A SanJun customer once shipped a 60-piece pre-production run of stainless-steel telecom enclosures for a European rollout. The geometry had been validated by three rapid-prototype rounds, the powder-coating looked right, and the fit on the bench was perfect. Six months later their field-install team started sending back photos: hairline cracks around the gland cutouts, weld discoloration along the side seams, and — on three enclosures — water ingress after the first autumn rain. Nothing on the prototype had predicted any of it.
The post-mortem came down to one sentence: the prototype and the production part were made in different ways by different people on different tooling. That sentence is what this guide is about. Moving sheet metal from prototype to production is not a matter of pressing "go" on a higher-volume order. It is a deliberate transition through three volume stages, three iteration rounds, and a twelve-point DFM delta review that has to happen before any part hits a progressive die.
By the end of this guide you will have a repeatable playbook: lock the material and tolerance before you cut anything, run a structured 1-2-3 prototype framework, score every part against the 12-point DFM delta review, and know exactly when to use bridge production tooling before committing to mass production. SanJun runs this playbook every week as an OEM/ODM sheet metal partner — and we will show you the exact checklist we send to customers who want their prototype to actually become a production part.
Why Sheet Metal Volume Stages Look Identical but Aren't
Walk into a sheet metal shop and ask for "one of these" and then for "a thousand of these." The finished parts will look identical — same bend radii, same edge quality, same finish. But the cost driver, the lead time, and the risk at each end of that range are completely different. Treating them as the same is how programs go off the rails at the 200-piece mark.
Three volume stages engineers should plan for
Most hardware programs fall into three volume stages, and the production strategy at each stage is different.
| Stage | Volume | Cost driver | Best production method | Typical SanJun lead time |
|---|---|---|---|---|
| Prototype | 1–50 pcs | programming, machine setup, hand fixturing | fiber laser + CNC press brake + manual welding | 5–10 working days |
| Low-volume / bridge | 50–1,000 pcs | operator hours, fixturing amortization | laser + brake + dedicated weld jigs | 2–4 weeks |
| Mass production | 1,000+ pcs | tool amortization, takt time, scrap | progressive / transfer die + automated welding + inline inspection | 4–8 weeks after die build |
A part that fits comfortably in the prototype stage at 30 pieces can completely change its economics at 1,000 pieces — not because the material got more expensive, but because the tooling amortization, fixturing labor, and tolerance-yield penalty start to dominate. This is why an OEM fabricator will often quote prototype and production parts at very different prices per piece, even though the parts themselves look the same.
What actually changes between stages
Five things shift simultaneously when you move from prototype to production:
- Bend tooling: prototype runs use whatever punch-and-die set gets closest to the target radius; production demands standardized tooling shared across part families.
- Nesting: prototypes are cut from whatever sheet is on hand; production requires optimized nesting, often standardized on 1–2 alloys and 1–2 gauges per product family.
- Fixture strategy: prototypes are hand-clamped; production uses modular fixture pallets with repeatable datum locations and poka-yoke features.
- Tolerance policy: tight tolerances everywhere cost time but remain feasible at low volume; at production, only critical dimensions should be tightly called out — everything else should reference ISO 2768 medium or coarse.
- Secondary operations: every deburr, sub-assembly weld, or finishing step multiplies cycle time at scale. Production DFM eliminates or combines these steps wherever possible.
If you skip these five transitions, your "production part" is just a prototype that someone agreed to make 500 times. It will look right at first article inspection and start drifting on unit 200.
Step 1 — Lock the Material, Gauge and Tolerance Before You Cut
The single most expensive mistake we see on prototype-to-production transitions is switching alloys or gauges between prototype and production. The customer picked 5052-H32 for fast prototypes, then "upgraded" to 304 stainless for production "because it's outdoor," without telling the fabricator. The bend tooling, weld parameters, and finish chemistry all change. Every quote, every tooling setup, every weld schedule has to start over.
Stock-gauge rule, alloy-vs-formability tradeoffs
Lock the material specification (alloy + temper + gauge + finish) in the same drawing revision you use to source the prototype. If you are undecided between two alloys, prototype both — but decide before the second iteration round. The four families our customers actually use:
| Family | Common alloys | Strength | Formability | Corrosion | Typical application |
|---|---|---|---|---|---|
| Mild / carbon steel | CRS, HRPO, galvanneal | ★★★★★ | ★★★★★ | ★★ (needs coating) | indoor cabinets, brackets |
| Stainless | 304, 316L | ★★★★ | ★★★ | ★★★★★ | outdoor / washdown / food / coastal |
| Aluminum | 5052, 6061-T6 | ★★★ | ★★★★★ | ★★★★ | lightweight enclosures, HMI bezels |
| Copper / brass | C110, C260 | ★★★ | ★★★★ | ★★★★ | busbars, EMI/RFI shielding |
For more on material tradeoffs, see our sheet metal material selection guide. The key point: the material you prototype in must be the material you produce in.
