Floor Framing with Single Angles

[StrEng007]

Good afternoon,
Is anyone familiar with floor framing utilizing single steel angles? Specifically where the angle is used as a joist and the vertical leg of the angle is turned up and stich welded to a flat plate used as a platform? I'm assuming this creates some sort of composite system where the plate becomes a flange (similar to a t-section, except you have a horizontal lip at the bottom of the T). I've seen these details used in naval architecture for the design of steel platforms.

Looking for any codes/design resources if anyone is familiar with this.

 

[KootK]

I think that what you're seeking is guidance on the design of a "stiffened plate". Resources that come to mind:

a) Blodgett.

b) Industrial codes for stuff like tanks, stacks, hoppers etc.

I imagine that the key parameter of interest is the effective width of the plate in compression.

 

[StrEng007]

KootK,
Yes, the effective width is a key question. The other part is incorporating the design checks from ASIC 360 Chapter F10 into the mix. I believe the floor plate will allow me to utilize the geometric axis approach only. However, my concern is the applied force will not be concentric to the shear center of the angle. That'll introduce torsion and likely force the angle into principal axes bending.

I'll check out the references you mentioned.

 

[KootK]

However, my concern is the applied force will not be concentric to the shear center of the angle. That'll introduce torsion and likely force the angle into principal axes bending.

If I understand you situation correctly, you're going to wind up with a bunch of channels rather than angles. And those channels will wind up being torsionally restrained by the floor plate. I would expect those two things to ameliorate your concerns substantially.

 

[StrEng007]

It's likely that I'll need to make this work with angles to match existing conditions.

Blodgett provides great information for stiffening a panel and determining the geometric properties of the section. The stiffener spacing is used as the effective width unless that spacing becomes too wide. I'm going to determine an effective width based on a reasonable width at which the plate would buckle if there was no angle leg present.

With Blodgett, everything calculates down to allowable stresses. The AISC approach, being Allowable Strength, makes it extremely difficult to pull out an equivalent stress unless I calculate my worst case limit state and divide out the section modulus of the angle.

 

[KootK]

I'm going to determine an effective width based on a reasonable width at which the plate would buckle if there was no angle leg present.

The plate wouldn't buckle if there were no angle leg present.

You might use the AISC b/t ratios as a means of determining effective width in a way that's conducive to LRFD checking.

 

[StrEng007]

I was referring to the plate compression checks from Blodgett. There is a theoretical width at which a plate that is simply supported on each end begins to buckle. I would use the effective width as the simple span. This is shown in Blodgett's welded structures book, Section 2.12. I realize the entire plate wouldn't buckle without the angle, but I'd like to assign some sort of limit and value to the effective width I'm willing to use.

Good call on the b/t ratios as well. I'll probably explore both of them.

Situations like these make me appreciate the old ASD methods even more.

Any thoughts on shear centers not lining up? Or concerns about twisting?

 

[KootK]

I was referring to the plate compression checks from Blodgett.

Ah.. got it.

Any thoughts on shear centers not lining up? Or concerns about twisting?

Nope. I'm comfortable with the assumption that the conversion to channels and rotational restraint provided by the plate will rectify that. I do, however, think that I would prefer two side fillet welding for this over single sided given that there will be some, nominal, torsion stuff going on.

 

[StrEng007]

So I’m clear… you mean the conversion to a channel in lieu of an angle? Or are you saying the composite section that “lies” within the floor plate will give me a theoretical channel?

 

[KootK]

Or are you saying the composite section that “lies” within the floor plate will give me a theoretical channel?

That one. Not that I don't also like small channels for this.

 

[le99]

I suggest flipping the alternate angles so the effective width is simply equal to twice the distance of the adjacent angles.