INSPECTION · September 14, 2026 · 6 MIN READ

Measuring Tools for Forged Parts: A Gauge Selection Matrix

The right gauge is set by the tolerance, not by what is on the bench. A ±0.5 mm forged dimension needs a caliper or a hard gauge; ±0.05 mm needs a micrometer or a bore gauge; ±0.01 mm needs a CMM or a air gauge in a temperature-controlled room. Using a caliper to accept a ±0.05 mm dimension is the most common source of "it passed here and failed there" disputes.

WHERE THIS APPLIES AT FORGEPRESS CHINA

ForgePress China reports dimensions with the gauge and resolution used, so the acceptance decision can be reproduced by your incoming inspection. Capacity: HKM3 hot-die forging line 1,000–5,000 t, part weight 0.1–500 kg, maximum envelope 1,200 × 800 mm, as-forged to ISO 8062 class C with critical dimensions machined to ±0.2 mm, process control to ISO 9001, first article 4–8 weeks from die design release.

KEY NUMBERS

Rule of ten: the gauge should resolve at least one tenth of the tolerance being checked.
Caliper resolution 0.02 mm; micrometer 0.001 mm; bore gauge 0.001 mm; CMM 0.001 mm with 0.003 mm uncertainty.
Steel expands about 0.012 mm per metre per °C — a 500 mm part at 28 °C is 0.048 mm larger than at 20 °C.
Hard go/no-go gauges check one feature in 2–5 s; a CMM takes 30–90 s per part.

The rule of ten

Measurement uncertainty should be no more than one tenth of the tolerance band, and never more than one fifth. For a ±0.5 mm tolerance the band is 1.0 mm, so a gauge resolving 0.1 mm is adequate and a 0.02 mm caliper is comfortable. For ±0.05 mm the band is 0.1 mm, so you need 0.01 mm resolution — which rules out every caliper on the market.

Applying this rule at the planning stage prevents most inspection disputes. Applying it after a rejection usually means re-inspecting with the right tool and discovering the parts were fine, which costs a week and a customer relationship.

Selection matrix

Feature / tolerance	Gauge	Resolution	Throughput
Overall length, ±0.5 mm	Vernier or digital caliper	0.02 mm	10–20 s
Overall length, ±0.05 mm	Micrometer	0.001 mm	20–30 s
Forged thickness, ±0.3 mm	Caliper or snap gauge	0.02 mm	5–10 s
Machined thickness, ±0.02 mm	Micrometer or snap gauge	0.001 mm	15–25 s
Bore diameter, ±0.05 mm	Two-point bore gauge	0.001 mm	30–45 s
Bore diameter, high volume	Plug gauge (go/no-go)	Attribute	2–5 s
Hole position, ±0.05 mm	CMM or functional gauge	0.001 mm	30–90 s
Draft angle	Protractor or CMM	0.1°	10–60 s
Radius, as-forged	Radius gauge (visual)	Attribute	2 s
Surface roughness	Stylus roughness tester	0.01 µm	60 s

Throughput figures are for one operator on one feature, excluding part handling. Hard gauges are the only option above about 300 parts per shift.

Temperature: the error that nobody records

Steel expands about 0.012 mm per metre per °C. A 500 mm forging measured at 28 °C reads 0.048 mm larger than it would at the 20 °C reference, which is the entire tolerance band on a ±0.05 mm dimension. Measuring a hot part straight off the press is worse by an order of magnitude.

The fix is procedural: let parts stabilise for at least 30 minutes in the inspection area, record the temperature, and apply the correction for long dimensions. For anything under 100 mm the correction is below 0.01 mm and can be ignored; above 300 mm it cannot.

Choosing between attribute and variable gauges

A go/no-go plug gauge tells you whether the hole is within limits in 2–5 seconds and needs no skill to use. It cannot tell you where in the band the hole sits, so it cannot detect a trend toward the limit.

The practical combination is variable gauging on the first pieces and every hundredth piece, with attribute gauges on everything in between. That gives process control and 100 % screening without the cost of measuring every part.

Frequently asked questions

Can I use a caliper for a ±0.05 mm tolerance?

No. A caliper resolves 0.02 mm and its measurement uncertainty is typically 0.03–0.05 mm, so it consumes the whole tolerance band. Use a micrometer, which resolves 0.001 mm and has uncertainty around 0.003 mm.

When do I need a CMM rather than hand tools?

When the callout is geometric (position, profile, concentricity), when the tolerance is below ±0.05 mm, or when you need a permanent record with a point cloud. Hand tools are faster and perfectly adequate for dimensions above ±0.1 mm.

How much does temperature actually matter?

About 0.012 mm per metre per °C for steel. Under 100 mm the effect is below 0.01 mm and can be ignored. Above 300 mm, a 5 °C difference exceeds 0.018 mm, which is significant for any tolerance below ±0.05 mm.