wood floor joist with notch needs shear reinforcement 1

[HouseBoy]

I have (a client with) an old row house (circa 1910) that has notched joists that are splitting at the notch.
Joists are approx. 1.75" x 10.5" at 16" o.c. spanning 16 ft
Notch is about mid depth with the split nearby (see pic)

I'm wondering about a way to provide shear reinforcement much like a external stirrup.
I can't think of a good way to install a metal strap around the joist that would clamp it together snug (needing to be tight and durable/without slipping).

Has anyone used long screws with small diameter continuous thread screws (Simpson makes 6" long 5/32" diam.) that would be installed from the bottom up?
Joist would be clamped together first and the screws would designed for shear flow.

Plywood on both sides would be another option but I'm wondering about the screw strategy as an option.
I have seen it for larger timbers but not for 2x material and wondering if others have done this.

 

[phamENG]

What's below this? Are you in a crawl/basement?

If so, unload the floor as much as possible, use a jack to push the split joist back together if needed, and install a face mounted Simpson hanger.

If not, do the same the thing but be careful where the jack is sitting.

 

[HouseBoy]

Thanks but.... Not all locations will accommodate a hanger.

 

[PropertyGuy67]

There is a way to repair a member with screws from below, but it won't bring you back to 100% capacity. If I recall correctly, the screw method gets you into the 80%-90% range. This is why my preferred recommendation is replacement or sistering with hangers. Also, the screw method results rely on specific spacing, which is contingent on the member's dimensions.

Edit: Also, the screws are installed at an angle, not straight up and down.

 

[PropertyGuy67]

What's below this? Are you in a crawl/basement?

If so, unload the floor as much as possible, use a jack to push the split joist back together if needed, and install a face mounted Simpson hanger.

If not, do the same the thing but be careful where the jack is sitting.

This method won't get you back to 100% capacity. There is still a shear plane failure present.

 

[HouseBoy]

PG67 - Do you have a reference for the "screw method" you are referring to?

 

[PropertyGuy67]

ASCE paper describing the screw method:

Link

 

[nmerr]

You could probably do a partial sistering either with 3/4" plywood or 2x members on one or either side depending on how much room you have and how much force you need to transfer and a connector to the beam (either hanger or angle bracket). Timberlok screws come in 2.5" and 4" length and if I remember correctly are stronger than Simpson SDWS (3" long) screws but I always did like the head of the SDWS. That's the way I've typically done it in the past.

 

[gte447f]

PropertyGuy67, interesting article. I only read the abstract, but it is interesting that the authors undertook a modeling and testing program for what seems like basic shear flow problem as insinuated by the OP. I am curious what is the limiting factor that prevents reaching 100% of the "composite" section? Probably not curious enough to spend the $26 though.

 

[phamENG]

You mention that some instances won't accommodate a hanger. What do those situations look like? Do they dominate the conditions through the structure?

Sorry - with the myriad hanger options out there I'm having trouble envisioning a condition at bearing for a notched joist that couldn't be corrected with a hanger of some kind. I'm really curious to see what you've got.

 

[PropertyGuy67]

Plywood has a pretty low shear strength compared to undamaged lumber. Repairing a shear rupture in wood with a material that has a lower shear strength will help some, but it's not the best choice. Example: in HouseBoy's joist, (assuming the joist is lifted back to the point where the split mates back up before any repair) a plywood panel needs to resist the tensile forces imparted as a the joist tries to re-split, and the shear forces as the lower cord of the joist tries to move to the right as it re-splits. Plywood is great in tension, but poor in shear. It would do a good job of holding the joist in its original spot, but under load it would only offer about 1/3rd the shear strength of a 2-by sister member. In other words, under load a plywood rectangle would fail quicker than 2-by lumber rectangle as you try to pull it into a parallelogram. If the wood member has already failed, plywood isn't going to help much. 2-by sistering is a better option.

 

[PropertyGuy67]

PropertyGuy67, interesting article. I only read the abstract, but it is interesting that the authors undertook a modeling and testing program for what seems like basic shear flow problem as insinuated by the OP. I am curious what is the limiting factor that prevents reaching 100% of the "composite" section? Probably not curious enough to spend the $26 though.

