Mirror-finish stainless panels should not be accepted through one undefined instruction such as check flatness or reflection must be undistorted. Geometric flatness, reflected-image appearance and installed condition are different decision characteristics. Each needs its own panel identity, support state, reference, method, exclusions, evidence and authorized owner.
A dimensional check may establish a stated geometric result while a panel is free, supported, clamped, bonded or otherwise held in a named factory state. A reflection check may establish how a defined reference appears from a defined observer or camera position under defined lighting. Installed acceptance may include the actual substrate, fastening, seams, adjacent panels, protection state and site viewing relationship. None of those states automatically proves the others.
Use this release rule:
- Factory geometry evidence is valid only for the identified panel, support/restraint state, reference setup, characteristic, locations, method and decision rule recorded.
- Factory reflection evidence is valid only for the identified surface state, reflected reference, light, observer/camera geometry, represented seams and declared purpose.
- Installed evidence is valid only for the identified location, substrate/support, fastening, adjacency, seam, cleaning/protection and viewing state represented.
A shop-supported panel can pass its dimensional requirement and still leave installed reflection open. A visually straight reflected line is not a calibrated flatness result. A first article cannot approve an unrepresented substrate or installation condition. Hold acceptance when the evidence is being promoted beyond its named characteristic or state.
Name the characteristic before choosing the check
Begin with the customer's decision. Is the controlled characteristic a geometric form requirement on one panel face, the relationship of a visible field to a project reference, the continuity of reflected imagery across a seam, local waviness or print-through, overall installed appearance, or several separately accepted characteristics? The word flatness should not be asked to cover all of them.
Identify the functional reason. A dimensional requirement may protect fit, clearance, attachment, seam behavior or a downstream assembly. A reflected-image requirement may protect an architectural appearance at a named viewing location. Those purposes can point to different support states, spatial scales and evidence methods. Record both when both matter.
ISO 1101:2017 formally addresses geometrical tolerancing of form, orientation, location and run-out, but its public record does not select a flatness value, panel state, reference setup, instrument, sampling pattern or acceptance method for a StelPanel project.[1] If that framework is invoked, the complete controlled requirement and competent project interpretation remain necessary. This article does not reproduce protected clauses or claim conformity.
Separate a panel-face characteristic from an installed-field relationship. A requirement on one unassembled panel does not include seam steps, substrate variation or adjacent-panel alignment unless the documents explicitly say so. Conversely, a field appearance decision may be sensitive to connected installation conditions that are not properties of the isolated sheet.
Use the StelPanel colour and finish selection pages to identify mirror or decorative options for project review, not as a geometric or reflection specification. A displayed finish does not provide the panel tolerance, polishing route, support condition or accepted appearance.
Freeze the panel and support state
State changes can begin before inspection. Record whether the evidence concerns incoming sheet, cut blank, formed part, fabricated panel, panel with backing or stiffeners, temporarily supported first article, packed item, installed panel or completed multi-panel field. Name any protective film, cleaning, handling, temperature or conditioning state that affects the observation.
The support record should identify actual contact locations, fixtures, frames, pads, clamps, fasteners, bonding state and applied adjustment or force. Do not write only on inspection table or installed normally. Identify what physically supports the panel and which movements remain possible. If the panel is measured in more than one state, keep separate results rather than replacing them with one pass status.
The accepted publisher abstract for Boldsaikhan and coauthors describes robotic metrology and a least-squares flatness estimate used to compare sheet condition before and after refill friction stir spot-welding sequences in an aluminum lap-panel study.[2] Their materials, joining process, fitting method and numerical findings do not transfer to decorative stainless panels. The bounded workflow lesson is to define the measurement method, capture a named baseline and compare like with like after a controlled state change.
A first article should therefore have a state chain such as fabricated free panel -> defined factory support -> represented backing or frame -> seam mockup if included -> installed state if represented. Use only the states that exist. For each transition, identify which earlier result remains meaningful and which characteristic needs a new check.
The related StelTherm article explains why held geometry and fixture-released alignment are different evidence states. It does not supply a mirror-panel tolerance, support arrangement, measurement method or result.
