This paper aims to propose an equivalent cylinder radius for the buckling behavior of relatively thick steel conical shells subjected to axial compression. This is both experimental and numerical/theoretical work. Test data on this subject is scarce and the existing design recommendations have not been validated experimentally, especially in the plastic buckling region. In order to authenticate the appropriateness of these existing rules and suggest others, collapse tests were conducted at the University of Liverpool and Universiti Teknikal Malaysia Melaka on cones and their equivalent cylinder counterparts. The first set of experiments consisted of two nominally identical cones and one equivalent cylinder. These mild steel models were CNC-machined, with the cones having the following geometric parameters: r₂ = 99 mm, r₁ = 49 mm, h = 100 mm, β = 26.56°, wall thickness of 3 mm. The equivalent cylinder was machined using the DNV CV 30.1 equivalent cylinder radius. The results show that the equivalent cylinder overestimates the conical models by 35% and 44% respectively.

The second set of four steel models consisted of: two cones having, r₂ = 70 mm, r₁ = 40 mm, h = 112 mm, β = 15°, wall thickness of 2 mm, and two equivalent cylinders designed according to ‘ECCS 2008 equivalent cylinder radius’. All four shells were manufactured in pairs using the traditional welding process. It has become evident here that the experimental collapse load predicted by the equivalent cylinders underestimated that of the cones by about 9%. Furthermore, numerical calculation indicated that the equivalent cylinder radius, suggested by the ECCS 2008, mimics its conical counterpart with good accuracy for cones with cone angle, 5° ≤ β ≤ 20°. However, for cones with large cone angle, β > 20°, one obtains large discrepancies between the cone and the equivalent cylinder. This highlights the weakness of the ECCS 2008 recommended equivalent cylinder radius. Finally, a new equivalent cylinder radius for axially compressed cones is suggested with better accuracy as compared to the ECCS current recommendation, and this in turn could be useful for design purposes.

本文旨在为承受轴向压缩的较厚钢锥壳的屈曲行为提出等效圆柱半径。这既是实验性的，也是数值/理论的工作。关于这个问题的测试数据很少，现有的设计建议还没有经过实验验证，特别是在塑料屈曲区域。为了验证这些现有规则的适当性并提出其他建议，利物浦大学和马来西亚马六甲理工大学对锥体及其等效圆柱体进行了坍塌试验。第一组实验由两个名义上相同的锥体和一个等效的圆柱体组成。这些低碳钢模型经过 CNC 加工，锥体具有以下几何参数： r₂  = 99 mm，r ₁  = 49 mm，h = 100 mm，β = 26.56°，壁厚 3 mm。使用 DNV CV 30.1 等效圆柱半径加工等效圆柱。结果表明，等效圆柱分别高估了圆锥模型 35% 和 44%。

第二组四个钢模型包括：两个锥体，r ₂  = 70 mm，r ₁ = 40 mm，h = 112 mm，β = 15°，壁厚为 2 mm，两个等效圆柱根据“ECCS 2008 等效圆柱半径”设计。所有四个外壳都是使用传统的焊接工艺成对制造的。很明显，等效圆柱体预测的实验坍塌载荷低估了锥体的坍塌载荷约 9%。此外，数值计算表明，ECCS 2008 建议的等效圆柱半径模拟其圆锥对应物，对于锥角为 5° ≤ β ≤ 20° 的锥具有良好的精度。然而，对于大锥角的锥体，β > 20°，在锥体和等效圆柱体之间会获得很大的差异。这突出了 ECCS 2008 推荐的等效圆柱半径的弱点。最后，