Test Standards and Batch Release

Why standardised test methods exist

A rubber property is not an absolute number. Tensile strength depends on specimen shape, pull rate, temperature, and how the sample was prepared. Change any of those and you get a different answer from the same material.

This is not a flaw in the tests. It reflects the fact that rubber is viscoelastic — its response genuinely depends on how you deform it. But it means that a number without a method attached is close to meaningless, and that two laboratories cannot compare results unless they are doing the same thing.

Standardised methods solve this by specifying everything that matters: specimen geometry, preparation, conditioning, equipment, rate, temperature, and how the result is calculated and reported. When a specification cites a method, it is buying that whole package, not just the property name.

The standards you will encounter

A practical caution: ASTM and ISO versions of the same test are frequently similar enough to look interchangeable and different enough to give different numbers. Specimen dimensions, pull rates, and conditioning periods vary. If a specification cites one and your laboratory runs the other, that is a real discrepancy rather than a technicality.

What a test method actually specifies

Reading a method for the first time, the level of detail can seem excessive. Every clause is there because someone found that varying it changed the answer.

• Specimen geometry and how it is cut or moulded.
• Sample preparation, including how long after cure the test may be run.
• Conditioning — temperature, humidity, and duration before testing.
• Equipment specification and calibration requirements.
• Test conditions: rate, temperature, applied load or strain.
• How the result is calculated, rounded, and reported.
• Precision statements — how much scatter to expect within and between laboratories.

The precision statement is the most neglected section and often the most useful. It tells you how much two results can legitimately differ before the difference means anything, which is exactly the question you need answered when a batch reads slightly out of specification.

Sample preparation and conditioning

More testing errors come from preparation than from the test itself.

Specimens are normally cut from moulded sheets or slabs cured under specified conditions, not from production parts, because a part has variable thickness and cure state. Where a part must be tested, the method usually says how and where to take the specimen.

Cut edge quality is worth singling out. A blunt die or a torn edge produces low tensile results that look like a material problem and are not. When tensile results come in low and scattered with no other explanation, the die is worth inspecting before the compound is questioned.

What gets tested and when

Testing happens at several points, and the purpose differs at each.

The distinction between batch release testing and qualification testing matters. Release tests are fast, run on every batch, and answer whether this batch is like the last one. Qualification tests are slow, run rarely, and answer whether the compound meets the specification at all. Confusing the two leads either to unnecessary cost or to unwarranted confidence.

Batch release

Batch release is the decision to allow a mixed compound to be used in production. It is normally based on a small number of fast tests, chosen because they detect the most likely errors quickly.

The usual set is compound Mooney viscosity and a rheometer cure curve, sometimes with specific gravity. Between them, these catch most weighing errors, most gross mixing errors, and most contamination. They are notably poor at one thing — dispersion — which is covered in Part C, because a badly dispersed batch can show entirely normal viscosity and cure behaviour.

The logic is worth understanding. Mooney viscosity reflects polymer, filler, and oil — the large additions. The rheometer curve reflects the cure system and interactions with the filler. Specific gravity reflects the overall composition and is very sensitive to a mineral filler error. Together they cover most of what can go wrong, in under half an hour, without waiting for a cure and a tensile test.