Wood Truss Buckling in Existing Building

[CBSE]

Have an investigative project. A 2-story wood structure that the owner called and said the second story wall shifted outward from the building. I spent about 2 hours looking at different structural aspects and then decided to look at the trusses a little further after not finding any thing structurally out of order or looking stressed (other than the obvious fact that the wall had rotated outward). What appears to have happened is most likely during the heavy snow we had last year, the bottom chords of these trusses buckled under the weight and pushed the wall out. The owner is just now noticing the movement due to a light fixture cover falling down.

The wall that has rotated outward is leaning about 3/8" in 48", so it's a pretty good sized deflection. The total height of the studs is 10ft. The trusses span 54ft, there is a single rat run down the middle of the trusses that runs nearly perfectly perpendicular. Has anyone seen this type of failure before? Questions:

1) Is it possible to "pull" the trusses back into place?
2) If it's not possible, is the best solution to add bracing to the trusses to keep them from buckling more? Or replace the trusses all together?
3) It doesn't appear that the tenants are in imminent danger at this time because the roof is not loaded, other than dead load, however, it seems if there was a sizable wind event, the suction force on that end of the building could potentially exacerbate the situation. Evacuating the end of the building that has the issue may be in order until it is fixed?

Attached is a section of the trusses.

 

[atrizzy]

Are you sure the tension chord 'buckled' and didnt' just slide? Is there any way this could be racking?

 

[jayrod12]

And this ladies and gentlemen is why specifying your truss bracing, specifically the diagonal bracing from the bottom chord to the top chord is so important. Is there some form of rigid ceiling attached?

1) I would say yes you could theoretically push the trusses back into place, however the feasibility of that being at the top of a second floor wall may be questionable. It will likely take a good chunk of force to push them back in place.

2) Whether or not number 1 is possible, you should be adding bracing, especially if there is no rigid ceiling. Take a look at BCSI publication titled "Guide to Good Practice for Handling, Installing, Restraining & Bracing of Metal Plate Connected Wood Trusses". This document has everything you need regarding bracing. Please note (everyone not just OP), temporary bracing of the trusses during erection is the responsibility of the contractor, PERMANENT BRACING is the responsibility of the EOR to specify on the drawings.

3) This is a judgement call. I don't think you'll get any guidance one way or the other on this item.

 

[bigengineer]

Is the roof diaphragm anchored well? I have a hard time seeing the trusses buckling in that way due to snow, as the bottom chords are in tension, and "buckling" is inherently due to compression. I would worry more about the lateral stability of the entire roof system to generate these type deflections. The cause is often important to try and pinpoint to determine the best solution.

For the wall to move, the roof diaphragm has displaced some with it, or the wall is not anchored well to the roof diaphragm. So I still would investigate connections to MWFRS if you haven't already. Pulling them back into place might be an option, but I would think about pulling from the wall. Hard to say without seeing the entire roof framing system.

Also, look at a clear load path in the bottom chord bracing all the way to roof diaphragm. Could it possibly be from the two trusses with the rafter overhead? Are their top chords not braced at all? That with a combination of poor bottom chord lateral brace anchorage could be the possible cause. As those couple trusses tend to buckle and rotate, if the btm chord bracing anchorage is inadequate, the entire truss system could begin to rotate as you've shown. From the drawing, something appears unstable and a load path is missing somewhere for resisting out-of-plane and lateral forces (maybe???)

 

[CBSE]

Racking or rolling may be a better term. There isn’t any rigid attachment at the ceiling such as gyp board. If I brace the trusses, they will be left in a stressed state. The walls perpendicular to the trusses that moved haven’t been noticeably affected.

I guess my question about the wall leaning, is what is too much? I don’t recall right off hand tolerances for walls being plum.

I’m having a hard time with evacuating and going back and forth. I will probably have them evacuate as my gut instinct is usually right.

I have the BCSI document so will be digging into that again to update my own standard details!

 

[dcarr82775]

I am skeptical of your sketch in that I find it hard to believe the bottom shifted over as indicated (not saying you are wrong, just that I am skeptical). Since there is no ceiling, can you confirm the trusses are bowed (string line between the bearings) and were not just installed out of plumb? I have run across trusses in new construction out of plumb by 3"+, several in a row. The erection crew comes along and sets the trusses a little off. Then the sheathing crew comes along they pull the trusses over to hit the sheathing joint.

There are erection and manufacturing tolerances on the trusses, even with the bow/rack/shift/whatever it is are you in tolerance?

The bottom chord needs some kind of bracing, for uplift is nothing else.

 

[CBSE]

Here is the truss plan. There isn't much detail from the existing drawings that shows the connections.

Upon inspection, each truss had (2)Simpson H2.5A's from truss to double top plate. I didn't find any that looked to be bent or stressed in any way, so I'm not necessarily too worried about the out-of-plane loading. The big concern is the truss bracing, which is the likely culprit.

