CNC Machining China Quality Control

Why is quality the biggest TCO lever?

Scrap, rework, re-measurement, schedule slip, and field failure are all quality costs, and they are all hidden in the TCO of a part. A process that is capable — Ppk ≥ 1.67 at the initial study on key characteristics — produces so little scrap that these costs essentially disappear. The cheapest part is the one that is made right the first time.

Which certifications back the quality system?

Standard	Certificate number	Scope
ISO 9001:2015	116024/A/0001/UK/En	Machining of metal parts
IATF 16949:2016	131941/A/0001/SM/En	Manufacture of CNC metal machining parts (excludes product design, Clause 8.3)
ISO 14001:2015	116024/B/0001/UK/En	Environmental management related to machining of metal parts

Honest scope note: IATF 16949 excludes product design under Clause 8.3. We are a build-to-print manufacturer.

How are dimensions verified before shipping?

• Hexagon GLOBAL S CMMs (2) for dimensional verification, plus vision measuring machines and a gear measuring center.
• KEYENCE VK-X3000 laser microscope and SURFCOM 1800D for surface finish and profile.
• XRF (iCHEQ) spectrometer for incoming material composition checks against the mill certificate.
• HVS-1000A micro-Vickers hardness tester, Fischer coating thickness gauge, salt spray and tensile testers.
• SPC: control charts on key characteristics, with Ppk ≥ 1.67 at the initial study on critical characteristics and Cpk ≥ 1.33 in production on general characteristics.

How is process capability measured?

Characteristic class	Cpk target	What it means
Key characteristics	≥ 1.67	The process is stable and centered well inside the tolerance band
General characteristics	≥ 1.33	The process meets standard capability expectations

On a representative length characteristic (76.1–76.5 mm), our SPC shows Cpk = 1.851 (one characteristic on one part, not a blanket capability claim).

One more check on that number: the 0.4 mm band behind it is wide by design. Hold the same process spread against a ±0.005 mm feature and it fails outright, so read this figure as evidence about one characteristic on one run, not as a blanket tolerance claim.

How does material verification cut TCO?

A substituted alloy is a field failure — the most expensive cost, because it happens at your customer's site. XRF verification at the door removes that entire class of TCO before it can happen.

How are parts traced back to the raw material?

Our MES/ERP system records raw material source, process parameters, machine data, batch, machining time, operator, and inspection results on every lot, so a recall is contained, not a full-line shutdown.

What documentation ships with a quality part?

• CMM dimensional reports for critical dimensions.
• SPC control charts for process-critical characteristics.
• Material certificates, with XRF verification.

Where a defect is caught decides what it costs you

The same non-conformance carries a different price depending on how far it travels. This is the escalation a quality system is built to stop.

Where it is found	What it costs	What catches it here
At the machine, during the run	One part and a tool change	First-part measurement before the batch continues, and control charts on key characteristics
At final inspection	The batch is sorted, and the schedule moves	Hexagon GLOBAL S CMM verification of critical dimensions
At your incoming inspection	Your labor, a rejection process, and a replacement lead time	Inspection records shipping with the lot so you can reduce or skip duplicate checking
On your production line	Line stoppage plus expedited freight	Capability held at Ppk ≥ 1.67 at the initial study on critical characteristics rather than sampling for hope
In the field, after assembly	Warranty, teardown, and your customer's trust	XRF material verification against the mill certificate, hardness on the HVS-1000A, coating thickness on the Fischer gauge, salt spray where corrosion matters

Every row down that table multiplies the cost of the same underlying mistake. That is the entire economic argument for measuring capability instead of inspecting outcomes.

The document set that settles a dispute instead of starting one

When a lot is questioned, the argument is won or lost by records that already exist. The set available here is deliberately specific: CMM dimensional reports naming nominal, actual and deviation per feature; SPC control charts naming the characteristic, the tolerance band and the subgroup count, against Cpk targets of 1.67 on critical and 1.33 on general characteristics; material certificates with the XRF verification that was run against them; coating thickness, hardness, salt spray or tensile results where the print calls for them; and PPAP or FAI packages for automotive and regulated work on request.

