Shear Centre of B2B Steel Angles with Gap

[EngDM]

Hey all,

I was wondering if anyone knew an appropriate approach for the shear centre of back to back angles with a gap, for instance in an OWSJ application. I have shop drawings for an existing joist that I am trying to check for a new load applied, however in the CSA S16 one of the components for the compressive resistance equation is Ω which is based off the principal coordinates of the shear centre. I've checked with a solver and it provides a shear centre for a single angle but not for back to back angles with a gap.

I've ran through the calculation in clause 13.3.2 using the shear centre for a single angle in the equation to find Ω and receive a similar result to what is shown on the shop drawings, but if I follow this same approach for an angle provided in the double angle strut tables I do not receive a similar result.

I've also ran the compressive resistance of this shape in RISA-2D and the compressive resistance is not the same as the shops provide.

Joists were provided by CAN-AM if that is a help to anything.

Thanks,

 

[rb1957]

the same angle, back-to-back ?

shear center at the centroid, no?

"Hoffen wir mal, dass alles gut geht !"
General Paulus, Nov 1942, outside Stalingrad after the launch of Operation Uranus.

 

[KootK]

Horizontally, I'd expect the shear center to be located at the geometric center of the combined cross section. That, owing to symmetry.

Vertically, I imagine that the gap does change things a bit as the spread between the vertical legs allows those legs to for a more effective torsion couple.

This may be somewhat of a moot point in that I would not expect conventional connections between back to back angles to produce a composite torsional response.

 

[rb1957]

surely both orientations would have shear center at the centroid ?

"Hoffen wir mal, dass alles gut geht !"
General Paulus, Nov 1942, outside Stalingrad after the launch of Operation Uranus.

 

[KootK]

surely both orientations would have shear center at the centroid ?

I don't believe so. Vertically, I anticipate that the shear center would be skewed towards the horizontal angle legs.

 

[EngDM]

I agree with KootK. I agree that the X distance (assuming X axis is parallel to the horizontal legs) would be at the centroid, aka x=0, but as for Y, a single angle has it thru the horizontal leg for equal leg members. Here is a link where one of the responses illustrates it:

https://engineering.stackexchange.c...ation-of-shear-flow-shear-center-of-l-profile

.

My angles would be back to back with the legs facing downwards towards the webs of the joist.

 

[rather_be_riding]

I would expect the shear centre to be located midway between the two angles horizontally and roughly midway through the thickness of the horizontal leg. I suspect you should be able to calculate the shear centre for a single angle in the same orientation and use the vertical location for your back-to-back arrangement but haven't had time to verify this to be the truth. As it is the point about which a load will not cause section twisting, it seems logical to me that there should be no vertical change of that location for a back-to-back arrangement unless there's some means of coupling them together. If they're coupled together, you have a different cross-section to consider anyway.

 

[EngDM]

As an update, upon researching further and seeing what RISA 3D uses in it's shape property calculations, it appears that for back-to-back angles (gap or no gap) it considers the distance from the underside of the horizontal flange to the centroid of the combined shape for the yo (shear centre in Y) distance. For xo, for symmetrical shapes about the Y axis the xo shear centre is zero as it lies on the axis.