EN 1998 Seismic Design — Steel Structures per Eurocode 8

Complete reference for seismic design of structural steel buildings per EN 1998-1 (Eurocode 8: Design of structures for earthquake resistance — Part 1: General rules, seismic actions and rules for buildings). Ground types A through E with spectral parameters, behaviour q-factors for moment-resisting frames (MRF), concentrically braced frames (CBF), and eccentrically braced frames (EBF), ductility classes DCL (low), DCM (medium), and DCH (high), capacity design principles for dissipative steel structures, and a worked design example for a 6-storey steel CBF office building.

Quick access: EN 1993 Steel Design → | EN 1993 Column Buckling → | EN 1990 Load Combinations → | Seismic Load Calculator →

EN 1998-1 Seismic Design Philosophy

EN 1998-1 adopts a performance-based seismic design philosophy with two fundamental requirements:

  1. No-collapse requirement (Ultimate Limit State): The structure shall be designed to sustain the design seismic action without local or global collapse, retaining its structural integrity and a residual load-bearing capacity after the seismic event (EN 1998-1 Cl. 2.1(1)P).
  2. Damage limitation requirement (Serviceability Limit State): The structure shall withstand a seismic action having a larger probability of occurrence than the design seismic action, without damage and associated limitations of use, the costs of which would be disproportionately high compared with the costs of the structure itself.

For ordinary buildings (Importance Class II), the reference return period for the no-collapse requirement is 475 years (10% probability of exceedance in 50 years), and for the damage limitation requirement 95 years (10% in 10 years). EN 1998-1 Annex A provides maps and seismic zonation for each country through the National Annex.

Importance Classes and Importance Factors γI (EN 1998-1 Cl. 4.2.5)

Importance Class Building Type γI (Recommended)
I Minor buildings not for human occupancy except access (agricultural sheds, temporary structures) 0.80
II Ordinary buildings — residential, commercial, office (default class) 1.00
III Buildings important for civil protection — schools, assembly halls, cultural institutions 1.20
IV Buildings whose integrity during earthquakes is vital — hospitals, fire stations, power plants 1.40

The design ground acceleration ag = γI × agR, where agR is the reference peak ground acceleration on Type A ground at the site. Importance Class III and IV buildings effectively have 20% and 40% higher design seismic forces respectively.

Ground Types and Elastic Response Spectrum

EN 1998-1 Cl. 3.1.2 defines five ground types (A-E) plus two special site categories (S1, S2) based on stratigraphic profile and shear wave velocity:

Ground Type Stratigraphic Description vs,30 (m/s) NSPT (blows/300 mm) cu (kPa)
A Rock or other rock-like geological formation, including at most 5 m of weaker material at the surface > 800 — —
B Deposits of very dense sand, gravel, or very stiff clay, several tens of metres thick — gradual increase of strength with depth 360-800 > 50 > 250
C Deep deposits of medium-dense sand, gravel, or stiff clay of thickness from several tens to many hundreds of metres 180-360 15-50 70-250
D Deposits of loose-to-medium cohesionless soil (with or without soft cohesive layers) or predominantly soft-to-firm cohesive soil < 180 < 15 < 70
E Soil profile consisting of surface alluvium layer (vs < 360 m/s, thickness 5-20 m) over stiffer material (vs > 800 m/s) — — —
S1 Deposits containing a layer ≥ 10 m thick of soft clays/silts with high plasticity index (PI > 40) and high water content (vs < 100 m/s) < 100 — 10-20
S2 Deposits of liquefiable soils, sensitive clays, or any soil profile not included in Types A-E or S1 (organic soil, peat, > 30 m soft soils) — — —

Ground Types S1 and S2 require special studies and site-specific seismic hazard analysis. The standard response spectrum should not be applied without verification.

