GB 4806: Chinese Food Standard for Pet Bag Parts

Wholesale pet bag sourcing desk · Updated 2026-10-06 · 14 min read

Wholesale heavy duty pet carrier builds start at the base panel, and Material & Technology documentation should state the board thickness before the fabric is chosen. Wholesale scratch resistant pet carrier and wholesale tear resistant pet carrier claims only hold when the test method is named, otherwise the coating alone is doing the work.

How the GB 4806 Series Is Structured

The national food contact standard series is built in layers. A general safety part sets the framework - basic requirements, restricted substances, labelling and the principle that materials must not endanger health or alter food unacceptably. Material-specific parts then address individual groups: plastics, rubber, coatings, paper and board, metals, ceramics, glass and plant-fibre materials among them. Test methods are published separately in the GB 31604 series, which is where laboratories find the procedures for migration, residue and elemental analysis.

That separation matters for a buyer drafting a test request. The requirement and the method live in different documents, and a laboratory instructed to test "to GB 4806" will ask which part and which method. Naming both - the material part and the specific method numbers - produces a report that a reviewer can accept without further questions.

The series is also revised periodically, and revisions change both limits and method references. A report citing a superseded version is not wrong about the sample but is out of date about the requirement, which is enough for a reviewer to reject it. Recording the version cited on every report is a one-line habit that prevents an annual problem.

Finally, the series applies to materials and articles intended to contact food. As in other jurisdictions, scope follows intent: a bowl, a liner, a spout and a gasket are in scope, while a shell fabric and a shoulder strap are not. Establishing that scope in the technical file is the first and cheapest step in the whole exercise.

Buyers new to the series should expect a short learning curve on numbering. Parts are referenced by number rather than by name in most documentation, and a supplier's certificate may cite only the number. Keeping an internal map of which number corresponds to which material group saves repeated lookup and, more usefully, makes an obviously wrong citation visible at a glance.

General Safety Part Versus Material-Specific Parts

A useful way to read the series is as a general part plus a material part, applied together. The general part supplies the principles and the cross-cutting requirements - basic safety, production control, labelling, and the restricted-substance framework. The material part supplies the numbers: which substances are permitted, which are restricted, and what the migration or residue limits are for that material group.

For pet bag components, plastics and rubber are the parts that usually apply, with coatings relevant where a liner or bowl has a surface treatment. Silicone sits within the rubber grouping in Chinese practice, which surprises buyers who assume it is handled like a plastic. Getting that classification right determines which limits apply and therefore which tests are ordered.

Two rows generate the most queries. Silicone grouping, because buyers order plastics tests for a material assessed under the rubber part. And internal prints, because decoration applied to a contact surface needs its own evidence and is routinely omitted from the request.

Where a component combines materials - a plastic body with a rubber seal - each material is assessed under its own part and the article is documented as an assembly. The weakest material determines the file, so identifying the combination at RFQ prevents a late discovery during review.

Where a component combines several materials, the file should describe the assembly rather than the dominant material. A bowl with a plastic rim and a silicone body needs evidence for both, and a report describing only the body leaves a documented gap at the rim. Assemblies are where national-standard files most often fail review, because the buyer requested evidence for a component and the supplier answered for a material.

GB 31604 Test Methods: What Laboratories Actually Run

The method series is the operational heart of the system. It contains procedures for overall migration, specific migration of named substances, residue determination, heavy metal and elemental analysis, and sensory assessment, among others. A competent laboratory maps a material and a use profile onto the relevant method numbers and reports results with the method cited.

The buyer's contribution, as in every jurisdiction, is the use profile. Food type, contact temperature, contact time and whether contact is repeated determine simulant choice and conditions. Pet components see warm aqueous food, fatty treats and repeated washing, so the profile should say so explicitly rather than leaving the laboratory to apply a default.

Simulant selection is defined by the method rather than chosen freely, and the definitions differ slightly from European and US practice in places. A buyer running a multi-market programme should not assume that one set of conditions covers all three jurisdictions; the efficient approach is to test to the strictest profile and check that the result set satisfies each jurisdiction's reporting requirements.

Reporting format matters more than buyers expect. A report should cite the method number, the simulant, the conditions and the applicable limit, with the sample identified as production-intent. Reports lacking any of those four elements generate follow-up questions, and follow-up questions in a component programme tend to arrive during the sample window rather than before it.

