A universal beam is taller than it is wide (depth-to-flange-width ratio greater than 1.0, usually in the 1.3 to 2.0 range), with a relatively narrow flange on a deep web; a universal column is squatter, with depth and flange width close to equal [S1][S3]. That single proportion difference drives the rest of the engineering.
UB and UC are both hot-rolled, parallel-flange I/H sections used in buildings and bridges, with flange thicknesses commonly running 9 to 30 mm and web thicknesses 6.5 to 20 mm on the wider UC sizes [S2]. Reading the designation matters more than reading the shape: a 203x133 UB 25 and a 203x203 UC 46 share a 203 mm depth, but the UC is wider, heavier per metre, and built for compression, not spanning [S3].
Flange-to-depth proportion, the headline number
UB serial sizes run 127 mm to 1016 mm in depth, with flange widths deliberately kept narrower so the section carries more material away from the neutral axis and resists bending efficiently across a span [S3]. A 533x210 UB 82, for example, is 533 mm deep but only 210 mm across the flanges, giving a depth-to-flange-width ratio of roughly 2.5, which is what makes it efficient as a flexural member [S1][S3].
UC serial sizes keep depth close to flange width on purpose, with common 152x152, 203x203, 254x254, 305x305, and 356x368 sizes in regular stock, and section depths from about 152 mm up to 356 mm or more [S3][S4]. A 203x203 UC 46 is 203 mm deep and 206 mm wide, ratio ~0.99, which is the geometric signal that the section is designed to buckle evenly about both axes under axial load [S3].
Web thickness and flange thickness, the secondary signal
For wide-flange UC sections, flange thickness commonly ranges 9 to 30 mm and web thickness 6.5 to 20 mm across the standard size band 4 in to 16 in (10.2 to 40.6 cm) wide [S2]. Flanges are thicker than the web on purpose: the web's job is to carry shear and hold the flanges apart, while the flanges carry bending stress and most of the axial load [S2][S5].
UB webs are typically slimmer relative to depth because bending efficiency, not shear or compression, governs the design [S3][S5]. UC webs and flanges are both beefed up, and the fillet radii where web meets flange are broader, which raises load capacity and improves local-buckling behaviour under compression [S3][S4]. For local-buckling checks, the governing width-to-thickness ratios are flange B/2T and web d/t, applied separately to UB and UC tables [S6].
Weight per metre and serial-size variants

Both UB and UC are stocked in multiple mass-per-metre variants for the same serial size, and the 152x152 UC family alone shows three standard weights with slightly different overall dimensions to suit different axial loads [S3]. UB serial sizes run from about 13 kg/m at the light end (127x76x13 UB) up to roughly 222 kg/m for heavy 1016x305 sections, so a single designation can vary by a factor of more than 15 in linear mass [S3].
UC mass per metre scales with section size, and heavy UC members (for example 305x305x240 or larger) are the typical choice where the column carries multi-storey axial load and needs both axes of buckling resistance [S3][S4]. For reference, the flange and web catalogue pages on most UK and Australian mill sites lay out B, T, d, t and mass per metre in a single row, which is the only safe way to compare two sections of similar depth [S6].
Structural role: bending versus compression
UB are selected to carry bending and shear across a horizontal span, so the deep web lifts the flanges away from the neutral axis and maximises second moment of area about the strong axis [S1][S3][S5]. The shape is weak in torsion and weak about its weak axis, which is why UB rarely gets used as a column without explicit engineering checks [S5].
UC are designed to carry axial compression, often with superimposed bending, and the near-equal depth and flange width gives similar radii of gyration about both axes, which directly improves slenderness and buckling capacity in both directions [S1][S3][S4]. In multi-storey frames, this balanced behaviour is what keeps drift and second-order effects within design limits without oversized base plates [S3].
Decision matrix: pick UB or UC by criterion

Across the four decision criteria below, UB and UC trade off directly rather than compete on the same axis [S1][S2][S3][S7]:
1. Bending efficiency (strong-axis Ixx per kg): UB wins, because material sits far from the neutral axis.
2. Axial compression and bi-axial buckling: UC wins, because the squarer B/d ratio gives similar capacity both ways.
3. Connection simplicity at beam-to-column joints: UC wins, because the wider, thicker flanges offer a bigger target for end-plate and fin-plate connections.
4. Floor-zone depth for a given span: UB wins, only because the section is shallower across the flanges (lower B) for a given depth, which can free up ceiling void; UC is wider and heavier for the same depth, so it eats more floor zone.
Rule of thumb from one supplier's preliminary sizing notes: a starting steel column of about 12 width-to-thickness ratio and 3 to 5 height-to-width ratio, with axial capacity taken at roughly 5% of yield strength times cross-sectional area, will usually resolve to a UC, not a UB [S2].
When to substitute and when not to
UB can be used as a column when its calculated capacity, slenderness, and connections work out, but the engineer must check weak-axis buckling, which is the usual failure mode because UB flanges are narrow [S1][S5]. UC can be used as a beam, but the heavier self-weight per metre and lower strong-axis efficiency mean it is only worth it where the section is already on site or the loading favours a stocky profile [S1][S7].
Neither substitution should be made on the basis of shape or depth alone; the full designation (serial size plus mass per metre) on the approved drawing is the only safe reference, and even then the engineer should re-run slenderness and connection checks against the actual section properties [S1][S6]. For related spec-driven comparisons on adjacent equipment categories, see the construction equipment reference and the flange spec page, which use the same read-the-table discipline [S6].
Material and standards context

Common structural steel grades behind both UB and UC include ASTM A992 (50,000 to 65,000 psi yield, 340 to 450 MPa) and ASTM A572 (42,000 to 60,000 psi, 290 to 410 MPa), with A992 having largely replaced A572 and A36 in modern US W-section practice [S5]. The AISC Steel Construction Manual covers both ASD and LRFD approaches for these sections from the 13th edition onward [S5].
On the UK/European side, UC and UB are commonly rolled in grades S275 and S355 to EN 10025, and the same cross-section designations are used in Australian practice to AS 3679.1 [S3][S4][S6]. Section property tables from any of these regions report serial size, mass per metre, B, T, d, t, root radius, and local-buckling ratios in a consistent format, so a UB 533x210x82 in one region corresponds directly to a comparable section in another [S6][S7]. For lab and field measurement gear used to verify flange and web thickness on received stock, see the checking and testing apparatus catalogue.
Related analysis: CMM robot loading: ROI structure, integration options, and where the payback actually.