Brick Lintel Design - Torsional effects

[Robert216]

In sizing brick lintels, I have only been considering bending in the vertical leg (the "x-axis"). I started thinking and asked myself this question: At what point do I start considering torsional effects (or bending about the "z-axis") due to the eccentric loading of the brick? Is this a legitimate concern or am I just crazy? In school we learned about lateral torsional bending, but I haven't considered it in this application. Thanks in advance for any help.

 

[271828]

I don't think it's crazy at all. I've never understood this either. The angle is not loaded through its shear center, and is an open section, so should twist very easily between restraints.

Just some random thoughts:

It's interesting that the 3rd Ed. LRFD Single Angle Spec. doesn't mention torsion. It does mention angles with and without lateral-torsional restraint and gives some equations in 5.2.2, 5.2.3, and 5.3.

I guess one could calculate the pure torsion shear stress and add it to the shear stress from flexure. There also seems to be warping, but I've never seen equations for how to calc the associated normal stress. It could be approximated using mechanics, though. Imagine the section trying to twist--the legs would be bending out of their planes.

Now the tough one: twisting angle. I guess one could calculate the pure torsion part using dphi/dz=T/GJ and then calc the warping part using the physical analogy of the legs bending out of their planes.

Here's a weird one that I've been wondering about. The 13th Ed. Manual lists Cw properties for angles. Anybody here know how to calc that?!

 

[hokie66]

It's not crazy, but it is not worth worrying about. Just stick to what has been proven to work. The brick restrains the upstanding leg, and the brick is restrained by ties, assuming of course that the wall is built correctly. And in the completed case, the load on the lintel usually reduces from the construction case because of arching. So one for the use of standard lintel tables which have been successful. No need to sweat over calculations.

 

[jike]

"The brick restrains the upstanding leg"

Hokie66:

Could you please explain what you mean here?

 

[RARWOOD]

This topic has always interested me. I would say 95% of the engineers I know neglect torsion when designing angle lintels and wide flange lintels with extended bearing plates that support brick and cmu. sometimes I feel most engineers don't understand torsion design so they neglect it.

Some comments about arching action and brick restraint. Often there are control joints at one or both ends of the lintel which eliminates or reduces arching action. Charles W. Allen, M. ASCE indicated in an article, published in Civil Engineering-ASCE March 1981, that cracking often occurs in the lintel area reducing the masonry's ability to brace the lintel.

His article provideds a lintel table and some design equations for sizing lintels. I have used his table many times. However when I did, my lintels were alway larger then what others were commonly using.

 

[dik]

Jike... for torsion to 'really' develop, it is necessary for the deformation causing it to occur... often, if something is restrained the anticipated forces cannot occur. For compatibility, the wall has to rotate, too. As soon as an angle starts to rotate, the load from the brick is applied closer to the fillet and if the upright leg wants to rotate, then then it presses against the masonry... strip windows for a mall are a different problem and I don't treat them as lintels.

Dik

 

[csd72]

This has been discussed recently, check out this thread:

javascriptpenindex(450,450,'

http://www.tipmaster.com/includes/refinfo.cfm?w=450&h=450')

 

[csd72]

Sorry that was thread679-177455

 

[271828]

"Jike... for torsion to 'really' develop, it is necessary for the deformation causing it to occur... often, if something is restrained the anticipated forces cannot occur. For compatibility, the wall has to rotate, too. "

This is the EXACT argument I made (and was castigated for by many LOLOL) for CMU walls restraining torsion of a steel beam lintel below. Gotta watch those BCs in both cases. One big difference between this and the CMU is that the CMU has a good connection to the steel beam, LOL.

 

[csd72]

what 271828 says is basically right.

When the angle starts to rotate under the 'torsion', the brickwork will tend to stay vertical. The result is that the center of load moves closer to the inside face of the brick.

 

[hokie66]

In the thread csd72 mentioned, JAE said that depending on the brick to brace the lintel torsionally was like two drunks holding each other upright. He didn't like the idea, but I do. Maybe I have had more experience with that analogy.

 

[csd72]

hokie66,

depends on the length of the lintel, I wouldnt rely on the lateral restraint from the brick for anything over about 10'.

 

[AUCE98]

I was taught that the brick ties attached to the wall/substrate would prohibit torsion from occuring.

 

[DaveAtkins]

I always design a loose angle lintel for torsion. I don't think the vertical leg of the angle will be tight to the back side of the brick, so the brick will not restrain the angle from twisting.

DaveAtkins

 

[hokie66]

csd72,
I agree, longer spans need another solution. I normally would not go beyond 8 ft with a loose angle lintel.

DaveAtkins,
How do you restrain the angle at the ends?

As a matter of interest, in Australia we tend to use lintels with ribs which are supposed to help the lintel and brick to work compositely. Just Google "galintel"

 

[DaveAtkins]

hokie66,

The lintel bears on the brick at each jamb. At the end of the lintel, the reaction will consist of a vertical force, which is located at the heel of the angle, and a torque. If you divide this torque by the vertical force, and use this eccentricity to relocate the vertical force, you will find that the resultant force is located at the center of the brick at the jamb.

DaveAtkins

 

[csd72]

hokie66,

I thought the ribs in the galintel were for rendering I didnt think it was for composite.

Dave Atkins,

By this theory the same applies to when you have a backing member, the reaction still ends up at the center of the brick (which is what I was trying to get at in the last post). The one thing this ignores is strain compatability. At mid span the angle will rotate and the load will be closer to the shear center. Reaction at the center of the wall is conservative for the lintel but not necessarily for the brcks at the ends.

Pedantics I know, but worth a thought.

 

[ChipB]

I'm sure we have all done a ton of these w/o a torsional failure. And as there are so many thoughts on this, I thought I'd add mine as well:

Torsion is going to occur.

So what's stopping it from rolling over?

First: I believe the mortar bed is such, the leg of the upright angle is pretty much against the face of the brick. Masons I have seen want to make sure the brick sticks so they place a pretty good mortar bed under and behind the brick. Whether that is typical, I don't know. I'm just basing it on what I've seen.

Second: The brick ties are typically spaced vertically @ 18" o.c., and horizontally @ 32" o.c., thus providing some lateral restraint.

Third: We assume a point load at the center of the brick. It would actually be a bearing pressure. As the angle begins to rotate, the bearing pressure goes from uniform pressure, to triangular pressure, thus moving the ecentricity closer to the centroid of the angle, until it balances with the restraining forces.

That's my thoughts anyway.

However, what do you do at the supports? I haven't researched this so it might be straight forward. We always assume pinned, however, it's more partially fixed, and creates a prying action at the end. Numerous times I've seen cracks develop at this location, coinciding with vertical displacement of the brick.

 

[JKW05]

Years ago I went through various exercises trying to calculate the torsional stresses and rotations and the results indicated that lintels that worked for ages should have failed a long time ago. I concluded that the load redistributions noted in some of the posts above made the more "exact" analysis meaningless. For openings up to about six feet I usually ignore any torsion, and for openings in the 8 to 10 foot range I will often fasten the lintel to the back-up at it's mid-point to reduce the potential for excessive rotation. For openings much longer than that with masonry back-up, I use tubes with a bottom plate or angle welded to it in lieu of WF's, for better torsional performance. Just my 2 cents....

JW

 

[hokie66]

csd72,
The literature definitely says that the ribs are for composite action. They are on both the horizontal and vertical face at the mortar-brick interface, and although parallel to the wall, would give some benefit, I think.