Capacity of a bolt through a flitch plate

[SeanTee]

I have a U bracket connecting to a flitch plate with bolts and I'm trying to understand what my bolt capacity will be.

To me it seams that I cant just use a standard shear capacity because the steel members are separated by 1.75" wood members.
I haven't been able to find any guidance on this. Any suggestions?

Thanks!

 

[phamENG]

NDS has the values you need. Steel isn't going to control the connection (unless you're bolting a 3/16" plate to an 8" thick piece of wood - then maybe). NDS has reference design values for bolting wood to steel for use in their dowel type connection design equations that account for the eccentricity.

 

[SeanTee]

Sorry I don't think I gave a good description of my system. I have a beam made up of (2) 1.75" LVL with a 3/8" flitch plate sandwiched in the middle. That beam is connecting to a steel U bracket made up with 1/2" steel.

Wouldn't the flitch plate make it so that I could get a greater capacity out of the beam connection than just a wood beam?

 

[Agent666]

I'm assuming the u bracket is on the outside of the timber members. So the connection interface is between the bracket and the timber, of which the timber will govern the capacity.

Now you can cut back one of the timber members and connect steel to steel which will then get you more capscity, you may need sufficient bolts at the end of the timber to transfer the loads in the timber to the steel.

 

[SeanTee]

How would you calculate the capacity of a similar connection but instead of wood you had Styrofoam?

 

[Celt83]

review TR12 from the American Wood Council: Link

Is this connection for uplift, if not the "U" saddle would be a bearing support and you'd only need lateral positive attachment which can be done with vertical slotted holes in the "U" and lag screws into the wood side members.

Open Source Structural Applications:

https://github.com/buddyd16/Structural-Engineering

 

[phamENG]

SeanTee - I think your logic is a little flawed there. You would have to design the joint such that when the bolt bends (because it will) none of the Styrofoam is crushed. So you'd end up with a massive bolt - larger, even, than with wood plies.

Celt83 - that's an awesome reference. Thanks for posting that.

Here's a clip from the NDS. It doesn't apply perfectly (not sure this is specifically covered in a "code" anywhere), but I think it's a sound approach to the problem.

 

[KootK]

If it is truly the case that you mean to transfer your beam reaction through the bolt(s) and into the sides of the U-bracket then I would view it like this:

1) If you're assuming that any of the load is carried by the wood plies, you'd need to transfer that load over to the steel plate ahead of the U-bracket.

2) The bolt ought to be evaluated for beam bending assuming a central point load an ~4" span length.

I don't feel that it's wise to consider both the LVL's and the steel plate as working together in delivering the load to your U-bracket bolts. The wood will have more give and, owing to constructional processes, is likely to have holes not in contact with the bolts anyhow.

 

[SeanTee]

I think I understand now.
That TR12 really helps to visualize what is happening for the different limit states, and doing a calculation for the bolt acting as a beam gives me the comparison I was looking for.

Thank you!

 

[JLNJ]

I agree with KootK that your system will be trying to transition from a shared, composite system in the interior of the span to the plate taking all the load toward the ends and delivering the shear to the support bracket in the plate only.

 

[dhengr]

SeanTee:
Instead of a “U” shaped end bracket, why not use two .25” thk. angles, back to back, but with about .44” (that’s .375” + .0625”) btwn, the angle legs, to fit over/around the .375’ flitch pl. Then, bolt through the flitch pl., the angles and the two LVL’s. The LVL’s must be notched out, by .25”, on their inside faces to fit around the angle legs, but now the faying surfaces btwn. the flitch pl. and the end connection steel are actually mating and should act like a shear/bolt bearing connection. And, the LVL’s may provide some reaction too, as long as the canti. bolts allow. The angles may be welded to an end pl. which then bolts to the other structure.

 

[SeanTee]

dhenger: I like that idea I'm going to give that a go and see if Architect and builder will go for it. Thanks!

 

[KootK]

That is clever but it basically takes us back to Agent666's solution, right? Just trim the LVL back and connect to the plate for gravity and uplift?

Will this connection be visible and have aesthetic requirements necessitating that the LVL run the whole way?

What are we tying into anyhow? Sitting on top of a timber post? Hitting the side of a steel column?

 

[SeanTee]

The beam extends beyond the post for a large cantilevered overhang. That is why I didn't think that cutting it back would work. I also was trying to understand how the connection worked, which was why I brought Styrofoam into the picture.

It will be connected to the top of a steel column.

 

[KootK]

The cantilever changes things as the negative moment region over your column will then be the location of greatest torsional stability demand. The last thing that you'll want to do there is disrupt the LVLs which effectively are your LTB resistance. I'm liking your original saddle connection best a present.

 

[KootK]

Is your difficulty in absorbing the gravity reaction, the uplift reaction, or both? You got stuff sitting on top of this thing or tying into the side?

 

[SeanTee]

Its both gravity and uplift. Its a rafter that will be fully laterally supported on the top and bottom from 2x roof purlins and celling joists.

There aren't many walls to speak off that can provide shear capacity so I'm trying to get some moment capacity out of the connection. That's how the flitch plate entered the equation. Does that seam like a screwy thing to do?

 

[KootK]

Does that seam like a screwy thing to do?

I'll reserve judgment until I see your answer to my nest question.

Does that seam like a screwy thing to do?

There aren't many walls to speak off that can provide shear capacity so I'm trying to get some moment capacity out of the connection.

So this flitch beam / steel column / joint assembly is really part of a moment frame used to resist lateral loads?

 

[SeanTee]

That's the idea.
By the way if it wasn't obvious I'm an EIT, and I appreciate that you are taking time with this.

 

[KootK]

Whaat???? You mean you don't have a PhD in flitch beam moment frame design?

No sweat. Helping out rookies is a big part of what this forum is all about. Additionally, their fresh eyes on problems helps us all to hone our skills.

Yeah, what you're trying to do is a little screwy. Stated with greater compassion, it's creative and highly unconventional.

Where I think the concept may break down is in the stiffness of the moment frame joint. You may well be able to make a go of things for connection strength but, for drift control etc, you really also need connection rotational stiffness. And that will be tough to do with bolts etc given the natural over sizing of the bolt holes and the general detailing requirement for the flitch plate to be recessed into the beam at both the top and the bottom. With respect to stiffness, you'd probably be better of with distributed, smaller fasteners like nails or timber rivets. I think that even those solutions may prove difficult to execute though.

All that said, if there's a slick way to do this, you can bet that somebody here will upload a sketch of it by week's end.