AS 4100 vs AISC 360 Column Design — Key Differences

Australian structural engineers use AS 4100; American engineers use AISC 360. Both codes design steel columns for axial compression, but the underlying formulas, buckling curve systems, section classification, and section-specific parameters differ in ways that produce different capacity predictions for the same section under the same length and loading. This page provides a point-by-point comparison, including a fully worked example where the same UB/W section is checked under both codes to quantify the difference.

PRELIMINARY — NOT FOR CONSTRUCTION. All information is for educational and reference use only. Must be independently verified by a licensed Professional Engineer (PE) or Chartered Structural Engineer (CPEng / NER) before use in any project.

Reliability Framework and Resistance Factors

Both codes use limit states design, but with different resistance factors for compression members.

Parameter AS 4100:2020 AISC 360-22 (LRFD)
Capacity reduction factor phi = 0.90 for compression phi_c = 0.90 for compression
Load combinations AS/NZS 1170.0 (1.2G + 1.5Q) ASCE 7-22 (1.2D + 1.6L)
Reliability calibration ISO 2394 target beta = 3.8 (50 yr) Target beta = 3.0 (members)
Serviceability delta_s per AS 1170.0 Appendix C Deflection limits per Table C-C2.1

Although both codes converged on phi = 0.90 for compression, the underlying capacity equations are different, so the same section can have significantly different design capacity.

Column Buckling — alpha_c vs F_cr

The core of column design in both codes is the buckling curve that relates slenderness to reduced strength.

AS 4100 — Section 6.3.3 Member Capacity

The nominal member capacity in compression is:

N_c = alpha_c * N_s, with N_c <= N_s

Where:

The member slenderness reduction factor alpha_c is determined from Table 6.3.3(1) or (2) based on:

lambdan = (L_e / r) * sqrt(kf) * sqrt(f_y / 250)

And the modified slenderness:

lambda = lambda_n + alpha_a * alpha_b

Where:

The alpha_c values are then read from Table 6.3.3(1) or computed as:

alpha*c = xi * {1 - sqrt(1 - [90 / (xi _ lambda)]^2)}

Where xi = (lambda/90)^2 + 1 + eta, and eta = (alphaa * alphab) / (2 * lambda^2) for lambda > 20.

AISC 360-22 — Section E3 Flexural Buckling

The nominal compressive strength is:

P_n = F_cr * A_g

Where:

F_cr depends on whether the column is elastic or inelastic:

For L_c / r <= 4.71 * sqrt(E / F_y) (inelastic buckling):

F_cr = [0.658^(F_y / F_e)] * F_y

For L_c / r > 4.71 * sqrt(E / F_y) (elastic buckling):

F_cr = 0.877 * F_e

Where F_e = pi^2 * E / (L_c / r)^2

Design strength: phic * Pn = 0.90 * F_cr * A_g

Critical Difference — alpha_b (Section Geometry Constant)

This is the most significant conceptual difference between the two codes.

AS 4100's alpha_b explicitly accounts for the effect of section shape and manufacturing method on column buckling behavior. Hot-rolled I-sections (UB, UC) have alpha_b = 0.0. Cold-formed hollow sections can have alpha_b = -0.5 or -1.0 (a NEGATIVE value that REDUCES the modified slenderness, effectively shifting the column into a MORE favorable buckling curve).

This means AS 4100 recognizes that a cold-formed RHS column, if stress-relieved, buckles at a higher stress than a hot-rolled UB of the same slenderness — because the residual stress pattern is more favorable. This is captured through alpha_b.

AISC 360 does not have an equivalent parameter. All rolled I-shapes follow the same single buckling curve (Section E3). The distinction between different section types is implicitly handled by the F_e/F_y ratio through the 0.658 exponent in the inelastic range, but there is no explicit section-constant parameter like alpha_b.

Buckling Curve Families

Code Number of Curves Basis of Distinction
AS 4100 5 (via alpha_b) alpha_b ranges from -1.0 to +1.0, effectively creating 5 curves
AISC 360 1 (Section E3) Single curve: 0.658^(F_y/F_e) inelastic / 0.877F_e elastic
EN 1993-1-1 5 (a0 through d) Explicit 5-curve system with different imperfection factors
CSA S16 2 (SSRC 1 & 2) Two curves: rolled and welded, similar to AS 4100 approach

The AS 4100 approach is closest to CSA S16, which also uses a two-curve system distinguishing between rolled and welded sections. EN 1993-1-1 uses the most detailed system with five explicit buckling curves, similar in spirit to AS 4100's alpha_b-modulated approach.

Section Classification — Slenderness Limits

Both codes classify cross-sections for local buckling, but the limits differ.

