Wooden I-Joist cracked flange

[isepousti]

I can't find much information on cracked flanges on wooden i-joists. I have a picture of a joist attached. It has a crack in the flange (bottom chord) extending from the web to bearing. Any thoughts? Is this an issue? Should these be replaced? Can/should they be reinforced?

 

[jayrod12]

There should be a manufacturer's stamp somewhere on the I-joist. I'd be contacting the manufacturer for their recommended repair process.

Alternatively, replace the joist.

By the way, I say yes that's an issue.

If I didn't have a manufacturer's recommended repair, and replacement wasn't an option, I'd be tempted to have a custom bent steel shoe made that was a snug fit for the flange and thru-bolt it. Alternatively a couple of angles with thru bolts would likely work. The only tricky part would be creating the connection to the wood plate.

 

[KootK]

I'll cast my vote for letting it stand as is so long as the crack doesn't extend very far beyond the interior edge of the bearing surface. The system has found its own equilibrium and I don't feel that the crack seriously compromises the flange's role as a bearing element. Were I to hazard a guess, I would say that the crack was likely present in some form at the time of installation and was the result of some rough handling.

If you can get the joist supplier to sign off on it, that would clearly be ideal from a liability perspective.

I like to debate structural engineering theory -- a lot. If I challenge you on something, know that I'm doing so because I respect your opinion enough to either change it or adopt it.

 

[Triangled]

could be field installation damage, or maybe could be load duress ...any signs of similar damage on adjacent joists?
If you determine it was just field installation damage, and like kootk said if the crack doesn't extend....I'd consider something like adding wood each side of IJoist at bearing, glued and screwed, something like a 2x8 oriented vertically with the bottom edge hard against the top of the bottom chord and the top edged 1/8" shy of the bottom of the top chord. The idea being that as the web tries to "knife through" the damaged bottom chord under load, the new web configuration will have to take the 2xs with it, ie the bottom of the 2xs are creating additional bearing on the top of the bottom chord so that damaged groove joint doesn't have to do all the work.

 

[jayrod12]

Koot,

What's your definition of very far?

My feeling is if the crack extends the entire length of bearing then it's an issue. I envision the plywood web wedging itself through that crack which would cause a significant drop in bearing elevation at that one joist. It's likely that the floor sheathing would span to adjacent joists however the deficiency is still there and a dip in the floor would be inevitable.

 

[jayrod12]

I like triangled's solution to the web knifing problem. I'd be comfortable with that approach.

 

[KootK]

I like triangled's solution to the web knifing problem. I'd be comfortable with that approach.

Ditto.

Koot, What's your definition of very far?

Imagine an extreme scenario where the crack is really a vertical slit sawn right below the web such that no bearing stress could be transmitted directly from the bottom flange to the web. For some value of load and length of slit, a different load transfer mechanism would be viable. The two halves of the split bottom flange would cantilever from the support reaction to an interior location where the slit no longer existed and shear could be transmitted up into the web as intended. Conceptually, it wold be like an OWSJ seat but with the joist flipped upside down.

In answer to the queston "how far is very far", I would say that "too far" would be reached when the mechanism described above ceased to work by the numbers.

Out of curiosity, what is the intended load path for bearing stress at the web/flange groove under normal circumstances? Direct bearing? Glue? Bearing and glue in concert? I feel like locally, at the bearings, you might overwhelm glue capacity and settle into direct bearing.



I like to debate structural engineering theory -- a lot. If I challenge you on something, know that I'm doing so because I respect your opinion enough to either change it or adopt it.

 

[jayrod12]

My feeling is direct bearing, and that is why I feel a crack such as this is an issue. If it were glue only, or some combo of glue and bearing than I could see the argument for the current situation being acceptable.

I grasp your premise for the definition of too far. However, in practical circumstances would you ever go that far to run those calculations, or would you jump straight to a quick easy repair such as the one triangled has suggested. My money is on the latter.