Conditioning and sample preparation deserve explicit instruction. Whether a sample is tested as received, washed or pre-conditioned changes results, particularly for residue determinations. Stating the preparation in the test request prevents the situation where a buyer's report and a supplier's report disagree because each laboratory prepared the sample differently.

Using the Standard When Components Are Sourced in China

The most common situation for export-focused pet bag buyers is not selling into China but buying a component from a Chinese manufacturer. In that case the national series is a useful baseline rather than a legal requirement: a component maker that can produce evidence against it has a functioning laboratory relationship, documented material control and a quality system capable of meeting customer requirements elsewhere.

The practical method is to request the national-series evidence as a first filter, then add destination-specific testing. Where a supplier cannot produce any national-series documentation, that is informative about its maturity, and the buyer should expect to commission more testing rather than less. Our production team uses exactly this filter when qualifying component makers for export programmes.

There is a negotiation benefit as well. Suppliers with current national-series evidence can demonstrate it quickly, and that demonstration is a credible signal of capacity and discipline during price discussion. Buyers who ask for it routinely find that the suppliers able to produce it are also the ones whose delivery and consistency records are strongest, which is not a coincidence but a consequence of the same underlying controls.

Consolidation follows naturally. A single qualified silicone article, evidenced once and used across a range, removes repeated qualification work. Buyers who consolidate components across SKUs typically reduce annual testing spend by a meaningful margin and, more importantly, reduce the number of files that can go stale.

Change control is the discipline that makes consolidation safe. Any change of grade, cure system, pigment or moulding supplier is a re-qualification event, and the change log is what keeps the consolidated file accurate. This is administrative work with a direct financial return: it prevents a silent substitution from invalidating evidence that several SKUs depend on.

Using the Standard When Selling Into the China Market

Where a pet bag programme is sold into China, the national series becomes a requirement rather than a filter, and the buyer's obligations expand. Components must meet the applicable parts, the article must carry correct labelling and instructions in Chinese, and the operator placing the product on the market must be able to produce documentation on request. Domestic market surveillance in this category is active, and e-commerce platforms apply their own document checks before listing.

Platform checks are often the practical gate. Major marketplaces require a documentary set before a listing goes live, and incomplete documentation delays launch rather than blocking it permanently - but a delay during a promotional window costs more than the testing would have. Collecting the set at RFQ is the only reliable protection.

Labelling deserves specific attention for the domestic market. Product name, material, use conditions, manufacturer or agent identity and any required warnings must appear in Chinese and must match the documentation. A mismatch between a declaration and a listing description is a common ground for a platform rejection.

Buyers should also identify the responsible operator early. Whether the importer, a local agent or a distributor holds the documentation obligation changes who must keep the file and who answers an enquiry. Deciding that at contract stage avoids the situation where every party assumes another is holding the evidence.

Channel choice affects documentation burden. Marketplace listing, distributor supply and direct retail each carry different document expectations, and the marketplace route is usually the most demanding in volume of paperwork if not in substance. Identifying the channel before commissioning testing allows the document set to be assembled once in the right format rather than twice in the wrong one.

Documentation and Declaration Practice in China-Facing Programmes

The documentation set for a China-facing programme mirrors other jurisdictions in structure: material identification, a supplier declaration or certificate of conformity citing the applicable parts, laboratory reports with methods and conditions, and lot traceability linking tested material to shipped goods. The differences are in detail - Chinese-language labelling, the identity of the responsible operator, and platform-specific document formats.

Supplier declarations in this context are frequently issued as a certificate of conformity or a test-report package rather than as a narrative declaration. Either is workable provided it names the material, cites the parts and methods, and carries a date and an issuing entity. A certificate without a material identity is decorative, and buyers should ask for the material line to be added rather than accept it.

Retention and access matter practically. Authorities and platforms may ask for documents months after a listing goes live, and the request window is usually short. Keeping a single indexed pack per component - identity, declaration, reports, lots - allows a response in hours. Our production team maintains exactly that pack, indexed by component article rather than by bag SKU, which is what allows one file to serve a whole range.

Where the programme involves a local agent or distributor, agree in writing who holds the master file and who responds to enquiries, and make sure both parties hold the same version. Divergent files between principal and agent are a recurring cause of failed submissions in this market.

Where declarations are issued per lot rather than per article, buyers should confirm that the lot numbering appears on the physical packaging as well as on the paperwork. A lot number that exists only on a certificate cannot be checked at goods-in, which defeats its purpose. Requiring the marking on the inner bag or carton label is a small specification change with real operational value.

