should a 3-in-1 hold flatness after setup from a sample?

Within 0.05 mm over 300 mm. If the feeler enters, the setting is still 5 to 10 percent short on penetration.

How many parts does a sample-based setup usually waste?

5 to 12, against 40 to 80 when starting from a blank sheet on an unfamiliar machine.

Does a 0.03 mm crown change the gap setting?

Yes. Set from the centre reading and open 0.01 mm, or the edges will be under-rolled by the crown amount.

Can a part from a 5-roll machine be set on a 7-roll machine?

Only as a starting point. Two extra rolls remove more curvature at the same gap, so reduce penetration by 5 to 10 percent before the first run.

Where to go next

A sample gets you to the first good part. Holding flatness across a whole coil needs the roll geometry behind it.

The flatness pillar covers roll count, pitch, gap and penetration as one system rather than four settings.

Read the flatness guide

Follow us to Get Latest Updates

subscription