Australian Column Effective Length Factor (k_e) — AS 4100 Clause 6.3
Complete reference for the effective length factor k_e for steel columns per AS 4100:2020 Clause 6.3. Covers theoretical effective lengths, alignment chart methods, the distinction between braced (non-sway) and sway frames, end restraint coefficients, and worked calculations for typical Australian multi-storey steel framing.
Quick access: Australian steel properties | Combined loading (beam-columns) | Column capacity calculator | Brace connection design
The Effective Length Concept — Clause 6.3.1
The effective length L_e of a column is the length of a pin-ended column of identical cross-section that would have the same elastic buckling load (Euler load) as the actual column with its actual end restraints. The effective length factor k_e relates the effective length to the geometric column length:
L_e = k_e x L
where L is the clear column length between floors or between braced points, and k_e is the effective length factor.
The elastic flexural buckling load about a given axis is:
N_om = pi^2 x E x I / (k_e x L)^2
The effective length factor k_e depends on:
- The rotational restraint provided by the beams framing into the column at each end
- The relative stiffness of the column to the restraining beams
- Whether the frame is braced (non-sway) or unbraced (sway)
- The pattern of axial load in the column (uniform vs stepped)
Braced versus Sway Frames — Clause 6.3.2
AS 4100 distinguishes between two fundamentally different frame behaviours:
Braced (Non-Sway) Frame
A frame in which lateral stability is provided by a bracing system (cross-bracing, shear walls, or cores) that is sufficiently stiff to resist all horizontal loads and limit the inter-storey drift to negligible values. In a braced frame, the columns buckle in single-curvature mode between floor levels without lateral displacement of the ends.
For a braced frame, the effective length factor is bounded: 0.5 <= k_e <= 1.0. The theoretical minimum k_e = 0.5 corresponds to a column fully fixed at both ends. The practical minimum adopted in design is typically k_e = 0.70 to 0.85 for columns with nominal end restraints.
Sway Frame
A frame in which lateral stability depends on the flexural stiffness of the columns and beams acting as rigid or semi-rigid frames. Lateral drift is resisted by frame action rather than a separate bracing system. In a sway frame, columns buckle in double-curvature mode with lateral displacement of the ends.
For a sway frame, the effective length factor is: 1.0 <= k_e <= infinity (theoretically), with practical values typically 2.0 to 3.0 for unbraced multi-storey frames. Very slender sway frames can have k_e >> 3.0.
Alignment Chart Method (Column Nomograph) — Clause 6.3.3
The alignment chart provides a graphical/numerical method for determining k_e from the relative stiffness ratios at each column end. The method was developed by Julian and Lawrence (1959) and adopted in AS 4100 via reference to the SSRC Guide.
End Restraint Coefficients
At each column end, the restraint coefficient G is calculated as:
G = sum(I_c / L_c) / sum(I_b / L_b)
where:
- I_c = second moment of area of the columns meeting at the joint (in the plane of buckling)
- L_c = length of the column segment
- I_b = second moment of area of the beams meeting at the joint (in the plane of buckling)
- L_b = length of the beam segment
The summations include all columns and beams rigidly connected at the joint in the plane of buckling.
At a fixed base: G = 1.0 (nominally pinned) to G = 0.0 (perfectly fixed). In Australian practice, typical column bases are modelled as nominally pinned (G = 10) unless the base plate and holding-down bolts are specifically designed for moment transfer.
k_e Values for Braced Frames (Non-Sway)
| G_A (top) | G_B = 0.0 | G_B = 0.5 | G_B = 1.0 | G_B = 2.0 | G_B = 5.0 | G_B = 10.0 |
|---|---|---|---|---|---|---|
| 0.0 | 0.50 | 0.57 | 0.63 | 0.67 | 0.71 | 0.73 |
| 0.5 | 0.57 | 0.65 | 0.71 | 0.76 | 0.80 | 0.82 |
| 1.0 | 0.63 | 0.71 | 0.78 | 0.83 | 0.87 | 0.90 |
| 2.0 | 0.67 | 0.76 | 0.83 | 0.89 | 0.93 | 0.95 |
| 5.0 | 0.71 | 0.80 | 0.87 | 0.93 | 0.97 | 1.00 |
| 10.0 | 0.73 | 0.82 | 0.90 | 0.95 | 1.00 | 1.00 |
For braced frames, k_e is always less than or equal to 1.0.
