Search for a slitter clearance chart and you will find numbers that contradict each other. One source puts knife overlap for thin film at 0.5 to 1.0 mm. Another puts it at 0.30 to 0.50 mm for the same material. A third gives clearance as a percentage of thickness and says 5 to 10 percent for PET. All three are published by companies that make slitting equipment, and all three are wrong for somebody, because they are not describing the same measurement.
That is the trap. In shear slitting there are three separate settings, they act in three different directions, and the word “clearance” is used loosely for at least two of them. Until the terms are separated, a chart is worse than useless: it gives you a number to argue about instead of a method to find the right one. This article separates the terms, explains why published figures differ, and then gives the starting windows and the adjustment procedure that actually get a line to a clean edge.
Three Settings, Three Directions
Picture the two knives as a pair of scissors mounted on shafts. Everything you can adjust acts along one of three axes.
| Setting | Direction | What it controls | Units |
|---|---|---|---|
| Overlap, penetration or depth of cut | Vertical, along the line of cut | How far the top knife passes the bottom knife’s cutting plane | mm, or percent of web thickness |
| Side clearance or axial gap | Horizontal, across the knife faces | Whether the two cutting faces stay in contact or leave a gap the web can fold into | mm, or side load in kgf |
| Cant angle or tilt | Angular, in the plane of the cut | Whether the knives meet at a point or across their whole faces | degrees |
Overlap creates the scissor action. Side clearance decides whether that action is a shear or a fold. Cant angle decides how much of the knife face is rubbing at any moment, which controls heat and dust. A line can have perfect overlap and still produce a turned edge because the side setting is wrong, and it can have perfect side contact and still produce dust because the cant angle is too small and the faces rub flat.
The logic transfers to any blade that has to hold a controlled contact. A doctor blade on a gravure press has the same three variables in a different order: contact pressure, edge angle and blade thickness, with the same consequence if one is set by feel. Where a blade is designed to remove material rather than separate a web, the V-grooving blade is a useful reminder that depth control and edge condition are separate failure modes.
This is why the numbers in different charts do not reconcile. A chart quoting 0.5 to 1.0 mm for thin film is describing vertical penetration. A chart quoting 5 to 10 percent of thickness for PET is describing axial clearance on a machine that runs a fixed gap rather than a side-loaded knife. Both can be correct for their own machine type and both are dangerous if applied to the other.
Why the Published Numbers Disagree
Three influences explain almost all of the variation between sources, and none of them is a mistake by the publisher.
Machine type. A side-loaded male knife in a spring holder runs with the faces in contact and effectively zero axial gap. A gang slitter with fixed spacers runs a deliberate axial gap set by shim. The first machine’s “clearance” is a force; the second’s is a dimension. Comparing them numerically is meaningless.
Knife diameter. Penetration is not independent of geometry. A 100 mm knife and a 250 mm knife set to the same vertical penetration have different contact arc lengths, so the same number produces different cutting conditions. Charts rarely state the diameter they were written for.
Definition of success. Some sources publish the minimum overlap that produces a complete cut. Others publish a range that has proven stable at high speed with long knife life. Those are different objectives and they produce different numbers on the same material.
| Source type | What its numbers usually mean | How to use it |
|---|---|---|
| Machine builder’s manual | Settings validated on that machine, at that knife diameter | Start here. This is the only figure written for your equipment |
| Knife maker’s data | Geometry the edge was designed to run at | Use to check the builder’s figure is achievable |
| Percentage-of-thickness charts | Axial gap on fixed-clearance gang slitters | Use only on machines with a set gap, not side-loaded holders |
| Penetration tables in millimetres | Vertical overlap on shear lines | Use as a starting window, then tune to edge quality |
| Converter forum advice | One plant’s experience on one substrate | Treat as a hypothesis, not a setting |
Starting Windows by Material Family
The table below is a starting point for shear slitting on a side-loaded top knife, with the understanding that the builder’s manual overrides it. The last column matters as much as the numbers: it names the failure mode you will see if you go too far in either direction.