Tolerance policy that scales — avoid ISO 2768 tight everywhere
The second most expensive mistake is calling out ±0.05 mm on every dimension on the prototype drawing and then asking the fabricator to "keep that tolerance" at production. At prototype volume, a fabricator can hit tight tolerances by slowing down and re-checking. At 1,000-piece volume, every ±0.05 mm callout adds machine time and rejection risk.
The right policy is functional tolerances:
- Use ISO 2768-m (medium) as the default for all unmarked dimensions on the production drawing.
- Call out tight tolerances only on mating features (PEM nut locations, gasket grooves, hinge pins, connector cutouts).
- Call out loose tolerances on cosmetic surfaces — most customers can't see ±0.3 mm on a 1.5 mm panel after powder coat.
Our sheet-metal tolerance guide walks through ISO 2768 medium vs fine vs the GD&T callouts we see on customer drawings, with examples from real production parts.
Step 2 — Prototype Iteration Rounds (the 1-2-3 Framework)
Compressing prototype validation into a single round is the second most common way programs stall at the production gate. You end up testing too many variables at once and can't isolate which change improved performance. We run sheet metal rapid prototyping programs on a structured three-round framework — not because customers ask for it, but because it is the minimum number of rounds that separates geometry validation from production-intent validation from first-article release.
| Round | Goal | What you test | What you decide |
|---|---|---|---|
| Round 1 | Validate basic geometry, fit, forming feasibility | 3D-printed bracket or laser-cut sample in any workable alloy | "Can this part be made in sheet metal at all, and does it fit?" |
| Round 2 | Validate production-intent material & finish | Production alloy, production finish, production gauge | "Will this material/finish combination survive the field?" |
| Round 3 (if needed) | DFM delta review + first-article inspection package | Production tooling assumptions, weld shrinkage, fixturing | "Is this part ready to release to production?" |
Most programs stop at Round 2. Programs that involve outdoor, washdown, or safety-critical applications almost always need Round 3.
Round 1 — geometry & fit
Round 1 answers the cheapest possible question: can this geometry be made? Use the fastest turnaround process available — laser cutting + manual bending + spot welding if joints are needed. Material choice is secondary; we often run Round 1 in 1.5 mm CRS for steel parts and 2 mm 5052 for aluminum parts purely because it is on the shelf. Don't invest in finish — leave it mill finish or bead-blasted. The goal of Round 1 is to confirm bend accessibility, hole alignment with mating parts, and overall dimensional feasibility. If Round 1 fails, you have lost a week. If you had caught the same failure in production, you would have lost a die build.
Round 2 — production-intent material & finish
Round 2 is where the material spec gets locked. Same part, but in the production alloy (304 vs 5052 vs CRS), production gauge (typically 1.5–3.0 mm for enclosures), and production finish (powder coat, anodize, or chemical film). This is also the round where you validate surface treatment adhesion — a bead-blasted prototype with no coating hides a host of adhesion problems that surface only after a real finish is applied. For a deeper dive on coating selection, see our surface finishing guide.
Round 3 — DFM delta review & first-article inspection
Round 3 exists for one reason: to catch the changes that happen between prototype and production. The Round 3 part should be built using the same tooling assumptions, nesting rules, fixture strategy, and tolerance policy that the production part will use. This is where the twelve-point DFM delta review (next section) gets applied, and where the first-article inspection report is generated. At SanJun we run Round 3 as part of the standard OEM hand-off for any production run over 500 pieces — see the DFM guide for the design rules Round 3 enforces.
Step 3 — The 12-Point DFM Delta Review Checklist
This is the part of the playbook that separates a fabricator who ships 50 sample parts from one who ships 5,000 production parts without field failures. The 12 points below are not design-for-manufacturability rules in the abstract — they are the specific changes a part has to absorb between Round 1/2 and a real production run.
