Most swiss turned parts business is program business. The same small, precise component is released month after month, and your risk is not whether the supplier can make the part once; it is whether the supplier can make the part the same way every time, lot after lot, while the bar lots change, the tools wear, and the setups are re-done. That is a different risk from a one-off order, and it needs a different verification.
Sourcing a swiss turned parts program therefore focuses on the repeatability layer rather than the novelty layer. The entity and the certificates are checked as with any supplier, and then the program-relevant questions take over: how the bar is verified and segregated, what the process evidence shows across production lots, how the first article anchors the configuration, and what controls catch drift before it ships.

What is different about sourcing a program rather than a one-off?
A one-off order tests whether the supplier can interpret a drawing and machine a part. A program tests whether the supplier can hold a proven process steady across months of releases, and the difference shows up in every part of the relationship. The drawing is settled, the process is proven, and the supplier's job becomes repetition with control: the same bands, the same finish, the same documentation, lot after lot, while the variables that can move a process — tool wear, bar lots, setups, operators — are managed rather than eliminated.
your verification shifts to match. The questions are no longer only "can you make this?" but "how do you keep it the same?" How is bar stock verified when the lot changes? What is measured through the run, not just at its start? What triggers a new first article? How is drift caught before it becomes a failed lot? The answers define the supplier's control layer, and the control layer is what a program buyer is actually purchasing.
The economics follow the same logic. The per-part price matters, but the cost of drift matters more — a lot that drifts out of band late in a program can stop an assembly line, and the cost of that stop dwarfs the saving on any single release. A program supplier is therefore chosen on its controls, and the price is negotiated within the population of suppliers whose controls pass the file. The verification in this guide is how the file is built.
Which supplier checks matter most for a long-running turned-part program?
The desk checks run as with any supplier: the entity and the payment account reconcile across the documents, and the certificates — ISO 9001:2015 as the baseline, with IATF 16949:2016 and ISO 14001:2015 where relevant — resolve against their registrars with a scope that covers machining. Those checks confirm who is being paid and that a documented quality system exists. On a program, they are the entrance ticket, not the qualification.
The program-relevant checks come next. Does the shop hold the bar sizes and grades the program needs, in the quantities a rolling release requires? Does it run the machine class that fits the part's geometry — the Swiss-type machines for slender components, the turning centers for the stiffer ones? And does it keep process data across lots, so you can audit the second release against the first? A program supplier is a supplier whose process can be audited repeatedly, and the evidence should exist to support each audit.
The audit itself is different for a program. Instead of one deep look before the first order, the program deserves a review rhythm: a look at the material area and the bar store, a read of the capability data at a fixed cadence, and a check that the lots arriving now match the configuration approved at the start. The factory behind this site, shown on the factory page, is used to being audited this way, because a program buyer who audits once and disappears has stopped managing the relationship.
How do you verify bar-stock discipline before paying?
Bar stock is the raw material identity of swiss work — the part inherits the bar's grade, condition, and heat — so bar discipline is the material-control heart of the program. Ask how bar is received and verified: does the incoming check tie the mill certificate to the physical bar before it is cut, and is certified stock segregated by heat so a mixed lot cannot reach the machine? The answer should describe a procedure with a consequence, not a hope that the bar supplier sent what the invoice says.
On a repeat program, the bar lot changes between releases, and each new lot should be verified before it runs. You should ask how a bar lot change is handled: is the new lot's certificate reviewed, is the material checked against it, and does the lot record show which bar supplied which parts? The chain from bar lot to finished lot is what makes a field issue bounded — when a problem appears, the shop can name the bar and the release it came from.
The certificate requirement belongs in the standing agreement. The drawing names the grade and condition, and the order should require the mill certificate with the heat number for each bar lot the program consumes. At receiving, you check that the certificate on the shipment matches the bar the program specified. Bar discipline verified once and monitored across lots is what keeps a long-running program from discovering a material problem in service.
What does process evidence look like for high-speed turned production?
At high cycle rates, the risk on a swiss program is not a single bad part; it is a slow drift across a long run. Tool wear moves a diameter a micron at a time; a bar lot with a slightly different hardness shifts the process; a setup change re-establishes the datum slightly differently. The process evidence that matters is therefore the evidence that shows the shop measured through the lot, not just at its start and end — the in-process checks that caught a drift while it was still small.
