CNC machining Kenosha: how a drawing becomes a finished part

• 1
  Key takeaways
• 2
  What CNC machining Kenosha buyers are really paying for
• 3
  How tolerance and surface finish interact
• 4
  Three-axis, four-axis and five-axis: picking the right setup
• 5
  Material behaviour on the shop floor
• 6
  From quote to shipment: what happens in between
• 7
  Which process route fits your part

A CNC machine does not decide the price of a part. The drawing does. When a job arrives as a STEP file and a 2D print, the shop reads four things before quoting: the tightest tolerance on the part, the number of setups needed, the material, and how the finished surfaces will be verified. Everything else follows from those four answers.

Take a simple aluminium bracket with one flat face and four drilled holes. Rough stock removal, one setup, drill, tap, deburr. That part is cheap because almost none of the work is uncertain. Now add a ±0.005 mm bore and a true position callout of 0.01 mm to two datums. The cutting time barely moves. What moves is the fixturing, the in-process checks and the scrap risk if a tool wears halfway through the run.

That is the core idea behind CNC machining in Kenosha and anywhere else. Precision is purchased in the inspection room and at the fixture bench as much as at the spindle. A shop quoting a tight tolerance without asking about datums, material condition or surface finish is either very good or not being honest about the risk.

So the useful question is not "how much per part". It is "which features on this part are actually tight, and what does each one cost me". Most parts have two or three critical features and a dozen free ones. Moving the free ones to a general tolerance of ±0.1 mm often removes more cost than changing suppliers.

• 1
  Tight toleranceDrives fixturing, tool selection and inspection frequency.
• 2
  Setup countEach re-fixturing adds a datum shift and a check.
• 3
  MaterialSets cutting speed, tool life and stock lead time.
• 4
  VerificationWhat gets measured, how often, and on which report.

Tolerance and finish are separate callouts, but they are produced by the same cut. A light finishing pass with a sharp tool can hold ±0.01 mm and reach Ra 1.6 μm. Push to Ra 0.4 μm and you are likely running a second pass at lower feed, sometimes with a different tool, and checking the surface with a profilometer instead of a visual.

The practical range at GreatLight runs from Ra 0.2–0.8 μm for fine finishes, Ra 0.8–1.6 μm for high-quality fits, and Ra 1.6–3.2 μm as-machined. Most functional mating surfaces sit in the middle band. Sealing faces and bearing bores move to the fine band. Cosmetic covers rarely need better than the as-machined range once they are anodized.

Here is the trap. A print that calls Ra 0.4 μm on every face will be quoted high, and half of those faces do not touch anything. Specify finish only where the function needs it. Same with tolerance. A general note of ±0.1 mm plus three or four toleranced dimensions is cheaper to make and easier to inspect than a blanket ±0.02 mm.

Also check whether a tight dimension is measured in the free state or clamped. Thin walls and long slender parts move after unclamping. If the print does not say, ask. That single question avoids most of the arguments that happen at incoming inspection.

• 1
  Fine finishRa 0.2–0.8 μm. Extra pass, slower feed, profilometer check.
• 2
  High finishRa 0.8–1.6 μm. Standard finishing pass on most alloys.
• 3
  As-machinedRa 1.6–3.2 μm. Fine for non-mating and coated faces.

A three-axis machine moves the tool in X, Y and Z while the part stays fixed. It is the cheapest way to cut a part whose features are reachable from a small number of faces. Flat plates, housings with open pockets and most turned parts fall here. GreatLight runs 27 three-axis machines and 12 four-axis mills for exactly this kind of work.

A fourth axis adds rotation about one axis, usually the X or the table. That lets you cut four sides of a prismatic part in one setup, which removes datum shifts and shortens the queue. If your part has features on four faces and a positional relationship between them, four-axis is often the cheapest way to hold that relationship.

