Designing Wood Splice for Section Capacity 1

[BridgEI]

I am currently designing a repair for a wooden structure located in a park. The wooden member being repaired is composed of (2) 2x12 timbers with 1/2" plywood in the middle that has deteriorated due to the flashing coming off and the prolonged exposure to moisture. The repair would replace the bottom section of each member with an equivalent section that is spliced onto the existing. My problem is that I would like to design the splice for the section capacity of the existing to avoid having to do a full analysis of the structure.

I have created a sheet that will calculate all of the dowel yield modes for nearly every connector in the NDS and calculate the LRFD design stresses of the member. The tension, compression, and shear are all very straight-forward, but I am having trouble with the bending. If I use the elastic method and calculate the force based on a moment derived from the F'b x Sx, an equivalent shear at the connection's eccentricity, and the polar moment of inertia, then the shear demand in the bolt is an order of magnitude larger than the design tension & compression and 2 orders larger than the design shear. I tried using the instantaneous center of rotation method based on the fastner's capacity, and although the shear demand is decreased, it still exceeds the capacity of the bolts in the NDS dowel failure modes. If I take the moment calculated from F'b x Sx and divide it by the vertical spacing between the top & bottom bolts, I am getting something much more acceptable that falls below the NDS dowel failure modes, but I understand this method may be unconservative, even though this moment would likely not act in conjunction with another force besides possibly the compression.

Any advice on how to resolve the above issue? I did a preliminary analysis and got a lower bending demand stress, but there's not really a way to take that design stress and use the ICR method with it like I can for the elastic method.

 

[jayrod12]

Wood is one of those materials, especially when it's just a couple 2x12, just replace the entire thing. You're going to spend a couple hundred in fees, and another hundred bucks in screws and steel plates, just to make a splice that won't perform well from a stiffness standpoint. To save having to shore up a bit more than originally anticipated.

Remove the rotted beam, replace in full. That's the easiest and cheapest solution here.

 

[BridgEI]

jayrod12: Thank you for the response. That may ultimately be what the park decides to do after we've given them a recommendation. The members being replaced are really beam columns, where all of the members are inclined up together to form essentially a wooden teepee with a roof on it. I think your point about a lack of stiffness has a lot of merit, but it also makes me consider whether essentially a steel sleeve would work for the splice, where the bolts aren't really doing the work for bending. As long as the sleeve covers enough of the old/new, wouldn't it act to transfer the bending stress if positively attached for the tension/compression?

 

[jayrod12]

Theoretically yes you could design it to transfer the tension compression loads appropriately. The issue still is there regarding stiffness. Wood splices are often lacking stiffness due to the inherent looseness in the connections. That is even more of a concern if the connections can get wet. The likelihood of them loosening over time is high. Therefore any splice detail would want many many small fasteners as opposed to just a couple large bolts.

 

[BridgEI]

Based on my initial design, uncoupling the moment between the top & bottom bolts and designing for the greater of the resultant force of that moment or the largest magnitude of the tension or compression & shear, I ended up with 3 rows & 3 columns of 1 in. bolts. If other fasteners were used, you're right that there would be many, many of them (not that (18) total 1 in. bolts isn't), probably fairly widely spaced with a steel plate connecting them all, which probably isn't economical and likely doesn't provide a reliable connection anyway in terms of overall member stiffness. I appreciate the advice. Now, I just need to find a good way to break the news as it is a fairly unique structure that park-goers would likely miss.

 

[phamENG]

What if you worked with an architect to reuse the part above the damaged section? Not knowing what it is or what it really looks like, I'm not sure what it could be, but there are probably options available.

 

[BridgEI]

phamENG: That was the original intent. We aren't working with an architect at this time, but the structure was originally drawn by an architect and then engineered. The working idea was to cut off the structure below the edge of the roof, which is where the weathering line is located, and replace the entire bottom section of the member with something equivalent, splicing the new with the existing. Obviously, that is the preferred option, but only if we can design something that is structurally sound as well as somewhat economical ("somewhat" because it is a fairly unique structure)

 

[phamENG]

Here's a quick and ugly sketch. Rather than trying to re-establish stability through a wood moment splice (not likely to last too long in an outdoor park environment with a history of inadequate maintenance practices), create a new structure beneath it that can prove a stable mounting point. Whether you form some sort of space truss like the one in my sketch or use concrete or masonry for cantilevered walls, there are ways to do it that don't rely on wood to wood splices (yuck).

 

[BridgEI]

At this point, I wonder if the simplest option isn't just to sister either side for the entire length of each member and provide enough fasteners for load transfer throughout the whole length. The bottom main member on each one could be replaced, and you would still get continuity with the inner members being sufficient on both sides of the "splice" location and the rigidity that comes with that continuity. Essentially, turning the entire existing inner portion into what that 1/2" plywood is for the members now.

 

[phamENG]

If there's room that could work. Or just go one at a time and replace them so you don't have a bunch of built in termite bait. Why is there plywood in there anyway?

But without a picture of the thing we're just throwing stones down a well and hoping they land on top of a hill somewhere.

 

[BridgEI]

The below is essentially what we are looking at. In the center of the structure, there is a fire pit with a chimney that vents out of the top. The weathering line is where the member would be spliced. At the bottom are inclined concrete pedestals with steel bent plates that retain the base of each post. The number of posts is not accurate. There are actually 12 posts, which is why we figured we could repair them one at a time since the tributary for each post is relatively small.

 

[jerseyshore]

No way I'd do a splice job for something like this. Replace the members or if you can't, sister with new pieces full length. You can get 3.5" PSLs that will fit nicely to replace the 2-2x's + 1/2" plywood setup you already have there.

 

[KootK]

I feel that these things ought to be able to be spliced using a scheme on the order of what I've shown below if we get your design approach dialed in. With these sistered wood moment connections, usually the key is splitting your moment connection into two, spread out shear connections as shown.

This would be butt ugly but, at the same time, I suspect that it would be the cheapest of your available options, including outright replacement. Mucking with the connections at either end of the full member will usually cost $$$ for something like this.

 

[Tomfh]

Don't get too smart. Go with full length splices.

 

[jerseyshore]

This is related, but has anyone actually seen any true splices that have held up for a long time? I'm not talking about staggering of a ply or two for a typical 3-2x8 basement girder (which I've seen plenty of issues with those too), but an actual moment type splice like this.

 

[KootK]

I feel like I've probably done this a couple of times a year for the past 20+ years. Lots and lots back in my wood truss days in the late nineties. None have come back to haunt me but, at the same time, it's not as though I've been touring North America annually to check in on them.

On the spectrum of "ballsy things I do for cash", this barely registers.

 

[dik]

Just as a caution... You have to be careful about including material that may contain dry-rot. It can lose 15% to 20% of its strength, and appear to be normal.

-----*****-----
So strange to see the singularity approaching while the entire planet is rapidly turning into a hellscape. -John Coates

-Dik

 

[jerseyshore]

I'm not sure whether or not we need to see the full rankings of "ballsy things KootK does for cash"...probably should, just for science.