A tooling engineer in Ohio once sent us a bracket drawing with a single line attached: "our last progressive die design stalled at station six." The die was fine. The strip layout was not — three bending stations had been stacked consecutively with no idle station between them, so peak tonnage spiked past the press rating and the ram stopped short of bottom dead centre on every hard part. One empty station, moved two positions, fixed it. That is the kind of failure this guide is written to prevent.
Progressive die design is the discipline of turning a flat coil of metal into a finished part in one continuous feed — and doing it for a million strokes without the tool destroying itself. Five numbers decide whether that happens: strip yield, station sequence, punch and die clearance, tonnage budget and tool life. Get them right and the tool runs for years. Get one wrong and you own a very expensive paperweight.
What a progressive die design actually has to solve
A progressive die is not one tool. It is a chain of stations, each performing one operation, all sharing one strip, one press stroke and one feed pitch. Designing one means satisfying four constraints at the same time:
- Geometry decomposition. Every feature on the finished part — holes, slots, bends, embosses, the final cut-off — has to be assigned to a station in a sequence that does not distort features already made.
- Strip stability. The carrier web tying parts together must survive the weakest forming stage without buckling, twisting or sagging between lifters.
- Force budget. The sum of every station's load, plus stripper, lifter, nitrogen and cam loads, has to fit inside the press tonnage and its energy envelope at the target stroke rate.
- Tool life. Clearance, tool steel, coating and lubrication have to survive the ordered quantity on a regrind schedule the maintenance team can actually execute.
These constraints fight each other. Tight nesting improves material yield but weakens the carrier. Idle stations cut peak tonnage but lengthen the die and raise its price. Harder tool steel extends life but costs more and takes longer to grind. A good progressive die design is simply the record of those trade-offs, written down before anyone cuts steel.
If you have not yet decided whether stamping is the right process at all, read metal stamping vs CNC machining before committing to tooling — for low volumes the honest answer is frequently no die at all.
Step 1 — Progressive die strip layout: orientation, nesting and pitch
In progressive die design, the strip layout is the roadmap of the whole tool. Three numbers define it.
Part orientation. Rotate the part on the strip to minimise strip width, but respect grain direction. Bending parallel to the rolling grain invites cracking; bending across it typically cuts crack risk by 20–40% at tight radii. Once the bend radius drops below 1× material thickness, grain direction stops being a preference and becomes a requirement.
Nesting and yield. Material utilisation is net part area divided by the strip area consumed per pitch:
utilisation (%) = net part area / (pitch × strip width) × 100
Worked example — a flat blank 42 mm × 26 mm with two Ø4 mm holes, 1.5 mm thick:
- Net part area = 42 × 26 − 2 × π × 2² = 1,092 − 25 = 1,067 mm²
Option A — feed the 42 mm side along the strip:
- Strip width = 26 + 2 × 5 mm trim = 36 mm
- Pitch = 42 + 2.5 mm bridge = 44.5 mm
- Strip area per part = 36 × 44.5 = 1,602 mm²
- Utilisation = 1,067 / 1,602 = 66.6%
Option B — rotate 90° and feed the 26 mm side:
- Strip width = 42 + 2 × 5 mm trim = 52 mm
- Pitch = 26 + 2.5 mm bridge = 28.5 mm
- Strip area per part = 52 × 28.5 = 1,482 mm²
- Utilisation = 1,067 / 1,482 = 72.0%
That is a 5.4-point gain from orientation alone, and it comes from the pitch, not the width — Option B uses a wider strip and still wins because it advances less material per stroke. On a 40-tonne annual coil buy, 5.4 points is a bit over two tonnes of scrap you stop paying for. This is why layout is settled before anything else in the job.
Pitch and bridge. Pitch is the distance the strip advances per stroke:
pitch = part length in feed direction + bridge bridge ≈ 1.25 to 1.5 × material thickness
Thicker stock needs a wider bridge, because the bridge is what carries the part through the die. Carrier width should be at least 2× material thickness, and large dies often need more purely to push the strip.
| Carrier type | Formed by | Best for | Trade-off |
|---|---|---|---|
| Solid / outside carrier | Edge trim | Flat parts, simple bends | Widest strip, lowest risk |
| Centre carrier | Strip near part centre | Forming around full perimeter | Restricts forming if too wide |
| Inboard carrier | Two ties inside part width | Minimum scrap | Weak at heavy draws |
| Lance carrier | Lance instead of trim | Minimum scrap | Creates shaving, needs vacuum |
| Stretch web carrier | Deliberate web loop | Deep draws, embossing | Needs vertical clearance |
| One-sided carrier | Single-side tie | Access on three sides | Lifting and alignment risk |
Feed height deserves its own line: the strip must lift to clear the previous station's form, and every extra millimetre of lift costs cycle time and adds bounce risk. Design for the minimum lift that clears the tallest form, and specify general tolerances on the drawing so the trim allowance is not left to interpretation.