Five-axis adds a second rotary axis. The tool can approach a contoured surface from an angle instead of straight down. That matters for impeller blades, turbine housings, complex brackets and any geometry where a ball nose tool would otherwise need a long reach. Short, stiff tools cut faster and chatter less. GreatLight has 16 simultaneous five-axis centers, including a Ø400 mm rotary table and a 4,000 mm maximum processing size.

Five-axis is not automatically better. On a simple plate it adds nothing and the hourly rate is higher. The honest test is this: how many setups would the part need on a three-axis machine, and can a shorter tool reach the critical surfaces if the part can be tilted? If the answer is three or more setups, or yes, five-axis usually wins.

• 1
  Three-axisSimple geometry, features on few faces, lowest rate.
• 2
  Four-axisFour-sided prismatic parts, one setup, tight face-to-face position.
• 3
  Five-axisContoured surfaces, undercuts, short tools, complex angles.

Material choice changes cutting data, tool life and sometimes the whole process route. Aluminium 6061 and 7075 cut fast and hold good finishes. 6061 is the general-purpose choice; 7075 is stronger but gummier and needs sharper tools and better chip evacuation. Both are stocked widely, so stock lead time is rarely the constraint.

Stainless 303 and 304 machine reasonably. 316 and 316L work-harden, so a light feed that rubs instead of cutting will destroy the surface and the tool. 17-4PH in the H900 condition is harder again. On these grades we slow the surface speed, keep the feed per tooth up and accept a longer cycle. That is a real cost, and it should show up in the quote rather than arrive as a surprise.

Titanium and Inconel sit at the difficult end. Ti-6Al-4V has low thermal conductivity, so heat goes into the tool edge. Inconel work-hardens aggressively and eats carbide. Both are machinable to ±0.005 mm, but cycle times are several times that of aluminium and tool changes are frequent. If a design can use 17-4PH or a 400-series stainless instead, the saving is usually large.

Plastics are their own case. POM and PEEK hold tolerance well; ABS and PP flex and can melt if the feed is too slow. Carbon fibre machines cleanly but the dust is abrasive and needs extraction. None of this is exotic, but it does mean the material line on a drawing is a process decision, not just a purchasing one.

• 1
  Aluminium6061, 7075, 2024, 6082. Fast, good finish, widely stocked.
• 2
  Stainless303, 304, 316L, 17-4PH. Watch work-hardening on 316.
• 3
  Titanium and InconelMachinable but slow. Heat and tool wear dominate cost.
• 4
  PlasticsPOM, PEEK hold tolerance. ABS and PP need care.

When a file arrives, the first output is usually a DFM note. It flags features that are hard to reach, walls that are too thin for the tool, tolerances that cannot be verified, and anywhere the drawing is ambiguous. At GreatLight the quotation and free DFM analysis come back within 12 hours, and production can start within 24 hours of approval. Parts ship in 3–5 days on standard work.

The route through the shop is fixed: raw material check, then machining, then any heat treatment or finishing, then final inspection. Inspection is 100% before shipment, with raw material, in-process and final checks recorded. Reports are available on request. If a dimension is critical, say so before the run starts so it can be measured on the machine and again at final.

Finishing is where schedules slip if it is not planned. Anodizing, electroless nickel, zinc, silver and gold plating, powder coating, black oxide, bead blasting, tumbling, brushing and polishing are all available, and laser marking needs a minimum character height of 1.5 mm. A hardcoat anodize adds a few days and can change dimensions on tight bores, so it belongs in the plan from the start.

Order size is flexible. There is no minimum order quantity, from one prototype to runs of 10,000 or more. That matters for Kenosha teams who need a first article to test before committing to a production batch. Uploads are kept secure and confidential, and an NDA is available on request.

• 1
  Quote and DFMWithin 12 hours, including manufacturability notes.
• 2
  Production startCan begin within 24 hours of approval.
• 3
  ShippingStandard parts ship in 3–5 days.
• 4
  Inspection100% before shipment, reports on request.