Define the dimensional method and its boundary
For the dimensional check, record the panel ID and revision, surface and perimeter included, support state, reference plane or datum realization, measurement volume, locations or sampling strategy, instrument category, alignment or fitting rule, material and environmental conditions where relevant, the retained raw result, and the decision authority. State any edge, opening, seam, weld, attachment or local-feature exclusions.
Method categories may include contact points, a straightedge-based project procedure, coordinate measurement, scanning or another approved route. Naming a category is not enough. Each method observes and processes the surface differently, so the project plan must define the actual setup and how a result near a limit is decided. This article does not rank methods or prescribe point density, span, uncertainty or tolerance.
The reference plane deserves explicit control. If a best-fit operation is used, identify what data it includes and excludes. If physical datum features establish the setup, identify them and their state. Do not subtract support behavior, edge zones or local marks through an unstated fitting choice. The reported result must be interpretable against the drawing and acceptance rule.
Document the panel surface before and after any temporary support is removed when the free state is relevant. If installation forces or adjustments are part of the intended system, record the authorized installed condition separately. A factory team should not infer a site requirement, and a site team should not reject a factory state against an unstated installed reference.
Treat reflected-image distortion as a separate observation
Mirror appearance depends on the relationship among surface, reflected scene, light and observer. Gonzalez-Leal and coauthors examined optical and surface characteristics of bright-annealed AISI 430 flat sheets and reported observations tied to illumination and anisotropic appearance.[3] Their specimens are not mirror-finish fabricated panels, seams or installations, and the paper supplies no reflection grid, viewing distance, polishing route, acceptance threshold or StelPanel result. It supports only keeping optical and illumination context with the appearance claim.
Define what reflected-image question is being asked. A project may care about continuity of a straight reference, local waviness, a broad field reflection, a seam transition, edge pull, fastening print-through or another named effect. Each characteristic needs a represented scale and location. Do not use one attractive photograph to approve every visible condition.
The OSTI public abstract for Huang and coauthors' review describes phase-measuring deflectometry as a full-field three-dimensional shape-measurement technique for specular reflecting surfaces and says the review addresses principles, developments, challenges and limitations.[4] That evidence does not make an informal reflected grid a calibrated measurement system and does not provide a project acceptance rule. It establishes a useful boundary: formal specular-surface metrology needs a defined technique, setup and interpretation; casual visual reflection remains a different evidence type.
A project reflection setup should identify the reference object or pattern, its geometry and position, the panel orientation and support, light sources and ambient conditions, observer or camera location, distance and angle, lens or capture settings when photographs are retained, field of view, seams and adjacent panels represented, and the comparison or decision rule. No universal grid spacing, viewing distance or light level is supplied here.
Photographs can preserve evidence of the represented setup, but they can also change apparent geometry through viewpoint, lens, crop, focus, exposure and later display. Retain an overall setup view, panel and reference IDs, and the original record where photography is part of the approved method. Do not accept a cropped image with unknown geometry as a substitute for the project procedure.
Control viewing without borrowing a foreign acceptance rule
ISO 3664:2009 addresses viewing conditions for graphic technology and photography, not acceptance of fabricated mirror-finish stainless panels.[5] Its public scope does not supply a project illuminance, spectrum, surround, observer position, viewing distance or pass/fail criterion. Its bounded relevance is that viewing conditions must be named when repeatable visual comparison is intended.
For a factory reflection review, state whether the purpose is route development, first-article comparison, pre-pack evidence or customer appearance approval. For site review, name the installed location, expected operational lighting state, observer zone, adjacent finishes, cleaning state and whether temporary construction reflections are excluded. The project authority decides which conditions represent acceptance.
Use the existing StelPanel article for a deeper discussion of physical references and controlled viewing conditions. Hairline direction and mirror reflection are not interchangeable, but both require the evidence state and observer/light relationship to remain attached to the decision.
Separate local workmanship review from field reflection. Local inspection may address scratches, dents, contamination, edge damage, protective-film marks or other listed conditions under its own procedure. Field review may address how the reflected reference behaves across a larger visible area. A local pass does not close the field decision, and a distant field view does not waive local criteria.
