Nesting & Sheet Optimization

Nesting is the craft of arranging your parts on the sheet to use as little material - and often as little cutting time - as possible. It sounds like housekeeping, but it is one of the biggest levers on the cost of 2D fabrication, because sheet stock and machine time are the two things you actually pay for. A layout that leaves generous margins and gaps everywhere might use two sheets where a tight nest would use one; that is a doubling of material cost before a single part is any different.

The measure is utilisation (or yield): the area of your finished parts divided by the area of sheet consumed. A loose layout might sit at 40-50% utilisation - half the sheet becomes offcut and dust. A well-nested layout of mixed parts can reach 70-85%, and simple rectangular parts can go higher still. Every point of utilisation you gain is material you did not buy and waste you did not create.

You can nest by hand - dragging parts around the sheet in your CAD or laser software, rotating them to fit, tucking small parts into the gaps between big ones - and for most studio jobs that is enough. Larger or repeat production uses automatic nesting software that packs shapes algorithmically, honouring rotation limits and spacing rules. Either way the principles are the same: keep a small consistent spacing between parts (enough that adjacent cuts do not overlap or char each other), push parts to the sheet edges, and fill the holes in big parts with small ones. Think of it as packing a suitcase, not scattering it.

The most powerful nesting trick is the shared (or common) cut line. When two straight-edged parts sit right next to each other, the single cut that forms one part's edge can also form its neighbour's edge - one pass of the beam does the job of two. Line up a whole row of rectangles this way and you cut each internal edge only once instead of twice, which both halves that cut length (less machine time) and removes the gap between the parts (less wasted sheet). It is the closest thing to a free lunch in laser cutting.

There is a subtlety, and it goes back to kerf. On a shared line, the single kerf sits between the two parts, so each part loses half a kerf on that shared edge - both come out slightly smaller than if cut separately. For most parts this is negligible, but where the shared edge is a critical dimension, compensate for it. There is also a small heat consideration: cutting long adjacent lines concentrates heat, so very thin or delicate materials can char more along a shared edge.

Shared-line cutting works best for straight-edged, rectilinear parts - panels, strips, box faces, grids - which is exactly the geometry a lot of architectural and furniture work uses. Curved or irregular parts cannot truly share a line, but they can still be nested tight, tucked concave-into-convex so one part's bulge fits another's dish. Recognising when parts can share edges, and orienting them so they do, is one of the fastest ways to cut a job's cost.

Nesting is not a free-for-all, because many sheet materials have a direction that constrains how you may orient parts. Plywood has a face grain, and both strength and appearance follow it: a thin part is far stronger cut along the grain than across it, and the visible grain direction matters on show surfaces. Wood veneer is entirely about grain direction for looks. Extruded acrylic and some plastics have a slight directional character from how they were made, and cardboard and corrugated board have a distinct grain (and flute direction) that decides how they fold and bend.

This collides with pure area-efficiency. The tightest possible nest might want to rotate a part 90 degrees to fill a gap - but if that part is a thin plywood leg that must be strong, or a veneer panel whose grain must run vertically, you cannot. So real nesting is constrained packing: pack as tight as possible within the rotations each part is allowed. In practice you tag parts as grain-locked or free-to-rotate, place the locked ones first in their required orientation, then fill around them with the free ones.

This matters most for anything that will bend or take load. A living hinge (next lesson) flexes best when its cuts run a particular way relative to the grain; a shelf or bracket resists load best along the grain; a folded card model creases cleanly only with the grain. Getting utilisation up is good, but never at the cost of a part that snaps or a fold that tears - direction is a hard constraint, efficiency is the optimisation inside it.

As parts finish cutting they become loose in the sheet, and loose parts cause trouble: small pieces can drop through the honeycomb bed, shift out of position (so a later cut lands wrong), or lift and flare in the beam. The fix is the micro-tab (or bridge): a tiny uncut gap - often just 0.2-0.5 mm - deliberately left in the cut outline so the part stays tacked to the surrounding sheet until you snap it free by hand afterwards.

Tabs are especially worth it for small or intricate parts, for jobs you want to lift off the bed as one intact sheet, and for delicate materials that curl. You place a few small tabs around each part's perimeter (avoiding show edges and fragile points), cut the job, then break the parts out and lightly sand or file the little nub the tab leaves. Some laser software adds tabs automatically; otherwise you draw the tiny gaps into the outline yourself. The trade-off is that little clean-up nub, so use tabs where you need holding, not everywhere.

Two related habits belong here. Lead-ins/lead-outs are less common on lasers than routers but the idea of where a cut starts and ends still matters for clean edges. And the cut order from Lesson 2.1 works hand-in-hand with tabs: engrave first, cut inner features next, and cut outlines last - with tabs ensuring that even the last outline does not let a part escape before the job is done. Together, tabs and cut order are how a busy sheet of many parts comes off the bed complete and in register, not as a scatter of shifted, half-cut pieces.

Put it together and the cost of a laser job is easy to reason about: it is mostly the sheet material consumed plus the machine time, and machine time is driven by total cut length (every millimetre the beam travels while firing), the number of pierce points (starts), and any engraving area (raster engraving is slow because it sweeps a whole area). Nesting attacks all of these at once - tighter packing uses less sheet, shared lines cut total length, and sensible ordering reduces wasted travel.

Batching multiplies the gains. If you need ten of a part, nesting all ten together (and across the whole sheet) beats cutting them one at a time, because you amortise setup, share more edges, and fill the sheet more completely. When several small jobs share a material and thickness, nest them onto one sheet together rather than starting a fresh sheet for each. This is exactly why file-to-factory makes variation cheap, from Module 0: a hundred different parts nested on a sheet cost about the same to lay out as a hundred identical ones - the machine does not care that the files differ, only how much it must cut and how much sheet it eats.

A rough worked feel: imagine a sheet costs a fixed amount and the machine bills by the minute. Loosely laid out, a set of parts spans two sheets and forty minutes of cutting; nested tight with shared lines, the same set fits one sheet and twenty-eight minutes. That is half the material and a third less time for the same parts - purely from the arrangement. Nesting is the cheapest optimisation in fabrication because it costs only a few minutes of thought before the machine ever runs.