Characteristic vs Design Loads: 1.35G + 1.5Q in a Load Takedown

Characteristic loads size the ground, factored loads size the member. One wall taken down both ways: 45 kN/m unfactored, about 57 kN/m at ULS.

---

Characteristic loads are the unfactored loads you expect in service. Design loads are the characteristic loads multiplied by partial factors, typically 1.35 on permanent actions and 1.5 on variable actions, so a member can be checked at the ultimate limit state. In a takedown you carry both: characteristic loads for foundation bearing pressure, design loads for the members.

Source: BS EN 1990 (expressions 6.10, 6.10a, 6.10b, 6.14b) with the UK National Annex (Table NA.A1.2(B), ξ = 0.925). Favourable permanent action is taken as 1.0 and favourable variable action as zero. Values are typical and must be checked by a competent engineer for design.

What is a characteristic load?

A characteristic value is the load you would reasonably expect to see, before any safety factor is added. For a permanent action it is usually the mean value: the unit weight times the volume. For a variable action it is an upper value with a low chance of being exceeded, which is what the imposed and snow load tables in the Eurocodes give you.

The floor in the manual takedown at 5.0 kN/m² permanent plus 2.0 kN/m² imposed is a characteristic load. So is every figure in a bearing pressure check. Nothing has been factored yet.

What is a design load?

A design load is a characteristic load multiplied by a partial factor, so that the member is checked against a deliberately pessimistic version of what may happen. The factors cover variation in the loads, uncertainty in the model, and how bad it is if the element fails.

You write it as Ed, the design effect of actions. It is what goes into the bending, shear, buckling or crushing check. Design resistance (Rd) comes from the material strengths with their own partial factors, so the two sides of the check are both made pessimistic.

What are permanent (G) and variable (Q) actions?

Permanent actions (G) are there all the time and change very little: self-weight of structure, floor finishes, fixed partitions, cladding. They are well known, so the factor is smaller.

Variable actions (Q) come and go: imposed loads from people and furniture, snow, wind. They are harder to predict, so the factor is larger.

That is the whole reason for 1.35 against 1.5. If you can weigh it, it gets the smaller factor. If it depends on how the building is used or what the weather does, it gets the larger one.

Which loads go into which check?

This is where many online answers go wrong. Fixings guidance and some software pages use "design load" to mean a factored value for a bracket or anchor. In a takedown the question is different: which loads feed the ground, and which feed the structure?

• Foundation bearing pressure against a presumed allowable value: characteristic (unfactored) loads. Presumed bearing values already carry a margin against failure, and they are used with service loads.
• Masonry, beams, lintels, slabs, footing reinforcement: ULS design loads, using 6.10 or the 6.10a and 6.10b pair.
• Settlement and deflection: characteristic or quasi-permanent combinations, not ULS.

Eurocode 7 is the exception that catches people out. If you design the ground to a limit state, you use factored actions and factored soil strengths. That is a different method from comparing an unfactored load with a presumed allowable pressure, and the two should not be mixed on one sheet.

The same wall taken down both ways

This is wall W1 from the worked example: a two-storey cavity wall carrying a 3.0 m loaded width of first floor and roof. Characteristic line loads at the base:

Characteristic (SLS): 36.7 + 7.8 = 44.5 kN/m, rounded to 45 kN/m. On a 600 mm strip that is about 75 kN/m² before footing and backfill, and about 95 kN/m² with them. That is the number you compare with a presumed bearing pressure.

ULS, 6.10, imposed floor leading: 1.35 × 36.7 + 1.5 × (6.0 + 0.5 × 1.8) = 49.5 + 10.4 = 59.9 kN/m. The snow is the accompanying action, so it takes ψ0 = 0.5.

If you factor both variable loads at 1.5 with no ψ0, you get 1.35 × 36.7 + 1.5 × 7.8 = 61 kN/m, which is what the earlier post used. It is slightly conservative and perfectly acceptable for a quick check.

ULS, 6.10a: 1.35 × 36.7 + 1.5 × (0.7 × 6.0 + 0.5 × 1.8) = 49.5 + 7.7 = 57.2 kN/m.