Keep substrate, fastening and seams inside the installed state
An installed mirror panel is part of an assembly. Record the substrate or support source and revision, survey state, datum or setting-out basis, support locations, fastening or bonding concept, adjustment state, joint and corner details, adjacent panels, backing, edge conditions and installation sequence. Identify who owns each input and who may disposition a receiving condition outside the agreed basis.
Li and Hou modeled dimensional and geometric tolerance propagation through assembly relationships in their mechanical CAD study.[6] Their model is not a decorative-panel substrate, fastener, adhesive, installer or building, and it predicts no mirror distortion or project tolerance. Its bounded relevance is procedural: connected member relationships belong in the review instead of treating an isolated panel result as the complete installed system.
Do not make the fabricator responsible for an unspecified substrate, and do not make the installer responsible for changing the panel design without authority. The customer or design authority defines the installed requirement and allocation of tolerances. StelPanel may provide part and factory-assembly evidence within confirmed scope. The substrate authority supplies current receiving-condition evidence. The installer records the actual support, adjustment and seam state. The named appearance authority accepts the represented installed condition.
Seams need their own scope. A single-panel first article cannot prove reflected continuity across multiple panels. If seam reflection matters, define a representative assembly, adjacent panel identities, gap or reveal definition, support condition, finish orientation, fastening state, reflected reference and viewing geometry. State whether the decision covers one seam, one corner, a repeated family or the completed field. General elevation, section, first-article and mockup approval for visible joints and corners remains a separate topic; P2 uses seam state only where it changes mirror-reflection or installed evidence.
The StelPanel project-review checklist can help collect geometry, visible faces, joints, backing, finish and written scope. It does not select support, flatness, reflection or installation criteria. A controlled drawing handoff should also retain revision and precedence; the StelTank article on connecting drawing identity to manufacturing inputs provides adjacent workflow context without defining mirror panels.
Use the Mirror-Panel Geometry and Reflection Matrix
Create one paired matrix for each panel or represented panel family. Use separate rows when support, seam or installed locations differ materially.
Part A - Dimensional geometry state
| Field | Record for the identified panel | Hold when |
|---|---|---|
| Identity and requirement | Panel/assembly ID, revisions, visible face, characteristic, function and governing source | Flatness is undefined or asked to cover several different decisions |
| Support/restraint state | Contacts, fixture/frame, clamps, fasteners/bond, adjustment and applied force state | Evidence says only shop or installed without physical support definition |
| Reference and method | Datum/reference-plane realization, method, locations, fit/alignment rule, environment and exclusions | Setup or data treatment cannot be reconstructed |
| Result and evidence | Raw/processed result, uncertainty or decision treatment where required, record ID and owner | Only a pass label remains or a near-limit rule is missing |
| State transition | Free, supported, formed, assembled, packed and installed states that actually apply | Earlier evidence is promoted across an unverified state change |
Part B - Reflection and installed state
| Field | Record for the same panel or family | Hold when |
|---|---|---|
| Surface state | Finish/sample ID, processing state, protection/cleaning, orientation and local exclusions | The surface represented differs from delivery without a recorded boundary |
| Reflected reference | Reference object/pattern, geometry, position, scale, locations and characteristic reviewed | No distortion has no physical reference or represented scale |
| Light and observer/camera | Light state, ambient conditions, position, distance, angle, field of view and capture record | A photograph or visual opinion has unknown geometry |
| Assembly representation | Substrate/support source, fastening/adjustment, seams, corners, adjacent panels and location | A single supported panel is being used to approve an installed field |
| Decision | Factory or installed purpose, method, result, limitations, status and authorized owner | Geometry, reflection, workmanship and installation decisions are collapsed into one approval |
| Change trigger | Panel, finish, backing, support, fastener, seam, substrate, installation, light, reference or method change | Prior evidence is reused without impact review |
The matrix is an editorial decision tool, not a flatness standard, optical-distortion formula, prescribed grid, viewing protocol, installation tolerance, inspection plan, capability study or project result.