Spectral Parameters (EN 1998-1 Table 3.2 — Type 1, Ms > 5.5)

Ground Type S (Soil Factor) TB (s) TC (s) TD (s)
A 1.00 0.15 0.40 2.00
B 1.20 0.15 0.50 2.00
C 1.15 0.20 0.60 2.00
D 1.35 0.20 0.80 2.00
E 1.40 0.15 0.50 2.00

Note: Ground Type C actually reduces the spectral amplification (S = 1.15) compared with Type B (S = 1.20) because deeper, softer deposits (Type C, hundreds of metres thick) do not amplify short-period motion as much as a shallow stiff layer over rock (Type B). The UK NA adopts Type 2 spectrum for low-seismicity areas (M ≤ 5.5), which has shorter corner periods.

Type 1 Elastic Response Spectrum (EN 1998-1 Cl. 3.2.2.2):

For 0 ≤ T ≤ TB: Se(T) = ag × S × [1 + T/TB × (η × 2.5 − 1)]

For TB ≤ T ≤ TC: Se(T) = ag × S × η × 2.5

For TC ≤ T ≤ TD: Se(T) = ag × S × η × 2.5 × [TC / T]

For TD ≤ T ≤ 4.0s: Se(T) = ag × S × η × 2.5 × [TC × TD / T²]

Where ag = γI × agR (design ground acceleration on Type A), S = soil factor, η = damping correction factor = √[10/(5+ξ)] ≥ 0.55 (for 5% viscous damping typical in steel structures, ξ = 5 → η = 1.0), and TB, TC, TD = corner periods defining the spectral shape.

Behaviour Factors q for Steel Structures (EN 1998-1 Cl. 6.3)

The behaviour factor q is the central parameter of EN 1998 design. It reduces the elastic seismic force demand to a design-level force, reflecting the structure's ductility and hysteretic energy dissipation capacity. Higher q means lower design forces but more stringent ductility detailing.

q = q0 × kw ≥ 1.5

Where q0 is the basic behaviour factor (depends on structural system and ductility class) and kw is the prevailing failure mode factor (kw = 1.0 for frame or braced frame systems; kw = 0.5 to 1.0 for inverted pendulum systems).

Behaviour Factors — Moment-Resisting Frames (MRF)

Ductility Class q0 Possible Multiplier Maximum q Application
DCL ≤ 1.5 None 1.5 Non-dissipative — elastic design, no capacity design required
DCM 4.0 αu/α1 4.0 × 1.3 = 5.2 Standard mid-rise frame — capacity design mandatory
DCH 5.0 αu/α1 5.0 × 1.3 = 6.5 High-seismicity — stringent capacity design + material requirements

The multiplier αu/α1 (EN 1998-1 Cl. 6.3.1(3)) is the ratio of base shear at global mechanism to base shear at first plastic hinge. Conservative default values: single-storey single-bay = 1.0, multi-storey single-bay = 1.1, multi-storey multi-bay (regular) = 1.2, multi-storey multi-bay (irregular) = 1.3.

Behaviour Factors — Steel Braced Frames

Structural Type Ductility Class q0 Max q Key Limitation
CBF — Diagonal bracing DCM 3.0 — Brace buckling in compression — tension-only design not permitted
CBF — Diagonal bracing DCH 4.0 4.8 (×αu/α1) Class 1 sections for dissipative braces
CBF — V-bracing (chevron) DCM 2.0 2.5 Reduced q — beam must resist full unbalanced brace force
CBF — V-bracing (chevron) DCH 2.5 3.0 Beam must be designed for (Npl,Rd,tension + 0.3×Npl,Rd,comp) × sinθ
EBF (Eccentrically Braced) DCM 4.0 5.0 (×αu/α1) Links dissipate energy — short links in shear, long links in bending
EBF (Eccentrically Braced) DCH 5.0 6.5 (×αu/α1) Compact links with web stiffeners mandatory
Inverted Pendulum DCM/DCH 2.0 × kw 2.0 Cantilever structures — limited energy dissipation

Critical design note on V-bracing: The reduced q-factor for V-braces is because the beam must resist the full unbalanced vertical force when the compression brace buckles. The tension brace yields, delivering Npl,Rd upward while the compression brace delivers only its post-buckling capacity (≈0.3 × Npl,Rd) downward. This net upward force on the beam is enormous and the beam is non-dissipative (must remain elastic). This limitation is widely misunderstood and a common source of non-conservative designs.

