bearing capacity of thru-bolts on brick

[abusementpark]

Ok, here is the scenario: I need to put a channel to span across two existing brick walls and need to thru-bolt into the brick walls to support the channel.

Firstly, what would you consider a good bearing stress on the brick to use as an upper limit? Also, how would you distribute the reaction to the thru-bolts? For instance if I have 4 thru-bolts at each support side spaced at 6" o.c, the first one will clearly take the most load, but how much can I count on the other ones to take? I know this isn't an exact science, but I would like to hear what you guys think.

 

[concretemasonry]

Solid masonry wall, bonded cavity wall or brick veneer?

Brick veneer is not normally capable of resisting loads unless constructed in a special maner.

Dick

 

[abusementpark]

Well, this is an old building. It is a 4-wythe structural brick wall, approximately 14-16" thick.

 

[hokie66]

I recall using between 125 and 250 psi for bearing stress on brickwork, but that was for bearing plates, not bolts. The load taken by through bolts in shear should not differ much from that taken by bolts set in epoxy, so you might use the values by Hilti.

For a channel, remember the shear center is outside the web, so I would orient the bolts so the centroid of bolt group approximates the shear centre of the channel.

 

[youngstructural]

I assume that you are using a face-mounted beam, correct?

If so, First thing's first: Don't allow moment to be input into the wall. You should also try to ensure that you use a relatively stiff beam...

After that you should be looking into the value of bearing on the brick based on either testing or local period and condition specific stress grades. I'm affraid that we would need a lot more info to help on that end.

Three useful points for you:

1. Use rocker plates at your beam to wall end connection (to permit rotation at the support) and size your beam for a tight deflection limit, I would suggest SPAN/360 at a minimum, preferably SPAN/500.
2. If you cannot test the wall, look into visual stress grading of the brick to local codes. Often times where people want to add beams the bricks are at their weakest. Internal whythes and the uppermost courses where often made up of under-fired bricks. These bricks are much weaker and are generally not considered structural, even by the masons placing them. Perhaps especially so. They are often a lighter colour than the surrounding bricks, most often pink or sometimes slightly tan. A tell tale sign is the bonding courses (bricks turned to lay accross one or more whythes, tying them together) are darker than the surrounding bricks.
3. Hilti's HIT HY-20 or HIT HY-50 products are excellent choices; Ask for assistance selecting the correct one for your specific application. Rumour has it they are changing over to HIT HY-70 to replace both, however you need to check with your local dealer.
4. I would humbly suggest you use four bolts maximum, spaced an even distance appart, in two vertical rows. The bolts should be at least two full bricks appart, no matter what your stresses (wall capacity) tell you the spacing should be.
5. You may need to employ some repointing to get a reliable wall strength. Also, you should make a full condition assessment of the wall (including looking for broken tie bricks). Re-tying the wall together can be accomplished with several different products, but my preference is stainless steel bolts in grout tube socks with matching grouts.

Happy to help further if I can, but need more info: Pictures, location, general age, condition assessment, exposure to environment, etc.

Cheers,

YS

B.Eng (Carleton)
Working in New Zealand, thinking of my snow covered home...

 

[Lion06]

It looks like the bearing stress issue has been addressed, but I don't think anyone addressed how much load the first bolt will be taking.
You are correct in thinking that the first bolt will take most of the load. You can set up equations to find out what it will be, but you will need to know certain paramters that will not be easily available.
I asked my professor during my advanced structural mechanics course what some of these parameters represent in the equations being used. I asked because they were always given and seemed like they could be hard to come across if not given. He was at a loss. Anyway, I might assume that the first bolt takes 50% - 60% of the load, but would not use less than 25% for any of the remaining 3 bolts. I know that gives you a higher load than you actually have, but that's what I would do short of doing an analysis using the principles of mechanics.

 

[dougantholz]

A quick solution might be to open a hole in each wall and bear the channel on the brick. Quick clean, and not likely to degrade over time.

 

[msquared48]

A single channel to support a 12 to 14" brick wall seems light to me. Why not use 1 or two I beams and avoid the bolting issue, using end bearing? Shoring could be an issue, but could be done with needle beams.

How much wall is the channel suppporting? Or is it a drag link between the brick walls? Or both?

Mike McCann
MMC Engineering

 

[youngstructural]

Hold on there Kids; This new beam isn't going to be bearing brick wall itself, is it? I would be very hesitant to support long term sustained dead load on a connection as I described above...

IF that is the case (which I doubt from reading the OP) you should certainly be looking towards a combination of Dougan's and Mike's suggestions.

Otherwise I stand by my recommendation. Also note that detailed correctly all four of my bolts take an equal load, so there is no need to get complicated in your analysis.

Cheers,

YS

B.Eng (Carleton)
Working in New Zealand, thinking of my snow covered home...

 

[abusementpark]

Thanks for all the responses,

hokie66,
How do you align the bolts to approximate the shear center? The reaction forces will be transmitted from the channel to the thru-bolts at the web of the channel where they bear. I don't see what changing the bolt group configuration will do. The range you mentioned for the allowable bearing stress is what I intuitively had in mind.

youngstructural,
When you say don't allow moment to get into the wall, you mean design the beam as simply supported? I was planning on doing that and limiting the deflection as you mentioned. I am not sure if understand the rocker plate you mentioned and how they help the connection. Also, I will have to look into those Hilti products you have proposed. Is there something that would make them more fit for use than a typical thru-bolt? I understand your concerns about the wall strength, but it was definitely designed to be the main load-bearing element for the floor system, so should have constructed as such.

