Need a book? Engineering books recommendations...

Return to index: [Subject] [Thread] [Date] [Author]

RE: Torsion on a steel tube and AISC WINTORQ program

[Subject Prev][Subject Next][Thread Prev][Thread Next]
Items 1, 2 and 3 below are essentially the same since Ao = b^2 (for a square HSS) and C = 2*t*Ao. Let me delve into the other question a bit as I can't answer it off the top of my head. It's been a while since I looked into the details there and I will have to reacquaint myself. I think what I will find is that the torsional function theta prime reduces to a function of the same constants above for HSS. But I have to re-derive it to be sure (and here I was just reminiscing in a psot yesterday about whiteboards and chalkboards).
 
Charlie
-----Original Message-----
From: Jesse Gobeli [mailto:jgobeli(--nospam--at)jgaeng.com]
Sent: Tuesday, March 18, 2003 8:02 AM
To: seaint(--nospam--at)seaint.org
Subject: Torsion on a steel tube and AISC WINTORQ program

List,
 
I have a question regarding calculating the shear stress in a steel tube.  I have the Steel Design Guide Torsional Analysis of Structural Steel Members and the WINTORQ program from AISC. 
 
Typically, I would calculate the shear stress in a tube using the formula: 
1.  shear stress = T / (2*b^2*t) (from table 4.1 of the Design Guide)
2.  shear stress = T / (2*t*Ao)  for constant T (Design Guide Eq. 4.4)  (This is the method used in example 5.2 of the Design Guide).
3.  shear stress = T / C (from Hollow Structural Sections Connection Manual spec. eq. 6-1, C is tabulated for each shape).
 
Methods 1,2 and 3 seem to give the same result. 
 
The WINTORQ program gives a "wall shear stress" value which seems to come from the following equation: 
4.  shear stress = G*t*theta'   (Design Guide Eq. 4.1)  (theta' from WINTORQ program)
This result can also be obtained using: 
5.  shear stress = T*t/J 
 
The problem is, the results from 1,2 and 3 are somewhat higher than the values from 4 and 5.  Since the Design Guide shows an example based on 2, but the WINTORQ program provides a value based on 4, I am not sure which value to use.  WINTORQ does not give any of the stresses obtained by 1,2 and 3.
 
(By the way, I have worked through the examples in the design guide for Wide Flanges, and along with the  WINTORQ program, they have been a HUGE help in understanding torsion on a wide flange beam.)
 
Jesse Gobeli, P.E.
Dublin, OH