| Material family | Typical thickness | Starting overlap | Approx. share of thickness | If too little | If too much |
|---|---|---|---|---|---|
| Thin PET, BOPP, metallised film | 8 to 25 micron | 0.3 to 0.6 mm | Low single digits as a gap equivalent | Uncut fibres, angel hair, web not separated | Heat marks, melted build-up, hard edge |
| Medium packaging film, CPP, laminates | 30 to 80 micron | 0.45 to 0.75 mm | Low single digits | Tagging, ragged edge | Dust, edge curl, knife wear |
| Aluminium foil | 6 to 30 micron | 0.3 to 0.6 mm | Low single digits | Torn edge, pinholes at the edge | Raised lip, folded edge |
| Copper and aluminium electrode foil | 6 to 20 micron plus coating | Set by machine builder, often below 0.1 mm | Micron-level, gap-controlled | Uncut coating, delamination | Burr, coating damage, dust |
| Coated paper and label stock | 60 to 200 gsm | 0.8 to 1.5 mm | Mid single digits to low teens | Fibre pull-out, fuzz | Crushed edge, heavy dust |
| Board and heavy coated stock | 200 to 400 gsm | 1.2 to 2.0 mm | Low teens | Uncut, torn edge | Delamination, edge crush |
| Nonwoven and laminates | 20 to 150 gsm | 1.0 to 2.5 mm | Mid teens | Fibre protrusion, stringers | Heat-sealed edge, fused fibres |
| Corrugated and heavy laminates | Above 1 mm | 1.5 to 3.0 mm | Low teens | Incomplete cut, crushed flute | Edge damage, dust, knife load |
Two cautions on reading this table. First, the “share of thickness” column is not a conversion factor for the overlap column; it is there to show the order of magnitude, because a 0.5 mm overlap on a 12 micron film is forty times the web thickness and a 0.5 mm overlap on 1 mm board is half of it. The same absolute number is a different cut. Second, the windows widen as material gets thicker and more compressible, because a thicker web tolerates more penetration before quality collapses. Thin film is unforgiving in both directions, which is why it is the material that exposes a badly set line first.
Converting Thickness to a Setting
Where the machine uses a fixed axial gap rather than side load, the convention is to express the gap as a percentage of web thickness. This is the method behind most of the contradictory charts, and it is useful once you know it applies to a gap and not to overlap.
| Material | Gap as share of thickness | Reason |
|---|---|---|
| Paper and board | Highest of the common ranges, roughly a quarter to a third | Fibrous structure needs room to separate without crushing |
| Nonwoven | Middle range, roughly a sixth to a quarter | Compressible, but fibres must be severed not pulled |
| PET and other brittle film | Lowest, roughly a twentieth or less | Clean fracture needs tight control; too much gap produces micro-cracks |
| Aluminium foil | Around a tenth | Ductile, deforms easily; a wide gap leaves a sharp lip |
| Battery electrode foil | Micron-level, set by machine | Burr limit is driven by separator thickness, not by the foil |
Worked example. A 12 micron PET film on a fixed-gap gang slitter at 5 percent gives a 0.6 micron gap. That is below what most mechanical spacer stacks can hold repeatably, which is exactly why thin film is usually run on side-loaded holders where the faces are held in contact and the gap is a force rather than a dimension. If a chart tells you to set 0.6 microns with shims, the chart was written for a different machine.
Shear, Crush and Score Are Not the Same Process
A single chart cannot cover all slitting because the three modes separate material by different mechanisms. Applying shear numbers to a crush line is one of the most common setup errors in plants that run both.
| Mode | How it separates | What is adjustable | Dominant burr cause |
|---|---|---|---|
| Shear | Two edges pass each other like scissors | Overlap, side load, cant angle | Wrong overlap or open side clearance |
| Crush or score | A knife presses into an anvil or a hardened ring | Pressure, knife edge condition, anvil condition | Pressure used as a substitute for sharpness |
| Razor or tangential | A static or oscillating blade draws through the web | Blade angle, draw speed, blade condition | Blade angle and edge wear |
Crush cutting has no overlap to set. It has pressure, and pressure is not a quality control. When operators raise pressure to compensate for a dull knife, the edge compresses before it separates and the result is a hard, raised edge that looks like a burr but cannot be fixed by any geometric adjustment. The correct response is a knife change, not a pressure change. The same trap exists on a carton slotter knife, where operators add anvil pressure to recover a worn edge and end up crushing the flute instead of cutting it. Score cutting behaves the same way: the edge does the work and pressure only maintains contact.
Symptoms Point to a Specific Parameter
This is the practical use of the three-parameter model. Each symptom narrows the adjustment to one axis, which is what prevents the “change everything and hope” loop.
| Symptom | Parameter to adjust first | Direction |
|---|---|---|
| Burr on one side only | Overlap | Increase in small steps until the cut is complete, then stop |
| Turned or wavy edge | Side load | Increase until the faces stay in contact and the edge stops folding |
| Fine dust on rolls and idlers | Overlap or cant angle | Reduce overlap; verify cant so the knives meet at a point |
| Heat marks or melted build-up | Overlap | Reduce overlap, then reduce side load |
| Uncut tags or strings | Overlap | Increase overlap, or check edge condition first |
| Repeating defect once per revolution | Neither | Stop: measure runout and holder play |
| Burr varies across the web width | Neither | Stop: check spacer stack, parallelism and deflection |
Note how many rows end in “stop”. Geometry cannot compensate for runout, and no overlap figure fixes a stack with contaminated spacers. Roughly half of the lines we are asked about have a holder or stack problem behind a symptom that looks like a setting problem. That is also why the choice of supplier matters before the first delivery: a matched knife set with a documented bore and face tolerance removes one variable from every future changeover, while a mixed set from three sources adds three.