| # | DFM delta item | Prototype default | Production requirement |
|---|---|---|---|
| 1 | Inside bend radius | whatever punch is on the bench | standard tooling set (1×, 1.5×, 2× material thickness) |
| 2 | Flange length | 3× thickness acceptable | ≥4× thickness + inside radius (see bending guide) |
| 3 | Hole-to-bend distance | hand-set, often tight | ≥2× thickness + bend radius |
| 4 | Hole diameter | as small as the CAD allows | ≥ material thickness (smaller = punch tool premium) |
| 5 | Material spec | whatever is in stock | locked alloy + temper + gauge + finish |
| 6 | Tolerance callouts | tight everywhere | functional tolerances (ISO 2768-m default) |
| 7 | Weld access | hand-fit | repeatable weld jig, consistent gap (see welding guide) |
| 8 | Weld shrinkage allowance | none | ≥0.2 mm per weld for carbon steel, ≥0.3 mm for stainless |
| 9 | Nesting strategy | leftover scrap | optimized nesting, ≥70% utilization |
| 10 | Deburr / edge break | as-needed | standard deburr spec for the entire family |
| 11 | Hardware (PEM / studs / rivet nuts) | hand-pressed where needed | press-in cycle time budgeted per assembly |
| 12 | Surface finish prep | bead blast only | full pre-treatment: blast + iron phosphate + powder coat (or anodize) |
Score every part before production release. If any item is still at "prototype default," it is either because the volume does not justify the change or because you have an open issue with engineering. Both need a written justification.
Bend tooling standardization
The most common delta review failure is bend radius. A prototype is bent on whatever punch gets close — maybe a 1.0 mm radius on a 1.5 mm part. The production part has to be bent on a standardized punch that is already loaded in the press brake. If the production part calls for a 2.4 mm radius that nobody in the shop owns, every piece requires a tool change, and tool change time at production volume is a multi-thousand-dollar line item. Standardize first, design second. See the bending guide for the punch-and-die library most fabricators actually stock. Before you freeze that radius, run our CNC bending force and allowance calculator — changing tooling moves both tonnage and bend allowance, which prototype numbers will not show.
Nesting & material yield
At prototype, a part is nested on whatever sheet is on hand. At production, the sheet is specified (1.2 mm CRS, 1.5 m × 3 m standard sheet, etc.), the parts are nested with common-line cutting where possible, and the scrap is minimized. Nesting alone can swing material cost by 15–25%. Our standard practice at SanJun: ≥70% material utilization per nest, common-line cuts on every shared edge, and one alloy per product family per nest. For a method comparison (laser cutting vs stamping for high volume), see our laser vs stamping guide.
Fixture & datum strategy
Hand-clamping a part for a single prototype weld is fine. Clamping 500 of them is not. Production fixtures need three repeatable datums (typically two pin holes + one edge), quick-mount clamping, and poka-yoke features so a misoriented part physically cannot be loaded wrong. Fixtures are amortized across the production run — the fixture cost is real but it is paid back over hundreds or thousands of pieces, not over ten.
Weld access & weld shrinkage
A prototype weld can be hand-positioned and ground flush afterward. A production weld runs on a fixture with a fixed gap, and the heat input is repeated hundreds of times. Two specific issues to engineer for at the production stage:
- Weld access: every weld must be reachable by the chosen welding process — MIG torch, TIG torch, or spot-welding electrode. If a seam is sealed inside a closed box section, MIG/TIG cannot reach it.
- Weld shrinkage: carbon steel shrinks ~0.2 mm per weld, stainless ~0.3 mm. The first article will be in spec; the 500th article may not if the welding sequence was not designed to balance shrinkage across the assembly. The welding guide has worked examples on distortion control.
Step 4 — Bridge Production: the Forgotten Middle
Most programs skip from prototype (1–50 pcs) straight to mass production tooling (1,000+ pcs). That is a mistake. The 50-to-1,000-piece zone is where a program's economics change the most, and where tooling decisions made for 5,000 pieces are over-investments. Bridge production tooling is the answer.
Three common bridge-production patterns we run at SanJun:
- Modular hard tooling: replace one or two soft tooling steps with a small dedicated punch or weld jig, while keeping the rest on laser + brake. Costs less than a full progressive die and reduces per-piece labor by 30–50% in the 100–1,000 zone.
- Dedicated weld fixture: the part still gets laser-cut and bent on standard tooling, but the welding moves to a dedicated jig with repeatable gap and shrinkage compensation. Weld time per piece drops from ~12 minutes to ~3 minutes.
- Semi-automated finishing: bead-blasting and powder-coating move from manual batch to a small conveyor loop. Per-piece finishing time drops from ~5 minutes to ~90 seconds.
The right time to introduce bridge tooling is around the 100-piece mark — not after the 1,000th piece, when the labor cost has already eaten the savings. Talk to your fabricator about bridge tooling options before you commit to a progressive die.
Bridge tooling answers a volume question inside the metal route. If the open question is which route the program belongs on at all, and where the two cost curves cross, that arithmetic is set out in where the sheet metal and injection molding cost curves cross.
