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]

Just kidding BA (trying to get dhengr in here to tear me up)

 

[Lion06]

Sorry your daughter is sick. I hope she gets better soon.

I was saying that each plate takes Ipl/Icombined of the total load in say _klf. I was saying the load is applied to the WF and merely shares the necessary load with the plates via the weld. I say shares the load because the neutral axis is the same for all. So say Ipl/Icombined =0.2 and there is a load of 3klf, I would size the weld to transfer 0.2*3klf = 0.6klf = 0.05k/in. This is a very small number, and your approach yield more weld, so you've made me a believer.

 

[ToadJones]

with that, I am going to sleep on the floor in my daughters room.
Good night all.
Hope we helped the OP.

 

[ToadJones]

http://www.eng-tips.com/viewthread.cfm?qid=258223&page=1

GREATEST THREAD OF ALL TIME

 

[dhengr]

In order....,
Lion06 is a purrrrrr fectly nice name, nothing Toady about that. And, congrats on the P.E.
As usual, I agree with BA, except I think Q is (bf * tf/2)(h/2 + tf/4). And, the side plates certainly will improve the torsional properties of the total section, maybe that’s what he’s trying to do, but that’s a another, different, shear flow, and maybe additive.
TJ.... keep repeating that and I’m liable to get P.Oed...., and so will BA, & just ain’t worth it.
Now, I’m going back to working on my 13,000 word harangue about and for new patent holders.

 

[WillisV]

This was the original thread Lion06 was thinking of: thread507-172412 where I referenced my response here: thread507-168270

I find myself still agreeing with myself on the methods I recommended in those =)

 

[ToadJones]

dhengr
I am just having a little fun with ya...no offense intended.

 

[struc100]

Baretired,
using the shear flow k/in by formula you provided in order to design the weld would you ratio the shear flow into 3 webs based on the thicknes of plate, the center web is cont connected the remainder will design the welds for outside plates.

 

[ToadJones]

that seems reasonable, but I would try to do it by ignoring the web altogether first. If the result is a reasonable size intermittent weld, I'd call it a day.

 

[a2mfk]

struc100- what are you trying to do with this beam? Like others said, curious as to the position of the plates on the side, are you having shear or torsion problems? For moment and deflection you'd gain a lot more Ix and Sx by placing the beams on the top/bot flanges. Sure you know that, just wondering what your case is.

 

[SAIL3]

Lets say the web of the WF is not connected initially to the fla.
So I have to design this weld for the two outside plates and also the web.
Outside plate....Q1=bf/4x(tf)x(h/2-tf/2)
Web......Q2=bf/2x(tf)x(h/2-tf/2)
shear=VQi/I.......I=combined
Maybe?

 

[ToadJones]

If you don't know what you are designing, design what you know.
Considering the fact that the beam is probably already loaded, you'll have to estimate the amount of shear flow due to the loading conditions anyway. I would think the safe way to connect the plates is to ignore the web contribution, for design purposes.

 

[Lion06]

From a practical standpoint, I hear you Toad. From a purely theorectical standpoint, however, you should only need enough weld to transfer the plates portion of the load to it. I would analogize this to a flitch plate beam. The bolts are designed only to transfer the steel's portion of the load to it, not for some shear flow from composite behavior, because there is none.

In this particular case, there is no shear flow - at least not in the sense that we generally think about when considering shear flow. Shear flow happens when component parts are forced to act together, in a more efficient manner than they will if working independently. These pieces, while deflecting together, are all working independently.

 

[SAIL3]

Here is an example of a sheave beam for an overhead crane for a steel mill that I designed about 15yrs ago. It had the added compleity of a varying moment of inertia.
It was a 110ton crane...design load =2x110=220kips

 

[a2mfk]

Holy crap Sail! I thought I had a headache from my allergies and then I took a look at that....

 

[Lion06]

Sail-

I agree that you have shearflow in that instance because you're building up your "WF". That's not the case if you start with a WF.

Also, just out of curiosity, why did you use a full pen weld for the center web at the bottom and fillets at the top of the same web?

 

[ToadJones]

Lion-
I agree with you 200%.
Well said.

 

[ToadJones]

Lion-
Very common in crane industry to use full pen welds in tension zones for fatigue reasons.
I use them for crane girders routinely.

 

[Lion06]

Is your daughter feeling any better?

 

[BAretired]

struc100,

I now believe I was completely wrong about the shear in the welds. There is no horizontal shear in the welds under gravity loading because each side plate and the beam are carrying load in proportion to their stiffness. The only function of the weld is to prevent the plates from buckling away from the beam. The concept of VQ/I is not applicable in this situation.

It is entirely different when plates are added to the top and bottom flanges in the horizontal position. In that case, shear flow exists.

Also, if you have torsion or weak axis bending, there may be shear flow, bot not under pure gravity load.



BA