HOMAG nested CNC efficiency: what actually transfers to metal parts

• 1
  Key takeaways
• 2
  What HOMAG nested CNC efficiency actually measures
• 3
  Which parts suit a nested workflow
• 4
  Toolpath order is where the efficiency is won or lost
• 5
  What tolerance a nested run can hold
• 6
  How we run nested metal work at GreatLight
• 7
  Designing a part for a nested run
• 8
  When nesting pays, when it does not

A nested machine does not cut faster than any other machine of the same spindle power. It wins by cutting more useful parts per hour of machine time. The software packs outlines onto one sheet, orders the toolpaths so the sheet stays rigid until the last pass, and keeps the spindle moving instead of waiting on an operator to load the next blank.

That last point is the whole idea. On a conventional vertical mill, a 40-part order means 40 clamps, 40 zero-points and 40 chances to scrap a part. On a nested router, it means one program, one datum and one unload. Setup time is not reduced by a percentage. It is removed from the workflow.

So when we talk about HOMAG nested CNC efficiency on this page, we mean three measurable things: sheet yield, spindle uptime and setup minutes per part. Everything else is marketing. A nested run that saves material but adds an hour of vacuum prep has saved nothing.

Metal changes the arithmetic. Wood panel is 18 mm MDF with a stable, uniform density. Aluminum 6061 plate at 6 mm will spring, chatter and lift under a climb cut. The nesting logic survives the move to metal. The fixturing assumptions do not.

Start with geometry. If a part can be cut from flat stock with a Z depth under roughly 50 mm and most of its features reachable from one face, nesting is a candidate. Brackets, plates, housings, gaskets, brackets with pockets, covers, manifolds with shallow channels and prototype panels all fit this description.

Then check the second face. Nested work assumes one setup or a flip onto a dedicated fixture. A part that needs five different approaches, deep cavities on both sides and a contoured 3D surface is a 5-axis job, not a nested job. Forcing it into a nest adds handling steps and fixture error.

Material cost pushes the decision too. A 300 × 300 × 12 mm titanium plate costs real money per square meter. Offcut from that plate is scrap you paid for twice. When the material is 7075 aluminum or Ti-6Al-4V, sheet yield of 70% versus 55% changes the part price more than any spindle-time saving.

Quantity matters less than variety. Nested runs are excellent for a mixed batch: 12 of one bracket, 30 of another, 8 of a third, all cut in one program from one plate. That is the case where HOMAG nested CNC efficiency is easiest to see, because the alternative is three separate setups on three separate days.

In a nested program, the sequence of cuts decides whether the plate stays flat. Cut the outer profile of every part first and the sheet turns into a loose pile of parts that chatter on the last pass. Cut the deepest pockets first and the plate weakens before the finishing tools arrive.

The working order we use is: drill and tap, then rough pockets, then finish pockets, then cut the outer profiles in an order that leaves a spine of material connecting the parts until the end. That spine keeps the plate rigid. It also keeps the vacuum seal intact.

Tabs are the practical version of this. A tab is a small uncut bridge, usually 0.5-1.5 mm of remaining material, that holds a part in place until you snap it off. On a 2 mm aluminum cover, tabs replace the vacuum entirely. On a 12 mm steel bracket, you may not need them at all.

Climb milling matters more on a nested job than on a one-off. A conventional cut lifts thin plate and pushes the tool away from the wall. A climb cut on a rigid spine of material gives a cleaner wall and less re-cutting. On aluminum we run climb with air blast; on stainless, climb with flood coolant.

Nesting does not set the tolerance. The machine, the fixture and the thermal state of the plate do. On a well-clamped 10 mm aluminum plate, a nested 3-axis run holds ±0.05 mm comfortably and ±0.02 mm with a finishing pass and a warm-up cycle. Pushing to ±0.005 mm on a nested part is possible, but only with a stable fixture and small depth of cut.

The failure mode is not the toolpath, it is the material moving after the cut. A 2 mm aluminum cover milled from a 6 mm plate will bow when 4 mm of stressed material comes off one side. The part was in tolerance on the machine. It is out of tolerance on the inspection table.

That is why we normally leave a finishing allowance of 0.3-0.5 mm per side, then take a light spring pass. Stress relief before machining, or a rough and re-clamp cycle, both solve the same problem. On titanium and 17-4PH, a rough cycle followed by a stress-relief pause is standard.

Surface finish follows the same rule. Nested routing with a 12 mm end mill on aluminum lands around Ra 1.6-3.2 μm as machined. A finishing ball or bull nose tool brings that to Ra 0.8-1.6 μm. Bead blasting after tabs are removed hides the tab witness marks.

GreatLight runs 127 high-precision CNC machines across three plants, 7,600 m² in Dongguan and a plant in Singapore. Sixteen of those are simultaneous 5-axis centers, twelve are 4-axis mills and twenty-seven are 3-axis machines. Flat nested plate work usually lands on the 3-axis and 4-axis machines, with a Ø400 mm rotary table handling the flip.

Maximum processing size is 4,000 mm, and the largest travel envelope is 4,000 × 400 × 150 mm. That covers long extruded housings and large panels that would not fit a standard VMC table. Smaller nests run in 750 × 1,150 × 550 mm or 600 × 600 × 600 mm envelopes.

Inspection is 100% before shipment, with raw material check, in-process monitoring and final inspection. Reports are available on request. Quotation and a free DFM analysis come back within 12 hours, and production can start within 24 hours. Parts ship in 3-5 days on a standard nested run.

Certifications covering this work are ISO 9001:2015, IATF 16949:2016, ISO 13485:2016 and ISO 27001:2022. There is no minimum order quantity, so a single nested prototype and a 10,000-part run use the same workflow.