For the process view around this tooling, see progressive die stamping: process, design rules and cost.
Step 2 — Station sequencing and pilots in progressive stamping die design
Three rules govern the sequence, and all three exist to protect accuracy.
Pierce before form. Holes, slots and notches are made early, while the strip is still flat and stiff. A hole pierced after a bend inherits that bend's angular error.
Pilot early, pilot always. A pilot hole pierced at station 1 or 2 engages hardened pilot pins at every downstream station, correcting cumulative feed error. Pilot diameter is normally sized around 1.5× material thickness, and pilots must engage before the punches descend — not after.
Distribute the load. Heavy forming is spread across stations, and idle stations are inserted between them to drop peak tonnage and leave physical room for tooling. An idle station is also the cheapest engineering change you will ever buy, because it gives you somewhere to add an operation later.
Here is a realistic seven-station plan for a 1.5 mm SUS304 bracket with two bends, four holes and a cut-off:
| Station | Operation | Feature | Notes |
|---|---|---|---|
| 1 | Pierce | 2 × pilot hole Ø2.4 | Sets strip alignment for stations 2–7 |
| 2 | Notch | Edge relief + partial outline | Slug sensor active |
| 3 | Pierce | 4 × mounting hole Ø4.0 | Strip still fully flat |
| 4 | Idle | — | Tonnage relief + tooling space |
| 5 | Form | Pre-bend 45° | Pre-form to control springback |
| 6 | Form | Final bend 90° | Restrike, form-height sensor |
| 7 | Cut-off | Part separation | Part-out sensor, scrap chopper |
Freeze the geometry before this plan is cut into steel, because station sequence is the most expensive thing in progressive die design to change after the fact. A DFM review at drawing stage costs a day and can remove a station outright.
Springback is the reason station 5 exists. Stainless springs back roughly 2–5° per 90° bend depending on thickness and temper; pre-forming and then restriking is far more repeatable than trying to over-bend in one hit. The bend allowance side of that calculation is covered in sheet metal fabrication.
Step 3 — Punch and die clearance: the number that decides burr and die life
Punch and die clearance — the gap between punch and die button, per side — is expressed as a percentage of material thickness. It controls the shear zone, burr height, edge quality and, critically, how fast the tool wears.
| Material (our standard grades) | Condition | Clearance per side (% of t) | Typical burr limit |
|---|---|---|---|
| SPCC / low-carbon steel | Annealed | 5–8% | 10% of t, max 0.10 mm |
| S45C / SAE 1045 carbon steel | Annealed | 6–9% | 10% of t, max 0.12 mm |
| SUS304 stainless | 1/2H | 7–10% | 8% of t, max 0.10 mm |
| SUS430 stainless | Annealed | 6–9% | 10% of t, max 0.10 mm |
| C2600 cartridge brass | Annealed | 4–6% | 6% of t, max 0.08 mm |
| C5191 phosphor bronze | H temper | 4–6% | 6% of t, max 0.08 mm |
| AL5052 / AL6061 aluminium | H32 / T6 | 6–10% | 10% of t, max 0.12 mm |
| 42CrMo alloy steel | Annealed | 8–12% | 12% of t, max 0.15 mm |
Two patterns are worth memorising. Tight clearance (3–5%) produces a larger burnished zone and a smaller burr — what precision connectors and electrical contacts need — at the cost of faster tool wear. Generous clearance (8–12%) extends die life on high-volume structural stampings where edge quality is secondary.
Carbon steel progressive stamping (SPCC, S45C/1045) sits in the middle of that window and is the most forgiving family to tool. Brass progressive stamping (C2600, plus the free-cutting C370 and C443 families) wants tighter clearance and produces a much cleaner edge, which is why brass is specified for contacts and terminals — see brass stamping parts for the alloy behaviour in detail.
Those contacts and terminals usually end up inside consumer electronics assemblies, where edge quality is a fit issue rather than a cosmetic one.