Ask for reports from production lots that show the measured values on the controlled diameters and features, the method behind each reading, and the points in the run where the checks were taken. A report that shows a single value per characteristic per lot is a report that cannot demonstrate control across the run; a report that shows the checks distributed through the lot is evidence the process was monitored. The second kind is what a program buyer should require.
The evidence should also show the shop's reaction to drift when it occurred. A capability figure such as Ppk ≥ 1.67 initial on a controlled diameter is meaningful when the data behind it includes the lots that were caught and corrected; a shop whose history shows no corrections at all either runs an exceptional process or did not measure closely enough to see the drift. The quality practice behind the evidence is documented on the quality page, and it is the layer a program buyer audits repeatedly.
How is the first article anchored on a repeat-program part?
The first article is the anchor of the whole program. It is measured on every characteristic against the released drawing revision, with the method named and the values reported, and its approval defines the configuration the shop is authorized to reproduce. Every later lot is compared against that approved first article — its data is the reference, its method is the method, and its approval is the license to run. A program without a solid first-article anchor is a program without a definition of "the same."
you should review the first article independently before production releases. Measure the critical features with your own equipment or an independent reviewer, compare the results with the supplier's report, and confirm the material certificate reconciles to the lot. A first article that matches its paperwork has established the anchor the program needs; one that drifts has shown where the process will need attention before the releases begin.
The first article's data also sets the capability baseline. The spread across the first production lot, measured against the approved first article, tells the buyer whether the process holds the band with margin or sits at its edge. A process at the edge of the band is a process that will drift out of it as the tools wear, and the program terms should account for that risk with the inspection frequency and the new-first-article triggers that the anchor defines.
What controls keep a repeat program from drifting?
The controls that keep a program steady are the measurements that run through every lot. The controlled diameters and features are checked against the approved first article at a frequency the capability data justifies; the capability figures are reviewed at a fixed cadence; and the trigger rule names the events that require a new first article — a tool change, a bar lot change, a setup change, or a material condition change. Each trigger is a moment where the process could move, and each is a moment where the control re-establishes the baseline.
Ask the supplier to state its trigger rule in the standing agreement. Which changes to the process require a new first article, and which are managed with in-process checks? How is a bar lot change handled between releases? The answers decide whether the program's second year looks like its first month or drifts away from it. A supplier that can name its triggers is managing the risk of drift; one that cannot is hoping the process stays put.
The reporting rhythm completes the controls. The supplier reports the lot data at a fixed cadence — the capability figures, the exceptions, the trigger events — and you review it against the first-article baseline. A drift that the control catches is a managed event with a correction; a drift that the review catches at delivery is a failed lot. The difference between the two is the difference between a program that is controlled and one that is merely repeated.
How should payment terms match a rolling call-off?
A rolling program needs terms that match its rhythm. The standing agreement fixes the drawing revision, the material and certificate requirements, the inspection level and report format, and the incoterm, so each release is administrative rather than negotiated. The call-off quantities are released against the schedule, and the payment terms should keep both sides current without either side carrying the other far into the future.
Gate each release on the evidence of the previous one. The lot report that clears the prior release — the measured values against the first article, the certificate reconciled to the bar, the statement of conformity — is the document that authorizes the next quantity. The settlement rhythm follows the releases, with each invoice settling against the evidence that was approved. A supplier that performs under this rhythm is a supplier the program can scale; one that resists it is asking the buyer to pay ahead of the proof.
The terms should also cover the exception cases that a long program will eventually see: a lot that drifts, a revision to the drawing, a bar supply problem, an expedite. Each exception has a written path — the correction, the cost owner, and the restart point — so the program survives its own incidents without renegotiating from scratch. The purchase order that anticipates the exceptions is the purchase order that keeps a multi-year program manageable.
Which findings should end the relationship before volume begins?
The standard findings end a swiss program relationship exactly as they end any sourcing process: an entity that shifts between documents, a bank account under a different name, certificate numbers that do not resolve, no process evidence, and payment pressure before the pilot or the first article exists. Each is visible in the file, and each is sufficient to stop. The program amplifies their cost, because a bad supplier is paid across many releases rather than once.