Step 5 — Production Hand-Off Checklist
Before you release a part to production, sign off on the following hand-off items with your fabricator. We send this checklist to every OEM/ODM customer at SanJun as part of our custom sheet metal enclosure manufacturer process.
- Final drawing revision locked, with all DFM delta items resolved.
- Material spec confirmed against the actual incoming stock certificate.
- First-article inspection report signed, with at least 5 measured samples.
- Tolerance report showing all functional dimensions within spec.
- Finish adhesion test (cross-hatch + tape pull for powder coat; salt spray for plating/anodize where applicable). The programme that these checks belong to is described in our enclosure qualification test plan guide.
- Weld inspection report (visual + dye-pen or X-ray for critical welds).
- Production routing sheet — laser / brake / weld / finish sequence agreed.
- Fixture plan — datums, clamp points, weld jig location confirmed.
- Packaging spec — foam, bag, carton, pallet, label agreed.
- Cost guide reviewed — see our sheet metal enclosure cost guide for the per-piece model our customers use for budgeting.
A clean hand-off package turns the first production run from a risk event into a scheduled delivery.
How SanJun Runs Prototype-to-Production Projects
SanJun is an OEM/ODM sheet metal manufacturer in Shenzhen running the playbook above for B2B customers across six product lines: EV charging enclosures, energy storage cabinets, industrial HMI displays, telecom network cabinets, industrial control panels, and outdoor security / municipal cabinets.
What this means in practice:
- Free DFM review on every RFQ. You send a STEP file + a 2D drawing; we come back with bend-radius, hole-to-bend, weld-access, and tolerance feedback within 48 hours.
- Prototype runs from 1 piece. No minimum order quantity on prototypes, no setup fee for first article.
- First-article inspection included on every production run. Five-sample measurement report, finish adhesion test, weld visual.
- Bridge production tooling available for the 100–1,000 piece zone — see Step 4 above.
- Mass production on progressive dies with inline inspection for runs over 5,000 pieces.
- Single-vendor accountability — we own the laser, brake, weld, finish, and assembly under one roof, which is why our customers can move from Round 1 prototype to production release without re-explaining their project to a different supplier.
If you have a STEP file and a working concept, send it to our engineering team and ask for the free DFM review. We will come back with a prototype-to-production plan, a per-piece cost model, and a realistic lead time — usually within one business day.
When the part moves from prototype to production tooling, piercing limits move with it; hole-to-hole and hole-to-bend minimums are listed in the turret press punching guide.
Add a colour and gloss reading against a signed powder coating colour match limit sample. It is the only record that settles an argument two batches later.
Frequently Asked Questions
What changes between a sheet metal prototype and a production part?
The geometry is identical; the manufacturing system is not. Bend tooling, fixturing, tolerance policy, weld shrinkage compensation, nesting, and finish pre-treatment all change between prototype and production. Skipping those changes is what causes production parts to drift from first article to unit 500.
How many prototype rounds do I need before production?
For most programs, two rounds (geometry validation, then production-intent material/finish) are sufficient. For outdoor, washdown, or safety-critical applications, three rounds (geometry + production-intent + DFM delta + first article) are recommended. See the 1-2-3 framework in Step 2 above.
What is bridge production tooling?
Bridge tooling is the production tooling you introduce in the 50–1,000 piece zone — between prototype and full progressive-die mass production. It is modular (a dedicated weld jig, a small hard punch, a finishing conveyor) and pays back over hundreds of pieces rather than thousands.
Should I switch alloys between prototype and production?
No. Lock the material spec before Round 2 and do not change it. Switching alloys invalidates your Round 1–2 prototype validation, forces new bend tooling, new weld parameters, and new finish chemistry — every change a fabricator has to absorb at production release.
Conclusion
Moving sheet metal from prototype to production is not a single event — it is a structured three-stage transition: lock material and tolerance, run the 1-2-3 iteration framework, and apply the 12-point DFM delta review before production release. Programs that follow this playbook ship on time and survive their first field deployment; programs that skip it ship late and fail in service.
SanJun runs it every week as an OEM/ODM partner. Send us your STEP file and a 2D drawing — we will come back with the prototype-to-production plan, a per-piece cost model, and a lead time, usually within one business day.
Get a free DFM review → Send your files here and we'll respond within one business day.
Tooling belongs at the end of that sequence, not the start: our tooling investment guide covers what to build with soft tooling first.
The interfaces that cause the most rework between prototype and production — doors, openings, cable entry — are grouped in the structure and interfaces cluster.
The application designs, and the commercial decisions that sit around them, are collected in the applications and sourcing cluster.