EN 12004-2 Ceramic Tile Adhesive Testing

What the test does

Part 2 carries the test methods for the adhesive classes Part 1 defines, and tensile adhesion strength is the central one. Tiles are bedded into adhesive on a concrete slab under a defined load, cured under a specified regime, and a pull-head is bonded to each tile. Each tile is then pulled perpendicular to the slab until the joint fails, and the force divided by the bonded area is the adhesion strength. Shear adhesion, open time and slip are measured by their own procedures.

What it measures, and why it matters

The primary figure is tensile adhesion strength in newtons per square millimetre, measured after several conditioning regimes: standard air storage, water immersion, heat ageing and freeze-thaw cycling. The set of them is the result rather than any single value, because an adhesive that performs well dry and poorly after immersion is unsuitable for a wet area whatever its dry figure. Open time — how long the adhesive stays workable — and slip decide whether it can be used on a wall at all.

The classification system in Part 1 turns these measurements into the letters a specifier actually writes — C1 or C2 for cementitious adhesives, with additional characteristics for rapid setting, reduced slip and extended open time. Each of those is a threshold on a specific measurement from this part, so an adhesive is not C2 in general; it is C2 because named figures from named conditioning regimes cleared named limits.

Slab, tiles and conditioning

Substrate

Concrete slab to the standard's own specificationPrescriptive, because the substrate moves the result more than most adhesive differences do.

Bedding

Tiles set into the adhesive under a defined load

Conditioning regimes

Air storage, water immersion, heat ageing, freeze-thawThe SET is the result. Reporting only the dry figure is the commonest way a specification gets satisfied on paper and not in the bathroom.

Pull-heads

Bonded square to the tile faceWith an adhesive stronger than the mortar under test.

Open time and slip

Measured by their own proceduresThey decide whether the adhesive can be used on a wall at all, which no strength figure addresses.

Loading

The tensile adhesion test is force-controlled rather than crosshead-controlled, so the figure to set is a force rate. The shear procedures in the same part are the other way round.

Direction

Perpendicular to the slab

Tensile adhesion rate

250 ± 50 N/sApplied until the joint fails. The rate used is reported with the result.

As a stress rate

0,1 N/mm² per secondDak250 N/s over the 2 500 mm² bonded area the method's 50 mm square tile and 50 mm square pull-head give. An adhesive at the 1,0 N/mm² class threshold therefore fails about ten seconds into the pull. Recompute it if a tile of another size is used.

Shear adhesion rate

Crosshead at 5 ± 0,5 mm/minDisplacement-controlled, not force-controlled. The two procedures in this part are not run the same way.

Many tiles per slab

UsualWhich is why a positioning arrangement reaching each tile in turn is more practical than moving a heavy slab under a fixed head.

What the machine must be capable of

A perpendicular pull with no bending, at a force rate of 250 ± 50 N/s, on a tile that may be one of many on a slab — which is why a positioning arrangement covering the whole slab is usual rather than moving the slab under a fixed head. The tile and the metallic pull-head are both 50 mm square, so the bonded area is 2 500 mm² and 250 N/s is a stress rate of 0,1 N/mm² per second: an adhesive at the 1,0 N/mm² threshold fails about ten seconds into the pull. The shear procedures in the same part are displacement-controlled instead, at 5 ± 0,5 mm/min. Forces are modest and the requirement is repeatability and squareness rather than capacity.

Force capacity is not the constraint — a 50 mm tile at a few newtons per square millimetre needs only a few kilonewtons — but travel and clearance to reach every tile on the slab are. Conditioning capability is as much a part of the method as the frame: water immersion, heat ageing and freeze-thaw cycling all have to be delivered to schedule or the results are not the ones the classification is written against.

How the test runs

1. 01Prepare concrete slabs to the standard's specification.
2. 02Mix the adhesive to the manufacturer's instructions and apply it.
3. 03Bed the tiles under the defined load.
4. 04Cure and then condition each set — air, water immersion, heat ageing, freeze-thaw.
5. 05Bond pull-heads square to each tile face with a stronger adhesive.
6. 06Load perpendicular to the slab at 250 ± 50 N/s.
7. 07Record the failure force and the bonded area for each tile.
8. 08Classify the failure — adhesive at the tile, adhesive at the substrate, or cohesive in the mortar.
9. 09Report each conditioning regime separately, never averaged together.

Watch the test

Calculations

What the report has to contain

• Reference to EN 12004-2
• Adhesive identification, class claimed and mixing instructions followed
• Substrate preparation and slab specification
• Tile type and bedding load
• CONDITIONING REGIME for every result
• Rate of loading
• Tensile adhesion strength in N/mm² per regime
• Failure mode for every tile
• Open time and slip results where measured
• Number of tiles per regime

What goes wrong in practice

Failing to score or otherwise define the bonded area lets the surrounding adhesive share load. A pull-head bonded out of square peels one edge first. Substrate preparation that drifts from the standard's procedure changes the answer more than the adhesive formulation does. And quoting a dry-storage figure against a specification that requires the water-immersion value is a common and consequential substitution.

Slabs that vary in porosity between batches introduce scatter that is easily read as adhesive variability. Pull-heads bonded with an adhesive that has not fully cured fail at the head rather than at the mortar. And tiles at the edge of a slab behave differently from those in the middle, because the bed dries at a different rate — which is why placement across the slab is part of the procedure rather than left to convenience.