Put the burr limit on the drawing, not in an email. "Burr ≤ 0.08 mm, die side, no secondary deburr" is a specification a toolmaker can design to. "Burr minimal" is not. The same discipline applies to the finish that follows: CNC parts finishing explains what each coating does to a stamped edge.
Step 4 — Tonnage budget for progressive die stamping
The cutting force for any pierce or blank is:
F = L × t × τ
- L = cut perimeter (mm)
- t = material thickness (mm)
- τ = shear strength (MPa), approximately 0.8 × tensile strength
Worked example — station 3 of the bracket above, four Ø4.0 mm holes in 1.5 mm SUS304 (τ ≈ 480 MPa):
- L = 4 × π × 4.0 = 50.3 mm
- F = 50.3 × 1.5 × 480 = 36,216 N ≈ 3.7 t
Now sum the whole die. A realistic budget has more lines than cutting alone, and the extra lines are where most quotes go wrong:
| Load source | Basis | Worked value |
|---|---|---|
| Station 1 pilots (2 × Ø2.4, L = 15.1 mm) | L·t·τ | 1.1 t |
| Station 2 notch (L = 58 mm) | L·t·τ | 4.3 t |
| Station 3 holes (4 × Ø4.0, L = 50.3 mm) | L·t·τ | 3.7 t |
| Stations 5–6 bends (52 mm total bend line) | bend line × t × UTS (600 MPa) | 4.8 t |
| Station 7 cut-off (L = 60 mm) | L·t·τ | 4.4 t |
| Spring stripper force | ~8% of cutting force | 1.1 t |
| Lifter pins + nitrogen pads | Pad pressure × area | 1.8 t |
| Scrap chopper | Vendor data | 0.6 t |
| Subtotal | 21.8 t | |
| Safety factor 25–30% | × 1.3 | 28.3 t |
Specify a press above 28 t — a 35 t or 40 t class in practice — and then check two things most quotes skip.
Tonnage is not energy. A press rated 40 t can still stall at bottom dead centre if the flywheel cannot deliver the required energy at the target strokes per minute, and loads applied very close to BDC are the worst case. The Fabricator's worked explanation of tonnage in progressive stamping is the clearest reference on the distinction.
Off-centre loading kills dies. Press ratings assume load spread over roughly two-thirds of the bed. A short die in a wide bed deflects the slide unevenly, shifting clearance from one flank of the tool to the other and accelerating wear on one side only. Resequence the stations or re-centre the die on the bed.
Shear ground onto the punch face can cut peak cutting force by 30–50%, but it spreads the load over a longer stroke, which changes the energy calculation. It is a real tool, not a free lunch. Above roughly 400 strokes per minute these margins tighten further — high speed metal stamping covers what changes.
Step 5 — Die materials, heat treatment and coating
Tool steel selection is an economic decision wearing a metallurgical disguise. Grades below follow the ISO 4957 tool steel designation.
| Tool material | Working hardness | Expected life (strokes) | Regrind interval | Relative insert cost |
|---|---|---|---|---|
| A2 air-hardening | 58–60 HRC | 100k–300k | 60k–100k | 1.0 (baseline) |
| D2 / SKD11 | 58–62 HRC | 500k–1M | 100k–200k | 1.6–2.0 |
| DC53 (toughened D2) | 60–62 HRC | 700k–1.2M | 150k–250k | 2.2–2.8 |
| ASP23 / M2 HSS (PM) | 62–64 HRC | 1M–2M | 200k–350k | 3.5–4.5 |
| Tungsten carbide (WC-Co) | 70–75 HRC | 2M–5M+ | 300k–500k | 6–10 |
D2 and SKD11 are the workhorses and are correct for most precision progressive dies. Carbide earns its 6–10× price only where abrasive wear dominates — stainless, high-strength steel, coated or glass-filled stock, or programs long enough that regrind downtime costs more than the insert.
Two supporting decisions decide whether the steel performs as specified:
Which steel the cutting edge is made of, and how long it holds that price, is the subject of our stamping die material guide.
- Stress relieve after rough machining, before hardening. This is what stops a die block moving during finish grind. Skipping it is the most common cause of a die that measures correct on the bench and then drifts in the press.
- Surface treatment. Nitriding and PVD coatings (TiN, TiCN, AlTiN) reduce galling and adhesive wear, especially on aluminium and stainless. Expect meaningful life extension on the punch, less dramatic gains on the die button.
