Torsion in open steel sections (I beam/H beam) - neglecting the St Venant torsion 1

[icebloom]

From mechanics, torsion in I beams has 2 components, the warping component (force couple between flanges causing flanges to bend about their local major axes) and St Venant torsion. In the Eurocode for the design of torsion in an open section, it states that the St Venant component of the torsion can be neglected. Does anyone know why the Eurocode 3 (

https://www.phd.eng.br/wp-content/uploads/2015/12/en.1993.1.1.2005.pdf)

allows you to neglect the St Venant torsion in design? My understanding from SCI P385 (

https://www.steelconstruction.info/images/6/6f/Sci_p385.pdf)

is that the relative proportions of warping torsion vs St Venant torsion depend upon the span of the beam, and for longer beams the St Venant torsion is the dominant mode. Is there some inherent conservativism in only considering warping torsion? Are there cases where this is not accurate?

Eurocode 3

SCI P385 (orange curve is long beam, red curve is short beam

 

[DaveAtkins]

According to the text you cite, it is a simplification, not a conservative assumption. My understanding is if you design for the warping torsion, the section will be OK, even neglecting St. Venant torsion.

DaveAtkins

 

[CDLD]

If you neglect the St. Venant torsion, your warping moment will be higher, which will be conservative with respect to normal stress (in my experience normal stress usually controls in design, but not always).

Neglecting St. Venant torsion can be done as per AISC design guide 9:

Note that this method is conservative with respect to:
1. Warping normal stress (perhaps overly conservative for long spans).
2. Warping shear stress
3. Unconservative for shear stress in pure torsion (St Venant) especially in long span members. Note that this approximation is intended for short span members, since the shear stress from pure torsion is low.