3D Printing First Article Approval: What to Check Before Production

First article approval checklist: what the sample must prove

Write the approval questions before the part arrives. Otherwise the review tends to reward whatever is easiest to notice - usually appearance - while fit, assembly, load, temperature, handling, or packaging risk stays untested.

If the part is still exploring geometry, material direction, or use conditions, it is probably a prototype rather than a release sample. Back up to the prototype-versus-production guide before turning an experiment into a purchasing approval.

A six-step first article approval process

1. Identify the exact sample. Record the part number, file revision, variant, sample ID, material, color, and date received.
2. Confirm the delivered condition. Note orientation-sensitive surfaces, support removal, finishing, installed hardware, labels, and packaging included in the proof.
3. Run the planned checks. Test the sample with the agreed mating parts, gauges, use conditions, and cosmetic standard.
4. Record evidence and exceptions. Capture pass/fail results, photos, measured values where useful, and every condition that was not actually tested.
5. Choose the release state. Approve another sample, a limited pilot, or the named production quantity; do not let commercial approval silently become full release.
6. Issue one controlled approval note. Name the approved baseline, remaining conditions, release quantity, and events that require reapproval.

The process can be short for a simple low-risk part. The discipline is the same: the approval must point to evidence and authorize a defined next action.

Freeze the exact file, revision, and sample identity

A physical sample without identity is difficult to control. Mark the unit, bag, traveler, or approval record with a sample ID and connect it to the controlling model and drawing revision. If an STL or 3MF, STEP file, drawing, and written note disagree, state which source governs geometry, tolerances, appearance, and assembly.

Ask whether the shop repaired, scaled, split, oriented, or otherwise changed the supplied file to make the sample. A legitimate repair is not a problem, but the production input should be retained and named. Approval should not depend on an unrecorded operator edit living on one workstation.

For fit-sensitive work, use the fit, tolerance, and file-version guide to separate critical acceptance features from general dimensions that only describe the model.

Test fit and function in the real assembly condition

Measure when measurement answers the requirement. Test the assembly when the assembly is the requirement. A snap feature, shaft fit, screw boss, gasket land, sliding guide, or replacement-part interface can pass a nominal dimension and still fail because the mating part, insertion direction, surface texture, or load condition was not represented.

Record the setup that produced the pass:

• the exact mating part, hardware revision, gauge, or fixture;
• the part condition when tested, including cleanup, conditioning, or installed inserts;
• the assembly direction and any force or torque limit that matters;
• the temperature, load, exposure, or duty cycle actually applied; and
• whether the check is destructive and therefore needs a separate retained sample.

If only one uncontrolled mating part was available, say so. It may prove that one combination works, but not that production will fit the full variation of field assemblies. For replacement work, the replacement-part printing service guide explains when the original, mating hardware, or a fit gauge should travel with the job.

Approve finish by zone, not with the word "clean"

FDM layer lines, seams, top-surface texture, bridging, support contact, and minor color variation can all be normal. Resin parts can show support marks, drainage evidence, and surface changes from post-processing. Approval should identify which faces are cosmetic, which are functional, which will be hidden, and what cleanup is included.

Photographs help document appearance, but use consistent lighting, distance, and orientation. If color matters, a screen image is weak evidence by itself. If texture or gloss matters, identify the approved physical reference or viewing condition. See what surface finish to expect from a custom 3D printed part before rejecting normal process evidence or accidentally accepting an expensive hand-finished hero sample as the production standard.

Separate sample approval from production release

These are related decisions, but they are not always the same decision. A sample can pass fit while labels, production quantity, material availability, packaging, or purchasing authority remains open.

Use the custom 3D printing quote-approval guide if the commercial baseline is still unclear. A sample cannot repair a quote that never defined the real job.

Ask what one sample does not prove about repeatability

One passing unit proves that the shop produced one passing unit under a particular set of conditions. It does not automatically prove batch consistency, multiple-machine equivalence, long-run thermal stability, color continuity, packaging throughput, or every unit in a family.

A production-minded supplier should be able to explain:

• which critical checks carry from the sample into production;
• what is checked on the first production units, at intervals, or on every part;
• whether production may move across machines, operators, orientations, or material lots;
• which changes require buyer notice or a new sample;
• how nonconforming units are contained before the full batch is exposed; and
• what evidence is retained for the next reorder.

This does not mean every job needs a formal inspection report. The control should match the consequence of failure. A low-risk spacer may need a fast gauge check. A fit-critical assembly part may justify a first-production check, retained reference, or pilot quantity. The production operator behind GoodPrints, JC Print Farm, should be judged by the same practical standard: can the team connect the approved sample to a realistic repeat-production plan?

Use a pilot batch when the open risk is production behavior

A pilot is useful when one sample can prove geometry but cannot efficiently prove queue behavior, process spread, assembly time, label flow, packaging labor, transit protection, or receiving usability. The pilot should be large enough to expose the open risk and small enough to limit the cost of being wrong.

Define the pilot question before choosing the quantity. Examples include:

• Can multiple units hold the mating fit without selective assembly?
• Can the shop install inserts or magnets without damaging the cosmetic face?
• Do labels, kit counts, and carton groupings work for receiving?
• Does a second machine or material lot stay inside the accepted functional window?
• Can the part survive normal packing and transport without edge or surface damage?

Then define who reviews the pilot and what releases the balance. The guide to when a pilot batch should replace a direct jump to production helps size that decision around risk rather than habit.

Control variants, handed parts, and supplied components

Do not let approval of one size, color, left-hand part, or material quietly authorize every related SKU. Group variants by what they genuinely share. A common material and process may justify family approval for low-risk cosmetic changes, while different mating geometry, wall thickness, load direction, or hardware can require separate proof.

