Prototype tooling materials: what to use for a short run

A 1,500-piece bracket order does not need a D2 die. It needs a block that survives 1,500 strokes without a regrind, holds the hole position for the length of the run, and costs less than the parts it produces. On 1.6 mm CRCA that block is usually a 40 mm plate of EN31 hardened to 58–62 HRC. On a 300-piece trial it is a mild steel plate with no heat treatment at all.

Prototype tooling materials are chosen in a fixed order: how many strokes the tool must survive, what material it will cut, how thick that material is, and whether the tool has to survive a second programme. Get the order wrong and the money is wasted in both directions. A hardened D2 tool on a 300-piece run spends two weeks in grinding and heat treatment to make 300 parts. An unhardened block on a 20,000-piece run puts a burr on every part after the first 2,000 and nobody notices until assembly.

The arithmetic below is the one the tool room runs before it draws the die. It is also the arithmetic a buyer can run from the RFQ, which is what makes the tooling line on a quotation defensible.

How many strokes does the tool actually carry?

A tool is sized by total strokes, not by annual volume. Take the annual quantity, multiply by the operations per part, multiply by the years the part will run, and add the tryout. A part at 4,000 pieces a year for three years on a single-stage tool with one stroke per part carries 12,000 production strokes. Add 5% for setup scrap and tryout and the number is 12,600.

That figure, not the 4,000, decides the steel. Three further numbers move it:

• Operations per part. A blank-and-form tool is two operations. If the blank is cut on the laser and only the form is in the press, it is one. Cutting the blank on the laser removes the largest wear surface from the tool entirely.
• Stations in the tool. A two-station tool with a hand-loaded blank does two strokes per part, because the press cycles once per station.
• Programme changes. A tool built for a bracket that may be revised twice in a year needs grinding stock on the cutting edges, not a harder steel.

Where the part is likely to change, the cheaper decision is a laser-cut blank and a simple form tool. Low-volume stamping without overbuilding tooling works through the split between what stays on the die and what moves to the laser, and the same logic sets the material budget: a tool that will be modified should be easy to machine, not hard.

Which material suits which stroke band?

The table is the selection chart used at quotation. Lives are for 1.5–2 mm CRCA at 6–8% clearance per side, with the punch and die both in the listed material and normal lubrication. Thicker, harder or more abrasive stock moves the figure down, not up.

The band that matters most is the first one. Below about 3,000 strokes the heat treatment is a larger cost than the steel, and an unhardened block is faster to make and easier to alter. Above 3,000 strokes the edge starts to round, the burr climbs, and the part that passed first article no longer meets the drawing by the end of the run.

Hardness is a trade, not a target

Every extra point of hardness comes out of toughness, and a chipped cutting edge fails faster than a worn one. D2 at 62 HRC is the hardest sensible setting for a cold-work die; at 64 HRC it holds a keener edge on thin stock and chips when the strip is 4 mm or the press is out of parallel. For stock above 3 mm, or for a part with an interrupted cut, drop to 58 HRC or move to a shock-resisting grade such as S7 at 56–58 HRC.

The same rule applies to punches. A small punch in hardened D2 is strong in compression and brittle in bending, so the punch body is often made from a tougher steel and only the tip is hardened, or the punch is supported by a stripper plate within one punch diameter of the work. A 6 mm punch piercing 3 mm CRCA needs a guided stripper; a 6 mm punch with 10 mm of unsupported length will bend before it wears.

Hardness also decides how the tool is finished. Above about 45 HRC the die opening cannot be milled after heat treatment, so it is wire-cut or ground. That pushes the die design toward a bolted insert rather than a pocket cut into a large block: an insert is easier to wire-cut, easier to replace, and its heat treatment distortion is confined to a small piece.

Where the grinding allowance goes

A hardened die is not finished when it comes out of the quench. EN31 through-hardens with 0.1–0.3 mm of movement on a 200 mm block, and the die face and the cutting edges are ground back to size afterwards. The tool drawing therefore carries grinding stock on every surface that will be ground: 0.3–0.5 mm per face is the usual figure, and 0.5 mm where the block is long or the section changes sharply.

The allowance sets the number of times the tool can be sharpened. Take a die insert 40 mm tall with a 6 mm hardened band above the minimum height, ground 0.2 mm per sharpening. Six millimetres divided by 0.2 mm gives 30 sharpening cycles. At 20,000 strokes per cycle that is 600,000 strokes of total life from a block that would otherwise be scrapped after the first wear band.

