Design Problem
Problem: Verify a 254 x 254 UC 73 in S355 steel acting as an internal column in a multi-storey braced frame. The column is simply connected (pinned base, pinned top). Storey height = 5.0 m. The column forms part of a sway-frame where sway effects are resisted by a concrete core (braced in both directions per EN 1993-1-1 Clause 5.2.1).
Loading:
- Permanent axial: NGk = 800 kN (from floors above via beam reactions)
- Variable axial: NQk = 450 kN (imposed load from floors)
Design parameters per UK NA:
- gamma_M0 = 1.00, gamma_M1 = 1.00
- Steel: S355JR to BS EN 10025-2, fy = 355 MPa for t <= 16 mm
- Buckling length factor k = 1.0 (pinned-pinned), Lcr = 5.0 m
- UK NA uses Eq. 6.10b for ULS: NEd = 1.35 x 800 + 1.5 x 450 = 1,080 + 675 = 1,755 kN
Section Properties — 254 x 254 UC 73
From SCI P363 (Blue Book):
| Property | Symbol | Value | Units |
|---|---|---|---|
| Depth of section | h | 254.1 | mm |
| Flange width | b | 254.6 | mm |
| Web thickness | tw | 8.6 | mm |
| Flange thickness | tf | 14.2 | mm |
| Root radius | r | 12.7 | mm |
| Area | A | 93.1 | cm2 |
| Iy | Iy | 11,400 | cm4 |
| Iz | Iz | 3,910 | cm4 |
| iy | iy | 11.1 | cm |
| iz | iz | 6.48 | cm |
| h/b ratio | h/b | 1.00 | — |
Step 1: Cross-Section Classification — Clause 5.5
Flange classification (outstand in compression):
c = (b - tw - 2r) / 2 = (254.6 - 8.6 - 25.4) / 2 = 110.3 mm c / tf = 110.3 / 14.2 = 7.77 epsilon = sqrt(235 / 355) = 0.814
Class 1 limit for flange in compression: 9ÃÂ÷epsilon = 9 x 0.814 = 7.32 Class 2 limit: 10ÃÂ÷epsilon = 10 x 0.814 = 8.14
7.77 > 7.32 but 7.77 < 8.14 => Flange is Class 2
Web classification (pure compression):
c = d = h - 2tf - 2r = 254.1 - 28.4 - 25.4 = 200.3 mm (approximately) c / tw = 200.3 / 8.6 = 23.3
Class 1 limit for web in compression: 33ÃÂ÷epsilon = 33 x 0.814 = 26.9 23.3 < 26.9 => Web is Class 1
Overall classification: Class 2 (governed by flange).
Step 2: Cross-Section Compression Resistance — Clause 6.2.4
For Class 2 section, use the plastic resistance with no local buckling reduction:
Nc,Rd = A x fy / gamma_M0 = 9,310 x 355 / 1.00 = 3,305 kN
Gross utilisation (ignoring buckling): NEd / Nc,Rd = 1,755 / 3,305 = 0.531
Step 3: Flexural Buckling Resistance — Clause 6.3.1
Effective length (pinned-pinned):
Lcr,y = k_y x L = 1.0 x 5,000 = 5,000 mm (buckling about y-y axis) Lcr,z = k_z x L = 1.0 x 5,000 = 5,000 mm (buckling about z-z axis, weaker)
Non-dimensional slenderness:
lambda_y = Lcr,y / iy = 5,000 / 111 = 45.0 lambda_1 = 93.9ÃÂ÷epsilon = 93.9 x 0.814 = 76.4 lambda_bar_y = lambda_y / lambda_1 = 45.0 / 76.4 = 0.589
lambda_z = Lcr,z / iz = 5,000 / 64.8 = 77.2 lambda_bar_z = lambda_z / lambda_1 = 77.2 / 76.4 = 1.010 (governing)
The z-z axis governs — buckling is about the weaker axis.
Step 4: Buckling Curve Selection — UK NA Table 6.2
Buckling about y-y axis:
h/b = 254.1 / 254.6 = 1.00 <= 1.2, tf = 14.2 mm <= 100 mm => Buckling curve b (UK NA Table 6.2) alpha_y = 0.34 (imperfection factor for curve b)
Buckling about z-z axis:
h/b = 1.00 <= 1.2, tf = 14.2 <= 100 mm => Buckling curve c (UK NA Table 6.2 — minor axis) alpha_z = 0.49 (imperfection factor for curve c)
Step 5: Reduction Factor chi — Clause 6.3.1.2
Buckling about y-y axis (curve b):
Phi_y = 0.5 x [1 + alpha_y x (lambda_bar_y - 0.2) + lambda_bar_y2] Phi_y = 0.5 x [1 + 0.34 x (0.589 - 0.2) + 0.5892] = 0.5 x [1 + 0.132 + 0.347] = 0.740
chi_y = 1 / (Phi_y + sqrt(Phi_y2 - lambda_bar_y2)) chi_y = 1 / (0.740 + sqrt(0.7402 - 0.5892)) = 1 / (0.740 + 0.448) = 0.842
Buckling about z-z axis (curve c) — governing:
Phi_z = 0.5 x [1 + alpha_z x (lambda_bar_z - 0.2) + lambda_bar_z2] Phi_z = 0.5 x [1 + 0.49 x (1.010 - 0.2) + 1.0102] = 0.5 x [1 + 0.397 + 1.020] = 1.209
chi_z = 1 / (Phi_z + sqrt(Phi_z2 - lambda_bar_z2)) chi_z = 1 / (1.209 + sqrt(1.2092 - 1.0102)) = 1 / (1.209 + 0.663) = 0.534
Step 6: Buckling Resistance
Governing buckling resistance (z-z axis):
Nb,Rd = chi_z x A x fy / gamma_M1 Nb,Rd = 0.534 x 9,310 x 355 / 1.00 = 1,765 kN
Utilisation:
NEd / Nb,Rd = 1,755 / 1,765 = 0.994 (99.4 %)
The 254 x 254 UC 73 is very tightly utilised. In practice, a designer would likely select the next size up (254 x 254 UC 89) to provide a margin of safety, or verify that the effective length assumptions are conservative.
