Question about an x-brace detail

[Cap07]

Hi,
I was wondering if I could get a quick opinion on the detail shown in the attached jpg. It's the connection to a concrete footing for the tension member of an x-brace. The steel rod is bent approximately 11° and welded to a steel plate that is attached to a concrete footing with wedge anchors.

I can show the rod and the weld good for the tensile load, but are there any further checks I should do because of the bend in the rod?

Thanks,
Cap

 

[csd72]

Also apply the vertical component of the force at the end of the welds to get a bending moment around the weld center.

divide this by 2d^2/6 to get the perpendicular shear stress in the welds then check the weld size for the resultant of the 2.

You will also get some tensile load in the anchors.

 

[Cap07]

Thanks for your answer csd72, I appreciate it very much.

 

[JAE]

Time out.

csd72 - I see what you are saying - but doesn't your 2d^2/6 concept assume that the tension rod is stiff in bending?...which it is not.

The rod isn't a rigid body and I would think that you'd get a very high stress concentration at the end of the weld - much higher than assuming the weld lines receive shear and orthogonal stress from a reaction eccentricity.

I see the weld simply unzipping starting at the left end and progressing down -

 

[JLNJ]

What if the angle were larger? It just doesn't look right.

If the force in the rod is big you might "unzip" the weld before the rod can distriute the force along the length of the weld.

 

[JAE]

Yes....I'd prefer to see a vertical tab plate welded to the horizontal plate - with the rod welded concentrically on the vertical tab plate.

 

[JLNJ]

Sorry about that JAE, my spidey sense didn't tell me that you were typing "unzip" at the same time I was!

 

[Cap07]

Thanks for your responses JAE and JLNJ - I can see what you're saying about the possibility of the connection just unzipping... Do you know of any way to estimate the force it would take to unzip the weld? I've got about 2300 lb tension in the rod, it's angled at 11°, giving a vertical component of about 440 lb.

Thanks for your help, I appreciate it.

 

[JAE]

I don't know how you'd analyze it as it would be dependent on the relative stiffness of the rod vs. the weld and supporing plate.

I would "guess" that all the force would initially all go into the first 1 inch of weld.

 

[Cap07]

Thanks JAE.... I'll take a look at it assuming the first inch of the weld carries the entire load.

 

[Cap07]

Would this be an acceptable way to analyze the weld:

Assume the length of the weld is 1 inch (it's actually 2+ inches). Resolve the tensile force (T) in the rod into horizontal and vertical components:

H = T cos (theta)
V = T sin (theta)
where theta = angle of rod with the horizontal (please see the jpg file attached to my initial post)

Shear stress in the weld, fv = H/Aweld

Bending stress in the weld, fb = M/S = Ve/S
where e = the distance from the edge of the weld to the center of the weld

Resultant stress, f = (fv² + fb²)^0.5

I would then compare the resultant stress to the allowable stress, Fw.

 

[Hanif08]

Cap07

You are on the right track. Here are few suggestions:

- Yes, resolve the force in horizontal and vertical components. These components will act on respective shear planes of the weld. The size of weld on these planes is a/sqrt(2), where 'a' is the effective throat thickness (ie) weld size.

- Calculate the shear stress on two planes:

fv=V/(a/sqrt(2)*Lw)
fh=H/(a/sqrt(2)*Lw)

where Lw is the length of weld

- Combine them to find resultant stress using huber-henky criteria.

f = sqrt(3*(fv^2+fh^2)) to be less than allowable stress.

In my opinion bending is not a problem once you have considered stress in two orthogonal planes and combined correctly. Although there will be vertical shear stress concentration, however vertical component of force is quite small.

 

[Cap07]

I'll check it out. Thanks for your advice Hanif08, I appreciate it very much.

 

[csd72]

JAE makes a very good point, my method is more appropriate for a rigid plate such as a gusset.

I disagree with hanif08, the vertical component is applied eccentrically and this needs to be taken into account.

 

[JAE]

Cap07 - I stated the first inch might take all the load, but honestly I don't know how long your weld is. With a very very stiff and brittle weld, along with a flexible rod like you show, the load might go into the first 1/4" of weld.

It depends on the rod diameter as well.

In all, to me it is not a very good detail and one I wouldn't waste my time on. I'd use a vertical gusset-type plate which would have more vertical rigidity to establish load distribution over the full length of weld. It would still have the eccentricity that csd72 states.

 

[LobstaEata]

Cap07

Considering that this weld is a two sided flare groove, its strength could be somewhat questionable especially if it was not backgouged properly. I'm not sure what kind of weld inspection was done, but lot of inclusions can result if it was not done properly.

If the first inch of weld cracks as a result of poor welding practice, maybe the unzipping of weld theory is possible.

One other question. Is this a wind brace subject to stress reversal? If so, perhaps its fatigue stress range would come into play?

 

[Cap07]

Thanks to everyone who weighed in on my question. This detail was part of an lateral bracing system that I was inspecting... I convinced the homeowner to replace this bracing with a pre-fabricated system, and I feel a lot better about it.