That is why the tool steel and the insert height are the same decision. A die built with 2 mm of hardened band can be sharpened ten times, and a programme that expects 300,000 strokes will run out of die before it runs out of parts. The regrind record is the evidence: die wear and tool life shows how the logged strokes per regrind turn into the maintenance line on a piece price.

Two smaller materials belong in the same conversation. Guide pillars and bushes are case-hardened to 58–60 HRC on the wearing surface and left tough in the core, because a through-hardened pillar snaps rather than wears. Stripper plates are usually EN8 at 45–50 HRC with hardened wear plates at the strip contact, since the plate guides the strip rather than cutting it.

When a cast zinc alloy tool is the right material

A zinc alloy die is cast from a pattern rather than machined from steel, and it is the cheapest way to make a large forming tool for a low tonnage. It suits form-only work on panels where the blank is laser-cut, the draw is shallow, and the press is at the low end of its range. Life is 1,000–10,000 parts depending on the alloy, the lubrication and how much the blank moves over the die radius.

Two constraints decide whether it works. The pattern must be made oversize by the shrinkage, about 0.6–0.7% for the common zinc alloys, so a 200 mm cavity is cast from a 201.2–201.4 mm pattern. And the tool must be supported evenly, because zinc alloy is soft enough to deflect under an off-centre load and the part comes out with a taper that changes along the run.

Aluminium is not a substitute for either. A 6082 or 7075 block is excellent for a form block, a check fixture or a handling aid, and it is wrong for a cutting die: the edge collapses, the burr is heavy from the first stroke, and the block cannot be sharpened enough times to matter. Where the quantity is under 500 and the part is flat, the honest route is laser cutting plus a press brake, not a press tool at all.

What Ganesh runs for short-run press work

The tool room sits inside the press shop, which is the reason the material choice can be made against a real stroke count rather than a rule of thumb. Press tools, progressive dies, welding fixtures and checking fixtures are designed and built in-house, and the tooling is quoted separately from the piece price so the buyer can see what the life band is paying for.

The presses run from 20 T to 300 T, with a 250 T punching press for louvres, perforations and heavy pierce work. Blanks up to 12 kW fibre laser capacity are cut on a 40 ft × 16 ft bed, which is what allows a single-stage form tool to start from an accurate blank instead of a strip. Inserts in D2 at 58–62 HRC are the default for stainless and for CRCA programmes above roughly 50,000 strokes; EN31 is used where the run is shorter and the part is simple.

Custom parts run from 0.8 mm to 40 mm, with ±0.5 mm typical on formed features and tighter callouts agreed per feature. Where the customer needs PPAP or a first article, the dimensional report is built from the same tryout parts the tool was signed off on, not from a later production sample.

What to put in the RFQ

The tooling line on a quotation is only as good as the inputs. Give the tool room these six numbers and the material choice follows:

1. Total quantity over the programme, not the annual quantity, and the number of years.
2. Operations you expect in the tool, or state that the blank will be laser-cut.
3. Material and thickness, including the grade and the temper, because 304 at 2 mm wears a die differently from CRCA at 2 mm.
4. Burr and edge condition, as a height in millimetres or as a deburr operation.
5. Whether the tool will be transferred to another plant or kept for service parts, which changes the insert height and the spares list.
6. Revision risk, meaning how likely the part is to change within the programme.

A drawing that carries the material, the thickness and the tolerance but no volume will get a tool sized for a generic run, and the price will be wrong in one direction or the other. The stamping and presses hub collects the rest of the process decisions around this one.

When the hardened prototype tool is the wrong buy

A hardened tool is the wrong answer in four cases: the quantity is under about 500 and the part is flat, the part is a one-off fixture or a prototype for fit checks, the design is still moving, or the geometry needs a draw that a single-stage tool cannot hold. In the last case the choice is not a harder steel but a different tool: a draw die with a blank holder and a radiused entry, or a two-piece fabricated housing.

Deep drawing is the process that most often gets mis-sized. A cup deeper than half its diameter needs a blank holder, often two stages, and a die material chosen for polishing rather than for abrasion. That is a different quotation from a pierce-and-form tool, and treating it as the same job is how a prototype tool ends up with a torn cup and a rework loop.

For the parts that sit between a hard tool and no tool, the decision is usually a single-stage tool fed by laser blanks, which is covered in progressive die versus single-stage press tool. Send the drawing, the total volume and the years to our engineering team and we will price the tool against the stroke count it actually has to carry, with the insert height and the grinding allowance stated on the tool drawing.