GreatLight CNC Machining Factory logo
CNC Machining
Rapid Prototyping
Materials
Industries
News
About GL

Get Instant Quote

CNC buyer guide

5 Checks to Make when Selecting CNC Drilling and Milling Capacity

This guide is for engineers and sourcing staff who must pick a drilling and milling supplier, not a catalog number. It covers the numbers worth asking for, the ones that look impressive but tell you little, and the shop-floor details that decide whether your parts come back in spec.

±0.005 mm tolerance4,000 mm max sizeNo MOQISO 9001 / IATF 16949
Several indicators to check when selecting CNC drilling and milling machines
Quick answer

Key takeaways

Start with the part, not the machineTravel, spindle speed and axis count only matter against your largest and tightest feature.
Rigidity shows up on the finishLight passes and chatter at Ra 3.2 μm usually mean the setup, not the program.
Ask for measured numbersA supplier quoting ±0.005 mm should show you the last CMM report, not a spec sheet.
Certificates are a floor, not a ceilingISO 9001 and IATF 16949 tell you a system exists, not that it runs every shift.
Judgement table

What to ask and what the answer tells you

Use this as a shopping list when selecting CNC drilling and milling capacity.

IndicatorWhat to askValue to watchWhy it matters
TravelX × Y × Z for each machine4,000 × 400 × 150 mm for long partsDecides whether a part fits at all
Axis count3-axis, 4-axis or simultaneous 5-axis16 five-axis centers for contoured facesOne setup instead of four
Spindle speedMax rpm and taper sizeLow rpm for Ø30 mm holes in steelSpeed is not accuracy
Position accuracyMeasured repeatability, not catalog±0.005 mm on the tight facesDrives hole-to-hole location
RigidityBed material and massCast iron or polymer concreteSets chatter and tool life
InspectionWho measures, on what, how often100% inspection before shipmentCatches drift before shipping
Lead timeQuote, first cut, ship3–5 days after production startsPlan your assembly line
MOQMinimum batch and setup chargeNo minimum order quantityPrototype cost stays sane

The short version

Pick the shop that answers your part questions with measurements, not the one with the biggest catalog numbers. Travel, axis count, spindle torque, measured accuracy and inspection records decide the result.

Indicator 1

Work envelope and axis count when selecting CNC drilling and milling

The first question is not how fast the machine is. It is whether your part fits. A shop with 27 three-axis machines and 12 four-axis mills can handle most bracket work, but a 1,800 mm gearbox housing needs a machine with 4,000 mm of X travel. Ask for the travel of the specific machine that will run your job, not the largest number on the website.

Axis count changes the setup plan. A part with holes on five faces and a contoured pocket usually goes on a 5-axis center. The same part on a 3-axis machine needs multiple fixtures, and each refixture adds stack-up error. On a ±0.05 mm part that is fine. On a ±0.005 mm part it is not.

Four-axis mills cover the middle ground: round parts, slots on a diameter, holes on a cylindrical face. If your drawing shows a bolt circle on a turned flange, a 4-axis mill or a mill-turn center is often cheaper than a 5-axis job. We keep 16 mill-turn centers for exactly that shape.

One more check: the rotary table size. A Ø400 mm rotary table limits the swing of a turned-and-milled part. If your flange is Ø450 mm, that machine is out before the quote starts.

Indicator 2

Spindle speed, torque and the material you are cutting

Spindle speed gets quoted like a car's top speed. For drilling and milling, torque at low rpm matters more. A 20,000 rpm spindle with a small taper will drill Ø8 mm in aluminium all day. Put a Ø30 mm indexable drill into 4140 steel and it stalls or chatters.

Match the spindle to the hole. Deep holes in stainless 316L need lower surface speed and steady feed, otherwise the drill walks and the hole comes out oversize or tapered. High-speed spindles help with small tools and finishing passes, not with hogging out pockets in tool steel.

Air-bearing spindles at 120,000 rpm exist, but they suit PCB drilling and graphite, not general parts. If a supplier leads with rpm and never asks about material, they are selling a spec sheet. A short call about material, hole depth and thread size tells you more than the spindle rating.

Indicator 3

Rigidity, bed material and surface finish you can measure

Rigidity is the quiet indicator. It does not appear in a catalog, but you can see it in the finish. A light 0.2 mm finishing pass that comes out at Ra 0.8–1.6 μm means the machine and the setup are stiff. The same pass that chatters and reads Ra 3.2 μm means something is moving.

Bed material is part of this. Cast iron and polymer concrete absorb vibration better than a welded steel frame. Some builders mix materials: a steel base with a polymer-concrete column. The mix is not a problem if the measured finish holds across a long cut.

Ask how the shop checks rigidity in practice. A simple test: run a long thin rib in aluminium and measure the wall thickness top to bottom. If the wall varies by more than 0.05 mm, the tool is pushing away and the setup needs work.

Workholding belongs here too. Pneumatic chucks and vacuum chucks are fast, but they wear. Worn jaws let a part shift mid-cut, and the error shows up as a step on a face that should be flat. Ask how often chucks are checked and rebuilt.

