Are the figures on this site measured by a third party?

No. The values on the model pages are the manufacturer’s typical values from the printed catalogue, produced to the GB/T method named on each page. Third-party reports exist only where a customer or a project paid for one; we do not present them as routine.

What is the difference between a typical value and a batch result?

A typical value describes what the product is designed to do. A batch result describes what was in the drum on a given date. Two batches can differ by several percent in tensile strength and both still be in specification, so a specification written against the typical value alone is a specification with no tolerance.

Can I get the raw test data?

Ask for the batch record for the lot you are buying: batch number, test date, method, result, specification range and the operator. Where a test was subcontracted, the report names the laboratory. What we will not do is send a curve without the method behind it.

How the numbers on this site are made

Last updated: 2026-11-12

A sealant page is full of numbers: tensile strength, elongation, skin time, density, movement class. They look interchangeable and they are not. Some are design targets, some are measurements of a specific lot, some are field observations with no instrument behind them. This page explains which is which, how each is produced, and where the limits are — so that a buyer can tell what a figure is worth before it goes into a specification.

Three kinds of number

Kind	What it describes	Where it comes from	What it is good for
Typical value	The product as designed	Printed catalogue, measured to the named GB/T method	Comparing grades and writing a first specification
Batch result	The lot you are actually buying	Factory test record for that batch number	Accepting or rejecting a shipment
Field result	One joint, one substrate, one applicator	Observation after cure, sometimes destructive testing	Diagnosing a failure, not predicting performance

Most specification mistakes come from mixing the three. A typical value used as a minimum rejects good batches. A field result used as a product claim turns one installer’s experience into a promise the next installer cannot keep.

How a typical value is produced

Every figure on a model page names its method — GB/T 528 for tensile and elongation, GB/T 7124 for tensile shear, GB/T 531.1 for Shore A, the GB/T 13477 series for skin time, sag, extrusion rate and elastic recovery, GB/T 2793 for volatile content, HG/T 4363 for cure rate. The method decides the specimen shape, the conditioning time, the pull rate and the arithmetic; change any of them and the number moves.

Values are reported as the manufacturer’s typical figure, rounded to the precision the method supports. They are not averages of every batch ever made, and they are not a guarantee that the next batch lands on the same point. What the batch record adds is the spread: the result for that lot, against the specification range the product is released to.

The full method table, with the reason each property matters at the joint rather than on the data sheet, is on the quality page.

What a batch record should contain

If a supplier cannot produce these six fields for the lot in front of you, the certificate they sent is decoration.

Field	Why it matters
Batch number and production date	Ties the drum to a record; without it nothing else can be checked
Test method	A result without a method cannot be compared with anything
Measured result	The actual figure for that lot, not the catalogue value
Specification range	What the batch was released against; a result with no range proves nothing
Operator or laboratory	Who ran the test, in-house or subcontracted
Retention sample reference	Lets a dispute be settled by re-testing the same material

What these numbers cannot tell you

Whether it will bond to your substrate. Tensile strength is measured on a specimen, not on your anodised aluminium, your plastic or your oily steel. Adhesion is a separate test on a separate coupon.
Whether the joint geometry is right. A 25LM sealant in a joint that moves 40% will fail at the same rate as a cheaper grade. Movement capability is a property; joint design is an engineering decision.
Whether the applicator will get it right. Surface preparation, tooling time and temperature decide more outcomes than the data sheet does.
How it ages in your climate. Lab ageing is accelerated and finite. A ten-year façade is not a 1,000-hour test.
Whether two products are compatible. A primer, a previous sealant or a coating can change the result; compatibility is tested, not assumed.
What the batch will do next month. Shelf life and storage conditions are separate claims with their own evidence.

How to reproduce or challenge a figure

Take the method from the model page and the batch number from the drum.
Ask for the batch record and the retention sample reference.
If the result matters commercially, send a sample to a laboratory you chose and pay for — not one the supplier chose.
Compare like with like: same method, same conditioning, same specimen geometry. A result produced to a different standard is a different measurement.

If our figure does not survive that process, we will correct the page and say what changed. That is the point of publishing the method next to the number.

Limits of this site’s own method

This is a manufacturer’s site, and the evidence has the limits that implies:

The typical values originate with the manufacturer. We name the method; we do not claim an independent audit of every batch.
We do not publish customer case data, site photographs of installed joints, or failure statistics, because we cannot verify them to the standard this page sets.
Third-party test reports exist only where a customer or project commissioned one. Where a page says a report exists for a model, ask for it by number.
Translations are checked against the English source for figures and method numbers; if a translated page and this one disagree, the English page and the batch record govern.
Where a claim depends on a certificate we do not hold a copy of, the page says so rather than printing the claim.

Corrections

If a figure here is wrong, or a method is misattributed, write to [email protected] with the page URL and the source you are comparing against. Corrections that change a number or a method are dated on the page; wording changes are not. We do not delete a page to avoid a correction — a retracted claim with a date is more useful to the next buyer than a silently edited one.

Questions this page usually raises

Are the figures on this site measured by a third party?: No. The values on the model pages are the manufacturer’s typical values from the printed catalogue, produced to the GB/T method named on each page. Third-party reports exist only where a customer or a project paid for one; we do not present them as routine.
What is the difference between a typical value and a batch result?: A typical value describes what the product is designed to do. A batch result describes what was in the drum on a given date. Two batches can differ by several percent in tensile strength and both still be in specification, so a specification written against the typical value alone is a specification with no tolerance.
Can I get the raw test data?: Ask for the batch record for the lot you are buying: batch number, test date, method, result, specification range and the operator. Where a test was subcontracted, the report names the laboratory. What we will not do is send a curve without the method behind it.