Connections to Brick Walls

[abusementpark]

What do you guys do when you are working on a renovation job and need to attach to an existing load-bearing, multi-wythe brick wall? It seems that it is very hard find well-established nominal values.

It seems that companies like Simpson or Hilti will deem a lot of their fastener products suitable for connection to brick, yet do not offer any allowable values like they do for connections to other materials. I think the only fastener I have seen them give values for is a drill-and-epoxy threaded anchor with a screen tube. What about using expansion anchors or screw anchors? I haven't seen any published values on those.

Also, what about thru-bolt connections to brick? Anyone have a way getting allowable values for that?

Any help would be greatly appreciated. Architects are constantly created situations where you are forced to attach to brick.

 

[JAE]

I think you have hit the right items: screen tube/adhesive anchors (such as Hilti HIT HY20) and through-bolting.

I think the key on attaching to another building is that if the load is small, the little 1/2" screen tube anchors might work for you.

For more substantial loads, you often are really attempting to tie two diaphragms together and perhaps through-bolting where you attach to structure on the other side of the wall is better.

Brick, especially old brick, has very high variable competence and strengths so relying on that is, to me, frightening at times.

 

[concretemasonry]

One of the reasons you will not find good figures is that the properties of brick are highly variable. Meeting the ASTM minimums do not give a good picture of the real works.

Because the raw materials (clay or even shale) are highly variable the manufacturing processes make it very easy for many brick to far surpass the minimal ASTM, but the old system of minimums still are used, but the averages or maximum for "brick" can be much higher. Attaching to brick depends on the brittleness, shear and tensile strength of the brick in place. Through bolts or epoxy/sleves do transfer the loads, but some brick (strong and hard)can be very difficult to drill and may split later in the attachment process.

When it comes to attachment to a brick wall there are many factors affecting the results and mortar is a key factor. Shear and bond do not always follow and have a correlation with the compressive strength. Even the ASTM C270 spec for mortar recommend using the weakest mortar necessary to carry the structural loads, but with an added attachment, the loads are changed.

The most uniform material in the brick wall may be the mortar, but few engineers will attach to it because it seems weaker, but does hold the assembly together to create structural element (wall, column, etc.).

Dick





Engineer and international traveler interested in construction techniques, problems and proper design.

 

[concretemasonry]

I assume this is a bond brick wall and not a wall of many wythes that may be connected with some type of a shear connector (collar joint, brick or steel) that is a part of the load resistance mechanism.

Dick

Engineer and international traveler interested in construction techniques, problems and proper design.

 

[BAretired]

If the load is appreciable, you may want to remove a few bricks and provide a concrete pad to spread the load over the existing brick. An allowable bearing pressure of 100 psi is a value we used in Toronto, Ontario years ago and is pretty conservative.

BA

 

[JAE]

concretemasonry - his original post stated multi-wythe.

 

[concretemasonry]

Some walls referred to as "multi-wythe" because of regional differences in terms that often do not agree with the code language or terminology.

Multi-wythe (multi-leaf) walls are common, but the Flemish bond (or similar) is usually recognized as a bonded wall. Multi-wythe walls can be classified as solid masonry walls depending on the construction and details. In some areas/juridictoions, a full collar joint and some light horizontal reinforcement qualifies as a solid bonded masonry wall.

I used the second post since he did not refer to the wall as a clay brick masonry veneer, but recognize that there could be some assumptions made.

It is important to determine (not assume) the properties of the units in the wall that you are attaching to. If you are through-bolting, it is not as critical. Certainly, a detailed description of the wall materials and construction would eliminate any confusion or miss-interpretation.

Dick

Engineer and international traveler interested in construction techniques, problems and proper design.

 

[msquared48]

My rule is never, but never use expansion type anchors in a brick wall - except CMU. I just do not trust old mortar.

Epoxy, epoxy, epoxy.

Mike McCann
MMC Engineering
Motto: KISS
Motivation: Don't ask

 

[abusementpark]

Thanks for the responses guys. To clarify, I was referring to an unreinforced, multi-wythe brick wall (i.e. just a build up of clay brick units and mortar).

I know that thru-bolts are widely used for brick questions. How do you determine the capacity of these connections though? It seems like the assumption is that a bearing failure of the brick is going to be the governing failure mechanism. However, I have two questions about that: What is a safe value to use for the allowable bearing stress on brick? And more importantly, what kind of the bearing stress distribution do you assume for a simple shear connection?

 

[JAE]

For bearing stress I go to ACI 530.

 

[abusementpark]

Does ACI 530 give guidance what kind of bearing stress distribution to assume?

 

[JAE]

They define bearing distribution for "concentrated loads" equalling 4 times the wall thickness...this is mostly used for vertical concentrated loads on a wall....using a strip 4T wide as a beam-column in design.

For your case, pulling or pushing perpendicular to the wall, I don't believe 530 has the answer.

For brick, I could see very limited distribution beyond a bearing plate with through-bolts. The vertical mortar joints in the wall aren't always staggered such that you would get a 1:1 spread in the load from a back-plate pressure.