Tooling tryout: what a press tool should prove before production

A press tool that has run fifty good parts on the toolmaker's bench press has proved that the tool exists. It has not proved that it will make the part at rate, on the press it was bought for, with the coil the customer will actually ship.

Tooling tryout is the stage where four things are measured rather than assumed: the peak tonnage the tool pulls, the way the strip behaves over a few hundred strokes, the dimensions of the first-off parts against the drawing, and the capability of the process over a consecutive run. Everything else in a tryout report is commentary.

The failure this prevents is specific. A tool released at 90% of press capacity runs for a month, then a coil arrives 0.05 mm thicker, the press deflects, the clearance opens, the burr grows on one side and a batch of 4,000 brackets reaches the paint line with an edge that fails the coating check.

What should a tooling tryout prove before release?

Four things, in this order.

That the tool fits the press. Peak tonnage, shut height, bed size, cushion or die-spring capacity and the stroke rate the tool was designed for. A tool that fits at 60% of rating has room for material variation; one at 90% does not.

That the strip runs. In a progressive tool this is the longest part of the tryout. Feed pitch, pilot entry, strip lift, scrap clearing, slug retention and sensor positions all have to work for hundreds of strokes, not five.

That the part is to drawing. Every characteristic on the drawing, measured on parts from the first, the tenth and the thirtieth stroke, with the measurement method stated.

That the tool repeats. Thirty consecutive parts from one coil, plotted, with a capability index for the characteristics that carry the function.

Only the third item is what most buyers see. The first, second and fourth are what decide whether the tool is still making good parts in month nine.

How much tonnage does the tool actually need?

Cutting force is the cut length times the thickness times the shear strength of the material. For CRCA, shear strength is roughly 0.7–0.8 of tensile strength, so a D-grade sheet at 400 MPa tensile shears at about 320 MPa.

A bracket with 1,200 mm of cut perimeter in 2 mm CRCA:

• F = 1,200 mm × 2 mm × 320 MPa = 768,000 N = 768 kN
• 768 kN ÷ 9.807 = 78.3 tonne-force for the cutting stroke
• Add 20% for the peak at penetration and for stripping: about 94 tf

On a 160 T press that is 59% of rating, which is a comfortable tool. The same part in 3 mm at 1,600 mm perimeter gives 1,536 kN, or 157 tf before margin — it does not fit a 160 T press at all, and the quotation that assumed it would is the one that fails at tryout.

Measure the number rather than trusting the calculation. A load cell or the press's tonnage monitor gives a peak per stroke, and the two figures should agree within about 10%. A measured peak well above the calculation usually means the die clearance is too small, the stripper is over-tensioned, or the tool is cutting more material than the layout suggests.

What belongs in the tryout record?

The last row is the one that gets skipped and the one that costs the most. A 0.05 mm shim added under a punch at tryout is a fact about the tool. Six months later, when the die is stripped for regrinding, nobody knows it was there.

Why does the strip test matter more than the first-off?

Because a press tool fails in the strip before it fails in the part.

A single good first-off proves the punch and die were made to the right shape. It says nothing about whether the pilot enters cleanly at stroke 300, whether the scrap chute carries the slugs away, whether the strip lifts off the pilots, or whether the material work-hardens enough at the first station to crack at the fourth.

A strip test runs the coil through at production speed for 200–500 strokes, with the operator watching rather than adjusting. Three things are checked at the end: the burr height on parts from the first, middle and last fifty strokes; the strip itself, for pilot elongation and edge cracking; and the die, for slug build-up and marking.

Burr height is the best single indicator. It should stay within 0.05 mm across the run. A tool whose burr doubles between stroke 50 and stroke 500 has a clearance problem or a punch edge that is already breaking down, and neither will improve in production.

What does capability have to look like before release?

Thirty consecutive parts from one coil, measured on the characteristics that matter, give a standard deviation. That deviation decides whether the tool can hold the drawing, not the average.

For a ±0.5 mm tolerance the total width is 1.0 mm, so:

• Cpk 1.33 needs σ ≤ 0.5 ÷ (3 × 1.33) = 0.125 mm
• Cpk 1.67 needs σ ≤ 0.5 ÷ (3 × 1.67) = 0.100 mm

A tool with a mean exactly on nominal and σ of 0.12 mm passes the first and fails the second. A tool with σ of 0.08 mm but a mean 0.3 mm off nominal fails both despite being more consistent, which is why the mean and the spread are reported together.

Thirty parts is a small sample. It is enough to catch a tool that cannot repeat and not enough to prove a process, so treat the tryout Cpk as a gate rather than a guarantee, and confirm it again over the first 500 production parts.

