ASTM D1004

What the test does

A specimen is die cut to a shape carrying a 90° angle at one edge, honed sharp so the tearing force is concentrated into a very small region rather than spread along a cut. It is clamped at a short initial grip separation and pulled at a constant, deliberately slow rate until it fails. The peak force, normally reached close to the onset of tearing, is the tear resistance. Extension is taken from grip separation rather than from an extensometer, because this geometry has no gauge length worth instrumenting.

What it measures, and why it matters

The force required to start a tear, as distinct from the force required to keep one running. That distinction is the reason two tear methods exist for film: a sealed pouch, a liner or a carrier bag most often fails because something initiates a tear at an edge or a fold, and initiation is what this geometry probes. The method's own scope is careful about what the number can be used for — it furnishes comparative information for ranking specimens of similar composition, and service performance need not follow it. It is reliable for materials that neither fail brittlely nor stretch enormously; the method excludes brittle failures and specimens whose maximum extension exceeds 101.6 mm.

Two limits govern how the number may be quoted. Tear resistance depends partly on thickness but has no simple relationship with it, so the result is reported as a force and never normalised per unit thickness unless a correlation has been established for that material. Results are therefore comparable only across specimens whose thickness stays within ±10 % of the average for the set.

Specimen and die

The whole method rests on the die. The geometry exists to force the tear to start in one tiny region, and a die that has lost its edge changes the answer without changing anything you can see.

Geometry

Die-cut, with a 90° angle at the tearing edgeThe angle is honed sharp, with no radius or the smallest radius the cutting process can achieve.

Material covered

Flexible plastic film and sheetingFilm is defined in the method as sheeting of nominal thickness not greater than 0.25 mm.

Not applicable where

Brittle failure occurs, or extension exceeds 101.6 mmThe method also notes poorer reliability for materials elongating more than about 200 % during the test.

Specimens

At least ten; ten in each direction where the film is anisotropicExtruded film is anisotropic, so twenty is the working minimum. Materials that vary across the web need considerably more before the mean is meaningful.

Directions tested

Machine direction and transverse direction, separatelyPracticeExtruded film is anisotropic. A single averaged figure describes neither direction.

Thickness spread within a set

Within ±10 % of the averageOutside that, the specimens are not comparable with one another, let alone with another material.

Conditioning

Standard laboratory atmosphere of Practice D618Testing runs in the same atmosphere.

Check the die before a campaign

Cut and measure a set of blanksDakA worn die raises apparent tear resistance steadily over months. Nothing in the trace reveals it.

The exact die dimensions and tolerances are tabulated in the standard. A die that has been reground has not necessarily been returned to the drawing, and results before and after are not comparable until it has been checked.

Test speed and grip separation

Unlike most tensile methods, this one gives a single rate. It is deliberately slow, so the tear initiates under a controlled stress concentration rather than under an impact.

Rate of grip separation

51 mm/min (2 in./min)

Initial grip separation

25.4 mm (1 in.)

Loading mode

Constant rate of grip separationSpecimen extension is measured as grip separation, not with an extensometer.

Test atmosphere

The conditioning atmosphere

Zero the force with the specimen slack

DakPre-tensioning to take up slack puts load into a specimen whose result is a single peak force.

51 mm/min is far below any rate a film meets in service. That is the design of the method, and it is also why a D1004 figure is a ranking rather than a performance prediction.

Calculations

There is almost nothing to calculate, and the one temptation — dividing by thickness — is the thing the method warns against.

How the test runs

1. 01Cut specimens with the die, taking machine-direction and transverse-direction sets separately.
2. 02Measure the thickness at several points in the notched area of each specimen and record the average; data are comparable only across specimens within ±10 % of the set average.
3. 03Condition the specimens to Practice D618 and keep them in that atmosphere.
4. 04Set the grips to an initial separation of 25.4 mm.
5. 05Mount the specimen square, with the 90° angle correctly oriented and the film free of wrinkles.
6. 06Close the jaws to a clamping force that holds the film without cutting or crushing it.
7. 07Zero the force with the specimen in place and slack.
8. 08Separate the grips at 51 mm/min, recording force against grip separation continuously.
9. 09Take the maximum force and the extension at which it occurred.
10. 10Discard any specimen that breaks at an obvious flaw or in or at the edges of the grips, and test a replacement.
11. 11Run at least ten valid specimens in each direction.
12. 12Report the mean and standard deviation for each direction, never combined.

A specimen that slips at the jaw produces a curve that looks entirely normal and a peak that arrives late. If maximum extension scatters far more than the force does, suspect the grips before the material.

Grips and fixtures for this method

HJ-42

Light Duty Pneumatic Grip

25 mm square flat faces and air-driven closure. Every specimen sees the same clamping force, which matters here because grip separation is also the extension measurement.

Specifications

Rubber facedTJ-34

25mm Square Vice Action Grip

Rubber-coated faces hold thin film and slip-additive film without the serration marks that start a tear at the jaw rather than at the die angle.

Specifications

Self-identifying

Load Cells

A low-range cell. Tearing a thin film is a few newtons of work, and a cell sized for the frame throws away the resolution the peak is read from.

Specifications

What the report has to contain

• Reference to ASTM D1004 and the edition used
• Complete material identification, including thickness and how the film was produced
• Direction tested — machine or transverse — reported separately, never averaged together
• Measured thickness of each specimen, and confirmation that the set falls within ±10 %
• Conditioning procedure, and the temperature and humidity at test
• Rate of grip separation and initial grip separation used
• Maximum tear resistance in newtons for each specimen
• Maximum extension in millimetres
• Number of specimens, the mean and the standard deviation, per direction
• Any specimen discarded, with the reason

What the machine must be capable of

Very little force and a great deal of resolution. Tearing a thin film is a few newtons of work, so the load cell has to be chosen for the specimen rather than for the frame — a cell sized to the machine's headline capacity cannot resolve a peak of this size, and the peak is the entire result. A constant rate of grip separation of 51 mm/min is required, with an initial grip separation of 25.4 mm, so the machine needs stable low-speed control rather than range.

Grips must hold thin, slippery film without cutting it. Rubber-faced or flat-faced vice-action jaws closing on a light, repeatable clamping force are the usual answer; serrations that suit a rigid dumbbell will start a tear at the jaw on film. Because the whole extension is measured as grip separation, any slip is read as extension and quietly corrupts the record of where the peak fell.

What goes wrong in practice

A worn die, which is invisible and systematic. Clamping pressure set by feel, so the film creeps at the jaw and the trace acquires a shoulder that is not the specimen. Averaging machine-direction and transverse specimens into a single figure, which describes no direction the film actually has. Dividing the force by the thickness to make two materials comparable, which the method specifically warns against. And running a set whose thicknesses straddle more than ±10 % of their own average, which makes the specimens incomparable with each other.

Initiation, propagation and the tear methods around them

Two questions are asked about a tear, and they need different specimens. Buying one number when the specification wanted the other is a common and expensive mistake.

An easy-open pack that tears too readily and one that will not start tearing at all are opposite defects. D1004 speaks to the second, D1938 and ISO 6383-1 to the first.

Questions we are asked about this test