Prerequisites Before You Set Anything
Settings are only meaningful if the mechanical platform is sound. These seven items should be checked before a chart is consulted, because a line that fails any of them will not hold any setting repeatably.
| Prerequisite | Typical acceptance | What fails without it |
|---|---|---|
| Mounted knife runout | Around 0.01 mm or better | Effective overlap changes once per revolution |
| Holder radial play | Below roughly 0.02 mm | The knife wanders laterally under load; gap opens and closes |
| Shaft straightness and bearing condition | No measurable lift or end play | Whole stations drift; outer positions worst |
| Spacer and mounting face cleanliness | No contamination, no raised burrs | Cumulative error across the stack |
| Knife diameter consistency | Matched set within drawing tolerance | Overlap differs station to station |
| Parallelism of top and bottom shafts | Per builder’s alignment spec | One side of the line cuts differently from the other |
| Edge condition | No nicks, uniform edge radius | No setting produces a clean cut |
How to Tune Without Losing the Answer
The adjustment procedure matters as much as the starting value. Most failed trials fail because two variables moved at once.
- Record the current settings before touching anything, including how each was measured. A number without a method cannot be repeated.
- Set overlap conservatively, at the low end of the window for the material, and run one roll.
- Judge the edge under magnification, not by eye or by feel. Photograph it.
- Increase overlap in small increments only until the cut is complete. The moment it is complete, stop: additional overlap buys nothing and costs knife life.
- Add side load last, and only to the point where the edge stops folding. Excess side load accelerates wear on both knives and loads the holder bearings.
- Verify cant angle if the holder allows adjustment. Point contact is what keeps dust and heat down.
- Re-check at production speed, because a setting that works at 150 m/min may not hold at 400 m/min. Speed changes the thermal balance, not the geometry.
- Write the result into the changeover sheet with the material, thickness, speed and measured burr result, then check the consumable cost against current carbide pricing before the interval is set for the year.
| Variable | Practical step size | Why not larger |
|---|---|---|
| Overlap | Small increments, around 0.02 to 0.05 mm | Edge quality changes non-linearly near the threshold |
| Side load | Minimum increment the holder allows | Excess load increases wear without improving the cut |
| Cant angle | Tenths of a degree | Large cant concentrates load and shortens knife life |
| Line speed | Steps of 50 m/min during trials | Speed affects heat, not geometry; isolate it |
Recording the Setting So It Repeats
The value of a chart is that it makes the next setup faster. That only happens if the chart is replaced by your own record of what worked, on your machine, with your material.
| Record field | Why it earns its place on the sheet |
|---|---|
| Material, grade, thickness, supplier batch | Material variation is the most common reason a good setting stops working |
| Knife set ID, diameter, hours run | Ties edge quality to knife life and to regrind history |
| Overlap and how it was measured | A figure without a method is not repeatable |
| Side load or axial gap, with machine type noted | Prevents a force being mistaken for a dimension |
| Cant angle where adjustable | Often unknown in plants, and often the cause of dust |
| Line speed and tension | Needed to reproduce a good run, not just a good edge sample |
| Burr result at start, middle and end | Shows whether quality drifts within a roll |
| Ambient temperature or web temperature notes | Explains seasonal variation that otherwise looks like tooling failure |
Six Mistakes That Keep Coming Back
| Mistake | What it produces | Correct approach |
|---|---|---|
| Copying a setting from another material | Crushed edge on thin film, uncut fibres on board | Start from the window for that material family, then tune |
| Treating clearance as one number | Adjusting the wrong axis for hours | Name the direction: vertical penetration, axial gap, cant |
| Using pressure to cover a dull knife | Hard raised edge on crush lines | Change the knife, keep pressure at the minimum that holds contact |
| Chasing burrs with overlap when runout is the cause | Five knife sets, same defect | Measure runout and holder play first |
| Setting up at low speed and releasing at production speed | A process that passes trials and fails in production | Verify at the speed the line will actually run |
| Never recording the winning combination | Every changeover is an experiment | Changeover sheet with measured values |
When the Answer Is the Knife, Not the Setting
Geometry can only work with an edge that is capable of shearing. If the knife shows a rounded edge under magnification, a nick, or heat discolouration on the body, no overlap value will recover edge quality, and continued adjustment will damage the mating knife as well. The blade troubleshooting guide covers the recognition side of that decision, and the guide to slitting line burrs sets out the diagnostic order that separates knife faults from geometry faults.