Read Part A from the characteristic to the measured state. Read Part B from the reflected reference to the represented installation. Then compare their boundaries. The two records may support the same release only when the project explicitly defines how both characteristics and states are accepted.
Define the first-article and change loop
A first article should identify panel and assembly revisions, material and finish reference, manufacturing route, support/backing state, seams represented, dimensional method, reflection setup, local-workmanship check, results, limitations and decision owners. A representative label is not enough; state what the article does and does not represent.
Before review, define who may change forming, polishing, backing, stiffening, bonding, fastening, support, seam detail, protective film, correction, dimensional method or reflection setup. Identify which changes require a new first article, which reopen installed review and which need design approval. An accepted factory panel does not authorize an undocumented change that alters its represented state.
Use statuses such as geometry definition pending, factory support verified, reflection setup agreed, first article accepted for named characteristics, installed state not represented, held for substrate review and accepted at identified location. Avoid an unqualified approved, because it hides the characteristic, state and authority.
The StelPanel delivery-stage guide can help locate sample, drawing, first-article, packing and site handoff decisions. It does not prove geometry, reflection, substrate, installation or acceptance for this project.
Issue a controlled review package
Send the controlled panel drawings and elevations, part and face IDs, mirror-finish sample/reference, material and surface condition, panel construction, backing/stiffener data, support and fastening concept, seams and corners, substrate/survey source, expected installed lighting/viewing state, dimensional requirement, reflection question and acceptance authorities.
Ask the supplier to confirm the proposed manufacturing and evidence scope for the actual project: represented panel size and construction, support state, available measurement route, reflection setup, sample/first-article plan, protection, records and exclusions. Capability to confirm is valid while first-party evidence is missing. Do not replace it with an unsupported equipment, flatness, polishing or appearance promise.
No cited paper or standard proves StelPanel sheet source, mirror-polishing process, equipment, panel envelope, flatness, reflection quality, backing, installation, substrate or accepted result. Every such claim needs controlled project evidence.
Final stop line
Mirror-panel geometry, reflected imagery and installed appearance can inform one another, but they are not interchangeable. A result is defensible only for the identified characteristic, surface, support state, reference setup, method, represented assembly and authority.
If the team cannot name the panel and support state, dimensional characteristic and method, reflection reference and viewing geometry, seam and substrate representation, and final decision owner, the first article is not ready for acceptance. Complete both parts of the Mirror-Panel Geometry and Reflection Matrix before production or installation is released.
References
- ISO 1101:2017, Geometrical product specifications (GPS) - Geometrical tolerancing - Tolerances of form, orientation, location and run-out. Official ISO record. Public normative subject only; no panel value, state, setup, method or conformity claim. Back to citation
- E. Boldsaikhan et al., "Metrology of Sheet Metal Distortion and Effects of Spot-Welding Sequences on Sheet Metal Distortion," Journal of Manufacturing and Materials Processing, 2023. DOI. The aluminum lap-panel method and results do not transfer to decorative stainless; use is limited to named-state baseline and comparison discipline. Back to citation
- J. M. Gonzalez-Leal et al., "Analysis of the Visual Appearance of AISI 430 Ferritic Stainless Steel Flat Sheets Manufactured by Cool Rolling and Bright Annealing," Metals, 2021. DOI. The flat bright-annealed specimens supply bounded optical/surface/illumination context only, not a mirror-panel criterion or result. Back to citation
- L. Huang, M. Idir, C. Zuo and A. Asundi, "Review of Phase Measuring Deflectometry," Optics and Lasers in Engineering, 2018. DOI. The review covers formal specular-surface measurement principles, developments and limitations; it supplies no project grid, visual shortcut or acceptance rule. Back to citation
- ISO 3664:2009, Graphic technology and photography - Viewing conditions. Official ISO record. The scope is not fabricated mirror-panel acceptance and supplies no project viewing geometry or limit. Back to citation
- C. Li and W. Hou, "Analysis of Assembly Tolerance Based on Assembly Constraint Information Model," Mathematical Problems in Engineering, 2021. DOI. The mechanical CAD model supports connected assembly review only; it predicts no panel, substrate, installation or reflection result. Back to citation