-------------------- | -------------------- | ------------------------ | --------------------- | ------------------- | | Design method | LRFD (primary) / ASD | Limit State Design | Limit State Design | Limit State Design | | Load standard | ASCE 7-22 | AS/NZS 1170 | EN 1990/1991 | NBC 2020 | | Resistance factor approach | phi on resistance | phi on resistance | gamma_M on resistance | phi on resistance | | Load factors (gravity) | 1.2D + 1.6L | 1.2D + 1.5L | 1.35D + 1.5L | 1.25D + 1.5L | | Wind/seismic combination | 1.2D + 1.0W + 0.5L | 1.2D + 1.0W + 0.4L | 1.0D + 1.5W | 1.0D + 1.4W | | Units | kip, in., ksi | kN, mm, MPa | kN, mm, MPa | kN, mm, MPa |

Resistance factors (phi / gamma)

The Eurocode uses gamma_M = 1.00 for member resistance (flexure, compression) but gamma_M2 = 1.25 for connection resistance (bolts, welds, net section). This makes Eurocode members appear to have higher capacity, but the higher load factors compensate, producing similar overall safety levels.

Column buckling comparison

The codes use different mathematical models for the column curve:

• AISC: Single curve based on SSRC Curve 2P. Transition at KL/r = 4.71 x sqrt(E/F_y). Simple to apply, conservative for some section types.
• AS 4100: Modified Perry-Robertson with alpha_b section constant (-1.0 to +0.5). Multiple curves depending on section type and fabrication method.
• EN 1993: Five explicit curves (a0, a, b, c, d) with imperfection factors alpha = 0.13 to 0.76. The curve selection depends on section type, axis of buckling, and fabrication (hot-rolled vs. welded).
• CSA S16: Single curve very similar to AISC (both based on SSRC research).

For a W10x49 column with KL/r = 70 (A992/Grade 350/S355), the design capacities are: AISC = 441 kip, AS 4100 = 435 kip, EN 1993 (curve b) = 446 kip, CSA S16 = 438 kip. The variation is less than 3% for this standard case.

Beam flexural capacity comparison

Note the Eurocode uses E = 210 GPa while the other three codes use E = 200 GPa. This 5% difference affects all stiffness-related calculations (deflection, buckling, LTB transition lengths).

Worked example — W14x48 beam, L = 20 ft, uniform load

Comparing the four codes for the same physical beam (equivalent sections used for AS/EN/CSA):

The Eurocode gives slightly higher values due to the higher E and gamma_M1 = 1.00. However, when combined with the higher Eurocode load factors (1.35D vs. 1.2D for AISC), the required member sizes are very similar across all four codes.

Key differences to watch

• Bolt hole deductions — AISC deducts 1/16 in. beyond the nominal hole diameter for net area calculations (effective hole = nominal + 1/16). AS 4100 uses the nominal hole diameter directly. EN 1993 uses d_0 = d_hole (nominal). This difference can affect net section rupture calculations by 5-10%.
• Weld directional strength — AISC allows a 50% increase in fillet weld strength for transversely loaded welds (1.0 + 0.50 x sin^1.5(theta)). AS 4100 does not allow this directional increase. EN 1993 uses a different directional formula. Using the AISC directional enhancement in an AS 4100 design is unconservative.
• Second-order analysis — AISC Chapter C provides the Direct Analysis Method (DAM) with notional loads and stiffness reductions, allowing K = 1.0. AS 4100 and EN 1993 use amplified first-order analysis with effective length factors. The methods give similar results but require different analysis model setups.
• Seismic provisions — AISC 341 provisions do not apply outside the U.S. Each code jurisdiction has its own seismic standard (AS 1170.4, EN 1998, NBC/CSA). The capacity design principles are similar but the detailing requirements, R-factors, and ductility classes differ significantly.

Common mistakes to avoid

• Mixing code provisions — using AISC bolt capacity tables with Eurocode load combinations (or vice versa) invalidates the calibrated safety level. Every element of the design — loads, resistance, detailing — must come from the same code framework.
• Using the wrong E value — AISC uses 29,000 ksi (200 GPa), Eurocode uses 210 GPa. A 5% error in E shifts all buckling calculations, deflection checks, and LTB transition lengths. Always verify which E the code specifies.
• Assuming phi factors are interchangeable — AISC uses phi = 0.75 for bolt shear while AS 4100 uses phi = 0.80. These differences are calibrated to their respective load factor systems and are not interchangeable.
• Ignoring unit systems — AISC is in kip-in-ksi, while the other three codes use kN-mm-MPa. A factor-of-25.4 error in converting inches to millimeters propagates through every calculation. Always convert at the start and verify units at each step.
• Overlooking material grade naming differences — A992 (AISC, Fy = 50 ksi) is equivalent to Grade 350 (AS, fy = 350 MPa) and S355 (EN, fy = 355 MPa). Using the wrong yield strength because of grade naming confusion introduces a 1-3% error that propagates through every capacity calculation. Always verify the actual fy value in the material standard, not just the grade name.