Behind those documents sits the lot record in the MES and ERP system, covering raw material source, process parameters, machine data, batch, machining time, operator and inspection results. That record is what turns a disagreement into a traceable question with one answer, and it is why containment can be limited to the affected lot rather than applied to every part you hold.

How to specify inspection so you pay for it in the right place

• Mark critical characteristics on the drawing. Capability tracking and CMM time then concentrate where function lives instead of spreading evenly across features that do not matter.
• State sampling expectations. A 100% check on a feature that a capable process holds at Ppk ≥ 1.67 at the initial study on key characteristics buys very little; a 100% check on a safety feature buys a great deal.
• Give datums, not only sizes. A relationship stated as a size stack forces conservative machining and heavier inspection.
• Separate cosmetic from functional finish. Ra called across the whole part adds inspection everywhere.
• Name the documents you want with the lot, so they are produced with the parts rather than reconstructed afterwards.

A drawing built this way lowers total cost twice: once in machining, because effort follows function, and once at your goods-in bench, because the paperwork that arrives is the paperwork you needed. To agree a document set for a specific part, email [email protected] with the characteristics you want covered.

What Cpk means in parts per million

Capability is usually quoted as a Cpk number, but the number only becomes useful when you translate it into what it predicts about scrap. Cpk is the ratio of the tolerance band to the process spread, and a higher number means the spread fits the band more times over, which collapses the non-conformance rate.

Cpk	Rough non-conformance rate	What it means for you
1.00	On the order of 2,700 parts per million	The process just fills the band; scrap is a regular event
1.33	On the order of 63 parts per million	Standard capability expectation; occasional scrap
1.67	On the order of 0.6 parts per million	The process is deep inside the band; scrap is rare enough to plan around

These figures assume a centered, stable, normally distributed process, so they are a guide rather than a guarantee, but the direction is what matters. Moving from 1.33 to 1.67 is not a small refinement; it is roughly two orders of magnitude less scrap, which is why the floor holds Ppk ≥ 1.67 at the initial study on critical characteristics. On a representative length characteristic (76.1–76.5 mm), the running SPC shows Cpk = 1.851 (one characteristic on one part, not a blanket capability claim), comfortably past the target.

How SPC turns a one-off check into a prediction

A single measurement tells you about one part. Statistical process control tells you about the next thousand, because it watches the process rather than the product. The mechanism is a control chart fed by small subgroups taken through the run.

• First-part measurement before the batch continues catches a setup error at the cost of one part instead of the whole lot.
• Subgroup samples through the run catch drift as it happens — a tool wearing, a temperature shift — before it produces out-of-spec parts.
• Control limits computed from the process's own history flag when the process has changed, not just when a part has failed.

The difference from inspection is the direction of time. Inspection sorts parts after they are made; SPC predicts before the bad part happens. A control chart that names the characteristic, the tolerance band, and the subgroup count is the evidence that a process is under control, which is the cheapest possible place to catch a defect, because nothing has been shipped yet.

What PPAP and FAI give a buyer

For automotive and regulated work, a lot is not just shipped — it is documented against a submission that proves the process can make the part. Two packages matter most, and they answer different questions.

• FAI (First Article Inspection) measures every dimension on a first-off part against the print and records each one. It answers "did the process produce a correct part?" before volume starts.
• PPAP (Production Part Approval Process) adds the process evidence around that first article — the control plan, the capability study, the material certification, and the measurement system — so the customer can approve the whole route, not just one part.

These packages are available on request here, and they matter to TCO for a simple reason: the cost of proving a process once, before volume, is a fraction of the cost of discovering a process problem after ten thousand parts are in your inventory. The documentation is not bureaucracy — it is the cheapest insurance you can buy.

How traceability follows a part from raw stock to dock

Traceability is the ability to answer "where did this lot come from and what happened to it" without interviewing people or guessing. The answer lives in the MES and ERP lot record, which carries the part's history as data rather than as memory.

• Raw material source and mill certificate, plus the XRF composition check run against it at the door.
• Process parameters and machine data for each operation, so a deviation can be traced to the exact step and machine.
• Batch, machining time, and operator, so a question about the run has a specific answer, not a recollection.
• Inspection results, so the measurements that shipped with the lot are tied to the lot itself.

