Slitting scrap: where the coil goes, and how to keep more of it

A slit coil leaves the line as three things: product, edge trim, and whatever the pattern could not use. Only the first is sold, and the other two are decided before the first cut, by the slit pattern and the coil you put on the uncoiler.

That is worth saying plainly, because scrap on a slitting line is mostly not a quality problem. A line running perfectly still turns a fixed percentage of every coil into trim, and that percentage is arithmetic you can do in advance. Rejects on top of that are a separate question, covered in the defect guide.

The whole of it, in one line:

Yield is the first term over the coil width. Everything below is about the other two.

The outer edge of a master coil is not product. It carries the mill edge, which is out of tolerance on width, often wavy, and sometimes damaged in transit. It gets cut off and wound onto a scrap winder.

How much comes off is your decision, not the machine's. In OptiStack the trim is whatever the pattern leaves over, split according to the trim mode:

• Both edges. The leftover is split evenly, with any odd fraction going to the right edge so the pattern stays stable against the datum.
• One edge. All of it comes off one side, which is what you want when the other edge is already good and you are working to a datum.
• None. Only valid when the pattern consumes the coil to within half a millimetre. Ask for it on a coil that does not and the solver refuses rather than quietly trimming anyway.

The practical point: trim is the difference between the coil you ordered and the pattern you are cutting. If it is bigger than you want, the coil width and the pattern are the two things to change, in that order.

A customer ordering 152 mm strip is ordering a band, not a number. Where you aim inside that band costs width, and over a full pattern it adds up.

Aim high and every strip is a little wider than nominal, which eats the trim you were relying on. Aim low and you risk strip that measures under and comes back. On an eight strip pattern, targeting 0.05 mm over nominal consumes 0.4 mm of coil that would otherwise have been trim. That is small, and it is exactly the kind of small that turns a pattern that fits into one that does not.

The width you actually get is the sum of the tooling stack that produced it, which is the subject of the slitting calculator. For where tolerance comes from mechanically, see cutting accuracy on the machine guide.

Same coil as the one on the coil calculator, so you can check every figure there: cold rolled steel, 1500 mm outside diameter on a 508 mm mandrel, 1250 mm wide, 1.00 mm gauge, density 7.85 g/cm³. That comes to 15,351 kg and 1,564 m of strip.

Cut it into 152 mm strips:

Weight scales with width at constant gauge, which is why the weight split is the width split. Nothing here depends on how well the line runs. 417 kg of that coil was never going to be product.

Now change one thing. A 1244 mm coil cuts the same eight strips with 28 mm of trim instead of 34, and yield goes to 97.75%. A 1500 mm coil fits nine strips with 132 mm left over, and yield falls to 91.2%. The pattern did not change and the line did not change. The coil width did.

• Coil width against the pattern. The biggest lever by a distance, and the one decided at purchasing rather than on the line. Work the pattern out before you buy the coil, not after it arrives.
• Combining orders on one pass. Two widths that together consume the coil beat one width that leaves a wide remnant. This is a planning decision, and it is where most of the recoverable width usually is.
• Where you aim inside the tolerance band. Small per strip, real across a pattern, and free to change.
• Arithmetic errors. A pack that does not come out to width is scrap that had nothing to do with the coil. This is a common cause of width problems and an avoidable one, which is what OptiStack removes: it builds the stack from the tooling you actually hold and refuses to print one your inventory cannot make.
• Edge condition on arrival. A damaged mill edge forces more trim than the pattern needs. Worth recording per supplier, because it is a cost you can put on a purchase order.

Published yield figures are not much use, because yield is a statement about your coil widths and your order book rather than about slitting. Work out your own, which takes one job:

1. Coil width as measured, not as ordered.
2. Sum of the strip widths in the pattern.
3. The first over the second is your yield by width for that job.
4. Multiply the coil weight by that fraction for the product weight, and the remainder is trim.
5. Record it per customer width. After a month you will know which orders pay and which ones you are trimming away.

The coil calculator does steps one to four from the numbers on the coil tag, in metric or imperial.