Indicator 4

Accuracy claims and the inspection behind them

Position accuracy is the number buyers argue about most, and the one most often copied from a brochure. The useful question is not the catalog value. It is what the shop measured on the last similar job, on a warm machine, after a full shift.

A ±0.005 mm claim on a small precision part is credible on a good machine with the right setup and a temperature-stable room. The same claim on a 4,000 mm long part needs a different conversation about thermal growth and how the part is supported.

Inspection is where the claim becomes real. Raw material check, in-process monitoring, final inspection: three stages, each with a record. Reports should be available on request, and a first article report on the first batch is normal. A shop that cannot show a CMM report for a tight part is asking you to trust a number.

Do not accept a single sample measurement. Ask how many pieces were measured, with what instrument, and what the spread was. A tight average with a wide spread is a process problem waiting for your next order.

Indicator 5

Lead time, MOQ and the questions a quote should answer

Lead time has three parts: quote, first cut, ship. A quote that arrives in 12 hours with a DFM note is useful. A quote that arrives in a week with no comment on a thin wall is not. Ask when production can start and when parts ship, as two separate numbers.

Batch size changes the answer. A shop set up for 10,000-part runs may not want a single prototype, and a prototype shop may not hold tolerance at volume. No minimum order quantity matters for a one-off fixture; a stable process matters for a 5,000-part run. Ask both.

On price, ask what is included. Setup, fixtures, inspection, surface finish and packing are separate cost lines at most shops. A low number that excludes anodizing or a CMM report is not a low price, it is a partial quote.

Finally, ask about confidentiality. Drawings for a new product are sensitive. A shop should be able to sign an NDA and keep files secure. That is a process question, not a paperwork question.

How to run the check

Step by step: shortlisting a drilling and milling supplier

Work through these in order. The first two steps usually remove half the list.

  • 1
    List your critical featuresWrite down the largest overall size, the tightest tolerance and the deepest hole. These three numbers drive machine selection.
  • 2
    Match travel and axis countCompare the list against machine travel. Long parts need 4,000 mm of X. Multi-face work needs 4-axis or 5-axis.
  • 3
    Check spindle and tooling fitConfirm the spindle can run your largest drill and tap. Ask for the taper size and the low-rpm torque.
  • 4
    Ask for measured accuracyRequest a first article report or a recent CMM result on a similar part. Judge the spread, not the average.
  • 5
    Confirm inspection stagesRaw material, in-process and final. Ask who signs off and whether reports ship with the parts.
  • 6
    Agree lead time and MOQGet quote time, production start and ship date in writing. Confirm whether a single prototype is accepted.
  • 7
    Settle finish and packingState the surface finish and whether anodizing, plating or marking is included in the price.
  • 8
    Sign the NDA before files moveSend drawings only after the confidentiality terms are agreed.
FAQs

Questions buyers ask before placing an order

Is a 5-axis machine always better than a 3-axis machine?

No. For a flat plate with holes on one face, a 3-axis machine is faster and cheaper. Five-axis pays off when the part has features on multiple faces or a contoured surface that would otherwise need several fixtures.

Each refixture adds error and handling time. Count the setups first, then decide.

How do I judge a ±0.005 mm tolerance claim?

Ask for a measured result on a similar part, not a catalog figure. A credible answer includes the instrument, the number of pieces measured and the spread.

Also ask about temperature. A tight tolerance on a long part depends on how the shop controls heat during the cut.

What surface finish should I expect from milling?

As-machined surfaces usually land around Ra 1.6–3.2 μm. A controlled finishing pass can reach Ra 0.8–1.6 μm, and fine finishing on a stable setup can reach Ra 0.2–0.8 μm.

Specify the finish only on the faces that need it. Calling out a fine finish everywhere adds cost with no function.

Does a low MOQ mean lower quality?

Not by itself. A shop can run one prototype on the same machine and inspection path as a production part. What matters is whether the process is documented and the first article is measured.

Ask how the prototype and the production run are set up. If the answer is the same machine and the same fixture plan, the low MOQ is a scheduling choice, not a quality shortcut.

Which certifications should I look for?

ISO 9001:2015 covers general quality systems. IATF 16949:2016 applies to automotive work, ISO 13485:2016 to medical devices and ISO 27001:2022 to information security.

Match the certificate to your industry. A certificate outside your sector tells you less than one inside it.

How early should I involve the machine shop?

Before the design is frozen. A DFM review can move a feature to a face that needs one setup instead of three, or relax a tolerance that costs money and adds nothing.

Send the model and the critical dimensions. A short review at that stage usually saves a revision later.

Send your drawing and get a real answer

Upload your model and critical dimensions. We return a quote and a free DFM analysis within 12 hours, with the machine plan and inspection path written out.

12-hour quote100% inspectionNDA on request

Follow our work

More from the shop floor

We publish setup notes, tooling trials and inspection data from the factory floor.

FacebookTikTokYouTubeLinkedInInstagramThreadsPinterest

Trusted by engineers and manufacturers worldwide

Tesla Ford Motor Company BYD Auto Denso Magna International Boeing Airbus Medtronic KUKA FANUC