k_e Values for Sway Frames
The sway frame values are significantly larger due to the lateral displacement of column ends during buckling.
Simplified k_e Values for Design — Clause 6.3.4
For preliminary and routine design where a full alignment chart analysis is not warranted, AS 4100 permits the following simplified effective length factors:
| Frame Type | End Conditions | k_e (simplified) |
|---|---|---|
| Braced | Both ends nominally pinned | 1.00 |
| Braced | One end fixed, one nominally pinned | 0.85 |
| Braced | Both ends nominally fixed | 0.75 |
| Sway | Both ends nominally pinned | 2.50 |
| Sway | One end fixed, one nominally pinned | 2.00 |
| Sway | Both ends nominally fixed | 1.50 |
Story Buckling Method — Clause 6.3.5
For multi-storey sway frames with regular geometry and loading, the story buckling method provides a more rational and typically less conservative effective length than the alignment chart method. The effective length factor for the storey is:
k_e = pi x sqrt(E / f_y) x sqrt(sum(N*) / (sum(N_om_storey)))
where sum(N*) is the total factored axial load in all columns in the storey, and sum(N_om_storey) is the sum of the Euler buckling loads of all columns in the storey with k_e = 1.0.
The story buckling method is preferred for frames with more than three storeys because it accounts for the load-sharing and restraint interaction between columns that the simple alignment chart ignores.
Column Effective Length in Both Principal Axes
A column must be checked for buckling about BOTH principal axes. The effective length factor k_e is generally different about the major (x) and minor (y) axes because the beam restraints at each floor level differ:
Major axis (k_ex): Beams framing into the column flanges provide rotational restraint. The restraint coefficient G_x depends on the ratio of column I_x/L_c to beam I_x/L_b in the major-axis plane.
Minor axis (k_ey): Only connections to the column web (typically simple shear tabs or end plates) provide restraint. These are frequently modelled as pinned, resulting in k_ey = 1.0 for braced frames, unless the web connection is designed as a moment connection.
For a typical braced frame with moment connections to the column flanges and simple shear connections to the web:
k_ex = 0.75 to 0.85 (beams framing into flanges provide restraint) k_ey = 1.00 (web connections are nominally pinned)
The column capacity is governed by buckling about the axis with the higher non-dimensional slenderness ratio: lambda_n = (L_e / r) x sqrt(f_y / (pi^2 x E)).
Worked Example: k_e for Multi-Storey Braced Frame Column
Problem: An internal column in a 4-storey braced steel frame consists of a 250UC89.5 Grade 300 section continuous through all floors. The column is subjected to major-axis bending from beams framing into the column flanges. Determine the effective length factor k_ex for the ground-to-first-floor segment, which has a clear height of 4.2 m.
