Direct Answer
A molded pulp mold trial report is an evidence document, and it must contain eight sections: trial purpose and part identity, tool and press configuration, process settings, measured dimensional and mass results, visual and functional results, defects with location and frequency, a conclusion judged against pre-agreed acceptance criteria, and a disposition such as approved, conditionally approved or returned for rework. A report without process settings cannot be reproduced, and one without acceptance criteria cannot be judged at all. Sampling must cover the tool's variation, across cavities, across the run and across the hardest-to-form positions, rather than the easiest handful. Version control matters because a conclusion reached on a different tool revision than the one on the drawing is not valid evidence.
Opening Hook
The conclusion first: a mold trial report that cannot be reproduced is not evidence, it is a memory. Picture a tool trial that produces excellent parts, a signed approval, and mass production that then yields parts outside tolerance for weeks. The trial report exists, but it records only dimensions and a "pass," not the press, the drying profile or the pulp consistency used on the day. The pain is that nobody can repeat the trial conditions, so the production team cannot tell whether the tool changed or the process did, and the argument runs until someone re-runs the trial from scratch. The fix is a report template that captures settings, sampling and acceptance criteria before the trial begins. At yisenpulp, every mold trial is reported against a fixed template so the result can be reproduced, challenged and audited rather than merely remembered.
The Eight Sections of a Trial Report
A trial report is a form, and the form decides what the trial can prove.
| Section | Content | Consequence if missing |
|---|---|---|
| Purpose and part identity | What is trialed, drawing revision | Results cannot be attributed |
| Tool and press configuration | Tool revision, press used | Trial cannot be reproduced |
| Process settings | Pulp, vacuum, drying profile | Causes cannot be separated |
| Dimensional and mass data | Measured values per sample | Pass judgement unsupported |
| Visual and functional results | Surface, fit, protection | Customer fit unknown |
| Defect list | Type, location, frequency | Improvement targets unclear |
| Conclusion vs criteria | Judged against pre-agreed limits | Opinion replaces evidence |
| Disposition | Approved or rework with actions | Next step undefined |
The two sections that most often get compressed are process settings and the conclusion criteria, and they are exactly the two that decide whether the report has value. Process settings make a result reproducible; criteria make it judgeable. A report with rich data but no criteria invites a debate at approval, and a report with criteria but no settings invites a repeat of the same debate during production.
The disposition section is what turns a report into a workflow step. Stating approved, conditionally approved with named conditions, or returned for rework with specific actions gives the next stage a trigger, which is how a trial connects to production release rather than ending in a folder. The tooling context behind the trial is described in our guide to molded pulp tooling design.
Data: ASTM paper, board and packaging standards define the dimensional and mass measurement methods used to verify formed fiber parts, which is the measurement basis a trial report's data section rests on.
Judgment: Record measured values against pre-agreed acceptance criteria rather than a pass or fail impression, because a numeric result can be re-checked while an impression cannot be challenged or reproduced.
Source: ASTM International - Paper, Board & Packaging Standards (2024)
Sampling Rules That Expose Variation
Sampling exists to find variation, not to confirm the best case.
| Sampling dimension | What it must cover | Reason |
|---|---|---|
| Cavities | Every cavity on the tool | Deposit varies by position |
| Time in run | Startup, middle, end | Warm tool behaves differently |
| Forming difficulty | Hardest corners and relief | Failures concentrate there |
| Drying position | Edges and center of dryer | Drying is not uniform |
| Batch and furnish | More than one pulp lot | Recipe variation is real |
Sampling only the first parts off a warm tool reports a result the production run will not reproduce, because a cold tool on startup forms differently from a warm one. Sampling only the easiest cavity hides the positional variation that a multi-cavity tool is prone to. A trial that follows these five dimensions produces a result that anticipates production rather than flattering the tool.
The temporal dimension also matters for the trial's process settings. If the trial runs only at startup, its recorded settings describe startup conditions, and a production team applying them mid-run will get a different result. Recording settings at each sampling point, rather than once, makes the report a map of the run instead of a single snapshot.
Data: TAPPI molded fiber resources describe the forming and drying steps in which deposit and moisture vary with tool position and process conditions, which is why sampling must cover position and time rather than a single point.
Judgment: Sample across cavities and across the run, because a trial that reports only startup parts understates the variation the production run will actually deliver.
Source: TAPPI - Molded Fiber & Pulp Resources (2024)
Acceptance Criteria, Statistics and Honesty
A conclusion is only as good as the criteria and the sample behind it.
| Element | Weak version | Defensible version |
|---|---|---|
| Criteria | Decided at review | Agreed before trial |
| Sample size | Convenient handful | Covers tool variation |
| Measurement | Single reading | Repeated per sample |
| Reporting | Average only | Average and spread |
| Comparison | Against drawing only | Against drawing and spec limits |
Reporting an average without the spread is the most common way a trial report overstates a result. A tool that produces parts with an acceptable average but a wide spread will generate field complaints, because the spread is exactly what the customer will see. Reporting the range alongside the average is what makes the report honest, and it costs nothing beyond a second column in the data table.
Repeated readings matter for the same reason. A single measurement per sample cannot distinguish measurement noise from part variation, so the report loses the ability to say which is which. Two or three readings per sample, reported individually rather than averaged, preserve that distinction and give the reader the evidence needed to judge whether a marginal result is stable or simply lucky.
Data: ISO quality and organization standards describe documented requirements for records, control of monitoring and measurement, and traceability, which support a trial report that can be audited rather than merely reviewed.
Judgment: Version-control the trial report against the tool revision it tested, because a conclusion drawn on one revision cannot support a decision about a tool that has since changed.
Source: International Organization for Standardization - Standards Catalogue, Quality & Organization (2024)
From Trial Report to Production Release
A good report answers the questions production will ask.
| Production question | Report section that answers it | If absent |
|---|---|---|
| Can we run this tool? | Disposition | Production hesitates |
| What settings do we use? | Process settings | Settings re-invented |
| What defects to expect? | Defect list | Defects treated as new |
| What varies and how much? | Data with spread | First complaints unnoticed |
| When do we re-check? | Conditions on approval | Drift goes unmeasured |
The last row is the one that keeps a trial relevant after launch. A conditional approval with named re-check conditions, such as a review after a set number of cycles or after the first full production lot, converts the trial from a one-time event into the first point of a monitoring series. The production parameters that follow from it are the same ones a first-article inspection verifies, as described in our guide to first-article inspection.
Trials run on a prototype or printed tool follow the same reporting rules, since the purpose is identical: to record what was tested, under what settings, with what spread, against what criteria. That consistency is what lets a prototype trial and a production trial be compared rather than treated as unrelated exercises.
Data: ISTA transit test protocols define how packaged goods are tested against distribution hazards, which is the functional test a protective molded pulp part must ultimately satisfy after tool release.
Judgment: Add transit performance to the trial's functional section where the part is protective, because a tool that passes dimensional checks but fails distribution testing has not been validated for its actual job.
Source: International Safe Transit Association - Transit Test Protocols (2024)
The Bottom Line
A mold trial report is proof, so it needs settings, sampling across the tool's variation, measurements with their spread, pre-agreed criteria and a clear disposition. Version it against the tool revision, and let it launch production rather than end in a folder.