Snow and wind are two separate calculations under two Eurocode parts, EN 1991-1-3 and EN 1991-1-4. A supplier who answers both with one number hasn't done either.
Snow load on a roof is s = μi × Ce × Ct × sk. Four inputs, and only the shape coefficient belongs to the sauna.
A flat or shallow roof takes a shape coefficient of 0.8 up to a 30° pitch. On a cylindrical roof the recommended upper value is 2.0, two and a half times as much.
Wind is decided by terrain, not by country. Roughness length runs from 0.003 m for coast exposed to open sea to 0.3 m for a village or permanent forest.
Ground snow load and basic wind velocity both come from the National Annex of the country the sauna stands in. Neither comes from the factory.
The second-generation snow part, EN 1991-1-3:2025, was published on 26 February 2025, and BSI withdraws the first generation on 30 March 2028.
A manufacturer supplies geometry, weights and anchoring details. The structural check belongs to the customer's engineer.
A sauna that fails in its first winter usually fails because nobody looked up two numbers before the crane arrived. We build in Klaipėda on the Baltic coast, where wind off open water and freeze-thaw run outside the workshop door all winter. By the end you'll know the two loads, the four inputs behind the snow number, and the sheet to send your customer's engineer.
An outdoor sauna should be designed for the snow and wind loads of the plot it stands on, not for a generic climate claim. Snow comes from EN 1991-1-3, wind from EN 1991-1-4. Both start with a site value taken from the National Annex of that country, the characteristic ground snow load and the basic wind velocity. The sauna itself contributes only shape. Roof geometry for snow, exposed elevation and terrain for wind. No single kN/m² figure travels with the product across a border.
So the honest answer to a customer is a question back. Which address, and which annex.
On its own, nothing. It's a sentence, not an input. Ask a supplier instead what snow load their roof is designed for. You'll get one of three answers, and each carries its own verdict.
No number, only a climate claim. That supplier has never been asked by an engineer.
A number with no country and no clause behind it. That's a marketing figure.
A number with the country, the ground snow load and the shape coefficient shown. That's a supplier whose product has been through a real building control file.
EN 1991-1-3, listed by the European Commission as Eurocode 1, Actions on structures, Part 1-3: Snow loads, first published in 2003. Wind is Part 1-4, from 2005. Both appear on the Commission's Eurocode 1 page (accessed July 2026).
The second generation is already here for snow. CEN published EN 1991-1-3:2025 on 26 February 2025, and its scope says the standard “gives principles and rules to determine the values of loads due to snow to be used for structural design of buildings and civil engineering works” (CEN catalogue entry, accessed July 2026). The same entry notes it doesn't apply above 1,500 m altitude unless specified otherwise.
In the UK, BS EN 1991-1-3:2025 supersedes BS EN 1991-1-3:2003+A1:2015, and BSI withdraws the first generation on 30 March 2028 (NBS publication index, accessed July 2026). So as of July 2026 your customer's engineer may work from either edition. Ask which one, and write it on the sheet.
Two loads, snow and wind. Each has its own standard, its own site value and its own list of things that move the answer. Run it before you quote a plot, not after the delivery date is agreed.
| Load 1: snow | Load 2: wind |
|---|---|---|
Eurocode part | EN 1991-1-3 (2003, second generation 2025) | EN 1991-1-4 (2005) |
Site value from the National Annex | Characteristic ground snow load sk | Fundamental basic wind velocity vb,0 |
What that site value means | 0.02 annual probability of exceedance, a 50-year return | 10-minute mean at 10 m over open country, same 0.02 probability |
What shape adds | Shape coefficient μ: 0.8 up to a 30° pitch, recommended upper value 2.0 on a cylindrical roof | Terrain category 0 to IV, roughness length z0 from 0.003 m to 1.0 m |
Correction factors | Exposure Ce (0.8 windswept, 1.0 normal, 1.2 sheltered) and thermal Ct | Orography, height above ground, directional and seasonal factors |
What the manufacturer supplies | Roof geometry, plan area, self-weight, fixing details | Elevation geometry, glazed area, anchoring details |
Now the arithmetic. Say the annex gives a characteristic ground snow load of 2.0 kN/m² for your customer's site. Shallow roof, so the shape coefficient is 0.8. Normal topography, so Ce is 1.0. Cold roof, so Ct is 1.0.
0.8 × 1.0 × 1.0 × 2.0 = 1.6 kN/m². That's about 163 kg on every square metre of roof plan.
Move the same sauna into a wooded hollow and Ce goes to 1.2. The number becomes 1.92 kN/m². Nothing about the sauna changed. The trees did.
Not the way most people assume. Under EN 1991-1-3 a flat or monopitch roof up to 30° carries a shape coefficient of 0.8, applied uniformly. For cylindrical roofs the coefficient depends on the rise-to-span ratio, and the recommended upper value is 2.0.