Classification AS 4100 (Table 5.2) AISC 360-22 (Table B4.1a/b)
Compact (AS) / b/t <= lambda_ep (plastic limit) b/t <= lambda_p (compact limit, fully plastic)
Non-compact / b/t <= lambda_ey (yield limit) b/t <= lambda_r (noncompact limit, yield at flange)
Slender b/t > lambda_ey b/t > lambda_r (elastic buckling)

Example — flange slenderness of a UB/W-section:

For AS 4100 (hot-rolled I-section flange, one longitudinal edge supported):

For AISC 360 (rolled I-shape flange):

For Grade 300 (AS 4100) / A572 Gr. 50 (AISC), f_y = 300 MPa / F_y = 50 ksi:

The compact/plastic limits are nearly identical. The yield/noncompact limits differ significantly (14.6 vs 24.1), with AISC being more permissive for thin-flange sections — but this is rarely governing for standard rolled shapes which typically have flange b/t well under either limit.

Form Factor (k_f) — Unique to AS 4100

AS 4100 introduces the form factor k_f, which has no direct equivalent in AISC 360. The form factor is defined as:

k_f = A_e / A_g

Where A_e is the effective area accounting for local buckling of slender elements.

For sections with compact flanges and webs (b/t <= lambda_ep for all elements), k_f = 1.0 and has no effect. For sections with slender elements, k_f < 1.0 and reduces the effective area and slenderness (since lambda_n includes sqrt(k_f), a lower k_f reduces the effective slenderness, partially offsetting the area reduction).

AISC 360 accounts for local buckling of slender elements through the effective width method in Section E7, using a reduced effective area. However, AISC applies the reduction directly to the gross area in the P_n = F_cr * A_g formula (using A_e in place of A_g for slender-element members), without the intermediate k_f concept.

Effective Length Factors (k_e)

Both codes use the same theoretical effective length factors for idealized end conditions:

End Condition k_e (both codes)
Both ends fixed (no sway) 0.50
One fixed, one pinned (no sway) 0.70
Both pinned 1.00
One fixed, one free (cantilever) 2.10 (AS 4100) / 2.00 (AISC 360)
Both fixed (sway permitted) 1.20
One fixed, one pinned (sway permitted) 2.00

Note the subtle difference for cantilever columns: AS 4100 uses k_e = 2.1 (Clause 4.6.3.1: member with one end fixed and the other end free, the effective length factor may be taken as 2.1), while AISC 360 uses k_e = 2.0. The AS 4100 value of 2.1 is slightly more conservative, accounting for the fact that true fixity is never perfect.

For columns in rigid frames, both codes provide alignment charts (AS 4100 Figure 4.6.3.2, AISC 360 Appendix 7) with similar formulations based on member stiffness ratios (gamma = sum(I/L)_columns / sum(I/L)_beams).

Worked Comparison — 250UB37.3 vs W10x26 (Same Section, Different Capacity)

To quantify the difference in predicted column capacity, we compare a nearly identical section under both codes.

Section Properties

Property 250UB37.3 (AS 4100) W10x26 (AISC 360)
Nominal depth 256 mm 10.3 in. (262 mm)
Flange width 146 mm 5.77 in. (147 mm)
Flange thickness 10.9 mm 0.440 in. (11.2 mm)
Web thickness 6.4 mm 0.260 in. (6.6 mm)
Area A_g 4760 mm^2 4.95 in.^2 (3194 mm^2)
r_x (radius of gyration) 109 mm 4.35 in. (110 mm)
r_y (radius of gyration) 34.8 mm 1.37 in. (34.8 mm)
Steel grade Grade 300 (f_y = 300 MPa) A992 (F_y = 50 ksi = 345 MPa)
E (Young's modulus) 200,000 MPa 29,000 ksi (200,000 MPa)

Note: The W10x26 has higher yield strength (345 MPa vs 300 MPa) which partially accounts for its different capacity, but the shape geometry is nearly identical (less than 2% difference in any dimension).

Design Condition

AS 4100 Calculation (250UB37.3, Grade 300)

Step 1 — Section capacity (N_s)

Step 2 — Modified slenderness

Step 3 — alpha_b and alpha_a

Step 4 — Modified slenderness with alpha_b

Step 5 — alpha_c from Table 6.3.3(1)

Step 6 — Member capacity

AISC 360-22 Calculation (W10x26, A992)

Step 1 — Cross-section check

Step 2 — Elastic buckling stress

Step 3 — Critical buckling stress

Step 4 — Column capacity

Comparison Summary

Parameter AS 4100 (250UB37.3 G300) AISC 360 (W10x26 A992) AS 4100 / AISC 360
Nominal capacity 720 kN 419 kN 1.72
Design capacity 648 kN 377 kN 1.72
Utilization at N* = 300 0.46 0.80 —

