Mold Acceptance: The Three Data Points in an SPC First Article Report That Buyers Most Often Overlook

When a new LSR mold is ready for acceptance, most buyers focus on one question: did the first article samples pass dimensional inspection? If the report shows "Pass" next to every dimension, it's easy to sign off and move to production. But a passing first article report and a mold that will run reliably for the life of an order are not the same thing — and the difference usually shows up in three data points that get far less attention than the pass/fail column.

1. Cpk (Process Capability Index) — Not Just Whether It's Within Tolerance

A dimension can measure well within tolerance on a single first-article sample and still come from a process that isn't stable enough to stay there run after run. This is exactly what Cpk is designed to catch, and it's the single most important number in an SPC report that buyers commonly skip past.

• A dimension sitting comfortably inside spec on one sample tells you almost nothing about repeatability.
• Cpk, calculated from a run of consecutive samples rather than one part, tells you how much margin the process actually has — and whether that margin is centered or drifting toward one edge of the tolerance band.
• A low Cpk on a critical dimension — even if every individual sample technically passed — is often an early warning that the mold, process, or material combination will produce out-of-spec parts once running at full production volume and speed, rather than under the careful, slowed-down conditions of a first article run.

What to ask for: request the Cpk value for every critical-to-function dimension, not just the raw measurement of a single sample, and ask what sample size and sampling method were used to calculate it.

2. Sample Size and Sampling Conditions — Not Just the Result

A first article report is only as meaningful as the conditions under which it was collected, and this context is frequently left out of the summary page buyers actually read.

• How many shots were sampled, and from where in the run — the first few shots off a freshly qualified mold behave differently than shots taken after a longer, more representative run.
• Which cavities were measured, in a multi-cavity mold. A report that only samples one or two cavities out of eight tells you nothing about cavity-to-cavity variation, which is often where real production problems appear.
• Process conditions during sampling — whether the machine was running at the actual intended production speed and cycle time, or at a slower, more controlled rate specifically to produce a clean first-article report.

What to ask for: a clear statement of sample size per cavity, and confirmation that samples were taken at production-representative cycle conditions, not an artificially slowed-down qualification run.

3. Trend Data Across the Sampling Run — Not Just the Average

An SPC report that shows only average, minimum, and maximum values for a dimension can hide a trend that matters more than any single statistic: is the dimension drifting in one direction over the course of the run?

• A dimension that starts within tolerance and drifts toward the edge of spec over a run of 30 or 50 shots is showing a trend that averages and single-point pass/fail checks won't reveal — but that trend is often the first sign of mold wear, thermal instability, or a material issue that will worsen at production volume.
• Reviewing the actual run chart or control chart, rather than only the summary statistics, is the only way to catch this before it becomes a field issue months into an order.

What to ask for: the raw run chart or control chart for critical dimensions, not just the summary average — and ask the mold maker to flag any dimension showing a directional trend, even if it's still technically within tolerance.

Why This Matters More for LSR Than for Rigid Plastic Parts

Liquid silicone parts tend to have tighter functional tolerances relative to their nominal size than many rigid plastic parts — particularly for sealing surfaces, medical components, and precision consumer electronics parts — which means a process with marginal Cpk or an undetected trend is more likely to produce a real functional failure, not just a cosmetic one.

A Simple Acceptance Checklist

Before signing off on a new mold based on its first article report, it's worth confirming:

• Cpk values are provided for every critical dimension, not just pass/fail results
• Sample size and cavity coverage are documented and adequate for the mold's cavitation
• Samples were taken at production-representative speed and cycle time
• Run/control charts are available on request, not just summary statistics
• Any dimension showing a directional trend has been flagged and explained, even if within tolerance

A first article report that includes this level of detail is usually a sign of a mold supplier who understands that "passing" and "production-ready" are two different standards — and that the gap between them is exactly what a thorough SPC report is meant to close.