Shear Flow for welds 1

[struc100]

I have a beam need to be reinforced , I have to use two side plates as shown in the attach sketch , the plates are the same size welded to a wide flange beam ,is the weld calcs shown in sketch correct ? specialy the Q calcs for VQ/I


Thank You

 

[ToadJones]

Hmmmm.....
Trickier than it seems at first glance, huh?

I do not think your "Q" calc is correct. The first moment of area of the side plate about the N.A. of the composite section = ZERO.
Which means your shear flow would be zero.

For design purposes, I would ignore the wideflange web completely and design the welds as if you were building a box shape member. That way you can use "Q" of the top/bottom flange. Y-bar would be to the centroid of "connected part" which would no be the flange.

I may be off base here....getting late.

 

[Lion06]

This is a trick question. There is no shear flow along that plane. This is not a composite section in the way you typically think of it. The neutral axis of all three parts are the same, therefore the welds only ensure that the pieces deflect together, nothing more.

To convince yourself, just look at Ipl + Ipl + IWF and you'll get the same answer as if you try to look at this as a composite section.

If you want to put numbers to this weld, I would determine how much of the load each plate takes (in plf), then size a skip weld to transfer that force (plf) into each plate.

 

[ToadJones]

Lion- a quick search of the forum and I see why you know of this trick!!!

I'd still use my method to design the intermittent weld (so you can design to something) unless it comes out unreasonably large. But then again, that's just me.

 

[ToadJones]

Lion are you the man formerly known as StructuralEIT/ PAStructuralPE ?

 

[Lion06]

Yes, I had a similar question about 4 years ago when I first started working.

I think your approach is reasonable. It will probably result in very low shear stresses since the plane is so far from the neutral axis.

I think I would probably still check it for how much force needs to be transferred into the plates. I only say that because that's the force transfer mechanism that I see.

 

[Lion06]

Yes, I was getting some flack for the long name PAStructuralPE, so I decided to shorten it.

 

[ToadJones]

I thought my method would do just that no?
I thought it would result in a conservatively high shear flow value, no?

When did the PE come?
I never congratulated you
Next Name "PAStructuralNittanyLionPE,SE" ?

 

[ToadJones]

I can't believe you caught flack for that name....I thought it was reasonable....I never really caught any flack for my name.
Of course, mine has a really nice ring to it.

 

[Lion06]

I might check it both ways and use the higher of the two. The shear flow check might be very low because the shear plane under consideration is so far from the neutral axis.

I got my PE results in January. No new names for me. I just wanted to get rid of the EIT, but StructralPE was already taken.

 

[ToadJones]

Lion-
Wouldnt having the plane far from the N.A. increase the shear flow......VQ/I?
Larger y-bar = Larger Q ---> Larger shear flow?

 

[JAE]

ToadJones, you have to have area "upstream" from the connection (i.e. on the opposite side of the weld from the NA).

In this case there is no area upstream. The weld is at the outermost corners.

 

[ToadJones]

we are on different pages here.... I am suggesting the you design the weld as if this were a regular built-up box section.

 

[Lion06]

Toad, the farther you get from the neutral axis of the composite section the lower the shear flow is. Remember the parabolic shear stress diagram with the max shear stress at the ENA?

 

[Lion06]

I did just do a quick comparison for a 40' beam with 2 klf of loading. Your method gives higher stresses in the weld under these conditions, and it's by a large enough margin that I believe it would be true for most conditions.

I think I'll use your approach in the future, and would only check my suggestion above if I have a short, highly loaded beam.

 

[BAretired]

He chose Lion because Toad was already taken.

The two added plates are not a very efficient way of reinforcing the beam in bending. Shear flow is the correct approach in determining the weld requirement, but struc100 has the wrong expression for Q.

Assuming the side plates terminate at the middle of the top and bottom flanges, Q is (bt_f/2)*(h/2-t_f/4). The shear in both top (or bottom) welds at any point along the beam is VQ/I where I is the moment of inertia of the combined section.

BA

 

[ToadJones]

to be honest, I am not sure what method you were using (excuse me here...i am popping in and out,...my two yr old daughter is terribly sick and i have been watching over her here)

 

[ToadJones]

BA-
I swear dhengr is rubbing off on you, you never used to be a smartA$$.

 

[ToadJones]

BA-
I swear dhengr is rubbing off on you, you never used to be a smartA$$.

 

[BAretired]

Sorry, Toad.

BA