Die blocks are roughed, stress relieved and finish-ground on the same CNC milling cells we run for production parts, so a die block behaves like any other large hardened workpiece once the layout drawing is signed off.
Hardness verification follows the discipline we apply to heat treated CNC parts.
The alloy behaviour behind those hardness targets is catalogued in CNC machining materials.
Step 6 — Progressive die tooling cost: when the die pays back
This is the section competitors leave qualitative, so here is the model we actually use.
Die cost scales with station count and part envelope. As a relative index, with a simple 1–3 station blanking die at 1.0:
| Die class | Stations | Part envelope | Relative cost index |
|---|---|---|---|
| Simple blank / pierce | 1–3 | < 60 mm | 1.0 |
| Short progressive | 4–6 | < 100 mm | 2.4–3.6 |
| Standard progressive | 7–12 | 100–250 mm | 4.5–7.0 |
| Complex progressive with cams or draws | 12–20 | 250–400 mm | 8.0–14.0 |
| Transfer or tandem line | 20+ | > 400 mm | 18.0–35.0 |
Break-even quantity. The die pays back when the per-part saving covers the tooling:
break-even quantity = die cost / (unit cost without die − unit cost with die)
Worked example — the 1.5 mm SUS304 bracket. A 7-station die sits at index ≈ 4.8. Laser-cut plus press-brake gives a unit cost of 1.00 in relative terms; the progressive die gives 0.22, because material yield is higher, cycle time is around 1.5 seconds and one operator runs the press. Saving per part = 0.78. Break-even lands in the low tens of thousands of pieces — typically 8,000–20,000 depending on the die index and the baseline process cost.
Below roughly 5,000 parts a year, hard tooling rarely wins. Above 50,000 it almost always does.
On automotive programs the tooling is amortised against a multi-year release, which usually pushes the answer further towards hard tooling than the raw arithmetic suggests.
In between the answer is genuinely part-specific, and low volume CNC machining or laser-and-brake usually beats both hard tooling and pure CNC.
The same crossover arithmetic applies to die casting, where the tooling is cheaper but the per-part material and cycle economics are different.
The same arithmetic also applies to precision fasteners, where material is an even larger share because the parts are small and the tooling is comparatively simple.
Per-part cost decomposition at volume, for a stamped part:
| Cost element | Share of unit cost | Main lever |
|---|---|---|
| Raw material | 35–55% | Nesting and yield (Step 1) |
| Press time | 15–30% | SPM, station count, coil handling |
| Tooling amortisation | 5–20% | Die life, total program volume |
| Secondary operations | 8–25% | Burr specification, finishing choice |
| Inspection and packaging | 3–10% | Cpk, sampling plan |
Material dominates — the same conclusion our CNC machining cost guide reaches from the machining side — which is the whole argument for spending engineering hours on the strip layout rather than haggling over press rate.
Our general CNC machining guide covers the process options when the volume answer turns out to be "no die".
Step 7 — Tryout, first article and die life in production
A finished progressive die design does not go straight into production. The standard ramp:
- Bench and setup — die assembled, shut height set, strip threaded, sensors mapped.
- Tryout — slow strokes, one station at a time, measuring after each. Forming simulation typically cuts tryout iterations by 50–70% versus pure trial and error.
- First article inspection — full dimensional report on the first production-representative parts, dimensioned to ASME Y14.5 convention.
- Capability run — a short production run establishing Cpk on the critical characteristics. Automotive programs want Cpk ≥ 1.67 on CTQ features, evidenced through the IATF 16949 documentation package.
- Production with preventive maintenance — the schedule below.
| Maintenance task | Interval (strokes) | Why |
|---|---|---|
| Visual and burr check | 10k–20k | Burr growth is the earliest wear signal |
| Cleaning, lubrication check | 20k–50k | Slug buildup is the number one jam cause |
| Pilot and bushing inspection | 50k–100k | Wear here shows up as hole position drift |
| Punch regrind | 100k–500k by steel | 0.02–0.05 mm removed per grind |
| Full teardown and refit | 300k–1M | Restores shut height and alignment |
Regrind is cheap — a few hundredths of a millimetre off the cutting edge — and skipping it is expensive, because a dull punch raises tonnage, which raises burr, which eventually cracks the punch. Budget regrind cost per stroke rather than per incident.