 

[kmart30]

Im not sure about tolerances for walls being plumb but I tend to gravitate to the building code definition of "substantial structural damage" which gives tolerances compared to the pre-damaged condition. Ultimately its your call based on what is happening in the field and the potential for safety issues. You should be able to determine if that amount of lean meets the criteria set by the codes structural damage definition.

Ive never run into this situation and this might be a crazy idea for a fix if you determine the wall is leaning too much....Cut the joists that are bracing the wall at the top (shoring required), push the wall back to plumb (not as easy as it sounds), repair cut joists, brace trusses properly. This is obviously a shot in the dark without seeing anything but maybe an option.

 

[CBSE]

Here is my over-exaggerated diaphragm deflection image on the roof plan.

What else would cause the diaphragm to deflect at the top, which is what has caused the wall to rotate outwards, if it's not the trusses racking/rolling? The side walls are in great shape with no signs of distress. There are parts of the acoustical ceiling as well that appear to have "crushed" at the partition walls from apparent movement.

 

[dik]

The Truss Plate Institute has some excellent publications for bracing of wood trusses.

Dik

 

[bigengineer]

CBSE - I don't see an image with you last post. All I can think of to answer your question is poor connection to the side walls/shear walls. But most likely, it's insufficient lateral, permanent bracing. Maybe it took a really good straight-line wind and couldn't quite handle it without some displacement.

 

[KootK]

Based on what you've shared, my guess is that the truss racking/buckling is a failure of the roof diaphragm as it passes to the perimeter shear walls at the heels. Are the heels tall? Is there competent blocking or bracing out there between diaphragm and wall?

If my hypothesis is correct:

- Conventional truss bracing may not help you here. Typically, after erection, that bracing is about restraining individual members from buckling. It does little to address whole truss, rigid body buckling unless specifically designed to do so. And I'm not even sure how you'd do that. With nothing restraining the bottom chord, bracing the top chords to the bottom chords alone isn't likely to restrain the racking. It would all come back to the heel condition.

- I really don't know how you'd re-plumb such a large roof without at least resetting the original trusses somehow. I'm sure that the trusses themselves are fine once they've been righted. One approach might be to assess the diaphragm demands considering the out of plumbness as an initial imperfection. If the heel to wall connection can be improved such that things check out considering that initial imperfection, perhaps you can leave the out of plumbness as is.

- I would't be inclined to evacuate the facility unless there's another snow storm on the way. Chances are, the structure just moved to where it needed to go in order to develop the resistance that it needed. This just a judgment call of course. I expect that you'll get a diversity of opinion on this.

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.

 

[CBSE]

Image attached now

 

[KootK]

Actually, my theory doesn't suit the direction of wall movement that you're seeing relative to the truss rotation in your sketch. Based on the rotation shown, my theory would predict that the wall would move inwards (outwards at the other end of the building). Have you observed or measured the truss out of plumbness or is that just the speculated cause at this time? Any chance the trusses are leaning the other way relative to wall movement?

It seems to me that anything causing the bottom chords to kick out laterally would generate a force that would eventually be transferred back up to the diaphragm without producing a net force on the diaphragm or the walls.

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.

 

[CBSE]

I ran a plumbob as well as a level on the trusses shown to be leaning, and all were leaning as shown in the section view the same direction the wall was leaning. It looked as if the roof was overloaded at one time and the single ratrun of a 2x6 at the bottom chord shifted all the trusses in this particular section outward. I haven't seen this before. I do think it all comes back to the bottom chord of the trusses were not braced appropriately...i.e. someone assumed there would be a gypsum ceiling attached maybe. It was built in 1992.

 

[bigengineer]

KootK - I agree with you and still feel like it is possibly a roof diaphragm anchorage issue.

 

[kmart30]

Im interested to see what your final recommendation/repairs will be with a situation like this....

 

[CBSE]

Another possibility that popped into my mind was maybe the tension on the double top plate in the last diagram attached was too high for the load applied...i.e. they didn't provide enough nails in their double top late splice, or they didn't splice the double top plate appropriately and it's flexing more than normal.

Not sure. It's interesting. I need to do some destructive investigation at this point of the roof to plate connections and also see if I can find what the diaphragm nailing is.

 

[KootK]

It looked as if the roof was overloaded at one time and the single ratrun of a 2x6 at the bottom chord shifted all the trusses in this particular section outward.

In this case, you're claiming that the BC are displaced, not that the trusses are rotating due to diaphragm movement, right? If so:

1) What would be the mechanism for the BC displacement given that it's a tension member under snow? Tension chord buckling is all that I can think of and that's pretty exotic. It also wouldn't aggregate much force in any bottom chord bracing.

2) Any force developed in the bottom chord bracing would make its way out to the truss heels at the end of the building and back up into the diaphragm wouldn't it? If so, that's a net force of zero on both the roof and the walls.

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.

 

[dik]

Any signs of distortion at the hip portion of the roof? and, what sort of damage has occurred to the wall-ceiling interface? Is the movement recent? With the existing roof diaphragm (appears to be relatively rigid) unless there is a separation at the peak of the hip, I don't see how it could have moved. There has to be distress elsewhere for that sort of movement.

Dik