StructuralEIT,
Your load estimates are about what I had in mind. I just wish there were a more scientific approach I could take.

dougantholz,
The solution you have proposed is a good one, just not within the architectural requirements.

msquared48,
The channel isn't supporting a brick wall. They are adding a balcony between two brick walls.

 

[youngstructural]

Abusementpark:

I will reply to your questions point by point.

- Absolutely design the beam as simply supported, but also detail to acheive a true simple support. A rocker plate is a continuous washer, say 40mm depth by 5mm thick, which runs the length of your connection. The beam is welded or hung from the plate, and the gap between the flanges and the wall guarentees no momen input to the connection. Limiting the deflection helps, but the rocker plate ensures.

- The Hilti products have been suggested so that you get good contact between the bolt and the surrounding brick, without undue stress on the brick during construction. A tight-fit bolt is difficult to install and does not have uniform contact; Grout is almost impossible to install reliably in a horizontal hole. It's constructability you're looking for and the Hilti (epoxy-like) installation gives you this.

- Even if a wall was designed to be load bearing, the engineers of the day (and Master Masons before them) understood that their stresses were highest at the base of the wall. They therefore logically used their poorest bricks for the highest portions of the wall. Depending on where your part of the wall is (and the age of the structure) you may have very poor brick indeed, load bearing or not).

Hope that helps/clarifies,
Cheers,

YS

B.Eng (Carleton)
Working in New Zealand, thinking of my snow covered home...

 

[Lion06]

There is a scientific way to do it. I've never done it for work, but I do have an example from my advanced structural mechanics course.
Is there any chance you can post a very basic sketch so I can verify how applicable my solution is to your situation?

 

[youngstructural]

StructuralEIT:

I don't think we're picturing the same joint, and would definately like to see a sketch as well!

Either way I'd be very keen to see your solution with what you think the joint looks like.

Cheers,

YS

B.Eng (Carleton)
Working in New Zealand, thinking of my snow covered home...

 

[hokie66]

abusementpark,

Sorry for any confusion about the shear center. I thought you were spanning a channel perpendicular to the two walls, not in plane with the walls.

 

[msquared48]

Thinking outside of the box here:

What about using the channel, but removing the lower flange and embedding the top flange into the wall by removing 2 to 3" of the mortar at one level, epoxying the top flange into the motar void, then bolting the modified channel to the brick wall - no bending on the bolts here, or bearing. Could also accomplish the same thing with a bent 1/4" plate or two welded plates. Metal tab plates could be welded to the plate to suopport the structure you are attaching to the wall.

Mike McCann
MMC Engineering

 

[hokie66]

This thread illustrates the old adage of the picture being worth more than words. I have read all the posts, and the way I now understand the problem, only SEIT has made sense of the geometry.

To connect a channel to the face of a wall, I have sometimes done as abusementpark intended, with a line of horizontal bolts, not worrying about the distribution if a stiff channel is used. Another way is to weld vertical plates to the channel, with the line of bolts vertical. Still have some prying, dependent on the stiffness of the plate, but set back far enough from the jamb, that is manageable.

 

[youngstructural]

I'm with Hokie... I really wish we'd had a sketch! Half of my points do not apply, and the other half are just not as useful the way they are written.

More than a little frustrating.

Cheers,

YS

B.Eng (Carleton)
Working in New Zealand, thinking of my snow covered home...

 

[hokie66]

Modifying my last response a bit, I think this issue could actually have been described quite readily in words without a sketch. But it required thought on the part of the OP as to how others would read the post. Abusementpark probably thought it was perfectly clear, but the principle should be to assume others who read it are complete dunces.

 

[Lion06]

YS-

Here is my calculation from a homework problem. Nothing too amazing, really, just time consuming. I will say that I assumed his joint was a face plate on the wall with all 4 bolts being in a line ABOVE the channel. If they are in line above the channel, then this example is pretty similar. I didn't include the Excel plot, but as KL gets larger, the load is more "front end loaded". As KL gets smaller, the bolts are more evenly loaded.
Just for reference, when looking at the calcs,

k = the spring stiffness of the bolts (in this case). I am assuming that this would be a shear spring stiffness, which I am not sure where you would get the value from in the real world, but it's all academic when in class.

K^2 = the ratio of k/(EA of the plate).

u(x) = the displacement function

u'(x) = the strain function


Abusmentpark-

Just as a side note, the NDS recognizes this sort of condition with a reduction factor for connections. I can't tell you exactly which it is at the moment because someone has borrowed my NDS and I can't find it right now.

Also, I just want to note that if the channel is centered on the bolt group and there are four bolts that resemble an anchor bolt assembly, the bolts should take equal load. I would also like to say that if you take that approach you should be sure to detail it so that no end moment is being taken into the wall.

 

[hokie66]

And there I was giving SEIT credit for understanding the problem!