Plan traceability at this stage, not later. If parts need serialisation, a marking station in the die or a downstream laser marking cell is far cheaper than retrofitting one after PPAP, and the mark has to stay readable after whatever coating follows.
Step 8 — What to send for an accurate progressive die quote
A die quote is only as good as the package behind it. Send these eight items and the quote comes back in about 24 hours instead of three weeks:
- 3D model (STEP) and a 2D drawing with GD&T — the drawing is where burr limits, grain direction and CTQ flags live.
- Material specification: grade, temper and thickness range, for example SUS304 1/2H at 1.5 mm per JIS G 4305.
- Annual volume and expected program life — these drive tool steel and station count, not just price.
- Batch size and release cadence, which set coil buying and press scheduling.
- Critical characteristics and the capability evidence required, meaning Cpk target and PPAP level.
- Surface finish and secondary operations: anodizing, plating, powder coat or marking.
- Burr limit and edge condition, stated numerically.
- Sample or production timeline, and whether you want a CNC prototype run before hard tooling is cut.
Missing items 3 and 7 cause more re-quotes than everything else combined.
FAQ
What is progressive die design, in one sentence?
Progressive die design is the engineering of a multi-station tool that converts coil stock into a finished part in one continuous feed, station by station, with each stroke advancing the strip by one fixed pitch.
How long does progressive die design and build take?
Expect 2–4 weeks for a simple 1–3 station tool and 8–16 weeks for a complex multi-station progressive die including tryout. Forming simulation and a clean RFQ package are the two biggest schedule levers.
What tolerances can a progressive die hold?
Standard production tooling holds ±0.05 mm comfortably, and precision-ground tooling with restrike stations reaches ±0.025 mm on well-supported features. Anything tighter usually needs a secondary machining operation such as five-axis CNC machining.
How much does progressive die tooling cost?
Relative to a simple 1–3 station blanking die at index 1.0, a standard 7–12 station progressive runs 4.5–7.0× and a complex cam or draw tool 8–14×. We quote against your actual part rather than publishing prices that will be stale next quarter.
Progressive, compound or transfer — which one do I need?
Progressive for most high-volume flat parts up to roughly 250 mm. Compound for flat parts needing concentricity between a hole and the blanked outline; compound die stamping covers that case. Transfer for large, deep-drawn or geometrically awkward parts.
Why does my die produce burrs after only 50,000 strokes?
Almost always clearance drift or a skipped regrind. Check pilot and bushing wear first, because positional drift changes the effective clearance long before the punch looks dull.
How is punch and die clearance chosen for stainless?
Stainless work-hardens, so it runs a wider clearance window than carbon steel — typically 7–10% per side for SUS304 — and galling, not wear, is the failure mode that ends tool life. Carbide inserts and PVD coating are usually worth it on stainless programs above a few hundred thousand strokes.
Tool steel grades, heat treatment and clearance by work material are set out in our stamping die material guide.
Can one progressive die make several part numbers?
Yes, with family tooling and interchangeable inserts, but the volumes must be compatible — you do not want to change the die weekly. It works best for mirror-image parts or a product family sharing one strip width.
Which materials are easiest on tooling?
Low-carbon steel and aluminium. Stainless steel work-hardens and demands carbide or coating.
Titanium is rarely worth stamping at all. Brass and phosphor bronze are gentle on tooling and give excellent edge quality.
Can I prototype before committing to hard tooling?
Yes, and you usually should. Soft tooling or a CNC prototype run validates form, fit and assembly before you spend on a die, and it often changes the strip layout for the better once real parts are in hand.
Conclusion
Progressive die design comes down to five numbers: strip yield, station sequence, punch and die clearance, tonnage budget and tool life. Everything else is documentation. Get the layout right and material cost drops several points; get clearance right and burr complaints disappear; get the tonnage budget right and the press never stalls; get the tooling economics right and the die pays for itself long before the program ends.
Ruijin Fenghui has run 23+ years of precision manufacturing with 200+ CNC machines under IATF 16949, ISO 9001 and ISO 13485, and we build and run progressive dies in the same facility that machines the tooling — which means the die, the tryout and the production part are engineered by one team, and a layout change does not turn into a three-week handoff.
Send the eight-item package above and you will have a tooling quote and a DFM reply within 24 hours.
Contact us to start.
If you are still comparing suppliers, how to choose a stamping partner lists the questions worth asking before you release an RFQ.