I bought the paper when it first came out. This is the type of failure I see the most of in older wood structures. The basis of their 80%-90% strength ceiling was empirical testing. They tested dozens of beam set ups with different screw patterns, all with the same simulated wood split, with the split located at different heights and locations. Screws just didn't get them back to 100%. Also of note, this type of split - without any repair - had less than half the load capacity of an undamaged member.

 

[gte447f]

PropertyGuy67, thanks for the information on the ASCE paper. That seems like a good reference to have for repairing split joists like the OP's.

Also, why do you say that plywood has a pretty low shear strength compared to undamaged lumber? Can you site some allowable stress values for each material that support this comment? Fv for southern yellow pine is 175 psi, and I thought that plywood is similar? What am I missing?

 

[structSU10]

for those interested, a google search of the title led me to the researchgate.net website where the paper is free:

Link

 

[gte447f]

Thanks for the link structSU10.

 

[PropertyGuy67]

Good point to bring up. I used the design value of a plywood repair (roughly 300 psi) versus the published test results on maximum shear strength (not design values) for kiln dried Douglas Fir (1100 psi), since you'd be using a designed plywood panel on a piece of fir that has already failed.

In-plane design shear capacity of structural plywood is roughly 300 psi: Link

Shear capacity of dried Douglas Fir lumber in the same plane as above is roughly 1100 psi: Link

 

[nmerr]

Example: in HouseBoy's joist, (assuming the joist is lifted back to the point where the split mates back up before any repair) a plywood panel needs to resist the tensile forces imparted as a the joist tries to re-split, and the shear forces as the lower cord of the joist tries to move to the right as it re-splits.

The plywood will resist the tensile forces trying to split the joist and I cant imagine the longitudinal shear would be the controlling factor in the side plate at the end of a member. If this is a typical uniformly loaded joist the longitudinal shear pulling on the bottom of the joist will be minimal at the ends of the member. This is assuming the member is being left supported by the notch.

The member didn't fail due to the shear parallel to the grain but due to the tension perpendicular to the grain.

 

[PropertyGuy67]

The member didn't fail due to the shear parallel to the grain but due to the tension perpendicular to the grain.

I'd tack on to the above statement as follows: At any cubic element in a wood beam in bending, the horizontal and vertical shear stresses will be equal, in order to maintain equilibrium. So if there is stress in a horizontal plane of a finite element of the wood, there will be an equal vertical stress in the wood, in order to maintain equilibrium. The wood failed how it did because the horizontal shear stress at the end of the joist (with a value of QV/Ib, driven by the governing shear at the end of the joist) grew to the point where the corresponding vertical component of stress exceeded the wood's ability to resist failure perpendicular to its grain (where it is weakest). There was no external tension force at the bottom of the joist that pulled the wood apart. The failure was driven by the shear stresses within the wood - which are maximum at the neutral axis - and the wood's inability to resist that shear perpendicular to the grain.

 

[dhengr]

HouseBoy:
I believe you can get lt. ga. repair pls. which have pre-punched nail holes, in various sizes. Some of these are shaped like an angle iron, with a large (tall) vert. leg, and a smaller horiz. leg which fits up under the bottom of your fl. jst. These repair pls. are of various lengths, and will take the shear and the splitting tension/shear perp. to the grain. You place them on the jst., jack up the jst. to close the crack (jack at several locations, not just the bearing end), and nail it off on the sides of the jsts. I would epoxy the cracks also. Even with the notches, has this happened on only a few jsts., indicating inferior jsts. at those notches; or on all of the jsts. indicating that the notches are the real design killer? I am not condoning the notches, we know they cause design and end splitting problems just trying to get you to think through the whole cause of the problem. How old are these cracks, just recently occurring, and due to new loadings, etc? When you edge nail or screw on a 2x, you really have to watch not to split the 2x, and you also want to pay some attention to staggering side nailing too.

 

[nmerr]

There was no external tension force at the bottom of the joist that pulled the wood apart.

I didn't say there was an external tensile force that split the joist. I think we are saying the same thing it is an internal tension splitting the joist when it is not supported from below. The nice thing using plywood as a repair is it can take tension and shear in both directions where a wood side member has the same weakness perpendicular to grain if you are not re- supporting the member.