Ductility Classes — Steel Material Requirements (EN 1998-1 Cl. 6.2)

DCL — Ductility Class Low

DCM — Ductility Class Medium

Steel requirements for dissipative structural elements (beams in MRF, braces in CBF):

Property Requirement Purpose
fy,max / fy ≤ 1.25 Ensures plastic hinges form at intended locations — avoids premature yielding in adjacent members
fu / fy ≥ 1.10 Provides strain-hardening capacity beyond yield — essential for redistribution
εu (elongation at failure) ≥ 15% Adequate deformation capacity for cyclic plastic straining
Charpy for welded dissipative zones 27 J at service temp Fracture resistance in cyclically loaded weldments
Section class for dissipative elements Class 1 or 2 Plastic hinge or plastic resistance development

DCH — Ductility Class High

More stringent requirements for the highest energy dissipation:

Property DCM DCH Additional
εu ≥ 15% ≥ 20% for open sections, ≥ 15% for hollow sections
Charpy for welded zones 27 J at Tservice 27 J at Tservice − 10°C (more demanding for cold climates)
Section width-to-thickness limits Class 1 or 2 Class 1 mandatory for plastic hinge zones in MRF
Beam-to-column connection certification By calculation Qualification testing in some National Annexes

Capacity Design Principles for Steel Frames

Capacity design is the foundational concept of EN 1998 dissipative design. The principle:

Dissipative elements yield and absorb seismic energy. Non-dissipative elements (columns, connections, foundations) remain elastic under the maximum credible forces the dissipative elements can deliver.

Capacity Design for MRF (EN 1998-1 Cl. 6.6)

Step 1 — Weak beam / strong column:

At every beam-column joint:

ΣMRc ≥ 1.3 × ΣMRb

Where ΣMRc = sum of column plastic moment resistances at the joint and ΣMRb = sum of beam plastic moment resistances. The 1.3 factor provides an overstrength margin against column hinging.

Step 2 — Column design moment from capacity design:

The column is designed for:

MEd,column = MEd,G + 1.1 × γov × Ω × MEd,E

Where MEd,G is the moment from the gravity load combination, MEd,E is the seismic analysis moment, γov = 1.25 (material overstrength factor), and Ω = minimum(Mpl,Rd,i / MEd,i) across all beam plastic hinge locations in the frame.

Step 3 — Panel zone design: The beam-to-column panel zone (column web between beam flanges) must resist the shear from plastic hinge moments: Vwp,Ed = ΣMpl,Rd,beam / (db − tbf). This often governs the column web thickness or requires doubler plates.

Capacity Design for CBF (EN 1998-1 Cl. 6.7)

Step 1 — Brace design: Diagonal braces are sized for the seismic action Sd(T) = Se(T) / q. Braces yield in tension and may buckle in compression — post-buckling resistance ≥ 0.3 × Npl,Rd is required.

Step 2 — Beam and column design from overstrength:

NEd,column = NEd,G + 1.1 × γov × Ω × NEd,E

Where Ω is the minimum overstrength ratio across all braces: Ω = min(Npl,Rd,i / NEd,E,i). γov = 1.25 (recommended for steel).

Step 3 — Brace connections: Designed for the maximum force the brace can deliver: Nconnection ≥ 1.1 × γov × Npl,Rd,brace. This drives substantial gusset plate thicknesses and bolt counts.

Eurocode 8 Seismic Analysis Methods

Lateral Force Method (EN 1998-1 Cl. 4.3.3.2)

Permitted for buildings meeting vertical and plan regularity criteria and with T1 ≤ min(4TC, 2.0s).

Fundamental period: T1 = Ct × H^(3/4)

Base shear: Fb = Sd(T1) × m × λ, where m = total seismic mass (Gk + ΣψE,i × Qk,i) and λ = 0.85 for T1 ≤ 2TC and number of storeys ≥ 2, otherwise λ = 1.0.