Matched sets matter here. A new top knife running against a worn bottom ring will not hold a setting, because the reference face it is loaded against has changed. This is why experienced buyers replace the pair, and why a wear-resistant slitting lower blade is often the missing half of a fix that was being chased on the top knife alone.
For shear lines, the working geometry is defined by the circular slitting knife and its matching bottom ring. Abrasive or highly filled substrates are better served by a tungsten carbide slitter blade, which holds edge radius longer and therefore holds the setting longer. Wide lines with many active stations benefit from large round slitting blades that resist deflection at the outer positions. The range is listed under the round blade category, with grades compared in the tungsten and steel blade category.
Dekeer quotes slitting knives against the substrate, the machine type and the target edge standard rather than against a catalogue number, because the same diameter behaves differently across three materials. Send the material and thickness, the slitting mode, the knife diameter and bore, the current settings, and a magnified photo of the edge, and the geometry can usually be specified in one exchange. The contact page reaches the technical team, and the product overview lists the full range by application.
Related Reading
Two companion articles cover the decisions either side of this one. The slitting knife selection guide works through diameter, thickness and edge geometry for a new line, while our guide to extending blade service life covers how regrind history changes the geometry a holder can still achieve. On the material side, the material comparison and the coating overview explain what a harder or coated edge actually changes. For packaging lines where slitting and cross-cutting share the same web, the packaging machine blade guide covers the cut-off side of the process, and the heat treatment guide explains why nominal hardness alone does not predict edge life.
Frequently Asked Questions
| Question | Answer |
|---|---|
| What is the difference between overlap and clearance? | Overlap is vertical penetration along the cut line. Clearance is the horizontal gap between the knife faces. They act in different directions and are set by different mechanisms. |
| What overlap should I use for 12 micron PET? | Start at the low end of the window for thin film and increase until the cut is complete. On a side-loaded holder this is typically a fraction of a millimetre; on a fixed-gap gang slitter the equivalent gap is only a few microns. |
| Should side clearance ever be measurable? | On a side-loaded shear holder, no. The faces should run in light contact under a controlled side load. A measurable gap lets the web fold instead of shear. |
| Why does my machine builder’s figure differ from an online chart? | Different knife diameter, different holder type, or a different definition of the parameter. The builder’s figure is the one written for your machine. |
| Can I use the same setting for paper and film? | No. Paper needs more room to separate fibres; film needs tighter control to fracture cleanly. Using a paper setting on film produces heat and edge curl. |
| Does more side load improve the cut? | Only up to the point where the edge stops folding. Beyond that it accelerates wear and loads the holder bearings without improving quality. |
| How do I know if the setting or the knife is at fault? | Inspect the edge under magnification first. A rounded edge, nick or heat mark means the knife; a sharp edge with a bad result means the setting. |
| Is a coated knife worth it for slitting? | It changes wear rate and therefore how long the edge stays in the good zone. It does not substitute for correct geometry, which is covered in the coating overview. |
| How often should settings be verified? | At every material or thickness change, after any knife change, and at the start of a shift on lines that run several products. Verification takes minutes; a bad roll takes hours. |
Where to Read Next
If the symptom is an edge defect rather than a setting question, start with the guide to causes of slitting line burrs, which separates tooling faults from process faults. For the hardware side, the supplier qualification guide covers what to ask before ordering, and the custom wear-resistant blade page explains how a non-standard geometry is quoted. Lines where slitting feeds a sealing station should also read the heat sealing blade notes, because a slit edge that passes inspection can still fail at the seal.
Buying the Knives Behind the Setting
A setting is only as repeatable as the hardware it is applied to. Buyers specifying a new line should start from the slitting knife selection guide for diameter and edge geometry, then confirm the grade against the tungsten and steel blade range. The full catalogue sits under the round blade category, and a technical question about a specific substrate is best sent through the contact page with the eight parameters listed above.
The One-Paragraph Version
Name the axis before you name the number. Vertical penetration, axial gap and cant angle are three different settings, and published charts disagree because they describe different ones on different machines. Start from your builder’s figure or from the window for the material family, hold the mechanical prerequisites first, then tune one variable at a time in small steps until the cut is complete and the edge stops folding. Record what worked, with the method used to measure it, because the setting only has value the second time you need it.