Low-volume stamping without overbuilding tooling

A bracket at 2,500 pieces a year arrives quoted with a four-station progressive die, and the tooling line is larger than three years of piece price. Nothing in that quote is wrong on its own. The die will make a good part at 60 strokes a minute. It is simply sized for a volume the programme does not have, and the buyer pays for the difference up front.

What makes low-volume stamping different from serial stamping?

Low-volume stamping is a tool-structure decision: fewer stations, a blank that stays on the laser, and money spent on the two or three features that actually need a die. At 2,500 pieces a year the press is rarely the constraint; the tool, the setup and the inspection are.

Which tool structure suits 500–5,000 pieces a year?

One forming tool fed by laser-cut blanks suits 500–5,000 pieces a year, with a fabricated part below that and a progressive die only above about 20,000 pieces with a frozen design. Each band is planned differently:

• Under 500 pieces a year. Laser-cut blank, press brake, fixture-welded details. No press tool. The variation is wider and the piece price is labour-heavy, but there is nothing to amortise.
• 500 to 5,000 a year. One forming tool or a compound blank-and-form tool, fed by laser-cut blanks. The blank stays free of tool cost, so a revision to the outline or a hole pattern costs a DXF edit.
• 5,000 to 20,000 a year. Two or three stations in one tool, or a compound tool with a separate pierce tool. Enough volume to justify dedicated blanking.
• Above 20,000 a year, design frozen. Progressive die. The strip does the handling, the part is made in one pass, and the labour per piece stops being the main cost.

Those bands are planning starting points, not rules. The real answer comes from the tool cost divided by the saving per piece, and the tooling cost versus piece price guide and the laser cutting versus stamping comparison run that arithmetic for the volume bands above. What this page decides is the tool you buy, not the number at which it pays back.

Maintenance belongs in the same comparison. Every regrind takes die height out of the tool and press time out of the schedule, and a die that runs 5,000 strokes a year will spend more of its life in storage than on the press. That storage has a cost too: rust prevention, handling damage and the re-tryout when production restarts after a long gap.

Which features should come off the die?

Tool cost is driven by feature count and feature difficulty far more than by part size. Moving one difficult feature into a second operation can remove a station and a cam from the tool.

The first line in that table is the one that changes most quotes. A hole 3 mm from a bend in 1.5 mm steel sits far inside R + V/2 + t, which on a 12 mm V-die with a 1.5 mm inside radius is 9 mm and forces the pierce station ahead of the form. The 1.5 × t + r shortcut gives about 3.8 mm for the same part and is only safe when the die opening is about 8t. Move the hole to 10 mm out and the same part can be formed first and pierced after, in a tool half the size; the hole-to-bend distance guide owns the rule.

Does a low-volume route cost more in material?

Usually yes, and the amount is worth calculating before the tool is discussed.

Take a bracket with a 220 mm × 150 mm blank in 1.5 mm CRCA. The blank volume is 49,500 mm³, or 49.5 cm³, and at 7.85 g/cm³ that is 389 g of steel per part before scrap.

Nested on a 1,250 mm × 2,500 mm sheet with a 3 mm web, the sheet takes 8 blanks across and 11 along, so 88 blanks from 3,125,000 mm² of sheet. The blanks use 2,904,000 mm², which is 92.9% utilisation, and the material per part is 389 ÷ 0.929, or 419 g.

A progressive strip for the same part needs a carrier. At 8 mm each side the strip is 166 mm wide and the pitch 224 mm, so each part consumes 37,184 mm² and utilisation falls to 88.7%. Material per part rises to 439 g, about 4.8% more steel than the nested blank, and the strip cannot be re-nested around a design change.

At low volume that 20 g matters less than the tool cost. At 200,000 parts a year it is 4 tonnes of steel, which is why the progressive die is the right answer there and the wrong one at 2,500.

How much of a low-volume piece price is setup?

At low volume the press is not the constraint. Setup, handling and inspection are, and they are priced per lot rather than per part.

Take a four-hour setup and put it over different lot sizes. At 250 pieces the setup alone carries 0.96 minutes of press time per part. At 1,000 pieces it is 0.24 minutes. At 2,500 pieces it is 0.096 minutes, and the tooling and material lines become the visible part of the price. Same press, same tool, a tenfold difference in the labour content of the quote.

