CNC Machining Automotive Lamps and Hand Plate Models

• 1
  Key takeaways
• 2
  Why a lamp housing is not just a pocket
• 3
  Hand plates and model parts: flatness under clamping
• 4
  Which aluminum suits a lamp housing
• 5
  Surface finish and how it is actually set
• 6
  Five-axis setup versus three-axis plus fixtures
• 7
  Matching the part to the process

A lamp housing looks simple on a screen. It is a bowl with a lens face, a mounting flange, sometimes a vent boss and a reflector seat. The cut is not simple. The wall is usually 1.5 mm to 3 mm. That wall has to survive the cutting force of a Ø10 mm end mill pushing sideways. When the wall is thin, the wall bends away from the cutter instead of the cutter shearing the material. You hear it before you see it: a ringing sound, then a gouge on the second pass.

The second problem is depth. A housing 120 mm deep with a 40 mm wide opening leaves very little room for a tool holder. A Ø6 mm end mill in a shrink-fit holder needs roughly 4× diameter in gauge length to reach the bottom without rubbing the shank on the rim. That is 24 mm of reach past the holder nose. Anything deeper needs a necked tool or a different setup angle.

This is where 5-axis work earns its place. Tilting the table 30° to 45° lets a short, stiff tool reach a wall that a 3-axis machine would need a long, flexible tool to reach. Short tool, less deflection, better wall thickness control. The trade is programming time and a slower feed on the tilted passes.

So the first question about any lamp housing is not tolerance. It is access. If the tool cannot reach the feature with a stiff setup, the tolerance on the drawing will not hold no matter what the machine can do.

• 1
  Wall under 1.5 mmExpect to rough with light radial engagement and finish in two passes.
• 2
  Depth over 4× tool diameterPlan a necked tool or a tilted 5-axis approach.
• 3
  One open face onlyAll other features must be reached from that side or from a second setup.

A hand plate is a different animal. It is long, thin and usually visible. A typical model hand plate might be 200 mm × 60 mm × 4 mm with a few holes and a chamfer. The cut itself is easy. Holding it is not. Six clamps on a 4 mm plate will bow it, and the bow stays in the part after unclamping. The plate springs back and the flatness you measured in the vise is gone.

The fix is to stop fighting the part. Vacuum fixturing or a low-melt wax bed spreads the holding force over the whole face. Where a vise is the only option, use soft jaws machined to the part profile and clamp on a thick sacrificial boss that gets cut off later. Clamp light. Take a 0.2 mm finishing pass after the last clamp adjustment so the final surface is cut in a relaxed state.

Model-scale parts add a second wrinkle: they are often judged by eye. A 0.05 mm step on a 1:18 scale model hand plate reads as a visible line under a raking light. That is not a tolerance issue, it is a surface continuity issue. The fix is usually a single continuous finishing pass with a constant stepover, not a tighter tolerance callout.

Both part families share one rule. Decide the holding method before you decide the cut. On thin work, the fixture is the process.

• 1
  Vacuum or wax bedBest for plates under 5 mm thick and over 150 mm long.
• 2
  Soft jaws on a bossClamp on material that will be removed, then finish after relaxing.
• 3
  Constant stepoverPrevents visible witness lines on cosmetic faces.

Most lamp housings and hand plates are aluminum, and the alloy choice changes the cut more than most designers expect. 6061-T6 is the default. It machines clean, holds a thread and takes anodizing well. Hardness is moderate, so a sharp carbide tool with a polished flute leaves Ra 0.8–1.6 μm without much effort.

2024 cuts faster and finishes brighter, but it is less corrosion resistant and anodizes to a duller, darker tone. If the part will be visible and clear anodized, 2024 is usually the wrong call. 7075 gives the best strength-to-weight and the best machined finish, but it is gummy at low cutting speeds and costs more. Use it where the housing carries load, not where it only holds a lens.

For heat sinks or brackets inside a lamp, ADC12 die casting alloy is sometimes supplied as bar stock for prototypes. It machines poorly compared to 6061. Chips pack, the surface tears, and you spend time on deburring. If the production part is die cast, a 6061 prototype is still the better choice for form and fit checks.

One more note. Aluminum moves with temperature. A 200 mm plate measured at 25 °C and then inspected in a 20 °C room can show 0.02 mm of apparent bow. Agree on the inspection temperature before you argue about the number.

• 1
  6061-T6Default for housings. Good finish, good anodizing, predictable.
• 2
  7075Highest strength and finish. Use where the part carries load.
• 3
  2024Fast cutting but duller anodized color. Avoid on visible clear-anodized parts.

Finish is not a number you request at the end. It is the result of four choices made before the last pass: stepover, tool runout, stock left for the finisher, and coolant. Get any one wrong and the Ra callout will not be met.

Stepover is the biggest lever. On a ball nose tool, a stepover of 0.1 mm on a Ø6 mm cutter leaves scallops around 0.4 μm high. Push the stepover to 0.3 mm and the scallops grow to about 3.7 μm. Same tool, same speed, seven times the roughness. If the drawing says Ra 0.8 μm, the finishing stepover has to be small and the toolpath has to keep it constant.

Runout matters more than most people think. A tool with 0.02 mm of runout cuts with one flute doing most of the work. That flute wears, the edge radius grows, and the surface starts to smear instead of shear. Check runout at the tool tip, not at the holder. Anything above 0.01 mm on a finishing tool is worth fixing before the pass starts.

For cosmetic surfaces, bead blasting hides a lot. A 120-grit glass bead blast turns a Ra 1.6 μm machined face into a uniform matte that reads as one surface. It also hides small tool marks. If the part is visible, blast it. If it is a sealing face, do not.

• 1
  Stepover drives scallop height0.1 mm stepover on Ø6 ball nose ≈ 0.4 μm scallop.
• 2
  Check runout at the tipKeep finishing tools under 0.01 mm.
• 3
  Blast cosmetic facesUniform matte hides witness lines. Never blast a seal face.

The usual argument for 5-axis is fewer setups. That is true, but it is not the main benefit on lamp work. The main benefit is tool stiffness. A 3-axis machine cutting a deep housing wall needs a long tool. A 5-axis machine tilts the part and uses a short tool. Short tool equals less deflection equals a wall that stays at 2.0 mm instead of drifting to 1.7 mm.

The cost is programming and cycle time. Tilted passes cut at a lower effective feed because the tool engages the wall at an angle. On a housing with a 1.5 mm wall, a 5-axis cycle can run 20% to 40% longer than a 3-axis cycle with the same toolpath length. You trade cycle time for dimensional control.

Where 3-axis still wins: flat plates, shallow pockets and any part where all features are reachable from one direction. A model hand plate with holes and a chamfer does not need a rotary table. Putting it on a 5-axis machine adds setup time and buys nothing.

The practical rule is simple. If the part has an undercut, a return flange or a swept surface that a straight tool cannot reach, use 5-axis. If it does not, use 3-axis and spend the time on the fixture.

• 1
  Choose 5-axis whenUndercuts, deep walls, swept surfaces, or one-setup datum control.
• 2
  Choose 3-axis whenFlat plates, shallow pockets, all features from one direction.
• 3
  Cycle time tradeTilted passes can add 20–40% to cycle time on thin walls.