TJI Issue

[XR250]

Looked at a house today with 20' span, 11 7/8" TJI230 spaced 19.2" O.C. In an open room with vinyl plank flooring, the joists were sagging 1/2".
I looked at an old TJI table and they are rated at 19'-11 at this depth and span for L/360. So basically meet minimum code - Yay!
In a current TJI catalog, they are only rated for 17'-9" (vibration controlled).
The house is 4 years old. Does anyone know when TJI changed their recommendations from span/deflection to vibration?
There was also a footnote in the older catalog about long-term creep not being considered (apparently not!)

1) This is why I f'n despise I-joists.
2) This is why i f'n despise production builders

The house supposedly has a 10 year structural warranty. I imagine they will do their best not to help this homeowner.

 

[XR250]

I agree. I don't like them in crawlspaces (or in general). Most crawlspaces around here are encapsulated these days so moisture is typically not an issue.

 

[gte447f]

@XR250 I understand where the homeowner and you are coming from, but in my opinion you haven't yet actually established that there is a problem with this floor. The only code requirement is L/360 for live load. There is no IRC code requirement for immediate or long term dead load deflection. I know that you know that, and I understand that you are basically operating under the assumption that if the floor has deflected L/480 under dead load only, then it will definitely deflect more that L/360 under full live load. You might be right (you probably are), but it is not easy to prove without a load/deflection test (which is never going to happen), and it isn't always as obvious as it sometimes seems at first anyway, depending on the amount of dead load and existing live load and the amount of creep deflection already present.

I used to use this logic (if it has deflected "x" under just dead load then it will deflect way more than "y" under full live load), but I have refined my approach to this problem over the years, and especially my approach to how I communicate this to homeowners. Now, unless the deflection is grossly excessive, I treat the subject pretty delicately, focusing on the fact that it is likely only a serviceability issue, and the building code is primarily concerned with safety rather than serviceability.

P.S. I hold a pessimistic suspicion that the lack of a limit on dead load deflection in the IRC is intentional, so that it will be difficult for an owner or end user to prove that a floor deflects too much.

 

[XR250]

@XR250 I understand where the homeowner and you are coming from, but in my opinion you haven't yet actually established that there is a problem with this floor. The only code requirement is L/360 for live load. There is no IRC code requirement for immediate or long term dead load deflection. I know that you know that, and I understand that you are basically operating under the assumption that if the floor has deflected L/480 under dead load only, then it will definitely deflect more that L/360 under full live load. You might be right (you probably are), but it is not easy to prove without a load/deflection test (which is never going to happen), and it isn't always as obvious as it sometimes seems at first anyway, depending on the amount of dead load and existing live load and the amount of creep deflection already present.

I used to use this logic (if it has deflected "x" under just dead load then it will deflect way more than "y" under full live load), but I have refined my approach to this problem over the years, and especially my approach to how I communicate this to homeowners. Now, unless the deflection is grossly excessive, I treat the subject pretty delicately, focusing on the fact that it is likely only a serviceability issue, and the building code is primarily concerned with safety rather than serviceability.

P.S. I hold a pessimistic suspicion that the lack of a limit on dead load deflection in the IRC is intentional, so that it will be difficult for an owner or end user to prove that a floor deflects too much.

Fair enough although I don't think I ever said there was a problem with the floor. I usually set a hard limit of L/240 total load & 360 LL - mainly to try to get something existing to work but never anything close to that in new designs.