This is why release size belongs in the RFQ next to annual volume. A buyer who orders 2,500 pieces a year in ten releases of 250 pays for ten setups, and no amount of negotiation on the piece price recovers that. One release of 1,000 and two of 750 usually costs less in total than ten small ones, at the price of carrying stock and the risk of a revision landing on a full pallet.

The same arithmetic applies to coating. A powder batch has a minimum practical load, and a small lot either waits for a colour change or runs with the booth partly empty. Neither is wrong; both are cheaper to plan around than to discover on the invoice.

How long does a low-volume tool actually last?

Tool life is rarely the limit on a low-volume programme. A die that runs 5,000 strokes a year for ten years will see 50,000 strokes in total, which is a fraction of what a progressive die does in a single good year.

The wear that matters at this volume is storage wear. A tool that sits for eight months gathers moisture, its springs take a set, its guide pins pick up rust, and the first run after the gap produces the sort of variation that gets blamed on the process. Grease the working faces, keep the tool off the floor on a pallet, record the die height, and re-try the tool before the production order rather than during it.

For a tool that runs once a year, put a sample part and the setup card in the tool box. The next run then starts from a known die height and a known first article instead of from a guess.

How do you split a tool so changes stay cheap?

Blank on the laser, pierce in a second smaller tool, and stage the stations: those three construction choices keep a low-volume tool adjustable.

Blank on the laser, form in the die. The outline, holes and slots come from a laser-cut blank, and the tool only forms. Changing a hole means editing a DXF, not stripping a die. For a part whose hole pattern is still moving, this is the cheapest structure available.

Pierce in a second, smaller tool. Cut the pierce stations into their own tool built on a standard die set. It costs less than the forming tool and can be re-cut or replaced when the hole pattern changes, while the forming tool stays untouched.

Stage the stations. Build the first two stations now and leave the pad and the mounting for a second phase once the design is frozen. A tool with an idle station and a reserved pad can grow without a rebuild, as long as the strip layout is planned for it at the start.

The one thing not to compromise is the datum scheme. A tool that locates on a different face from the customer's checking fixture will pass its own first article and fail at goods-in, whatever the station count.

What Ganesh runs for low-volume press work

Low-volume programmes here are built as one or two stations on a standard die set, with the blank supplied by the laser, so the tool never carries a pierce station the design has not frozen. The tool room builds the die, the checking fixture and the spare punch set from the same drawing revision, and the same team does the tryout on the press that will run the order. Tooling stays a separate line from the piece price.

Blanks come off a 12 kW fibre laser with a 40 ft × 16 ft bed, with 6 kW and 3 kW machines for thinner work, in mild steel, CRCA, stainless, aluminium, brass and copper from 0.8 mm to 40 mm. Tolerances run ±0.5 mm typical with tighter features by agreement, and first article inspection or PPAP documentation is available.

The sheet metal stamping capability page covers the press range and the route, and the tool room page explains how dies, fixtures and checking gauges are built and maintained. The stamping and press tool hub collects the related design material.

What to put in a low-volume RFQ

• Annual volume and the release size you actually place, not the forecast.
• Expected programme life in years, so the tool can be amortised against something real.
• The revision, and a note on which features are likely to change.
• Which features must be formed and which could be a second operation.
• Whether a laser-cut blank and a forming tool is acceptable, or whether the customer requires a full blanking die.
• Sample quantity and the acceptance method for the first article.
• Tool ownership, and what happens to the tool if volume moves to another part or another plant.

The RFQ checklist turns those into a one-page brief, and the tooling cost versus piece price guide sets out how the amortisation moves with tool complexity. For a first pass at press size on the forming operation, use the press tonnage calculator before sending the drawing for a tool-structure review.

When a press tool is the wrong answer

Below roughly 500 pieces a year, the tool is usually the wrong purchase, and the honest answer is a fabricated part with more welding and a higher piece price. A prototype should never carry a die unless the design has already been validated by other means.

A press tool also stops where the geometry stops. Panels longer than a brake bed are better folded on the 200 T, 4,450 mm press brake than formed in a die. Plate above 10 mm is a brake and laser job. A part with thirty holes at different angles is a laser job with a drill follow-up, because a die for that pattern would be a maintenance liability at any volume this article is about.

And a tool built before the design is frozen is the most expensive version of a cheap tool. The progressive die versus single-stage tool comparison sets out the crossover between tool types once the volume is known.