Platform document reviews are frequently triggered by listing updates rather than by new listings. Changing a product image or a material description can reopen a review that was previously closed, so any listing edit should be checked against the file before it is published. Buyers who treat listing maintenance as a compliance event avoid the surprise of a live product suddenly requiring documents that were never re-collected.

Cost, Timing and Consolidation for a Component File

A full component evidence set typically runs CNY 4,000-14,000 per material family, depending on how many method determinations are required and whether pigment or coating work is added. Silicone with residue and extractables testing sits at the upper end; a simple plastic film sits at the lower end. The range is wide because scope varies with material and colour, so the test request should be agreed with the laboratory rather than copied from a previous season.

Timing is the binding constraint. Laboratory work occupies two to five weeks depending on conditioning and queue, and capacity tightens ahead of domestic promotional seasons. The correct plan freezes the component at RFQ and runs testing during the 6-10 working day sample window so that results precede the 35-50 day bulk run. Programmes that commission testing after sample approval routinely add a month.

Consolidation remains the main lever. One qualified silicone article across several bag SKUs costs one evidence set; separate articles cost several. Standard component colours avoid both moulding minimums above the bag MOQ of 500 and additional colourant screening, which is why our production team recommends differentiating at bag level rather than at component level.

Finally, plan re-testing. Evidence ages, standards are revised and suppliers change formulations. A simple annual review - check versions cited, check material identity unchanged, check report dates - is enough to keep a consolidated file valid and is far cheaper than rebuilding it after a failed submission.

Budget behaviour differs between the first season and later ones. First-season cost is dominated by testing and supplier qualification; later seasons are dominated by maintenance - version checks, refresh testing and change reviews. Buyers who budget only for the first year routinely find the second-year file stale, which is the point at which a previously clean programme starts generating rejected submissions.

Common Gaps in a National-Standard File

Four gaps appear repeatedly. The first is wrong grouping: plastics tests ordered for a material assessed under the rubber part, most often silicone. The second is a missing layer: the weld layer of a laminated film, or an internal print, omitted because the buyer asked about the base material.

The third is version drift: reports citing a superseded standard version, which reviewers reject even though the sample result was valid. The fourth is traceability absence: reports with no lot reference, so the tested material cannot be linked to the shipped components. Each gap is administrative and each is preventable with a checklist applied before submission.

A fifth gap is scope creep in the other direction - claiming the standard applies to the whole bag because one component meets it. That is a claim no document supports and it invites a platform or authority enquiry that the file cannot answer.

The remedy is a one-page file index reviewed before each submission: component identity, material grouping, parts and methods cited, report dates and versions, lot reference, and the claim wording actually used in listings. Five minutes of review prevents most of what goes wrong in this category.

One further gap is worth naming: absent use conditions. A report that states results without the food type, temperature and contact time cannot be matched to the product, and a reviewer will ask. Requiring conditions on every report, in the same format, is a one-line instruction in the test request that eliminates an entire category of follow-up.

Coordinating a China Set With US and EU Evidence

A component sold in three markets does not need three unrelated test programmes. Migration science, simulant logic and method structure overlap substantially across jurisdictions, so a well-designed programme tests once to the strictest plausible profile and issues jurisdiction-specific documentation from the same results. The differences that remain are reporting format, declaration content and labelling, not the underlying measurements.

The coordination work is mostly at the request stage. Identify the strictest simulant set, the highest temperature and the longest contact time among the jurisdictions involved, and instruct the laboratory accordingly. Where one jurisdiction requires a determination the others do not - a residue test, an extractables set - add it once rather than commissioning a separate programme later.

Documentation then diverges by design. The material supplier issues the declaration or certificate appropriate to each market from one result set, and the buyer maintains an index showing which report supports which claim. That index is what makes a multi-market file reviewable, and it is the artefact most often missing from the files buyers bring to us.

The payoff is a smaller annual testing bill and a shorter launch sequence. For wholesale pet bag programmes selling into more than one region, this is the highest-return compliance decision available: one tested component, three declarations, one traceable lot record.

The organisational requirement for a multi-market file is modest but non-negotiable: one person or function must own the index. Where component documentation is held separately by sourcing, quality and regulatory colleagues, the index fragments and each market sees only part of the picture. A single owner, with a single index showing which report supports which claim in which market, is what makes a consolidated programme auditable. Independent support for multi-jurisdiction programmes is available through organisations such as SGS, and process expectations are documented by ISO. Verification bodies including OEKO-TEX also publish limit catalogues useful for textile-adjacent components.