Press time is the other constraint. A tryout on a 160 T press typically takes half a day to a day for a single-stage tool and one to three days for a progressive tool, most of it spent on the strip rather than on the parts. That press time is booked against production, which is why a clear tryout agenda — every check written down before the tool is cut — saves money twice over.

Which tool faults show up as part faults?

Four tryout faults have a physical cause that the part reveals before the tool does.

Slug pulling shows as a pierced hole with a torn or partly missing edge, and the slug is usually found on top of the strip or inside the next part. It happens when the slug is held by friction in the punch rather than pushed out, so the fix is a slug ejector, a taper on the punch bore, or less oil on the punch face.

Galling shows as a scored or pick-up-marked surface on the drawn wall. The cause is adhesive wear between the sheet and the die, driven by insufficient clearance, low-viscosity lubricant or a die radius that is too tight for the material.

Pilot marks show as a witness ring around a pierced hole, slightly off position. The pilot is entering before the strip has settled, which distorts the hole and pulls the pitch. Adjusting pilot timing by a few tenths of a millimetre is a common tryout correction and must be logged.

Ejection faults show as a part that is bent or marked on the way out. At 40 strokes a minute, an ejector that is 5 ms late will hit the part on the next stroke, and the fault will not appear when the press is inched.

Each of these is a tool condition, not a process capability, and none of them is visible in a first-off that was hand-fed.

What changes during tryout, and why must each change be recorded?

Almost every tool is adjusted at tryout. The changes fall into five groups.

Clearance changes come from measuring the burr and re-cutting or shimming the die insert. Each change alters the burr height and the edge condition, so it has to be re-measured after.

Pressure and stripping changes come from parts lifting with the punch. A heavier pressure pad or a longer stripper travel fixes the lift and increases the tonnage, which is why the tonnage is re-measured after the fix.

Form changes come from springback. A 90° bend in 2 mm CRCA typically needs 1–3° of overbend; in 2 mm 304 stainless, 3–6°. The correction belongs in the tool, not in the operator's technique.

Timing changes come from cams, sensors and ejectors. A cam that returns 0.2 s late will be hit by the next strip at speed, and it will not show at 20 strokes per minute.

Material changes are not tool changes but they behave like them. If the tryout coil is 2.00 mm and the production coil runs 1.95–2.05 mm, the clearance that was right at nominal is 5% larger at the top of the range.

The adjustment record is what turns a tryout into a specification. Without it the next setup engineer starts from zero, and the setup time that was quoted on the strength of a proven tool quietly doubles.

What Ganesh runs in the tool room and the press shop

Tools are designed and built in-house — press tools, progressive dies and checking fixtures — and tried out on the presses that will run them: mechanical presses from 20 T to 300 T, with 250 T punching. The tool room sits alongside the press shop, so a clearance change or a form correction is made and re-measured the same day rather than shipped between sites.

Tooling is quoted separately from the piece price, which keeps the tryout, the tool card and the maintenance plan visible instead of buried in a unit rate. First article inspection and PPAP are available, and the plants hold ISO 9001:2015 and ISO 14001:2015. Parts are generally held to ±0.5 mm with tighter features by agreement, and the stamping capability page sets out the press range and the tool room scope.

The progressive die maintenance article explains what happens to the same tool after 100,000 strokes, the inspection guide for repeat production sets out what should be measured once the tool is in the press every day, and the stamping hub collects the press and tooling guides.

What to put on the tryout agenda and the purchase order

1. The press the tool will run on, by capacity and ideally by asset number, and whether the tryout must happen on that press.
2. Material grade and the thickness range expected in production, not just the nominal.
3. Every characteristic to be measured, the method, and which are critical for capability.
4. The capability target per characteristic, with the sample size and the coil to be used.
5. Burr height limit and which face the burr is allowed on.
6. The adjustment record and the tool card as deliverables, with the tool drawing revision.
7. Whether the customer witnesses the tryout and signs the first article.
8. Spare punches, inserts and springs to be handed over with the tool.

Most of this is one page. A purchase order that says "tool to be proved on samples" hands the definition of proved to whoever is holding the micrometer.

When tryout is the wrong instrument

Tryout proves a tool against one material lot on one press. It cannot prove the drawing. If the part has a tolerance the process cannot hold — a ±0.05 mm hole position on a 3 mm progressive tool, for example — the tryout will show it late and expensively, after the tool is cut.

It also cannot prove demand. A tool built for 50,000 parts a year that runs at 40 strokes per minute on a 20 T press is not the same investment as a progressive die for 500,000. The progressive versus single-stage comparison covers that decision, and it belongs before the tool is designed, not after it is tried out.

Send the drawing with the annual volume and the target press through the quotation route. The tryout plan comes back with the tooling cost, so the acceptance criteria are agreed before metal is cut.