Put that through the same 2.0 kN/m² ground load. 2.0 × 1.0 × 1.0 × 2.0 = 4.0 kN/m², roughly 408 kg per square metre where the drift sits.
Same plot. Same winter. Two and a half times the load, from roof shape alone.
That's the practical reason our barrel line runs from 2 m to 6 m rather than one universal shell. A 6 m barrel in a snow region and a 2 m barrel on a sheltered terrace aren't the same structural problem, even with identical timber and the same heater. Our cube line has the opposite issue. The flat roof takes the tidier snow coefficient, then the full-height glazed elevation hands the problem to the wind calculation.
Timber choice changes none of this arithmetic, which surprises people. Thermowood earns its place against moisture and movement, and we've written on what thermowood actually does elsewhere. Snow doesn't care. Snow weighs what it weighs.
Wind starts with the fundamental basic wind velocity from the national annex, then gets adjusted for the ground the sauna sits on. EN 1991-1-4 Table 4.1 sets five terrain categories. Category 0 is “Sea or coastal area exposed to the open sea”, with a roughness length z0 of 0.003 m. Category III covers villages, suburban terrain and permanent forest at z0 = 0.3 m.
That's a hundredfold difference in roughness, landing on the same product. A sauna on a dune and one behind a treeline are the same cabin with two different anchoring answers. What moves the number is exposed elevation area, height above ground, local orography and the hold-down detail into the base. Glass changes where the stiffness sits, because a full-height glazed wall doesn't brace a frame the way a closed timber wall does.
Sea. Slope. Trees. Each one moves the number, and none of them is written on the sauna.
We don't do the structural calculation for your architect, and you shouldn't want us to. We're not the engineer of record on your customer's plot, and we don't hold the site survey.
What a manufacturer owes you is the input side. Geometry, roof plan area, self-weight, the anchoring detail, the material specification. That's the half of the file only the factory has. The site half belongs to the engineer.
When a supplier offers a finished structural verdict for any address in Europe, that isn't service. That's a supplier who hasn't read the annex system. The wider version of that test is here: how to tell a reliable sauna manufacturer from a confident one.
Four things, judging by what comes back through partners. The site values used, with the annex they came from. The coefficients and load combination applied. Drawings showing roof geometry and plan area. The anchoring detail, because uplift is where small outdoor buildings actually fail.
Under EN 1990 a small ancillary building normally sits in consequence class CC1, the lowest of the three. That doesn't mean no calculation. It means the partial factors suit a structure whose failure carries low consequence for human life.
The National Annex is where each country's own decisions live. The UK annex to the snow part states plainly that “The recommended value for exposure coefficient Ce in the UK is 1,0 for all topographies” and that “The recommended value for thermal coefficient Ct in the UK is 1,0 for all roofing materials” (UK National Annex to BS EN 1991-1-3:2003, clauses NA.2.15 and NA.2.16, accessed July 2026). Another country sets those differently. That's why one European datasheet can't answer for every plot.
Copy this, send it with every enquiry on a snowy or exposed site, and get it back before you promise a delivery date.
Full plot address and altitude above sea level.
Ground snow load sk for that address, with the annex and clause it came from.
Basic wind velocity vb,0 for that address, same evidence.
Terrain category, 0 to IV, with one sentence on what surrounds the plot.
Nearest taller building or wall and its distance, because drift off a neighbouring roof is a separate load case.
Roof form to be quoted: flat, monopitch or cylindrical, plus the covering.
Base type, and who is setting it, because the anchoring detail depends on it.
Name of the engineer who will sign the check.
Eight lines, none of which needs a lab. All of them are cheaper to collect in an email than to argue about in March.
You don't need to run both checks this week. Do one thing. Take the next enquiry on your desk, look up the ground snow load and the basic wind velocity for that address in the national annex, and write both on the enquiry sheet before you quote. Two numbers, one sheet. That sheet is what the inspector asks your customer for, and it's your name on the roof after the snow lands.
Get an email when Giedrius publishes
More from
Giedrius Patlaba →How Do You Verify Where a Sauna Was Actually Manufactured?
Key Takeaways Three documents carry origin: the origin declaration on the commercial invoice, the transport paperwork showing where the load was collected, and the manufacturer's entry in a public business register. A CE mark is not an origin mark. The European Commission puts it…
What Does an Insurer Ask For Before Covering a Commercial Sauna?
Key Takeaways Insurers judge a commercial sauna on paperwork, not on how it looks. The British Sauna Society operator guide (accessed July 2026) puts it plainly: "They care less about perfection and more about evidence of competence." Six of the documents an insurer wants come…
Will a Standard Sauna Heater Work on a Norwegian 230 V IT Grid?
Key Takeaways A 230 V IT supply carries 230 V between the phases and brings no neutral into the building, so a heater wired for 400 V 3N~ has nothing to connect to. NEK reported in June 2022 that 70–80% of Norwegian buildings are still fed by the old 230 V IT system, and that…