ULS, 6.10b, imposed floor leading: 1.25 × 36.7 + 1.5 × (6.0 + 0.5 × 1.8) = 45.9 + 10.4 = 56.2 kN/m.

The design value to carry down is the less favourable of 6.10a and 6.10b, which is 57.2 kN/m. That is about 5% lower than 6.10, and about 28% above the characteristic load. Neither number is "the" load on W1. Each one answers a different question.

Which of 6.10, 6.10a and 6.10b governs?

The UK National Annex lets you use 6.10, or the less favourable of 6.10a and 6.10b. Because that pair always gives a result at or below 6.10, it is the economical route, and 6.10 is the safe fallback if you do not want to run two combinations.

For a single leading variable action and one ψ0, the pair compare like this:

• 6.10a governs when permanent load dominates, because it keeps the full 1.35 on G.
• 6.10b governs when variable load is large relative to permanent, because it applies the full 1.5 to the leading Q.

The crossover for one variable action is where 1.5·Q·(1 − ψ0) equals 0.1·G, roughly Q = 0.22 G with ψ0 = 0.7. W1 sits just below that, at Q/G = 0.21, which is why 6.10a edges it.

Heavy masonry and concrete buildings are permanent-load dominated, so 6.10a usually governs. Light-framed buildings and storage floors tip the other way.

Two cautions. Where permanent action is favourable (uplift, overturning, a counterweight), it takes 1.0 instead, and you check that case separately. And 6.10b needs a leading variable action chosen: you run it once for each candidate and take the worst.

What are ψ factors and when do I need them?

When more than one variable action acts at once, you do not apply the full characteristic value of all of them together. The chance of every one peaking at the same moment is small. The leading action takes its full value. The others are reduced by ψ0.

Source: BS EN 1990 UK National Annex, Table NA.A1.1. Check against the current BSI publication before relying on them.

ψ0 is used in the ULS combinations above. ψ1 and ψ2 are for the frequent and quasi-permanent combinations used at SLS, such as vibration and long-term deflection. For a simple gravity takedown you normally only need ψ0, and only when floor and snow (or wind) are both present.

Where this fits in a takedown

Do the takedown once in characteristic loads, keeping G and Q in separate columns at every level. That gives you the foundation numbers directly. Then apply the factors to each column to get design loads for the members. Never factor as you go, because you can no longer recover the characteristic load without undoing the arithmetic.

If your spreadsheet or software carries a single "total load" per wall, check which one it is. Mixing factored and unfactored figures in one column is the most common takedown error, and the foundation load example shows how it reaches the footing size.

Do the whole building in LoadTakedown: G and Q stay separate at every level, so characteristic and ULS results come from the same model.

FAQ

Do I use factored or unfactored loads for a foundation?

For a presumed allowable bearing pressure, use characteristic (unfactored) loads. The presumed value already includes a margin. Use factored loads for the structural design of the footing itself, such as bending and shear in the base, and for any Eurocode 7 limit state check where soil strengths are factored too.

Why 1.35 for permanent and 1.5 for variable?

Permanent loads can be calculated fairly accurately, so they carry a lower factor. Variable loads depend on use and weather, so the uncertainty is greater and the factor is higher. The values are set by the Eurocodes and confirmed in the UK National Annex for building structures.

Can I just use 1.35G + 1.5Q for everything?

Yes, as a safe simplification. Factoring every permanent load by 1.35 and every variable load by 1.5 is more conservative than 6.10 and is widely used for quick checks. The refinements, ψ0 on accompanying loads and the 6.10a and 6.10b pair, save material but need more combinations. For a small domestic job the saving is often not worth the extra work.

What is the difference between ULS and SLS loads?

ULS loads are factored to check strength and stability, so the structure does not fail. SLS loads are unfactored or lightly factored to check deflection, cracking, vibration and settlement, so the structure remains usable. The same takedown supplies both, with different multipliers applied to the same G and Q.

Does the factor apply to self-weight?

Yes. Self-weight of the structure is a permanent action and takes 1.35 in the unfavourable case. It matters most at the base of a masonry wall, where self-weight is often a third or more of the total. Make sure the wall's own weight is inside the factored figure, not added afterwards at 1.0.