The TCO payoff is containment. When a problem is found, traceability limits the response to the affected lot instead of a full recall of everything you hold. Without it, the same defect forces you to quarantine and sort every part from every batch — a cost that multiplies with your inventory instead of staying proportional to the problem.

What this shop actually runs

Ask for capability evidence and the answer should be specific. The facility behind this guide runs 100+ CNC machines on a 6,800 m2 floor in Dongguan, China, with 100+ staff. The machine list, by category:

• 5-axis machining centres (10+) - Mazak VARIAXIS I-600 (510/910/510 mm travel, 12,000 rpm, 22 kW, 71.6 N·m, 30-tool BT40) and Mazak INTEGREX I-250H S (B-axis 0.0001°, 24 kW, 124 N·m, 38-tool HSK-A63). Run on compound-angle parts, single-setup multi-face prismatic work, and aerospace fittings where datum transfer loss is unacceptable.
• 3-axis / 4-axis machining centres (50+) - Brother SPEEDIO S500Z2N / S700Z2N (10,000 rpm, 14/21-tool ATC, 0.7 s tool-to-tool, Z-axis 2.2 G) and Mazak VCE-430A L / VCE-570C L (10,000 rpm, 118 N·m, 24-tool ATC). Run on prismatic parts, plates, brackets, housings.
• Swiss-type turning centres (10+) - Citizen Cincom A20-3F7 (Φ20 mm, 10,000 rpm), TSUGAMI B0206-III / B0205-III (Φ20 mm, 0.1 µm), STAR SR-32JII / SR-32J Type N (Φ32 mm, 8,000 rpm). Run on long, slender rotational parts at ±0.002 mm.
• CNC turning centres (20+) - HEADMAN T70M/1250 (720 mm swing, 117 mm bore, 3,000 rpm, BMT75 12-station turret). Run on shafts, bushings, flanges up to ±0.002 mm.
• Wire EDM (5+) - Sodick AQ535L (550/350/300 mm travel, ±20° taper, 800 kg workpiece). Run on hardened tool steel, internal splines, sharp internal corners that milling cannot reach.
• Inspection: Hexagon CMM + 20+ gauges - Hexagon CMM for coordinate measurement, plus vision system, profilometer, hardness tester, coating thickness gauge, XRF for material verification. The 10:1 rule (gauge 10x more accurate than the band) is applied to every feature.
• Machine highlights - VARIAXIS I-600: one-setup multi-face 5-axis + MAZA-CHECK auto-calibration; INTEGREX I-250H S: DONE-IN-ONE mill-turn, B-axis 0.0001° + Ai thermal shield; SPEEDIO S500Z2N / S700Z2N: Z-axis 2.2 G + 0.7 s ATC; Swiss-type: guide-bush switchable + 0.1 µm resolution.

Tolerances held: milling ±0.003 mm, turning ±0.002 mm, grinding and wire EDM ±0.002 mm. Capability: Ppk >= 1.67 initial and Cpk >= 1.33 in production on key characteristics.

Certifications: ISO 9001:2015 (116024/A/0001/UK/En), IATF 16949:2016 (131941/A/0001/SM/En, clause 8.3 design excluded, disclosed openly) and ISO 14001:2015 (116024/B/0001/UK/En), issued to Dongguan Zhuohang Ltd, USCC 91441900MA4UWT6X39, established 2005.

Service area: Austria, Belgium, the Czech Republic, Denmark, France, Germany, Greece, Hungary, Ireland, Italy, the Netherlands, Spain, Sweden and the United Kingdom in Europe; Brazil, Canada, Mexico and the United States in the Americas; China, Israel, Japan, Singapore and Sri Lanka in Asia; Australia and New Zealand in Oceania; and South Africa in Africa — built in China, shipped worldwide.

Contact: Felipe Iglesias, [email protected], +86 13126536955.

Frequently Asked Questions

Standards referenced on this page

• IATF 16949:2016 — Automotive quality management system standard
• ASTM A582/A582M-12 — Standard Specification for Free-Machining Stainless Steel Bars
• ISO 14001 — Environmental management systems — Requirements with guidance for use