Given:
- Column: 250UC89.5; I_x = 143 x 10^6 mm^4; L_c = 4.2 m
- Beam at top of segment (first floor): two 310UB40.4 beams framing into column; I_x = 86.4 x 10^6 mm^4 each; L_b = 8.0 m each
- Beam at bottom of segment (ground): column base is nominally pinned
- Frame type: braced (lateral stability from concrete core)
Solution:
Step 1: Calculate restraint coefficient at top joint (G_top)
sum(I_c / L_c) = 2 x 143 x 10^6 / 4200 = 68,095 mm^3 (two column segments, above and below the joint)
sum(I_b / L_b) = 2 x 86.4 x 10^6 / 8000 = 21,600 mm^3 (two beams)
G_top = 68,095 / 21,600 = 3.15
Step 2: Restraint coefficient at bottom (G_bottom)
Nominally pinned base: G_bottom = 10.0
Step 3: Determine k_e from alignment chart
For braced frame with G_A = 3.15 and G_B = 10.0, interpolating from the table:
G_A = 2.0, G_B = 10.0: k_e = 0.95 G_A = 5.0, G_B = 10.0: k_e = 1.00
Interpolating for G_A = 3.15: k_e = 0.95 + (3.15 - 2.0)/(5.0 - 2.0) x (1.00 - 0.95) = 0.95 + 0.019 = 0.97
Step 4: Effective length
L_ex = k_ex x L = 0.97 x 4200 = 4074 mm
Step 5: Slenderness ratio
r_x = sqrt(I_x / A) = sqrt(143 x 10^6 / 11400) = 112 mm
L_ex / r_x = 4074 / 112 = 36.4
lambda_n = 36.4 x sqrt(280 / (pi^2 x 200000)) = 36.4 x sqrt(1.418 x 10^(-4)) = 36.4 x 0.01191 = 0.434
Result: k_ex = 0.97, L_ex = 4.07 m, slenderness ratio 36.4. The column has significant reserve against Euler buckling (lambda_n << 1.0), and the design will be governed by section capacity or combined actions rather than pure buckling.
Frequently Asked Questions
What is the difference between k_e and k in AS 4100?
In AS 4100, k_e is the effective length factor (Clause 6.3), while k_f is the form factor (Clause 6.2) that accounts for local buckling in slender sections. The notation k is ambiguous and should be avoided -- always use k_e for effective length and k_f for form factor. The effective length factor k_e multiplies the geometric length to get the Euler buckling length. Typical values range from 0.5 (fixed-fixed) to 2.5+ (sway frame).
When can I use k_e = 1.0 for a braced frame column?
k_e = 1.0 should be used for braced frame columns when both ends are nominally pinned or when the beam-to-column connections at both ends are designed as simple (shear-only) connections with no moment transfer. In Australian practice, most braced frame columns use k_e = 1.0 for the minor axis (web connections are typically simple) and k_e = 0.85 to 1.0 for the major axis depending on the flange connection type.
How does the alignment chart account for semi-rigid connections?
The standard alignment chart assumes fully rigid connections at the column ends. For semi-rigid connections, the beam stiffness I_b/L_b should be reduced by a factor that accounts for the connection flexibility. An equivalent beam stiffness can be calculated as (I_b/L_b)_eff = (I_b/L_b) / (1 + 6EI_b/L_b/G_k), where G_k is the secant rotational stiffness of the connection. For typical Australian end-plate connections, G_k is between 10,000 and 50,000 kNm/rad, and the effect on k_e is generally less than 5% for braced frames.
What effective length factor applies to a column in a portal frame?
For a portal frame column (sway frame), the effective length factor depends on the base fixity and the rafter stiffness. With a nominally pinned base (typical for Australian portal frame construction), k_e ranges from 2.0 to 3.0 for the in-plane direction. The exact value should be determined by a rational buckling analysis (computational) or by the alignment chart for sway frames. Many Australian portal frame designers conservatively assume k_e = 2.5 for pinned-base portal frame columns.
Does the effective length factor apply to both axes of a column?
Yes, k_e must be determined independently for buckling about each principal axis. The beams and connections that provide restraint in the major-axis plane are different from those in the minor-axis plane, and the effective length factors are generally different. The design column capacity is governed by the more critical axis, which is typically the minor axis (higher k_e and smaller r) for I-section columns in braced frames, or the major axis (higher k_e due to sway) for I-section columns in sway frames.
Educational reference only. All design values must be verified against the current edition of AS 4100:2020 and the project specification. This information does not constitute professional engineering advice. Always consult a qualified structural engineer for design decisions.