Vertical force distribution: Fi = Fb × (zi × mi) / Σ(zj × mj) — inverted triangular, applying 85-100% of total mass at the effective height.

Pushover Analysis (EN 1998-1 Annex B)

For existing structures and irregular new structures, nonlinear static (pushover) analysis is recommended. Two lateral load distributions are required: uniform (acceleration proportional to mass) and modal (proportional to first mode shape). The capacity curve (base shear vs roof displacement) is bilinearised and compared with the demand spectrum in ADRS (Acceleration-Displacement Response Spectrum) format using the N2 method or Coefficient Method.

Worked Example — 6-Storey Steel CBF Office Building

Building description: 6 storeys, h = 3.5 m per storey (total H = 21 m), plan 30 m × 20 m, braced bays on perimeter. Site: agR = 0.25g (moderate European seismicity — e.g., southern Italy, Greece, Romania). Ground Type C, γI = 1.0, DCM CBF with diagonal bracing (q = 3.0).

Step 1 — Seismic Mass

mstorey = (Gk + ψE,i × Qk) per floor. Dead = 5.5 kN/m², Imposed = 3.0 kN/m² (Category B office), ψ2 = 0.3. Floor area = 600 m².

Seismic mass per floor: mstorey = 600 × (5.5 + 0.3 × 3.0) / 9.81 = 391 tonnes. Roof: mroof = 600 × (5.0 + 0.3 × 1.5) / 9.81 = 333 tonnes. Total M = 5 × 391 + 333 = 2,288 tonnes.

Step 2 — Fundamental Period and Spectrum

T1 = 0.075 × 21^0.75 = 0.075 × 10.05 = 0.75 s.

Elastic spectrum: ag = 1.0 × 0.25g = 2.45 m/s². Type C: S = 1.15, TC = 0.6 s.

Since TB < T1 < TC: Se(T1) = 2.45 × 1.15 × 1.0 × 2.5 = 7.04 m/s².

Design spectrum: Sd(T1) = 7.04 / 3.0 = 2.35 m/s².

Step 3 — Base Shear

Fb = Sd(T1) × M × λ = 2.35 × 2,288 × 0.85 = 4,570 kN. (Approximately 20% of seismic weight — reasonable for DCM CBF in moderate seismicity).

Step 4 — Brace Design and Capacity Design Overstrength

With 4 braced bays, 2 braces per bay = 8 braces at ground floor. Brace angle ≈ 45°. Seismic force per brace: NEd,E = 4,570 / (8 × cos45°) = 4,570 / 5.657 = 808 kN.

Select SHS 200 × 200 × 10.0 in S355J2: A = 7,410 mm², Npl,Rd = 7,410 × 355 / 1.0 = 2,630 kN. Buckling check (Lcr = 5.7 m, i = 76.9 mm, λbar ≈ 0.75) gives χ ≈ 0.72 → Nb,Rd = 0.72 × 2,630 = 1,894 kN > 808 kN. OK.

Ω = Npl,Rd / NEd,E = 2,630 / 808 = 3.25.

Step 5 — Column Design (Capacity Design)

Gravity load at ground column: NEd,G ≈ 6 × 600 × 5.5 × (tributary fraction) ≈ 7,920 kN (per column, simplified). Seismic component: NEd,E,column ≈ 808 × cos45° × 2 braces per column ≈ 1,143 kN.

Capacity design axial force: NEd,col = 7,920 + 1.1 × 1.25 × 3.25 × 1,143 = 7,920 + 5,105 = 13,025 kN.

This extreme force would require a very heavy column. In practice, the designer iterates by reducing the brace sections to bring Ω closer to 1.5-2.0, which reduces the capacity-design force proportionally. This is the fundamental economic driver in CBF design: select the smallest possible brace that satisfies the seismic demand to minimise Ω and the capacity-design forces in the non-dissipative elements.

Frequently Asked Questions

What is the difference between DCL, DCM, and DCH in steel seismic design?