The AS 4100 capacity is approximately 72% higher than the AISC 360 capacity. This large difference is driven by three factors:

  1. Area difference: The 250UB37.3 has A_g = 4760 mm^2 vs W10x26 A_g = 3194 mm^2. This alone accounts for a 1.49x ratio.
  2. Steel grade: The W10x26 uses F_y = 50 ksi (345 MPa) vs the 250UB37.3 using f_y = 300 MPa. All else equal, higher yield strength reduces F_cr/F_y faster because F_e/F_y decreases.
  3. Buckling curve: AS 4100's alpha_b = 0.0 and alpha_c interpolation yields alpha_c = 0.504, while AISC's single curve via F_e/F_y yields F_cr/F_y = 19.0/50 = 0.38. The AS 4100 curve is less punitive for this slenderness range.

If we normalize for area and yield strength differences, the AS 4100 / AISC 360 capacity ratio reduces to approximately 1.08 — meaning AS 4100 still gives about 8% higher capacity for this column. This remaining 8% is attributable to the different buckling curve shapes.

Key Column Design Differences — Summary Table

Aspect AS 4100:2020 AISC 360-22
Resistance factor phi = 0.90 phi_c = 0.90
Basic capacity formula N_c = alpha_c * N_s P_n = F_cr * A_g
Section constant alpha_b (geometry effect, Table 6.3.3) None (single E3 curve)
Form factor k_f = A_e / A_g Implicit in effective area (Section E7)
Slenderness parameter lambdan = (L_e/r) * sqrt(kf * f_y/250) L_c/r compared to 4.71*sqrt(E/F_y)
Inelastic buckling range Table 6.3.3(1) for lambda <= ~170 F_cr = 0.658^(F_y/F_e) * F_y
Elastic buckling range Table 6.3.3(1) for lambda > ~170 F_cr = 0.877 * F_e
Cantilever effective length k_e = 2.1 K = 2.0 (theoretical, AISC App. 7)
Torsional-flexural buckling Section 6.3.4 (separate check) Section E4 (combined with flexural)
Bending + compression Section 8.4 (separate in-plane/out-of-plane) Chapter H (combined interaction equations)

Design Implications for Practitioners

When AS 4100 is More Conservative

When AISC 360 is More Conservative

Practical Recommendation

When designing a column that must satisfy BOTH codes (e.g., an international project with dual compliance), design to the MORE conservative of the two. For intermediate slenderness columns (KL/r 80–130), this is typically AISC 360. For cantilever columns or stress-relieved hollow sections, AS 4100 may govern.

Frequently Asked Questions

1. Why does AS 4100 use the alpha_b section constant?

alpha_b accounts for differences in residual stress patterns between manufacturing methods. Hot-rolled sections cool unevenly, creating a residual stress field that reduces buckling capacity. Stress-relieved hollow sections have lower residual stresses and thus buckle at a higher stress for the same slenderness. The alpha_b parameter shifts the buckling curve to reflect this — a concept also used in CSA S16 and EN 1993-1-1 (via the 5-curve system), but absent from AISC 360's single E3 curve.

2. Can I convert an AISC 360 column design to AS 4100 by applying a simple factor?

No. As the worked example shows, the difference varies with slenderness, section type, steel grade, and area. A single conversion factor would be inaccurate across the full design space. Always perform a complete AS 4100 check when Australian compliance is required.

3. Is AS 4100's form factor (k_f) the same as AISC's effective width reduction?

Conceptually similar, but the mechanics differ. AS 4100's kf reduces the effective AREA (N_s = k_f * An * f_y) AND the slenderness parameter (lambda_n includes sqrt(k_f)), a dual effect that partially offsets the area loss. AISC's effective width method in Section E7 directly replaces A_g with A_e in the capacity equation, but does not modify the slenderness calculation.

Try it now: Check your column design with our free AU Steel Column Capacity calculator →

Related Pages

Disclaimer

This page provides an educational comparison of AS 4100:2020 and AISC 360-22 column design provisions. It is not a substitute for the actual code documents. Always design to the legally adopted building code in your jurisdiction. All design must be independently verified by a licensed Professional Engineer (PE) or Chartered Professional Engineer (CPEng / NER) before use in any project. Steel Calculator does not reproduce the full text of any copyrighted standard — refer to the official AS 4100 (Standards Australia) or AISC Steel Construction Manual for the complete provisions. This tool is for preliminary use only; final designs require professional engineering certification.

Codes referenced: AS 4100:2020 (Steel Structures), AISC 360-22 (Specification for Structural Steel Buildings), AS/NZS 1170.0 (Structural Design Actions — General Principles), ASCE 7-22 (Minimum Design Loads).