Reading the Data Sheet: What V and W Classes Really Mean

04.08.2026 6 min read

A facade paint’s data sheet reads “V1, W3 to DIN EN 1062-1”. That sounds like a test the product has passed. It isn’t — it’s a classification, and one where the two figures actually count in opposite directions. Anyone who doesn’t know that is reading the data sheet backwards.

The classification standard sets no requirement

DIN EN 1062-1 (2004-08 edition) is a classification standard. It describes a grid into which coating materials for exterior walls can be sorted: gloss, dry film thickness, particle size, water vapour diffusion rate, water permeability, crack-bridging ability, CO₂ permeability.

Nowhere, however, does it say which class is needed on a particular wall. The rating “W3” therefore does not mean “suitable” — only “measured and classified”.

What is actually required is set out in a different standard: DIN 4108-3, current edition 2024-03. Only this standard links a building’s exposure to a numerical value that must be met.

Two classes, two directions

The water vapour diffusion rate V describes how well moisture escapes outward through the coating:

Class	Permeability	Diffusion rate	corresponds to s_d
V1	high	> 150 g/(m²·d)	< 0.14 m
V2	medium	≤ 150, > 15 g/(m²·d)	≥ 0.14, < 1.40 m
V3	low	≤ 15 g/(m²·d)	≥ 1.40 m

Water permeability W describes how much liquid water gets in during driving rain:

Class	Permeability	Water absorption coefficient
W1	high	> 0.5 kg/(m²·h^0.5)
W2	medium	≤ 0.5, > 0.1 kg/(m²·h^0.5)
W3	low	≤ 0.1 kg/(m²·h^0.5)

Both sets of class numbers describe the same thing: the degree of permeability. The difference is that a lot of water vapour permeability is desirable, while a lot of water permeability is not. That creates the trap:

V1 is the best diffusion class — the wall dries out well.
W1 is the worst rain-protection class — the wall soaks up water.

“V1, W1” is therefore not a double top mark, but a coating that lets moisture pass through freely — in both directions.

What DIN 4108-3 actually requires

The standard assigns every building to one of three driving-rain exposure groups, based essentially on annual rainfall: Group I under 600 mm, Group II from 600 to 800 mm, Group III over 800 mm. An overview map using German Weather Service (Deutscher Wetterdienst) data provides the starting value.

Two further rules are regularly overlooked, even though they override the map:

Building geometry raises the group. High-rise buildings and buildings in exposed locations move up one level, regardless of the regional climate.
When in doubt, the higher group applies. The standard states this explicitly: if the group is not clearly evident from the map, the higher value must be assumed.

For Groups II and III, the standard requires a water-repellent exterior render. And “water-repellent” is precisely defined — by three values simultaneously:

Parameter	Limit value
Water absorption coefficient w	≤ 0.5 kg/(m²·h^0.5)
Diffusion-equivalent air layer thickness s_d	≤ 2.0 m
Product w · s_d	≤ 0.2 kg/(m·h^0.5)

The three values are linked — and that changes everything

These limit values are known as the Künzel criteria; they go back to research carried out in 1968 at the Fraunhofer Institute for Building Physics in Holzkirchen. The name only describes their origin: they are not criteria alongside the standard — they are written verbatim into it. Phrases such as “doesn’t meet the standard, but does meet the Künzel criteria” are meaningless.

The decisive factor is the third criterion. The product prevents both individual limit values from being used to the full at the same time:

Anyone using the full w = 0.5 may have at most s_d = 0.4 m — not 2.0 m.
Anyone using the full s_d = 2.0 m may have at most w = 0.1 — not 0.5.

This is why the literature seems contradictory, sometimes stating “s_d ≤ 2 m” and sometimes “s_d ≤ 0.4 m”. Both are correct; they are two points on the same curve.

Behind this lies a piece of building-physics logic worth understanding: a coating with higher water absorption is not automatically bad — as long as the water that gets in can leave again quickly after the rain. What is assessed is the balance between absorption and release, not absorption on its own.

Four things a data sheet doesn’t tell you

The test methods are not the same. DIN 4108-3 requires w to DIN EN ISO 15148 and s_d to DIN EN ISO 12572. Coating data sheets typically state w to DIN EN 1062-3 and s_d to DIN EN ISO 7783. The methods differ in test specimen, substrate and test duration. In practice the values are treated as equivalent — strictly speaking, that isn’t correct.

The s_d value depends on film thickness. It scales proportionally: double the dry film thickness, double the s_d value. The data-sheet figure applies to the test thickness stated there. Anyone applying a thicker coat, or a third coat, leaves the tested condition — and on an existing facade, any earlier coats already present simply add on top.

There are two calculation formulas. DIN EN ISO 7783 contains two ways of calculating the s_d value from the same measurement. A test report from the Institut für Lackprüfung Andreas Keiner documents 0.227 m for the same product using one formula and 0.421 m using the other — a factor of 1.9. For products close to the V1/V2 boundary of 0.14 m, the choice of formula can determine the class.

The criteria apply to the build-up, not the paint. The standard speaks of “renders and coatings” as the rain-protecting layer. What matters is what is actually on the wall: render plus coating plus any earlier coatings. A coating that on its own achieves w ≤ 0.5 says little on its own about the overall build-up.

Watch out for two common claims

“Vapour-open” is not a standardised term. There is no technical definition for it. Some manufacturers call s_d < 1 m vapour-open, others s_d < 2 m. Only the V class, or an s_d figure stated together with the test film thickness, is reliable. The same applies to “breathable”.

Very small s_d values are noise. Figures around 0.01 m are within the order of magnitude of measurement error. Anyone claiming superiority based on s_d = 0.01 rather than 0.02 m is arguing within the measurement uncertainty.

And a warning about a commonly cited source: the freely accessible reproduction of DIN 4108-3 on umwelt-online.de states “< 2.0” instead of the correct ≤ 0.2 for the product w · s_d — apparently a slip from the neighbouring column. The error is a factor of 10 and makes any assessment based on it worthless. The page is frequently linked in forums.

Sources

The class tables for V and W are taken from a test report by Institut für Lackprüfung Andreas Keiner GmbH, which reprints them verbatim as the basis for classification. The figures on exposure groups and on Table 6 come from the full text of DIN 4108-3:2018-10. The current edition is 2024-03; its published amendment notes concern the periodic balance method and a new section on ground slabs, not the driving-rain chapter. A change to Table 6 is therefore unlikely, but has not been checked against the original text of the 2024-03 edition.

The gloss and CO₂-permeability classes are not covered here: for both, reliable source confirmation of the numerical limits could not be obtained. A gap is preferable to a plausible-sounding figure.