DCL (Ductility Class Low) is non-dissipative with q ≤ 1.5 — the structure is designed for essentially elastic seismic forces with no special ductility provisions. DCM (Medium) uses q up to 4.0 (MRF) or 3.0 (CBF), requires capacity design with γov = 1.25, and imposes material requirements including fy,max/fy ≤ 1.25 and fu/fy ≥ 1.10. DCH (High) uses q up to 6.5 (MRF) or 4.8 (CBF), with more stringent material requirements (εu ≥ 20% for open sections), Class 1 sections mandatory, and typically requires Charpy at Tservice − 10°C. Most European buildings are DCM; DCH is reserved for high-seismicity regions where the economic benefit of reduced forces outweighs the additional fabrication and material costs.

When can I use q = 1.5 (non-dissipative design) instead of a higher q-factor?

Use q ≤ 1.5 in three circumstances: (1) Low seismicity regions where agR × S < 0.10g — seismic forces are modest and the fabrication cost of ductility detailing exceeds material savings from reduced sections. (2) Highly irregular structures (plan or vertical irregularity exceeding EN 1998-1 Cl. 4.2.3 limits) where dissipative behaviour cannot be reliably predicted; EN 1998-1 requires q to be reduced by 20% for irregular structures, and DCL is a safe default for extreme irregularity. (3) Client or jurisdiction specification for elastic design of critical infrastructure (hospitals, nuclear, defence). The penalty: base shear is typically 2.5-4× higher than DCM, driving substantially larger sections.

How do I apply capacity design to an eccentrically braced frame (EBF)?

In an EBF, the seismic links are the dissipative elements. Short links (e < 1.6 × Mpl,link / Vpl,link) dissipate energy through shear yielding; long links through bending. The link is designed for Sd(T) forces. Braces, columns, and beam segments outside the link are non-dissipative and designed for: NEd,brace = 1.1 × γov × Ω × NEd,E, where Ω is calculated at the link. The link resistance Vpl,link (short link) or Mpl,link × 2/e (long link) determines Ω. Link web stiffeners are mandatory — spacing ≤ 30tw − d/5 for short links, ≤ 30tw − d/5 at link ends, ≤ 52tw − d/5 at link centre for long links.

What magnification factor do I apply for P-Delta effects in a steel MRF?

The inter-storey drift sensitivity coefficient θ = Ptot × dr / (Vtot × h) determines P-Delta treatment (EN 1998-1 Cl. 4.4.2.2). If θ ≤ 0.1: negligible, may be ignored. If 0.1 < θ ≤ 0.2: multiply seismic effects by 1/(1-θ) for all storeys. If θ > 0.2: full second-order analysis required (P-Δ geometric stiffness in the analysis model). θ > 0.3 is not permitted — the structure must be stiffened. For a typical 6-8 storey steel MRF, θ is typically 0.06-0.12, so the amplification factor is modest (1.06-1.14).

How does EN 1998 handle seismic design of steel connections?

Connection design follows capacity design. In dissipative zones, connection resistance must exceed the maximum force the connected dissipative element can deliver including strain hardening: Rd,connection ≥ 1.1 × γov × Rpl,Rd,member. For MRF beam-to-column connections in DCH, this means the welded beam-flange-to-column connection must resist 1.1 × 1.25 × Mpl,Rd,beam (≥ 1.375 Mpl,Rd). For CBF brace-to-gusset connections, the connection must resist 1.1 × 1.25 × Npl,Rd,brace. Bolted connections in dissipative zones must be preloaded (Category B or C slip-resistant to EN 1993-1-8) to prevent bolt slip and ensure energy dissipation through member yielding rather than friction sliding. Full-penetration butt welds in dissipative zones require Charpy certification.

Related Pages

Educational reference only. Seismic design parameters (agR, ground type, q-factor, importance class) must be verified against the current National Annex for the building jurisdiction. Seismic design is a specialised discipline — all designs must be independently verified by a qualified structural engineer with seismic experience. Results are PRELIMINARY — NOT FOR CONSTRUCTION without professional structural engineering review.