Step 7: Alternative — 254 x 254 UC 89
Let us verify the next UC section for comparison:
| Property | UC 73 | UC 89 |
|---|---|---|
| Area | 93.1 cm2 | 113 cm2 |
| iz | 6.48 cm | 6.55 cm |
| lambda_bar_z | 1.010 | 0.998 |
| chi_z (curve c) | 0.534 | 0.545 |
| Nb,Rd | 1,765 kN | 2,187 kN |
| Utilisation | 99.4 % | 80.2 % |
The UC 89 provides a much more comfortable 80 % utilisation with minimal weight penalty (89 versus 73 kg/m).
UK Buckling Curve Selection Rules — Summary
Per UK NA to BS EN 1993-1-1, Table 6.2:
| Section type | Limits | Buckling about y-y | Buckling about z-z |
|---|---|---|---|
| UC (h/b <= 1.2, tf <= 100 mm) | S235-S460 | b | c |
| UB (h/b > 1.2, tf <= 100 mm) | S235-S355 | b | c |
| UB (h/b > 1.2, tf <= 100 mm) | S420-S460 | a | b |
| RHS hot-finished | All | a | a |
| CHS hot-finished | All | a | a |
| Welded box sections | tf <= 40 mm | c | c |
Key UK NA differences from EN 1993-1-1 recommended values:
- The UK NA is generally more conservative for minor-axis buckling of UC sections (curve c instead of b)
- UK NA requires curve c for minor-axis buckling of all rolled I and H sections (EN 1993-1-1 allows curve b for h/b <= 1.2)
- UK research demonstrated that residual stress patterns in UK-produced UC sections justify curve c
Frequently Asked Questions
Why does the UK NA use curve c for minor-axis UC buckling?
Extensive UK research (SCI Report RT953 and University of Sheffield full-scale testing) demonstrated that residual stress patterns in domestically produced UC sections reduce the minor-axis buckling capacity compared to the EN 1993-1-1 default curve b. The UK NA conservatively adopts curve c to reflect UK production practices. For UK-designed buildings, curve c must be used unless project-specific testing justifies otherwise.
How does effective length factor k differ between braced and unbraced frames?
For braced frames (Clause 5.2.1), k <= 1.0 for columns that do not contribute to sway stability. Pinned-pinned columns use k = 1.0; fixed-fixed columns use k = 0.5 (theoretical) or 0.7 (practical, allowing for imperfect fixity by a factor of 1.4). For sway frames (Clause 5.2.2), k > 1.0 is typical — the SCI P362 stability guidance provides charts. The UK NA recommends using second-order analysis for k > 1.5 to avoid conservatism in the effective length method.
What is the minimum UC size for a given UK storey height?
For pinned-pinned UC columns in typical multi-storey braced frames: 152UC for heights to 3.0 m, 203UC for 3-4 m, 254UC for 4-6 m, 305UC for 6-8 m. These are approximate and depend on loading. Deep columns (356UC) are used when high axial loads exceed 2,500 kN or when 152/203UC slenderness exceeds limits. UK practice favours UC sections over fabricated box or compound sections wherever possible.
When does torsional or torsional-flexural buckling govern?
For doubly-symmetric hot-rolled UC and UB sections, flexural buckling (Clause 6.3.1) almost always governs. Torsional or torsional-flexural buckling (Clause 6.3.1.4) governs only for: (1) open sections such as angles, tees, and channels; (2) cruciform sections; (3) sections where the shear centre and centroid do not coincide. For UK UC sections, buckling is always flexural about the minor axis.
Related Pages
- EN 1993 Column Buckling — Full Guide
- UK Column Design Guide
- UK Column K Factor Reference
- UK Steel Grades
- UK Compact Section Limits
- UK Combined Loading
- UK Steel Section Properties
Educational reference only. All design values are per BS EN 1993-1-1:2005 + UK National Annex and BS EN 10025-2:2019. Verify all values against the current editions of the standards and the applicable UK National Annex for your project jurisdiction. Designs must be independently verified by a Chartered Structural Engineer registered with the Institution of Structural Engineers (IStructE) or the Institution of Civil Engineers (ICE). Results are PRELIMINARY — NOT FOR CONSTRUCTION without independent professional verification.