Connection design methodology comparison

The four codes handle connection design with different levels of explicit guidance. AISC provides the most comprehensive connection design provisions with extensive design guides (DG4 for end plates, DG13 for stiffened seats, DG16 for multiple-row end plates). EN 1993-1-8 uses the T-stub analogy for end plate connections, which is an elegant analytical framework that identifies three distinct failure modes. AS 4100 Section 9 provides compact provisions but relies on supplemental references (Hogan, Murray) for detailed connection design. CSA S16 references the CISC Handbook of Steel Construction for worked connection design examples.

A key difference: AISC allows bearing at deformations of 0.34 in. (using 2.4dtFu) or at larger deformations (using 1.5dtFu with a lower phi). EN 1993 and AS 4100 do not make this distinction explicitly but achieve similar results through their respective gamma_M or phi factors. The Eurocode T-stub method (EN 1993-1-8 Section 6.2.4) classifies connection failure into three modes: Mode 1 (complete plate yielding with full prying), Mode 2 (combined bolt failure with plate yielding), and Mode 3 (bolt failure without prying). This systematic approach provides clearer insight into the failure mechanism than the AISC prying formulas alone.

Detailed design procedure comparison

Beam design procedure differences

The four codes follow the same general philosophy for beam design — determine plastic or elastic moment capacity, reduce for lateral-torsional buckling if the beam is laterally unbraced, and check local buckling — but the specific calculations differ.

AISC 360-22 Chapter F:

1. Determine Mu from structural analysis (factored loads)
2. Classify section as compact, noncompact, or slender (Table B4.1b)
3. Calculate Mn based on three zones: Mp (compact, Lb <= Lp), inelastic LTB (Lp < Lb <= Lr), elastic LTB (Lb > Lr)
4. Apply moment gradient factor Cb to inelastic and elastic zones
5. Design capacity = phi × Mn (phi = 0.90)

EN 1993-1-1 Section 6.3.2:

1. Determine MEd from analysis with partial factors on loads
2. Classify section as Class 1-4 (Table 5.2, based on flange and web slenderness)
3. Calculate Mc,Rd = fy × Wpl (Class 1-2) or fy × Wel (Class 3)
4. Reduce for LTB: Mb,Rd = chi_LT × Wy × fy / gamma_M1
5. chi_LT from buckling curves (a, b, c, d) based on section type
6. Verify Mc,Rd >= MEd and Mb,Rd >= MEd

AS 4100-2020 Section 5:

1. Determine M* from analysis (factored loads)
2. Check section slenderness: compact, noncompact, slender
3. Calculate Ms = fy × Zs (effective section modulus)
4. Reduce for LTB: phi × Mob = phi × alpham × alphac × Ms (phi = 0.90)
5. alpham = moment modification factor (Table 5.6.1 or formula)
6. alphac = slenderness reduction factor from Figure 5.6.3
7. Verify phi × Mob >= M*

CSA S16:24 Clause 13.6:

1. Determine Mf from analysis (factored loads)
2. Check section class (Class 1-4)
3. Calculate Mr = phi × Mp = phi × Fy × Zx (Class 1-2, phi = 0.90)
4. Reduce for LTB using linear interpolation between Mp and elastic buckling
5. Apply omega_2 moment gradient factor
6. Verify Mr >= Mf

Column design approach differences

The Eurocode's five-column-curve system provides the most refined approach, with different curves for hot-rolled H-sections about each axis, welded sections, hollow sections, and angles. AISC and CSA use a single curve, which is slightly conservative for some section types and slightly unconservative for others. AS 4100 offers intermediate refinement with five alpha_b values.

Which code to use where

Engineers working internationally must verify which code is required by the local building authority. Many countries accept multiple codes but may require local adaptations (wind speed maps, seismic hazard maps, material grades). Software like the Steel Calculator supports multiple codes to facilitate international design verification.

Run this calculation

• Section Properties Database
• Beam Calculator

Try it now: Check your code comparison with our free Steel Beam Capacity calculator →

Related references

• How to Verify Calculations
• Bolted Connections
• Beam Design Guide
• load combinations guide
• Beam Capacity Design Tool
• structural engineering unit converter
• Steel Seismic Design

Disclaimer

This page is for educational and reference use only. It does not constitute professional engineering advice. All design values must be verified against the applicable standard and project specification before use. The site operator disclaims liability for any loss arising from the use of this information.

Frequently Asked Questions

Which code gives the most conservative design for a typical steel beam?

For simply supported beams under gravity load, the four codes produce member sizes within one W-shape depth increment of each other. EN 1993 tends to give slightly higher capacity for flexure (gamma_M1 = 1.00 vs AISC phi = 0.90) but uses higher load factors (1.35D + 1.5L vs 1.2D + 1.6L), roughly offsetting. AS 4100 and CSA S16 sit between AISC and EN 1993. The bolt and connection provisions show wider variation: AS 4100 uses phi = 0.80 for bolts vs AISC's phi = 0.75, making AS 4100 connection designs appear 7% more efficient for identical bolt groups, though the load factors differ.

Can I use AISC steel section properties with Eurocode load combinations?

No. Every element of a design — loads, resistance factors, material properties, and detailing — must come from a single code framework. Mixing codes invalidates the calibrated reliability. For example, using AISC phi = 0.90 with Eurocode load factors 1.35D + 1.5L would be unconservative because Eurocode's higher load factors are calibrated against its own resistance factors (gamma_M1 = 1.00). The Steel Calculator enforces a single-code selection per analysis session to prevent this.