Buyer-supplied items need identity too. Record the hardware, label, insert, magnet, adhesive, or mating component used for the approved sample. If final components were unavailable and a substitute was used, write the approval as conditional and state who owns the final verification.

Packaging belongs in the same boundary. A loose sample may approve the printed part but not the label, kit, bag, protection, or carton. Route those open items through the guide to packaging, labeling, and inspection before a batch starts.

Keep a usable approval baseline

The baseline can include a retained physical sample, but it should never rely on the physical part alone. Samples age, get damaged, disappear into an assembly, or become ambiguous when several versions sit on the same shelf.

Keep a compact record with:

• part number, revision, sample ID, approval date, and approver;
• controlling production files and any supplier-created derivative;
• material, color, finish, hardware, and packaging condition;
• critical check results and the exact mating references or gauges used;
• photos of cosmetic zones, orientation, markings, and accepted exceptions;
• the release state and quantity authorized; and
• changes that require a new sample, deviation, or buyer approval.

If the part will reorder, decide who keeps the physical reference, how it is labeled, whether it is allowed to be consumed in testing, and when it stops being trustworthy. Use the guide to keeping custom 3D printing reorders consistent to carry that record forward.

Send one buyer-ready approval note

A clear note can be brief because it points to a controlled record instead of retelling the whole project.

First-article approval template

We approve sample [ID] for part [part number/name], revision [revision], produced from files [filenames], in [material/color] with [finish, hardware, and packaging condition]. It passed [critical fit, function, appearance, or pack-out checks] using [mating part, gauge, setup, or test]. Accepted exceptions are [list]. This approval releases [pilot or production quantity] under the quoted process and inspection plan. It does not approve [open variants, labels, hardware, packaging, or other conditions]. Please acknowledge any production change or conflict before starting.

If the sample passes but production remains held, say that directly: "The design baseline is approved; full production remains blocked until separate written release." Purchasing language, a purchase order, and technical approval should all point to the same release state.

Know what should reopen approval

Reapproval is justified when a change can alter the accepted outcome. It is not necessary for every administrative correction.

• Usually reopen: geometry or revision changes, material or grade substitution, different critical hardware, meaningful orientation or machine-family change, altered finishing, new cosmetic surfaces, changed inspection method, or a pack-out change that affects the part.
• Evaluate case by case: color-lot change, equivalent machine assignment, alternate supplier for noncritical hardware, or a quantity increase that changes run conditions.
• Usually administrative: corrected contact details, unchanged ship-to formatting, or documentation cleanup that does not alter the product or delivery promise.

If production parts miss the baseline, stop multiplying exposure. Contain affected units, compare evidence against the released requirement, determine whether the cause is supplier execution or a changed buyer assumption, and agree on correction before restarting. The guide to what to do when production parts do not match the approved sample covers that escalation path.

Red flags before you release the production run

• The sample cannot be tied to one file revision or named production input.
• Approval is based on appearance even though fit or function is the real failure risk.
• The sample used temporary hardware or a substitute mating part, but the approval does not say so.
• One sample is being stretched across materially different sizes, variants, or handed parts.
• The supplier cannot state which sample conditions will be repeated in production.
• A heavily hand-finished hero sample is presented as representative of normal batch output.
• Packaging, labels, kits, or supplied components remain assumed after the part itself passes.
• Nobody has clear authority to approve, reject, or release the next quantity.
• The approval lives only in a chat, inbox, or unlabeled physical part.
• Any process change is treated as harmless, or every small note is treated as a reason to reset the entire job.

A red flag is a reason to narrow the release or ask for evidence. It is not automatically proof that a shop cannot do the work. Use the production-readiness guide when the broader supplier-control question remains open.

Choose the next step that matches the result

If the files, quantity, sample result, and release notes are controlled, use the JC Print Farm quote intake for a production-minded next step. The useful outcome is not an aggressive release; it is a release that the supplier can repeat and the buyer can verify.

Common questions

Is a first article the same as a prototype?

Not usually. A prototype helps discover or revise the design. A first article or approval sample should represent a defined production candidate and prove named acceptance conditions. One physical part can serve both roles only if the transition and final production baseline are documented clearly.

Does approving one sample approve the full batch?

Only if the written approval says it releases the full named quantity and all other production conditions are closed. A safer record distinguishes design/sample approval from pilot or full-production release.

Should the buyer keep the approved sample?

Keep or identify a physical sample when visual, tactile, assembly, or fit evidence is hard to capture completely in records. Also retain the files, photos, checks, exceptions, and sample ID so the baseline survives if the physical unit is lost or damaged.

Can a sample be approved from photos or video?

Yes for limited questions that remote media can prove, such as general appearance, markings, or a demonstrated motion. It is weaker for color, texture, exact fit, force, rigidity, sealing, or hidden assembly conditions. State what remote approval did and did not prove.

How many first articles are needed?

One may be enough for a low-risk, single-variant part with a stable process. Multiple units or a pilot are more useful when the risk involves repeatability, multiple machines, variants, assembly labor, packaging, or transit rather than the geometry of one unit.

What if the approved sample was made with special hand work?

Record that work and decide whether it is included in production. If normal batch economics cannot support the same finishing or tuning, the sample is not a fair production baseline. Approve a representative process or change the quote and acceptance standard.

Related buyer guides

• How to Approve a Custom 3D Printing Quote
• Do You Need a Prototype Before Ordering a Small Batch?
• When Should a Pilot Batch Replace a Direct Jump to Production?
• Packaging, Labeling, and Inspection Details to Confirm
• How to Check a Custom 3D Printing Order When It Arrives
• How to Keep Custom 3D Printing Reorders Consistent