Abstract:To study the mechanical performance of cold-formed steel web-stiffened double-limb built-up short columns, the effects of parameters, such as the web height-to-thickness ratio, stiffening angle, flange width-to-thickness ratio, and screw spacing, on the mechanical properties of members were systematically investigated by combining experimental research, numerical simulation, and theoretical analysis. In addition, the prediction results of bearing capacity from the effective width method and the direct strength method in the Chinese code GB/T 50018—2025 and the North American code AISI S100—2016 were compared. The results show that the specimens mainly undergo distortional buckling and distortional-local interactive buckling; the ultimate bearing capacity increases with the increase of stiffening angle and flange width-to-thickness ratio, and the maximum improvements reach 22.09% and 19.46% at stiffening angles of 10° and 20°, respectively; the increase in web height-to-thickness ratio leads to a significant decrease in bearing capacity, with a decrease ranging from 9.45% to 22.42%; the screw spacing has a more significant impact on the ultimate bearing capacity of medium-to-long columns (such as 2 000 mm and 2 500 mm specimens); the calculation results of bearing capacity using the effective width method in the North American code AISI S100—2016 are closer to the experimental results, while the effective width method in the Chinese code GB/T 50018—2025 and the direct strength method in the North American code AISI S100—2016 are both conservative. On this basis, the code methods were modified by combining the experimental and finite element data, a more accurate calculation method for bearing capacity was proposed, and its reliability was verified. The proposed calculation method takes into account both the stiffening effect and the built-up effect and has good prediction accuracy and applicability, providing a theoretical basis for the revision of relevant codes and design practice.