 

[KootK]

I grasp your premise for the definition of too far. However, in practical circumstances would you ever go that far to run those calculations, or would you jump straight to a quick easy repair such as the one triangled has suggested

In practical circumstances, it would go one of two ways for me:

1) if the crack didn`t extend beyond the interior face of bearing, I`d just okay it without ever having run any calcs.
2) if the crack extended beyond the interior face of bearing, I`d skip straight to the easy repair.

I like to debate structural engineering theory -- a lot. If I challenge you on something, know that I'm doing so because I respect your opinion enough to either change it or adopt it.

 

[racookpe1978]

Well, we have only one view.

This particular joist has ALREADY failed - the web is pressing down into a failed flange, and will - as pointed out above several times - only continue to push the two separate halves of the flange further apart. Jayrod12's point is a good one, but without knowing what the rest of the web is like, a solution can only be a guess.

We need a side view of the beam (another photo from the side at 90 degrees from the first), and then "might" be able to recommend two additional long thick plates glued and bolted on both sides of the original web to carry the resulting sandwich of plate + web + plate from the center of the original beam down to the support underneath the original web.

 

[isepousti]

The house is being built and they just installed the main floor joists. Only 2 of the 20+ joists are visible and both are cracked. The rest aren't visible because they've been boxed-in / crowned with LSL. The two joists are next to each other. There is almost no load on them currently and I can't tell how far the crack extends.

Although I doubt it, every single joist is potentially all cracked. I've also lost some faith in the team framing the house.

Many thanks for all the feedback.

 

[EngineeringEric]

I would block the web solid if 2x material (leaving 1/8" gap at top). say 2x6 for argument. then provide an additional 2x on the outside of entire blocked joist full height (no gap). But the best would be to get something from manufacturer and then tell us what they recommend.

And make sure to specify glue/liquid nails for all connections even if it isn't required.

 

[dcarr82775]

Could be the nail that did the splitting. I wouldn't drive in at the web/flange joint as was done, although I am surprised that would crack it. If the only two you can see are cracked, it seems likely several others are as well. I would do something similar to Eric. Block out the web each side and glue/nail it together enough to get the reaction out of the web, into the blocking, and down to the flange

 

[boo1]

looks like they missed the anchor bolt too

 

[KootK]

If it's new and it's systemic, I'd definitely a) get the supplier involved and b) establish a definitive cause.

I like to debate structural engineering theory -- a lot. If I challenge you on something, know that I'm doing so because I respect your opinion enough to either change it or adopt it.

 

[msquared48]

Or this was a bolt for something that has been removed...

I would check out the bearing stress at the end since only about 60% is in contact with the wall plate. The wall plate will be critical here.

It looks like the joist could have rolled over at some point in time. Where is the required blocking between the trusses to prevent rollover? Has that been (temporarily or permanently) removed?

Mike McCann, PE, SE (WA)

 

[Triangled]

i was kinda thinking this was a remodel and that these joists had experienced load in their history....
the 2 joists in a row is a bit of a concern, at least a clue...
if this is brand new construction such that the joists haven't really been loaded yet, seems like the damage is due to contractor mishandling. I think IJoists are often delivered nested and bundled, and if fork-lifted into place as a bundle, and while lifting, the top of the bottom flange of the leading joist strikes something or catches on something as it is being lifted - forcing the flange downward, I could see that flange damaged as this one is, and the one next to it too, depending how nested. A telltale sign might be something like a ding on the top of the bottom flange. That one corner does look a little chewed up....

 

[isepousti]

We heard back from the Weyerhaeuser engineer. Apparently this is "normal" and not an issue. They are prepared to sign-off on it and will, "for our peace of mind", give us a sketch of how they would like to see the blocking installed.

I guess I find that to be a little bit of a contradiction. It either is an issue, or it's not and doesn't require anything special.

 

[msquared48]

If this is "normal", I am, well, I can't say it here.

Mike McCann, PE, SE (WA)

 

[boo1]

normal ????