Program-specific findings belong in the same column. Bar stock that cannot be traced to its certificate, or a bar lot change that was not verified, is a material-control failure that compounds across the releases it touches. A first article that does not match its paperwork has failed to anchor the configuration. Capability data that does not exist, or a drift between lots that the shop did not catch, shows a control layer that cannot hold a program steady. On repeat work, a weak control layer is not a risk for one lot; it is a risk for every lot the relationship will ship.
Program stop list: bar stock that cannot be traced to its certificate · unverified bar lot changes · a first article that does not reconcile with its paperwork · no capability data across lots · drift the shop did not catch · the standard findings — entity shifts, unresolved certificates, no process evidence, payment pressure. Any single item pauses the program; two or more end it.
A swiss turned parts program is bought on the controls that run through every lot, and the verification in this guide builds those controls before the first release. The entity and certificates confirm who is being paid; the bar-stock checks confirm the material identity of every part; the process evidence confirms the shop measures through its runs; the first article anchors the configuration; and the drift controls keep the program steady across months of releases. A supplier that clears the file is a supplier whose program can be trusted at a distance, and the factory behind this site is used to being qualified exactly this way, with the quality page and the factory page written so the checks can be run against them.
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). Set up for 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). Set up for 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). Set up for 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). Set up for shafts, bushings, flanges up to ±0.002 mm.
- Wire EDM (5+) - Sodick AQ535L (550/350/300 mm travel, ±20° taper, 800 kg workpiece). Set up for 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.
Planning a swiss turned parts program and want a supplier that will hold it steady?
Send the drawing, the bar specification, and the forecast through the contact page — the response includes the entity details, the bar-stock plan, the process evidence, and the program terms structured around the controls this guide describes.
Frequently Asked Questions
What is different about sourcing swiss turned parts for a program rather than a one-off?
In short, A program repeats the same part on a schedule, so the supplier's job is stability rather than novelty: the bar stock must be dependable, the process must hold the band across many lots, and the controls must catch drift before it ships. Verification therefore focuses on the repeatability layer — material discipline, process evidence, and the first-article anchor that every lot is measured against.
Which supplier checks matter most for a long-running turned-part program?
In short, The entity and certificates as with any supplier, and then the program-relevant checks: does the shop hold the bar sizes and grades the program needs, does it run the machine class that fits the geometry, and does it keep process data across lots? A program supplier is a supplier whose process you can audit repeatedly, not once.
How do I verify bar-stock discipline before paying?
In short, Ask how bar is received, verified, and stored: does the incoming check tie the certificate to the bar before cutting, and is certified stock segregated by heat? On a repeat program the bar lot changes between releases, and each new lot should be verified before it runs.
What does process evidence look like for high-speed turned production?
In short, Reports from production lots that show the measured values on the controlled diameters and features, the method behind each reading, and in-process checks that caught drift during the run. At high cycle rates the risk is not a single bad part but a slow drift across a long run, so the evidence should show the shop measures through the lot, not just at its start and end.
How is the first article anchored on a repeat-program part?
In short, The first article is measured on every characteristic against the released drawing and becomes the reference every later lot compares against. Its data, its method, and its approval define the configuration the shop is authorized to reproduce.
What controls keep a repeat program from drifting?
In short, The controls are the measurements that run through each lot: the critical diameters checked against the approved first article, the capability data reviewed at a fixed cadence, and the trigger rule for when a new first article is required — after a tool change, a bar lot change, or a setup change. A drift that is caught by the control is a managed event; one caught at delivery is a failed lot.
How should payment terms match a rolling call-off?
In short, Terms for a rolling program should gate each release on the evidence of the previous one: the approved lot report and the pre-shipment check clear the release for the next quantity, with a settlement rhythm that keeps both sides current without prepaying far ahead. The standing agreement fixes the drawing revision, the material, the inspection level, and the incoterm, so each release is administrative rather than negotiated.
Which findings should end the relationship before volume begins?
In short, The standard findings — entity shifts, unresolved certificates, no process evidence, payment pressure — end it as with any supplier. Program-specific findings do too: bar stock that cannot be traced to its certificate, a first article that does not match its paperwork, capability data that does not exist, or drift between lots that the shop did not catch.
Standards referenced on this page
- ASTM A582/A582M-12 — Standard Specification for Free-Machining Stainless Steel Bars
- ISO 9001 — Quality management systems — Requirements
- ASTM E8/E8M-21 — Standard Test Methods for Tension Testing of Metallic Materials
